Corporate Average Fuel Economy Standards for Passenger Cars and Light Trucks for Model Years 2027 and Beyond and Fuel Efficiency Standards for Heavy-Duty Pickup Trucks and Vans for Model Years 2030 and Beyond

AGENCY:

National Highway Traffic Safety Administration (NHTSA).

ACTION:

Final rule.

SUMMARY:

NHTSA, on behalf of the Department of Transportation (DOT), is finalizing Corporate Average Fuel Economy (CAFE) standards for passenger cars and light trucks that increase at a rate of 2 percent per year for passenger cars in model years (MYs) 2027-31, 0 percent per year for light trucks in model years 2027-28, and 2 percent per year for light trucks in model years 2029-31. NHTSA is also finalizing fuel efficiency standards for heavy-duty pickup trucks and vans (HDPUVs) for model years 2030-32 that increase at a rate of 10 percent per year and model years 2033-35 that increase at a rate of 8 percent per year.

DATES:

This rule is effective August 23, 2024.

ADDRESSES:

For access to the dockets or to read background documents or comments received, please visit https://www.regulations.gov, and/or Docket Management Facility, M-30, U.S. Department of Transportation, West Building, Ground Floor, Rm. W12-140, 1200 New Jersey Avenue SE, Washington, DC 20590. The Docket Management Facility is open between 9 a.m. and 4 p.m. Eastern time, Monday through Friday, except Federal holidays.

FOR FURTHER INFORMATION CONTACT:

For technical and policy issues, Joseph Bayer, CAFE Program Division Chief, Office of Rulemaking, National Highway Traffic Safety Administration, 1200 New Jersey Avenue SE, Washington, DC 20590; email: joseph.bayer@dot.gov. For legal issues, Rebecca Schade, NHTSA Office of Chief Counsel, National Highway Traffic Safety Administration, 1200 New Jersey Avenue SE, Washington, DC 20590; email: rebecca.schade@dot.gov.

SUPPLEMENTARY INFORMATION:

Table of Acronyms and Abbreviations

Does this action apply to me?

This final rule affects companies that manufacture or sell new passenger automobiles (passenger cars), non-passenger automobiles (light trucks), and heavy-duty pickup trucks and vans (HDPUVs), as defined under NHTSA's Corporate Average Fuel Economy (CAFE) and medium and heavy duty (MD/HD) fuel efficiency (FE) regulations. Regulated categories and entities include:

This list is not intended to be exhaustive, but rather provides a guide regarding entities likely to be regulated by this action. To determine whether particular activities may be regulated by this action, you should carefully examine the regulations. You may direct questions regarding the applicability of this action to the persons listed in FOR FURTHER INFORMATION CONTACT .

Table of Contents

I. Executive Summary

II. Overview of the Final Rule

A. Summary of the NPRM

B. Public Participation Opportunities and Summary of Comments

C. Changes to the CAFE Model in Light of Public Comments and New Information

D. Final Standards—Stringency

E. Final Standards—Impacts

1. Light Duty Effects

2. Heavy Duty Pickup Trucks and Vans Effects

F. Final Standards Are Maximum Feasible

G. Final Standards Are Feasible in the Context of EPA's Final Standards and California's Standards

III. Technical Foundation for Final Rule Analysis

A. Why is NHTSA conducting this analysis?

1. What are the key components of NHTSA's analysis?

2. How do requirements under EPCA/EISA shape NHTSA's analysis?

3. What updated assumptions does the current model reflect as compared to the 2022 final rule and the 2023 NPRM?

B. What is NHTSA analyzing?

C. What inputs does the compliance analysis require?

1. Technology Options and Pathways

2. Defining Manufacturers' Current Technology Positions in the Analysis Fleet

3. Technology Effectiveness Values

4. Technology Costs

5. Simulating Existing Incentives, Other Government Programs, and Manufacturer ZEV Deployment Plans

a. Simulating ZEV Deployment Unrelated to NHTSA's Standards

b. IRA Tax Credits

6. Technology Applicability Equations and Rules

D. Technology Pathways, Effectiveness, and Cost

1. Engine Paths

2. Transmission Paths

3. Electrification Paths

4. Road Load Reduction Paths

a. Mass Reduction

b. Aerodynamic Improvements

c. Low Rolling Resistance Tires

5. Simulating Air Conditioning Efficiency and Off-Cycle Technologies

E. Consumer Responses to Manufacturer Compliance Strategies

1. Macroeconomic and Consumer Behavior Assumptions

2. Fleet Composition

a. Sales

b. Scrappage

3. Changes in Vehicle Miles Traveled (VMT)

4. Changes to Fuel Consumption

F. Simulating Emissions Impacts of Regulatory Alternatives

G. Simulating Economic Impacts of Regulatory Alternatives

1. Private Costs and Benefits

a. Costs to Consumers

(1) Technology Costs

(2) Consumer Sales Surplus

(3) Ancillary Costs of Higher Vehicle Prices

b. Benefits to Consumers

(1) Fuel Savings

(2) Refueling Benefit

(3) Additional Mobility

2. External Costs and Benefits

a. Costs

(1) Congestion and Noise

(2) Fuel Tax Revenue

b. Benefits

(1) Climate Benefits

(a) Social Cost of Greenhouse Gases Estimates

(b) Discount Rates for Climate Related Benefits

(c) Comments and Responses About the Agency's Choice of Social Cost of Carbon Estimates and Discount Rates

(2) Reduced Health Damages

(3) Reduction in Petroleum Market Externalities

(4) Changes in Labor Use and Employment

3. Costs and Benefits Not Quantified

H. Simulating Safety Effects of Regulatory Alternatives

1. Mass Reduction Impacts

2. Sales/Scrappage Impacts

3. Rebound Effect Impacts

4. Value of Safety Impacts

IV. Regulatory Alternatives Considered in This Final Rule

A. General Basis for Alternatives Considered

B. Regulatory Alternatives Considered

1. Reference Baseline/No-Action Alternative

2. Alternative Baseline/No-Action Alternative

3. Action Alternatives for Model Years 2027-2032 Passenger Cars and Light Trucks

a. Alternative PC1LT3

b. Alternative PC2LT002—Final Standards

c. Alternative PC2LT4

d. Alternative PC3LT5

e. Alternative PC6LT8

f. Other Alternatives Suggested by Commenters for Passenger Car and LT CAFE Standards

4. Action Alternatives for Model Years 2030-2035 Heavy-Duty Pickups and Vans

a. Alternative HDPUV4

b. Alternative HDPUV108—Final Standards

c. Alternative HDPUV10

d. Alternative HDPUV14

V. Effects of the Regulatory Alternatives

A. Effects on Vehicle Manufacturers

1. Passenger Cars and Light Trucks

2. Heavy-Duty Pickups and Vans

B. Effects on Society

1. Passenger Cars and Light Trucks

2. Heavy-Duty Pickups and Vans

C. Physical and Environmental Effects

1. Passenger Cars and Light Trucks

2. Heavy-Duty Pickups and Vans

D. Sensitivity Analysis, Including Alternative Baseline

1. Passenger Cars and Light Trucks

2. Heavy-Duty Pickups and Vans

VI. Basis for NHTSA's Conclusion That the Standards Are Maximum Feasible

A. EPCA, as Amended by EISA

1. Lead Time

a. Passenger Cars and Light Trucks

b. Heavy-Duty Pickups and Vans

2. Separate Standards for Passenger Cars, Light Trucks, and Heavy-Duty Pickups and Vans, and Minimum Standards for Domestic Passenger Cars

3. Attribute-Based and Defined by a Mathematical Function

4. Number of Model Years for Which Standards May Be Set at a Time

5. Maximum Feasible Standards

a. Passenger Cars and Light Trucks

(1) Technological Feasibility

(2) Economic Practicability

(3) The Effect of Other Motor Vehicle Standards of the Government on Fuel Economy

(4) The Need of the U.S. To Conserve Energy

(a) Consumer Costs and Fuel Prices

(b) National Balance of Payments

(c) Environmental Implications

(d) Foreign Policy Implications

(5) Factors That NHTSA Is Prohibited From Considering

(6) Other Considerations in Determining Maximum Feasible CAFE Standards

b. Heavy-Duty Pickups and Vans

(1) Appropriate

(2) Cost-Effective

(3) Technologically Feasible

B. Comments Regarding the Administrative Procedure Act (APA) and Related Legal Concerns

C. National Environmental Policy Act

1. Environmental Consequences

a. Energy

(1) Direct and Indirect Impacts

(2) Cumulative Impacts

b. Air Quality

(1) Direct and Indirect Impacts

(a) Criteria Pollutants

(b) Toxic Air Pollutants

(c) Health Impacts

(2) Cumulative Impacts

(a) Criteria Pollutants

(b) Toxic Air Pollutants

(c) Health Impacts

c. Greenhouse Gas Emissions and Climate Change

(1) Direct and Indirect Impacts

(a) Greenhouse Gas Emissions

(b) Climate Change Indicators (Carbon Dioxide Concentration, Global Mean Surface Temperature, Sea Level, Precipitation, and Ocean pH)

(2) Cumulative Impacts

(a) Greenhouse Gas Emissions

(b) Climate Change Indicators (Carbon Dioxide Concentration, Global Mean Surface Temperature, Sea Level, Precipitation, and Ocean pH)

(c) Health, Societal, and Environmental Impacts of Climate Change

(d) Qualitative Impacts Assessment

2. Conclusion

D. Evaluating the EPCA/EISA Factors and Other Considerations To Arrive at the Final Standards

1. Passenger Cars and Light Trucks

2. Heavy-Duty Pickups and Vans

3. Severability

VII. Compliance and Enforcement

A. Background

B. Overview of Enforcement

1. Light Duty CAFE Program

a. Determining Compliance

b. Flexibilities

c. Civil Penalties

2. Heavy-Duty Pickup Trucks and Vans

a. Determining Compliance

b. Flexibilities

c. Civil Penalties

C. Changes Made by This Final Rule

1. Elimination of OC and AC Efficiency FCIVs for BEVs in the CAFE Program

2. Addition of a Utility Factor for Calculating FCIVs for PHEVs

3. Phasing Out OC FCIVs by MY 2033

4. Elimination of the 5-Cycle and Alternative Approval Pathways for CAFE

5. Requirement To Respond To Requests for Information Regarding Off-Cycle Requests Within 60 Days for LDVs for MYs 2025 and 2026

6. Elimination of OC Technology Credits for Heavy-Duty Pickup Trucks and Vans Starting in Model Year 2030

7. Technical Amendments for Advanced Technology Credits

8. Technical Amendments to Part 523

a. 49 CFR 523.2 Definitions

b. 49 CFR 523.3 Automobile

c. 49 CFR 523.4 Passenger Automobile

d. 49 CFR 523.5 Non-Passenger Automobile

e. 49 CFR 523.6 Heavy-Duty Vehicle

f. 49 CFR 523.8 Heavy-Duty Vocational Vehicle

9. Technical Amendments to Part 531

a. 49 CFR 531.1 Scope

b. 49 CFR 531.4 Definitions

c. 49 CFR 531.5 Fuel Economy Standards

10. Technical Amendments to Part 533

a. 49 CFR 533.1 Scope

b. 49 CFR 533.4 Definitions

11. Technical Amendments to Part 535

a. 49 CFR 535.4 Definitions

b. 49 CFR 535.7 Average, Banking, and Trading (ABT) Credit Program

12. Technical Amendments to Part 536

13. Technical Amendments to Part 537

a. 49 CFR 537.2 Scope

b. 49 CFR 537.3 Applicability

c. 49 CFR 537.4 Definitions

d. 49 CFR 537.7 Pre-Model Year and Mid-Model Year Reports

D. Non-Fuel Saving Credits or Flexibilities

E. Additional Comments

1. AC FCIVs

2. Credit Transfer Cap AC

3. Credit Trading Between HDPUV and Light Truck Fleets

4. Adjustment for Carry Forward and Carryback Credits

5. Increasing Carryback Period

6. Flex Fuel Vehicle Incentives

7. Reporting

8. Petroleum Equivalency Factor for HDPUVs

9. Incentives for Fuel Cell Electric Vehicles

10. EV Development

11. PHEV in HDPUV

VIII. Regulatory Notices and Analyses

A. Executive Order 12866, Executive Order 13563, and Executive Order 14094

B. DOT Regulatory Policies and Procedures

C. Executive Order 14037

D. Environmental Considerations

1. National Environmental Policy Act (NEPA)

2. Clean Air Act (CAA) as Applied to NHTSA's Final Rule

3. National Historic Preservation Act (NHPA)

4. Fish and Wildlife Conservation Act (FWCA)

5. Coastal Zone Management Act (CZMA)

6. Endangered Species Act (ESA)

7. Floodplain Management (Executive Order 11988 and DOT Order 5650.2)

8. Preservation of the Nation's Wetlands (Executive Order 11990 and DOT Order 5660.1a)

9. Migratory Bird Treaty Act (MBTA), Bald and Golden Eagle Protection Act (BGEPA), Executive Order 13186

10. Department of Transportation Act (Section 4(f))

11. Executive Order 12898: “Federal Actions To Address Environmental Justice in Minority Populations and Low-Income Populations”; Executive Order 14096: “Revitalizing Our Nation's Commitment to Environmental Justice for All”

12. Executive Order 13045: “Protection of Children From Environmental Health Risks and Safety Risks”

E. Regulatory Flexibility Act

F. Executive Order 13132 (Federalism)

G. Executive Order 12988 (Civil Justice Reform)

H. Executive Order 13175 (Consultation and Coordination With Indian Tribal Governments)

I. Unfunded Mandates Reform Act

J. Regulation Identifier Number

K. National Technology Transfer and Advancement Act

L. Department of Energy Review

M. Paperwork Reduction Act

N. Congressional Review Act

I. Executive Summary

NHTSA, on behalf of the Department of Transportation, is finalizing new corporate average fuel economy (CAFE) standards for passenger cars and light trucks for model years 2027-2031, setting forth augural standards for MY 2032, and finalizing new fuel efficiency standards for heavy-duty pickup trucks and vans (HDPUVs) for model years 2030-2035. This final rule responds to NHTSA's statutory obligation to set CAFE and HDPUV standards at the maximum feasible level that the agency determines vehicle manufacturers can achieve in each MY, in order to improve energy conservation. Improving energy conservation by raising CAFE and HDPUV standard stringency not only helps consumers save money on fuel, but also improves national energy security and reduces harmful emissions.

Based on the information currently before us, NHTSA estimates that relative to the reference baseline this final rule will reduce gasoline consumption by 64 billion gallons relative to reference baseline levels for passenger cars and light trucks and will reduce fuel consumption by approximately 5.6 billion gallons relative to reference baseline levels for HDPUVs through calendar year 2050. If compared to the alternative baseline, which has lower levels of electric vehicle penetration than the reference baseline, fuel savings will be greater at approximately 115 billion gallons. Reducing gasoline consumption has multiple benefits—it improves our nation's energy security, it saves consumers money, and reduces harmful pollutant emissions that lead to adverse human and environmental health outcomes and climate change. NHTSA estimates that relative to the reference baseline, this final rule will reduce carbon dioxide (CO₂ ) emissions by 659 million metric tons for passenger cars and light trucks, and by 55 million metric tons for HDPUVs through calendar year 2050. Again, these relative reductions are greater if the rule is compared to the alternative baseline, but demonstrating a similar level of absolute carbon dioxide emissions. While consumers could pay more for new vehicles upfront, we estimate that they would save money on fuel costs over the lifetimes of those new vehicles—in the reference baseline analysis lifetime fuel savings exceed modeled regulatory costs by roughly $247, on average, for passenger car and light truck buyers of MY 2031 vehicles, and roughly $491, on average, for HDPUV buyers of MY 2038 vehicles. By comparison, in the No ZEV alternative baseline analysis, lifetime fuel savings exceed modeled regulatory costs by roughly $400, on average, for passenger car and light truck buyers of MY 2031 vehicles. Net benefits for the preferred alternative for passenger cars and light trucks are estimated to be $35.2 billion at a 3 percent discount rate (DR), and $30.8 billion at a 7 percent DR, and for HDPUVs, net benefits are estimated to be $13.6 billion at a 3 percent DR, and $11.8 billion at a 7 percent DR. Net benefits are higher if the final rules are assessed relative to the alternative baseline, estimated to be $44.9 billion at a 3 percent DR and $39.8 billion at 7 percent DR. (For simplicity, however, all projections presented in this document use the reference baseline unless otherwise stated.)

The record for this action is comprised of the notice of proposed rulemaking (NPRM) and this final rule, a Technical Support Document (TSD), a Final Regulatory Impact Assessment (FRIA), and a Draft and Final EIS, along with extensive analytical documentation, supporting references, and many other resources. Most of these resources are available on NHTSA's website, and other references not available on NHTSA's website can be found in the rulemaking docket, the docket number of which is listed at the beginning of this preamble.

The final rule considers a range of regulatory alternatives for each fleet, consistent with NHTSA's obligations under the Administrative Procedure Act (APA), National Environmental Policy Act (NEPA), and E.O. 12866. Specifically, NHTSA considered five regulatory alternatives for passenger cars and light trucks, as well as the No-Action Alternative. Each alternative is labeled for the type of vehicle and the rate of increase in fuel economy stringency based on changes for each model year, for example, PC1LT3 represents a 1 percent increase in Passenger Car standards and a 3 percent increase in Light Truck standards. We include four regulatory alternatives for HDPUVs, each representing different possible rates of year-over-year increase in the stringency of new fuel economy and fuel efficiency standards, as well as the No-Action Alternative. For example, HDPUV4 represents a 4 percent increase in fuel efficiency standards applicable to HDPUVs. The regulatory alternatives are as follows:

After assessing these alternatives against the reference baseline and the alternative baseline, and evaluating numerous sensitivity cases, NHTSA is finalizing stringency increases at 2 percent per year for passenger cars for MYs 2027 through 2031, and at 0 percent per year for light trucks for MYs 2027 and 2028, and 2 percent per year for MYs 2029-2031. NHTSA is also setting forth an augural MY 2032 standard that increases at a rate of 2 percent for both passenger cars and light trucks. NHTSA is finalizing stringency increases at 10 percent per year for HDPUVs for MYs 2030-2032, and 8 percent per year for MYs 2033-2035. The regulatory alternatives representing these final stringency increases are called “PC2LT002” for passenger cars and light trucks, and “HDPUV108” for HDPUVs. These standards are also referred to throughout the rulemaking documents as the “preferred alternative” or “final standards.” NHTSA concludes that these levels are the maximum feasible for these model years as discussed in more detail in Section VI of this preamble, and in particular given the statutory constraints that prevent NHTSA from considering the fuel economy of battery electric vehicles (BEVs) in determining maximum feasible CAFE standards.

NHTSA notes that due to the statutory constraints that prevent NHTSA from considering the fuel economy of dedicated alternative fueled vehicles, the full (including electric-only operation) fuel economy of dual-fueled alternative fueled vehicles, and the availability of over-compliance credits when determining what standards are maximum feasible, many aspects of our analysis are different from what they would otherwise be without the statutory restrictions—in particular, the technologies chosen to model possible compliance options, the estimated costs, benefits, and achieved levels of fuel economy, as well as the current and projected adoption of alternative fueled vehicles. NHTSA evaluates the results of that constrained analysis by weighing the four enumerated statutory factors to determine which standards are maximum feasible, as discussed in Section VI.A.5.

For passenger cars and light trucks, NHTSA notes that the final year of standards, MY 2032, is “augural,” as in the 2012 final rule which established CAFE standards for model years 2017 and beyond. Augural standards mean that they are NHTSA's best estimate of what the agency would propose, based on the information currently before it, if the agency had authority to set CAFE standards for more than five model years in one action. The augural standards do not, and will not, have any effect in themselves and are not binding unless adopted in a subsequent rulemaking. Consistent with past practice, NHTSA is including augural standards for MY 2032 to give its best estimate of what those standards would be to provide as much predictability as possible to manufacturers and to be consistent with the time frame of the Environmental Protection Agency (EPA) standards for greenhouse gas (GHG) emissions from motor vehicles. Due to statutory lead time constraints for HDPUV standards, NHTSA's final rule for HDPUV standards must begin with MY 2030. There is no restriction on the number of model years for which NHTSA may set HDPUV standards, so none of the HDPUV standards are augural.

The CAFE standards remain vehicle-footprint-based, like the current CAFE standards in effect since MY 2011, and the HDPUV standards remain work-factor-based, like the HDPUV standards established in the 2011 “Phase 1” rulemaking used in the 2016 “Phase 2” rulemaking. The footprint of a vehicle is the area calculated by multiplying the wheelbase times the track width, essentially the rectangular area of a vehicle measured from tire to tire where the tires hit the ground. The work factor (WF) of a vehicle is a unit established to measure payload, towing capability, and whether or not a vehicle has four-wheel drive. This means that the standards are defined by mathematical equations that represent linear functions relating vehicle footprint to fuel economy targets for passenger cars and light trucks, and relating WF to fuel consumption targets for HDPUVs.

The target curves for passenger cars, light trucks, and compression-ignition and spark-ignition HDPUVs are set forth in Sections II and IV; curves for model years prior to the years of the rulemaking time frame are included in the figures for context. NHTSA underscores that the equations and coefficients defining the curves are the CAFE and HDPUV standards, and not the mpg and gallon/100-mile estimates that the agency currently estimates could result from manufacturers complying with the curves. We provide mpg and gallon/100-mile estimates for ease of understanding after we illustrate the footprint curves, but the equations and coefficients are the actual standards. NHTSA is also finalizing new minimum domestic passenger car CAFE standards (MDPCS) for model years 2027-2031 as required by the Energy Policy and Conservation Act of 1975 (EPCA), as amended by the EISA, and applied to vehicles defined as manufactured in the United States. Section 32902(b)(4) of 49 U.S.C. requires NHTSA to project the minimum domestic standard when it promulgates passenger car standards for a MY; these standards are shown in Table I-3 below. NHTSA retains the 1.9 percent offset first used in the 2020 final rule, reflecting prior differences between passenger car footprints originally forecast by the agency and passenger car footprints as they occurred in the real world, such that the minimum domestic passenger car standard is as shown in the table below.

Recognizing that many readers think about CAFE standards in terms of the mpg values that the standards are projected to eventually require, NHTSA currently estimates that the standards would require roughly 50.4 mpg in MY 2031, on an average industry fleet-wide basis, for passenger cars and light trucks. NHTSA notes both that real-world fuel economy is generally 20-30 percent lower than the estimated required CAFE level stated above, and also that the actual CAFE standards are the footprint target curves for passenger cars and light trucks. This last note is important, because it means that the ultimate fleet-wide levels will vary depending on the mix of vehicles that industry produces for sale in those model years. NHTSA also calculates and presents “estimated achieved” fuel economy levels, which differ somewhat from the estimated required levels for each fleet, for each year. NHTSA estimates that the industry-wide average fuel economy achieved in MY 2031 for passenger cars and light trucks combined could increase from about 52.1 mpg under the No-Action Alternative to 52.5 mpg under the standards.

To the extent that manufacturers appear to be over-complying in our analysis with required fuel economy levels in the passenger car fleet, NHTSA notes that this is due to the inclusion of several all-electric manufacturers in the reference baseline analysis, which affects the overall average achieved levels. Manufacturers with more traditional fleets do not over-comply at such high levels in our analysis, and our analysis considers the compliance paths for both manufacturer groups. In contrast, while it looks like some manufacturers are falling short of required fuel economy levels in the light truck fleet (and choosing instead to pay civil penalties), NHTSA notes that this appears to be an economic decision by a relatively small number of companies. In response to comments from vehicle manufacturers, in particular manufacturers that commented that they cannot stop manufacturing large fuel inefficient light trucks while also transitioning to manufacturing electric vehicles, NHTSA has reconsidered light truck stringency levels and notes that manufacturers no longer face CAFE civil penalties as modeled in the NPRM. Please see Section VI.D of this preamble for more discussion on these topics and how the agency has considered them in determining maximum feasible standards for this final rule.

For HDPUVs, NHTSA currently projects that the standards would require, on an average industry fleet-wide basis for the HDPUV fleet, roughly 2.851 gallons per 100 miles in MY 2035. HDPUV standards are attribute-based like passenger car and light truck standards, so here, too, ultimate fleet-wide levels will vary depending on what industry produces for sale.

For all fleets, average requirements and average achieved CAFE and HDPUV fuel efficiency levels would ultimately depend on manufacturers' and consumers' responses to standards, technology developments, economic conditions, fuel prices, and other factors.

Our technical analysis for this final rule keeps the same general framework as past CAFE and HDPUV rules, but as applied to the most up-to-date fleet available at the time of the analysis. NHTSA has updated technologies considered in our analysis (removing technologies which are already universal or nearly so and technologies which are exiting the fleet, adding certain advanced engine technologies); updated macroeconomic input assumptions, as with each round of rulemaking analysis; improved user control of various input parameters; updated our approach to modeling manufacturers' expected compliance with states' Zero Emission Vehicle (ZEV) programs and deployment of additional electric vehicles consistent with manufacturer commitments; accounted for changes to DOE's Petroleum Equivalency Factor (PEF), for the reference baseline assumptions; expanded accounting for Federal incentives such as Inflation Reduction Act programs; expanded procedures for estimating new vehicle sales and fleet shares; updated inputs for projecting aggregate light-duty Vehicle Miles Traveled (VMT); and added various output values and options.

NHTSA concludes, as we explain in more detail below, that Alternative PC2LT002 is the maximum feasible alternative that manufacturers can achieve for model years 2027-2031 passenger cars and light trucks, based on a variety of reasons. Energy conservation is still paramount, for the consumer benefits, energy security benefits, and environmental benefits that it provides. Moreover, although the vehicle fleet is undergoing a significant transformation now and in the coming years, for reasons other than the CAFE standards, NHTSA believes that a significant percentage of the on-road (and new) vehicle fleet may remain propelled by internal combustion engines (ICEs) through 2031. NHTSA believes that the final standards will encourage manufacturers producing those ICE vehicles during the standard-setting time frame to achieve significant fuel economy, improve energy security, and reduce harmful pollution by a large amount. At the same time, NHTSA is finalizing standards that our estimates project will continue to save consumers money and fuel over the lifetime of their vehicles while being economically practicable and technologically feasible for manufacturers to achieve.

Although all of the other alternatives, except for the no-action alternative, would conserve more energy and provide greater fuel savings benefits and certain pollutant emissions reductions, NHTSA's statutorily-constrained analysis currently estimates that those alternatives may not be achievable for many manufacturers in the rulemaking time frame. Additionally, the analysis indicates compliance with those more stringent alternatives would impose significant costs (under the constrained analysis) on individual consumers without corresponding fuel savings benefits large enough to, on average, offset those costs. Within that framework, NHTSA's analysis suggests that the more stringent alternatives could push more technology application than would be economically practicable, given anticipated reference baseline activity that will already be consuming manufacturer resources and capital and the constraints of planned manufacturer redesign cycles. In contrast to all other action alternatives, except for the no-action alternative, Alternative PC2LT002 comes at a cost we believe the market can bear without creating consumer acceptance or sales issues, appears to be much more achievable, and will still result in consumer net benefits on average. The alternative also achieves large fuel savings benefits and significant reductions in emissions compared to the no-action alternative. NHTSA concludes Alternative PC2LT002 is the appropriate choice given this record.

For HDPUVs, NHTSA concludes, as explained in more detail below, that Alternative HDPUV108 is the maximum feasible alternative that manufacturers can achieve for model years 2030-2035 HDPUVs. It has been seven years since NHTSA revisited HDPUV standards, and our analysis suggests that there is much opportunity for cost-effective improvements in this segment, broadly speaking. At the same time, we recognize that these vehicles are primarily used to conduct work for a large number of businesses. Although Alternatives HDPUV10 and HDPUV14 would conserve more energy and provide greater fuel savings benefits and CO₂ emissions reductions, they are more costly than HDPUV108, and NHTSA currently estimates that Alternative HDPUV108 is the most cost-effective under a variety of metrics and at either a 3 percent or a 7 percent DR, while still being appropriate and technologically feasible. NHTSA is allowed to consider electrification in determining maximum feasible standards for HDPUVs. As a result, NHTSA concludes that HDPUV108 is the appropriate choice given the record discussed in more detail below, and we believe it balances EPCA's overarching objective of energy conservation while remaining cost-effective and technologically feasible.

For passenger cars and light trucks, NHTSA estimates that this final rule would reduce average fuel outlays over the lifetimes of MY 2031 vehicles by about $639 per vehicle relative to the reference baseline, while increasing the average cost of those vehicles by about $392 over the reference baseline, at a 3 percent discount rate; this represents a difference of $247. With climate benefits discounted at 2 percent and all other benefits and costs discounted at 3 percent, when considering the entire CAFE fleet for model years 1983-2031, NHTSA estimates $24.5 billion in monetized costs and $59.7 billion in monetized benefits attributable to the standards, such that the present value of aggregate net monetized benefits to society would be $35.2 billion. Again, the net benefits are larger if the final rule is assessed relative to the alternative baseline.

For HDPUVs, NHTSA estimates that this final rule could reduce average fuel outlays over the lifetimes of MY 2038 vehicles by about $717 per vehicle, while increasing the average cost of those vehicles by about $226 over the reference baseline, at a 3 percent discount rate; this represents a difference of $491. With climate benefits discounted at 2 percent and all other benefits and costs discounted at 3 percent, when considering the entire on-road HDPUV fleet for calendar years 2022-2050, NHTSA estimates $3.4 billion in monetized costs and $17 billion in monetized benefits attributable to the standards, such that the present value of aggregate net monetized benefits to society would be $13.6 billion.

These assessments do not include important unquantified effects, such as energy security benefits, equity and distributional effects, and certain air quality benefits from the reduction of toxic air pollutants and other emissions, among other things, so the net benefit estimate is a conservative one. In addition, the power sector emissions modeling reflected in this analysis is subject to uncertainty and may be conservative to the extent that other components that influence energy markets, such as recently finalized Federal rules and additional modeled policies like Federal tax credits, are incorporated in those estimates. That said, NHTSA performed additional modeling to test the sensitivity of those estimates and found that in the context of total emissions, any changes from using different power sector forecasts are extremely small. This is discussed in more detail in FRIA Chapter 9.

Table I-6 presents aggregate benefits and costs for new vehicle buyers and for the average individual new vehicle buyer.

NHTSA recognizes that EPA has recently issued a final rule to set new multi-pollutant emissions standards for model years 2027 and later light-duty (LD) and medium-duty vehicles (MDV). EPA describes its final rule as building upon EPA's final standards for Federal GHG emissions standards for passenger cars and light trucks for model years 2023 through 2026 and leverages advances in clean car technology to unlock benefits to Americans ranging from reducing pollution, to improving public health, to saving drivers money through reduced fuel and maintenance costs. EPA's standards phase in over model years 2027 through 2032.

NHTSA coordinated with EPA in developing our final rule to avoid inconsistencies and produce requirements that are consistent with NHTSA's statutory authority. The final rules nevertheless differ in important ways. First, NHTSA's final rule, consistent with its statutory authority and mandate under EPCA/EISA, focuses on improving vehicle fuel economy and not directly on reducing vehicle emissions—though reduced emissions are a follow-on effect of improved fuel economy. Second, the biggest difference between the two final rules is due to EPCA/EISA's statutory prohibition against NHTSA considering the fuel economy of dedicated alternative fueled vehicles, including BEVs, and including the full fuel economy of dual-fueled alternative fueled vehicles in determining the maximum feasible fuel economy level that manufacturers can achieve for passenger cars and light trucks, even though manufacturers may use BEVs and dual-fueled alternative fuel vehicles (AFV) like PHEVs to comply with CAFE standards. EPA is not prohibited from considering BEVs or PHEVs as a compliance option. EPA's final rule is informed by, among other considerations, trends in the automotive industry (including the proliferation of announced investments by automakers in electrifying their fleets), tax incentives under the Inflation Reduction Act (IRA), and other factors in the rulemaking record that are leading to a rapid transition in the automotive industry toward less-pollutant-emitting vehicle technologies. NHTSA, in contrast, may not consider BEVs as a compliance option for the passenger car and light truck fleets even though manufacturers may, in fact, use BEVs to comply with CAFE standards. This constraint means that not only are NHTSA's stringency rates of increase different from EPA's but also the shapes of our standards are different based upon the different scopes.

Recognizing these statutory restrictions and their effects on NHTSA's analysis (and that EPA's analysis and decisions are not subject to such constraints) NHTSA sought to optimize the effectiveness of the final CAFE standards consistent with our statutory factors. Our statutorily constrained simulated industry response shows a reasonable path forward to compliance with CAFE standards, but we want to stress that our analysis simply shows feasibility and does not dictate a required path to compliance. Because the standards are performance-based, manufacturers are always free to apply their expertise to find the appropriate technology path that best meets all desired outcomes. Indeed, as explained in greater detail later on in this final rule, it is entirely possible and reasonable that a vehicle manufacturer will use technology options to meet NHTSA's standards that are significantly different from what NHTSA's analysis for this final rule suggests given the statutory constraints under which it operates. NHTSA has ensured that these final standards take account of statutory objectives and constraints while minimizing compliance costs.

As discussed before, NHTSA does not face the same statutory limitations in setting standards for HDPUVs as it does in setting standards for passenger cars and light trucks. This allows NHTSA to consider a broader array of technologies in setting maximum feasible standards for HDPUVs. However, we are still considerate of factors that allow these vehicles to maintain utility and do work for the consumer when we set the standards.

Additionally, NHTSA has considered and accounted for the electric vehicles that manufacturers' have indicated they intend to deploy in our analysis, as part of the analytical reference baseline. Some of this deployment would be consistent with manufacturer compliance with California's Advanced Clean Cars (ACC) I and Advanced Clean Trucks (ACT). We find that manufacturers will comply with ZEV requirements in California and a number of other states in the absence of CAFE standards, and accounting for that expected compliance allows us to present a more realistic picture of the state of fuel economy even in the absence of changes to the CAFE standards. In the proposal, we also included the main provisions of California's Advanced Clean Cars II program (ACC II), which California has adopted but which has not been granted a Clean Air Act preemption waiver by EPA. Because ACC II has not been granted a waiver, we have not included it in our analysis as a legal requirement applying to manufacturers. However, manufacturers have indicated that they intend to deploy additional electric vehicles regardless of whether the waiver is granted, and our analysis reflects these vehicles. Reflecting this expected deployment of electric vehicles for non-CAFE compliance reasons in the analysis improves the accuracy of this reference baseline in reflecting the state of the world without the revised CAFE standards, and thus the information available to decision-makers in their decision as to what standards are maximum feasible, and to the public. However, in order to ensure that the analysis is robust to other possible futures, NHTSA also prepared an alternative baseline—one that reflected none of these electric vehicles (No ZEV Alternative Baseline). The net benefits of the standards are larger under this alternative baseline than they are under the reference baseline, and the technology deployment scenario is reasonable under the alternative baseline, further reinforcing NHTSA's conclusion that the final standards are reasonable, appropriate, and maximum feasible regardless of the deployment of electric vehicles that occurs independent of the standards.

NHTSA notes that while the current estimates of costs and benefits are important considerations and are directed by E.O. 12866, cost-benefit analysis provides only one informative data point in addition to the host of considerations that NHTSA must balance by statute when determining maximum feasible standards. Specifically, for passenger cars and light trucks, NHTSA is required to consider four statutory factors—technological feasibility, economic practicability, the effect of other motor vehicle standards of the Government on fuel economy, and the need of the United States to conserve energy. For HDPUVs, NHTSA is required to consider three statutory factors—whether standards are appropriate, cost-effective, and technologically reasonable—to determine whether the standards it adopts are maximum feasible. As will be discussed further below, NHTSA concludes that Alternatives PC2LT002 and HDPUV108 are maximum feasible on the basis of these respective factors, and the cost-benefit analysis, while informative, is not one of the statutorily-required factors. NHTSA also considered several dozen sensitivity cases varying different inputs and concluded that even when varying inputs resulted in changes to net benefits or (on rare occasions) changed the relative order of regulatory alternatives in terms of their net benefits, those changes were not significant enough to outweigh our conclusion that Alternatives PC2LT002 and HDPUV108 are maximum feasible.

NHTSA further notes that CAFE and HDPUV standards apply only to new vehicles, meaning that the costs attributable to new standards are “front-loaded” because they result primarily from the application of fuel-saving technology to new vehicles. By contrast, the impact of new CAFE and HDPUV standards on fuel consumption and energy savings, air pollution, and GHGs—and the associated benefits to society—occur over an extended time, as drivers buy, use, and eventually scrap these new vehicles. By accounting for many model years and extending well into the future to 2050, our analysis accounts for these differing patterns in impacts, benefits, and costs. Given the front-loaded costs versus longer-term benefits, it is likely that an analysis extending even further into the future would find additional net present benefits.

The bulk of our analysis for passenger cars and light trucks presents a “model year” (MY) perspective rather than a “calendar year” (CY) perspective. The MY perspective considers the lifetime impacts attributable to all passenger cars and light trucks produced prior to MY 2032, accounting for the operation of these vehicles over their entire lives (with some MY 2031 vehicles estimated to be in service as late as 2050). This approach emphasizes the role of the model years for which new standards are being finalized, while accounting for the potential that the standards could induce some changes in the operation of vehicles produced prior to MY 2027 (for passenger cars and light trucks), and that, for example, some individuals might choose to keep older vehicles in operation, rather than purchase new ones.

The calendar year perspective we present includes the annual impacts attributable to all vehicles estimated to be in service in each calendar year for which our analysis includes a representation of the entire registered passenger car, light truck, and HDPUV fleet. For this final rule, this calendar year perspective covers each of calendar years 2022-2050, with differential impacts accruing as early as MY 2022. Compared to the MY perspective, the calendar year perspective includes model years of vehicles produced in the longer term, beyond those model years for which standards are being finalized.

The tables below summarize estimates of selected impacts viewed from each of these two perspectives, for each of the regulatory alternatives considered in this final rule, relative to the reference baseline.

Our net benefit estimates are likely to be conservative both because (as discussed above) our analysis only extends to MY 2031 and calendar year 2050 (LD) and calendar year 2050 (HDPUV), and because there are additional important health, environmental, and energy security benefits that could not be fully quantified or monetized. Finally, for purposes of comparing the benefits and costs of CAFE and HDPUV standards to the benefits and costs of other Federal regulations, policies, and programs under the Regulatory Right-to-Know Act, we have computed “annualized” benefits and costs relative to the reference baseline, as follows:

It is also worth emphasizing that, although NHTSA is prohibited from considering the availability of certain flexibilities in making our determination about the levels of CAFE standards that would be maximum feasible, manufacturers have a variety of flexibilities available to aid their compliance. Section VII of this preamble summarizes these flexibilities and what NHTSA has finalized for this final rule. NHTSA is finalizing changes to these flexibilities as shown in Table I-13 and Table I-14.

The following sections of this preamble discuss the technical foundation for the agency's analysis, the regulatory alternatives considered in this final rule, the estimated effects of the regulatory alternatives, the basis for NHTSA's conclusion that the standards are maximum feasible, and NHTSA's approach to compliance and enforcement. The extensive record supporting NHTSA's conclusion is documented in this preamble, in the TSD, the FRIA, the Final EIS, and the additional materials on NHTSA's website and in the rulemaking docket.

II. Overview of the Final Rule

A. Summary of the NPRM

In the NPRM, NHTSA proposed new fuel economy standards for LDVs for model years 2027-2031 and new fuel efficiency standards for HDPUVs for model years 2030-2035. NHTSA also set forth proposed augural standards for LDVs for model year 2032. NHTSA explained that it was proposing the standards in response to the agency's statutory mandate to improve energy conservation and reduce the nation's energy dependence on foreign sources. NHTSA also explained that the proposal was also consistent with Executive Order (E.O.) 14037, “Strengthening American Leadership in Clean Cars and Trucks,” (August 5, 2021), which directed the Secretary of Transportation (by delegation, NHTSA) to consider beginning work on rulemakings under the Energy Independence and Security Act of 2007 (EISA) to establish new fuel economy standards for LDVs beginning with model year 2027 and extending through at least model year 2030, and to establish new fuel efficiency standards for HDPUVs beginning with model year 2028 and extending through at least model year 2030, consistent with applicable law.

NHTSA discussed the fact that EPA issued a proposal to set new multi-pollutant emissions standards for model years 2027 and later for light-duty and medium-duty vehicles. NHTSA explained that we coordinated with EPA in developing our proposal to avoid inconsistencies and produce requirements that are consistent with NHTSA's statutory authority. The proposals nevertheless differed in important ways, described in detail in the NPRM. EPA has since issued a final rule associated with its proposal, and the interaction between EPA's final standards and NHTSA's final standards is discussed in more detail below.

NHTSA also explained that it had considered and accounted for manufacturers' expected compliance with California's Advanced Clean Cars (ACC I) program and Advanced Clean Trucks (ACT) regulations in our analysis, as part of the analytical reference baseline. We stated that manufacturers will comply with current ZEV requirements in California and a number of other states in the absence of CAFE standards, and accounting for that expected compliance allows us to present a more realistic picture of the state of fuel economy even in the absence of changes to the CAFE standards. NHTSA also incorporated deployment of electric vehicles that would be consistent with California's ACC II program, which has not received a preemption waiver from EPA. However, automakers have indicated their intent to deploy electric vehicles consistent with the levels that would be required under ACCII if a waiver were to be granted, and as such its inclusion similarly makes the reference baseline more accurate. Reflecting expected compliance with the current ZEV programs and manufacturer deployment of EVs consistent with levels that would be required under the ACC II program in the analysis helps to improve the accuracy of the reference baseline in reflecting the state of the world without the revised CAFE standards, and thus the information available to policymakers in their decision as to what standards are maximum feasible and to the public in commenting on those standards. NHTSA also described several other improvements and updates it made to the analysis since the 2022 final rule based on NHTSA analysis, new data, and stakeholder meetings for the NPRM.

NHTSA proposed fuel economy standards for model years 2027-2032 (model year 2032 being proposed augural standards) that increased at a rate of 2 percent per year for both passenger cars and 4 percent per year for light trucks, and fuel efficiency standards for model years 2030-2035 that increased at a rate of 10 percent per year for HDPUVs. NHTSA also took comment on a wide range of alternatives, including no-action alternatives for both light duty vehicles and HDPUVs (retaining the 2022 passenger car and light truck standards and the 2016 final rule for HDPUV standards) and updates to the compliance flexibilities. The proposal was accompanied by a Preliminary Regulatory Impact Analysis (PRIA), a Draft Environmental Impact Statement (Draft EIS), Technical Support Document (TSD) and the CAFE Model software source code and documentation, all of which were also subject to comment in their entirety and all of which received significant comments.

NHTSA tentatively concluded that Alternative PC2LT4 was maximum feasible for LDVs for model years 2027-2031 and Alternative HDPUV10 was maximum feasible for HDPUVs for model years 2030-2035. NHTSA explained that average requirements and achieved CAFE levels would ultimately depend on manufacturers' and consumers' responses to standards, technology developments, economic conditions, fuel prices, and other factors. NHTSA estimated that the proposal would reduce gasoline consumption by 88 billion gallons relative to reference baseline levels for LDVs, and by approximately 2.6 billion gallons relative to reference baseline levels for HDPUVs through calendar year 2050. NHTSA also estimated that the proposal would reduce carbon dioxide (CO₂) emissions by 885 million metric tons for LDVs, and by 22 million metric tons for HDPUVs through calendar year 2050.

In terms of economic effects, NHTSA estimated that while consumers would pay more for new vehicles upfront, they would save money on fuel costs over the lifetimes of those new vehicles—lifetime fuel savings exceed modeled regulatory costs by roughly $100, on average, for model year 2032 LDVs, and by roughly $300, on average, for buyers of model year 2038 HDPUVs. NHTSA estimated that net benefits for the preferred alternative for LDVs would be $16.8 billion at a 3 percent discount rate, and $8.4 billion at a 7 percent discount rate, and for the preferred alternative for HDPUVs would be $2.2 billion at a 3 percent discount rate, and $1.4 billion at a 7 percent discount rate.

NHTSA also addressed the question of harmonization with other motor vehicle standards of the Government that affect fuel economy. Even though NHTSA and EPA issued separate rather than joint notices, NHTSA explained that it had worked closely with EPA in developing the respective proposals, and that the agencies had sought to minimize inconsistency between the programs where doing so was consistent with the agencies' respective statutory mandates. NHTSA emphasized that differences between the proposals, especially as regards programmatic flexibilities, were not new in the proposal, and that differences were often a result of the different statutory frameworks. NHTSA reminded readers that since the agencies had begun regulating concurrently in 2010, these differences have meant that manufacturers have had (and will have) to plan their compliance strategies considering both the CAFE standards and the GHG standards and assure that they are in compliance with both. NHTSA was also confident that industry would still be able to build a single fleet of vehicles to meet both the NHTSA and EPA standards. NHTSA sought comment broadly on all aspects of the proposal.

B. Public Participation Opportunities and Summary of Comments

The NPRM was published on NHTSA's website on July 28, 2023, and published in the Federal Register on August 17, 2023, beginning a 60-day comment period. The agency left the docket open for considering late comments to the extent practicable. A separate Federal Register notice, published on August 25, 2023, announced a virtual public hearing taking place on September 28 and 29, 2023. Approximately 155 individuals and organizations signed up to participate in the hearing. The hearing started at 9:30 a.m. EDT on September 28th and ended at approximately 5:00 p.m., completing the entire list of participants within a single day, resulting in a 141-page transcript. The hearing also collected many pages of comments from participants, in addition to the hearing transcript, all of which were submitted to the docket for the rule.

Including the 2,269 comments submitted as part of the public hearings, NHTSA's docket received a total of 63,098 comments, with tens of thousands of comments submitted by individuals and over 100 deeply substantive comments that included many attachments submitted by stakeholder organizations. NHTSA also received five comments on its Draft EIS to the separate EIS docket NHTSA-2022-0075, in addition to 17 comments on the EIS scoping notice that informed NHTSA's preparation of the Draft EIS.

Many commenters supported the proposal. Commenters supporting the proposal emphasized the importance of increased fuel economy for consumers, as well as cited concerns about climate change, which are relevant to the need of the United States to conserve energy. Commenters also expressed the need for harmonization and close coordination between NHTSA, EPA, and DOE for their respective programs. Many citizens, environmental groups, some States and localities, and some vehicle manufacturers stated strong support for NHTSA finalizing the most stringent alternative.

Many manufacturers urged NHTSA to consider the impact of EPA's standards as well as the impact of DOE's Petroleum Equivalency Factor (PEF) rule on fleet compliance (discussed in more detail below). Many manufacturers supported alignment with EPA's and DOE's standards. Manufacturers were also supportive of keeping the footprint-based standards for LD vehicles and work factor-based standards for HDPUVs. Manufacturers and others were also supportive of continuing the HD Phase 2 approach for HDPUVs by having separate standards for compression ignition (CI) and spark ignition (SI) vehicles, as well as continuing to use a zero fuel consumption value for alternative fuel vehicles such as battery electric vehicles.

In other areas, commenters expressed mixed views on the compliance and flexibilities proposed in the notice. Manufacturers were supportive of maintaining the Minimum Domestic Passenger Car Standard (MDPCS) offset relative to the standards. Most manufacturers and suppliers did not support phasing out off-cycle and AC efficiency fuel consumption improvement values (FCIVs), whereas NGOs and electric vehicle manufacturers supported removing all flexibilities. Many fuel and alternative fuel associations opposed the regulation due to lack of consideration for other types of fuels in NHTSA's analysis.

NHTSA also received several comments on subjects adjacent to the rule but beyond the agency's authority to influence. NHTSA has reviewed all comments and accounted for them where legally possible in the modeling and qualitatively, as discussed below and throughout the rest of the preamble and in the TSD.

NHTSA received a range of comments about the interaction between DOE's Petroleum Equivalency Factor (PEF) proposal and NHTSA's CAFE proposal, mainly from vehicle manufacturers. Several stakeholders commented in support of the proposed PEF, while others commented that the PEF should remain at the pre-proposal level, or even increase. The American Automotive Policy Council (AAPC), the policy organization that represents the “Detroit Three” or D3—Ford, General Motors, and Stellantis—commented that DOE's proposed PEF reduction inappropriately devalues electrification, and accordingly “a devalued PEF yields a dramatic deficiency in light-duty trucks, that make up 83% of the D3's product portfolio.” The AAPC also commented that “NHTSA's inclusion of the existing PEF for EVs in 2026 creates an artificially high CAFE compliance baseline, and the proposed PEF post-2027 removes the only high-leverage compliance tool available to auto manufacturers.” Relatedly, as part of their comments generally opposing DOE's proposed PEF level, other automakers provided alternative values for the PEF, or supported a phase-in of the PEF to better allow manufacturers to restructure their product mix. Other stakeholders urged NHTSA to delay the CAFE rule until DOE adopts a revised PEF, or stated that NHTSA should reopen comments on its proposal following final DOE action on the PEF. Finally, some commenters recommended that NHTSA apply a PEF to the HDPUV segment.

Regarding comments that were supportive of or opposing the new PEF, those comments are beyond the scope of this rulemaking. By statute, DOE is required to determine the PEF value and EPA is required to use DOE's value for calculation of a vehicle's CAFE value. NHTSA has no control over the selection of the PEF value or fuel economy calculation procedures; accordingly, the PEF value is just one input among many inputs used in NHTSA's analysis. While NHTSA was in close coordination with DOE during the pendency of the PEF update process, stakeholder comments about the PEF value and whether the value should be phased in were addressed in DOE's final rule.

As NHTSA does not take a position on the PEF value, the agency believes it was appropriate to use the most up-to-date input assumption at each stage of the analysis to provide stakeholders the best information about the effects of different levels of CAFE standards. NHTSA also included sensitivity analyses in the NPRM with DOE's pre-proposal PEF value so that all stakeholders had notice of and the opportunity to comment on a scenario where the PEF did not change. NHTSA accordingly disagrees that the agency needed to reopen comments on the proposal following final DOE action on the PEF.

NHTSA agrees with AAPC that when a manufacturer's portfolio consists predominantly of lower fuel economy light trucks, as in the particular case of the D3, averaging the fuel economy of those vehicles with high fuel economy BEVs would help them comply with fuel economy standards more so than if BEVs had a lower fuel economy due to a lower PEF. However, this concern is somewhat ameliorated by the changes in DOE's final PEF rule, including a gradual reduction of the fuel content factor. Furthermore NHTSA has determined that the final standards are the maximum feasible fuel economy level that manufacturers can achieve even without producing additional electric vehicles. And, NHTSA disagrees that including in the modeling the old PEF in 2026 and prior and the new PEF in 2027 and beyond “removes the only high-leverage compliance tool available to auto manufacturers” (emphasis added), as there are several compliance tools available to manufacturers, including increasing the fuel economy of their ICE vehicles. As discussed further in Section VI, NHTSA believes that the standards finalized in this rule explicitly contemplate the concerns expressed by and the capability of all manufacturers.

NHTSA will not use a PEF for HDPUV compliance at this time. NHTSA will continue to use the framework that was put in place by the HD Phase 2 rule, and in coordination with EPA's final rule, by using zero upstream energy consumption for compliance calculations (note that NHTSA does consider upstream effects of electricity use in its effects modeling). Any potential future action on developing PEF for HDPUV compliance would most likely occur in a standalone future rulemaking after NHTSA has a more thorough opportunity to consider the costs and benefits of such an approach and all stakeholders can present feedback on the issue.

NHTSA also received a range of comments about BEV infrastructure. Comments covered both the amount and quality of BEV charging infrastructure and the state of electric grid infrastructure. Some stakeholders, including groups representing charging station providers and electricity providers, commented that although additional investments will be required to support future demand for public chargers and the electricity required for BEV charging, their preparation and planning for the BEV transition is already underway. Many stakeholders emphasized the role of a robust public charging network to facilitate the BEV transition, and broadly urged the Administration to work amongst the agencies and with automakers, utilities, and other interested parties to ensure that BEV charging infrastructure buildout, including developing minimum standards for public charging efficiency, and BEV deployment happen hand in hand.

In contrast, some stakeholders emphasized the current lack of public BEV charging infrastructure as a barrier to EV adoption. Stakeholders also highlighted mechanical problems with existing charging stations, which they stated contributes to dissatisfaction with public charging stations among electric vehicle owners. Other stakeholders commented that the country's electricity transmission infrastructure is not currently in a position to support the expected electricity demand from the BEV transition and may not be in the future for several reasons, such as the lack of materials needed to expand and upgrade the grid. To combat those concerns, other stakeholders recommended that administration officials and congressional leaders prioritize policies that would strengthen transmission systems and infrastructure and speed up their growth. Stakeholders also recommended that NHTSA capture some elements of charging and grid infrastructure issues in its analysis, and outside of the analysis and this rulemaking, identify ways to assist in the realization of adequate BEV infrastructure.

NHTSA acknowledges and appreciates all the comments received on charging infrastructure, which include both broad comments on future grid infrastructure needs, as well as increased deployment of reliable and convenient charging stations. NHTSA agrees with commenters in that infrastructure is an important aspect of a successful transition to BEVs in the future. We also agree that infrastructure improvements are necessary and directly related to keeping pace with projected levels of BEV supply and demand as projected by other agencies and independent forecasters.

With that said, NHTSA projects that manufacturers will deploy a wide variety of technologies to meet the final CAFE standards that specifically are not BEVs, considering NHTSA's statutory limitations. As discussed further throughout this preamble, NHTSA does not consider adoption of BEVs in the LD fleet beyond what is already in the reference baseline. Results in Chapter 8 of the FRIA show increased technology penetrations of more efficient conventional ICEs, increased penetration of advanced transmissions, increased mass reduction technologies, and other types of electrification such as mild and strong hybrids.

In addition, as discussed further below, NHTSA has coordinated with DOE and EPA while developing this final rule, as requested by commenters. Experts at NHTSA's partner agencies have found that the grid and associated charging infrastructure could handle the increase in BEVs related to both EPA's light- and medium-duty vehicle multi-pollutant rule and the HD Phase 3 GHG rule —significantly more BEVs than NHTSA projects in the LD and HDPUV reference baselines examined in this rule. Thus, infrastructure beyond what is planned for buildout in the rulemaking timeframe, accounting not only for electricity generation and distribution, but considering load-balancing management measures, as well, to improve grid operations, would not be required. It should also be noted that expert projections show an order of magnitude increase in available (domestic) public charging ports between the release of the final rule and the rulemaking timeframe, not accounting for the additional availability of numerous residential and depot chargers. Battery energy storage integration with DC fast chargers can further expedite deployment of necessary infrastructure, reducing lead time for distribution upgrades while increasing the likelihood of meeting public charging needs in the next decade. The National Electric Vehicle Infrastructure (NEVI) program is also investing $5 billion in federal funding to deploy a national network of public EV chargers. Additionally, federally funded charging stations are required to adhere to a set of nationally recognized standards, such as a minimum of 97% annual-uptime, which is anticipated to greatly improve charging reliability concerns of today.

For the HDPUV analysis, NHTSA does consider adoption of BEVs in the standard setting years, and we do see an uptake of BEVs; however, the population of the HDPUV fleet is extremely small, consisting of fewer than 1 million vehicles, compared to the LD fleet that consists of over 14 million vehicles. This means that any potential impact of HDPUV BEV adoption on the electric grid would be similarly small. We also want to note that the adoption of these HDPUV BEVs is driven primarily by factors other than NHTSA's standards, including the market demand for increased fuel efficiency and state ZEV programs, as shown in detail in Section V of this preamble and FRIA Chapter 8.3.2. However, as with LD standards examined in this rule, most manufacturers could choose to meet the preferred standards with limited BEVs. There are still opportunities in the advanced engines, advanced transmissions, and strong hybrid technologies that could be used to meet the HDPUV preferred standards starting in model year 2030.

Although NHTSA does not consider BEVs in its analysis of CAFE stringency, and there is minimal BEV adoption driven by the HDPUV FE standards, NHTSA coordinated with both DOE and EPA on many of the challenges raised by commenters to understand how the infrastructure will be developing and improving in the future. Our review of efforts taking place under the NEVI Program and consultation with DOE and EPA leads us to conclude that (1) there will be sufficient EV infrastructure to support the vehicles included in the light-duty reference baseline and in the HDPUV analysis; and (2) it is reasonable to anticipate that the power sector can continue to manage and improve the electricity distribution system to support the increase in BEVs. DOE and EPA conducted analyses that evaluate potential grid impacts of LD and HD fleet that contain significantly more BEVs than NHTSA's light-duty reference baseline and HDPUV fleets. Their analyses conclude that the implementation of EPA's LD and HD rules can be achieved. DOE and EPA found that sufficient electric grid charging and infrastructure can be deployed, numerous federal programs are providing funding to upgraded charging and grid infrastructure, and managed charging and innovative charging solutions can reduce needed grid updates. The analyses conducted for this assessment of the power sector section covered multiple inputs and assumptions across EPA and DOE tools, such as PEV adoption and EVSE access and utilization, to make sure that all aspects of the grid scenarios modeled are analyzed through 2050 between the no action and action alternative in EPA's rule.

NHTSA also received several comments regarding critical materials used to make EV batteries. In support of its comments that the EV supply chain is committed to supporting full electrification, ZETA provided a thorough recitation of policy drivers supporting critical minerals development, projected demand for critical minerals, and ongoing investments and support from its members for critical mineral production, refining, and processing. Similarly, stakeholders commented about different federal and industry programs, incentives, and investments to promote the production and adoption of electric vehicles. Similar to comments on EV infrastructure, many stakeholders commented that federal agencies should work together to ensure a reliable supply chain for critical minerals.

Other stakeholders commented about several critical minerals issues they perceived to be barriers to a largescale transition to EVs. Stakeholders commented generally on a limited or unavailable supply of certain critical minerals, and more specifically the lack of mineral extraction and production in the United States, stating that domestic production of critical minerals is insufficient to meet projected demands. Stakeholders also commented on the potential environmental impact of mining critical minerals, particularly as vehicle manufacturers produce EVs with increasing battery pack sizes. Other stakeholders commented that all of these factors (including costs and environmental impact) should be considered in NHTSA's analysis. Finally, several stakeholders commented on how critical minerals' energy security issues interact with NHTSA's balancing factors to set maximum feasible standards and those comments are addressed in Section VI.5; other stakeholders commented on how critical minerals sourcing interacts with NHTSA's assumptions about tax credits and those comments are addressed in Section III.C.

We appreciate the commenters' feedback in this area and believe that the comments are important to note. However, as we have discussed earlier in this section, the CAFE standards final rulemaking analysis does not include adoption of BEVs beyond what is represented in the reference baseline. We do allow adoption of BEVs in the HDPUV fleet, as EPCA/EISA does not limit consideration of HDPUV technologies in the same way as LD technologies; however, as discussed above, BEV adoption is driven primarily by reasons other than NHTSA's fuel efficiency standards and the number of vehicles that adopt BEV technology in our analysis is relatively (compared to the LD fleet) small. That said, NHTSA believes that commenters' concerns are either currently addressed or are being actively addressed by several public and private endeavors.

NHTSA, in coordination with DOE and EPA, reviewed current supply chain and updated analyses on critical materials. In particular, the DOE, through Argonne National Laboratory, conducted an updated assessment of developing and securing mineral supply for the U.S. electric vehicle industry, the Securing Critical Minerals report. The Argonne study focuses on five materials identified in a previous assessment, including lithium, nickel, cobalt, graphite, and manganese. The study collects and examines potential domestic sources of materials, as well as sources outside the U.S. including Free Trade Agreement (FTA) partners, members of the Mineral Security Partnership (MSP), economic allies without FTAs (referred to as “Non-FTA countries” in the Argonne study), and Foreign Entity of Concern (FEOC) sources associated with covered nations, to support domestic critical material demand from anticipated electric vehicle penetration. The assessment considers geological resources and current international development activities that contribute to the understanding of mineral supply security as jurisdictions around the world seek to reduce emissions. The study also highlights current activities that are intended to expand a secure supply chain for critical minerals both domestically and among U.S. allies and partner nations; and considers the potential to meet U.S. demand with domestic and other secure sources. The DOE Securing Critical Minerals report concluded that the U.S. is “well-positioned to meet its lithium demand through domestic production.” In the near- and medium-term there is sufficient capacity in FTA and MSP countries to meet demand for nickel and cobalt; however, the U.S. will likely need to rely at least partly on non-FTA counties given expected competition for these minerals from other countries' decarbonization goals. In the near-term, meeting U.S. demand with natural graphite supply from domestic FTA and MSP sources is unlikely. In the medium-term, there is potential for new capacity in both FTA and non-FTA countries, and for synthetic graphite production to scale. The U.S. can rely on FTA and MSP partners, as well as other economic and defense partners, to fill supply gaps; countries with which the U.S. has good trade relations are anticipated to have the ability to assist the U.S. in securing the minerals needed to meet EV and ESS (energy storage system) deployment targets set by the Biden Administration. NHTSA considers Argonne's assessment to be thorough and up to date. In addition, it should be noted that DOE's assessments consider critical minerals and battery components to support more than ten million EVs by 2035 —significantly more than we project in our reference baseline.

NHTSA also received a wide variety of comments on alternative fuels including ethanol and biofuels. A group of commenters representing ethanol and biofuel producers objected to NHTSA's handling of BEVs in the analysis, in part because of their views on NHTSA's ability to consider those vehicles under 49 U.S.C. 32902(h), raised energy security concerns with reduced demand for and reliance on U.S.-produced alternative fuels as a result of these regulations, and commented that BEVs would increase reliance on foreign supply chains. Other commenters shared similar sentiments regarding alternative fuels. These commenters stated that NHTSA failed to consider other fuels like ethanol and biofuels as a way to improve fuel economy in the analysis as part of a holistic approach to reducing the U.S.'s gasoline consumption, and therefore the proposed rule was arbitrary. Commenters also stated that NHTSA did not consider the Renewable Fuel Standard (RFS) regulation in this rulemaking, and argued that NHTSA's failure to do so was arbitrary. Finally, commenters recommended that NHTSA consider high octane renewable fuels as a way to improve fuel economy for conventional ICEs.

NHTSA believes that fuel producers' comments about NHTSA's purported inability to consider BEVs under 49 U.S.C. 32902(h) are somewhat misguided, considering that EPCA's definition of “alternative fuel” in 49 U.S.C. 32901 also includes ethanol, other alcohols, and fuels derived from biological materials, among other fuels. This means that if NHTSA were to adopt the fuel producers' interpretation of 49 U.S.C. 32902(h) to restrict BEV adoption in the reference baseline, NHTSA would have to take an analogous approach to limit the agency's consideration of vehicles fueled by other alternative fuels, for example, ethanol, in the reference baseline. This is because 49 U.S.C. 32902(h) does not just place guardrails on NHTSA's consideration of manufacturers producing BEVs in response to CAFE standards, but all dedicated alternative fueled automobiles, and fuels produced by the commenters here are, as listed above, considered alternative fuels. NHTSA does consider some alternative-fueled vehicle adoption in the reference baseline where that adoption is driven for reasons other than NHTSA's standards (see Section IV), and the commenters do mention the RFS as a driver of the increased use of renewable alternative fuels like ethanol and biofuels. However, the RFS is a regulation that increases the use of renewable fuels to replace petroleum derived fuels in motor gasoline, and to the extent that EPA has approved the use of E15 in all model year 2001 and newer gasoline vehicles produced for the U.S. market, we account for that in our analysis. NHTSA also considers flexible fuel vehicles (FFVs) that exist in the reference baseline fleet in the analysis, however FFVs are also subject to the restrictions in 49 U.S.C. 32902(h)(2). NHTSA applies the same CAFE Model restrictions in the standard-setting analysis to FFVs that apply to PHEVs to ensure that the agency is not improperly considering the alternative-fueled operation of dual-fueled vehicles when setting CAFE standards.

There is also a practical consideration that while blending ethanol or biofuels with gasoline has the potential to reduce U.S. reliance on petroleum, renewable fuels like ethanol and biofuels decrease fuel economy. The fuel economy of FFVs operating on high-ethanol blends are worse than when operating on conventional gasoline, because although ethanol has a higher octane rating than petroleum gasoline, it is less energy dense. For example, a model year 2022 Ford F150 4WD achieves a real world combined 20 mpg rating on conventional gas versus 15 mpg on alternative E85 fuel. FFVs do see a compliance boost in the CAFE program with a 0.15 multiplier, however, again NHTSA's consideration of those vehicles' fuel economy values to set higher fuel economy standards is limited by 49 U.S.C. 32902(h)(2).

Regarding comments about energy security, we discuss this further in preamble Section VI. As mentioned above, commenters suggested that consideration of BEVs also impacts NHTSA's statutory considerations of energy security. However, NHTSA does not consider BEVs in its standard-setting, and notes that this final rule is not a BEV mandate, as claimed by some commenters. Results in preamble Section V and FRIA Chapter 8 show that manufacturers have a wide variety of technology options to meet both LD and HDPUV standards, and the paths to compliance modeled in this analysis represent only a possible path, and not a required path. NHTSA does not mandate any one technology that manufacturers must use, hence why we have evaluated an array of technologies for manufacturers to use for meeting the standards. As with other technologies in the analysis, nothing prevents manufacturers from using FFVs or other dedicated alternative fueled vehicles to comply with CAFE standards.

Finally, NHTSA received a wide variety of comments on compliance aspects of the CAFE program. Although most of them have been summarized and discussed in Section VII of this preamble, we received comments regarding the fuel economy utility factor (UF) compliance calculation for plug-in hybrids. Mitsubishi commented that NHTSA failed to account for EPA's proposal to update the UF calculation for the combined fuel economy for PHEVs, stating that “[t]he result is that NHTSA overestimated the value of PHEV CAFE compliance and underestimated the costs of achieving compliance.” On the other hand, ICCT and the Strong PHEV Coalition supported NHTSA using EPA's new proposed UF approach for the rulemaking analysis. MECA supported NHTSA's continued use of SAE J2841 and recommended that, at a minimum, we should not reduce the UF from the current levels.

We appreciate stakeholders providing comments to NHTSA on PHEV fuel economy calculations. While in the CAFE modeling NHTSA uses SAE J2841 to calculate PHEV fuel economy, for CAFE compliance, NHTSA must use EPA's test procedures. This means that EPA will report fuel economy values to NHTSA beginning in model year 2031 consistent with the new PHEV UF finalized in EPA's final rule. NHTSA chose to use SAE J841 as a simplifying assumption in the model for this analysis to reduce analytical complexity and based on a lack of readily available data from manufacturers; however, choosing to use SAE J2841 versus another PHEV UF results in functionally no difference in NHTSA's standard setting analysis because for the purpose of setting fuel economy standards, NHTSA cannot consider the electric portion of PHEV operation, per statute. For more detailed discussion of modeled PHEV fuel economy values, see TSD Chapter 3.3.

Discussion and responses to other comments can be found throughout this preamble in areas applicable to the comment received.

Nearly every aspect of the NPRM analysis and discussion received some level of comment by at least one commenter. Overall, the comments received included both broad assessments and pointed analyses, and the agency appreciates the level of engagement of commenters in the public comment process and the information and opinions provided.

C. Changes to the CAFE Model in Light of Public Comments and New Information

Comments received to the NPRM were considered carefully within the statutory authority provided by the law, because they are critical for understanding stakeholders' positions, as well as for gathering additional information that can help to inform the agency about aspects or effects of the proposal that the agency may not have considered at the time of the proposal was issued. The views, data, requests, and suggestions contained in the comments help us to form solutions and make appropriate adjustments to our proposals so that we may be better assured that the final standards we set are reasonable for the rulemaking time frame. For this final rule, the agency made substantive changes resulting directly from the suggestions and recommendations from commenters, as well as new information obtained since the time the proposal was developed, and corrections both highlighted by commenters and discovered internally. These changes reflect DOT's long-standing commitment to ongoing refinement and improvement of its approach to estimating the potential impacts of new CAFE standards. Through further consideration and deliberation, and also in response to many public comments received since then, NHTSA has made a number of changes to the CAFE Model since the 2023 NPRM, including those that are listed below and detailed in Section II and III, as well as in the TSD and FRIA that accompany this final rule.

D. Final Standards—Stringency

NHTSA is establishing new CAFE standards for passenger cars (PCs) and light trucks (LTs) produced for model years 2027-2031, setting forth augural CAFE standards for PCs and LTs for model year 2032, and establishing fuel efficiency standards for HDPUVs for model years 2030-2035. Passenger cars are generally sedans, station wagons, and two-wheel drive crossovers and sport utility vehicles (CUVs and SUVs), while light trucks are generally 4WD sport utility vehicles, pickups, minivans, and passenger/cargo vans. NHTSA is establishing standards (represented by alternative PC2LT002, which is the preferred alternative in our analysis) that increase in stringency at 2 percent per year for PCs produced for model years 2027-2031 (and setting forth augural standards that would increase by another 2 percent for PCs produced in model year 2032), at 0 percent per year for LTs produced in model years 2027-2028 and 2 percent per year for LTs produced in model years 2029-2031 (and setting forth augural standards that would increase by another 2 percent for LTs produced in model year 2032). Passenger car and light truck standards are all attribute-based. NHTSA is setting CAFE standards defined by a mathematical function of vehicle footprint, which has an observable correlation with fuel economy. The final standards, and regulatory alternatives, take the form of fuel economy targets expressed as functions of vehicle footprint, which are separate for PCs and LTs. Section IV below discusses NHTSA's continued reliance on footprint as the relevant attribute for PCs and LTs in this final rule.

The target curves for the final passenger car and light truck standards are as follows; curves for model years 2024-2026 are included in the figures for context. NHTSA underscores that the equations and coefficients defining the curves are, in fact, the CAFE standards, and not the mpg numbers that the agency estimates could result from manufacturers complying with the curves. Because the estimated mpg numbers are an effect of the final standards, they are presented in Section II.E. To give context to what the passenger car footprint curve is showing in Figure II-1, for model year 2024, the target for the smallest footprint passenger cars is 55.4 mpg, and the target for the largest footprint passenger cars is 41.5 mpg. For model year 2031, the smallest footprint passenger cars have a target of 74.1 mpg and the largest passenger cars have a target of 55.4 mpg.

To give context to what the light truck footprint curve is showing in Figure II-2, the smallest footprint truck fuel economy target is 44.5 mpg, and the largest truck fuel economy target is 26.7 mpg. And in model year 2031, the smallest truck footprint target is 57.1 mpg, and the largest truck footprint target is 34.3 mpg.

NHTSA has also amended the minimum domestic passenger car standard (MDPCS) for model years 2027-2031 and set forth an augural MDPCS for model year 2032. Section 32902(b)(4) of 49 U.S.C. requires NHTSA to project the MDPCS when it promulgates passenger car standards for a model year, as a result the MDPCSs are established as specific mpg values. NHTSA retains the 1.9-percent offset to the MDPCS, first used in the 2020 final rule, to account for recent projection errors as part of estimating the total passenger car fleet fuel economy. The final MDPCS for model years 2027-2031 and the augural MDPCS for model year 2032 for the preferred alternative are presented in Table II-1.

Heavy-duty pickup trucks and vans are work vehicles that have GVWR between 8,501 pounds to 14,000 pounds (known as Class 2b through 3 vehicles) manufactured as complete vehicles by a single or final stage manufacturer or manufactured as incomplete vehicles as designated by a manufacturer. The majority of these HDPUVs are 3/4 -ton and 1-ton pickup trucks, 12- and 15-passenger vans, and large work vans that are sold by vehicle manufacturers as complete vehicles, with no secondary manufacturer making substantial modifications prior to registration and use. The final standards, represented by alternative HDPUV108 in NHTSA's analysis, increases at a rate of 10 percent per year for model years 2030-2032 and 8 percent per year for model years 2033-2035. The final standards, like the proposed standards, are defined by a linear work factor target function with two sets of sub-configurations with one for spark ignition (SI) that represents gasoline, CNG, strong hybrids, and PHEVs and the other for compression ignition (CI) that represents diesels, BEVs and FCEVs. The target linear curves for HDPUV are still in the same units as in Phase 2 final rule in gallons per 100 miles and for context both the SI and CI curves are shown for model years 2026-2035.

E. Final Standards—Impacts

As for past CAFE rulemakings, NHTSA has used the CAFE Model to estimate the effects of this final rule's light duty CAFE and HDPUV fuel efficiency standards and of other regulatory alternatives under consideration. Some inputs to the CAFE Model are derived from other models, such as Argonne National Laboratory's Autonomie vehicle simulation tool and Argonne's GREET fuel-cycle emissions analysis model, the U.S. Energy Information Administration's (EIA's) National Energy Modeling System (NEMS), and EPA's Motor Vehicle Emissions Simulator (MOVES) vehicle emissions model. Especially given the scope of NHTSA's analysis, these inputs involve a number of uncertainties. NHTSA underscores that all results of today's analysis simply represent the agency's best estimates based on the information currently before us and on the agency's reasonable judgment.

1. Light Duty Effects

NHTSA estimates that this final rule would increase the eventual average of manufacturers' CAFE requirements to about 50.4 mpg by 2031 rather than, under the No-Action Alternative ( i.e., the baseline standards issued in 2023 ending with model year 2026 standards carried forward indefinitely), about 46.9 mpg. For passenger cars, the standards in 2031 are estimated to require 65.1 mpg, and for light trucks, 45.2 mpg. This compares with 58.8 mpg and 42.6 mpg for cars and trucks, respectively, under the No-Action Alternative.

The model year 2032 augural CAFE standard is estimated to require a fleet average fuel economy of 51.4 mpg rather than, under the No-Action Alternative, about 46.9 mpg. For passenger cars, the average in 2032 is estimated to require 66.4 mpg, and for the light trucks, 46.2 mpg.

Because manufacturers do not comply exactly with each standard in each model year, but rather focus their compliance efforts when and where it is most cost-effective to do so, “estimated achieved” fuel economy levels differ somewhat from “estimated required” levels for each fleet, for each year. NHTSA estimates that the industry-wide average fuel economy achieved in model year 2031 could increase from about 52.1 mpg under the No-Action Alternative to 52.5 mpg under the final rule's standards.

The augural achieved CAFE level in model year 2032 is estimated to be 53.5 mpg rather than, under the No-Action Alternative, about 53 mpg. For passenger cars, the fleet average in 2032 is estimated to achieve 72.3 mpg, and for light trucks 47.3 mpg.

NHTSA's analysis estimates manufacturers' potential responses to the combined effect of CAFE standards and separate (reference baseline, model years 2024-2026) CO₂ standards, ZEV programs, and fuel prices. Together, the regulatory programs are more binding ( i.e., require more of manufacturers) than any single program considered in isolation, and today's analysis, like past analyses, shows some estimated overcompliance with the final CAFE standards for both the passenger car and light truck fleets.

NHTSA measures and reports benefits and costs from increasing fuel economy and efficiency standards from two different perspectives. First, the agency's “model year” perspective focuses on benefits and costs of establishing alternative CAFE standards for model years 2027 through 2031 (and fuel efficiency standards for HDPUVs for model years 2030 through 2035), and measures these over each separate model year's entire lifetime. The calendar year perspective we present includes the annual impacts attributable to all vehicles estimated to be in service in each calendar year for which our analysis includes a representation of the entire registered passenger car, light truck, and HDPUV fleet. For this final rule, this calendar year perspective covers each of calendar years 2022-2050, with differential impacts accruing as early as MY 2022. Compared to the model year perspective, the calendar year perspective includes model years of vehicles produced in the longer term, beyond those model years for which standards are being finalized. The strengths and limitations of each accounting perspective is discussed in detail in FRIA Chapter 5.

The table below summarizes estimates of selected impacts viewed from each of these two perspectives, for each of the regulatory alternatives considered in this final rule, relative to the reference baseline.

NHTSA estimates for the final standards are compared to levels of gasoline and electricity consumption NHTSA projects would occur under the No-Action Alternative (i.e., the reference baseline) as shown in Table II-8.

NHTSA's analysis also estimates total annual consumption of fuel by the entire on-road light-duty fleet from calendar year 2022 through calendar year 2050. On this basis, gasoline and electricity consumption by the U.S. light-duty vehicle fleet evolves as shown in Figure II-5 and Figure II-6, each of which shows projections for the No-Action Alternative, PC2LT002 (the Preferred Alternative), PC1LT3, PC2LT4, PC3LT5, and PC6LT8.

Accounting for emissions from both vehicles and upstream energy sector processes ( e.g., petroleum refining and electricity generation), which are relevant to NHTSA's evaluation of the need of the United States to conserve energy, NHTSA estimates that the final rule would reduce greenhouse gas emissions by about 659 million metric tons of carbon dioxide (CO₂), about 825 thousand metric tons of methane (CH₄), and about 24 thousand metric tons of nitrous oxide (N₂ O).

Emissions reductions accrue over time, as the example for CO₂ emissions shows in Figure II-7.

For the “standard setting” analysis, the FRIA accompanying today's notice provides additional detail regarding projected criteria pollutant emissions and health effects, as well as the inclusion of these impacts in today's benefit-cost analysis. For the “unconstrained” or “EIS” analysis, the Final EIS accompanying today's notice presents much more information regarding projected criteria pollutant emissions, as well as model-based estimates of corresponding impacts on several measures of urban air quality and public health. As mentioned above, these estimates of criteria pollutant emissions are based on a complex analysis involving interacting simulation techniques and a myriad of input estimates and assumptions. Especially extending well past 2050, the analysis involves a multitude of uncertainties.

To illustrate the effectiveness of the technology added in response to today's final rule, Table II-10 presents NHTSA's estimates for increased vehicle cost and lifetime fuel expenditures. For more detailed discussion of these and other results related to LD final standards, see Section V below.

With the SC-GHG discounted at 2.0 percent and other benefits and costs discounted at 3 percent, NHTSA estimates that monetized costs and benefits could be approximately $24.5 billion and $59.7 billion, respectively, such that the present value of aggregate monetized net benefits to society could be approximately $35.2 billion. With the SC-GHG discounted at 2.0 percent and other benefits and costs discounted at 7 percent, NHTSA estimates approximately $16.2 billion in monetized costs and $47.0 billion in monetized benefits could be attributable to vehicles produced during and prior to model year 2031 over the course of their lives, such that the present value of aggregate net monetized benefits to society could be approximately $30.8 billion.

2. Heavy Duty Pickup Trucks and Vans Effects

NHTSA estimates that the final rule would increase HDPUV fuel efficiency standards to about 2.851 gals/100 mile by 2035 rather than, under the No-Action Alternative ( i.e., the baseline standards issued in 2016 final rule for Phase 2 ending with model year 2029 standards carried forward indefinitely), about 5.023 gals/100mile. Unlike the light-duty CAFE program, NHTSA may consider AFVs when setting maximum feasible standards for HDPUVs. Additionally, for purposes of calculating average fuel efficiency for HDPUVs, NHTSA considers EVs, fuel cell vehicles, and the proportion of electric operation of EVs and PHEVs that is derived from electricity that is generated from sources that are not onboard the vehicle to have a fuel efficiency value of 0 gallons/mile. NHTSA estimates that the final rule would achieve an average fuel efficiency 2.565 gals/100 mile by 2035 rather than, under the No-Action Alternative, about 2.716 gals/100 mile.

NHTSA estimates that over the lives of vehicles subject to these final HDPUV standards, the final standards would save about 5.6 billion gallons of gasoline and increase electricity consumption (as the percentage of electric vehicles increases over time) by about 56 TWh (a 5.4 percent increase), compared to levels of gasoline and electricity consumption NHTSA projects would occur under the reference baseline standards ( i.e., the No-Action Alternative) as shown in Table II-15.

NHTSA's analysis also estimates total annual consumption of fuel by the entire on-road HDPUV fleet from calendar year 2022 through calendar year 2050. On this basis, gasoline and electricity consumption by the U.S. HDPUV fleet evolves as shown in Figure II-8 and Figure II-9, each of which shows projections for the No-Action Alternative, HDPUV4, HDPUV108 (the Preferred Alternative), HDPUV10, and HDPUV14.

Accounting for emissions from both vehicles and upstream energy sector processes ( e.g., petroleum refining and electricity generation), which are relevant to NHTSA's evaluation of the need of the United States to conserve energy, NHTSA estimates that the final HDPUV standards would reduce greenhouse gas emissions by about 55 million metric tons of carbon dioxide (CO₂), about 65 thousand metric tons of methane (CH₄), and about 3 thousand metric tons of nitrous oxide (N₂ O).

NHTSA's analysis also estimates annual emissions attributable to the entire on-road HDPUV fleet from calendar year 2022 through calendar year 2050. Also accounting for both vehicles and upstream processes, NHTSA estimates that CO₂ emissions from the HDPUV standards could evolve over time as shown in Figure II-10.

To illustrate the effectiveness of the technology added to HDPUVs in response to today's final rule and the overall societal effects of the HDPUV standards, Table II-17 presents NHTSA's estimates for increased vehicle cost and lifetime fuel expenditures and Table II-18 summarizes the benefit-cost analysis. For more detailed discussion of these and other results related to HDPUV final standards, see Preamble Section V and Section VI below.

F. Final Standards Are Maximum Feasible

NHTSA's conclusion, after consideration of the factors described below and information in the administrative record for this action, is that 2 percent increases in stringency for passenger cars for model years 2027-2031, 0 percent increases in stringency for light trucks in model years 2027-2028, and 2 percent increases in stringency for model years 2029-2031 for light trucks (Alternative PC2LT002) are maximum feasible. The Department of Transportation is deeply committed to working aggressively to improve energy conservation and reduce environmental harms and economic and security risks associated with energy use. NHTSA has concluded that Alternative PC2LT002 is technologically feasible, is economically practicable (based on manageable average per-vehicle cost increases, minimal effects on sales, and estimated increases in employment, among other considerations), and is complementary to other motor vehicle standards of the Government on fuel economy that are simultaneously applicable during model years 2027-2031, as described in more detail below.

After consideration of the technical capabilities, economic practicability, statutory requirements, and the Phase 2 final standards, NHTSA has concluded that a 10 percent increase in model years 2030-2032 and an 8 percent increase in model years 2033-2035 for the HDPUV fleet (HDPUV108) is maximum feasible. NHTSA's analysis shows that current Phase 2 standards do not require significant technological improvements through model year 2029, though we expect to see additional fuel efficient technology penetration in model years 2030 through 2035, which can be viewed in more detail in FRIA Chapter 8. Considering our statutory requirements, we have reduced the stringency to 8 percent increases in model years 2033-2035.

See preamble Section VI for more discussion on how we determined that the final CAFE and HDPUV standards are maximum feasible.

G. Final Standards Are Feasible in the Context of EPA's Final Standards and California's Standards

The NHTSA and EPA final rules remain coordinated despite being issued as separate regulatory actions. NHTSA is finalizing CAFE standards that represent the maximum feasible under our program's statutory constraints, which differ to varying degrees by vehicle classification and model year from the GHG standards set forth by the EPA. Overall, EPA's GHG standards, developed under their program's authorities, place a higher degree of stringency on manufacturers in part because of their ability to consider all vehicle technologies, including alternative fueled vehicles, in setting standards. As with past rules, NHTSA's and EPA's programs also differ in other respects, such as programmatic flexibilities. Accordingly, NHTSA's coordination with EPA was limited to areas where each agency's statutory framework allowed some level of harmonization. These differences mean that manufacturers have had (and will continue to have) to plan their compliance strategies considering both the CAFE standards and the GHG standards to ensure that they maintain compliance with both. Because NHTSA and EPA are regulating the same vehicles and manufacturers will use many of the same technologies to meet each set of standards, NHTSA performed appropriate analyses to quantify the differences and their impacts. Auto manufacturers have shown a consistent historical ability to manage compliance strategies that account for the concurrent implementation of multiple regulatory programs. Past experience with these programs indicates that each manufacturer will optimize its compliance strategy around whichever standard is most binding for its fleet of vehicles. If different agencies' standards are more binding for some companies in certain years, this does not mean that manufacturers must build multiple fleets of vehicles, but rather that they will have to be more strategic about how they build their fleet. More detailed discussion of this issue can be found in Section VI.A of this preamble. Critically, NHTSA has concluded that it is feasible for manufacturers to meet the NHTSA standards in a regulatory framework that includes the EPA standards.

NHTSA has also considered and accounted for manufacturers' expected compliance with California's ZEV program (ACC I and ACT) and its adoption by other states in developing the reference baseline for this final rule. We have also accounted for the Framework Agreements between manufacturers who have committed to meeting those Agreements. Finally, we accounted for additional ZEV deployment that manufacturers have committed to undertake, which would be consistent with the requirements of ACC II. NHTSA's assessment regarding the inclusion of ZEVs in the reference baseline is detailed in Preamble Section III.C.5 and Section IV.B.1, and well as in Chapter 3.1 of the accompanying FRIA.

NHTSA also conducted an analysis using an alternative baseline, under which NHTSA removed not only the electric vehicles that would be deployed to comply with ACC I, but also those that would be deployed consistent with manufacturer commitments to deploy additional electric vehicles regardless of legal requirements, consistent with the levels under ACC II. NHTSA describes this as the “No ZEV alternative baseline.” For further reading on this alternative baseline, see RIA Chapters 3 and 8 and Preamble Section IV.B for comparison of the baselines.

III. Technical Foundation for Final Rule Analysis

A. Why is NHTSA conducting this analysis?

NHTSA is finalizing CAFE standards that will increase at 2 percent per year for passenger cars during MYs 2027 through 2031, and for light trucks, standards that will not increase beyond the MY 2026 standards in MYs 2027 through 2028, thereafter increasing at 2 percent per year for MYs 2029 through 2031. The final HDPUV standards will increase at 10 percent per year during MYs 2030 through 2032, and then increase at 8 percent for MYs 2033 through 2035. NHTSA estimates these stringency increases in the passenger car and light truck fleets will reduce gasoline consumption through calendar year 2050 by about 64 billion gallons and increase electricity consumption by about 333 terawatt-hours (TWh). The stringency increases in the HDPUV fleet will reduce gasoline consumption by about 5.6 billion gallons and increase electricity consumption by about 56 TWh through calendar year 2050. Accounting for emissions from both vehicles and upstream energy sector processes ( e.g., petroleum refining and electricity generation), NHTSA estimates that the CAFE standards will reduce greenhouse gas emissions by about 659 million metric tons of carbon dioxide (CO₂), about 825 thousand metric tons of methane (CH₄), and about 23.5 thousand metric tons of nitrous oxide (N₂ 0). The HDPUV standards are estimated to further reduce greenhouse gas emissions by 55 million metric tons of CO₂, 65 thousand metric tons of CH₄ and 3 thousand metric tons of N₂ 0.

When NHTSA promulgates new regulations, it generally presents an analysis that estimates the impacts of those regulations, and the impacts of other regulatory alternatives. These analyses derive from statutes such as the Administrative Procedure Act (APA) and NEPA, from E.O.s (such as E.O. 12866 and 13563), and from other administrative guidance ( e.g., Office of Management and Budget (OMB) Circular A-4). For CAFE and HDPUV standards, EPCA, as amended by EISA, contains a variety of provisions that NHTSA seeks to account for analytically. Capturing all of these requirements analytically means that NHTSA presents an analysis that spans a meaningful range of regulatory alternatives, that quantifies a range of technological, economic, and environmental impacts, and that does so in a manner that accounts for EPCA/EISA's various express requirements for the CAFE and HDPUV programs ( e.g., passenger cars and light trucks must be regulated separately; the standard for each fleet must be set at the maximum feasible level in each MY; etc.).

NHTSA's standards are thus supported by, although not dictated by, extensive analysis of potential impacts of the regulatory alternatives under consideration. Together with this preamble, a TSD, a FRIA, and a Final EIS, provide a detailed enumeration of related methods, estimates, assumptions, and results. These additional analyses can be found in the rulemaking docket for this final rule and on NHTSA's website.

This section provides further detail on the key features and components of NHTSA's analysis. It also describes how NHTSA's analysis has been constructed specifically to reflect governing law applicable to CAFE and HDPUV standards (which may vary between programs). Finally, the discussion reviews how NHTSA's analysis has been expanded and improved in response to comments received on the 2023 proposal, as well as additional work conducted over the last year. The analysis for this final rule aided NHTSA in implementing its statutory obligations, including the weighing of various considerations, by reasonably informing decision-makers about the estimated effects of choosing different regulatory alternatives.

1. What are the key components of NHTSA's analysis?

NHTSA's analysis makes use of a range of data ( i.e., observations of things that have occurred), estimates ( i.e., things that may occur in the future), and models ( i.e., methods for making estimates). Two examples of data include (1) records of actual odometer readings used to estimate annual mileage accumulation at different vehicle ages and (2) CAFE compliance data used as the foundation for the “analysis fleets” containing, among other things, production volumes and fuel economy/fuel efficiency levels of specific configurations of specific vehicle models produced for sale in the U.S. Two examples of estimates include (1) forecasts of future Gross Domestic Product (GDP) growth used, with other estimates, to forecast future vehicle sales volumes and (2) technology cost estimates, which include estimates of the technologies' “direct cost,” marked up by a “retail price equivalent” (RPE) factor used to estimate the ultimate cost to consumers of a given fuel-saving technology, and an estimate of “cost learning effects” ( i.e., the tendency that it will cost a manufacturer less to apply a technology as the manufacturer gains more experience doing so).

NHTSA uses the CAFE Compliance and Effects Modeling System (usually shortened to the “CAFE Model”) to estimate manufacturers' potential responses to new CAFE, HDPUV, and GHG standards and to estimate various impacts of those responses. DOT's Volpe National Transportation Systems Center (often simply referred to as the “Volpe Center”) develops, maintains, and applies the model for NHTSA. NHTSA has used the CAFE Model to perform analyses supporting every CAFE rulemaking since 2001. The 2016 rulemaking regarding HDPUV fuel efficiency standards, NHTSA's most recent HDPUV rulemaking, also used the CAFE Model for analysis.

The basic design of the CAFE Model is as follows: The system first estimates how vehicle manufacturers might respond to a given regulatory scenario, and from that potential compliance solution, the system estimates what impact that response will have on fuel consumption, emissions, safety impacts, and economic externalities. In a highly summarized form, TSD Figure 1-1 shows the basic categories of CAFE Model procedures and the sequential logical flow between different stages of the modeling. The diagram does not present specific model inputs or outputs, as well as many specific procedures and model interactions. The model documentation accompanying this final rule presents these details.

More specifically, the model may be characterized as an integrated system of models. For example, one model estimates manufacturers' responses, another estimates resultant changes in total vehicle sales, and still another estimates resultant changes in fleet turnover ( i.e., scrappage). Additionally, and importantly, the model does not determine the form or stringency of the standards. Instead, the model applies inputs specifying the form and stringency of standards to be analyzed and produces outputs showing the impacts of manufacturers working to meet those standards, which become part of the basis for comparing different potential stringencies. A regulatory scenario, meanwhile, involves specification of the form, or shape, of the standards ( e.g., flat standards, or linear or logistic attribute-based standards), scope of passenger car, light truck, and HDPUV regulatory classes, and stringency of the CAFE or HDPUV standards for each MY to be analyzed. For example, a regulatory scenario may define CAFE or HDPUV standards for a particular class of vehicles that increase in stringency by a given percent per year for a given number of consecutive years.

Manufacturer compliance simulation and the ensuing effects estimation, collectively referred to as compliance modeling, encompass numerous subsidiary elements. Compliance simulation begins with a detailed user-provided initial forecast of the vehicle models offered for sale during the simulation period. The compliance simulation then attempts to bring each manufacturer into compliance with the standards defined by the regulatory scenario contained within an input file developed by the user.

Estimating impacts involves calculating resultant changes in new vehicle costs, estimating a variety of costs ( e.g., for fuel) and effects ( e.g., CO₂ emissions from fuel combustion) occurring as vehicles are driven over their lifetimes before eventually being scrapped, and estimating the monetary value of these effects. Estimating impacts also involves consideration of consumer responses— e.g., the impact of vehicle fuel economy/efficiency, operating costs, and vehicle price on consumer demand for passenger cars, light trucks, and HDPUVs. Both basic analytical elements involve the application of many analytical inputs. Many of these inputs are developed outside of the model and not by the model. For example, the model applies fuel prices; it does not estimate fuel prices.

NHTSA also uses EPA's Motor Vehicle Emission Simulator (MOVES) model to estimate “vehicle” or “downstream” emission factors for criteria pollutants, and uses four Department of Energy (DOE) and DOE-sponsored models to develop inputs to the CAFE Model, including three developed and maintained by DOE's Argonne National Laboratory (Argonne). The agency uses the DOE Energy Information Administration's (EIA's) National Energy Modeling System (NEMS) to estimate fuel prices, and uses Argonne's Greenhouse gases, Regulated Emissions, and Energy use in Transportation (GREET) model to estimate emissions rates from fuel production and distribution processes. DOT also sponsored DOE/Argonne to use Argonne's Autonomie full-vehicle modeling and simulation system to estimate the fuel economy/efficiency impacts for over a million combinations of technologies and vehicle types. The TSD and FRIA describe details of our use of these models. In addition, as discussed in the Final EIS accompanying this final rule, DOT relied on a range of models to estimate impacts on climate, air quality, and public health. The Final EIS discusses and describes the use of these models.

To prepare for the analysis that supports this final rule, DOT has refined and expanded the CAFE Model through ongoing development. Examples of such changes, some informed by past external comment, made since 2022 include:

• Updated analysis fleet
• Addition of HDPUVs, and associated required updates across entire model
• Updated technologies considered in the analysis

○ Addition of HCRE, HCRD and updated diesel technology models

○ Removal of EFR, DSLIAD, manual transmissions, AT6L2, EPS, IACC, LDB, SAX, and some P2 combinations

• User control of additional input parameters
• Updated modeling approach to manufacturers' expected compliance with states' ZEV programs
• Expanded accounting for Federal incentives, such as the IRA
• Expanded procedures for estimating new vehicle sales and fleet shares
• VMT coefficient updates

In response to feedback, interagency meetings, comments from stakeholders, as well as continued development, DOT has made additional changes to the CAFE Model for the final rule. Since the 2023 NPRM, DOT has made the following changes to the CAFE Model and inputs, including:

• Updated battery costs for electrified technologies
• Updated different phase-in penetration for different BEV ranges
• Updated ZEV State shares, credit values and projected ZEV requirements to inform the reference baseline
• Reclassified Rivian and Ford vehicles from HDPUV to LD based on official certification data submission
• Allow the user to directly input AC efficiency, AC leakage and off cycle credit limits for each MY, separately for conventional ICE vehicles and electric vehicles

Addressed issues with when road load technologies are applied to the fleet

• Updated and expanded model reporting capabilities
• Updated IRA Tax Credit implementation
• Updated input factors for economic models
• Updated input factors for the safety models
• Updated emission modeling

These changes reflect DOT's long-standing commitment to ongoing refinement of its approach to estimating the potential impacts of new CAFE and HDPUV standards. The TSD elaborates on these changes to the CAFE Model, as well as changes to inputs to the model for this analysis.

NHTSA underscores that this analysis uses the CAFE Model in a manner that explicitly accounts for the fact that in producing a single fleet of vehicles for sale in the United States, manufacturers make decisions that consider the combination of CAFE/HDPUV standards, EPA GHG standards, and various policies set at sub-national levels ( e.g., ZEV regulatory programs, set by California and adopted by many other states). These regulations have important structural and other differences that affect the strategy a manufacturer could pursue in designing a fleet that complies with each of the above. As explained, NHTSA's analysis reflects a number of statutory and regulatory requirements applicable to CAFE/HDPUV and EPA GHG standard-setting. As stated previously, NHTSA coordinated with EPA and DOE to optimize the effectiveness of NHTSA's standards while minimizing compliance costs, informed by public comments from all stakeholders and consistent with the statutory factors.

2. How do requirements under EPCA/EISA shape NHTSA's analysis?

EPCA contains multiple requirements governing the scope and nature of CAFE standard setting. Some of these have been in place since EPCA was first signed into law in 1975, and some were added in 2007, when Congress passed EISA and amended EPCA. EISA also gave NHTSA authority to set standards for HDPUVs, and that authority was generally less constrained than for CAFE standards. NHTSA's modeling and analysis to inform standard setting is guided and shaped by these statutory requirements. EPCA/EISA requirements regarding the technical characteristics of CAFE and HDPUV standards and the analysis thereof include, but are not limited to, the following:

Corporate Average Standards: Section 32902 of 49 U.S.C. requires standards for passenger cars, light trucks, and HDPUVs to be corporate average standards, applying to the average fuel economy/efficiency levels achieved by each corporation's fleets of vehicles produced for sale in the U.S. The CAFE Model calculates the CAFE and CO₂ levels of each manufacturer's fleets based on estimated production volumes and characteristics, including fuel economy/efficiency levels, of distinct vehicle models that could be produced for sale in the U.S.

Separate Standards for Passenger Cars, Light Trucks, and HDPUVs: Section 32902 of 49 U.S.C. requires the Secretary of Transportation to set CAFE standards separately for passenger cars and light trucks and allows the Secretary to prescribe separate standards for different classes of heavy-duty (HD) vehicles like HDPUVs. The CAFE Model accounts separately for differentiated standards and compliance pathways for passenger cars, light trucks, and HDPUVs when it analyzes CAFE/HDPUV or GHG standards.

Attribute-Based Standards: Section 32902 of 49 U.S.C. requires the Secretary of Transportation to define CAFE standards as mathematical functions expressed in terms of one or more vehicle attributes related to fuel economy, and NHTSA has extended this approach to HDPUV standards as well through regulation. This means that for a given manufacturer's fleet of vehicles produced for sale in the U.S. in a given regulatory class and MY, the applicable minimum CAFE requirement (or maximum HDPUV fuel consumption requirement) is computed based on the applicable mathematical function, and the mix and attributes of vehicles in the manufacturer's fleet. The CAFE Model accounts for such functions and vehicle attributes explicitly.

Separately Defined Standards for Each Model Year: Section 32902 of 49 U.S.C. requires the Secretary of Transportation (by delegation, NHTSA) to set CAFE standards (separately for passenger cars and light trucks) at the maximum feasible levels in each MY. Fuel efficiency levels for HDPUVs must also be set at the maximum feasible level, in tranches of (at least) 3 MYs at a time. The CAFE Model represents each MY explicitly, and accounts for the production relationships between MYs.

Separate Compliance for Domestic and Imported Passenger Car Fleets: Section 32904 of 49 U.S.C. requires the EPA Administrator to determine CAFE compliance separately for each manufacturer's fleets of domestic passenger cars and imported passenger cars, which manufacturers must consider as they decide how to improve the fuel economy of their passenger car fleets. The CAFE Model accounts explicitly for this requirement when simulating manufacturers' potential responses to CAFE standards, and combines any given manufacturer's domestic and imported cars into a single fleet when simulating that manufacturer's potential response to GHG standards (because EPA does not have separate standards for domestic and imported passenger cars).

Minimum CAFE Standards for Domestic Passenger Car Fleets: Section 32902 of 49 U.S.C. requires that domestic passenger car fleets meet a minimum standard, which is calculated as 92 percent of the industry-wide average level required under the applicable attribute-based CAFE standard, as projected by the Secretary at the time the standard is promulgated. The CAFE Model accounts explicitly for this requirement when simulating manufacturer compliance with CAFE standards and sets this requirement aside when simulating manufacturer compliance with GHG standards.

Civil Penalties for Noncompliance: Section 32912 of 49 U.S.C. (and implementing regulations) prescribes a rate (in dollars per tenth of a mpg) at which the Secretary is to levy civil penalties if a manufacturer fails to comply with a passenger car or light truck CAFE standard for a given fleet in a given MY, after considering available credits. Some manufacturers have historically chosen to pay civil penalties rather than achieve full numerical compliance across all fleets. The CAFE Model calculates civil penalties (adjusted for inflation) for CAFE shortfalls and provides means to estimate that a manufacturer might stop adding fuel-saving technologies once continuing to do so would effectively be more “expensive” (after accounting for fuel prices and buyers' willingness to pay for fuel economy) than paying civil penalties. The CAFE Model does not allow civil penalty payment as an option for EPA's GHG standards or NHTSA's HDPUV standards.

Dual-Fueled and Dedicated Alternative Fuel Vehicles: For purposes of calculating passenger car and light truck CAFE levels used to determine compliance, 49 U.S.C. 32905 and 32906 specify methods for calculating the fuel economy levels of vehicles operating on alternative fuels to gasoline or diesel, such as electricity. In some cases, after MY 2020, methods for calculating AFV fuel economy are governed by regulation. The CAFE Model can account for these requirements explicitly for each vehicle model. However, 49 U.S.C. 32902 prohibits consideration of the fuel economy of dedicated Alternative Fuel Vehicles (AFVs), and requires that the fuel economy of dual-fueled AFVs' fuel economy, such as plug-in electric vehicles (EVs), be calculated as though they ran only on gasoline or diesel, when NHTSA determines the maximum feasible fuel economy level that manufacturers can achieve, in a given year for which NHTSA is establishing CAFE standards. The CAFE Model therefore has an option to be run in a manner that excludes the additional application of dedicated AFVs and counts only the gasoline fuel economy of dual-fueled AFVs, in MYs for which maximum feasible standards are under consideration. As allowed under NEPA for analysis appearing in Environmental Impact Statements (EIS) that help inform decision makers about the environmental impacts of CAFE standards, the CAFE Model can also be run without this analytical constraint. The CAFE Model does account for dedicated and dual-fueled AFVs when simulating manufacturers' potential responses to EPA's GHG standards because the Clean Air Act (CAA), under which the EPA derives its authority to set GHG standards for motor vehicles, contains no restrictions in using AFVs for compliance. There are no specific statutory directions in EISA with regard to dedicated and dual-fueled AFV fuel efficiency for HDPUVs, so the CAFE Model reflects relevant regulatory provisions by calculating fuel consumption directly per 49 U.S.C. 32905 and 32906 specified methods.

ZEV Regulatory Programs: The CAFE Model can simulate manufacturers' compliance with state-level ZEV programs applicable in California and “Section 177” states. This approach involves identifying specific vehicle model/configurations that could be replaced with BEVs and converting to BEVs only enough sales count of the vehicle models to meet the manufacturer's compliance obligations under state-level ZEV programs, before beginning to consider the potential that other technologies could be applied toward compliance with CAFE, HDPUV, or GHG standards.

Creation and Use of Compliance Credits: Section 32903 of 49 U.S.C. provides that manufacturers may earn CAFE “credits” by achieving a CAFE level beyond that required of a given passenger car or light truck fleet in a given MY and specifies how these credits may be used to offset the amount by which a different fleet falls short of its corresponding requirement. These provisions allow credits to be “carried forward” and “carried back” between MYs, transferred between regulated classes (domestic passenger cars, imported passenger cars, and light trucks), and traded between manufacturers. However, credit use for passenger car and light truck compliance is also subject to specific statutory limits. For example, CAFE compliance credits can be carried forward a maximum of five MYs and carried back a maximum of three MYs. Also, EPCA/EISA caps the amount of credits that can be transferred between passenger car and light truck fleets and prohibits manufacturers from applying traded or transferred credits to offset a failure to achieve the applicable minimum standard for domestic passenger cars. The CAFE Model can simulate manufacturers' potential use of CAFE credits carried forward from prior MYs or transferred from other fleets. Section 32902 of 49 U.S.C. prohibits consideration of manufacturers' potential application of CAFE compliance credits when determining the maximum feasible fuel economy level that manufacturers can achieve for their fleets of passenger cars and light trucks. The CAFE Model can be operated in a manner that excludes the application of CAFE credits for a given MY under consideration for standard setting, and NHTSA operated the model with that constraint for the purpose of determining the appropriate CAFE standard for passenger cars and light trucks. No such statutory restrictions exist for setting HDPUV standards. For modeling EPA's GHG standards, the CAFE Model does not limit transfers because the CAA does not limit them. Insofar as the CAFE Model can be exercised in a manner that simulates trading of GHG compliance credits, such simulations treat trading as unlimited.

Statutory Basis for Stringency: Section 32902 of 49 U.S.C. requires the Secretary of Transportation (by delegation, NHTSA) to set CAFE standards for passenger cars and light trucks at the maximum feasible levels that manufacturers can achieve in a given MY, considering technological feasibility, economic practicability, the need of the United States to conserve energy, and the impact of other motor vehicle standards of the Government on fuel economy. For HDPUV standards, which must also achieve the maximum feasible improvement, the similar yet distinct factors of appropriateness, cost-effectiveness, and technological feasibility must be considered. EPCA/EISA authorizes the Secretary of Transportation (by delegation, NHTSA) to interpret these factors, and as the Department's interpretation has evolved, NHTSA has continued to expand and refine its qualitative and quantitative analysis to account for these statutory factors. For example, one of the ways that economic practicability considerations are incorporated into the analysis is through the technology effectiveness determinations: the Autonomie simulations reflect the agency's conservative assumption that it would not be economically practicable (nor, for HDPUVs, appropriate for vehicles with different use cases) for a manufacturer to “split” an engine shared among many vehicle model/configurations into myriad versions each optimized to a single vehicle model/configuration.

National Environmental Policy Act: NEPA requires NHTSA to consider the environmental impacts of its actions in its decision-making processes, including for CAFE standards. The Final EIS accompanying this final rule documents changes in emission inventories as estimated using the CAFE Model, but also documents corresponding estimates—based on the application of other models documented in the Final EIS—of impacts on the global climate, on air quality, and on human health.

Other Aspects of Compliance: Beyond these statutory requirements applicable to DOT, EPA, or both are a number of specific technical characteristics of CAFE, HDPUV, and/or GHG regulations that are also relevant to the construction of this analysis, like the “off-cycle” technology fuel economy/emissions improvements that apply for both CAFE and GHG compliance. Although too little information is available to account for these provisions explicitly in the same way that NHTSA has accounted for other technologies, the CAFE Model includes and makes use of inputs reflecting NHTSA's expectations regarding the extent to which manufacturers may earn such credits, along with estimates of corresponding costs. Similarly, the CAFE Model includes and makes use of inputs regarding credits EPA has elected to allow manufacturers to earn toward GHG levels (not CAFE or HDPUV) based on the use of air conditioner refrigerants with lower global warming potential, or on the application of technologies to reduce refrigerant leakage. In addition, the CAFE Model accounts for EPA “multipliers” for certain AFVs, based on current regulatory provisions or on alternative approaches. Although these are examples of regulatory provisions that arise from the exercise of discretion rather than specific statutory mandate, they can materially impact outcomes.

3. What updated assumptions does the current model reflect as compared to the 2022 final rule and the 2023 NPRM?

Besides the updates to the CAFE Model described above, any analysis of regulatory actions that will be implemented several years in the future, and whose benefits and costs accrue over decades, requires a large number of assumptions. Over such time horizons, many, if not most, of the relevant assumptions in such an analysis are inevitably uncertain. Each successive CAFE and HDPUV analysis seeks to update assumptions to better reflect the current state of the world and the best current estimates of future conditions.

A number of assumptions have been updated since the 2022 final rule and the 2023 NPRM. As discussed below, NHTSA continues to use a MY 2022 reference fleet for passenger cars and light trucks and continues to use an updated HDPUV analysis fleet (the last HDPUV analysis fleet was built in 2016). NHTSA has also updated estimates of manufacturers' compliance credit “holdings,” updated fuel price projections to reflect the U.S. EIA's 2023 Annual Energy Outlook (AEO), updated projections of GDP and related macroeconomic measures, and updated projections of future highway travel. While NHTSA would have made these updates as a matter of course, we note that the ongoing global economic recovery and the ongoing war in Ukraine have impacted major analytical inputs such as fuel prices, GDP, vehicle production and sales, and highway travel. Many inputs remain uncertain, and NHTSA has conducted sensitivity analyses around many inputs to attempt to capture some of that uncertainty. These and other updated analytical inputs are discussed in detail in the TSD and FRIA.

Additionally, as discussed in the TSD, NHTSA calculates the climate benefits resulting from anticipated reductions in emissions of each of three GHGs, CO₂, CH₄, and N₂ O, using estimates of the social costs of greenhouse gases (SC-GHG) values reported in a recent report from EPA (henceforward referred to as the “2023 EPA SC-GHG Report”). In the 2022 final rule and the 2023 NPRM, NHTSA used SC-GHG values recommended by the federal Interagency Working Group (IWG) on the SC-GHG for interim use until updated estimates are available. In this final rule, NHTSA has elected to use the updated values in the 2023 EPA SC-GHG Report to reflect the most recent scientific evidence on the cost of climate damages resulting from emission of GHGs. Those estimates of costs per ton of emissions (or benefits per ton of emissions reductions) are greater than those applied in the analysis supporting the 2022 final rule or the 2023 NPRM. Even still, the estimates NHTSA is now using are not able to fully quantify and monetize a number of important categories of climate damages; because of those omitted damages and other methodological limits, DOT believes its values for SC-GHG are conservative underestimates.

B. What is NHTSA analyzing?

NHTSA is analyzing the effects of different potential CAFE and HDPUV standards on industry, consumers, society, and the world at large. These different potential standards are identified as regulatory alternatives, and amongst the regulatory alternatives, NHTSA identifies which ones the agency is selecting. As in the past several CAFE rulemakings and in the Phase 2 HDPUV rulemaking, NHTSA is establishing attribute-based CAFE and HDPUV standards defined by either a mathematical function of vehicle footprint (which has an observable correlation with fuel economy) or a towing-and-hauling-based WF, respectively. EPCA, as amended by EISA, expressly requires that CAFE standards for passenger cars and light trucks be based on one or more vehicle attributes related to fuel economy, and be expressed in the form of a mathematical function. The statute gives NHTSA discretion as to how to structure standards for HDPUVs, and NHTSA continues to believe that attribute-based standards expressed as a mathematical function remain appropriate for those vehicles as well, given their similarity in many ways to light trucks. Thus, the standards (and the regulatory alternatives) for passenger cars and light trucks take the form of fuel economy targets expressed as functions of vehicle footprint (the product of vehicle wheelbase and average track width) that are separate for passenger cars and light trucks, and the standards and alternatives for HDPUVs take the form of fuel consumption targets expressed as functions of vehicle WF (which is in turn a function of towing and hauling capabilities).

For passenger cars and light trucks, under the footprint-based standards, the function defines a fuel economy performance target for each unique footprint combination within a car or truck model type. Using the functions, each manufacturer thus will have a CAFE average standard for each year that is almost certainly unique to each of its fleets, based upon the footprint and production volumes of the vehicle models produced by that manufacturer. A manufacturer will have separate footprint-based standards for cars and for trucks, consistent with 49 U.S.C. 32902(b)'s direction that NHTSA must set separate standards for cars and for trucks. The functions are mostly sloped, so that generally, larger vehicles ( i.e., vehicles with larger footprints) will be subject to lower mpg targets than smaller vehicles. This is because smaller vehicles are generally more capable of achieving higher levels of fuel economy, mostly because they tend not to have to work as hard (and therefore to require as much energy) to perform their driving task. Although a manufacturer's fleet average standard could be estimated throughout the MY based on the projected production volume of its vehicle fleet (and are estimated as part of EPA's certification process), the standards with which the manufacturer must comply are determined by its final model year (FMY) production figures. A manufacturer's calculation of its fleet average standards, as well as its fleets' average performance at the end of the MY, will thus be based on the production-weighted average target and performance of each model in its fleet.

For passenger cars, consistent with prior rulemakings, NHTSA is defining fuel economy targets as shown in Equation III-1.

Where:

TARGET_FE is the fuel economy target (in mpg) applicable to a specific vehicle model type with a unique footprint combination,

a is a minimum fuel economy target (in mpg),

b is a maximum fuel economy target (in mpg),

c is the slope (in gallons per mile (or gpm) per square foot) of a line relating fuel consumption (the inverse of fuel economy) to footprint, and

d is an intercept (in gpm) of the same line.

Here, MIN and MAX are functions that take the minimum and maximum values, respectively, of the set of included values. For example, MIN [40, 35] = 35 and MAX (40, 25) = 40, such that MIN [ MAX (40, 25), 35] = 35.

For the Preferred Alternative, this equation is represented graphically as the curves in Figure III-1.

For light trucks, also consistent with prior rulemakings, NHTSA is defining fuel economy targets as shown in Equation III-2.

Where:

TARGET_FE is the fuel economy target (in mpg) applicable to a specific vehicle model type with a unique footprint combination,

a, b, c, and d are as for passenger cars, but taking values specific to light trucks,

e is a second minimum fuel economy target (in mpg),

f is a second maximum fuel economy target (in mpg),

g is the slope (in gpm per square foot) of a second line relating fuel consumption (the inverse of fuel economy) to footprint, and

h is an intercept (in gpm) of the same second line.

For the Preferred Alternative, this equation is represented graphically as the curves in Figure III-2.

Although the general model of the target function equation is the same for passenger cars and light trucks, and the same for each MY, the parameters of the function equation differ for cars and trucks. The actual parameters for both the Preferred Alternative and the other regulatory alternatives are presented in Section IV.

The required CAFE level applicable to a passenger car (either domestic or import) or light truck fleet in a given MY is determined by calculating the production-weighted harmonic average of fuel economy targets applicable to specific vehicle model configurations in the fleet, as shown in Equation III-3.

Where:

CAFE_required is the CAFE level the fleet is required to achieve,

i refers to specific vehicle model/configurations in the fleet,

PRODUCTION_i is the number of model configuration i produced for sale in the U.S., and

TARGET_{FE, i} is the fuel economy target (as defined above) for model configuration i.

For HDPUVs, NHTSA has previously set attribute-based standards, but used a work-based metric as the attribute rather than footprint. Work-based measurements such as payload and towing capability are key among the parameters that characterize differences in the design of these vehicles, as well as differences in how the vehicles will be used. Since NHTSA has been regulating HDPUVs, these standards have been based on a work factor (WF) attribute that combines the vehicle's payload and towing capabilities, with an added adjustment for 4-wheel drive vehicles. Again, while NHTSA is not required by statute to set HDPUV standards that are attribute-based and that are described by a mathematical function, NHTSA continues to believe that doing so is reasonable and appropriate for this segment of vehicles, consistent with prior HDPUV standard-setting rulemakings. NHTSA is continuing the use of the work-based attribute and gradually increasing stringency (which for HDPUVs means that standards appear to decline, as compared to passenger car and light truck standards where increasing stringency means that standards appear to increase. This is because HDPUV standards are based on fuel consumption, which is the inverse of fuel economy, the metric that NHTSA is statutorily required to use when setting standards for light-duty vehicle (LDV) fuel use). NHTSA defines HDPUV fuel efficiency targets as shown in Equation III-4.

Where:

c is the slope (in gal/100-miles/WF)

d is the y-intercept (in gal/100-miles)

WF = Work Factor = [0.75 × ( Payload Capacity + Xwd)] + [0.25 × Towing Capacity]

Where:

Xwd = 4wd adjustment = 500 lbs. if the vehicle group is equipped with 4WD and all-wheel drive, otherwise equals 0 lbs. for 2wd

Payload Capacity = GVWR (lbs.)−Curb Weight (lbs.) (for each vehicle group)

Towing Capacity = GCWR (lbs.)−GVWR (lbs.) (for each vehicle group)

For the Preferred Alternative, this equation is represented graphically as the curves in Figure III-3 and Figure III-4.

Similar to the standards for passenger cars and light trucks, NHTSA (and EPA) have historically set HDPUV standards such that each manufacturer's fleet average standard is based on production volume-weighting of target standards for all vehicles, which are based on each vehicle's WF as explained above. Thus, for HDPUVs, the required fuel efficiency level applicable in a given MY is determined by calculating the production-weighted harmonic average of subconfiguration targets applicable to specific vehicle model configurations in the fleet, as shown in Equation III-5.

Where:

Subconfiguration Target Standard_i = fuel consumption standard for each group of vehicles with the same payload, towing capacity, and drive configuration (gallons per 100 miles), and

Volume_i = production volume of each unique subconfiguration of a model type based upon payload, towing capacity, and drive configuration.

Chapter 1 of the TSD contains a detailed description of the use of attribute-based standards, generally, for passenger cars, light trucks, and HDPUVs, and explains the specific decision, in past rules and for the current final rule, to continue to use vehicle footprint as the attribute over which to vary passenger car and light truck stringency, and WF as the attribute over which to vary HDPUV stringency. That chapter also discusses the policy and approach in selecting the specific mathematical functions.

Commenters expressed several concerns regarding the implementation of the fuel economy footprint target curves used for passenger cars and light trucks in this rule. Most concerns fell into one of four categories: the use of alternate or additional factors in generating the curves, the shape of the attribute curve, consideration of how footprint changes may be expressed or used by manufacturers, and considerations of changes made by the EPA in its own rulemaking.

Regarding the use of alternate or additional factors in generating the curves, Rivian commented that NHTSA should reconsider the National Academy of Sciences (NAS) recommendation for multi-attribute standards for CAFE and requested that the agency “more fully describe why” the alternative approach to including electrification as another attribute described in the MYs 2024-2026 proposal “would be inconsistent with its current legal authority.”

In the 2021 NAS Report, the committee recommended that if Congress did not act to remove the prohibition at 49 U.S.C. 32902(h) on considering the fuel economy of dedicated AFVs (like BEVs) in determining maximum feasible CAFE standards, then the Secretary (by delegation, NHTSA) should consider accounting for the fuel economy benefits of ZEVs by “setting the standard as a function of a second attribute in addition to footprint—for example, the expected market share of ZEVs in the total U.S. fleet of new light-duty vehicles—such that the standards increase as the share of ZEVs in the total U.S. fleet increases.” NHTSA remains concerned that adding electrification, specifically, as part of a multi-attribute approach to standards may be inconsistent with our current legal authority. The 49 U.S.C. 32902(h) prohibition against considering the fuel economy of electric vehicles applies to the determination of maximum feasible standards. The attribute-based target curves are themselves the standards. NHTSA therefore does not see how the fuel economy of electric vehicles could be incorporated as an attribute forming the basis of the standards. Moreover, NHTSA further explored and received comments on this issue in the final rule setting standards for MYs 2024-2026. While NHTSA considered this recommendation carefully as part of that rulemaking, NHTSA ultimately agreed with many commenters that including electrification as an attribute on which to base fuel economy standards for that rulemaking could introduce lead time concerns and uncertainty for industry needing to adjust their compliance strategies.

The Center for Environmental Accountability (CEA) also commented on considering the use of acceleration as an additional attribute in the attribute based standard function. The CEA was concerned with capturing the potential trade off manufacturers may make between improved vehicle performance or improved fuel economy. NHTSA provides discussion and reasoning for the agency's approach to performance trade-offs in Section III.C.3 and believes the approach of maintaining performance neutrality is a reasonable method for accounting for the variety of possible manufacturer decisions. Furthermore, to date, every time NHTSA has considered options for which attribute(s) to select, the agency has concluded that a properly designed footprint-based approach provides the best means of achieving the basic policy goals ( i.e., by increasing the likelihood of improved fuel economy across the entire fleet of vehicles) involved in applying an attribute-based standard.

Other commenters expressed concern about the possible influence of the shape, slope or cutpoints of the footprint curve on real-world vehicle footprint size. The Institute for Policy Integrity (IPI) and the Natural Resources Defense Council (NRDC) both argued that NHTSA should flatten the footprint curves to discourage upsizing, because larger vehicles consume more energy. NRDC also stated that “NHTSA should further reduce the footprint of the cutpoint for light trucks based on pickup certification.” Other commenters expressed similar concerns.

NHTSA appreciates these comments but based on the detailed discussion presented in Chapter 1.2.3.1 of the TSD, NHTSA is retaining the same curve shapes for passenger car and light truck standards in this final rule that NHTSA has used over the past several rulemakings—that is, at this time NHTSA is not changing the shape of the existing footprint curves. Based on the analysis of data presented by the EPA Trends Report discussed in the TSD, vehicle footprint size, by vehicle category, has in fact changed very little over the last decade. By sales-weighted average, the data examined showed that sedans and wagons increased their footprints the most, about 3.4% or a 2 ft increase, over 10 years. For context, a 1.5 ft increase in overall footprint increase would equate to about a 2 inch increase in the track width of a MY 2022 Toyota Corolla. NHTSA's assessment in the TSD shows that over the 10 years it took for manufacturers to increase sedan footprint by 3.4% on average, the fuel economy consequence was approximately a 3% reduction in the MY 2022 fuel economy target for a Toyota Corolla, compared to if it had retained its MY 2012 footprint size. Spread over each of those 10 years, the footprint increases for the example Corolla resulted in fuel economy targets that were lowered by approximately 0.3% per year. While NHTSA agrees that this number is greater than zero, for context, the fuel economy standard improvement from MY 2023 to MY 2024 will require approximately an 8% increase in fuel economy—in other words, the increases in CAFE stringency are decidedly outpacing manufacturers' current ability, or plans, to upsize individual vehicle footprints to obtain lower targets.

NHTSA notes, however, that while increases in footprint size by vehicle category are small, there is a separate phenomenon of aggregate footprint increase for the entire fleet, which NHTSA found to be about 5.4% over the same time period. This is due not to changes in individual vehicle size or vehicle-class-level size, but to changes in fleet share. The fleet share of generally-smaller-footprint sedans and wagons decreased by nearly 28.4% over 10 years, while the fleet share of generally-larger-footprint trucks, SUVs, and pickups increased by 29.5%. Simply put, manufacturers are selling more larger trucks and fewer smaller cars than they were 10 years ago—which is different from individual vehicle models (or vehicle classes) themselves increasing in size, as one might expect if the shape of the footprint curves or the use of footprint as an attribute were incentivizing upsizing. This evidence leads us to conclude that the use of footprint as an attribute and the current slopes and cutoff points for the existing curves for passenger car and light truck CAFE standards do not lead to manufacturers significantly altering the size of their vehicles, within vehicle classes.

In contrast, Mitsubishi argued that the current shape of the curves, and particularly the passenger car curve, discouraged manufacture of smaller footprint vehicles. As Mitsubishi stated,

Mitsubishi holds a unique position in the industry as the manufacturer with the smallest fleet-average vehicle footprint. As such, Mitsubishi also has the strictest GHG and CAFE standard among vehicle manufacturers. Despite having one of the highest fleet-average fuel economy ratings and the lowest fleet GHG emissions of any mass-market vehicle manufacturer, Mitsubishi has accrued CAFE and GHG deficits in recent years, while other manufacturers with lower CAFE and higher GHG fleet emissions have accrued credits. While we understand the math that delivers this result, we question whether this outcome is what the program set out to achieve. Mitsubishi supports the reevaluation of the shape and slope of the footprint curves to ensure fleetwide fuel economy increases and GHG reductions are done in a neutral manner.

NHTSA is aware of Mitsubishi's unique position in the industry as a manufacturer of smaller, highly fuel-efficient, affordably-priced vehicles and is sympathetic to these comments. Unfortunately, the standard is designed for the overall industry rather than for individual manufacturers. The format of NHTSA's standards, with target goals based on footprint, instead allows each manufacturer's compliance obligation to vary with their sales mix. This can cause difficulty for some manufacturers if their vehicles' average fuel economy does not meet the required average of their footprint targets. Mitsubishi is correct that the current curve shapes do not incentivize manufacturers to build smaller cars—but neither does NHTSA find, as discussed above, that they particularly incentivize manufacturers to build larger cars, perhaps contrary to expectation. Unfortunately, the overall structure of the target curves places Mitsubishi—like all other manufacturers—in a position where it must balance its need to increase the fuel economy of its fleet with marketing increasing vehicle costs to its consumer base.

IPI suggested that NHTSA add the use of increased footprint size as a potential compliance strategy used during the simulation of manufacturer behavior, stating that “This upsizing could be modeled either directly as a vehicle-level change ( i.e., a technology change) or approximated by applying a specific level of sales-weighted average increase to the vehicle class level. In the former case, NHTSA could include footprint technology options, such as increased footprint size by 0%, 5%, 7.5%, 10%, 15%, and 20%, much like NHTSA treats mass-reduction technologies.”

NHTSA disagrees that additional modeling approaches are required to capture the behavior of the manufacturers that appears to lead to increasing fleet footprint. The analysis of the EPA's Trends Data, discussed above and provided in detail in TSD Chapter 1.2.3.1, indicates that over the last 10 years vehicle footprint size has seen only small changes within vehicle classes. Sedans and wagons showed the greatest sales-weighted average increase between MY 2012 and MY 2022 at a 3.4% increase, minivans saw a 2.1% increase, car SUVs (or crossovers) saw a 1.6% increase, truck SUVs saw a 0.9% increase, and pickups saw the smallest increase at 0.5%. The increase in sales-weighted average footprint size for the aggregate fleet instead appears driven by a change in fleet shares between passenger cars and light trucks—a behavior that is captured by the CAFE model and is discussed in TSD Chapter 4.2.1.3, Modeling Changes in Fleet Mix.

Several commenters expressed concern that NHTSA had not followed EPA's proposed approach to reconfiguring their attribute-based CO₂ standard functions. Mitsubishi stated, “Unlike the EPA, NHTSA did not propose any changes to the slope or cut-points for the passenger car or light truck curves.” The Motor & Equipment Manufacturer's Association (MEMA) offered similar comments, stating, “NHTSA should follow EPA's lead in flattening the curves to further improve the fuel efficiency of the overall fleet and limit upsizing.” Other commenters also expressed concern about the departure in target curve shape between EPA's proposed standards and NHTSA proposed standards, arguing that NHTSA should have considered the same factors EPA used in their determinations.

NHTSA has explained our position on changing curve shape based on addressing concerns about upsizing above. That said, NHTSA is aware that EPA recently issued a final rule changing the shapes of its CO₂ standards curves for passenger cars and light-duty trucks, as compared to its prior set of standards. EPA explained that it chose to make the slopes of both curves, especially the car curves, flatter than those of prior rulemakings, stating that:

When emissions reducing technology is applied, such as advanced ICE, or HEV or PHEV or BEV electrification technologies, the relationship between increased footprint and tailpipe emissions is reduced. From a physics perspective, a positive footprint slope for ICE vehicles makes sense because as a vehicle's size increases, its mass, road loads, and required power (and corresponding vehicle-based CO₂ emissions) will increase accordingly [and its fuel economy will correspondingly decrease accordingly]. Moreover, as the emissions control technology becomes increasingly more effective, the relationship between tailpipe emissions and footprint decreases proportionally; in the limiting case of vehicles with 0 g/mile tailpipe emissions such as BEVs, there is no relationship at all between tailpipe emissions and footprint.

Since the Supreme Court's decision in Massachusetts v. EPA , NHTSA and EPA have both employed equivalent footprint-based CAFE and CO₂ target curves for PCs and LTs. In this final rule, NHTSA cannot reasonably promulgate target curves that are flatter, like EPA's new curves based on EPA's rationale, for two main reasons. First, EPA altered their curves based on considering the effects of emission reduction technologies such as PHEVs and BEVs as viable solutions to meet their standards. Given that the target curves are the CAFE standards, and given that 49 U.S.C. 32902(h) prohibits consideration of BEVs or even the electric only operation of PHEVs in determining maximum feasible CAFE standards, NHTSA does not believe that the law permits us to base target curve shapes in CAFE-standard-driven increases on the presence ( i.e., the fuel economy) of BEVs or the use of the electric operation of PHEVs in the vehicle fleets. Second, even if NHTSA could consider BEVs and full use of PHEV technology in developing target curve shapes, NHTSA would not consider them the same way as EPA does. BEV compliance values in the CAFE program are determined, per statute, using DOE's Petroleum Equivalency Factor. Moreover, the calculated equivalent fuel economies still vary with vehicle footprint and, in general, larger vehicles have lower calculated equivalent fuel economies. They are not the fuel-economy-equivalent of 0 g/mi, which would be infinite fuel economy. NHTSA, therefore, cannot adopt EPA's rationale that curve slopes should become flatter in response to increasing numbers of BEVs because our statutory requirements for how BEV fuel economy is calculated necessarily differ from how EPA chooses to calculate CO₂ emissions for BEVs. NHTSA understands that this divergence in curve shape creates inconsistency between the programs, but NHTSA does not agree that the agency currently has authority to harmonize with EPA's new approach to curve shape.

Regarding the fuel consumption work factor target curves proposed for HDPUVs, stakeholders expressed two types of comments. First, a group of commenters expressed support for the continued use of the work factor attribute, and second, some stakeholders expressed concern over NHTSA maintaining separate diesel and gasoline compliance curves.

On the use of the work factor attribute, the Alliance stated, “We agree with NHTSA's conclusion that work factor is a reasonable and appropriate attribute for setting fuel consumption standards. Work factor effectively captures the intent of these vehicles, which is to perform work, and has a strong correlation to fuel consumption.” These sentiments were echoed by other commenters. NHTSA agrees with the stakeholders, and after considering these comments, the agency has once again concluded that the work factor approach established in the 2011 “Phase 1” rulemaking and continued in the 2016 “Phase 2” rulemaking is reasonable and appropriate.

On the continued use of separate diesel and gasoline curves for the HDPUV standards, the American Council for an Energy-Efficient Economy (ACEEE) commented, “In further alignment with EPA, NHTSA should eliminate the different standards for diesel and gasoline ( i.e., compression-ignition and spark-ignition) HDPUVs.” ACEEE argued further that “Given NHTSA's acknowledgement of the emergence of van electrification and its history of alignment with EPA for HDPUVs, raising the stringency of the gasoline standards to match that of the diesel standards should be feasible.”

ACEEE requested that NHTSA align with EPA by developing a single standard curve for both SI and CI HDPUVs for MYs 2027 through 2032. As mentioned in the NPRM, NHTSA is statutorily required to provide at least four full MYs of lead time and three full MYs of regulatory stability for its HDPUV fuel consumption standards. As such, we are unable to align with EPA's change to its standard due to an insufficient amount of lead time. However, we believe the regulatory stability of the current HDPUV fuel consumption standards provide enough stability for the industry to continue to develop technologies needed to meet our standards. In addition, we believe retaining separate CI and SI curves will better balance NHTSA's statutory factors.

C. What inputs does the compliance analysis require?

The first step in our analysis of the effects of different levels of fuel economy standards is the compliance simulation. When we say, “compliance simulation” throughout this rulemaking, we mean the CAFE Model's simulation of how vehicle manufacturers could comply with different levels of CAFE standards by adding fuel economy-improving technology to an existing fleet of vehicles. At the most basic level, a model is a set of equations, algorithms, or other calculations that are used to make predictions about a complex system, such as the environmental impact of a particular industry or activity. A model may consider various inputs, such as emissions data, technology costs, or other relevant factors, and use those inputs to generate output predictions.

One important note about models is that a model is only as good as the data and assumptions that go into it. We attempt to ensure that the technology inputs and assumptions that go into the CAFE Model to project the effects of different levels of CAFE standards are based on sound science and reliable data, and that our reasons for using those inputs and assumptions are transparent and understandable to stakeholders. This section and the following section discuss at a high level how we generate the technology inputs and assumptions that the CAFE Model uses for the compliance simulation. The TSD, CAFE Model Documentation, CAFE Analysis Autonomie Model Documentation, and other technical reports supporting this final rule discuss our technology inputs and assumptions in more detail.

We incorporate technology inputs and assumptions either directly in the CAFE Model or in the CAFE Model's various input files. The heart of the CAFE Model's decisions about how to apply technologies to manufacturer's vehicles to project how the manufacturer could meet CAFE standards is the compliance simulation algorithm. The compliance simulation algorithm is several equations that direct the model to apply fuel economy-improving technologies to vehicles in a way that estimates how manufacturers might apply those technologies to their vehicles in the real world. The compliance simulation algorithm projects a cost-effective pathway for manufacturers to comply with different levels of CAFE standards, considering the technology present on manufacturer's vehicles now, and what technology could be applied to their vehicles in the future. Embedded directly in the CAFE Model is the universe of technology options that the model can consider and some rules about the order in which it can consider those options and estimates of how effective fuel economy improving-technology is on different types of vehicles, like on a sedan or a pickup truck.

Technology inputs and assumptions are also located in all four of the CAFE Model Input Files. The Market Data Input File is a Microsoft Excel file that characterizes the analysis automotive fleet used as the starting point for CAFE modeling. There is one Excel row describing each vehicle model and model configuration manufactured in the United States in a MY (or years), and input and assumption data that links that vehicle to technology, economic, environmental, and safety effects. Next, the Technologies Input File identifies approximately six dozen technologies we use in the analysis, uses phase-in caps to identify when and how widely each technology can be applied to specific types of vehicles, provides most of the technology costs (only battery costs for electrified vehicles are provided in a separate file), and provides some of the inputs involved in estimating impacts on vehicle fuel consumption and weight. The Scenarios Input File provides the coefficient values defining the standards for each regulatory alternative, and other relevant information applicable to modeling each regulatory scenario. This information includes, for example, the estimated value of select tax credits from the IRA, which provide Federal technology incentives for electrified vehicles, and the PEF, which is a value that the Secretary of Energy determines under EPCA that applies to EV fuel economy values. Finally, the Parameters Input File contains mainly economic and environmental data, as well as data about how fuel economy credits and California's Zero Emissions Vehicle program credits are simulated in the model.

We generate these technology inputs and assumptions in several ways, including by and through evaluating data submitted by vehicle manufacturers pursuant to their CAFE reporting obligations; consolidating public data on vehicle models from manufacturer websites, press materials, marketing brochures, and other publicly available information; collaborative research, testing, and modeling with other Federal agencies, like the DOE's Argonne National Laboratory; research, testing, and modeling with independent organizations, like IAV GmbH Ingenieurgesellschaft Auto und Verkehr (IAV), Southwest Research Institute (SwRI), NAS, and FEV North America; determining that work done for prior rules is still relevant and applicable; considering feedback from stakeholders on prior rules, in meetings conducted before the commencement of this rule, and feedback received during the comment period for this final rule; and using our own engineering judgment. When we say “engineering judgment” throughout this rulemaking, we are referring to decisions made by a team of engineers and analysts. This judgment is based on their experience working in the automotive industry and other relevant fields, and assessment of all the data sources described above. Most importantly, we use engineering judgment to assess how best to represent vehicle manufacturer's potential responses to different levels of CAFE standards within the boundaries of our modeling tools, as “a model is meant to simplify reality in order to make it tractable.” In other words, we use engineering judgment to concentrate potential technology inputs and assumptions from millions of discrete data points from hundreds of sources to three datasets integrated in the CAFE Model and four input files. How the CAFE Model decides to apply technology, i.e., the compliance simulation algorithm, has also been developed using engineering judgment, considering some of the same factors that manufacturers consider when they add technology to vehicles in the real world.

While upon first read this discussion may seem oversimplified, we believe that there is value in all stakeholders being able to understand how the analysis uses different sets of technology inputs and assumptions and how those inputs and assumptions are based on real-world factors. This is so that all stakeholders have the appropriate context to better understand the specific technology inputs and assumptions discussed later and in detail in all of the associated technical documentation.

1. Technology Options and Pathways

We begin the compliance analysis by defining the range of fuel economy-improving technologies that the CAFE Model could add to a manufacturer's vehicles in the United States market. These are technologies that we believe are representative of what vehicle manufacturers currently use on their vehicles, and that vehicle manufacturers could use on their vehicles in the timeframe of the standards (MYs 2027 and beyond for the LD analysis and MYs 2030 and beyond for the HDPUV analysis). The technology options include basic and advanced engines, transmissions, electrification, and road load technologies, which include mass reduction (MR), aerodynamic improvement (AERO), and tire rolling resistance (ROLL) reduction technologies. Note that while EPCA/EISA constrains our ability to consider the possibility that manufacturers would comply with CAFE standards by implementing some electrification technologies when making decisions about the level of CAFE standards that is maximum feasible, there are several reasons why we must accurately model the range of available electrification technologies. These are discussed in more detail in Section III.D and in Section VI.

We require several data elements to add a technology to the range of options that the CAFE Model can consider; those elements include a broadly applicable technology definition, estimates of how effective that technology is at improving a vehicle's fuel economy value on a range of vehicles ( e.g., sedan through pickup truck, or HD pickup truck and HD van), and the cost to apply that technology on a range of vehicles. Each technology we select is designed to be representative of a wide range of specific technology applications used in the automotive industry. For example, in MY 2022, eleven vehicle brands under five vehicle manufacturers used what we call a “downsized turbocharged engine with cylinder deactivation.” While we might expect brands owned by the same manufacturer to use similar technology on their engines, among those five manufacturers, the engine systems will likely be very different. Some manufacturers may also have been making those engines longer than others, meaning that they have had more time to make the system more efficient while also making it cheaper, as they make gains learning the development improvement and production process. If we chose to model the best performing, cheapest engine and applied that technology across vehicles made by all automotive manufacturers, we would likely be underestimating the cost and underestimating the technology required for the entire automotive industry to achieve higher levels of CAFE standards. The reverse would be true if we selected a system that was less efficient and more expensive. So, in reality, some manufacturers' systems may perform better or worse than our modeled systems, and some may cost more or less than our modeled systems. However, selecting representative technology definitions for our analysis will ensure that, on balance, we capture a reasonable level of costs and benefits that would result from any manufacturer applying the technology.

We have been refining the LD technology options since first developing the CAFE Model in the early 2000s. “Refining” means both adding and removing technology options depending on technology availability now and projected future availability in the United States market, while balancing a reasonable amount of modeling and analytical complexity. Since the last analysis we have reduced the number of LD ICE technology options but have refined the options, so they better reflect the diversity of engines in the current fleet. Our technology options also reflect an increase in diversity for hybridization and electrification options, though we utilize these options in a manner that is consistent with statutory constraints. In addition to better representing the current fleet, this reflects consistent feedback from vehicle manufacturers who have told us that they will reduce investment in ICEs while increasing investment in hybrid and plug-in BEV options.

Feedback on the past several CAFE rules has also centered thematically on the expected scope of future electrified vehicle technologies and how we should consider future developments in our analysis. We have received feedback that we cannot consider BEV options and even so, our costs underestimate BEV costs when we do consider them in, for example, the reference baseline. We have also received comments that we should consider more electrified vehicle options and our costs overestimate future costs. Consistent with our interpretation of EPCA/EISA, discussed further in Section III.D and VI, we include several LD electrified technologies to appropriately represent the diversity of current and anticipated future technology options while ensuring our analysis remains consistent with statutory limitations. In addition, this ensures that our analysis can appropriately capture manufacturer decision making about their vehicle fleets for reasons other than CAFE standards ( e.g., other regulatory programs and manufacturing decisions).

The technology options also include our judgment about which technologies will not be available in the rulemaking timeframe. There are several reasons why we may have concluded that it was reasonable to exclude a technology from the options we consider. As with past analyses, we did not include technologies unlikely to be feasible in the rulemaking timeframe, engines technologies designed for markets other than the United States market that are required to use unique gasoline, or technologies where there were not appropriate data available for the range of vehicles that we model in the analysis ( i.e. technologies that are still in the research and development phase but are not ready for mass market production). Each technology section below and Chapter 3 of the TSD discusses these decisions in detail.

The HDPUV technology options also represent a diverse range of both internal combustion and electrified powertrain technologies. We last used the CAFE Model for analyzing HDPUV standards in the Phase 2 Medium and Heavy-Duty Greenhouse Gas and Fuel Efficiency joint rules with EPA in 2016. Since issuing that rule, we refined the ICE technology options based on trends on vehicles in the fleet and updated technology cost and effectiveness data. The HDPUV options also reflect more electrification and hybridization options in that real-world fleet. However, the HDPUV technology options are also less diverse than the LD technology options, for several reasons. The HDPUV fleet is significantly smaller than the LD fleet, with five manufacturers building a little over 25 nameplates in one thousand vehicle model configurations, compared with the 20 LDV manufacturers building more than 250 nameplates in the range of over two thousand configurations. Also, by definition, the HDPUV fleet only includes two vehicle types: HD pickup trucks and work vans. These vehicle types have focused applications, which includes transporting people and moving equipment and supplies. As discussed in more detail below, these vehicles are built with specific technology application, reliability, and durability requirements in order to do work. We believe the range of HDPUV technology options appropriately and reasonably represents the smaller range of technology options available currently and for application in future MYs for the United States market.

Note, however, that for both the LD and HDPUV analyses, the CAFE Model does not dictate or predict the technologies manufacturers must use to comply; rather, the CAFE Model outlines a technology pathway that manufacturers could use to meet the standards cost-effectively. While we estimate the costs and benefits for different levels of CAFE standards estimating technology application that manufacturers could use in the rulemaking timeframe, it is entirely possible and reasonable that a vehicle manufacturer will use different technology options to meet our standards than the CAFE Model estimates and may even use technologies that we do not include in our analysis. This is because our standards do not mandate the application of any particular technology. Rather, our standards are performance-based: manufacturers can and do use a range of compliance solutions that include technology application, shifting sales from one vehicle model or trim level to another, and even paying civil penalties. That said, we are confident that the 75 LD technology options and 30 HDPUV technology options included in the analysis (in particular considering that for each technology option, the analysis includes distinct technology cost and effectiveness values for fourteen different types of vehicles, resulting in about a million different technology effectiveness and cost data points) strike a reasonable balance between the diversity of technology used by an entire industry and simplifying reality in order to make modeling tractable.

Chapter 3 of the TSD and Section III.D below describe the technologies that we used for the LD and HDPUV analyses. Each technology has a name that loosely corresponds to its real-world technology equivalent. We abbreviate the name to a short easy signifier for the CAFE Model to read. We organize those technologies into groups based on technology type: basic and advanced engines, transmissions, electrification, and road load technologies, which include MR, aerodynamic improvement, and low rolling resistance tire technologies.

We then organize the groups into pathways. The pathways instruct the CAFE Model how and in what order to apply technology. In other words, the pathways define technologies that are mutually exclusive ( i.e., that cannot be applied at the same time), and define the direction in which vehicles can advance as the model evaluates which technologies to apply. The respective technology chapters in the TSD and Section 4 of the CAFE Model Documentation for the final rule include a visual of each technology pathway. In general, the paths are tied to ease of implementation of additional technology and how closely related the technologies are.

As an example, our “Turbo Engine Path” consists of five different engine technologies that employ different levels of turbocharging technology. A turbocharger is essentially a small turbine that is driven by exhaust gases produced by the engine. As these gases flow through the turbocharger, they spin the turbine, which in turn spins a compressor that pushes more air into an engine's cylinder. Having more air in the engine's cylinder allows the engine to burn more fuel, which then creates more power, without needing a physically larger engine. In our analysis, an engine that uses a turbocharger “downsizes,” or becomes smaller. The smaller engine can use less fuel to do the same amount of work as the engine did before it used a turbocharger and was downsized. Allowing basic engines to be downsized and turbocharged instead of just turbocharged keeps the vehicle's utility and performance constant so that we can measure the costs and benefits of different levels of fuel economy improvements, rather than the change in different vehicle attributes. This concept is discussed further, below.

Grouping technologies on pathways also tells the model how to evaluate technologies; continuing this example, a vehicle can only have one engine, so if a vehicle has one of the Turbo engines the model will evaluate which more advanced Turbo technology to apply. Or, if it is more cost-effective to go beyond the Turbo pathway, the model will evaluate whether to apply more advanced engine technologies and hybridization path technology.

Then, the arrows between technologies instruct the model on the order in which to evaluate technologies on a pathway. This ensures that a vehicle that uses a more advanced technology cannot downgrade to a less advanced version of the technology, or that a vehicle would switch to technology that was significantly technically different. As an example, if a vehicle in the compliance simulation begins with a TURBOD engine—a turbocharged engine with cylinder deactivation—it cannot adopt a TURBO0 engine. Similarly, this vehicle with a TURBOD engine cannot adopt an ADEACD engine. As an example of our rationale for ordering technologies on the technology tree, an engine could potentially be changed from TURBO0 to TURBO2 without redesigning the engine block or requiring significantly different expertise to design and implement. A change to ADEACD would likely require a different engine block that might not be possible to fit in the engine bay of the vehicle without a complete redesign and different technical expertise requiring years of research and development. This change, which would strand capital and break parts sharing, is why the advanced engine paths restrict most movement between them. The concept of stranded capital is discussed further in Section III.C.6. The model follows instructions pursuant to the direction of arrows between technology groups and between technologies on the same pathway.

We also consider two categories of technology that we could not simulate as part of the CAFE Model's technology pathways. “Off-cycle” and air conditioning (AC) efficiency technologies improve vehicle fuel economy, but the benefit of those technologies cannot be captured using the fuel economy test methods that we must use under EPCA/EISA. As an example, manufacturers can claim a benefit for technology like active seat ventilation and solar reflective surface coatings that make the cabin of a vehicle more comfortable for the occupants, who then do not have to use other less efficient accessories like heat or AC. Instead of including off-cycle and AC efficiency technologies in the technology pathways, we include the improvement as a defined benefit that gets applied to a manufacturer's entire fleet instead of to individual vehicles. The defined benefit that each manufacturer receives in the analysis for using off-cycle and AC efficiency technology on their vehicles is located in the Market Data Input File. See Chapter 3.7 of the TSD for more discussion in how off-cycle and AC efficiency technologies are developed and modeled.

To illustrate, throughout this section we will follow the hypothetical vehicle mentioned above that begins the compliance simulation with a TURBOD engine. Our hypothetical vehicle, Generic Motors' Ravine Runner F Series, is a roomy, top of the line sport utility vehicle (SUV). The Ravine Runner F Series starts the compliance simulation with technologies from most technology pathways; specifically, after looking at Generic Motors' website and marketing materials, we determined that it has technology that loosely fits within the following technologies that we consider in the CAFE Model: it has a turbocharged engine with cylinder deactivation, a fairly advanced 10-speed automatic transmission, a 12V start-stop system, the least advanced tire technology, a fairly aerodynamic vehicle body, and it employs a fairly advanced level of MR. We track the technologies on each vehicle using a “technology key”, which is the string of technology abbreviations for each vehicle. Again, the vehicle technologies and their abbreviations that we consider in this analysis are shown in Table II-1 and Table II-2 above. The technology key for the Ravine Runner F Series is “TURBOD; AT10L2; SS12V; ROLL0; AERO5; MR3.”

2. Defining Manufacturers' Current Technology Positions in the Analysis Fleet

The Market Data Input File is one of four Excel input files that the CAFE Model uses for compliance and effects simulation. The Market Data Input File's “Vehicles” tab (or worksheet) houses one of the most significant compilations of technology inputs and assumptions in the analysis, which is a characterization of an analysis fleet of vehicles to which the CAFE Model adds fuel economy-improving technology. We call this fleet the “analysis fleet.” The analysis fleet includes a number of inputs necessary for the model to add fuel economy-improving technology to each vehicle for the compliance analysis and to calculate the resulting impacts for the effects analysis.

The “Vehicles” tab contains a separate row for each vehicle model. For LD, vehicle models are vehicles that share the same certification fuel economy value and vehicle footprint, and for HDPUVs they are vehicles that share the same certification fuel consumption and WF. This means that vehicle models with different configurations that affect the vehicle's certification fuel economy or fuel consumption value will be distinguished in separate rows in the Vehicles tab. For example, our Ravine Runner example vehicle comes in three different configurations—the Ravine Runner FWD, Ravine Runner AWD, and Ravine Runner F Series—which would result in three separate rows.

In each row we also designate a vehicle's engine, transmission, and platform codes. Vehicles that have the same engine, transmission, or platform code are deemed to “share” that component in the CAFE Model. Parts sharing helps manufacturers achieve economies of scale, deploy capital efficiently, and make the most of shared research and development expenses, while still presenting a wide array of consumer choices to the market. The CAFE Model was developed to treat vehicles, platforms, engines, and transmissions as separate entities, which allows the modeling system to concurrently evaluate technology improvements on multiple vehicles that may share a common component. Sharing also enables realistic propagation, or “inheriting,” of previously applied technologies from an upgraded component down to the vehicle “users” of that component that have not yet realized the benefits of the upgrade. For additional information about the initial state of the fleet and technology evaluation and inheriting within the CAFE Model, please see Section 2.1 and Section 4.4 of the CAFE Model Documentation.

Figure III-5 below shows how we separate the different configurations of the Ravine Runner. We can see by the Platform Codes that these Ravine Runners all share the same platform, but only the Ravine Runner FWD and Ravine Runner AWD share an engine. Even so, all three certification fuel economy values are different, which is common of vehicles that differ in drive type (drive type meaning whether the vehicle has all-wheel drive (AWD), four-wheel drive (4WD), front-wheel drive (FWD), or rear-wheel drive). While it would certainly be easier to aggregate vehicles by model, ensuring that we capture model variants with different fuel economy values improves the accuracy of our analysis and the potential that our estimated costs and benefits from different levels of standards are appropriate. We include information about other vehicle technologies at the farthest right side of the Vehicles tab, and in the “Engines”, “Transmissions”, and “Platforms” worksheets, as discussed further below.

Moving from left to right on the Vehicles tab, after including general information about vehicles and their compliance fuel economy value, we include sales and manufacturer's suggested retail price (MSRP) data, regulatory class information ( i.e., domestic passenger car, import passenger car, light truck, MDPV, HD pickup truck, or HD van), and information about how we classify vehicles for the effectiveness and safety analyses. Each of these data points are important to different parts of the compliance and effects analysis, so that the CAFE Model can accurately average the technologies required across a manufacturer's regulatory classes for each class to meet its CAFE standard, or the impacts of higher fuel economy standards on vehicle sales.

In addition, we include columns indicating if a vehicle is a “ZEV Candidate,” which means that the vehicle could be made into a zero emissions vehicle (ZEV) at its first redesign opportunity in order to simulate a manufacturer's compliance with California's ACC I or ACT program, or manufacturer deployment of electric vehicles on a voluntary basis consistent with ACC II, which is discussed further below.

Next, we include vehicle information necessary for applying different types of technology; for example, designating a vehicle's body style means that we can appropriately apply aerodynamic technology, and designating starting curb weight values means that we can more accurately apply MR technology. Importantly, this section also includes vehicle footprint data (because we set footprint-based standards).

We also set product design cycles, which are the years when the CAFE Model can apply different technologies to vehicles. Manufacturers often introduce fuel saving technologies at a “redesign” of their product or adopt technologies at “refreshes” in between product redesigns. As an example, the redesigned third generation Chevrolet Silverado was released for the 2019 MY, and featured a new platform, updated drivetrain, increased towing capacity, reduced weight, improved safety and expanded trim levels, to name a few improvements. For MY 2022, the Chevrolet Silverado received a refresh (or facelift as it is commonly called), with an updated interior, infotainment, and front-end appearance. Setting these product design cycles ensures that the CAFE Model provides manufacturers with a realistic duration of product stability between refresh and redesign cycles, and during these stability windows we assume no new fuel saving technology introductions for a given model.

During modeling, all improvements from technology application are initially realized on a component and then propagated (or inherited) down to the vehicles that share that component. As such, new component-level technologies are initially evaluated and applied to a platform, engine, or transmission during their respective redesign or refresh years. Any vehicles that share the same redesign and/or refresh schedule as the component apply these technology improvements during the same MY. The rest of the vehicles inherit technologies from the component during their refresh or redesign year (for engine- and transmission-level technologies), or during a redesign year only (for platform-level technologies). Please see Section 4.4 of the CAFE Model Documentation for additional information about technology evaluation and inheriting within the CAFE Model. We did receive comments on the refresh and redesign cycles employed in the CAFE Model, and those are discussed in detail below in Section III.C.6.

The CAFE Model also considers the potential safety effect of MR technologies and crash compatibility of different vehicle types. MR technologies lower the vehicle's curb weight, which may change crash compatibility and safety, depending on the type of vehicle. We assign each vehicle in the Market Data Input File a “safety class” that best aligns with the CAFE Model's analysis of vehicle mass, size, and safety, and include the vehicle's starting curb weight.

The CAFE Model includes procedures to consider the direct labor impacts of manufacturers' response to CAFE regulations, considering the assembly location of vehicles, engines, and transmissions, the percent U.S. content (that reflects percent U.S. and Canada content), and the dealership employment associated with new vehicle sales. Estimated labor information, by vehicle, is included in the Market Data Input File. Sales volumes included in and adapted from the market data also influence total estimated direct labor projected in the analysis. See Chapter 6.2.5 of the TSD for further discussion of the labor utilization analysis.

We then assign the CAFE Model's range of technologies to individual vehicles. This initial linkage of vehicle technologies is how the CAFE Model knows how to advance a vehicle down each technology pathway. Assigning CAFE Model technologies to individual vehicles is dependent on the mix of information we have about any particular vehicle and trends about how a manufacturer has added technology to that vehicle in the past, equations and models that translate real-world technologies to their counterparts in our analysis ( e.g., drag coefficients and body styles can be used to determine a vehicle's AERO level), and our engineering judgment.

As discussed further below, we use information directly from manufacturers to populate some fields in the Market Data Input File, like vehicle horsepower ratings and vehicle weight. We also use manufacturer data as an input to various other models that calculate how a manufacturer's real-world technology equates to a technology level in our model. For example, we calculate initial MR, aerodynamic drag reduction, and ROLL levels by looking at industry-wide trends and calculating—through models or equations—levels of improvement for each technology. The models and algorithms that we use are described further below and in detail in Chapter 3 of the TSD. Other fields, like vehicle refresh and redesign years, are projected forward based on historic trends.

Let us return to the Ravine Runner F Series with the technology key “TURBOD; AT10L2, SS12V; ROLL0; AERO5; MR3.” Generic Motor's publicly available spec sheet for the Ravine Runner F Series says that the Ravine Runner F Series uses Generic Motor's Turbo V6 engine with proprietary Adaptive Cylinder Management Engine (ACME) technology. ACME improves fuel economy and lowers emissions by operating the engine using only three of the engine's cylinders in most conditions and using all six engine cylinders when more power is required. Generic Motors uses this engine in several of their vehicles, and the specifications of the engine can be found in the Engines Tab of the Market Data Input File, under a six-digit engine code.

This is a relatively easy engine to assign based on publicly available specification sheets, but some technologies are more difficult to assign. Manufacturers use different trade names or terms for different technology, and the way that we assign the technology in our analysis may not necessarily line up with how a manufacturer describes the technology. We must use some engineering judgment to determine how discrete technologies in the market best fit the technology options that we consider in our analysis. We discuss factors that we use to assign each vehicle technology in the individual technology subsections below.

In addition to the Vehicles Tab that houses the analysis fleet, the Market Data Input File includes information that affects how the CAFE Model might apply technology to vehicles in the compliance simulation. Specifically, the Market Data Input File's “Manufacturers” tab includes a list of vehicle manufacturers considered in the analysis and several pieces of information about their economic and compliance behavior. First, we determine if a manufacturer “prefers fines,” meaning that historically in the LD fleet, we have observed this manufacturer paying civil penalties for failure to meet CAFE standards. We might designate a manufacturer as not preferring fines if, for example, they have told us that paying civil penalties would be a violation of provisions in their corporate charter. For the NPRM analysis, we assumed that all manufacturers were willing to pay fines in MYs 2022-2026, and that in MY 2027 and beyond, only the manufacturers that had historically paid fines would continue to pay fines. We sought comment on fine payment preference assumptions. Jaguar Land Rover NA commented that they do “not view fine payment as an appropriate compliance route or as a flexibility in the regulation.” In response to JLR's comment, NHTSA has changed their fine preference in the analysis from “prefer fines” to “not prefer fines” for MYs 2027 and beyond. Ford and the Alliance also commented on not using fines for HDPUV compliance. Both commenters agreed with NHTSA's approach of not including fines in the HDPUV analysis. NHTSA maintained the same approach from the NPRM for this final rule and intends to do so in the future.

However, as further discussed below in regard to the CAFE Model's compliance simulation algorithm in Section III.C.6, note that the model will still apply technologies for these manufacturers if it is cost-effective to do so, as defined by several variables.

Next, we designate a “payback period” for each manufacturer. The payback period represents an assumption that consumers are willing to buy vehicles with more fuel economy technology because the fuel economy technology will save them money on gas in the long run. For the past several CAFE Model analyses we have assumed that in the absence of CAFE or other regulatory standards, manufacturers would apply technology that “pays for itself”—by saving the consumer money on fuel—in 2.5 years. While the amount of technology that consumers are willing to pay for is subject to much debate, we continue to assume a 2.5-year payback period based on what manufacturers have told us they do, and on estimates in the available literature. This is discussed in detail in Section III.E below, and in the TSD and FRIA.

We also designate in the Market Data Input File the percentage of each manufacturer's sales that must meet Advanced Clean Car I requirements in certain states, and percentages of sales that manufacturers are expected to produce consistent with levels that would be required under the Advanced Clean Cars II program, if it were to be granted a Clean Air Action preemption waiver. Section 209(a) of the CAA generally preempts states from adopting emission control standards for new motor vehicles; however, Congress created an exemption program in section 209(b) that allows the State of California to seek a waiver of preemption. EPA must grant the waiver unless the Agency makes one of three statutory findings. Under CAA section 177, other States can adopt and enforce standards identical those approved under California's section 209(b) waiver.

Finally, we include estimated CAFE compliance credit banks for each manufacturer in several years through 2021, which is the year before the compliance simulation begins. The CAFE Model does not explicitly simulate credit trading between and among vehicle manufacturers, but we estimate how manufacturers might use compliance credits in early MYs. This reflects manufacturers' tendency to use regulatory credits as an alternative to applying technology.

Before we begin building the Market Data Input File for any analysis, we must consider what MY vehicles will comprise the analysis fleet. There is an inherent time delay in the data we can use for any particular analysis because we must set LD CAFE standards at least 18 months in advance of a MY if the CAFE standards increase, and HDPUV fuel efficiency standards at least 4 full MYs in advance if the standards increase. In addition to the requirement to set standards at least 18 months in advance of a MY, we must propose standards with enough time to allow the public to comment on the proposed standards and meaningfully evaluate that feedback and incorporate it into the final rule in accordance with the APA. This means that the most recent data we have available to generate the analysis fleet necessarily falls behind the MY fleets of vehicles for which we generate standards.

Using recent data for the analysis fleet is more likely to reflect the current vehicle fleet than older data. Recent data will inherently include manufacturer's realized decisions on what fuel economy-improving technology to apply, mix shifts in response to consumer preferences ( e.g., more recent data reflects manufacturer and consumer preference towards larger vehicles), and industry sales volumes that incorporate substantive macroeconomic events ( e.g., the impact of the Coronavirus disease of 2019 (COVID) or microchip shortages). We considered that using an analysis fleet year that has been impacted by these transitory shocks may not represent trends in future years; however, on balance, we believe that updating to using the most complete set of available fleet data provides the most accurate analysis fleet for the CAFE Model to calculate compliance and effects of different levels of future fuel economy standards. Also, using recent data decreases the likelihood that the CAFE Model selects compliance pathways for future standards that affect vehicles already built in previous MYs.

At the time we start building the analysis fleet, data that we receive from vehicle manufacturers in accordance with EPCA/EISA, and our CAFE compliance regulations in advance of or during an ongoing MY, offers the best snapshot of vehicles for sale in the US in a MY. These pre-model year (PMY) and mid-model year (MMY) reports include information about individual vehicles at the vehicle configuration level. We use the vehicle configuration, certification fuel economy, sales, regulatory class, and some additional technology data from these reports as the starting point to build a “row” ( i.e., a vehicle configuration, with all necessary information about the vehicle) in the Market Data Input File's Vehicle's Tab. Additional technology data come from publicly available information, including vehicle specification sheets, manufacturer press releases, owner's manuals, and websites. We also generate some assumptions in the Market Data Input File for data fields where there is limited data, like refresh and redesign cycles for future MYs, and technology levels for certain road load reduction technologies like MR and aerodynamic drag reduction.

For this analysis, the LD analysis fleet consists of every vehicle model in MY 2022 in nearly every configuration that has a different compliance fuel economy value, which results in more than 2,000 individual rows in the Vehicles Tab of the Market Data Input File. The HDPUV fleet consists of vehicles produced in between MYs 2014 and 2022, which results in a little over 1100 individual rows in the HDPUV Market Data Input File. We used a combination of MY data for that fleet because of data availability, but the resulting dataset is a robust amalgamation that provides a reasonable starting point for the much smaller fleet.

Rivian and ZETA commented that some of Rivian's vehicles were mis-classified between the light-duty and HDPUV analysis fleets. NHTSA was aware that some manufacturer's vehicles were erroneously included in the HDPUV fleet rather than the LD fleet. NHTSA stated in the TSD that “for this NPRM, vehicles were divided between light-duty and HDPUV solely on GVWR being above or below 8,500 lbs.” and that “the following will be reassigned to the LD fleet in the final rule: all Rivian vehicles.” Per Rivian's further clarification, NHTSA has reassigned all of Rivian's vehicles in accordance with their comments. NHTSA has also reassigned Ford F150 Lightnings and some Ford Transit Wagons to the LD fleet.

The Ford vehicles moved represent 3,199 total sales out of 1.6 million LD and 319.5 thousand HDPUV sales. The re-classification of Ford's and Rivian's vehicles does not materially affect the analysis results. Ford's vehicles moved represented a very small volume of either fleet, and each regulatory class is regulated based on average performance thus resulting in minor differences of manufacturer's compliance position in each analysis. Moving Rivian's vehicles does not materially affect the analysis results either because they always exceed the regulatory standards, in either fleet. Their vehicles are all electric and outperform the standards every year, regardless of which fleet they find themselves in. Their vehicles will have different technologies available to them in the LD fleet and thus the actual solution will vary. The average costs and pollutant levels of each regulatory class will have changed subtly as a result of moving the vehicles from one fleet to another, but their changes were also affected by the different preferred alternative. The only circumstance in which Rivian's inclusion in one fleet or another could materially sway the outcome is if we modeled credit trading between manufacturers, which is an analysis that EPCA/EISA restricts NHTSA from doing, as discussed further elsewhere in this preamble.

Furthermore, Rivian, ZETA, and Tesla commented about the lack of inclusion of Rivian's Class 2b vans and Tesla's Cybertruck. Rivian stated that in the case of the HDPUV program, “omitting Rivian's Class 2b vans could have material implications for the agency's final” regulation. Rivian also further explained these comments to the agency in a meeting on October 12, 2023. Tesla's Cybertruck is a 2023 or 2024 MY vehicle and the compliance data for that vehicle—which is essential to accurately characterizing the vehicle in the analysis fleet—was not available to the agency at the time of analysis. Rivian's electric delivery van launched in MY 2022 but the compliance data was not available to NHTSA at the time of fleet development.

NHTSA does not believe that the HDPUV analysis would change materially with the inclusion of Rivian's Class 2b vans or Tesla's Cybertruck. Both manufacturers would be able to demonstrate compliance with any stringency in that analysis, and their inclusion would not affect other manufacturers' ability to comply with their standards. This is because, once again, the analysis does not perform any form of credit trading between manufacturers and thus would not have allowed for other manufacturers to comply with higher stringencies. While NHTSA does examine the industry average performance when setting standards, NHTSA also looks at individual manufacturer performance with the standards as well. NHTSA discusses the results of the final HDPUV analysis in Section V. NHTSA will be happy to include all available manufacturers in any future analysis fleets if compliance data is available at the time the fleet is being developed.

The next section discusses how our analysis evaluates how adding additional fuel economy-improving technology to a vehicle in the analysis fleet will improve that vehicle's fuel economy value. Put another way, the next section answers the question, how do we estimate how effective any given technology is at improving a vehicle's fuel economy value?

3. Technology Effectiveness Values

How does the CAFE Model know how effective any particular technology is at improving a vehicle's fuel economy value? Accurate technology effectiveness estimates require information about: (1) the vehicle type and size; (2) the other technologies on the vehicle and/or being added to the vehicle at the same time; and (3) and how the vehicle is driven. Any oversimplification of these complex factors could make the effectiveness estimates less accurate.

To build a database of technology effectiveness estimates that includes these factors, we partner with the DOE's Argonne National Laboratory (Argonne). Argonne has developed and maintains a physics-based full-vehicle modeling and simulation tool called Autonomie that generates technology effectiveness estimates for the CAFE Model.

What is physics-based full-vehicle modeling and simulation? A model is a mathematical representation of a system, and simulation is the behavior of that mathematical representation over time. The Autonomie model is a mathematical representation of an entire vehicle, including its individual technologies such as the engine and transmission, overall vehicle characteristics such as mass and aerodynamic drag, and the environmental conditions, such as ambient temperature and barometric pressure.

We simulate a vehicle model's behavior over the “two-cycle” tests that are used to measure vehicle fuel economy. For readers unfamiliar with this process, measuring a vehicle's fuel economy on the two-cycle tests is like running a car on a treadmill following a program—or more specifically, two programs. The “programs” are the “urban cycle,” or Federal Test Procedure (abbreviated as “FTP”), and the “highway cycle,” or Highway Fuel Economy Test (abbreviated as “HFET”). For the FTP drive cycle the vehicle meets certain speeds at certain times during the test, or in technical terms, the vehicle must follow the designated “speed trace.” The FTP is meant roughly to simulate stop and go city driving, and the HFET is meant roughly to simulate steady flowing highway driving at about 50 miles per hour (mph). We also use the Society of Automotive Engineers (SAE) recommended practices to simulate hybridized and EV drive cycles, which involves the test cycles mentioned above and additional test cycles to measure battery energy consumption and range.

Measuring every vehicle's fuel economy values using the same test cycles ensures that the fuel economy certification results are repeatable for each vehicle model, and comparable across all of the different vehicle models. When performing physical vehicle cycle testing, sophisticated test and measurement equipment calibrated according to strict industry standards further ensures repeatability and comparability of the results. This can include dynamometers, environmental conditions, types and locations of measurement equipment, and precise testing procedures. These physical tests provide the benchmarking empirical data used to develop and verify Autonomie's vehicle control algorithms and simulation results. Autonomie's inputs are discussed in more detail later in this section.

Finally, “physics-based” simply refers to the mathematical equations underlying the modeling and simulation—the simulated vehicle models and all of the sub-models that make up specific vehicle components and the calculated fuel used on simulated test cycles are calculated mathematical equations that conform to the laws of physics.

Full-vehicle modeling and simulation was initially developed to avoid the costs of designing and testing prototype parts for every new type of technology. For example, Generic Motors can use physics-based computer modeling to determine the fuel economy penalty for adding a 4WD, rugged off-road tire trim level of the Ravine Runner to its lineup. The Ravine Runner, modeled with its new drivetrain and off-road tires, can be simulated on a defined test route and under defined test conditions and compared against the initial Ravine Runner simulated without the change. Full-vehicle modeling and simulation allows Generic Motors to consider and evaluate different designs and concepts before building a single prototype for any potential technology change.

Full vehicle modeling and simulation is also essential to measuring how all technologies on a vehicle interact. For example, if technology A improves a particular vehicle's fuel economy by 5% and technology B improves a particular vehicle's fuel economy by 10%, an analysis using single or limited point estimates may erroneously assume that applying both of these technologies together would achieve a simple additive fuel economy improvement of 15%. Single point estimates generally do not provide accurate effectiveness values because they do not capture complex relationships among technologies. Technology effectiveness often differs significantly depending on the vehicle type ( e.g., sedan versus pickup truck) and the way in which the technology interacts with other technologies on the vehicle, as different technologies may provide different incremental levels of fuel economy improvement if implemented alone or in combination with other technologies. As stated above, any oversimplification of these complex factors could lead to less accurate technology effectiveness estimates.

In addition, because manufacturers often add several fuel-saving technologies simultaneously when redesigning a vehicle, it is difficult to isolate the effect of adding any one individual technology to the full vehicle system. Modeling and simulation offer the opportunity to isolate the effects of individual technologies by using a single or small number of initial vehicle configurations and incrementally adding technologies to those configurations. This provides a consistent reference point for the incremental effectiveness estimates for each technology and for combinations of technologies for each vehicle type. Vehicle modeling also reduces the potential for overcounting or undercounting technology effectiveness.

Argonne does not build an individual vehicle model for every single vehicle configuration in our LD and HDPUV Market Data Input Files. This would be nearly impossible, because Autonomie requires very detailed data on hundreds of different vehicle attributes (like the weight of the vehicle's fuel tank, the weight of the vehicle's transmission housing, the weight of the engine, the vehicle's 0-60 mph time, and so on) to build a vehicle model, and for practical reasons we cannot acquire 4000 vehicles and obtain these measurements every time we promulgate a new rule (and we cannot acquire vehicles that have not yet been built). Rather, Argonne builds a discrete number of vehicle models that are representative of large portions of vehicles in the real world. We refer to the vehicle model's type and performance level as the vehicle's “technology class.” By assigning each vehicle in the Market Data Input File a “technology class,” we can connect it to the Autonomie effectiveness estimate that best represents how effective the technology would be on the vehicle, taking into account vehicle characteristics like type and performance metrics. Because each vehicle technology class has unique characteristics, the effectiveness of technologies and combinations of technologies is different for each technology class.

There are ten technology classes for the LD analysis: small car (SmallCar), small performance car (SmallCarPerf), medium car (MedCar), medium performance car (MedCarPerf), small SUV (SmallSUV), small performance SUV (SmallSUVPerf), medium SUV (MedSUV), medium performance SUV (MedSUVPerf), pickup truck (Pickup), and high towing pickup truck (PickupHT). There are four technology classes for the HDPUV analysis, based on the vehicle's “weight class.” An HDPUV that weighs between 8,501 and 10,000 pounds is in “Class 2b,” and an HDPUV that weighs between 10,001 and 14,000 pounds is in “Class 3.” Our four HDPUV technology classes are Pickup2b, Pickup3, Van2b, and Van3.

We use a two-step process that involves two algorithms to give vehicles a “fit score” that determines which vehicles best fit into each technology class. At the first step we determine the vehicle's size, and at the second step we determine the vehicle's performance level. Both algorithms consider several metrics about the individual vehicle and compare that vehicle to other vehicles in the analysis fleet. This process is discussed in detail in TSD Chapter 2.2.

Consider our Ravine Runner F Series, which is a medium-sized performance SUV. The exact same combination of technologies on the Ravine Runner F Series will operate differently in a compact car or pickup truck because they are different vehicle sizes. Our Ravine Runner F Series also achieves slightly better performance metrics than other medium-sized SUVs in the analysis fleet. When we say, “performance metrics,” we mean power, acceleration, handing, braking, and so on, but for the performance fit score algorithm, we consider the vehicle's estimated 0-60 mph time compared to an initial 0-60 mph time for the vehicle's technology class. Accordingly, the “technology class” for the Ravine Runner F Series in our analysis is “MedSUVPerf”.

Table III-1 shows how vehicles in different technology classes that use the exact same fuel economy technology have very different absolute fuel economy values. Note that, as discussed further below, the Autonomie absolute fuel economy values are not used directly in the CAFE Model; we calculate the ratio between two Autonomie absolute fuel economy values (one for each technology key for a specific technology class) and apply that ratio to an analysis fleet vehicle's starting fuel economy value.

Let us also return to the concept of what we call technology synergies. Again, depending on the technology, when two technologies are added to the vehicle together, they may not result in an additive fuel economy improvement. This is an important concept to understand because in Section III.D, below, we present technology effectiveness estimates for every single combination of technology that could be applied to a vehicle. In some cases, technology effectiveness estimates show that a combined technology has a different effectiveness estimate than if the individual technologies were added together individually. However, this is expected and not an error. Continuing our example from above, turbocharging technology and DEAC technology both improve fuel economy by reducing the engine displacement, and accordingly burning less fuel. Turbocharging allows a larger naturally aspirated engine to be reduced in size or displacement while still doing the same amount of work, and its fuel efficiency improvements are, in part, due to the reduced displacement. DEAC effectively makes an engine with a particular displacement intermittently offer some of the fuel economy benefits of a smaller-displacement engine by deactivating cylinders when the work demand does not require the full engine displacement and reactivating them as-needed to meet higher work demands; the greater the displacement of the deactivated cylinders, the greater the fuel economy benefit. Therefore, a manufacturer upgrading to an engine that uses both a turbocharger and DEAC technology, like the TURBOD engine in our example above, would not see the full combined fuel economy improvement from that specific combination of technologies. Table III-2 shows a vehicle's fuel economy value when using the first-level DEAC technology and when using the first-level turbocharging technology, compared to our vehicle that uses both of those technologies combined with a TURBOD engine.

As expected, the percent improvement in Table III-2 between the first and second rows is 1.7% and between the third and fourth rows is 0.3%, even though the only difference within the two sets of technology keys is the DEAC technology (note that we only compare technology keys within the same technology class). This is because there are complex interactions between all fuel economy-improving technologies. We model these individual technologies and groups of technologies to reduce the uncertainty and improve the accuracy of the CAFE Model outputs.

Some technology synergies that we discuss in Section III.D include advanced engine and hybrid powertrain technology synergies. As an example, we do not see a particularly high effectiveness improvement from applying advanced engines to existing parallel strong hybrid ( i.e., P2) architectures. In this instance, the P2 powertrain improves fuel economy, in part, by allowing the engine to spend more time operating at efficient engine speed and load conditions. This reduces the advantage of adding advanced engine technologies, which also improve fuel economy, by broadening the range of speed and load conditions for the engine to operate at high efficiency. This redundancy in fuel savings mechanism results in a lower effectiveness when the technologies are added to each other. Again, we intend and expect that different combinations of technologies will provide different effectiveness improvements on different vehicle types. These examples all illustrate relationships that we can only observe using full vehicle modeling and simulation.

Just as our CAFE Model analysis requires a large set of technology inputs and assumptions, the Autonomie modeling uses a large set of technology inputs and assumptions. Figure III-6 below shows the suite of fuel consumption input data used in the Autonomie modeling to generate the fuel consumption input data we use in the CAFE Model.

What are each of these inputs? For full vehicle benchmarking, vehicles are instrumented with sensors and tested both on the road and on chassis dynamometers ( i.e., the car treadmills used to calculate vehicle's fuel economy values) under different conditions and duty cycles. Some examples of full vehicle benchmark testing we did in conjunction with our partners at Argonne in anticipation of this rule include a 2019 Chevrolet Silverado, a 2021 Toyota Rav4 Prime, a 2022 Hyundai Sonata Hybrid, a 2020 Tesla Model 3, and a 2020 Chevrolet Bolt. We produced a report for each vehicle benchmarked which can be found in the docket. As discussed further below, that full vehicle benchmarking data are used as inputs to the engine modeling and Autonomie full vehicle simulation modeling. Component benchmarking is like full vehicle benchmarking, but instead of testing a full vehicle, we instrument a single production component or prototype component with sensors and test it on a similar duty cycle as a full vehicle. Examples of components we benchmark include engines, transmissions, axles, electric motors, and batteries. Component benchmarking data are used as an input to component modeling, where a production or prototype component is changed in fit, form and/or function and modeled in the same scenario. As an example, we might model a decrease in the size of holes in fuel injectors to see the fuel atomization impact or see how it affects the fuel spray angle.

We use a range of models to do the component modeling for our analysis. As shown in Figure III-6, battery pack modeling using Argonne's BatPaC Model and engine modeling are two of the most significant component models used to generate data for the Autonomie modeling. We discuss BatPaC in detail in Section II.D, but briefly, BatPaC is the battery pack modeling tool we use to estimate the cost of vehicle battery packs based on the materials chemistry, battery design, and manufacturing design of the plants manufacturing the battery packs.

Engine modeling is used to generate engine fuel map models that define the fuel consumption rate for an engine equipped with specific technologies when operating over a variety of engine load and engine speed conditions. Some performance metrics we capture in engine modeling include power, torque, airflow, volumetric efficiency, fuel consumption, turbocharger performance and matching, pumping losses, and more. Each engine map model has been developed ensuring the engine will still operate under real-world constraints using a suite of other models. Some examples of these models that ensure the engine map models capture real-world operating constraints include simulating heat release through a predictive combustion model, knock characteristics through a kinetic fit knock model, and using physics-based heat flow and friction models, among others. We simulate these constraints using data gathered from component benchmarking, and engineering and physics calculations.

The engine map models are developed by creating a base, or root, engine map and then modifying that root map, incrementally, to isolate the effects of the added technologies. The LD engine maps, developed by IAV using their GT-Power modeling tool and the HDPUV engine maps, developed by SwRI using their GT-Power modeling tool, are based on real-world engine designs. One important feature of both the LD and HDPUV engine maps is that they were both developed using a knock model. As noted above, a knock model ensures that any engine size or specification that we model in the analysis does not result in engine knock, which could damage engine components in a real-world vehicle. Although the same engine map models are used for all vehicle technology classes, the effectiveness varies based on the characteristics of each class. For example, as discussed above, a compact car with a turbocharged engine will have a different effectiveness value than a pickup truck with the same engine technology type. The engine map model development and specifications are discussed further in Chapter 3 of the TSD.

Argonne also compiles a database of vehicle attributes and characteristics that are reasonably representative of the vehicles in that technology class to build the vehicle models. Relevant vehicle attributes may include a vehicle's fuel efficiency, emissions, horsepower, 0-60 mph acceleration time, and stopping distance, among others, while vehicle characteristics may include whether the vehicle has all-wheel-drive, 18-inch wheels, summer tires, and so on. Argonne identified representative vehicle attributes and characteristics for both the LD and HDPUV fleets from publicly available information and automotive benchmarking databases such as A2Mac1, Argonne's Downloadable Dynamometer Database (D ), EPA compliance and fuel economy data, EPA's guidance on the cold start penalty on 2-cycle tests, the 21st Century Truck Partnership, and industry partnerships. The resulting vehicle technology class baseline assumptions and characteristics database consists of over 100 different attributes like vehicle height and width and weights for individual vehicle parts.

Argonne then assigns “reference” technologies to each vehicle model. The reference technologies are the technologies on the first step of each CAFE Model technology pathway, and they closely (but do not exactly) correlate to the technology abbreviations that we use in the CAFE Model. As an example, the first Autonomie vehicle model in the “MedSUVPerf” technology class starts out with the least advanced engine, which is “DOHC” (a dual overhead cam engine) in the CAFE Model, or “eng01” in the Autonomie modeling. The vehicle has the least advanced transmission, AT5, the least advanced MR level, MR0, the least advanced aerodynamic body style, AERO0, and the least advanced ROLL level, ROLL0. The first vehicle model is also defined by initial vehicle attributes and characteristics that consist of data from the suite of sources mentioned above. Again, these attributes are meant to reasonably represent the average of vehicle attributes found on vehicles in a certain technology class.

Then, just as a vehicle manufacturer tests its vehicles to ensure they meet specific performance metrics, Autonomie ensures that the built vehicle model meets its performance metrics. We include quantitative performance metrics in our Autonomie modeling to ensure that the vehicle models can meet real-world performance metrics that consumers observe and that are important for vehicle utility and customer satisfaction. The four performance metrics that we use in the Autonomie modeling for light duty vehicles are low-speed acceleration (the time required to accelerate from 0-60 mph), high-speed passing acceleration (the time required to accelerate from 50-80 mph), gradeability (the ability of the vehicle to maintain constant 65 mph speed on a six percent upgrade), and towing capacity for light duty pickup trucks. We have been using these performance metrics for the last several CAFE Model analyses, and vehicle manufacturers have repeatedly agreed that these performance metrics are representative of the metrics considered in the automotive industry. Argonne simulates the vehicle model driving the two-cycle tests ( i.e., running its treadmill “programs”) to ensure that it meets its applicable performance metrics ( e.g., our MedSUVPerf does not have to meet the towing capacity performance metric because it is not a pickup truck). For HDPUVs, Autonomie examines sustainable maximum speed at 6 percent grade, start/launch capability on grade, and maximum sustainable grade at highway cruising speed, before examining towing capability to look for the maximum possible vehicle weight over 40 mph in gradeability. This process ensures that the vehicle can satisfy the gradeability requirement (over 40 mph) with additional payload mass to the curb weight. These metrics are based on commonly used metrics in the automotive industry, including SAE J2807 tow requirements. Additional details about how we size light duty and HDPUV powertrains in Autonomie to meet defined performance metrics can be found in the CAFE Analysis Autonomie Documentation.

If the vehicle model does not initially meet one of the performance metrics, then Autonomie's powertrain sizing algorithm increases the vehicle's engine power. The increase in power is achieved by increasing engine displacement (which is the measure of the volume of all cylinders in an engine), which might involve an increase in the number of engine cylinders, which may lead to an increase in the engine weight. This iterative process then determines if the baseline vehicle with increased engine power and corresponding updated engine weight meets the required performance metrics. The powertrain sizing algorithm stops once all the baseline vehicle's performance requirements are met.

Some technologies require extra steps for performance optimization before the vehicle models are ready for simulation. Specifically, the sizing and optimization process is more complex for the electrified vehicles, which includes hybrid electric vehicle (HEVs) and plug-in hybrid electric vehicles (PHEVs), compared to vehicles with only ICEs, as discussed further in the TSD. As an example, a PHEV powertrain that can travel a certain number of miles on its battery energy alone (referred to as all-electric range (AER), or as performing in electric-only mode) is also sized to ensure that it can meet the performance requirements of the SAE standardized drive cycles mentioned above in electric-only mode.

Every time a vehicle model in Autonomie adopts a new technology, the vehicle weight is updated to reflect the weight of the new technology. For some technologies, the direct weight change is easy to assess. For example, when a vehicle is updated to a higher geared transmission, the weight of the original transmission is replaced with the corresponding transmission weight ( e.g., the weight of a vehicle moving from a 6-speed automatic (AT6) to an 8-speed automatic (AT8) transmission is updated based on the 8-speed transmission weight). For other technologies, like engine technologies, calculating the updated vehicle weight is more complex. As discussed earlier, modeling a change in engine technology involves both the new technology adoption and a change in power (because the reduction in vehicle weight leads to lower engine loads, and a resized engine). When a vehicle adopts new engine technology, the associated weight change to the vehicle is accounted for based on a regression analysis of engine weight versus power.

In addition to using performance metrics that are commonly used by automotive manufacturers, we instruct Autonomie to mimic real-world manufacturer decisions by only resizing engines at specific intervals in the analysis and in specific ways. When a vehicle manufacturer is making decisions about how to change a vehicle model to add fuel economy-improving technology, the manufacturer could entirely “redesign” the vehicle, or the manufacturer could “refresh” the vehicle with relatively more minor technology changes. We discuss how our modeling captures vehicle refreshes and redesigns in more detail below, but the details are easier to understand if we start by discussing some straightforward yet important concepts. First, most changes to a vehicle's engine happen when the vehicle is redesigned and not refreshed, as incorporating a new engine in a vehicle is a 10- to 15-year endeavor at a cost of $750 million to $1 billion. But, manufacturers will use that same basic engine, with only minor changes, across multiple vehicle models. We model engine “inheriting” from one vehicle to another in both the Autonomie modeling and the CAFE Model. During a vehicle “refresh”, one vehicle may inherit an already redesigned engine from another vehicle that shares the same platform. In the Autonomie modeling, when a new vehicle adopts fuel saving technologies that are inherited, the engine is not resized ( i.e., the properties from the reference vehicle are used directly). While this may result in a small change in vehicle performance, manufacturers have repeatedly and consistently told us that the high costs for redesign and the increased manufacturing complexity that would result from resizing engines for small technology changes preclude them from doing so. In addition, when a manufacturer applies MR technology ( i.e., makes the vehicle lighter), the vehicle can use a less powerful engine because there is less weight to move. However, Autonomie will only use a resized engine at certain MR application levels, as a representation of how manufacturers update their engine technologies. Again, this is intended to reflect manufacturer's comments that it would be unreasonable and unaffordable to resize powertrains for every unique combination of technologies. We have determined that our rules about performance neutrality and technology inheritance result in a fleet that is essentially performance neutral.

Why is it important to ensure that the vehicle models in our analysis maintain consistent performance levels? The answer involves how we measure the costs and benefits of different levels of fuel economy standards. In our analysis, we want to capture the costs and benefits of vehicle manufacturers applying fuel economy-improving technologies to their vehicles. For example, say a manufacturer that adds a turbocharger to their engine without downsizing the engine, and then directs all of the additional engine work to additional vehicle horsepower instead of vehicle fuel economy improvements. If we modeled increases or decreases in performance because of fuel economy-improving technology, that increase in performance has a monetized benefit attached to it that is not specifically due to our fuel economy standards. By ensuring that our vehicle modeling remains performance neutral, we can better ensure that we are reasonably capturing the costs and benefits due only to potential changes in the fuel economy standards.

For the NPRM, we analyzed the change in low speed acceleration (0-60 mph) time for four scenarios: (1) MY 2022 under the no action scenario ( i.e., No-Action Alternative), (2) MY 2022 under the Preferred Alternative, (3) MY 2032 under the no action scenario, and (4) MY 2032 under the Preferred Alternative. Using the MY 2022 analysis fleet sales volumes as weights, we calculated the weighted average 0-60 mph acceleration time for the analysis fleet in each of the four above scenarios. We identified that the analysis fleet under no action standards in MY 2032 had a 0.5002 percent worse 0-60 mph acceleration time than under the Preferred Alternative, indicating there is minimal difference in performance between the alternatives. Although we did not conduct the same analysis for the final rule preferred standard, we are confident that the difference in performance time would be insignificant, similar to the NPRM analysis, because the preferred standard falls between the no action and the proposal.

Autonomie then adopts one single fuel saving technology to the initial vehicle model, keeping everything else the same except for that one technology and the attributes associated with it. Once one technology is assigned to the vehicle model and the new vehicle model meets its performance metrics, the vehicle model is used as an input to the full vehicle simulation. This means that Autonomie simulates driving the optimized vehicle models for each technology class on the test cycles we described above. As an example, the Autonomie modeling could start with 14 initial vehicle models (one for each technology class in the LD and HDPUV analysis). Those 14 initial vehicle models use a 5-speed automatic transmission (AT5). Argonne then builds 14 new vehicle models; the only difference between the 14 new vehicle models and the first set of vehicle models is that the new vehicle models have a 6-speed automatic transmission (AT6). Replacing the AT5 with an AT6 would lead either to an increase or decrease in the total weight of the vehicle because each technology class includes different assumptions about transmission weight. Argonne then ensures that the new vehicle models with the 6-speed automatic transmission meet their performance metrics. Now we have 28 different vehicle models that can be simulated on the two-cycle tests. This process is repeated for each technology option and for each technology class. This results in fourteen separate datasets, each with over 100,000 results, that include information about a vehicle model made of specific fuel economy-improving technology and the fuel economy value that the vehicle model achieved driving its simulated test cycles.

We condense the million-or-so datapoints from Autonomie into three datasets used in the CAFE Model. These three datasets include (1) the fuel economy value that each modeled vehicle achieved while driving the test cycles, for every technology combination in every technology class (converted into “fuel consumption”, which is the inverse of fuel economy; fuel economy is mpg and fuel consumption is gallons per mile); (2) the fuel economy value for PHEVs driving those test cycles, when those vehicles drive on gasoline-only in order to comply with statutory constraints; and (3) optimized battery costs for each vehicle that adopts some sort of electrified powertrain (this is discussed in more detail below).

Now, how does this information translate into the technology effectiveness data that we use in the CAFE Model? An important feature of this analysis is that the fuel economy improvement from each technology and combinations of technologies should be accurate and relative to a consistent reference point. We use the absolute fuel economy values from the full vehicle simulations only to determine the relative fuel economy improvement from adding a set of technologies to a vehicle, but not to assign an absolute fuel economy value to any vehicle model or configuration. For this analysis, the absolute fuel economy value for each vehicle in the analysis fleet is based on CAFE compliance data. For subsequent technology changes, we apply the incremental fuel economy improvement values from one or more technologies to the analysis fleet vehicle's fuel economy value to determine the absolute fuel economy achieved for applying the technology change. Accordingly, when the CAFE Model is assessing how to cost-effectively add technology to a vehicle in order to improve the vehicle's fuel economy value, the CAFE Model calculates the difference in the fuel economy value from an Autonomie modeled vehicle with less technology and an Autonomie modeled vehicle with more technology. The relative difference between the two Autonomie modeled vehicles' fuel economy values is applied to the actual fuel economy value of a vehicle in the CAFE Model's analysis fleet.

Let's return to our Ravine Runner F Series, which has a starting fuel economy value of just over 26 mpg and a starting technology key “TURBOD; AT10L2; SS12V; ROLL0; AERO5; MR3.” The equivalent Autonomie vehicle model has a starting fuel economy value of just over 30.8 mpg and is represented by the technology descriptors Midsize_SUV, Perfo, Micro Hybrid, eng38, AUp, 10, MR3, AERO1, ROLL0. In 2028, the CAFE Model determines that Generic Motors needs to redesign the Ravine Runner F Series to reach Generic Motors' new light truck CAFE standard. The Ravine Runner F Series now has lots of new fuel economy-improving technology—it is a parallel strong HEV with a TURBOE engine, an integrated 8-speed automatic transmission, 30% improvement in ROLL, 20% aerodynamic drag reduction, and 10% lighter glider ( i.e., mass reduction). Its new technology key is now P2TRBE, ROLL30, AERO20, MR3. Table III-3 shows how the incremental fuel economy improvement from the Autonomie simulations is applied to the Ravine Runner F Series' starting fuel economy value.

Note that the fuel economy values we obtain from the Autonomie modeling are based on the city and highway test cycles ( i.e., the two-cycle test) described above. This is because we are statutorily required to measure vehicle fuel economy based on the two-cycle test. In 2008, EPA introduced three additional test cycles to bring fuel economy “label” values from two-cycle testing in line with the efficiency values consumers were experiencing in the real world, particularly for hybrids. This is known as 5-cycle testing. Generally, the revised 5-cycle testing values have proven to be a good approximation of what consumers will experience while driving, significantly better than the previous two-cycle test values. Although the compliance modeling uses two-cycle fuel economy values, we use the “on-road” fuel economy values, which are the ratio of 5-cycle to 2-cycle testing values ( i.e., the CAFE compliance values to the “label” values) to calculate the value of fuel savings to the consumer in the effects analysis. This is because the 5-cycle test fuel economy values better represent fuel savings that consumers will experience from real-world driving. For more information about these calculations, please see Section 5.3.2 of the CAFE Model Documentation, and our discussion of the effects analysis later in this section.

In sum, we use Autonomie to generate physics-based full vehicle modeling and simulation technology effectiveness estimates. These estimates ensure that our modeling captures differences in technology effectiveness due to (1) vehicle size and performance relative to other vehicles in the analysis fleet; (2) other technologies on the vehicle and/or being added to the vehicle at the same time; and (3) and how the vehicle is driven. This modeling approach also comports with the NAS 2015 recommendation to use full vehicle modeling supported by the application of lumped improvements at the sub-model level. The approach allows the isolation of technology effects in the analysis supporting an accurate assessment.

In our analysis, “technology effectiveness values” are the relative difference between the fuel economy value for one Autonomie vehicle model driving the two-cycle tests, and a second Autonomie vehicle model that uses new technology driving the two-cycle tests. We add the difference between two Autonomie-generated fuel economy values to a vehicle in the Market Data Input File's CAFE compliance fuel economy value. We then calculate the costs and benefits of different levels of fuel economy standards using the incremental improvement required to bring an analysis fleet vehicle model's fuel economy value to a level that contributes to a manufacturer's fleet meeting its CAFE standard.

In the next section, Technology Costs, we describe the process of generating costs for the Technologies Input File.

4. Technology Costs

We estimate present and future costs for fuel-saving technologies based on a vehicle's technology class and engine size. In the Technologies Input File, there is a separate tab for each technology class that includes unique costs for that class (depending on the technology), and a separate tab for each engine size that also contains unique engine costs for each engine size. These technology cost estimates are based on three main inputs. First, we estimate direct manufacturing costs (DMCs), or the component and labor costs of producing and assembling a vehicle's physical parts and systems. DMCs generally do not include the indirect costs of tools, capital equipment, financing costs, engineering, sales, administrative support or return on investment. We account for these indirect costs via a scalar markup of DMCs, which is termed the RPE. Finally, costs for technologies may change over time as industry streamlines design and manufacturing processes. We estimate potential cost improvements from improvements in the manufacturing process with learning effects (LEs). The retail cost of technology in any future year is estimated to be equal to the product of the DMC, RPE, and LE. Considering the retail cost of equipment, instead of merely DMCs, is important to account for the real-world price effects of a technology, as well as market realities. Each of these technology cost components is described briefly below and in the following individual technology sections, and in detail in Chapters 2 and 3 of the TSD.

DMCs are the component and assembly costs of the physical parts and systems that make up a complete vehicle. We estimate DMCs for individual technologies in several ways. Broadly, we rely in large part on costs estimated by the NHTSA-sponsored 2015 NAS study on the Cost, Effectiveness, and Deployment of Fuel Economy Technologies for LDVs and other NAS studies on fuel economy technologies; BatPaC, a publicly available battery pack modeling software developed and maintained by Argonne, NHTSA-sponsored teardown studies, and our own analysis of how much advanced MR technology ( i.e., carbon fiber) is available for vehicles now and in the future; confidential business information (CBI); and off-cycle and AC efficiency costs from the EPA Proposed Determination TSD. While DMCs for fuel-saving technologies reflect the best estimates available today, technology cost estimates will likely change in the future as technologies are deployed and as production is expanded. For emerging technologies, we use the best information available at the time of the analysis and will continue to update cost assumptions for any future analysis.

Our direct costs include materials, labor, and variable energy costs required to produce and assemble the vehicle; however, direct costs do not include production overhead, corporate overhead, selling costs, or dealer costs, which all contribute to the price consumers ultimately pay for the vehicle. These components of retail prices are illustrated in Table III-4 below.

To estimate total consumer costs ( i.e., both direct and indirect costs), we multiply a technology's DMCs by an indirect cost factor to represent the average price for fuel-saving technologies at retail. The factor that we use is the RPE, and it is the most commonly used to estimate indirect costs of producing a motor vehicle. The RPE markup factor is based on an examination of historical financial data contained in 10-K reports filed by manufacturers with the Securities and Exchange Commission (SEC). It represents the ratio between the retail price of motor vehicles and the direct costs of all activities that manufacturers engage in.

For more than three decades, the retail price of motor vehicles has been, on average, roughly 50 percent above the direct cost expenditures of manufacturers. This ratio has been remarkably consistent, averaging roughly 1.5 with minor variations from year to year over this period. At no point has the RPE markup based on 10-K reports exceeded 1.6 or fallen below 1.4. During this time frame, the average annual increase in real direct costs was 2.5 percent, and the average annual increase in real indirect costs was also 2.5 percent. The RPE averages 1.5 across the lifetime of technologies of all ages, with a lower average in earlier years of a technology's life, and, because of LEs on direct costs, a higher average in later years. Many automotive industry stakeholders have either endorsed the 1.5 markup, or have estimated alternative RPE values. As seen in Table III-5 all estimates range between 1.4 and 2.0, and most are in the 1.4 to 1.7 range.

An RPE of 1.5 does not imply that manufacturers automatically mark up each vehicle by exactly 50 percent. Rather, it means that, over time, the competitive marketplace has resulted in pricing structures that average out to this relationship across the entire industry. Prices for any individual model may be marked up at a higher or lower rate depending on market demand. The consumer who buys a popular vehicle may, in effect, subsidize the installation of a new technology in a less marketable vehicle. But, on average, over time and across the vehicle fleet, the retail price paid by consumers has risen by about $1.50 for each dollar of direct costs incurred by manufacturers. Based on our own evaluation and the widespread use and acceptance of the RPE by automotive industry stakeholders, we have determined that the RPE provides a reasonable indirect cost markup for use in our analysis. A detailed discussion of indirect cost methods and the basis for our use of the RPE to reflect these costs, rather than other indirect cost markup methods, is available in the FRIA for the 2020 final rule.

Finally, manufacturers make improvements to production processes over time, which often result in lower costs. “Cost learning” reflects the effect of experience and volume on the cost of production, which generally results in better utilization of resources, leading to higher and more efficient production. As manufacturers gain experience through production, they refine production techniques, raw material and component sources, and assembly methods to maximize efficiency and reduce production costs.

We estimated cost learning by considering methods established by T.P. Wright and later expanded upon by J.R. Crawford. Wright, examining aircraft production, found that every doubling of cumulative production of airplanes resulted in decreasing labor hours at a fixed percentage. This fixed percentage is commonly referred to as the progress rate or progress ratio, where a lower rate implies faster learning as cumulative production increases. J.R. Crawford expanded upon Wright's learning curve theory to develop a single unit cost model, which estimates the cost of the nth unit produced given the following information is known: (1) cost to produce the first unit; (2) cumulative production of n units; and (3) the progress ratio.

Consistent with Wright's learning curve, most technologies in the CAFE Model use the basic approach by Wright, where we estimate technology cost reductions by applying a fixed percentage to the projected cumulative production of a given fuel economy technology in a given MY. We estimate the cost to produce the first unit of any given technology by identifying the DMC for a technology in a specific MY. As discussed above and in detail below and in Chapter 3 of the TSD, our technology DMCs come from studies, teardown reports, other publicly available data, and feedback from manufacturers and suppliers. Because different studies or cost estimates are based on costs in specific MYs, we identify the “base” MYs for each technology where the learning factor is equal to 1.00. Then, we apply a progress ratio to back-calculate the cost of the first unit produced. The majority of technologies in the CAFE Model use a progress ratio ( i.e., the slope of the learning curve, or the rate at which cost reductions occur with respect to cumulative production) of approximately 0.89, which is derived from average progress ratios researched in studies funded and/or identified by NHTSA and EPA. Many fuel economy technologies that have existed in vehicles for some time will have a gradual sloping learning curve implying that cost reductions from learning is moderate and eventually becomes less steep toward MY2050. Conversely, newer technologies have an initial steep learning curve where cost reduction occurs at a high rate. Mature technologies will generally have a flatter curve and may not incur much cost reduction, if at all, from learning. For an illustration showing various slopes of learning curves, see TSD Chapter 2.4.4.

We assign groups of similar technologies or technologies of similar complexity to each learning curve. While the grouped technologies differ in operating characteristics and design, we chose to group them based on market availability, complexity of technology integration, and production volume of the technologies that can be implemented by manufacturers and suppliers. In general, we consider most base and basic engine and transmission technologies to be mature technologies that will not experience any additional improvements in design or manufacturing. Other basic engine technologies, like VVL, SGDI, and DEAC, do decrease in costs through around MY 2036, because those were introduced into the market more recently. All advanced engine technologies follow the same general pattern of a gradual reduction in costs until MY 2036, when they plateau and remain flat. We expect the cost to decrease as production volumes increase, manufacturing processes are improved, and economies of scale are achieved. We also assigned advanced engine technologies that are based on a singular preceding technology to the same learning curve as that preceding technology. Similarly, the more advanced transmission technologies experience a gradual reduction in costs through MY 2031, when they plateau and remain flat. Lastly, we estimate that the learning curves for road load technologies, with the exception of the most advanced MR level (which decreases at a fairly steep rate through MY 2040, as discussed further below and in Chapter 3.4 of the TSD), will decrease through MY 2036 and then remain flat.

We use the same cost learning rates for both LD and HDPUV technologies. This approach was used in the HDPUV analysis in the Phase 2 HD joint rule with EPA, and we believe that this is an appropriate assumption to continue to use for this analysis. While the powertrains in HDPUVs do have a higher power output than LD powertrains, the designs and technology used will be very similar. Although most HDPUV components will have higher operating loads and provide different effectiveness values than LD components, the overall designs are similar between the technologies. The individual technology design and effectiveness differences between LD and HDPUV technologies are discussed below and in Chapter 3 of the TSD.

For technologies that have been in production for many years, like some engine and transmission technologies, this approach produces reasonable estimates that we can compare against other studies and publicly available data. Generating the learning curve for battery packs for BEVs in future MYs is significantly more complicated, and we discuss how we generated those learning curves in Section III.D and in detail in Chapter 3.3 of the TSD. Our battery pack learning curves recognize that there are many factors that could potentially lower battery pack costs over time outside of the cost reductions due to improvements in manufacturing processes due to knowledge gained through experience in production.

Table III-6 shows how some of the technologies on the MY 2022 Ravine Runner Type F decrease in cost over several years. Note that these costs are specifically applicable to the MedSUVPerf class, and other technology classes may have different costs for the same technologies. These costs are pulled directly from the Technology Costs Input File, meaning that they include the DMC, RPE, and learning.

5. Simulating Existing Incentives, Other Government Programs, and Manufacturer ZEV Deployment Plans

Similar to the regulations that we are enacting, other government actions have the ability to influence the technology manufacturers apply to their vehicles. For the purposes of this analysis, we incorporate manufacturers' expected response to two other government actions into our analysis: state ZEV requirements and Federal tax credits. We also include ZEV deployment that manufacturers have committed to execute even though it goes beyond any government's legal requirements.

a. Simulating ZEV Deployment Unrelated to NHTSA's Standards

The California Air Resources Board (CARB) has developed various programs to control emissions of criteria pollutants and GHGs from vehicles sold in California. CARB does so in accordance with the federal CAA; CAA section 209(a) generally preempts states from adopting emission control standards for new motor vehicles; however, Congress created an exemption program in CAA section 209(b) that allows the State of California to seek a waiver of preemption related to adopting or enforcing motor vehicle emissions standards. EPA must grant the waiver unless the Agency makes one of three statutory findings. Under CAA section 177, other States can adopt and enforce standards identical to those approved under California's Section 209(b) waiver and other specified criteria in section 177 are met. States that do so are sometimes referred to as section 177 states, in reference to section 177 of the CAA. Since 1990, CARB has included a version of a Zero-Emission Vehicle (ZEV) program as part of its package of standards that control smog-causing pollutants and GHG emissions from passenger vehicles sold in California, and several states have adopted those ZEV program requirements. This section focuses on the way we modeled manufacturers' expected compliance with these ZEV program requirements as well as additional electric vehicle deployment that manufacturers have indicated they will undertake. See Section IV.B.1 for a discussion of the role of these electric vehicles in the reference baseline and associated comments and responses.

There are currently two operative ZEV regulations that we consider in our analysis: ACC I (LD ZEV requirements through MY 2025) and Advanced Clean Trucks (ACT) (requirements for trucks in Classes 2b through 8, from MYs 2024-2035). California has adopted a third ZEV regulation, ACC II (LD ZEV requirements for MYs 2026-2035). EPA is evaluating a petition for a waiver of Clean Air Act preemption for ACC II, but has not granted it. While ACC II is currently unenforceable while the waiver request is under consideration by EPA—in contrast to ACC I and ACT, which have already received waiver approvals—manufacturers have indicated that they intend to deploy additional electric vehicles consistent with (or beyond) what ACC II would require for compliance if a waiver were to be granted. We have therefore modeled compliance with ACC II as a proxy for these additional electric vehicles that manufacturers have committed to deploying in the reference baseline or No-Action Alternative. As discussed further below, we also developed a sensitivity case and an alternative baseline that included, respectively, some or none of the electric vehicles that would be expected to enter the fleet under ACC I, ACT, and manufacturer deployment commitments consistent with ACC II in order to ensure that our standards satisfy the statutory factors regardless of which baseline turns out to be the most accurate.

In the NPRM, we stated that we are confident that manufacturers will comply with the ZEV programs because they have previously complied with state ZEV programs, and they have made announcements of new ZEVs demonstrating an intent to comply with the requirements going forward. The American Fuel & Petrochemical Manufacturers (AFPM) objected to the use of the word “confident” given their concerns about manufacturers' ability to comply with ZEV standards. Valero and Kia commented that CARB historically has eased compliance for manufacturers by allowing for compliance via changing compliance dates, stringencies, and ZEV definitions. Valero also commented that our inclusion of ACT was premature given its 2024 start date and stated their doubts about its technological feasibility.

We focus on including the provisions that CARB and other states currently have in place in their regulations and that have received a Clean Air Act preemption waiver from EPA, and we have taken this into account by having incorporated changing standards and compliance landscapes in our past and current rulemakings. Valero further cited risks of ZEV programs such as varying compliance challenges across OEMs, consumer preferences, and affordability concerns, as well as general uncertainty in predicting future ZEV sales. NHTSA observes that companies have historically complied with California waivers and notes that even though industry entities such as Valero have previously made such comments about ZEV programs, historically, manufacturers have complied. Further, NHTSA notes that manufacturers have indicated that they intend to deploy electric vehicles consistent with the requirements of not just ACC I and ACT, but also ACC II. In this analysis, NHTSA has not assumed that the ACC II waiver will be granted. However, in the reference baseline, NHTSA has included electric vehicle deployment consistent with stated manufacturer plans to deploy such vehicles—and that level would result in full compliance with the ACC II program. Furthermore, many of the ZEVs that can earn credits from CARB are already present in the 2022 analysis fleet, leading the modeled MY 2022 analysis fleet to achieve 100% compliance with that years' ACC I requirement in MY 2022 (per CARB, the total ending year credit balances significantly exceed the annual credit requirements). NHTSA models manufacturers' compliance with ACC I and ACT and the additional electric vehicle deployment that manufacturers have announced they intend to execute because accounting for technology improvements that manufacturers would make even in the absence of CAFE standards allows NHTSA to gain a more accurate understanding of the effects of the final rule. Importantly, as noted above, NHTSA also developed an alternative baseline, the No ZEV alternative baseline, to test whether the standards remain consistent with the statutory factors regardless of the level of electrification that occurs in the reference baseline. NHTSA also modeled the HDPUV program assuming the ACT program was not included in the reference baseline, even though EPCA/EISA contains no limitations on the consideration of alternative fueled vehicles in that program.

The Zero Emission Transportation Association commented that NHTSA should include CARB's Advanced Clean Fleets (ACF) regulation as part of its modeling. We do not include the Advanced Clean Fleets regulation in our modeling at this time, due to the small number of HDPUV Class 2b/3 vehicles that would be affected by this regulation in the rulemaking time frame, and due to the analytical complexity of modeling this small amount of vehicles. We will continue to monitor this program to determine whether it should be featured in future analyses.

This is the fourth analysis where we have modeled compliance with the ACC program (and now the ACT program) requirements in the CAFE Model. In the MY 2024-2026 final rule, we received feedback from commenters agreeing or disagreeing with the modeling inclusion of the ZEV programs at all, however, the only past substantive comments on the ZEV program modeling methodology have been requesting the inclusion of more states that signed on to adopt California's standards in our analysis. As noted below, the inclusion or exclusion of states in the analysis depends on which states have signed on to the programs at the time of our analysis. While we are aware of legal challenges to some states' adoption of the ZEV programs, it is beyond the scope of this rulemaking to evaluate the likelihood of success of those challenges. For purposes of our analysis, what is important is predicting, using a reasonable assessment, how the fleet will evolve in the future. The following discussion provides updates to our modeling methodology for the ZEV programs in the analysis.

The ACC I and ACT programs require that increasing levels of manufacturers' sales in California and section 177 states in each MY be ZEVs, specifically BEVs, PHEVs, FCEVs. BEVs, PHEVs, and FCEVs each contribute a “value” towards a manufacturer's annual ZEV requirement, which is a product of the manufacturer's production volume sold in a ZEV state, multiplied by a “percentage requirement.” The percentage requirements increase in each year so that a greater portion of a manufacturer's fleet sold in ZEV states in a particular MY must be ZEVs. For example, a manufacturer selling 100,000 vehicles in California and 10,000 vehicles in Connecticut (both states that have ZEV programs) in MY 2025 must ensure that 22,000 ZEV credits are earned by California vehicles and 2,200 ZEV credits are earned by Connecticut vehicles. In MYs 2026 through 2030 of the ACC II program (if granted a waiver) would allow manufacturers to apply a capped amount of credits to the percentage requirement. In response to various commenters mentioning the pooled credits route, we added this option to our modeling, slightly scaling down the percent requirement assumed to be met by ZEV sales; this corresponds to the maximum pooled credits that would be allowed by CARB under ACC II, if granted a waiver.

At the time of our analysis, seventeen states in addition to California have either formally signed on to the ACC I or ACC II standards or are in the process of adopting them. Although a few states are adopting these requirements in future MYs, for the ease of modeling we include in the unified ACC II group every state that has regulations in place to adopt or is already in the process of adopting the requirements by the time of our analysis at the start of December 2023. A variety of commenters expressed concern with our NPRM approach of considering all the states as a group that adopted the programs in all the model years that CARB outlined. Hyundai noted in their comments that Nevada, Minnesota, and Virginia are “unlikely to adopt ACC II.” Commenters such as the AFPM and Nissan stated that several states have adopted only some model years of ACC II. NHTSA notes that its analysis does not assume legal enforcement of ACC II because it has not been granted a preemption waiver, but that manufacturers have nonetheless indicated they intend to deploy electric vehicles during these model years at levels that would be consistent with ACC II in both California and other states. However, to be appropriately conservative, NHTSA has updated its approach to reflect the variety in model years to which states have committed and in response to comments, we now include different state sales share groups in our modeling. Splitting these groups based on model years in which they have indicated their participation also allows us to distinguish between assumed future ACC I compliance and the deployment that manufacturers have indicated they are intending to execute that would be consistent with ACC II. The seventeen states included in our light-duty ZEV analysis have adopted ACC I and/or ACC II in at least one model year.

Some commenters such as the Center for Environmental Accountability and Nissan stated that many of the states included in our ZEV modeling had not actually adopted the ZEV programs. NHTSA disagrees; we include all states that have regulations in place to adopt or are already in the process of adopting ACC I, ACC II, or ACT, based on information available at the time of the analysis. Our final ZEV state assumptions are also consistent with those tracked by CARB on their website at the time of writing. This included adding states to our analysis that were not present in the NPRM ZEV modeling. Commenters such as ACEEE and the American Lung Association requested that we make these updates to the ZEV states list. We added the state of Colorado into our analysis, based on new information and their comment indicating their commitment to all three ZEV programs. Similarly, eleven states including California have formally adopted the ACT standards at the time of analysis. As this group is smaller and has somewhat less variety in start dates than the ACC I/ACC II states, we model ACT state shares without breaking out specific model year start dates.

It is also important to note in the context of all the above comments on ZEV adoption that NHTSA developed an alternative baseline, the No ZEV alternative baseline, in order to evaluate whether the standards are consistent with the statutory factors regardless of the amount of electrification that occurs in the absence of NHTSA's standards during the standard setting years. NHTSA further evaluated sensitivity cases, that one could certainly consider as additional alternative baselines, that precluded electric vehicles from being added to the fleet between Model Years 2027-2035; between 2027-2050; and 2022-2050.

It is important to note that not all section 177 states have adopted the ACC II or ACT program components. Furthermore, more states have formally adopted the ACC II program than the ACT program, so the discussion in the following sections will call states that have opted in “ACC I/ACC II states” or “ACT states.” Separately, many states signed a memorandum of understanding (MOU) in 2020 to indicate their intent to work collaboratively towards a goal of turning 100% of MD and HD vehicles into ZEVs in the future. For the purposes of CAFE analysis, we include only those states that have formally adopted the ACT in our modeling as “ACT states.” States that have signed the MOU but not formally adopted the ACT program are referred to as “MOU states” and are not included in CAFE modeling. When the term “ZEV programs” is used hereafter, it refers to both the ACC II and ACT programs.

Incorporating ACC I and ACT as applicable legal requirements and ACC II as a proxy for additional electric vehicle deployment expected to occur regardless of the NHTSA standards into the model includes converting vehicles that have been identified as potential ZEV candidates into BEVs at the vehicle's ZEV application year so that a manufacturer's fleet meets its required ZEV credit requirements. We focused on BEVs as ZEV conversions, rather than PHEVs or FCEVs, because, as for 2026-2035, manufacturers cannot earn more than 20% of their ZEV credits through PHEV sales. Similarly, PHEVs receive a smaller number of credits than BEVs and FCEVs under ACC I, and those with lower all-electric range values would receive a smaller number of credits under ACC II if it became legally enforceable. We determined that including PHEVs in the ZEV modeling would have introduced unnecessary complication to the modeling and would have provided manufacturers little benefit in the modeled program. In addition, although FCEVs can earn the same number of credits as BEVs, we chose to focus on BEV technology pathways since FCEVs are generally less cost-effective than BEVs and most manufacturers have not been producing them at high volumes. However, any PHEVs and FCEVs already present in the CAFE Model analysis fleets receive ZEV credits in our modeling.

Total credits are calculated by multiplying the credit value each ZEV receives by the vehicle's volume. In the ACC I program, until 2025, each full ZEV can earn up to 4 credits. In the ACC II program, from 2026 onwards, each full ZEV would earn one credit value per vehicle, while partial ZEVs (PHEVs) would earn credits based on their AER, if ACC II became legally enforceable. In the context of this section, “full ZEVs” refers to BEVs and FCEVs, as PHEVs can receive a smaller number of credits than other ZEVs, as discussed above. Based on comments from CARB and the Strong PHEV Coalition, we adjusted the number of ZEV credits received by PHEV50s in our analysis to 1 full credit under the ACC II proxy after determining with Argonne that the range of all the PHEVs marked as “PHEV50s” in our analysis fleet was sufficient to receive the full ZEV credit. Credit targets in the ACT program (referred to as deficits) are calculated by multiplying sales by percentage requirement and weight class multiplier. Each HDPUV full ZEV in the 2b/3 class earns 0.8 credits and each near-zero emissions vehicle (called PHEVs in the CAFE Model) earns 0.75 credits. We adjusted some of the explanations in this section and the TSD accompanying this rule in response to a comment from CARB requesting that we very clearly distinguish between the number of credits earned between different vehicle types and programs.

The CAFE Model is designed to present outcomes at a national scale, so the ZEV programs analysis considers the states as a group as opposed to estimating each state's ZEV credit requirements individually. However, in response to comments discussed above, we adjusted our ZEV modeling to reflect states' varying commitments to the ACC I and ACC II programs in different model years. To capture the appropriate volumes subject to the ACT requirements and that would be deployed consistent with ACC II, we still calculated each manufacturer's total market share in ACC II or ACT states but also expanded the market share inputs to vary across model year according to how many states had opted into the program in each year between 2022 and 2035. We used Polk's National Vehicle Population Profile (NVPP) from January 2022 to calculate these percentages. These data include vehicle characteristics such as powertrain, fuel type, manufacturer, nameplate, and trim level, as well as the state in which each vehicle is sold. At the time of the data snapshot, MY 2021 data from the NVPP contained the most current estimate of new vehicle market shares for most manufacturers, and best represented the registered vehicle population on January 1, 2022. We assumed that this source of new registrations data was the best approximation of new sales given the data options. For MY 2021 vehicles in the latest NVPP, the ACC II State group at its largest makes up approximately 38% of the total LD sales in the United States. The ACT state groups comprise approximately 22% of the new Class 2b and 3 (HDPUV) vehicle market in the U.S. We based the volumes used for the ZEV credit target calculation on each manufacturer's future assumed market share in ACC II and ACT states. We made this assumption after examining three past years of market share data and determining that the geographic distribution of manufacturers' market shares remained fairly constant.

We calculated total credits required for ACT compliance and consistent with ACC II implementation by multiplying the percentages from each program's ZEV requirement schedule by the ACC II or ACT state volumes. For the first set of ACC I requirements covering 2022 (the first modeled year in our analysis) through 2025, the percentage requirements start at 14.5% and ramp up in increments to 22 percent by 2025. For ACC II, the potential percentage requirements start at 35% in MY 2026 and would ramp up to 100% in MY 2035 and subsequent years if it became legally enforceable. For ACT Class 2b-3 Group vehicles (equivalent to HDPUVs in our analysis), the percentage requirements start at 5% in MY 2024 and increase to 55% in MYs 2035 and beyond. We then multiply the resulting national sales volume predictions by manufacturer by each manufacturer's total market share in the ACC II or ACT states to capture the appropriate volumes in the ZEV credits calculation. Credits consistent with ACC II by manufacturer, per year, are determined within the CAFE Model by multiplying the ACC II state volumes by CARB's ZEV credit percentage requirement for each program respectively. In the first five years of the ACC II program (as currently submitted to EPA), MYs 2026-2030, CARB would allow for a pooled credits allowance, capped at a specific percentage per year (which decreases in later years). We accounted for this in the final rule in response to comments by reducing the percent requirement in those years by the maximum pooled credit allowance.

To ensure that the ACT credit requirements are met in the reference baseline and deployment consistent with ACC II is reflected in the reference baseline in each modeling scenario, we add ZEV candidate vehicles to the reference baseline. We flag ZEV candidates in the `vehicles' worksheet in the Market Data Input File, which is described above and in detail in TSD Chapter 2.5. Although we identify the ZEV candidates in the Market Data Input File, the actual conversion from non-ZEV to ZEV vehicles occurs within the CAFE Model. The CAFE Model converts a vehicle to a ZEV during the specified ZEV application year.

We flag ZEV candidates in two ways: using reference vehicles with ICE powertrains or using PHEVs already in the existing fleet. When using ICE powertrains as reference vehicles, we create a duplicate row (which we refer to as the ZEV candidate row) in the Market Data Input File's Vehicles tab for the ZEV version of the original vehicle, designated with a unique vehicle code. The ZEV candidate row specifies the relevant electrification technology level of the ZEV candidate vehicle ( e.g., BEV1, BEV2, and so on), the year that the electrification technology is applied, and zeroes out the candidate vehicle's sales volume. We identify all ICE vehicles with varying levels of technology up to and including strong hybrid electric vehicles (SHEVs) with rows that have 100 sales or more as ZEV candidates. The CAFE Model moves the sales volume from the reference vehicle row to the ZEV candidate row on an as-needed basis, considering the MY's ZEV credit requirements. When using existing PHEVs within the fleet as a starting point for identifying ZEV candidates, we base our determination of ZEV application years for each model based on expectations of manufacturers' future EV offerings. The entire sales volume for that PHEV model row is converted to BEV on the application year. This approach allows for only the needed additional sales volumes to flip to ZEVs, based on the ACC II and ACT targets, and keeps us from overestimating ZEVs in future years. The West Virginia Attorney General's Office commented that “NHTSA programmed the CAFE model to assume that manufacturers will turn every internal combustion engine vehicle into a ZEV at the `first redesign opportunity.' ” This comment is a misunderstanding of the ZEV candidate modeling, where the model will shift only the necessary volumes to comply with the ZEV programs into ZEVs. As we stated in the NPRM and repeated above, this approach allows for only the needed additional sales volumes to flip to ZEVs, based on the ACC II and ACT targets, and keeps us from overestimating ZEVs in future years. See TSD Chapter 2.5 for more details on our ZEV program modeling.

We identify LD ZEV candidates by duplicating every row with 100 or more sales that is not a PHEV, BEV, or FCEV. We refer to the original rows as `reference vehicles.' Although PHEVs are all ZEV candidates, we do not duplicate those rows as we focus the CAFE Model's simulation of the ACC II and ACT programs on BEVs. However, any PHEVs already in the analysis fleet or made by the model will still receive the appropriate ZEV credits. While flagging the ZEV candidates, we identified each one as a BEV1, BEV2, BEV3, and BEV4 (BEV technology types based on range), based partly on their price, market segment, and vehicle features. For instance, we assumed luxury cars would have longer ranges than economy cars. We also assigned AWD/4WD variants of vehicles shorter BEV ranges when appropriate. See TSD Chapter 3.3 for more detailed information on electrification options for this analysis. The CAFE Model assigns credit values per vehicle depending on whether the vehicle is a ZEV in a MY prior to 2026 or after, due to the change in value after the update of the standards from ACC II (as currently submitted to EPA).

We follow a similar process in assigning HDPUV ZEV candidates as in assigning LD ZEV candidates. We duplicate every van row with 100 or more sales and duplicate every pickup truck row with 100 or more sales provided the vehicle model has a WF less than 7,500 and a diesel- or gasoline-based range lower than 500 miles based on their rated fuel efficiency and fuel tank size. This is consistent with our treatment of HDPUV technology applicability rules, which are discussed below in Section III.D and in TSD Chapter 3.3. Note that the model can still apply PHEV technology to HDPUVs because of CAFE standards, and like the LD analysis, any HDPUVs turned into PHEVs will receive credit in the ZEV program. When identifying ZEV candidates, we assign each candidate as either a BEV1 or a BEV2 based on their price, market segment, and other vehicle attributes.

The CAFE Model brings manufacturers into compliance with ACC II (as currently submitted to EPA) and ACT first in the reference baseline, solving for the technology compliance pathway used to meet increasing ZEV standards. Valero commented on the BEV sales shift in the HDPUV analysis being too large for ACT compliance purposes. Our ZEV modeling structure is designed to only convert ZEV candidates if needed for the ACT program requirements. However, the CAFE Model also incorporates many other factors into its technology and CAFE compliance pathways decisions, technology payback, including technology costs and sizing requirements based on vehicle performance. See the TSD Chapter 3.3 and Preamble Section III.D for further discussion of electrification pathways and sales volume results.

In the proposal, we did not include two provisions of the ZEV regulations in our modeling. First, while the ACC II program (as currently submitted to EPA) includes compliance options for providing reduced-price ZEVs to community mobility programs and for selling used ZEVs (known as “environmental justice vehicle values”), these are focused on a more local level than we could reasonably represent in the CAFE Model. The data for this part of the program are also not available from real world application. Second, under ACC II (as currently submitted to EPA), CARB would allow for some banking of ZEV credits and credit pooling. In the proposal, we did not assume compliance with ZEV requirements through banking of credits when simulating the program in the CAFE Model and focused instead on simulating manufacturer's deployment of ZEV consistent with ACC II fully through the production of new ZEVs, after conversations with CARB. In past rules, we assumed 80% compliance through vehicle requirements and the remaining 20% with banked credits. In this rule, due to the complicated nature of accounting for the entire credit program, we focus only on incorporating CARB's allowance (as outlined in the ACC II program currently submitted to EPA) for manufacturers to use pooled credits in MYs 2026-2030 as part of their ZEV compliance in our modeling. Based on guidance from CARB in the NPRM and assessment of CARB's responses to manufacturer comments, we expect impacts of banked credit provisions on overall volumes to be small.

TSD Chapter 2.5.1 includes more information about the process we use to simulate ACT program compliance and ZEV deployment consistent with ACC II in this analysis.

b. IRA Tax Credits

The IRA included several new and expanded tax credits intended to encourage the adoption of clean vehicles. At the proposal stage, the agency was presented with three questions on how to incorporate the IRA. First, identifying which credits should be modeled. Next, determining the responses of consumers and producers to the subsidies. And finally determining which vehicles would qualify and how to value the credits. In its proposal, NHTSA modeled two provisions of the IRA. The first was the Advanced manufacturing production tax credit (AMPC). This provision provides a $35 per kWh tax credit for manufacturers of battery cells and an additional $10 per kWh for manufacturers of battery modules (all applicable to manufacture in the United States). The second provision modeled in the proposal was the Clean vehicle credit (§ 30D), which provides up to $7,500 toward the purchase of clean vehicles with critical minerals extracted or processed in the United States or a country with which the United States has a free trade agreement or recycled in North America, and battery components manufactured or assembled in North America.¹⁷³

After NHTSA developed its methodology for incorporating the IRA tax credits into its analysis for the proposal, the Treasury Department clarified that leased vehicles qualify for the Credit for qualified commercial clean vehicles (§ 45W) and that the credit could be calculated based off of the DOE's Incremental Purchase Cost Methodology and Results for Clean Vehicles report for at least calendar year 2023 as a safe harbor, rather than having the taxpayer estimate the actual cost differential. As a result, EPA modified their approach to modeling the IRA tax credits prior to finalizing their Multi-Pollutant Emissions Standards for Model Years 2027 and Later Light-Duty and Medium-Duty Vehicles proposal, however NHTSA was unable to incorporate a similar methodology in time for its proposal.

NHTSA noted in the proposal that there are several other provisions of the IRA related to clean vehicles that were excluded from the analysis, including the Previously-owned Clean Vehicle credit, the Qualifying Advanced Energy Project credit (48C), IRA § 50142 Advanced Technology Vehicle Manufacturing Loan Program, IRA § 50143 Domestic Manufacturing Conversion Grants, IRA § 70002 USPS Clean Fleets, and IRA § 13404 Alternative Fuel Vehicle Refueling Property Credit. As NHTSA noted in the proposal, these credits and grants incentivize clean vehicles through avenues the CAFE Model is currently unable to consider as they typically affect a smaller subset of the vehicle market and may influence purchasing decisions through means other than price, e.g., through expanded charging networks. NHTSA also does not model individual state tax credits or rebate programs. Unlike ZEV requirements which are uniform across states that adopt them, state clean vehicle tax credits and rebates vary from jurisdiction to jurisdiction and are subject to more uncertainty than their Federal counterparts. Tracking sales by jurisdiction and modeling each program's individual compliance program would require significant revisions to the CAFE Model and likely provide minimal changes in the net outputs of the analysis.

NHTSA sought comment from the public about which credits should be included in its analysis, and in particular whether the agency should include § 45W. Rivian and the American Council for an Energy Efficient Economy (ACEEE) both suggested that NHTSA also include § 45W in its analysis, to avoid underestimating the impact of the IRA on reference baseline technology adoption. NHTSA did not receive any comments recommending either removing the AMPC or § 30D from its analysis, or advocating for other credits, Federal or State, to be included.

For the Final Rule, NHTSA models three of the IRA provisions in its analysis. NHTSA is again modeling the AMPC and, based on the recommendations of commenters and guidance from the Treasury Department indicating that § 45W applies to leased personal vehicles, NHTSA decided to jointly model § 30D and § 45W (collectively, the Clean Vehicle Credits or “CVCs”). Both credits are available at the time of sale and provide up to $7,500 towards the purchase of light-duty and HDPUV PHEVs, BEVs, and FCEVs placed in service before the end of 2032. § 30D is only available to purchasers of vehicles assembled in North America and which meet certain sourcing requirements for critical minerals and battery components manufactured in North America. § 45W is available for commercial purchasers of vehicles covered by this rule for a purpose other than resale. The credit value is the lesser of the incremental cost to purchase a comparable ICE vehicle or 15 percent of the cost basis for PHEVs or 30 percent of the cost basis for FCEVs and BEVs, up to $7,500 for vehicles with GVWR less than 14,000. Since only one of the CVCs may be claimed for purchasing a given vehicle, NHTSA modeled them jointly, employing a methodology similar to EPA's approach.

Interactions between producers and consumers in the marketplace tend to ensure that subsidies like the AMPC and the CVCs, regardless of whether they are initially paid to producers or consumers, are ultimately shared between the two groups. In the proposal, NHTSA assumed that manufacturers and consumers would each capture half the dollar value of each credit. NHTSA sought comment on its modeling assumptions related to how it modeled tax credits in the proposal. The Institute for Policy Integrity (IPI) suggested that NHTSA's assumptions about the incidence of tax credits were not compatible with its assumptions about the pass-through of changes in technology costs to consumers. AFPM commented that IRA tax credits may be eliminated or modified, and that manufacturers may not pass the cost savings from the AMPC through to consumers. NHTSA acknowledged uncertainty over its pass-through assumptions in its proposal and ran sensitivity cases which varied the degree to which these incentives are shared between consumers and manufacturers. NHTSA believes that changing the production quantities of these vehicles is a complex process that involves developing new supply chains and significant changes in production processes. As a result, NHTSA believes that manufacturers are likely to experience some motivation to recover these costs by attempting to capture some portion of IRA credits, for example, by raising prices of qualifying vehicles in response to availability of the 30D credit. On the other hand, NHTSA does not believe it is likely that manufacturers will be able to raise prices for these vehicles enough to fully capture the amount of credit in this way. NHTSA believes that the tax credits are likely to be a salient factor in the purchase decisions of consumers who purchase eligible vehicles and the § 30D credits have strict price eligibility constraints, which likely limits the ability of manufacturers to raise prices enough to fully capture the credits for vehicles whose sticker prices are close to the limit. NHTSA notes that the overall new vehicle market supply curve is the sum of all individual vehicle supply curves, which are presumed to be upward sloping. This means that the overall new vehicle supply curve will be more elastic than individual vehicle supply curves at all price levels. This means that any effective tax or subsidy that only hits a subset of vehicles will have a greater incidence on the producer. Finally, unlike technology improvements, the § 30D credits have income limits for eligibility. Thus, the effective price for buyers of these vehicles is not uniform since some potential buyers will be above this income limit and will not qualify for the credit (and may not wish to lease a vehicle in order to claim the § 45W credit). Since manufacturers cannot set different MSRP's based on the customer's income, the sticker prices they choose may reflect a balance between raising prices and not losing market share from potential customers who do not qualify for the credits. As a result, NHTSA believes that its split incidence of the credits represents a reasonable approach to modeling this policy. We believe that a similar logic applies to the AMPC where manufacturers operating in a competitive market will not be able to fully capture the tax credit. Many suppliers and OEMs work closely together through contractual agreements and partnerships, and these close connections promote fair pricing arrangements that prevent any one party from capturing the full value of the credit. With regard to the future existence of these tax credits, NHTSA conducted sensitivity analysis of a case in which the tax credits are not included in the analysis but does not believe that this should be treated as the central analysis since these incentives are currently being claimed and are scheduled to be available in the years that NHTSA analyzed.

For this analysis, the agency maintained its assumption from the proposal that manufacturers and consumers will each capture half of the dollar value of the AMPC and CVCs. The agency assumes that manufacturers' shares of both credits will offset part of the cost to supply models that are eligible for the credits—PHEVs, BEVs, and FCEVs. The subsidies reduce the costs of eligible vehicles and increase their attractiveness to buyers (however, in the LD fleet, the tax credits do not alter the penetration rate of BEVs in the regulatory alternatives). Because the AMPC credit scales with battery capacity, NHTSA staff determined average battery energy capacity by powertrain ( e.g., PHEV, BEV, FCEV) for passenger cars, light trucks, and HDPUVs based on Argonne simulation outputs. For a more detailed discussion of these assumptions, see TSD Chapter 2.3.2. In the proposal NHTSA explained that it was unable to explicitly account for all of the eligibility requirements of § 30D and the AMPC, such as the location of final assembly and battery production, the origin of critical minerals, and the income restrictions of § 30D. Instead, we account for these restraints through the credit schedules that are constructed in part based off of these factors and allow all PHEVs, BEVs, and FCEVs produced and sold during the time frame that tax credits are offered to be eligible for those credits subject to the MSRP restrictions discussed above.

To account for the agency's inability to dynamically model sourcing requirements and income limits for § 30D, NHTSA used projected values of the average value of § 30D and the AMPC for the proposal. The projections increased throughout the analysis due to the expectation that gradual improvements in supply chains over time would allow more vehicles to qualify for the credits. Commenters suggested that NHTSA's assumed values for the § 30D credit were too optimistic and did not reflect limitations that manufacturers face in adjusting their supply chains and component manufacturing processes to produce vehicles that qualify for the credit. Similarly, some commenters argued that NHTSA did not adequately explain how it arrived at the credit estimates, did not offer any data to support the estimates, and failed to properly account for foreign entities of concern.

To address the concerns raised by commenters, NHTSA is using an independent report performed by DOE for the Final Rule that provides combined values of the CVCs. These values consider the latest information of EV penetration rates, EV retail prices, the share of US EV sales that meet the critical minerals and battery component requirements, the share of vehicles that exclude suppliers that are “Foreign Entities of Concern”, and lease rates for vehicles that qualify for the § 45W CVC. The DOE projections are the most detailed and rigorous projections of credit availability that NHTSA is aware of at this time. According to DOE's analysis the average credit value for the CVCs across all PHEV, BEV, and FCEV sales in a given year will never reach its full $7,500 value for all vehicles, and instead project a maximum average credit value of $6,000. NHTSA is using the same projection for the average AMPC credit per kwh as in the proposal.

Similar to the proposal, the CAFE Model's approach to analyzing the effects of the CVCs includes a statutory restriction. The CAFE Model accounts for the MSRP restrictions of the § 30D by assuming that the CVCs cannot be applied to cars with an MSRP above $55,000 or other vehicles with an MSRP above $80,000, since these are ineligible for § 30D. § 45W does not have the same MSRP restrictions, however since NHTSA is unable to model the CVCs separately at this time, the agency had to choose whether to model the restriction for both CVCs or not to model the restriction at all. NHTSA chose to include the restriction for both CVCs to be conservative. See Chapter 2.5.2 of the TSD for additional details on how NHTSA implements the IRA tax credits.

As the agency was coordinating with EPA and DOE on tax credits, NHTSA discovered that it was using nominal values for tax credits in the proposal instead of real dollars. NHTSA uses real dollars for future costs and benefits, such as technology costs in future model years. Including the tax credits as nominal dollars instead of real dollars artificially raises the value of the credits in respect to other costs. For the Final Rule, NHTSA has converted the DOE projections to real dollars.

As explained in the proposal, the CAFE model projects vehicles in model year cohorts rather than on a calendar year basis. Given that model years and calendar years can be misaligned, e.g., a MY 24 vehicle could be sold in calendar years 2023, 2024, or even 2025, choosing which calendar year a model year falls into is important for assigning tax credits which are phased-out during the analytical period. In the proposal, NHTSA assumed that the majority of vehicles of a given model year would be sold in the calendar year that preceded it, e.g., MY 2024 would largely be sold in calendar year 2023. NHTSA also noted at the time that there was a possible incentive for manufacturers to pull-up sales in the last calendar years that tax credits are available. NHTSA reanalyzed the timing of new vehicle sales and new vehicle registrations and determined that for the Final Rule it was appropriate to change its assumption that credits available in a given calendar year be available to all vehicles sold in the following model year. Instead, NHTSA decided to model vehicles in a given model year as eligible for credits available in the same calendar year. As a result, NHTSA applies the credits to MYs 2023-2032 in the analysis for both LDVs and HDPUVs.

6. Technology Applicability Equations and Rules

How does the CAFE Model decide how to apply technology to the analysis fleet of vehicles? We described above that the CAFE Model projects cost-effective ways that vehicle manufacturers could comply with CAFE standards, subject to limits that ensure that the model reasonably replicates manufacturer's decisions in the real-world. This section describes the equations the CAFE Model uses to determine how to apply technology to vehicles, including whether technologies are cost-effective, and why we believe the CAFE Model's calculation of potential compliance pathways reasonably represents manufacturers' decision-making. This section also gives a high-level overview of real-world limitations that vehicle manufacturers face when designing and manufacturing vehicles, and how we include those in the technology inputs and assumptions in the analysis.

The CAFE Model begins by looking at a manufacturer's fleet in a given MY and determining whether the fleet meets its CAFE standard. If the fleet does not meet its standard, the model begins the process of applying technology to vehicles. We described above how vehicle manufacturers use the same or similar engines, transmissions, and platforms across multiple vehicle models, and we track vehicle models that share technology by assigning Engine, Transmission, and Platform Codes to vehicles in the analysis fleet. As an example, the Ford 10R80 10-speed transmission is currently used in the following Ford Motor Company vehicles: 2017-present Ford F-150, 2018-present Ford Mustang, 2018-present Ford Expedition/Lincoln Navigator, 2019-present Ford Ranger, 2020-present Ford Explorer/Lincoln Aviator, and the 2020-present Ford Transit. The CAFE Model first determines whether any technology should be “inherited” from an engine, transmission, or platform that currently uses the technology to a vehicle that is due for a refresh or redesign. Using the Ford 10R80 10-speed transmission analysis as applied to the CAFE Model, the above models would be linked using the same Transmission Code. Even though the vehicles might be eligible for technology applications in different years because each vehicle model is on a different refresh or redesign cycle, each vehicle could potentially inherit the 10R80 10-speed transmission. The model then again evaluates whether the manufacturer's fleet complies with its CAFE standard. If it does not, the model begins the process of evaluating what from our universe of technologies could be applied to the manufacturer's vehicles.

The CAFE Model applies the most cost-effective technology out of all technology options that could potentially be applied. To determine whether a particular technology is cost-effective, the model will calculate the “effective cost” of multiple technology options and choose the option that results in the lowest “effective cost.” The “effective cost” calculation is actually multiple calculations, but we only describe the highest levels of that logic here; interested readers can consult the CAFE Model Documentation for additional information on the calculation of effective cost. Equation III-6 shows the CAFE Model's effective cost calculation for this analysis.

Where:

TechCost_Total: the total cost of a candidate technology evaluated on a group of selected vehicles;

TaxCredits_Total: the cumulative value of additional vehicle and battery tax credits (or, Federal Incentives) resulting from application of a candidate technology evaluated on a group of selected vehicles;

FuelSavings_Total: the value of the reduction in fuel consumption (or, fuel savings) resulting from application of a candidate technology evaluated on a group of selected vehicles;

Δ Fines: the change in manufacturer's fines in the analysis year if the CAFE compliance program is being evaluated, or zero if evaluating compliance with CO₂ standards;

Δ ComplianceCredits: the change in manufacturer's compliance credits in the analysis year, which depending on the compliance program being evaluated, corresponds to the change in CAFE credits (denominated in thousands of gallons) or the change in CO₂ credits (denominated in metric tons); and

EffCost: the calculated effective cost attributed to application of a candidate technology evaluated on a group of selected vehicles.

For the effective cost calculation, the CAFE Model considers the total cost of a technology that could be applied to a group of connected vehicles, just as a vehicle manufacturer might consider what new technologies it has that are ready for the market, and which vehicles should and could receive the upgrade. Next, like the technology costs, the CAFE Model calculates the total value of Federal incentives (for this analysis, Federal tax credits) available for a technology that could be applied to a group of vehicles and subtracts that total incentive from the total technology costs. For example, even though we do not consider the fuel economy of LD BEVs in our standard-setting analysis, we do account for the costs of vehicles that manufacturers may build in response to California's ACC I program (and in the HDPUV analysis, the ACT program), and additional electric vehicles that manufacturers have committed to deploy (consistent with ACC II), as part of our evaluation of how the world would look without our regulation, or more simply, the regulatory reference baseline. If the CAFE Model is evaluating whether to build a BEV outside of the MYs for which NHTSA is setting standards (if applicable in the modeling scenario), it starts with the total technology cost for a group of BEVs and subtracts the total value of the tax credits that could be applied to that group of vehicles.

The total fuel savings calculation is slightly more complicated. Broadly, when considering total fuel savings from switching from one technology to another, the CAFE Model must calculate the total fuel cost for the vehicle before application of a technology and subtract the total fuel cost for the vehicle after calculation of that technology. The total fuel cost for a given vehicle depends on both the price of gas (or gasoline equivalent fuel) and the number of miles that a vehicle is driven, among other factors. As technology is applied to vehicles in groups, the total fuel cost is then multiplied by the sales volume of a vehicle in a MY to equal total fuel savings. This equation also includes an assumption that consumers are likely to buy vehicles with fuel economy-improving technology that pays for itself within 2.5 years, or 30 months. Finally, in the numerator, we subtract the change in a manufacturer's expected fines before and after application of a specific technology. Then, the result from the sequence above is divided by the change in compliance credits, which means a manufacturer's credits earned (expressed as thousands of gallons for the purposes of effective cost calculation) in a compliance category before and after the application of a technology to a group of vehicles.

The effective cost calculation has evolved over successive CAFE Model iterations to become increasingly more complex; however, manufacturers' decision-making regarding what fuel economy-improving technology to add to vehicles has also become increasingly more complex. We believe this calculation appropriately captures a number of manufacturers implicit or explicit considerations.

The model accounts explicitly for each MY, applying technologies when vehicles are scheduled to be redesigned or freshened and carrying forward technologies between MYs once they are applied. The CAFE Model accounts explicitly for each MY because manufacturers actually “carry forward” most technologies between MYs, tending to concentrate the application of new technology to vehicle redesigns or mid-cycle “freshenings,” and design cycles vary widely among manufacturers and specific products. Comments by manufacturers and model peer reviewers to past CAFE rules have strongly supported explicit year-by-year simulation. The multi-year planning capability, simulation of “market-driven overcompliance,” and EPCA credit mechanisms increase the model's ability to simulate manufacturers' real-world behavior, accounting for the fact that manufacturers will seek out compliance paths for several MYs at a time, while accommodating the year-by-year requirement. This same multi-year planning structure is used to simulate responses to standards defined in grams CO₂ /mile and utilizing the set of specific credit provisions defined under EPA's program, when applicable in the modeling scenario.

In addition to the model's technology application decisions pursuant to the compliance simulation algorithm, there are also several technology inputs and assumptions that work together to determine which technologies the CAFE Model can apply. The technology pathways, discussed in detail above, are one significant way that we instruct the CAFE Model to apply technology. Again, the pathways define technologies that are mutually exclusive ( i.e., that cannot be applied at the same time), and define the direction in which vehicles can advance as the modeling system evaluates specific technologies for application. Then, the arrows between technologies instruct the model on the order in which to evaluate technologies on a pathway, to ensure that a vehicle that uses a more fuel-efficient technology cannot downgrade to a less efficient option.

In addition to technology pathway logic, we have several technology applicability rules that we use to better replicate manufacturers' decision-making. The “skip” input—represented in the Market Data Input File as “SKIP” in the appropriate technology column corresponding to a specific vehicle model—is particularly important for accurately representing how a manufacturer applies technologies to their vehicles in the real world. This tells the model not to apply a specific technology to a specific vehicle model. SKIP inputs are used to simulate manufacturer decisions with cost-benefit in mind, including (1) parts and process sharing; (2) stranded capital; and (3) performance neutrality.

First, parts sharing includes the concepts of platform, engine, and transmission sharing, which are discussed in detail in Section II.C.2 and Section II.C.3, above. A “platform” refers to engineered underpinnings shared on several differentiated vehicle models and configurations. Manufacturers share and standardize components, systems, tooling, and assembly processes within their products (and occasionally with the products of another manufacturer) to manage complexity and costs for development, manufacturing, and assembly. Detailed discussion for this type of SKIP is provided in the “adoption features” section for different technologies, if applicable, in Chapter 3 of the TSD.

Similar to vehicle platforms, manufacturers create engines that share parts. For instance, manufacturers may use different piston strokes on a common engine block or bore out common engine block castings with different diameters to create engines with an array of displacements. Head assemblies for different displacement engines may share many components and manufacturing processes across the engine family. Manufacturers may finish crankshafts with the same tools to similar tolerances. Engines on the same architecture may share pistons, connecting rods, and the same engine architecture may include both six- and eight-cylinder engines. One engine family may appear on many vehicles on a platform, and changes to that engine may or may not carry through to all the vehicles. Some engines are shared across a range of different vehicle platforms. Vehicle model/configurations in the analysis fleet that share engines belonging to the same platform are identified as such, and we also may apply a SKIP to a particular engine technology where we know that a manufacturer shares an engine throughout several of their vehicle models, and the engine technology is not appropriate for any of the platforms that share the same engine.

It is important to note that manufacturers define common engines differently. Some manufacturers consider engines as “common” if the engines share an architecture, components, or manufacturing processes. Other manufacturers take a narrower definition, and only assume “common” engines if the parts in the engine assembly are the same. In some cases, manufacturers designate each engine in each application as a unique powertrain. For example, a manufacturer may have listed two engines separately for a pair that share designs for the engine block, the crank shaft, and the head because the accessory drive components, oil pans, and engine calibrations differ between the two. In practice, many engines share parts, tooling, and assembly resources, and manufacturers often coordinate design updates between two similar engines. We consider engines together (for purposes of coding, discussed in Section III.C.2 above, and for SKIP application) if the engines share a common cylinder count and configuration, displacement, valvetrain, and fuel type, or if the engines only differed slightly in compression ratio (CR), horsepower, and displacement.

Parts sharing also includes the concept of sharing manufacturing lines (the systems, tooling, and assembly processes discussed above), since manufacturers are unlikely to build a new manufacturing line to build a completely new engine. A new engine that is designed to be mass manufactured on an existing production line will have limits in number of parts used, type of parts used, weight, and packaging size due to the weight limits of the pallets, material handling interaction points, and conveyance line design to produce one unit of a product. The restrictions will be reflected in the usage of a SKIP of engine technology that the manufacturing line would not accommodate.

SKIPs also relate to instances of stranded capital when manufacturers amortize research, development, and tooling expenses over many years, especially for engines and transmissions. The traditional production life cycles for transmissions and engines have been a decade or longer. If a manufacturer launches or updates a product with fuel-saving technology, and then later replaces that technology with an unrelated or different fuel-saving technology before the equipment and research and development investments have been fully paid off, there will be unrecouped, or stranded, capital costs. Quantifying stranded capital costs accounts for such lost investments. One design where manufacturers take an iterative redesign approach, as described in a recent SAE paper, is the MacPherson strut suspension. It is a popular low-cost suspension design and manufacturers use it across their fleet. As we observed previously, manufacturers may be shifting their investment strategies in ways that may alter how stranded capital could be considered. For example, some suppliers sell similar transmissions to multiple manufacturers. Such arrangements allow manufacturers to share in capital expenditures or amortize expenses more quickly. Manufacturers share parts on vehicles around the globe, achieving greater scale and greatly affecting tooling strategies and costs.

As a proxy for stranded capital, the CAFE Model accounts for platform and engine sharing and includes redesign and refresh cycles for significant and less significant vehicle updates. This analysis continues to rely on the CAFE Model's explicit year-by-year accounting for estimated refresh and redesign cycles, and shared vehicle platforms and engines, to moderate the cadence of technology adoption and thereby limit the implied occurrence of stranded capital and the need to account for it explicitly. In addition, confining some manufacturers to specific advanced technology pathways through technology adoption features acts as a proxy to indirectly account for stranded capital. Adoption features specific to each technology, if applied on a manufacturer-by-manufacturer basis, are discussed in each technology section. We discuss comments received on refresh and redesign cycles, parts-sharing, and SKIP logic below.

The National Resources Defense Council (NRDC) commented about several aspects of the redesign and refresh cycles included in the model. NRDC commented that we did not clearly explain why manufacturers' historic redesign cadences “are representative of what manufacturers `can' do if required,” citing EPCA's command that each standard we set be the “maximum feasible” standard. NRDC gave several examples, like that “NHTSA's historical data show that Ford and GM have redesigned heavier pickups every 6 years on average, Draft TSD at 2-29, but show Toyota taking 9 years on average.” NRDC stated that “[i]f it is feasible and practicable for two full-line manufacturers to redesign on a 6-year cadence, it is unclear why it is infeasible for others to do so as well.” NRDC continued on to state that “[t]he disparity between assumed redesign cycles for different automakers also appears to violate NHTSA's interpretation of `economic practicability,' which “has long abandoned the `least capable manufacturer' approach. 88 FR at 56,314.” NRDC also took issue with our interpretation that redesign cycles help us to account for stranded capital costs, which we do not explicitly include in our modeling, stating that “[t]he possibility of even considerable stranded capital for some automakers-a reduced probability given the considerable lead time to MY2031 here-is not a per se `harsh' economic consequence for the `industry,' . . . that might render standards not economically practicable.” NRDC requested that an alternative with reduced time between redesigns/refreshes should be modeled to compare the sensitivity of key metrics. NRDC also expressed that NHTSA's sensitivity case allowing for annual redesigns is not instructive and questioned the reasons for including it and not a more realistic case.

NHTSA agrees with NRDC that refresh and redesign cycles are a significant input to the CAFE Model, and we understand that using refresh and redesign cycles to represent stranded capital that otherwise would be difficult to quantify has been a longstanding point of disagreement between the agency and NRDC. NHTSA continues to believe that the resources manufacturers spend on new vehicle technologies—including developing, testing, and deploying those technologies—represents a significant amount of capital, although that number may be declining because, like both NHTSA and NRDC mentioned, manufacturers are taking advantage of sharing suppliers and sharing parts (which NHTSA does model).

While NHTSA does observe different trends in development cycles for different manufacturers, the adoption of new technologies, particularly for major and advanced components, continues to require multiple years of investment before being deployed to production models. Table 2-9 in the TSD contains information about the percentage of a manufacturer's vehicle fleet that is expected to be redesigned. The contents reflect that each manufacturer has their own development schedules, which vary due to multiple factors including technological adoption trends and consumer acceptance in specific market segments. We also show the average redesign schedules for each technology class in the TSD, which similarly bears out this trend. On the other hand, as discussed further in Section VI, vehicle manufacturers in comment to the proposal reiterated that their ability to spend resources improving ICE vehicles between now and MY 2031 are limited in light of the need to spend resources on the BEV transition. NHTSA understands this to mean that the potential for the negative consequences of stranding capital is an even more important consideration to manufacturers than it may have been in previous rules. For purposes of this analysis, we believe that our refresh and redesign cycles are reasonable, for the reasons discussed in more detail below. If NHTSA were to reevaluate refresh/ redesign cycles, it would be as part of a future rulemaking action, in which all stakeholders would have the opportunity to comment.

That said, we disagree that the way that we apply refresh and redesign cycles in the model is contrary to EPCA and we disagree with the examples that NRDC provided to illustrate that point. Allowing some manufacturers to have longer product redesign cycles does not conflict with our statement that we should not be setting standards with reference to a least capable manufacturer. There are several reasons why a manufacturer could be the “least capable” in fuel economy space that have nothing to do with its vehicles' refresh or redesign cycles. Using the example of manufacturers that NRDC provided, NHTSA's analysis estimates that under the preferred alternative in MY 2031, Ford's light truck fleet achieves a fuel economy level of 42.6 mpg, exactly meeting their standard, GM's light truck fleet achieves a fuel economy level of 40.9 mpg, falling short of their standard by 0.9 mpg, while Toyota's light truck fleet achieves a fuel economy level of 50.2 mpg, exceeding their standard by 3.7 mpg. Each manufacturer takes a different approach to redesigning its pickup trucks—Ford and GM every six years and Toyota every nine years—but on a fleet average basis, which is the relevant metric when considering fuel economy standards, each manufacturer's pickup design cycles are not indicative of their fleets' performance.

NRDC also stated that using historical average redesign cadences “can obscure significant variation about the average,” using as an example the design window for the Ram 1500 and the Ram 1500 Classic in their comment—stating that “[i]t is not clear how the automaker can feasibly update the 1500 every six years but not upgrade the 1500 Classic any faster than every 9 years.” The most recent redesign of the Ram 1500 Classic was in 2009 and it will continue to be sold as-is for the 2024 model year. Ram did update the 1500 in 2019 with a BISG system, but for reasons unique to Ram they decided to keep making the existing 1500 Classic. Since the manufacturer chose to keep the same product for 15 years, we cannot assume there would be a “lost” redesign window for this particular product. Note that the Ram 1500 Classic example is an extremely fringe example with a handful of other vehicles; as we showed in the Draft TSD and again in the Final TSD accompanying this rule, on average across the industry, manufacturers redesign vehicles every 6.6 years.

NRDC also commented about the interaction between redesign cycles and shared components, citing the Dodge Challenger as example of when “a vehicle may go into a redesign window, yet not have major components such as engines upgraded, because the leader vehicle for that engine [the Ram 1500 Classic] has not yet entered its redesign window. NHTSA believes that NRDC's Dodge Challenger/Ram example to support using alternative redesign assumptions is an incomplete understanding of how the CAFE Model considers leader-follower relationships and redesigns. The CAFE Model considers each component separately when determining the most cost-effective path to compliance. Sticking to engines, the Dodge Challenger can accept four different engines, one of which is not used in any Ram truck.

NHTSA does consider the effect of reducing the time between redesigns and refreshes through a sensitivity case, the “annual redesigns case,” which, as mentioned above, NRDC also took issue with. Perhaps we were not clear enough in the PRIA about the relative importance of this sensitivity case to our decision making, so we will clarify here. When we look at the annual redesign sensitivity case, we are examining the most extreme case of potential redesigns, explicitly not counting for the development, integration and manufacturing costs associated with such a cadence. Thus, this scenario is instructive of the upper bound of potential benefits under the assumption of unrestrained expenditures for vehicle design. While we agree that there are model outliers that could conceivably redesign closer to the average of six years, or even on an accelerated schedule of five years, we do not believe that we would see redesigns occurring, for example, any faster than three or four years. This is why we include planned vehicle refreshes in the modeling as well. Thus, the annual redesigns case is instructive because it shows us that any further refining of our redesign cadences ( i.e., on a scale between what we currently use and what we might consider reasonable for a lower bound schedule, which presumably would not be any shorter than the refresh schedule) would not have a significant impact on the analysis.

Like we maintain in other aspects of our analysis, some manufacturers' redesign cycles may be shorter than we model, and some manufacturers' redesign cycles may be longer than we model. We believe that it is reasonable to, on average, have our analysis reflect the capability of the industry. NHTSA will continue to follow industry trends in vehicle refresh and redesigns—like moving sales volume of an ICE model to a hybrid model, for example, or evaluating which technologies are now more frequently being applied during refreshes than redesigns—and consider how the refresh and redesign inputs could be updated in future analyses.

NHTSA also received two comments related to parts sharing. The Institute for Policy Integrity (IPI) at New York University School of Law commented that “NHTSA assumes that manufacturers apply the same costly technology to multiple models that share the same vehicle platform ( i.e., the car's essential design, engineering, and production components), while also (as noted above) maintaining their market shares irrespective of these cost changes.” IPI stated that this assumption “restricts manufacturers from optimizing their technology strategies,” which leads the model to overstate compliance costs. Similarly, NRDC argued that “NHTSA should reevaluate categorical restrictions on upgrading shared components on separate paths.” NRDC included several examples of components shared on vehicles that it thought resulted in a vehicle not being updated with additional technology.

While the CAFE Model considers part sharing by manufacturers across vehicle platforms, this assumption is based on real-world observations of the latest vehicle markets (See TSD 2.2, The Market Data Input File). As mentioned in TSD Chapter 2.2.1, manufacturers are expected to share parts across platforms to take advantage of economies of scale. These factors prevent the CAFE Model from predicting the adoption of unreasonably costly technologies across vehicle fleets.

While use of parts sharing by the CAFE Model is described as a restriction, we do not believe this is an accurate characterization. By considering upgrades across all vehicles that share a particular component, we are able to capture the total volume of that component in a way analogous to the manufacturers. If a potential upgrade is not cost-effective in the aggregate, it is unlikely that it would be cost-effective for a subset with a smaller volume.

IPI points to Mazda's MY 2032 estimated per-vehicle technology costs under alternative PC6LT8 as an example of an unrealistic outcome resulting from parts sharing. NHTSA maintains that this is an accurate projection of the effects of that regulatory alternative. The high per-vehicle costs in this specific case are due to a confluence of factors. The CAFE Model calculates the least expensive total regulatory cost, which includes both technology costs and fines. Mazda's preference to avoid fines in MY 2032 means that they would spend more on technology in order to comply with the standards. As a manufacturer, Mazda has an uncommonly high level of platform commonality, which means that investments in platform technology are likely to be propagated throughout their fleet in order to amortize costs more quickly. Their relatively small sales volume also drives up the per-vehicle costs. Taken together, these explain why the projected technology cost for Mazda is high, yet it is still within the same order of magnitude as some of Mazda's peer manufacturers (see FRIA Chapter 8). In the next most stringent regulatory alternative, Mazda's per-vehicle costs are projected to be in the middle of the pack compared to their peers.

NRDC also gave the example that the Dodge Challenger “will be prevented from upgrading to any high-compression ratio (HCR) engine, because the [sales] leader Classic 1500 is categorically excluded from upgrading to an HCR engine in the CAFE model because it is a pickup truck” as another example of the pitfalls of part sharing. NHTSA believes that this is a misreading of how the CAFE Model handles upgrade paths for shared components. The model restricts certain upgrade paths on the component level based on technology paths defined in TSD Chapter 3 and in this case, both the 1500 and the Challenger are only prevented from upgrading to a non-hybrid HCR engine. In the specific NRDC example, Engine Code 123602, a DOHC engine meant for high torque, was selected by Stellantis for, amongst other models, a pickup truck (Ram 1500 Classic) and a high-performance car (Dodge Challenger). HCR engines have higher efficiency at the cost of lower torque and lower power density, making them an unsuitable replacement for either model or any other model in this engine family. TSD Chapter 2.2.1, Characterizing Vehicles and their Technology Content has further information on how the CAFE Model applies SKIP logic. Also see TSD Chapter 3.1.1.2.3 for more information about HCR and Atkinson cycle engines.

NRDC also cited [an] “example of an engine-sharing family in its 2018 fuel economy standards proposal included the Chevy Equinox SUV, which shared a 6-cylinder engine with the Colorado and Canyon pickups (along with other vehicles)” that in later years “did not maintain engine sharing.” NHTSA stands by its position that historical data show manufacturers typically maintain parts commonality. The MY 2018 Chevy Equinox was available with two engines, a 4-cylinder and 6-cylinder, both naturally aspirated. The 4-cylinder variant was shared with the GMC Terrain and several Buick models which have since been discontinued, but not with the Chevy Colorado or GMC Canyon pickup trucks. This lineage was replaced by a choice of 1.5L or 2.0L 4-cylinder turbo engines in MY 2020 and now a single 1.5L 4-cylinder turbo in MY 2022. This engine is still shared between the Chevy Equinox and the GMC Terrain. In contrast, the Colorado and Canyon Pickups continue to use naturally aspirated engines in the 4-cylinder and 6-cylinder varieties, but these 4-cylinder engines are from a different lineage that were never shared with the Equinox. Instead of showing an example of manufacturers fracturing an existing engine family, this example validates our approach of considering technology upgrades at the component level.

Finally, we ensure that our analysis is performance neutral because the goal is to capture the costs and benefits of vehicle manufacturers adding fuel economy-improving technology because of CAFE standards, and not to inappropriately capture costs and benefits for changing other vehicle attributes that may have a monetary value associated with them. This means that we “SKIP” some technologies where we can reasonably assume that the technology would not be able to maintain a performance attribute for the vehicle, and where our simulation over test cycles may not capture the technology limitation.

For example, prior to the development of SAE J2807, manufacturers used internal rating methods for their vehicle towing capacity. Manufacturers switched to the SAE tow rating standard at the next redesign of their respective vehicles so that they could mitigate costs via parts sharing and remain competitive in performance. Usually, the most capable powertrain configuration will also have the highest towing capacity and can be reflected in using this input feature. Separately, we also ensure that the analysis is performance neutral through other inputs and assumptions, like developing our engine maps assuming use with a fuel grade most commonly available to consumers. Those assumptions are discussed throughout this section, and in Chapters 2 and 3 of the TSD. Technology “phase-in caps” and the “phase-in start years” are defined in the Technology Cost Input File and offer a way to gradually “phase-in” technology that is not yet fully mature to the analysis. They apply to the manufacturer's entire estimated production and, for each technology, define a share of production in each MY that, once exceeded, will stop the model from further applying that technology to that manufacturer's fleet in that MY.

The influence of these inputs varies with regulatory stringency and other model inputs. For example, setting the inputs to allow immediate 100 percent penetration of a technology will not guarantee any application of the technology if stringency increases are low and the technology is not at all cost effective. Also, even if these are set to allow only very slow adoption of a technology, other model aspects and inputs may nevertheless force more rapid application than these inputs, alone, would suggest ( e.g., because an engine technology propagates quickly due to sharing across multiple vehicles, or because BEV application must increase quickly in response to ZEV requirements). For this analysis, nearly all of these inputs are set at levels that do not limit the simulation at all.

This analysis also applies phase-in caps and corresponding start years to prevent the simulation from showing unlikely rates of applying battery-electric vehicles (BEVs), such as showing that a manufacturer producing very few BEVs in MY 2022 could plausibly replace every product with a 300- or 400-mile BEV by MY 2026. Also, this analysis applies phase-in caps and corresponding start years intended to ensure that the simulation's plausible application of the highest included levels of MR (20 percent reductions of vehicle “glider” weight) do not, for example, outpace plausible supply of raw materials and development of entirely new manufacturing facilities.

These model logical structures and inputs act together to produce estimates of ways each manufacturer could potentially shift to new fuel-saving technologies over time, reflecting some measure of protection against rates of change not reflected in, for example, technology cost inputs. This does not mean that every modeled solution would necessarily be economically practicable. Using technology adoption features like phase-in caps and phase-in start years is one mechanism that can be used so that the analysis better represents the potential costs and benefits of technology application in the rulemaking timeframe.

D. Technology Pathways, Effectiveness, and Cost

The previous section discussed, at a high level, how we generate the technology inputs and assumptions used in the CAFE Model. We do this in several ways: by evaluating data submitted by vehicle manufacturers; consolidating publicly available data, press materials, marketing brochures, and other information; collaborative research, testing, and modeling with other Federal agencies; research, testing, and modeling with independent organizations; determining that work done for prior rules is still relevant and applicable; considering feedback from stakeholders on prior rules and meetings conducted prior to the commencement of this rulemaking; and using our own engineering judgment.

This section discusses the specific technology pathways, effectiveness, and cost inputs and assumptions used in the compliance analysis. As an example, interested readers learned in the previous section that the starting point for estimating technology costs is an estimate of the DMC—the component and assembly costs of the physical parts and systems that make up a complete vehicle—for any particular technology; in this section, readers will learn that our transmission technology DMCs are based on estimates from the NAS.

After spending over a decade refining the technology pathways, effectiveness, and cost inputs and assumptions used in successive CAFE Model analyses, we have developed guiding principles to ensure that the CAFE Model's compliance analysis results in impacts that we would reasonably expect to see in the real world. These guiding principles are as follows:

Technologies will have complementary or non-complementary interactions with the full vehicle technology system. The fuel economy improvement from any individual technology must be considered in conjunction with the other fuel economy-improving technologies applied to the vehicle, because technologies added to a vehicle will not result in a simple additive fuel economy improvement from each individual technology. In particular, we expect this result from engine and other powertrain technologies that improve fuel economy by allowing the ICE to spend more time operating at efficient engine speed and load conditions, or from combinations of engine technologies that work to reduce the effective displacement of the engine.

The effectiveness of a technology depends on the type of vehicle the technology is being applied to. When we talk about “vehicle type” in our analysis, we're referring to our vehicle technology classes— e.g., a small car, a medium performance SUV, or a pickup truck, among other classes. A small car and a medium performance SUV that use the exact same technology will start with very different fuel economy values; so, when the exact same technology is added to both of those vehicles, the technology will provide a different effectiveness improvement on both of those vehicles.

The cost and effectiveness values for each technology should be reasonably representative of what can be achieved across the entire industry. Each technology model employed in the analysis is designed to be representative of a wide range of specific technology applications used in industry. Some manufacturers' systems may perform better or worse than our modeled systems and some may cost more or less than our modeled systems; however, employing this approach will ensure that, on balance, the analysis captures a reasonable level of costs and benefits that would result from any manufacturer applying the technology.

A consistent reference point for cost and effectiveness values must be identified before assuming that a cost or effectiveness value could be employed for any individual technology. For example, as discussed below, this analysis uses a set of engine map models that were developed by starting with a small number of engine configurations, and then, in a very systematic and controlled process, adding specific well-defined technologies to create a new map for each unique technology combination. Again, providing a consistent reference point to measure incremental technology effectiveness values ensures that we are capturing accurate effectiveness values for each technology combination.

The following sections discuss the engine, transmission, electrification, MR, aerodynamic, ROLL, and other vehicle technologies considered in this analysis. The following sections discuss:

• How we define the technology in the CAFE Model,

• How we assigned the technology to vehicles in the analysis fleet used as a starting point for this analysis,
• Any adoption features applied to the technology, so the analysis better represents manufacturers' real-world decisions,
• The technology effectiveness values, and
• Technology cost.

Please note that the following technology effectiveness sections provide examples of the range of effectiveness values that a technology could achieve when applied to the entire vehicle system, in conjunction with the other fuel economy-improving technologies already in use on the vehicle. To see the incremental effectiveness values for any particular vehicle moving from one technology key to a more advanced technology key, see the CAFE Model Fuel Economy Adjustment Files that are installed as part of the CAFE Model Executable File, and not in the input/output folders. Similarly, the technology costs provided in each section are examples of absolute costs seen in specific MYs, for specific vehicle classes. Please refer to the Technologies Input File to see all absolute technology costs used in the analysis across all MYs.

For the LD analysis we show two sets of technology effectiveness charts for each technology type, titled “Unconstrained” and “Standard Setting.” For the Standard Setting charts, effectiveness values reflect the application of 49 U.S.C. 32902(h) considerations to the technologies; for example, PHEV technologies only show the effectiveness achieved when operating in a gasoline only mode (charge sustaining mode). The Unconstrained charts show the effectiveness values modeled for the technologies without the 49 U.S.C; 32902(h) constraints; when unconstrained, PHEV technologies show effectiveness for their full dual fuel use functionality. The standard setting values are used during the standard setting years being assessed in this analysis, and the unconstrained values are used for all other years.

1. Engine Paths

ICEs convert chemical energy in fuel to useful mechanical power. The chemical energy in the fuel is released and converted to mechanical power by being oxidized, or burned, inside the engine. The air/fuel mixture entering the engine and the burned fuel/exhaust by-products leaving the engine are the working fluids in the engine. The engine power output is a direct result of the work interaction between these fluids and the mechanical components of the engine. The generated mechanical power is used to perform useful work, such as vehicle propulsion. For a complete discussion on fundamentals of engine characteristics, such as torque, torque maps, engine load, power density, brake mean effective pressure (BMEP), combustion cycles, and components, please refer to Heywood 2018.

We classify the extensive variety of both LD and HDPUV vehicle ICE technologies into discrete Engine Paths. These paths are used to model the most representative characteristics, costs, and performance of the fuel economy-improving engine technologies most likely available during the rulemaking time frame. The paths are intended to be representative of the range of potential performance levels for each engine technology. In general, the paths are tied to ease of implementation of additional technology and how closely related the technologies are. The technology paths for LD and HDPUV can be seen in Chapter 3.1.1 of the TSD.

The LD Engine Paths have been selected and refined over a period of more than ten years, based on engines in the market, stakeholder comments, and our engineering judgment, subject to the following factors: we included technologies most likely available during the rulemaking time frame and the range of potential performance levels for each technology, and excluded technologies unlikely to be feasible in the rulemaking timeframe, technologies unlikely to be compatible with U.S. fuels, or technologies for which there was not appropriate data available to allow the simulation of effectiveness across all vehicle technology classes in this analysis.

For technologies on the HDPUV Engine Paths, we revisited work done for the HDPUV analysis in the Phase 2 rulemaking. We have updated our HDPUV Engine Paths based on that work, the availability of technology in the HDPUV analysis fleet, and technologies we believe will be available in the rulemaking timeframe. The HDPUV fleet is significantly smaller than the LD fleet with the majority of vehicles being produced by only three manufacturers, General Motors, Ford, and Stellantis. These vehicles include work trucks and vans that are focused on transporting people and moving equipment and supplies and tend to be more focused on a common need than that of vehicles in the LD fleet, which includes everything from sports cars to commuter cars and pickup trucks. The engine options between the two fleets are different in the real world and are accordingly different in the analysis. HDPUVs are work vehicles and their engines must be able to handle additional work such as higher payloads, towing, and additional stop and go demands. This results in HDPUVs often requiring larger, more robust, and more powerful engines. As a result of the HDPUV's smaller fleet size and narrowed focus, fewer engines and engine technologies are developed or used in this fleet. That said, we believe that the range of technologies included in the HDPUV Engine Paths and Electrification/Hybrid/Electrics Path discussed in Section III.D.3 of this preamble presents a reasonable representation of powertrain options available for HDPUVs now and in the rulemaking time frame.

The Engine Paths begin with one of the three base engine configurations: dual over-head camshaft (DOHC) engines have two camshafts per cylinder head (one operating the intake valves and one operating the exhaust valves), single over-head camshaft (SOHC) engines have a single camshaft, and over-head valve (OHV) engines also have a single camshaft located inside of the engine block (south of the valves rather than over-head) connected to a rocker arm through a push rod that actuates the valves. DOHC and SOHC engine configurations are common in the LD fleet, while OHV engine configurations are more common in the HDPUV fleet.

The next step along the Engine Paths is at the Basic Engine Path technologies. These include variable valve lift (VVL), stoichiometric gasoline direct injection (SGDI), and a basic level of cylinder deactivation (DEAC). VVL dynamically adjusts how far the valve opens and reduces fuel consumption by reducing pumping losses and optimizing airflow over broader range of engine operating conditions. Instead of injecting fuel at lower pressures and before the intake valve, SGDI injects fuel directly into the cylinder at high pressures allowing for more precise fuel delivery while providing a cooling effect and allowing for an increase in the CR and/or more optimal spark timing for improved efficiency. DEAC disables the intake and exhaust valves and turns off fuel injection and spark ignition on select cylinders which effectively allows the engine to operate temporarily as if it were smaller while also reducing pumping losses to improve efficiency. New for the NPRM and carried into this final rule analysis is that variable valve timing (VVT) technology is integrated in all non-diesel engines, so we do not have a separate box for it on the Basic Engine Path. For the LD analysis, VVL, SGDI, and DEAC can be applied to an engine individually or in combination with each other, and for the HDPUV analysis, SGDI and DEAC can be applied individually or in combination.

Moving beyond the Basic Engine Path technologies are the “advanced” engine technologies, which means that applying the technology—both in our analysis and in the real world—would require significant changes to the structure of the engine or an entirely new engine architecture. The advanced engine technologies represent the application of alternate combustion cycles, various applications of forced induction technologies, or advances in cylinder deactivation.

Advanced cylinder deactivation (ADEAC) systems, also known as rolling or dynamic cylinder deactivation systems, allow the engine to vary the percentage of cylinders deactivated and the sequence in which cylinders are deactivated. Depending on the engine's speed and associated torque requirements, an engine might have most cylinders deactivated ( e.g., low torque conditions as with slower speed driving) or it might have all cylinders activated ( e.g., high torque conditions as with merging onto a highway). An engine operating at low speed/low torque conditions can then save fuel by operating as if it is only a fraction of its total displacement. We model two ADEAC technologies, advanced cylinder deactivation on a single overhead camshaft engine (ADEACS), and advanced cylinder deactivation on a dual overhead camshaft engine (ADEACD).

Forced induction gasoline engines include both supercharged and turbocharged downsized engines, which can pressurize or force more air into an engine's intake manifold when higher power output is needed. The raised pressure results in an increased amount of airflow into the cylinder supporting combustion, increasing the specific power of the engine. The first-level turbocharged downsized technology (TURBO0) engine represents a basic level of forced air induction technology being applied to a DOHC engine. Cooled exhaust gas recirculation (CEGR) systems take engine exhaust gasses and passes them through a heat exchanger to reduce their temperature, and then mixes them with incoming air in the intake manifold to reduce peak combustion temperature and effect fuel efficiency and emissions. We model the base TURBO0 turbocharged engine with the addition of cooled exhausted recirculation (TURBOE), basic cylinder deactivation (TURBOD), and advanced cylinder deactivation (TURBOAD). Advancing further into the Turbo Engine Path leads to engines that have higher BMEP, which is a function of displacement and power. The higher the BMEP, the higher the engine performance. We model two levels of advanced turbocharging technology (TURBO1 and TURBO2) that run increasingly higher turbocharger boost levels, burning more fuel and making more power for a given displacement. As discussed above, we pair turbocharging with engine downsizing, meaning that the turbocharged downsized engines in our analysis improve vehicle fuel economy by using less fuel to power the smaller engine while maintaining vehicle performance.

NHTSA received a limited number of comments on forced induction gasoline engines. The comments seemed to highlight some misunderstandings of our forced induction pathway rather than the technology itself and how it was applied in our analysis for this rulemaking. In discussing the turbocharged pathway NRDC commented, “. . . NHTSA has not appropriately considered the relative efficiency of these engines with respect to each other when designing its technology pathways. As a result, the technology pathway does not reasonably reflect an appropriate consideration of the full availability of turbocharged engine improvements.”

NRDC assumed that the pathways are in order from least effective to most effective, however, this is not how the technologies are arranged in the pathway. The technology pathways represent an increase in the level or combinations of technologies being applied, with lower levels at the top and higher levels at the bottom of the path. Chapter 3.1.1 of the TSD shows the technology pathways for visualization purposes, however the CAFE Model could apply any cost-effective combinations of technologies from those given pathways. Levels of improvement are dependent upon the vehicle class and the technology combinations. As a reminder, we stated in the NPRM section describing the technology pathways just before the figure of the technology tree that “[i]n general, the paths are tied to ease of implementation of additional technology and how closely related the technologies are.” An example of how this applies to the TURBO family of technologies is described below. To the extent that the verbiage around the technology tree was confusing, we will endeavor to make that clearer moving forward. The pathways are not aligned from “least effective” to “most effective” because assuming so would ignore several important considerations, including how technologies interact on a vehicle, how technologies interact on vehicles of different sizes that have different power requirements, and how hardware changes may be required for a particular technology (see above, “ease of implementation of additional technology,” and the related example below that describes how once a manufacturer downsizes an engine accompanying the application of a turbocharger, it would most likely not then re-upsize the engine to add a less advanced turbocharger). The interaction of these technology combinations is discussed in more details in TSD Chapter 2.

While we have modeled TURBO0 with cooled EGR (TURBOE) and with DEAC (TURBOD), NRDC is correct that we do not apply these technologies to TURBO1 or TURBO2; this decision was intentional and not a lapse in engineering judgment, as NRDC seems to imply. We define TURBO1 in our analysis by adding VVL to the TURBO0 engine, and TURBO2 is our highest turbo downsized engine with a high BMEP. The benefits of cooled EGR and DEAC on TURBO1 and TURBO2 technologies would occur at high engine speeds and loads, which do not occur on the two-cycle tests. Because technology effectiveness in our analysis is measured based on the delta in improvements in vehicles' two-cycle test fuel consumption values, adding cooled EGR and DEAC to TURBO1 and TURBO2 would provide little effectiveness improvement in our analysis with a corresponding increase in cost that we do not believe manufacturers would adopt in the real world. These complex interactions among technologies are effectively captured in our modeling and this is an example of why we do not simply add effectiveness values from different technologies together. This potential for added costs with limited efficiency benefit is also an example of why we do not order our technology tree from least to most effective technology, and we choose to include particular technologies on the technology tree and not others. For more discussion on interactions among individual technologies in the full vehicle simulations, see TSD Chapter 2.

NRDC also believes the model is improperly constrained because it cannot apply lower levels of technology over higher levels, which results in a situation where vehicles in the analysis fleet that have been assigned higher levels of turbocharging technology cannot adopt what NRDC alleges to be a more efficient turbocharged engine technology. For example, the model does not allow a vehicle assigned a TURBO2 technology to adopt a TURBOE technology. A vehicle in the analysis fleet that is assigned the TURBO2 technology tells us a manufacturer made the decision to either skip over or move on from lower levels of force induction technology. Moving backwards in the technology tree from TURBO2 to any of the lower turbo technologies would require the engine to be upsized to meet the same performance metrics as the analysis fleet vehicle. As discussed further in Section III.C.6, we ensure the vehicles in our analysis meet similar performance levels after the application of fuel economy-improving technology because we want to measure the costs and benefits of manufacturers responding to CAFE standards in our analysis, and not the costs or benefits related to changing performance metrics in the fleet. Moving from a higher to a lower turbo technology works counter to saving fuel as the engine would grow in displacement requiring more fuel, adding frictional losses, and increasing weight and cost. While fuel economy is important to manufacturers, it is not the only parameter that drives engine or technology selection, and it goes against the industry trends for downsized engines. Accordingly, we believe that our Turbo engine pathway appropriately captures the ways manufacturers might apply increasing levels of turbocharging technology to their vehicles.

In this analysis, high compression ratio (HCR) engines represent a class of engines that achieve a higher level of fuel efficiency by implementing a high geometric CR with varying degrees of late intake valve closing (LIVC) ( i.e., closing the intake valve later than usual) using VVT, and without the use of an electric drive motor. These engines operate on a modified Atkinson cycle allowing for improved fuel efficiency under certain engine load conditions but still offering enough power to not require an electric motor; however, there are limitations on how HCR engines can apply LIVC and the types of vehicles that can use this technology. The way that each individual manufacturer implements a modified Atkinson cycle will be unique, as each manufacturer must balance not only fuel efficiency considerations, but emissions, on-board diagnostics, and safety considerations that includes the vehicle being able to operate responsively to the driver's demand.

We define HCR engines as being naturally aspirated, gasoline, SI, using a geometric CR of 12.5:1 or greater, and able to dynamically apply various levels of LIVC based on load demand. An HCR engine uses less fuel for each engine cycle, which increases fuel economy, but decreases power density (or torque). Generally, during high loads—when more power is needed—the engine will use variable valve actuation to reduce the level of LIVC by closing the intake valve earlier in the compression stroke (leaving more air/fuel mixture in the combustion chamber), increasing the effective CR, reducing over-expansion, and sacrificing efficiency for increased power density. However, there is a limit to how much the air-fuel mixture can be compressed before ignition in the HCR engine due to the potential for engine knock Engine knock can be mitigated in HCR engines with higher octane fuel, however, the fuel specified for use in most vehicles is not this higher octane fuel. Conversely, at low loads the engine will typically increase the level of LIVC by closing the intake valve later in the compression stroke, reducing the effective CR, increasing the over-expansion, and sacrificing power density for improved efficiency. By closing the intake valve later in the compression stroke ( i.e., applying more LIVC), the engine's displacement is effectively reduced, which results in less air and fuel for combustion and a lower power output. Varying LIVC can be used to mitigate, but not eliminate, the low power density issues that can constrain the application of an Atkinson-only engine.

When we say, “lower power density issues,” this translates to a low torque density, meaning that the engine cannot create the torque required at necessary engine speeds to meet load demands. To the extent that a vehicle requires more power in a given condition than an engine with low power density can provide, that engine would experience issues like engine knock for the reasons discussed above, but more importantly, an engine designer would not allow an engine application where the engine has the potential to operate in unsafe conditions in the first place. Instead, a manufacturer could significantly increase an engine's displacement ( i.e., size) to overcome those low power density issues, or could add an electric motor and battery pack to provide the engine with more power, but a far more effective pathway would be to apply a different type of engine technology, like a downsized, turbocharged engine.

Vehicle manufacturers' intended performance attributes for a vehicle—like payload and towing capability, features for off-road use, and other attributes that affect aerodynamic drag and rolling resistance—dictate whether an HCR engine can be a suitable technology choice for that vehicle. As vehicles require higher payloads and towing capacities, or experience road load increases from larger all-terrain tires, a less aerodynamic design, or experience driveline losses for AWD and 4WD configurations, more engine torque is required at all engine speeds. Any time more engine torque is required the application of HCR technology becomes less effective and more limited. For these reasons, and to maintain a performance-neutral analysis and as discussed further below, we limit non-hybrid and non-plug-in-hybrid HCR engine application to certain categories of vehicles. Also for these reasons, HCR engines are not found in the HDPUV analysis fleet nor are they available as an engine option in the HDPUV analysis.

For this analysis, our HCR Engine Path includes three technology options: (1) a first-level Atkinson-enabled engine (HCR) with VVT and SGDI, (2) an Atkinson enabled engine with cooled exhaust gas recirculation (HCRE), and finally, (3) the Atkinson enabled engine with DEAC (HCRD). This updated family of HCR engine map models also reflects our statement in NHTSA's May 2, 2022 final rule that a single engine that employs an HCR, CEGR, and DEAC “is unlikely to be utilized in the rulemaking timeframe based on comments received from the industry leaders in HCR technology application.”

These three HCR Engine Path technology options (HCR, HCRE, HCRD) should not be confused with the hybrid and plug-in hybrid electric pathway options that also utilize HCR engines in combination with an P2 hybrid powertrain ( i.e., P2HCR, P2HCRE, PHEV20H, and PHEV50H); those hybridization path options are discussed in Section III.D.3, below. In contrast, Atkinson engines in our powersplit hybrid powertrains (SHEVPS, PHEV20PS, and PHEV50PS) for this analysis run the Atkinson Cycle full time but are connected to an electric motor. The full-time Atkinson engines are also discussed in Section III.D.3.

The Miller cycle is another alternative combustion cycle that effectively uses an extended expansion stroke, similar to the Atkinson cycle but with the application of forced induction, to improve fuel efficiency. Miller cycle-enabled engines have a similar trade-off in power density as Atkinson engines; the lower power density requires a larger volume engine in comparison to an Otto cycle-based turbocharged system for similar applications. To address the impacts of the extended expansion stroke on power density during high load operating conditions, the Miller cycle operates in combination with a forced induction system. In our analysis, the first-level Miller cycle-enabled engine includes the application of variable turbo geometry technology (VTG), or what is also known as a variable-geometry turbocharger. VTG technology allows for the adjustment of key geometric characteristics of the turbocharging system, thus allowing adjustment of boost profiles and response based on the engine's operating needs. The adjustment of boost profile during operation increases the engine's power density over a broader range of operating conditions and increases the functionality of a Miller cycle-based engine. The use of a variable geometry turbocharger also supports the use of CEGR. The second level of VTG engine technology in our analysis (VTGE) is an advanced Miller cycle-enabled system that includes the application of at least a 40V-based electronic boost system. An electronic boost system has an electric motor added to assist the turbocharger; the motor assist mitigates turbocharger lag and low boost pressure by providing the extra boost needed to overcome the torque deficit at low engine speeds.

Variable compression ratio (VCR) engines work by changing the length of the piston stroke of the engine to optimize the CR and improve thermal efficiency over the full range of engine operating conditions. Engines that use VCR technology are currently in production as small displacement turbocharged in-line four-cylinder, high BMEP applications.

Diesel engines have several characteristics that result in better fuel efficiency over traditional gasoline engines, including reduced pumping losses due to lack of (or greatly reduced) throttling, high pressure direct injection of fuel, a combustion cycle that operates at a higher CR, and a very lean air/fuel mixture relative to an equivalent-performance gasoline engine. However, diesel technologies require additional systems to control NO_X emissions, such as a NO_X adsorption catalyst system or a urea/ammonia selective catalytic reduction system. We included two levels of diesel engine technology in both the LD and HDPUV analyses: the first-level diesel engine technology (ADSL) is a turbocharged diesel engine, and the more advanced diesel engine (DSLI) adds DEAC to the ADSL engine technology. The diesel engine maps are new for this analysis. The LD diesel engine maps and HD van engine maps are based on a modern 3.0L turbo-diesel engine, and the HDPUV pickup truck engine maps are based on a larger 6.7L turbo-diesel engine.

Finally, compressed natural gas (CNG) systems are ICEs that run on natural gas as a fuel source. The fuel storage and supply systems for these engines differ tremendously from gasoline, diesel, and flex fuel vehicles. The CNG engine option has been included in past analyses; however, the LD and HDPUV analysis fleets do not include any dedicated CNG vehicles. As with the last analyses, CNG engines are included as an analysis fleet-only technology and are not applied to any vehicle that did not already include a CNG engine.

We received several comments that gave examples of vehicle technologies that work in various ways to improve fuel efficiency, some of which we use in our analysis and some we do not. MECA gave us several examples of fuel efficiency technologies that we use in our analysis such as cylinder deactivation, VVT and VVL, VTG, and VTGe. MECA also discussed technologies we do not use in the analysis such as turbo compounding. Similarly, ICCT gave examples of technology such as negative valve overlap in-cylinder fuel reforming (NVO), passive prechamber combustion (PPC), and high energy ignition, that we also did not use in this analysis.

These technologies are in various stages of development and some like PPC are in very limited production; however, we did not include them in the analysis as we do not believe these technologies will gain enough adoption during the rulemaking timeframe. We had discussed this topic in detail in the 2022 final rule and we do not think that there has been any significant development since than that would indicate that manufacturers would pursue these costly technologies. If anything, manufacturers have indicated that they are willing to continue to research and develop more cost effective electrification technologies such as strong hybrids and PHEVs to meet current and future regulations from multiple agencies.

The Alliance for Vehicle Efficiency commented that they want to see stronger support for hydrogen combustion and fuel cell vehicles in the HDPUV fleet. Hydrogen powertrain technology has been in development for years and there are several roadblocks to more mainstream adoption such as system packaging, infrastructure, technology reliability and durability, and costs to name a few. While hydrogen powertrain technology has the possibility to provide improved efficiency and even with funding support from the IRA, these technologies still do not show up in the HDPUV fleet today and we do not believe the technology will gain enough market penetration in the rule making timeframe for us to include them in the pathway to compliance.

The first step in assigning engine technologies to vehicles in the LD and HDPUV analysis fleets is to use data for each manufacturer to determine which vehicle platforms share engines. Within each manufacturer's fleet, we develop and assign unique engine codes based on configuration, technologies applied, displacement, CR, and power output. While the process for engine assignments is the same between the LD and HDPUV analyses, engine codes are not shared between the two fleets, and engine technologies are not shared between the fleets, for the reasons discussed above. We also assign engine technology classes, which are codes that identify engine architecture ( e.g., how many cylinders the engine has, whether it is a DOHC or SOHC, and so on) to accurately account for engine costs in the analysis.

When we assign engine technologies to vehicles in the analysis fleets, we must consider the actual technologies on a manufacturer's engine and compare those technologies to the engine technologies in our analysis. We have just over 270 unique engine codes in the LD analysis fleet and just over 20 unique engine codes in the HDPUV fleet, meaning that for both analysis fleets, we must identify the technologies present on those almost 300 unique engines in the real world, and make decisions about which of our approximately 40 engine map models (and therefore engine technology on the technology tree) best represents those real-world engines. When we consider how to best fit each of those 300 engines to our 40 engine technologies and engine map models, we use specific technical elements contained in manufacturer publications, press releases, vehicle benchmarking studies, technical publications, manufacturer's specification sheets, and occasionally CBI (like the specific technologies, displacement, CR, and power mentioned above), and engineering judgment. For example, in the LD analysis, an engine with a 13.0:1 CR is a good indication that an engine would be considered an HCR engine in our analysis, and some engines that achieve a slightly lower CR, e.g., 12.5, may be considered an HCR engine depending on other technology on the engine, like inclusion of SGDI, increased engine displacement compared to other competitors, a high energy spark system, and/or reduction of engine parasitic losses through variable or electric oil and water pumps. Importantly, we never assign engine technologies based on one factor alone; we use data and engineering judgment to assign complex real-world engines to their corresponding engine technologies in the analysis. We believe that our initial characterization of the fleet's engine technologies reasonably captures the current state of the market while maintaining a reasonable amount of analytical complexity. Also, as a reminder, in addition to the 40 engine map models used in the Engine Paths Collection, we have over 20 additional potential powertrain technology assignments available in the Hybrid/Electric Paths Collection.

Engine technology adoption in the model is defined through a combination of technology path logic, refresh and redesign cycles, phase-in capacity limits, and SKIP logic. How does technology path logic define technology adoption? Once an engine design moves to the advanced engine tree it is not allowed to move to alternate advanced engine trees. For example, any LD basic engine can adopt one of the TURBO engine technologies, but vehicles that have turbocharged engines in the analysis fleet will stay on the Turbo Engine Path to prevent unrealistic engine technology change in the short timeframe considered in the rulemaking analysis. This represents the concept of stranded capital, which as discussed above, is when manufacturers amortize research, development, and tooling expenses over many years. Besides technology path logic, which applies to all manufacturers and technologies, we place additional constraints on the adoption of VCR and HCR technologies.

VCR technology requires a complete redesign of the engine, and in the analysis fleet, only two models have incorporated this technology. VCR engines are complex, costly by design, and address many of the same efficiency losses as mainstream technologies like turbocharged downsized engines, making it unlikely that a manufacturer that has already started down an incongruent technology path would adopt VCR technology. Because of these issues, we limited adoption of the VCR engine technology to original equipment manufacturers (OEMs) that have already employed the technology and their partners. We do not believe any other manufacturers will invest to develop and market this technology in their fleet in the rulemaking time frame.

HCR engines are subject to three limitations. This is because, as we have recognized in past analyses, HCR engines excel in lower power applications for lower load conditions, such as driving around a city or steady state highway driving without large payloads. Thus, their adoption is more limited than some other technologies.

First, we do not allow vehicles with 405 or more horsepower, and (to simulate parts sharing) vehicles that share engines with vehicles with 405 or more horsepower, to adopt HCR engines due to their prescribed power needs being more demanding and likely not supported by the lower power density found in HCR-based engines. Because LIVC essentially reduces the engine's displacement, to make more power and keep the same levels of LIVC, manufacturers would need to increase the displacement of the engine to make the necessary power. We do not believe manufacturers will increase the displacement of their engines to accommodate HCR technology adoption because as displacement increases so does friction, pumping losses, and fuel consumption. This bears out in industry trends: total engine size (or displacement) is at an all-time low, and trends show that industry focus on turbocharged downsized engine packages are leading to their much higher market penetration. Separately, as seen in the analysis fleet, manufacturers generally use HCR engines in applications where the vehicle's power requirements fall significantly below our horsepower threshold. In fact, the average horsepower for the sales weighted average of vehicles in the analysis fleet that use HCR Engine Path technologies is 179 hp, demonstrating that HCR engine use has indeed been limited to lower-hp applications, and well below our 405 hp threshold. In fringe cases where a vehicle classified as having higher load requirements does have an HCR engine, it is coupled to a hybrid system.

Second, to maintain a performance-neutral analysis, we exclude pickup trucks and (to simulate parts sharing) vehicles that share engines with pickup trucks from receiving HCR engines that are not accompanied by an electrified powertrain. In other words, pickup trucks and vehicles that share engines with pickup trucks can receive HCR-based engine technologies in the Hybridization Paths Collection of technologies. We exclude pickup trucks and vehicles that share engines with pickup trucks from receiving HCR engines that are not accompanied by an electrified powertrain because these often-heavier vehicles have higher low speed torque needs, higher base road loads, increased payload and towing requirements, and have powertrains that are sized and tuned to perform this additional work above what passenger cars are required to conduct. Again, vehicle manufacturers' intended performance attributes for a vehicle—like payload and towing capability, intention for off-road use, and other attributes that affect aerodynamic drag and rolling resistance—dictate whether an HCR engine can provide a reasonable fuel economy improvement for that vehicle. For example, road loads are comprised of aerodynamic loads, which include vehicle frontal area and its drag coefficient, along with tire rolling resistance that attribute to higher engine loads as vehicle speed increases. We assume that a manufacturer intending to apply HCR technology to their pickup truck or vehicle that shares an engine with a pickup truck would do so in combination with an electric system to assist with the vehicle's load needs, and indeed the only manufacturer that has an HCR-like engine (in terms of how we model HCR engines in this analysis) in its pickup truck in the analysis fleet has done so.

Finally, we restrict HCR engine application for some manufacturers that are heavily performance-focused and have demonstrated a significant commitment to power dense technologies such as turbocharged downsizing. When we say, “significant commitment to power dense technologies,” we mean that their fleets use near 100% turbocharged downsized engines. This means that no vehicle manufactured by these manufacturers can receive an HCR engine. Again, we implement this adoption feature to avoid an unquantified amount of stranded capital that would be realized if these manufacturers switched from one technology to another.

Note, however, that these adoption features only apply to vehicles that receive HCR engines that are not accompanied by an electrified powertrain. A P2 hybrid system that uses an HCR engine overcomes the low-speed torque needs using the electric motor and thus has no restrictions or SKIPs applied.

We received a limited number of comments disagreeing with the HCR restrictions we have in place, most of which had been received in previous rulemakings. To avoid repetition, previous discussions located in prior related documents are adopted here by reference.

We realize that engine technology, vehicle type, and their applications are always evolving, and we agree with both the States and Cities and the Joint NGOs that the Hyundai Santa Cruz, unibody pickup truck with a 4-cylinder HCR engine, is one example of a pickup truck with a non-hybrid HCR engine. However, we disagree that the Santa Cruz is comparable in capability to other pickup models like the Tacoma, Colorado, and Canyon, and that those pickup models should therefore be able to adopt non-hybrid HCR technology as well. Small unibody pickup trucks like the Santa Cruz and the Ford Maverick do not have the same capabilities and functionality as a body-on-frame pickup like the Toyota Tacoma. We believe our current restrictions for HCR are reasonable and appropriate and we have not been presented with any new information that would suggest otherwise. Our stance on this issue has also borne out in real-world trends. Manufacturers who had the potential to use HCR technologies for high utility capable vehicles like Toyota Tacoma and Mazda CX-90 (replacing CX-9) have incorporated turbocharged engines. We do not believe HCR in its current state can provide enough fuel efficiency benefit for us to remove our current HCR restrictions; however, this by no means precludes manufacturers from developing and deploying HCR technology for future iterations of their pickup trucks.

We would also like to emphasize in response to the Joint NGOs that manufacturers do not pursue technology pathways because we model them in our analysis supporting setting CAFE and HDPUV standards. We have stated multiple times that we give an example of a low-cost compliance pathway, and no manufacturer has to comply with the pathway as we have modeled it. In fact, it is more than likely they will not follow the technology pathways we project in our standard-setting analysis because of the standard setting restrictions we have in place. Also, we do not allege that manufacturers cannot use different technologies than we model in our analysis to meet their standard, we just do not believe that manufacturers will abandon investments in one technology pathway for another, particularly with respect to HCR technology for pickup trucks and high horsepower vehicles. If we were to model unrealistic pathways to compliance, manufacturers would incur more cost, and/or see less efficiency improvement than we estimate for any given level of CAFE standards, resulting in a standard that is more stringent than maximum feasible. For this and other reasons we endeavor to model our best estimates of a low-cost pathway to compliance.

We conducted a sensitivity case in which we removed all HCR restrictions, which is titled “Limited HCR skips” and is described in more detail in Chapter 9.2.2.4 of the RIA. By MY 2031 in this sensitivity case, we see a 7.5% increase in HCR technology penetration, but it corresponds with an additional 3 billion gallons of gasoline and 27 million metric tons more CO2 when compared to the reference baseline. The limited HCR skips sensitivity has a total social cost that is $500 million less than the reference baseline, however, the 2.50% discount rate of the net social benefits is $100 million more than the reference baseline. This sensitivity shows that without the HCR restrictions we use more gasoline and we do not see an appreciable societal benefit. With that, and in lieu of no new developments in HCR technology we have left our HCR restrictions in place for the final rule but will continue to monitor and assess the technology for future rulemakings.

How effective an engine technology is at improving a vehicle's fuel economy depends on several factors such as the vehicle's technology class and any additional technology that is being added or removed from the vehicle in conjunction with the new engine technology, as discussed in Section III.C, above. The Autonomie model's full vehicle simulation results provide most of the effectiveness values that we use as inputs to the CAFE Model. For a full discussion of the Autonomie modeling see Chapter 2.4 of the TSD and the CAFE Analysis Autonomie Documentation. The Autonomie modeling uses engine map models as the primary inputs for simulating the effects of different engine technologies.

Engine maps provide a three-dimensional representation of engine performance characteristics at each engine speed and load point across the operating range of the engine. Engine maps have the appearance of topographical maps, typically with engine speed on the horizontal axis and engine torque, power, or BMEP on the vertical axis. A third engine characteristic, such as brake-specific fuel consumption (BSFC), is displayed using contours overlaid across the speed and load map. The contours provide the values for the third characteristic in the regions of operation covered on the map. Other characteristics typically overlaid on an engine map include engine emissions, engine efficiency, and engine power. We refer to the engine maps developed to model the behavior of the engines in this analysis as engine map models.

The engine map models we use in this analysis are representative of technologies that are currently in production or are expected to be available in the rulemaking timeframe. We develop the engine map models to be representative of the performance achievable across industry for a given technology, and they are not intended to represent the performance of a single manufacturer's specific engine. We target a broadly representative performance level because the same combination of technologies produced by different manufacturers will have differences in performance, due to manufacturer-specific designs for engine hardware, control software, and emissions calibration. Accordingly, we expect that the engine maps developed for this analysis will differ from engine maps for manufacturers' specific engines. However, we intend and expect that the incremental changes in performance modeled for this analysis, due to changes in technologies or technology combinations, will be similar to the incremental changes in performance observed in manufacturers' engines for the same changes in technologies or technology combinations.

IAV developed most of the LD engine map models we use in this analysis. IAV is one of the world's leading automotive industry engineering service partners with an over 35-year history of performing research and development for powertrain components, electronics, and vehicle design. Southwest Research Institute (SwRI) developed the LD diesel and HDPUV engine maps for this analysis. SwRI has been providing automotive science, technology, and engineering services for over 70 years. Both IAV and SwRI developed our engine maps using the GT-POWER© Modeling tool (GT-POWER). GT-POWER is a commercially available, industry standard, engine performance simulation tool. GT-POWER can be used to predict detailed engine performance characteristics such as power, torque, airflow, volumetric efficiency, fuel consumption, turbocharger performance and matching, and pumping losses.

Just like Argonne optimizes a single vehicle model in Autonomie following the addition of a singular technology to the vehicle model, our engine map models were built in GT-POWER by incrementally adding engine technology to an initial engine—built using engine test data, component test data, and manufacturers' and suppliers' technical publications—and then optimizing the engine to consider real-world constraints like heat, friction, and knock. One of the basic assumptions we make when developing our engine maps is using 87 octane gasoline because it is the most common octane rating engines are designed to operate on and it is going to be the test fuel manufacturers will have to use for EPA fuel economy testing. We use a small number of initial engine configurations with well-defined BSFC maps, and then, in a very systematic and controlled process, add specific well-defined technologies to optimize a BSFC map for each unique technology combination. This could theoretically be done through engine or vehicle testing, but we would need to conduct tests on a single engine, and each configuration would require physical parts and associated engine calibrations to assess the impact of each technology configuration, which is impractical for the rulemaking analysis because of the extensive design, prototype part fabrication, development, and laboratory resources that are required to evaluate each unique configuration. We and the automotive industry use modeling as an approach to assess an array of technologies with more limited testing. Modeling offers the opportunity to isolate the effects of individual technologies by using a single or small number of initial engine configurations and incrementally adding technologies to those initial configurations. This provides a consistent reference point for the BSFC maps for each technology and for combinations of technologies that enables us to carefully identify and quantify the differences in effectiveness among technologies.

We received several comments regarding the use and benefits of high-octane and low carbon fuels in our analysis. The Missouri Corn Growers Association commented, “[t]he proposed rule, along with NHTSA's larger policy vision around vehicles ignores the widely diverse range of powertrain and liquid fuel options that could be more widely deployed to improve energy conservation . . . .” They go on to discuss the benefits of high-octane low carbon ethanol blended fuels and when combined with higher technology engines. Both the Alliance for Vehicle Efficiency and the Defour Group had similar comments on high octane low carbon fuels, particularly when used with HCR technology.

While we agree that a higher-octane fuel can work to improve engine fuel efficiency, we do not include it in our analysis. Our engine maps were developed with the use of 87 octane Tier 3 fuel, which represents the most commonly available fuel used by consumers. As we have stated previously, regulation of fuels is outside the scope of NHTSA's authority. Accordingly, we made no updates to the fuel assumed used in the engine map models.

Before use in the Autonomie analysis, both IAV and SwRI validated the generated engine maps against a global database of benchmarked data, engine test data, single cylinder test data, prior modeling studies, technical studies, and information presented at conferences. IAV and SwRI also validated the effectiveness values from the simulation results against detailed engine maps produced from the Argonne engine benchmarking programs, as well as published information from industry and academia. This ensures reasonable representation of simulated engine technologies. Additional details and assumptions that we use in the engine map modeling are described in detail in Chapter 3.1 of the TSD and the CAFE Analysis Autonomie Model Documentation chapter titled “Autonomie—Engine Model.”

Note that we never apply absolute BSFC levels from the engine maps to any vehicle model or configuration for the rulemaking analysis. We only use the absolute fuel economy values from the full vehicle Autonomie simulations to determine incremental effectiveness for switching from one technology to another technology. The incremental effectiveness is then applied to the absolute fuel economy or fuel consumption value of vehicles in the analysis fleet, which are based on CAFE or FE compliance data. For subsequent technology changes, we apply incremental effectiveness changes to the absolute fuel economy level of the previous technology configuration. Therefore, for a technically sound analysis, it is most important that the differences in BSFC among the engine maps be accurate, and not the absolute values of the individual engine maps.

While the fuel economy improvements for most engine technologies in the analysis are derived from the database of Autonomie full-vehicle simulation results, the analysis incorporates a handful of what we refer to as analogous effectiveness values. We use these when we do not have an engine map model for a particular technology combination. To generate an analogous effectiveness value, we use data from analogous technology combinations for which we do have engine map models and conduct a pairwise comparison to generate a data set of emulated performance values for adding technology to an initial application. We only use analogous effectiveness values for four technologies that are all SOHC technologies. We determined that the effectiveness results using these analogous effectiveness values provided reasonable results. This process is discussed further in Chapter 3.1.4.2 of the TSD.

The engine technology effectiveness values for all vehicle technology classes can be found in Chapter 3.1.4. of the TSD. These values show the calculated improvement for upgrading only the listed engine technology for a given combination of other technologies. In other words, the range of effectiveness values seen for each specific technology ( e.g., TURBO1) represents the addition of the TURBO1 technology to every technology combination that could select the addition of TURBO1.

These values are derived from the Argonne Autonomie simulation dataset and the righthand side Y-axis shows the number of Autonomie simulations that achieve each percentage effectiveness improvement point. The dashed line and grey shading indicate the median and 1.5X interquartile range (IQR), which is a helpful metric to use to identify outliers. Comparing these histograms to the box and whisker plots presented in prior CAFE program rule documents, it is much easier to see that the number of effectiveness outliers is extremely small.

We received a comment from the International Council on Clean Transportation (ICCT) regarding the application of the engine sizing algorithm, and when it is applied in relation to vehicle road load improvement technologies. ICCT stated that, “NHTSA continues to only downsize engines for large changes in tractive load,” which they assume artificially increases the overall performance of the fleet. These are incorrect assumptions and chapter 2.3.4 of the TSD discusses our approach of sizing powertrains by iteratively going through both low and high speed acceleration performance loops and adjusting powertrain size as needed based on the performance neutrality requirements.

We disagree with the comment implying that engine resizing is required for every technology change on a vehicle platform. We believe that this would artificially inflate effectiveness relative to cost. Manufacturers have repeatedly and consistently conveyed that the costs for redesign and the increased manufacturing complexity resulting from continual resizing engine displacement for small technology changes preclude them from doing so. NHTSA believes that it would not be reasonable or cost-effective to expect resizing powertrains for every unique combination of technologies, and even less reasonable and cost-effective for every unique combination of technologies across every vehicle model due to the extreme manufacturing complexity that would be required to do so. In addition, a 2011 NAS report stated that “[f]or small (under 5 percent [of curb weight]) changes in mass, resizing the engine may not be justified, but as the reduction in mass increases (greater than 10 percent [of curb weight]), it becomes more important for certain vehicles to resize the engine and seek secondary mass reduction opportunities.”

We also believe that ICCT's comment regarding Autonomie's engine resizing process is further addressed by Autonomie's powertrain calibration process. We do agree that the powertrain should be re-calibrated for every unique technology combination and this calibration is performed as part of the transmission shift initializer routine. Autonomie runs the shift initializer routine for every unique Autonomie full vehicle model configuration and generates customized transmission shift maps. The algorithms' optimization is designed to balance minimization of energy consumption and vehicle performance.

ICCT also submitted a comment regarding the validity of the continued use of our engine map models. ICCT stated that, “[a]lthough NHTSA scales its MY2010 hybrid Atkinson engine map to match the thermal efficiency of the MY2017 Toyota Prius, this appears to have been the only update made to the several engine maps that underpin all base and advanced engine technologies. The remaining engine maps are still primarily based on outdated engines ( e.g., from MY2011, 2013 and 2014 vehicles). Even with the updated hybrid engine, the newest Toyota Prius demonstrates an additional 10% improvement over the outgoing variant, due in part to improvements in engine efficiency.” ICCT also took issue with NHTSA not using two of EPA's engine map models, and for the perceived lack of effectiveness benefit for adding cylinder deactivation technology to turbocharged and HCR engines.

We disagree with statements that our engine maps are outdated. Many of the engine maps were developed specifically to support analysis for the current rulemaking timeframe. The engine map models encompass engine technologies that are present in the analysis fleet and technologies that could be applied in the rulemaking timeframe. In many cases those engine technologies are mainstream today and will continue to be during the rulemaking timeframe. For example, the engines on some MY 2022 vehicles in the analysis fleet have technologies that were initially introduced ten or more years ago. Having engine maps representative of those technologies is important for the analysis. The most basic engine technology levels also provide a useful consistent starting point for the incremental improvements for other engine technologies. The timeframe for the testing or modeling is unimportant because time by itself doesn't impact engine map data. A given engine or model will produce the same BSFC map regardless of when testing or modeling is conducted. Simplistic discounting of engine maps based on temporal considerations alone could result in discarding useful technical information.

We also disagree with ICCT's example that our hybrid engine map models are outdated and have even been provided comments that our hybrid effectiveness values exceed reasonable thermal efficiency. This is further discussed in the III.D.3 of this preamble. Finally, we responded to ICCT's criticisms that we did not employ EPA's engine map models in the 2020 final rule for MYs 2021-2026 standards, where we showed that our modeled engines provided similar incremental effectiveness values as the EPA engine map models. As far as we are aware, ICCT has not provided additional information showing that our engine map models are not reasonably similar to (if not providing a better effectiveness improvement than, in the case of the benchmarked Honda engine) EPA's engine map models.

Finally, in regard to engine effectiveness modeling, ICCT commented that “[t]he modeled benefit of adding cylinder deactivation (DEAC) to turbocharged and HCR engines appears to be only about 25% of the benefit of adding DEAC to the base engine. While DEAC added to turbo or HCR engines will have lower pumping loss reductions than when added to base naturally aspirated engines, DEAC can still be expected to provide significant pumping loss reductions while enabling the engine to operate in a more thermally efficient region of the engine map.”

In the NPRM we gave an example of the effects of adding DEAC to a turbocharged engine and discussed more about how fuel-efficient technologies have complex interactions and the effectiveness values of technology cannot be simply added together. Turbocharging and DEAC both work to reduce engine pumping losses and when working together they often provide a fuel-efficiency improvement greater then when they are working independently; however, much of these improvement happen in the same regions of engine operation where one or the other technology has a dominate effect which overshadows the benefits of the other. In other words, the benefits of the technologies are overlapping in the similar regions where the engine operates. These complex interactions among technologies are captured in our engine modeling.

The engine costs in our analysis are the product of engine DMCs, RPE, the LE, and updating to a consistent dollar year. We sourced engine DMCs from multiple sources, but primarily from the 2015 NAS report. For VTG and VTGE technologies ( i.e., Miller Cycle), we used cost data from a FEV technology cost assessment performed for ICCT, aggregated using individual component and system costs from the 2015 NAS report. We considered costs from the 2015 NAS report that referenced a Northeast States Center for a Clean Air Future (NESCCAF) 2004 report, but believe the reference material from the FEV report provides more updated cost estimates for the VTG technology.

All engine technology costs start with a base engine cost, and then additional technology costs are based on cylinder and bank count and configuration; the DMC for each engine technology is a function of unit cost times either the number of cylinders or number of banks, based on how the technology is applied to the system. The total costs for all engine technologies in all MYs across all vehicle classes can be found in the Technologies Input file.

2. Transmission Paths

Transmissions transmit torque generated by the engine from the engine to the wheels. Transmissions primarily use two mechanisms to improve fuel efficiency: (1) a wider gear range, which allows the engine to operate longer at higher efficiency speed-load points; and (2) improvements in friction or shifting efficiency ( e.g., improved gears, bearings, seals, and other components), which reduce parasitic losses.

We only model automatic transmissions in both the LD and HDPUV analyses. The four subcategories of automatic transmissions that we model in the LD analysis include traditional automatic transmissions (AT), dual clutch transmissions (DCT), continuously variable transmissions (CVT and eCVT), and direct drive (DD) transmissions. We also include high efficiency gearbox (HEG) technology improvements as options to the transmission technologies (designated as L2 or L3 in our analysis to indicate level of technology improvement). There has been a significant reduction in manual transmissions over the years and they made up less than 1% of the vehicles produced in MY 2022. Due to the trending decline of manual transmissions and their current low production volumes, we have removed manual transmissions from this analysis and have assigned vehicles using manual transmissions as DCTs in the analysis fleet.

We only model ATs in the HDPUV analysis because, except for DD transmissions that are only included as part of an electrified drivetrain, all HDPUV fleet analysis vehicles use ATs. In addition, from an engineering standpoint, DCTs and CVTs are not suited for HDPUV work requirements, as discussed further below. The HDPUV automatic transmissions work in the same way as the LD ATs and are labeled the same, but they are sized and mapped, in the Autonomie effectiveness modeling, to account for the additional work, durability, and payload these vehicles are designed to conduct. The HDPUV transmissions are sized with larger clutch packs, higher hydraulic line pressures, different shift schedules, larger torque converter and different lock up logic, and stronger components when compared to their LD counterparts. Chapter 3.2.1 of the TSD discusses the technical specifications of the four different AT subtypes in more detail. The LD and HDPUV transmission technology paths are shown in Chapter 3.2.3 of the TSD.

To assign transmission technologies to vehicles in the analysis fleets, we identify which Autonomie transmission model is most like a vehicle's real-world transmission, considering the transmission's configuration, costs, and effectiveness. Like with engines, we use manufacturer CAFE compliance submissions and publicly available information to assign transmissions to vehicles and determine which platforms share transmissions. To link shared transmissions in a manufacturer's fleet, we use transmission codes that include information about the manufacturer, drive configuration, transmission type, and number of gears. Just like manufacturers share transmissions in multiple vehicles, the CAFE Model will treat transmissions as “shared” if they share a transmission code and transmission technologies will be adopted together.

While identifying an AT's gear count is fairly easy, identifying HEG levels for ATs and CVTs is more difficult. We reviewed the age of the transmission design, relative performance versus previous designs, and technologies incorporated to assign an HEG level. There are no HEG Level 3 automatic transmissions in either the LD or the HDPUV analysis fleets. For the LD analysis we found all 7-speed, all 9-speed, all 10-speed, and some 8-speed automatic transmissions to be advanced transmissions operating at HEG Level 2 equivalence. We assigned eight-speed automatic transmissions and CVTs newly introduced for the LD market in MY 2016 and later as HEG Level 2. All other automatic transmissions are assigned to their respective transmission's initial technology level ( i.e., AT6, AT8, and CVT). For DCTs, the number of gears in the assignments usually match the number of gears listed by the data sources, with some exceptions (we assign dual-clutch transmissions with seven and nine gears to DCT6 and DCT8 respectively). We assigned vehicles in either the LD or HDPUV analyses fleets with a fully electric powertrain a DD transmission. We assigned any vehicle in the LD analysis fleet with a power-split hybrid (SHEVPS) powertrain an electronic continuously variable transmission (eCVT). Finally, we assigned the limited number of manual transmissions in the LD fleet as DCTs, as we did not model manual transmissions in Autonomie for this analysis.

Most transmission adoption features are instituted through technology path logic ( i.e., decisions about how less advanced transmissions of the same type can advance to more advanced transmissions of the same type). Technology pathways are designed to prevent “branch hopping”—changes in transmission type that would correspond to significant changes in transmission architecture—for vehicles that are relatively advanced on a given pathway. For example, any automatic transmission with more than five gears cannot move to a dual-clutch transmission. We also prevent “branch hopping” as a proxy for stranded capital, which is discussed in more detail in Section III.C and Chapter 2.6 of the TSD.

For the LD analysis, the automatic transmission path precludes adoption of other transmission types once a platform progresses past an AT8. We use this restriction to avoid the significant level of stranded capital loss that could result from adopting a completely different transmission type shortly after adopting an advanced transmission, which would occur if a different transmission type were adopted after AT8 in the rulemaking timeframe. Vehicles that did not start out with AT7L2 transmissions cannot adopt that technology in the model. It is likely that other vehicles will not adopt the AT7L2 technology, as vehicles that have moved to more advanced automatic transmissions have overwhelmingly moved to 8-speed and 10-speed transmissions.

CVT adoption is limited by technology path logic and is only available in the LD fleet analysis and therefore, not in the technology path for the HDPUV analysis. Vehicles that do not originate with a CVT or vehicles with multispeed transmissions beyond AT8 in the analysis fleet cannot adopt CVTs. Vehicles with multispeed transmissions greater than AT8 demonstrate increased ability to operate the engine at a highly efficient speed and load. Once on the CVT path, the platform is only allowed to apply improved CVT technologies. Due to the limitations of current CVTs, discussed in TSD Chapter 3.2, this analysis restricts the application of CVT technology on LDVs with greater than 300 lb.-ft of engine torque. This is because of the higher torque (load) demands of those vehicles and CVT torque limitations based on durability constraints. We believe the 300 lb.-ft restriction represents an increase over current levels of torque capacity that is likely to be achieved during the rule making timeframe. This restriction aligns with CVT application in the analysis fleet, in that CVTs are only witnessed on vehicles with under 280 lb.-ft of torque. Additionally, this restriction is used to avoid stranded capital. Finally, the analysis allows vehicles in the analysis fleet that have DCTs to apply an improved DCT and allows vehicles with an AT5 to consider DCTs. Drivability and durability issues with some DCTs have resulted in a low relative adoption rate over the last decade. This is also broadly consistent with manufacturers' technology choices. DCTs are not a selectable technology for the HDPUV analysis.

Autonomie models transmissions as a sequence of mechanical torque gains. The torque and speed are multiplied and divided, respectively, by the current ratio for the selected operating condition. Furthermore, torque losses corresponding to the torque/speed operating point are subtracted from the torque input. Torque losses are defined based on a three-dimensional efficiency lookup table that has the following inputs: input shaft rotational speed, input shaft torque, and operating condition. We populate transmission template models in Autonomie with characteristics data to model specific transmissions. Characteristics data are typically tabulated data for transmission gear ratios, maps for transmission efficiency, and maps for torque converter performance, as applicable. Different transmission types require different quantities of data. The characteristics data for these models come from peer-reviewed sources, transmission and vehicle testing programs, results from simulating current and future transmission configurations, and confidential data obtained from OEMs and suppliers. We model HEG improvements by modeling improvements to the efficiency map of the transmission. As an example, the AT8 model data comes from a transmission characterization study. The AT8L2 has the same gear ratios as the AT8, however, we improve the gear efficiency map to represent application of the HEG level 2 technologies. The AT8L3 models the application of HEG level 3 technologies using the same principle, further improving the gear efficiency map over the AT8L2 improvements. Each transmission (15 for the LD analysis and 6 for the HDPUV analysis) is modeled in Autonomie with defined gear ratios, gear efficiencies, gear spans, and unique shift logic for the technology configuration the transmission is applied to. These transmission maps are developed to represent the gear counts and span, shift and torque converter lockup logic, and efficiencies that can be seen in the fleet, along with upcoming technology improvements, all while balancing key attributes such as drivability, fuel economy, and performance neutrality. This modeling is discussed in detail in Chapter 3.2 of the TSD and the CAFE Analysis Autonomie Documentation chapter titled “Autonomie—Transmission Model.”

The effectiveness values for the transmission technologies, for all LD and HDPUV technology classes, are shown in Chapter 3.2.4 of the TSD. Note that the effectiveness for the AT5, eCVT, and DD technologies is not shown. The DD and eCVT transmissions do not have standalone effectiveness values because those technologies are only implemented as part of electrified powertrains. The AT5 has no effectiveness values because it is a reference-point technology against which all other transmission technologies are compared.

Our transmission DMCs come from the 2015 NAS report and studies cited therein. The LD costs are taken almost directly from the 2015 NAS report adjusted to the current dollar year or for the appropriate number of gears. We applied a 20% cost increase for HDPUV transmissions based on comparing the additional weight, torque capacity, and durability required in the HDPUV segment. Chapter 3.2 of the TSD discusses the specific 2015 NAS report costs used to generate our transmission cost estimates, and all transmission costs across all MYs can be found in CAFE Model's Technologies Input file. We have used the 2015 NAS report transmission costs for the last several LD CAFE Model analyses (since reevaluating all transmission costs for the 2020 final rule) and have received no comments or feedback on these costs. We again sought comment on our approach to estimating all transmission costs, but in particular on HDPUV transmission costs for this analysis, in addition to any publicly available data from manufacturers or reports on the cost of HDPUV transmissions. We received no comments or feedback on these costs, so we continue to use the NPRM estimates for the analysis supporting this final rule.

3. Electrification Paths

The electrification paths include a set of technologies that share common electric powertrain components, like batteries and electric motors, for certain vehicle functions that were traditionally powered by combustion engines. While all vehicles (including conventional ICE vehicles) use batteries and electric motors in some form, some component designs and powertrain architectures contribute to greater levels of electrification than others, allowing the vehicle to be less reliant on gasoline or other fuel.

Several stakeholders commented about general topics related to electrification technologies like the perceived merits or disadvantages of electric vehicles, OEM investments in electric vehicles, and infrastructure and supply chain considerations around electric vehicles. Additional comments stated that hybrids are “popular, cost effective” and that dozens of new electric vehicle models having reached “twice as many as before the pandemic” with highly efficient electric vehicle technology that “is scalable and increasingly accessible.” Stakeholders stated that “[n]early every automaker has publicly committed to transitioning model line-ups to new technologies with substantially less fuel consumption” and more electrified vehicles will enter the market “with the goal of making these mobility options more accessible for everyone . . . offering a diverse portfolio of EVs to meet varying customer needs.” Insofar as our electrification technology penetration rates reach into the rulemaking timeframe, several other commenters stated that our future electrification penetration rates are not realistic due to limitations/uncertainty with battery material acquisition, manufacturing/production, and the current state of infrastructure and are expecting PHEVs to “play a more prominent role over the near to mid-term.” On the other hand, ICCT stated that our penetration rates of electrification technologies in the no action and action alternatives “are reasonable and feasible.”

NHTSA thanks commenters for expressing their opinions and submitting relevant data on topics surrounding electrification technology adoption. We endeavor to reasonably model technologies that manufacturers use to respond to our standards, other government standards, and consumer preferences, and we believe that the inputs and assumptions that we selected to represent electrification technologies results in reasonable outcomes. The grounds for building the foundation to determine appropriate electrification technology effectiveness and cost values (therefore resulting in appropriate technology penetration rates) as these technologies affect the reference baseline and out years was based on numerous well-thought-out inputs and assumptions. Although time and resources limit consideration of each and every individual electrification technology, NHTSA focused on key inputs and assumptions ( e.g., the costs of batteries and applicability of specific electrified technologies for vehicles that do extensive work in the HDPUV fleet) to provide reasonable results for compliance pathways. While we recognize that stakeholders identified issues that they believed to be impediments to electrification technology adoption in particular fleets or market segments, we feel confident that we took the appropriate approach to determining the technologies applicable for vehicles in this analysis and that we capture many of these considerations explicitly in the analysis or qualitatively in additional technical support for this final rule. We have provided details of the inputs and assumptions in the TSD accompanying this final rule and provided more information to support our responses to comments throughout Section II and III of this preamble.

Unlike with other technologies in the analysis, including other electrification technologies, Congress placed specific limitations on how we consider the fuel economy of alternative fueled vehicles (such as PHEVs, BEVs, and FCEVs) when setting CAFE standards. We implement these restrictions in the CAFE Model by using fuel economy values that assume “charge sustaining” (gasoline-only) PHEV operation, and by restricting technologies that convert a vehicle to a BEV or a FCEV from being applied during “standard-setting” years. However, there are several reasons why we must still accurately model PHEVs, BEVs, and FCEVs in the analysis; these reasons are discussed in detail throughout this preamble and, in particular, in Sections IV and VI. In brief: we must consider the existing fleet fuel economy level in calculating the maximum feasible fuel economy level that manufacturers can achieve in future years. Accurately calculating the pre-existing fleet fuel economy level is crucial because it marks the starting point for determining what further efficiency gains will be feasible during the rulemaking timeframe. As discussed in detail above and in TSD Chapter 2.2, PHEVs, BEVs, and FCEVs currently exist in manufacturer's fleets and count towards manufacturer's reference baseline compliance fuel economy values.

In addition to accurately capturing an analysis, or initial, fleet of vehicles in a given MY, we must capture a regulatory “no action” reference baseline in each MY; that is, the regulatory reference baseline captures what the world will be like if our rule is not adopted, to accurately capture the costs and benefits of CAFE standards. The “no-action” reference baseline includes our representation of the existing fleet of vehicles ( i.e., the LD and HDPUV analysis fleets) and (with some restrictions) our representation of manufacturer's fleets in the absence of our standards. Specifically, we assumed that in the absence of LD CAFE and HDPUV FE standards, manufacturers will produce certain BEVs to comply with California's ACC I and ACT program. We further assumed, consistent with manufacturer comments, that they will (regardless of legal requirements) produce additional BEVs consistent with the levels that would be required by California's ACC II program, were it to be granted a Clean Air Act preemption waiver. Accounting for electrified vehicles that manufacturers produced in response to state regulatory requirements or will produce for their own reasons improves the accuracy of the analysis of the costs and benefits of additional technology added to vehicles in response to CAFE standards, while adhering to the statutory prohibition against considering the fuel economy gains that could be achieved if manufacturers create new dedicated automobiles to comply with the CAFE standards.

Next, the costs and benefits of CAFE standards do not end in the MYs for which we are setting standards. Vehicles produced in standard-setting years, e.g., MYs 2027 through MY 2031 in this analysis, will continue to have effects for years after they are produced as the vehicles are sold and driven. To accurately capture the costs and benefits of vehicles subject to the standards in future years, the CAFE Model projects compliance through MY 2050. Outside of the standard-setting years, we model the extent to which manufacturers could produce electrified vehicles, in order to improve the accuracy and realism of our analysis in situations where statute does not prevent us from doing so. Finally, due to NEPA requirements, we do consider the effects of electrified vehicle adoption in the CAFE Model under a “real-world” scenario where we lift EPCA/EISA's restrictions on our decision-making. On the basis of our NEPA analysis, we can consider the actual environmental impacts of our actions in the decision-making process, subject to EPCA's constraints.

For those reasons, we must still accurately model electrified vehicles. That said, PHEVs, BEVs, and FCEVs only represent a portion of the electrified technologies that we include in the analysis. We discuss the range of modeled electrified technologies below and in detail in Chapter 3.3.1 of the TSD.

Among the simpler configurations with the fewest electrification components, micro HEV technology (SS12V) uses a 12-volt system that simply restarts the engine from a stop. Mild HEVs use a 48-volt belt integrated starter generator (BISG) system that restarts the engine from a stop and provides some regenerative braking functionality. Mild HEVs are often also capable of minimal electric assist to the engine on take-off.

Strong hybrid-electric vehicles (SHEVs) have higher system voltages compared to mild hybrids with BISG systems and are capable of engine start/stop, regenerative braking, electric motor assist of the engine at higher speeds, and power demands with the ability to provide limited all-electric propulsion. Common SHEV powertrain architectures, classified by the interconnectivity of common electrified vehicle components, include both a series-parallel architecture by power-split device (SHEVPS) as well as a parallel architecture (P2). P2s—although enhanced by the electrification components, including just one electric motor—remains fundamentally similar to a conventional powertrain. In contrast, SHEVPS is considerably different than a conventional powertrain; SHEVPSs use two electric motors, which allows the use of a lower-power-density engine. This results in a higher potential for fuel economy improvement compared to a P2, although the SHEVPS' engine power density is lower. Or, put another way, “[a] disadvantage of the power split architecture is that when towing or driving under other real-world conditions, performance is not optimum.” In contrast, “[o]ne of the main reasons for using parallel hybrid architecture is to enable towing and meet maximum vehicle speed targets.” This is an important distinction to understand why we allow certain types of vehicles to adopt P2 powertrains and not SHEVPS powertrains, and to understand why we include only P2 strong hybrid architectures in the HDPUV analysis. Both concepts are discussed further below.

Plug-in hybrids (PHEVs) utilize a combination gasoline-electric powertrain, like that of a SHEV, but have the ability to plug into the electric grid to recharge the battery, like that of a BEV; this contributes to all-electric mode capability in both blended and non-blended PHEVs. The analysis includes PHEVs with an all-electric range (AER) of 20 and 50 miles to encompass the range of PHEV AER in the market today. BEVs have an all-electric powertrain and use only batteries for the source of propulsion energy. BEVs with ranges of 200 to more than 350 miles are used in the analysis. Finally, FCEVs are another form of electrified vehicle that have a fully electric powertrain that uses a fuel cell system to convert hydrogen fuel into electrical energy. See TSD Chapter 3.3 for more information on every electrification technology considered in the analysis, including its acronym and a brief description. For brevity, we refer to technologies by their acronyms in this section.

Readers familiar with previous LD CAFE analyses will notice that we have increased the number of engine options available for strong hybrid-electric vehicles and plug-in hybrid-electric vehicles. As discussed above, this better represents the diversity of different hybrid architectures and engine options available in the real world for SHEVs and PHEVs, while still maintaining a reasonable level of analytical complexity. In addition, we now refer to the BEV options as BEV1, BEV2, BEV3, and BEV4, rather than by their range assignments as in the previous analysis, to accommodate using the same model code for the LD and HDPUV analyses. Note that BEV1 and BEV2 have different range assignments in the LD and HDPUV analyses; further, within the HDPUV fleet, different range assignments exist for HD pickups and HD vans.

In the CAFE Model, HDPUVs only have one SHEV option and one PHEV option. The P2 architecture supports high payload and high towing requirements versus other types of hybrid architecture, which are important considerations for HDPUV commercial operations. The mechanical connection between the engine, transmission, and P2 hybrid systems enables continuous power flow to be able to meet high towing weights and loads at the cost of system efficiency. We do not allow engine downsizing in this setup in so that when the battery storage system is depleted, the vehicle is still able to operate while achieving its original performance. We picked the P2 architecture for HDPUV SHEVs because, although there are currently no SHEV HDPUVs in the market on which to base a technology choice, we believe that the P2 strong hybrid architecture would more likely be picked than other architecture options, such as ones with power-split powertrains. This is because, as discussed above, the P2 architecture “can be integrated with existing conventional powertrain systems that already meet the additional attribute requirements of these large vehicle segments.”

We only include one HDPUV PHEV option as there are no PHEVs in the HDPUV analysis fleet, and there are no announcements from major manufacturers that indicate this a pathway that they will pursue in the short term ( i.e., the next few years). We believe this is in part because PHEVs, which are essentially two separate powertrains combined, can decrease HDPUV capability by increasing the curb weight of the vehicle and reducing cargo capacity. A manufacturer's ability to use PHEVs in the HDPUV segment is highly dependent on the load requirements and the duty cycle of the vehicle. However, in the right operation, HDPUV PHEVs can have a cost-effective advantage over their conventional counterparts. More specifically, there would be a larger fuel economy benefit the more the vehicle could rely on its electric operation, with partial help from the ICE; examples of duty cycles where this would be the case include short delivery applications or construction trucks that drive between work sites in the same city. Accordingly, we do think that PHEVs can be a technology option for adoption in the rulemaking timeframe. We picked a 50-mile AER for this segment based on discussions with experts at Argonne, who were also involved in DOE projects and provided guidance for this segment. Additional information about each technology we considered is located in Chapter 3.3.1 of the TSD.

We sought comment on the range of electrification path technologies and received comment from stakeholders regarding electrified powertrain options for both the light-duty and HDPUV fleets.

Two commenters repeatedly referenced a Roush report and suggested that we should include more-capable, higher output 48-volt mild hybrid systems beyond P0 mild hybrids in our modeling, such as “P2, P3, or P4 configurations” which offer additional benefits of “electric power take-offs” ( i.e., launch assist) or “slow-speed electric driving” on the vehicle's drive axle(s). It was also noted in comment that P2 mild hybrids mated with more advanced engine technologies have the ability to increase system efficiency.

We agree with the commenters that these mild hybrid configurations, such as P2 (mild) and P4, could offer better improvements compared to P0 mild hybrids. Non-P0 powertrains, however, require significant changes to the powertrain and would require a higher capacity battery—both leading to increase powertrain cost; this is similar to what we observed in past rulemakings with the (P1) CISG system, with the non-P0 mild hybrid not being a cost-effective way for manufacturers to meet standards in the rulemaking time frame. Accordingly, we did not include additional mild hybrid technology for this final rule but will consider mild hybrid advancements, such as P2 through P4, in future analysis if they become more prevalent in the U.S. market.

To extent possible, for any analyses conducted for any new rulemaking, we update as much of the technical aspects as possible with available data and time allotted. For example, we have significantly expanded our strong hybrid and plug-in hybrid offering for adopting in the rulemaking time frame, we have also updated our full vehicle modeling based on the testing of Toyota RAV4 Prime, Nissan Leaf, and Chevy Bolt, for HDPUV we worked with SwRI to develop a new engine map for P2 Hybrids.

We also received a handful of comments on technologies considered for the HDPUV analysis. ICCT commended “NHTSA for incorporating [hybrid technologies, including PHEVs] into its modeling of the HD pickup and van fleet.” We received related supportive comment on PHEVs for HDPUV from MECA stating, “[p]lug-in hybrids (PHEVs) can be practical for light and medium- duty trucks ( e.g., Class 1 through 3) that do not travel long distances or operate for long periods of time without returning to a central location.”

NHTSA appreciates the comment and MECA's technological insight. NHTSA thanks other commenters, such as ICCT, for support of our underlying assumptions and providing insight into technology trends.

Related to the electrified HDPUV fleet, AFPM stated that we “do not distinguish between the less costly lower range BEV1 and BEV2 options, and the much more costly and virtually unavailable higher range BEV3 and BEV4 options” for HDPUVs and that “NHTSA should adjust its modeling to fully assess the real feasibility (and cost) of the BEVs that commercial HDPUV fleet operators really need.”

We believe that AFPM misunderstood our proposal documents. As was clear in the NPRM and outlined in TSD Chapter 3.3, there are no BEV3 or BEV4 options for HDPUVs. This is because we ensure that BEVs (and all vehicles) are modeled to meet sizing and utility (such as towing and hauling) requirements as described in Autonomie Model Documentation. Additionally, we do not allow high towing capable vehicles to be fully converted BEVs as they have utility requirements that far exceed driving range of BEVs. These and other considerations of vehicle's capabilities and utility have been further discussed in the TSD Chapter 3.3. However, NHTSA disagrees with AFPM that BEV HDPUVs analyzed by NHTSA for this rule have a more limited carrying capacity than their ICE counterparts. NHTSA examined HDPUV BEV configurations in conjunction with Argonne and meetings with stakeholders prior to finalizing inputs for the CAFE Model analysis and does not believe that battery pack sizes will limit cargo capacity for HDPUVs (as opposed to what may be seen for larger MD/HD vehicles). This is especially true with the relatively lower total mileage ranges needed for HDPUV delivery vehicles, which generally operate in a more limited spatial area (as opposed again to the long-distance requirements and larger cargo area needed with larger MD/HD vehicles). To reflect these considerations, NHTSA only modeled two HDPUV range configurations for HDPUVs (termed “BEV1” and “BEV2”). NHTSA disagrees that we should adjust our HDPUV modeling as we have conducted analysis based on available data on technologies and capabilities of vehicles within the fleet but appreciates AFPM's comment nonetheless; NHTSA has not made any changes to electrification pathways in the model for HDPUVs for this rulemaking.

We received comment from Alliance for Vehicle Efficiency (AVE) relating to the inclusion of FCEVs in the analysis, stating that, “NHTSA dismisses [FCEV] chances for meaningful market penetration” and that they encourage “NHTSA to fully assess the fuel economy benefits that hydrogen vehicles could achieve and how these vehicles could become cost-effective solutions for manufacturers.” We disagree—not only have we assessed each powertrain technology specifically for this analysis (which includes FCEVs), our market penetration for FCEVs is aligned with market projections during the rulemaking time frame.

As described in TSD Chapter 3.3, we assigned electrification technologies to vehicles in the LD and HDPUV analysis fleets using manufacturer-submitted CAFE compliance information, publicly available technical specifications, marketing brochures, articles from reputable media outlets, and data from Wards Intelligence. TSD Chapter 3.3.2 shows the penetration rates of electrification technologies in the LD and HDPUV analysis fleets, respectively. Over half the LD analysis fleet has some level of electrification, with the vast majority—over 50 percent of the fleet—being micro hybrids; BEV3 (>275 miles; ≤350 miles) is the most common LD BEV technology. The HDPUV analysis fleet has only a conventional non-electrified powertrain, currently; however, the first year of HDPUV standards in this analysis is MY 2030, and we expect additional electrification technologies to be applied in the fleet before then. Like the other technology pathways, as the CAFE Model adopts electrification technologies for vehicles, more advanced levels of electrification technologies will supersede all prior levels, while certain technologies within each level are mutually exclusive. The only adoption feature applicable to micro (SS12V) and mild (BISG) hybrid technology is path logic; vehicles can only adopt micro and mild hybrid technology if the vehicle did not already have a more advanced level of electrification.

The adoption features that we apply to strong hybrid technologies include path logic, powertrain substitution, and vehicle class restrictions. Per the technology pathways, SHEVPS, P2x, P2TRBx, and the P2HCRx technologies are considered mutually exclusive. In other words, when the model applies one of these technologies, the others are immediately disabled from future application. However, all vehicles on the strong hybrid pathways can still advance to one or more of the plug-in technologies, when applicable in the modeling scenario ( i.e., allowed in the model).

When the model applies any strong hybrid technology to a vehicle, the transmission technology on the vehicle is superseded; regardless of the transmission originally present, P2 hybrids adopt an advanced 8-speed automatic transmission (AT8L2), and PS hybrids adopt a continuously variable transmission via power-split device (eCVT). When the model applies the P2 technology, the model can consider various engine options to pair with the P2 architecture according to existing engine path constraints—taking into account relative cost effectiveness. For SHEVPS technology, the existing engine is replaced with a full time Atkinson cycle engine. For P2s, we picked the 8-speed automatic transmission to supersede the vehicle's incoming transmission technology. This is because most P2s in the market use an 8-speed automatic transmission, therefore it is representative of the fleet now. We also think that 8-speed transmissions are representative of the transmissions that will continue to be used in these hybrid vehicles, as we anticipate manufacturers will continue to use these “off-the-shelf” transmissions based on availability and ease of incorporation in the powertrain. The eCVT (power-split device) is the transmission for SHEVPSs and is therefore the technology we picked to supersede the vehicle's prior transmission when adopting the SHEVPS powertrain.

SKIP logic is also used to constrain adoption for SHEVPS and PHEV20/50PS technologies. These technologies are “skipped” for vehicles with engines that meet one of the following conditions: the engine belongs to an excluded manufacturer; the engine belongs to a pickup truck ( i.e., the engine is on a vehicle assigned the “pickup” body style); the engine's peak horsepower is more than 405 hp; or if the engine is on a non-pickup vehicle but is shared with a pickup. The reasons for these conditions are similar to those for the SKIP logic that we apply to HCR engine technologies, discussed in more detail in Section III.D.1. In the real world, performance vehicles with certain powertrain configurations cannot adopt the technologies listed above and maintain vehicle performance without redesigning the entire powertrain.

It may be helpful to understand why we do not apply SKIP logic to P2s and to understand why we do apply SKIP logic to SHEVPSs. Remember the difference between P2 and SHEVPS architectures: P2 architectures are better for “larger vehicle applications because they can be integrated with existing conventional powertrain systems that already meet the additional attribute requirements” of large vehicle segments. No SKIP logic applies to P2s because we believe that this type of electrified powertrain is sufficient to meet all of the performance requirements for all types of vehicles. Manufacturers have proven this now with vehicles like the Ford F-150 Hybrid and Toyota Tundra Hybrid. In contrast, “[a] disadvantage of the power split architecture is that when towing or driving under other real-world conditions, performance is not optimum.” If we were to size (in the Autonomie simulations) the SHEVPS motors and engines to achieve not “not optimum” performance, the electric motors would be unrealistically large (on both a size and cost basis), and the accompanying engine would also have to be a very large displacement engine, which is not characteristic of how vehicle manufacturers apply SHEVPS ICEs in the real-world. Instead, for vehicle applications that have particular performance requirements—defined in our analysis as vehicles with engines that belong to an excluded manufacturer, engines belonging to a pickup truck or shared with a pickup truck, or the engine's peak horsepower is more than 405hp—those vehicles can adopt P2 architectures that should be able to handle the vehicle's performance requirements.

NHTSA received general comments from ICCT related to the strong hybrid technology pathway restrictions. ICCT suggested that the analysis should allow strong “hybridization on all vehicle types” in the analysis, without further elaboration on what of the above explanation they disagreed with or any technical justification for making their proposed change. To be clear, strong hybridization is allowed on all vehicle types. However, we allow different types of strong hybrid powertrains to be applied to different types of vehicles for the reasons discussed above. We believe that allowing SHEVPS and P2 powertrains to be applied subject to the base vehicle's performance requirements is a reasonable approach to maintaining a performance-neutral analysis.

LD PHEV adoption is limited only by technology path logic; however, in the HDPUV analysis, PHEV technology is not available in the model until MY 2025 for HD vans and MY 2027 for HD pickups. As discussed above, there are no PHEVs in the HDPUV analysis fleet and there are no announcements from major manufacturers that indicate this a pathway that they will pursue in the short term; that said, we do believe this is a technology that could be beneficial for very specific HDPUV applications. However, the technology is fully available for adoption by HDPUVs in the rulemaking timeframe ( i.e., MYs 2030 and beyond). We sought comment on this assumption, and any other information available from manufacturers or other stakeholders on the potential that original equipment manufacturers will implement PHEV technology prior to MY 2025 for HD vans, and prior to MY 2027 for HD pickups. We did not receive any specific comments on this request and so we finalized the NPRM assumptions for PHEV availability in the HDPUV fleet.

The engine and transmission technologies on a vehicle are superseded when PHEV technologies are applied. For example, the model applies an AT8L2 transmission with all PHEV20T/50T plug-in technologies, and the model applies an eCVT transmission for all PHEV20PS/50PS and PHEV20H/50H plug-in technologies in the LD fleet and for more details on different system combinations of electrification see TSD Chapter 3.3. A vehicle adopting PHEV20PS/50PS receives a hybrid full Atkinson cycle engine, and a vehicle adopting PHEV20H/PHEV50H receives an HCR engine. For PHEV20T/50T, the vehicle receives a TURBO1 engine.

Adoption of BEVs and FCEVs is limited by both path logic and phase-in caps. They are applied as end-of-path technologies that supersede previous levels of electrification. Phase-in caps, which are defined in the CAFE Model Input Files, are percentages that represent the maximum rate of increase in penetration rate for a given technology. They are accompanied by a phase-in start year, which determines the first year the phase-in cap applies. Together, the phase-in cap and start year determine the maximum penetration rate for a given technology in a given year; the maximum penetration rate equals the phase-in cap times the number of years elapsed since the phase-in start year. Note that phase-in caps do not inherently dictate how much a technology is applied by the model. Rather, they represent how much of the fleet could have a given technology by a given year.

Because a BEV1 costs less and has slightly higher effectiveness values than other advanced electrification technologies, the model will have vehicles adopt it first, until it is restricted by the phase-in cap. However, this only applies during non-standard setting years as well as when the analysis is simulated for the EIS. The standard setting simulations do not consider BEVs; thus, phase-in caps are not applicable throughout this timeframe. TSD Chapter 3.3.3 shows the phase-in caps, phase-in year, and maximum penetration rate through 2050 for BEV and FCEV technologies.

The LD BEV1 phase-in cap is informed by manufacturers' tendency to move away from low-range passenger vehicle offerings in part because of potential consumer concern with range anxiety. In some cases, the advertised range on EVs may not reflect the actual real-world range in cold and hot ambient temperatures and real-world driving conditions, affecting the utility of these lower range vehicles. Many manufacturers, including comments from General Motors, as discussed further below, have told us that the portion of consumers willing to accept a vehicle with the lowest modeled range is small, with manufacturers targeting range values well above BEV1 range.

Furthermore, the average BEV range has steadily increased over the past decade, due to battery technological progress increasing energy density as well as batteries becoming more cost effective. EPA observed in its 2023 Automotive Trends Report that “the average range of new EVs has climbed substantially. In MY 2022, the average new EV is 305 miles, or more than four times the range of an average EV in 2011.” Based on the cited examples and basis described in this section, the maximum growth rate for LD BEV1s in the model is set accordingly low to less than 0.1 percent per year. While this rate is significantly lower than that of the other BEV technologies, the BEV1 phase-in cap allows the penetration rate of low-range BEVs to grow by a multiple of what is currently observed in the market.

For higher BEV ranges (such as that for BEV2 for both LD and HDPUVs), phase-in caps are intended to conservatively reflect potential challenges in the scalability of BEV manufacturing and implementing BEV technology on many vehicle configurations, including larger vehicles. In the short term, the penetration of BEVs is largely limited by battery material acquisition and manufacturing. Incorporating battery packs with the capacity to provide greater electric range also poses its own engineering challenges. Heavy batteries and large packs may be difficult to integrate for many vehicle configurations and require vehicle structure modifications. Pickup trucks and large SUVs, in particular, require higher levels of stored energy as the number of passengers and/or payload increases, for towing and other high-torque applications. In the LD analysis, we use the LD BEV3 and BEV4 phase-in caps to reflect these transitional challenges. For HDPUV analysis, we use similar phase-in caps for the BEV1 and BEV2 to control for realities of adoption of electrified technologies in work vehicles.

Recall that BEV phase-in caps are a tool that we use in the simulations to allow the model to build higher-range BEVs (when the modeling scenario allows, as in outside of standard-setting years), because if we did not, the model would only build BEV1s, as they are the most cost-effective BEV technology. Based on the analysis provided above, we believe there is a reasonable justification for different BEV phase-in caps based on expected BEV ranges in the future. We sought comment on the BEV phase-in caps for the LD and HDPUV analyses, and we received comment from several stakeholders that asked us to reevaluate our phase-in caps for BEVs: one comment from General Motors asserted a specific issue with the penetration rates of short-range BEVs, stating, “[t]he agency assumes a very large portion of the market will adopt BEVs with less than 300-mile range” and that we should adjust “phase-in caps to recognize that 100% of the market is unlikely to adopt BEVs with 300 miles range or less.”

We have modified the values of our phase-in caps for LD BEVs, as shown above in TSD Chapter 3.3.3, to “produce more realistic compliance pathways that project higher shares of longer-range BEVs and restrict or eliminate the projection of shorter-range BEVs in some applications;” the broad LD phase-in cap values adjust shorter-range BEV prevelance in the fleet.

MEMA commented that phase-in caps constrain “the ability of the industry to pursue all compliance options” and “keep the production volume of BEV/FCEV technologies low.” It was suggested that a delayed launch of some technologies (like BEVs and FCEVs, when they're more advanced) would be more practical. Similarly, we also received comment from Valero on HDPUV phase-in caps for BEVs, which stated, “NHTSA sets phase-in caps at unrealistically high values that ignore the actual penetration rates in the 2022 baseline fleet. Furthermore, NHTSA's application of fleetwide phase-in caps fails to account for the unique penetration hurdles of each tech class within the HDPUV fleet—Van 2b, Van 3, Pickup 2b, and Pickup 3.”

NHTSA disagrees, in general, that phase-in caps are constraining, as these limitations are applied based on market availability, cost, and consumer acceptance in the rulemaking timeframe. Our internal research, discussions with stakeholders, and other outreach has led us to not be too optimistic on these crucial technologies, but we believe the phase-in caps represent a reasonable middle ground between allowing for the application of technology at reasonable levels. The details of phase-in caps are discussed this further in TSD Chapter 3.3.3.4.

NHTSA also disagrees with the argument that HDPUV BEV penetration from the underlying phase-in caps is unrealistic, for a few reasons. First, NHTSA's HDPUV HDPUV analysis fleet contains vehicles that span a range of model years prior to and including MY 2022 vehicles, based on the most up-to-date compliance data we had at the time of modeling. Between the earliest MY vehicle in the analysis fleet and the first MY for which we are setting standards, MY 2030, in the absence of phase-in caps, the model will pick a cost-effective pathway for compliance that manufacturers themselves may not have selected, and we want the years prior to the first analysis year to reasonably reflect reality. There are already annoucements of HDPUV BEV production and sales that are not captured in the HDPUV analysis fleet but can be observed in the analysis years. Second, as discussed further in Section VI, NHTSA understands that there could be uncertatinty in looking out eight to thirteen MYs in the future; this affects new vehicle technology adoption, and so we applied some conservatatism in setting phase-in caps. Finally, when applying technologies to the HDPUVs, we considered the applications of the vehicle and what could be the limiting factors in allowing more advanced technologies to apply. For example, we maintain the engine size when a vehicle adopts PHEV technologies, and we do not allow HD pickups with work factors greater than 7500 and higher than 500 mile range to adopt BEVs, further discussed in TSD Chapters 2.3.2 and 3.3. However, we understand unique technological barriers to each of the HDPUV vehicle types, and we will continue to monitor this space and consider updating the phase-in cap modeling approach in the future.

The phase-in cap for FCEVs is assigned based on existing market share as well as historical trends in FCEV production for LDVs and HDPUVs. FCEV production share in the past five years has been extremely low and the lack of fueling infrastructure remains a limiting factor —we set the phase-in cap accordingly. As with BEV1, however, the phase-in cap still allows for the market share of FCEVs to grow several times over.

Autonomie determines the effectiveness of each electrified powertrain type by modeling the basic components, or building blocks, for each powertrain, and then combining the components modularly to determine the overall efficiency of the entire powertrain. The components, or building blocks, that contribute to the effectiveness of an electrified powertrain in the analysis include the vehicle's battery, electric motors, power electronics, and accessory loads. Autonomie identifies components for each electrified powertrain type and then interlinks those components to create a powertrain architecture. Autonomie then models each electrified powertrain architecture and provides an effectiveness value for each architecture. For example, Autonomie determines a BEV's overall efficiency by considering the efficiencies of the battery (including charging efficiency), the electric traction drive system (the electric machine and power electronics), and mechanical power transmission devices. Or, for a PHEV, Autonomie combines a very similar set of components to model the electric portion of the hybrid powertrain and then also includes the ICE and related power for transmission components. Argonne uses data from their Advanced Mobility Technology Laboratory (AMTL) to develop Autonomie's electrified powertrain models. The modeled powertrains are not intended to represent any specific manufacturer's architecture but act as surrogates predicting representative levels of effectiveness for each electrification technology. We discuss the procedures for modeling each of these sub-systems in detail in the TSD and in the CAFE Analysis Autonomie Documentation and include a brief summary below.

The fundamental components of an electrified powertrain's propulsion system—the electric motor and inverter—ultimately determine the vehicle's performance and efficiency. For this analysis, Autonomie employed a set of electric motor efficiency maps created by Oak Ridge National Laboratory (ORNL), one for a traction motor and an inverter, the other for a motor/generator and inverter. Autonomie also uses test data validations from technical publications to determine the peak efficiency of BEVs and FCEVs. The electric motor efficiency maps, created from production vehicles like the 2007 Toyota Camry hybrid, 2011 Hyundai Sonata hybrid, and 2016 Chevrolet Bolt, represent electric motor efficiency as a function of torque and motor rotations per minute (RPM). These efficiency maps provide nominal and maximum speeds, as well as a maximum torque curve. Argonne uses the maps to determine the efficiency characteristics of the motors, which includes some of the losses due to power transfer through the electric machine. Specifically, Argonne scales the efficiency maps, specific to powertrain type, to have total system peak efficiencies ranging from 96-98 percent —such that their peak efficiency value corresponds to the latest state-of-the-art technologies, opposed to retaining dated system efficiencies (90-93 percent).

Beyond the powertrain components, Autonomie also considers electric accessory devices that consume energy and affect overall vehicle effectiveness, such as headlights, radiator fans, wiper motors, engine control units, transmission control units, cooling systems, and safety systems. In real-world driving and operation, the electrical accessory load on the powertrain varies depending on how the driver uses certain features and the condition in which the vehicle is operating, such as for night driving or hot weather driving. However, for regulatory test cycles related to fuel economy, the electrical load is repeatable because the fuel economy regulations control for these factors. Accessory loads during test cycles do vary by powertrain type and vehicle technology class, since distinctly different powertrain components and vehicle masses will consume different amounts of energy.

The analysis fleets consist of different vehicle types with varying accessory electrical power demand. For instance, vehicles with different motor and battery sizes will require different sizes of electric cooling pumps and fans to optimally manage component temperatures. Autonomie has built-in models that can simulate these varying sub-system electrical loads. However, for this analysis, we use a fixed (by vehicle technology class and powertrain type), constant power draw to represent the effect of these accessory loads on the powertrain on the 2-cycle test. We intend and expect that fixed accessory load values will, on average, have similar impacts on effectiveness as found on actual manufacturers' systems. This process is in line with the past analyses. For this analysis, we aggregate electrical accessory load modeling assumptions for the different powertrain types (electrified and conventional) and technology classes (both LD and HDPUV) from data from the Draft TAR, EPA Proposed Determination, data from manufacturers, research and development data from DOE's Vehicle Technologies Office, and DOT-sponsored vehicle benchmarking studies completed by Argonne's AMTL.

Certain technologies' effectiveness for reducing fuel consumption requires optimization through the appropriate sizing of the powertrain. Autonomie uses sizing control algorithms based on data collected from vehicle benchmarking, and the modeled electrification components are sized based on performance neutrality considerations. This analysis iteratively minimizes the size of the powertrain components to maximize efficiency while enabling the vehicle to meet multiple performance criteria. The Autonomie simulations use a series of resizing algorithms that contain “loops,” such as the acceleration performance loop (0-60 mph), which automatically adjusts the size of certain powertrain components until a criterion, like the 0-60 mph acceleration time, is met. As the algorithms examine different performance or operational criteria that must be met, no single criterion can degrade; once a resizing algorithm completes, all criteria will be met, and some may be exceeded as a necessary consequence of meeting others.

Autonomie applies different powertrain sizing algorithms depending on the type of vehicle considered because different types of vehicles not only contain different powertrain components to be optimized, but they must also operate in different driving modes. While the conventional powertrain sizing algorithm must consider only the power of the engine, the more complex algorithm for electrified powertrains must simultaneously consider multiple factors, which could include the engine power, electric machine power, battery power, and battery capacity. Also, while the resizing algorithm for all vehicles must satisfy the same performance criteria, the algorithm for some electric powertrains must also allow those electrified vehicles to operate in certain driving cycles, like the US06 cycle, without assistance of the combustion engine and ensure the electric motor/generator and battery can handle the vehicle's regenerative braking power, all-electric mode operation, and intended range of travel.

To establish the effectiveness of the technology packages, Autonomie simulates the vehicles' performance on compliance test cycles. For vehicles with conventional powertrains and micro hybrid powertrains, Autonomie simulates the vehicles using the 2-cycle test procedures and guidelines. For mild HEVs and strong HEVs, Autonomie simulates the same 2-cycle test, with the addition of repeating the drive cycles until the final state of charge (SOC) is approximately the same as the initial SOC, a process described in SAE J1711; SAE J1711 also provides test cycle guidance for testing specific to plug-in HEVs. PHEVs have a different range of modeled effectiveness during “standard setting” CAFE Model runs, in which the PHEV operates under a “charge sustaining” (gasoline-only) mode—similar to how SHEVs function—compared to “EIS” runs, in which the same PHEV operates under a “charge depleting” mode—similar to how BEVs function. For BEVs and FCEVs, Autonomie simulates vehicles performing the test cycles per guidance provided in SAE J1634.

Chapters 2.4 and 3.3 of the TSD and the CAFE Analysis Autonomie Documentation chapter titled “Test Procedure and Energy Consumption Calculations” discuss the components and test cycles used to model each electrified powertrain type; please refer to those chapters for more technical details on each of the modeled technologies discussed in this section.

The range of effectiveness for the electrification technologies in this analysis is a result of the interactions between the components listed above and how the modeled vehicle operates on its respective test cycle. This range of values will result in some modeled effectiveness values being close to real-world measured values, and some modeled values that will depart from measured values, depending on the level of similarity between the modeled hardware configuration and the real-world hardware and software configurations. The range of effectiveness values for the electrification technologies applied in the LD fleets are shown in TSD Figure 3-23 and Figure 3-24. Effectiveness values for electrification technologies in the HDPUV fleet are shown in TSD Figure 3-25.

Some advanced engine technologies indicate low effectiveness values when paired with hybrid architectures. The low effectiveness results from the application of advanced engines to existing P2 architectures. This effect is expected and illustrates the importance of using the full vehicle modeling to capture interactions between technologies, and capture instances of both complimentary technologies and non-complimentary technologies. When developing our hybrid engine maps, we consider the engine, engine technologies, electric motor power, and battery pack size. We calibrate our hybrid engine maps to operate in their respective hybrid architecture most effectively and to allow the electric machine to provide propulsion or assistance in regions of the engine map that are less efficient. As the model sizes the powertrain for any given application, it considers all these parameters as well as performance neutrality metrics to provide the most efficient solution. In this instance, the P2 powertrain improves fuel economy, in part, by allowing the engine to spend more time operating at efficient engine speed and load conditions. This reduces the advantage of adding advanced engine technologies, which also improve fuel economy, by broadening the range of speed and load conditions for the engine to operate at high efficiency. This redundancy in fuel savings mechanism results in a lower effectiveness when the technologies are added to each other.

We received limited comment on ways to improve our strong hybrid effectiveness modeling in the analysis. Toyota commented that our strong hybrid fuel economy improvements are “unrealistic” because of “ICE and hybrid powertrains approaching the limits of diminishing returns”; Toyota also noted and disagreed with the associated rolling resistance and aerodynamic advancements producing “such dramatic fuel efficiency gains.” Conversely, ICCT commented that our hybrid engine effectiveness is “outdated” and that “NHTSA assumes no additional hybrid powertrain improvements,” mentioning “every subsequent generation of Toyota's hybrid system significantly improves upon the prior generation's efficiency.” A similar commenter suggested that we mischaracterize “how hybrid systems can improve engine efficiency,” also referencing a Roush report.

We disagree with comment that the electrification technology represented in this analysis is “outdated” or “unrealistic”—the majority of the technologies were developed specifically to support analysis for this rulemaking time frame. For example, the hybrid Atkinson engine peak thermal efficiency was updated based on 2017 Toyota Prius engine data. Toyota stated that their current hybrid engines achieve 41 percent thermal efficiency, which aligns with our modeling. Similarly, the electric machine peak efficiency for FCEVs and BEVs is 98 percent and based on the 2016 Chevy Bolt. Specifically, Argonne scales the efficiency maps, specific to powertrain type, to have total system peak efficiencies ranging from 96-98 percent —such that their peak efficiency value corresponds to the latest state-of-the-art technologies, as opposed to retaining dated system efficiencies (90-93 percent). The 2016 maps scaled to peak efficiency are equivalent to (if not exceed) efficiencies seen in vehicles today and in the future. Although the base references for these technologies are from a few years ago, we have worked with Argonne to update individual inputs to reflect the latest improvements. Accordingly, we have made no changes to the electric machine efficiency maps for this final rule analysis.

We also received comments on the interaction between vehicle weights in the Autonomie modeling and vehicle weights when transitioning to BEVs in the real world. Commenters spoke to EV batteries “creating a heavier product” and that “some of these electric vehicles will exceed 8,500 lbs. GVWR, even though they are substitutes for comparable internal combustion engine products that certify as light trucks” to meet customer demands. Another comment from Ford requested that NHTSA reconsider the classification of MDPVs in lieu of LTs that could have weights that would force them into the HDPUV regulatory class, but still have characteristics of the light truck regulatory class.

In regard to reclassifying or offering credits for MDPVs, NHTSA is bound by statute as to how these vehicles are classified for the purpose of CAFE program, and we discuss this concept further in response to these comments and other similar comments in Section VII of this preamble.

In regard to concerns that heavy vehicles could fall out of the light truck fleet into the HDPUV fleet because of the weight of batteries, and in response to comments we received on the MYs 2024-2026 analysis, for the NPRM and continued into this final rule analysis we coordinated with Argonne to conduct the Autonomie modeling in a way that maintained the vehicle regulatory class when a vehicle was upgraded to a BEV. This process was described further in the Autonomie Model Documentation. In some cases, this means some range was sacrificed, but we believe that is a tradeoff that manufacturers could make in the real world. In addition, we believe this situation where a vehicle would hop regulatory classes with the addition of a heavy battery pack only affects a very small subset of vehicles. While some manufacturers are choosing to make very large BEVs, other manufacturers have chosen to focus their efforts on BEVs with smaller battery packs. Our review of the MY 2022 market shows that these novelty vehicles that could toe regulatory class lines are being manufactured in lower volumes and that these moving to the HDPUV regulatory classes may have limited impact on manufacturer compliance.

When the CAFE Model turns a vehicle powered by an ICE into an electrified vehicle, it must remove the parts and costs associated with the ICE (and, potentially, the transmission) and add the costs of a battery pack and other non-battery electrification components, such as the electric motor and power inverter. To estimate battery pack costs for this analysis, we need an estimate of how much battery packs cost now ( i.e., a “base year” cost), and estimates of how that cost could reduce over time ( i.e., the “learning effect.”). The general concept of learning effects is discussed in detail in Section III.C and in Chapter 2 of the TSD, while the specific learning effect we applied to battery pack costs in this analysis is discussed below. We estimate base year battery pack costs for most electrification technologies using BatPaC, which is an Argonne model designed to calculate the cost of EV battery packs.

Traditionally, a user would use BatPaC to cost a battery pack for a single vehicle, and the user would vary factors such as battery cell chemistry, battery power and energy, battery pack interconnectivity configurations, battery pack production volumes, and/or charging constraints, just to name a few, to see how those factors would increase or decrease the cost of the battery pack. However, several hundreds of thousands of simulated vehicles in our analysis have electrified powertrains, meaning that we would have to run individual BatPaC simulations for each full vehicle simulation that requires a battery pack. This would have been computationally intensive and impractical. Instead, Argonne staff builds “lookup tables” with BatPaC that provide battery pack manufacturing costs, battery pack weights, and battery pack cell capacities for vehicles with varying power requirements modeled in our large-scale simulation runs.

Just like with other vehicle technologies, the specifications of different vehicle manufacturer's battery packs are extremely diverse. We, therefore, endeavored to develop battery pack costs that reasonably encompass the cost of battery packs for vehicles in each technology class.

In conjunction with our partners at Argonne working on the CAFE analysis Autonomie modeling, we referenced BEV outlook reports, vehicle teardown reports, and stakeholder discussions to determine common battery pack chemistries for each modeled electrification technology. The CAFE Analysis Autonomie Documentation chapter titled “Battery Performance and Cost Model—BatPaC Examples from Existing Vehicles in the Market” includes more detail about the reports referenced for this analysis. For mild hybrids, we used the LFP-G chemistry because power and energy requirements for mild hybrids are very low, the charge and discharge cycles (or need for increased battery cycle life) are high, and the battery raw materials are much less expensive than a nickel manganese cobalt (NMC)-based cell chemistry. We used NMC622-G for all other electrified vehicle technology base (MY 2022) battery pack cost calculations. While we made this decision at the time of modeling based on the best available information, while also considering feedback on prior rules, more recent data affirms that BEV batteries using NMC622 cathode chemistries are still a significant part of the market. We recognize there is ongoing research and development with battery cathode chemistries that may have the potential to reduce costs and increase battery capacity. In particular, we are aware of a recent shift by manufacturers to transition to lithium iron phosphate (LFP) chemistry-based battery packs as prices for materials used in battery cells fluctuate (see additional discussion below); however, we believe that based on available data, NMC622 is more representative for our MY 2022 base year battery costs than LFP, and any additional cost reductions from manufacturers switching to LFP chemistry-based battery packs in years beyond 2022 are accounted for in our battery cost learning effects. The learning effects estimate potential cost savings for future battery advancements (a learning rate applied to the battery pack DMC), this final rule includes a dynamic NMC/LFP cathode mix over each future model year, as discussed in more detail below. As discussed above, the battery chemistry we use is intended to reasonably represent what is used in the MY 2022 U.S. fleet, the DMC base year for our BatPaC calculations.

We also looked at vehicle sales volumes in MY 2022 to determine a reasonable base production volume assumption. In practice, a single battery plant can produce packs using different cell chemistries with different power and energy specifications, as well as battery pack constructions with varying battery pack designs—different cell interconnectivities (to alter overall pack power end energy) and thermal management strategies—for the same base chemistry. However, in BatPaC, a battery plant is assumed to manufacture and assemble a specific battery pack design, and all cost estimates are based on one single battery plant manufacturing only that specific battery pack. For example, if a manufacturer has more than one BEV in its vehicle lineup and each uses a specific battery pack design, a BatPaC user would include manufacturing volume assumptions for each design separately to represent each plant producing each specific battery pack. As a consequence, we examined battery pack designs for vehicles sold in MY 2022 to determine a reasonable manufacturing plant production volume assumption. We considered each assembly line designed for a specific battery pack and for a specific BEV as an individual battery plant. Since battery technologies and production are still evolving, it is likely to be some time before battery cells can be treated as commodity where the specific numbers of cells are used for varying battery pack applications and all other metrics remain the same.

Similar to previous rulemakings, we used BEV sales as a starting point to analyze potential base modeled battery manufacturing plant production volume assumptions. Since actual production data for specific battery manufacturing plants are extremely hard to obtain and the battery cell manufacturer is not always the battery pack manufacturer, we calculated an average production volume per manufacturer metric to approximate BEV production volumes for this analysis. This metric was calculated by taking an average of all BEV battery energies reported in vehicle manufacturer's PMY 2022 reports and dividing by the averaged sales-weighted energy per-vehicle; the resulting volume was then rounded to the nearest 5,000. Manufacturers are not required to report gross battery pack sizes for the PMY report, so we estimated pack size for each vehicle based on publicly available data, like manufacturer's announced specifications. This process was repeated for all other electrified vehicle technologies. We believe this gave us a reasonable base year plant production volume—especially in the absence of actual production data—since the PMY data from manufacturers already includes accurate related data, such as vehicle model and estimated sales information metrics. Our final battery manufacturing plant production volume assumptions for different electrification technologies are as follows: mild hybrid and strong hybrids are manufactured assuming 200,000 packs, PHEVs are manufactured assuming 20,000 packs, and BEVs are manufactured assuming 60,000 packs.

We believe it was reasonable to consider U.S. sales for purposes of this calculation rather than global sales based on the best available data we had at the time of modeling and based on our understanding of how manufacturers design BEVs for particular markets. A manufacturer may have previously sold the same vehicle with different battery packs in two different markets, but as the outlook for battery materials and global economic events dynamically shift, manufacturers could take advantage of significant design overlap and other synergies like from vertical integration to introduce lower-cost battery packs in markets that it previously perceived had different design requirements. To the extent that manufacturers' costs are based more closely on global volumes of battery packs produced, our base year battery pack production volume assumption could potentially be conservative; however, as discussed further below, our base year MY 2022 battery pack costs fall well within the range of reasonable estimates based on 2023 data. We sought comment on our approach to calculating base year cost estimates, and we also sought comment from manufacturers and other stakeholders on how vehicle and battery manufacturers take advantage of design overlap across markets to maintain cost reduction progress in battery technology; we did not receive comment on either of these particular issues.

As mentioned above, our BatPaC lookup tables provide $/kWh battery pack costs based on vehicle power and energy requirements. As an example, a midsized SUV with mid-level road load reduction technologies might require a 110-120kWh energy and 200-210kW power battery pack. From our base year BatPaC cost estimates, that vehicle might have a battery pack that costs around $123/kWh. Note that the total cost of a battery pack increases the higher the power/energy requirements, however the cost per kWh decreases. This represents the cost of hardware that is needed in all battery packs but is deferred across more kW/kWh in larger packs, which reduces the per kW/kWh cost. Table 3-78 in TSD Chapter 3.3.5 shows an example of the BatPaC-based lookup tables for the BEV3 SUV through pickup technology classes.

Note that the values in the table above should not be considered the total battery $/kWh costs that are used for vehicles in the analysis in future MYs. As detailed below, battery costs are also projected to decrease over time as manufacturers improve production processes, shift battery chemistries, and make other technological advancements. In addition, select modeled tax credits further reduce our estimated costs; additional discussion of those tax credits is located throughout this preamble, TSD Chapter 2.3, and the FRIA Chapters 8 and 9.

The CAFE Analysis Autonomie Documentation details other specific assumptions that Argonne used to simulate battery packs and their associated base year costs for the full vehicle simulation modeling, including updates to the battery management unit costs, and the range of power and energy requirements used to bound the lookup tables. Please refer to the CAFE Analysis Autonomie Documentation and Chapter 3.3 of the TSD for further information about how we used BatPaC to estimate base year battery costs. The full range of BatPaC-generated battery DMCs is located in the file ANL—Summary of Main Component Performance Assumptions_NPRM_2206. Note again that these charts represent the DMC using a dollar per kW/kWh metric; battery absolute costs used in the analysis by technology key can be found in the CAFE Model Battery Costs File.

Our method of estimating future battery costs has three fundamental components: (1) an estimate of MY 2022 battery pack costs ( i.e., our base year costs generated in the BatPaC model (version 5.0, March 2022 release) to estimate battery pack costs for specific vehicles, depending on factors such as pack size and power requirements, discussed above), (2) future learning rates estimated using a learning curve, and (3) the effect of changes in the cost of key minerals on battery pack costs, which are discussed below.

For the proposal, NHTSA estimated learning rates using a study by Mauler et al., in which the authors fit a central tendency curve to 237 published estimates of lithium-ion battery costs. To reflect the combination of fluctuating mineral costs and an increase in demand in the near-term, NHTSA also held the battery pack cost learning curve constant between MYs 2022 and 2025. We explained that this was a conservative assumption that was also employed by EPA in their proposed rule (and now final rule, as discussed further below) for light duty vehicles and medium duty vehicles beginning in MY 2027 at NPRM Preamble Section II.D.3 and Draft Technical Support Document Chapter 3.3.5.3.1. The assumption reflected increased lithium costs since 2020 that were not expected to decline appreciably to circa 2020 levels until additional capacity (mining, materials processing, and cell production) comes on-line, although prices had already fallen from 2022 highs at the time the NPRM was published. NHTSA stated that a continuation of high prices for a few years followed by a decrease to near previous levels is reasonable because world lithium resources are more than sufficient to supply a global EV market and higher prices should continue to induce investment in lithium mining and refining. NHTSA stated that the resulting battery cost estimates provided a reasonable representation of potential future costs across the industry, based on the information available to us at the time of the analysis for this proposal was completed. We also included a summary of current and future battery cost estimates from other government agencies, consulting firms, and manufacturers to both highlight the uncertainties in estimating future battery costs and to show that our estimated costs fell reasonably within the range of projections. NHTSA also examined several battery sensitivity cases that showed examples of how changing different battery pack assumptions could change battery pack costs over time. NHTSA also reminded commenters that because of NHTSA's inability to consider manufacturers building BEVs in response to CAFE standards during standard-setting years, net social costs and benefits do not change significantly between battery cost sensitivity cases, and similarly would not change significantly if much lower battery costs were used.

NHTSA also noted ongoing conversations with DOE and EPA on battery costs, and sought comment on a variety of topics surrounding future battery costs. We sought comment in particular from vehicle and battery manufacturers on any additional data they could submit (preferably publicly) to further the conversation about battery pack costs in the later part of this decade through the early 2030s. In addition, we sought comment on all aspects of our methodology for modeling base year and future year battery pack costs, and welcomed data or other information that could inform our approach for the final rulemaking. We specifically sought comment on how the performance metrics may change in response to shifts in chemistries used in vehicle models driven by global policies affecting battery supply chain development, total global production and associated learning rates, and related sensitivity analyses. Finally, NHTSA also recognized the uncertainty in critical minerals prices into the near future and sought comment on representation of mineral costs in the learning curve, and any other feedback relevant to incorporating these considerations into our modeling framework.

We received comments from several stakeholders regarding general trends and forecasts in battery costs, our battery cost curves, and underlying battery cost assumptions. Some stakeholders cited outside sources they said supported our battery cost values, and other stakeholders cited outside sources they claimed showed our battery cost values were too low. ZETA stated generally that, “[o]verall, the cost of lithium-ion batteries declined substantially between 2008 and 2022, down to $153 per kWh,” citing DOE's estimates as well as Benchmark Minerals information. ICCT commented that “there is evidence available to support lower BEV costs than NHTSA has modeled” and that automakers “are investing heavily in BEV R&D and manufacturing capacity and are achieving higher production volumes with more advanced technologies at lower costs.” ICCT continued to cite their research from 2022, also referenced by NHTSA in the NPRM, stating, “[c]ontinued technological advancements and increased battery production volumes mean that pack-level battery costs are expected to decline to about $105/kWh by 2025 and $74/kWh by 2030.”

NHTSA appreciates the extensive data on declining EV battery costs provided by ZETA, and we believe that the provided data and lines up with our estimates from the NPRM and now this final rule reasonably well. NHTSA agrees with ICCT that there is evidence to support lower BEV costs than what was modeled in the NPRM. NHTSA has since, in collaboration with DOE/Argonne and EPA, modified the battery learning curve used in this analysis, which ultimately reflects lower future battery costs compared to the NPRM. The methodology that NHTSA employed is discussed further below and in TSD Chapter 3.3.

On the other hand, comments from POET highlighted a BNEF reference from 2022, stating that our optimistic learning curve is contradictory to BNEF's analysis —citing “demand continues to grow, battery producers and automakers are scrambling to secure key metals such as lithium and nickel, battling high prices and tight supply” and stating we should “not rely on battery back [sic] learning curves, which have significant uncertainties.” Additional commenters stated that battery cost reduction curves have flattened and costs “rose 7 percent in 2022” with AFPM stating further, “BEV makers will need to increase prices by 25% to account for rising battery prices,” citing a March 2022 Bloomberg article on Morgan Stanley projections; Valero commented that some “forecasters have made naïve predictions that the cost declines will continue,” with Clean Fuels Development Coalition in agreement stating that the decline in battery costs “isn't realistic.” Valero commented that our “learning curve analysis ignores a host of pressures that will be pushing average battery prices higher between now and 2032,” which include “batteries that can power longer-range EVs” and “battery suppliers that can access lithium and other key raw materials at an affordable price.”

NHTSA disagrees with commenters that battery costs will continue to plateau indefinitely or increase in the rulemaking timeframe and believes that battery costs will continue to trend downward in the mid- and long-term. BNEF has since continued to predict a reduction in lithium-ion battery pack price since the BNEF article referenced in POET's comments, stating “[l]ithium prices reached a high point at the end of 2022, but fears that prices would remain high have largely subsided since then and prices are now falling again.” This is in agreement with expert interagency projections from our working group with DOE/Argonne and EPA, in addition to other recent trends and expert projections However, NHTSA does agree that mineral prices have remained elevated during the time of this rulemaking, which is reflected in us continuing to incorporate a learning plateau from MY 2022 to MY 2025 as we did in the NPRM—holding our battery learning rate constant to account for potential fluctuating mineral prices.

We have also considered many of these challenges identified by Valero to the extent possible for this final rule. In addition to continuing the learning curve plateau from MY 2022 to MY 2025 to account for materials-related uncertainties, mentioned above, we worked with DOE/Argonne and EPA to conduct an analysis that confirms the availability of raw materials for batteries, such as lithium. While the analysis from DOE is exogenous to our CAFE Model analysis for the final rule, it does confirm that the availability of battery materials necessary to support the BEVs projected to be built in NHTSA's reference baseline projection as a function of ZEV programs or expected manufacturer production at levels consistent with ACC II levels.

We received additional comment from Valero stating, “NHTSA should not embed chemistry changes into the `learning effect.' NHTSA should instead forecast between now and 2032 what fraction of new vehicles will have one battery design versus another and develop cost estimates for each battery design,” citing that the only major change in chemistry is likely towards LFP. We also received related comment from Rivian stating, “we encourage the agency to elaborate on the extent to which it considered battery cell chemistry trends as they relate specifically to the HDPUV fleet” and that it was unclear whether the NMC battery chemistry applied to the HDPUV fleet, specifically that the “logic of applying LFP in this market is so compelling that it could become the chemistry of choice in the very near term.”

We thank Valero and Rivian for providing comment and agree that LFP should be considered in our battery learning curve. Since our NPRM, we have updated our learning curves to accommodate these concerns—including in the HDPUV fleet. NHTSA and EPA worked with DOE/Argonne to distinguish a battery learning curve that is dynamic over the rulemaking period in the following ways: (1) there is a unique learning curve for each powertrain type (HEV or PHEV/BEV) and vehicle type (compact passenger car through the HDPUV space), which is based primarily on battery pack energy and power for the specific vehicle; (2) there is now a weighted mix between cathode chemistries (NMC vs LFP) throughout the rulemaking period to accommodate the increased prevalence of LFP in the market. NHTSA continues to collaborate with other agencies in developing battery-related metrics for rulemakings that are reflective of industry.

Finally, we received comment from POET on our battery cost curves where they cited comments on EPA's recent “vehicle GHG proposed rule” where POET commented that they found “substantial learning related to the production of BEV componentry has already occurred in the light-duty vehicle sector as evidenced by the current mass production of BEVs and further learning curve benefits would therefore be expected to be much smaller than those assumed by U.S. EPA.” Further, POET stated that NHTSA “should not rely on battery pack learning curves that have significant uncertainties to increase the stringency of the CAFE regulations.” POET gave no further guidance on how our battery learning curve could be changed to account for these uncertainties.

While we agree that there have been advancements in the battery production process, those advancements have been captured in our BatPaC-based circa-MY 2022 battery costs as well as our future battery costs. The BatPaC model is used to set our base year battery costs as well as our battery learning curve, which are dependent on vehicle/powertrain metrics as well as battery-related parameters (such as chemistry, production volume, production efficiency, labor rates, equipment costs and material costs, to name a few). Additionally, we examined several battery cost sensitivity cases, which explore variations of battery cost DMCs as well as material costs; more information on these sensitivities can be found in RIA Chapter 9.2.2 and the Final Rule Battery Costs Docket Memo. We believe our BatPaC-based circa-MY 2022 battery costs and future costs via the learning curve have been developed in a transparent way that involved feedback from stakeholders and expertise from leading government experts on battery-related issues. Despite high-granularity with modeling, there are still inherent uncertainties with modeling any metric (such as fuel prices, for instance); however, just because something is uncertain doesn't mean we shouldn't model it—this is why we sought comment from stakeholders on our inputs and assumptions and have incorporated that feedback in the final rule analysis as discussed in more detail.

For this analysis, to reflect the evolution of battery manufacturing, comments from stakeholders, and for better alignment of battery assumptions between government agencies, the Department of Energy and Argonne, with significant input from NHTSA and EPA, developed battery cost correlation equations from BatPaC for use in both the NHTSA CAFE and EPA GHG analyses. These cost equations—developed for use through MY 2035—were tailored for different vehicle segments, different levels of electrification, and anticipated plant production volumes. These equations represent cost improvements achieved from advanced manufacturing, pack design, and cell design with current and anticipated future battery chemistries, design parameters, forecasted market prices, and vehicle technology penetration. Please see Argonne's Cost Analysis and Projections for U.S.-Manufactured Automotive Lithium-ion Batteries report for a more detailed discussion of the inputs and assumptions that were used to generate these cost equations. The methodology outlined in the report is largely the same that we used in previous rules, which utilized the most up-to-date BatPaC model to estimate future battery costs based on current chemistries, production volumes, and projected material prices.

Similar to our past BatPaC-based estimates for a battery learning curve, the employed learning curve explicitly assumes particular battery chemistry is used; unlike in previous rulemakings, however, a dynamic NMC/LFP mix has been incorporated into the learning curve in collaboration with EPA and DOE/Argonne, which is discussed in more detail below. We believe that during the rulemaking time frame, based on ongoing research and discussions with stakeholders, the industry will continue to employ lithium-ion NMC as the predominant battery cell chemistry for the near-term but will transition more fully to advanced high-nickel battery chemistries like NMC811 or less-costly cell chemistries like LFP-G during the middle or end of the decade— i.e., during the rulemaking timeframe. We acknowledge there are other battery cell chemistries currently being researched that reduce the use of cobalt, use solid opposed to liquid electrolyte, use of silicon-dominant anodes or lithium-metal anodes, or even eliminate use of lithium in the cell altogether; however, at this time, we are limiting battery chemistry to NMC622, NMC811, and LFP for this rulemaking but will continue to monitor work from DOE and related government agencies as well as other developments in the advancement of battery cell chemistries.

As discussed above, due to the potential increasing prevalence of LFP displacing NMC cathodes in the U.S. EV market, especially in the rulemaking years, NHTSA uses a dynamic NMC/LFP mix between the battery cost correlation equations, referred to as a composite correlation equation; LFP market projections used for the mix are noted in TSD Chapter 3.3. LFP market share starts at 1 percent in MY 2021 and grows to 19 percent in MY 2028. For the model years that the composite cost equation covers (for MYs through 2035), NMC battery cathode chemistry is assumed for the remaining market share. Note the composite cost equation only corresponds with BEV and PHEV electrification technologies and not HEV or FCEV electrification technologies. For more information on the development of battery learning curves, please see TSD Chapter 3.3.5.3.1.

Beyond the extent of the battery cost correlation equation, starting in MY 2036, a constant 1.5% learning rate was used through MY 2050. NHTSA used this constant rate due to uncertainty associated with reducing the cost of the pack below the cost of the raw material to build the pack in that far out time frame.

As there are inherent uncertainties in projecting future technology costs such as battery pack due to several factors, including the timing of the analysis used to support this final rule, we performed several battery-related cost sensitivity analyses. These include cases increasing the battery pack DMCs by 25%, decreasing the battery pack DMC by 15%, high and low mineral costs, and a curve we used for the NPRM. These results are presented in Chapter 9 of the FRIA. One important point that these sensitivity case results emphasize is that because of NHTSA's inability to consider manufacturers building BEVs and consider the combined fuel economy of PHEVs in response to CAFE standards during standard-setting years ( i.e., MYs 2027-2031 for this final rule), net social costs and benefits do not change significantly between battery cost sensitivity cases, and similarly would not change significantly if much lower battery costs were used.

Additional discussion in TSD Chapter 3 shows that our projected costs fall fairly well in the middle of the range of other costs projected by various studies and organizations for future years. Using the same approach as the rest of our analysis—that our costs should represent an average achievable performance across the industry—we believe that the battery DMCs with the learning curve applied provide a reasonable representation of potential future costs across the industry, based on the information available to us at the time of the analysis for this final rule was completed. RIA chapter 9.2.2 shows how our reference and sensitivity case cost projections change over time using different base year and learning assumptions.

We received two other comments suggesting the price of BEVs are not accurately accounted for in our analysis. CEA and the Corn Growers Associations stated that NHTSA bases its technology costs on nominal prices or MSRP, which do not reflect actual costs to manufacturers. Both commenters stated that this does not reflect reality, as vehicle manufacturers have been reportedly cross-subsidizing electric vehicle costs to different extents since introducing their electrified vehicles.

NHTSA disagrees with these comments and believes that a fundamental misunderstanding of how technology costs are calculated in the analysis could have led to this mistake in the commenters' comprehension of this issue. While all of these concepts were described in detail in the NPRM and Draft TSD (and now this final rule and Final TSD), we will summarize the relevant concepts here. Please see Final TSD Chapter 2.4., Technology Costs, for more detailed information. Our technology costs are from real price teardowns and ground up assembly costs of the component being added to the vehicle. When vehicles adopt technologies in the reference baseline or in response to standards in the analysis, the costs for those technologies are based on the incremental addition of the ground up costs to the reference price, which in this case is the vehicle price. Note that we determine the direct manufacturing costs of the components first, then apply a retail price equivalent markup to that cost before incrementally applying the technology cost to the vehicle price. TSD Chapter 3.3 discusses in detail in how we have developed the ground up costs for BEV batteries and components, and TSD Chapter 2.4 discusses how we account for direct manufacturing costs and retail costs.

We also received several comments related to electric vehicle maintenance and battery replacement costs. For more information on repair/maintenance costs, please see Preamble Section III.G.3.

While batteries and relative battery components are the biggest cost driver of electrification, non-battery electrification components, such as electric motors, power electronics, and wiring harnesses, also add to the total cost required to electrify a vehicle. Different electrified vehicles have variants of non-battery electrification components and configurations to accommodate different vehicle classes and applications with respective designs; for instance, some BEVs may be engineered with only one electric motor and some BEVs may be engineered with two or even four electric motors within their powertrain to provide all wheel drive function. In addition, some electrified vehicle types still include conventional powertrain components, like an ICE and transmission.

For all electrified vehicle powertrain types, we group non-battery electrification components into four major categories: electric motors (or e-motors), power electronics (generally including the DC-DC converter, inverter, and power distribution module), charging components (charger, charging cable, and high-voltage cables), and thermal management system(s). We further group the components into those comprising the electric traction drive system (ETDS), and all other components. Although each manufacturer's ETDS and power electronics vary between the same electrified vehicle types and between different electrified vehicle types, we consider the ETDS for this analysis to be comprised of the e-motor and inverter, power electronics, and thermal system.

When researching costs for different non-battery electrification components, we found that different reports vary in components considered and cost breakdown. This is not surprising, as vehicle manufacturers use different non-battery electrification components in different vehicles systems, or even in the same vehicle type, depending on the application. In order of the component categories discussed above, we examined the following cost teardown studies discussed in TSD 3.3.5 on Table 3-82. Using the best available estimate for each component from the different reports captures components in most manufacturer's systems but not all; we believe, however, that this is a reasonable metric and approach for this analysis, given the non-standardization of electrified powertrain designs and subsequent component specifications. Other sources we used for non-battery electrification component costs include an EPA-sponsored FEV teardown of a 2013 Chevrolet Malibu ECO with eAssist for some BISG component costs, which we validated against a 2019 Dodge Ram eTorque system's publicly available retail price, and the 2015 NAS report. Broadly, our total BISG system cost, including the battery, fairly matches these other cost estimates.

While the majority of electric vehicle cost comments related to batteries, we did receive three comments pertaining to non-battery electrification costs or electrification costs more generally. The Strong PHEV Coalition asserted that despite agreeing with other costs in the analysis, our PHEV50 transmission costs (as shown in the Draft TSD Table 3-89) “disagrees with ANL's previous studies which show a transmission for about $1600 less than shown in the draft technical support document,” referencing an Argonne Light Duty Vehicle Techno-Economic Analysis and quoted, “ANL shows a PHEV transmission cost of $793.” Additionally, the Strong PHEV Coalition stated, “several additional technical modifications can lower the cost of PHEVs that most analyses do not consider,” without providing further specifics.

Upon inspection of the cited Argonne reference, the stated $793 value (or any PHEV50 transmission specific value) could not be found in documentation (in neither the Part One light-duty section nor the Part Two medium-heavy duty section); the only information on PHEV transmissions in the document relates to the number of transmission gears, and the only component-specific costs live in the medium-heavy duty section (without a specific transmission cost given). We use the cost of the AT8L2 transmission as a cost proxy for the hybrid transmission architecture in P2 hybrid systems and CVTL2 transmission architecture in SHEVPS hybrid systems, whose DMCs are based on estimates from Table 8A.2a of the 2015 NAS report; these transmissions are used for other powertrain configurations in the analysis and represents costs that have been agreed on by industry today.

John German argued that our power-split hybrid costs are “incomprehensively high compared with both NHTSA's own previous estimates and with independent cost assessments.” John German claimed that the teardown study conducted by FEV North America, Inc. “on 2013 hybrids found mid-size car powersplit hybrid direct manufacturing cost (DMC) is about $2,050—far below the estimated DMC of $2,946 for electrical components alone in Table 3-89 of the proposed rule TSD that excludes the battery cost.”

NHTSA has responded to this comment in prior rules, extensively detailing the agency's reasons for not relying on particular FEV studies to estimate hybrid costs. Upon further examination of the FEV document, the “Net Incremental Direct Manufacturing Cost” for a midsize passenger car for power-split HEVs was stated as “€2,230” (or approximately $2,943 in 2012$ and about $3,474 in 2021$). Taking a different approach, converting John German's stated value of $2,050 into Euros (which is approximately €1,553, used to search within the FEV study), it is found that this is a value that is listed for a subcompact power-split hybrid in Table E-5 titled “Power-Split Hybrid Electric Vehicle Case Study Results Eastern Europe Labor Rate Substitution.” As detailed extensively in the documentation supporting our analysis, we consider ten vehicle classes, and we believe a subcompact vehicle is only likely to represent vehicles covering a small portion of the vehicles we consider.

Further, the commenter oversimplifies a technology walk between powertrains in a given model year, stating a 2023 Toyota Camry “SE list price is $27,960 and SE hybrid is $30,390, for an increment of $2,430. If RPE is 1.5, then DMC is $1,620.” As discussed in more detail in Final TSD Chapter 2.4 and referenced in a comment response above, we do not use vehicle prices to estimate technology costs, rather we estimate technology costs from the ground-up. For a more-accurate representation of a technology walk from a conventional powertrain to a power-split powertrain, see RIA Chapter 4. We have not made any changes to power-split hybrid costs for this final rule.

As discussed earlier in Section III.C, our technology costs account for three variables: retail price equivalence (RPE), which is 1.5 times the DMC, the technology learning curve, and the adjustment of the dollar value to 2021$ for this analysis. While HDPUVs have larger non-battery electrification componentry than LDVs, the cost calculation methodology is identical, in that the $/kW metric is the same, but the absolute costs are higher. As a result, HDPUVs and LDVs share the same non-battery electrification DMCs.

For the non-battery electrification component learning curves, in both the LD and HDPUV fleets, we used cost information from Argonne's 2016 Assessment of Vehicle Sizing, Energy Consumption, and Cost through Large-Scale Simulation of Advanced Vehicle Technologies report. The report provides estimated cost projections from the 2010 lab year to the 2045 lab year for individual vehicle components. We considered the component costs used in electrified vehicles and determined the learning curve by evaluating the year over year cost change for those components. Argonne published a 2020 and a 2022 version of the same report; however, those versions did not include a discussion of the high and low-cost estimates for the same components. Our learning estimates generated using the 2016 report align in the middle of these two ranges, and therefore we continue to apply the learning curve estimates based on the 2016 report. There are many sources that we could have picked to develop learning curves for non-battery electrification component costs, however given the uncertainty surrounding extrapolating costs out to MY 2050, we believe these learning curves provide a reasonable estimate.

In summary, we calculate total electrified powertrain costs by summing individual component costs, which ensures that all technologies in an electrified powertrain appropriately contribute to the total system cost. We combine the costs associated with the ICE (if applicable) and transmission, non-battery electrification components like the electric machine, and battery pack to create a full-system cost. Chapter 3.3.5.4 of the TSD presents the total costs for each electrified powertrain option, broken out by the components we discussed throughout this section. In addition, the chapter discusses where to find each of the component costs in the CAFE Model's various input files.

4. Road Load Reduction Paths

No car or truck uses energy (whether gas or otherwise) 100% efficiently when it is driven down the road. If the energy in a gallon of gas is thought of as a pie, the amount of energy ultimately available from that gallon to propel a car or truck down the road would only be a small slice. So where does the lost energy go? Most of it is lost due to thermal and frictional loses in the engine and drivetrain and drag from ancillary systems (like the air conditioner, alternator generator, various pumps, etc.). The rest is lost to what engineers call road loads. For the most part, road loads include wind resistance (or aerodynamics), drag in the braking system, and rolling resistance from the tires. At low speeds, aerodynamic losses are very small, but as speeds increases these loses rapidly become dramatically higher than any other road load. Drag from the brakes in most cars is practically negligible. ROLL losses can be significant: at low speeds ROLL losses can be more than aerodynamic losses. Whatever energy is left after these road loads are spent on accelerating the vehicle anytime a its speed increases. This is where reducing the mass of a vehicle is important to efficiency because the amount of energy to accelerate the vehicle is always directly proportional to a vehicle's mass. All else being equal, reduce a car's mass and better fuel economy is guaranteed. However, keep in mind that at freeway speeds, aerodynamics plays a more dominant role in determining fuel economy than any other road load or than vehicle mass.

We include three road load reducing technology paths in this analysis: the MR Path, Aerodynamic Improvements (AERO) Path, and ROLL Path. For all three vehicle technologies, we assign analysis fleet technologies and identify adoption features based on the vehicle's body style. The LD fleet body styles we include in the analysis are convertible, coupe, sedan, hatchback, wagon, SUV, pickup, minivan, and van. The HDPUV fleet body styles include chassis cab, cutaway, fleet SUV, work truck, and work van. Figure III-7 and Figure III-8 show the LD and HDPUV fleet body styles used in the analysis.

As expected, the road load forces described above operate differently based on a vehicle's body style, and the technology adoption features and effectiveness values reflect this. The following sections discuss the three Road Load Reduction Paths.

a. Mass Reduction

MR is a relatively cost-effective means of improving fuel economy, and vehicle manufacturers are expected to apply various MR technologies to meet fuel economy standards. Vehicle manufacturers can reduce vehicle mass through several different techniques, such as modifying and optimizing vehicle component and system designs, part consolidation, and adopting materials that are conducive to MR (advanced high strength steel (AHSS), aluminum, magnesium, and plastics including carbon fiber reinforced plastics).

We received multiple comments on how this analysis evaluated mass reduction as a possible pathway for manufacturers to use to meet the standards. Raw aluminum supplier Arconic, the Aluminum Association, the American Chemistry Council and the California Attorney General commented generally about the benefits of mass reduction to increasing fuel economy. Stakeholders also commented broadly about mass reduction technology given the current state of the vehicle fleet and anticipated future fleet technology transitions. Even given the effectiveness of mass reduction as a pathway to CAFE compliance as well as tightening CAFE standards, multiple aluminum industry members noted that the average mass of vehicles continues to increase. They also noted that there are limited indications of adoption of aluminum primary structure in the fleet and that this will not change by 2032. They also pointed out that significant average mass increases are at least partially being driven by the higher masses associated with BEVs and their heavy batteries. Furthermore, they called on BEV manufacturers to use more aluminum to offset the higher masses associated with the batteries in these vehicles. Similarly, the States and Cities commented with research showing that potential fuel economy improvements from mass reduction have not been fully realized because manufacturers add weight back to the vehicle for other reasons, and because of increasing vehicle footprints. Additional discussion of how NHTSA considers various materials in the mass reduction analysis are given below and in TSD Chapter 3.4, and NHTSA's discussion of vehicle footprint trends is located in TSD Chapter 1.

For the LD fleet portion of this analysis, we considered five levels of MR technology (MR1-MR5) that include increasing amounts of advanced materials and MR techniques applied to the vehicle's glider. The subsystems that may make up a vehicle glider include the vehicle body, chassis, interior, steering, electrical accessory, brake, and wheels systems. We accounted for mass changes associated with powertrain changes separately. We considered two levels of MR (MR1-MR2) and an initial level (MR0) for the HDPUV fleet. We use fewer levels because vehicles within the HD fleets are built for a very different duty cycle than those in the LD fleet and tend to be larger and heavier. Moreover, there are different vehicle parameters, like towing capacity, that drive vehicle mass in the HD fleet rather than, for example, NVH (noise, vibration, and harshness) performance in the LD fleet. Similarly, HDPUV MR is assumed to come from the glider, and powertrain MR occurs during the Autonomie modeling. Our estimates of how manufacturers could reach each level of MR technology in the LD and HDPUV analyses, including a discussion of advanced materials and MR techniques, can be found in Chapter 3.4 of the TSD.

A coalition of NGOs stated that achieving the highest degree of mass reduction, MR5, can be achieved in the mainstream fleet with aluminum alone and carbon fiber technology is not necessary. We disagree with this conclusion. While aluminum technology can be a potent mass reduction pathway, it does have its limitations. First, aluminum, does not have a fatigue endurance limit. That is, with aluminum components there is always some combination of stress and cycles when failure occurs. Automotive design engineering teams will dimension highly stressed cross sections to provide an acceptable number of cycles to failure. But this often comes at mass savings levels that fall short of what would be expected purely based on density specific strength and stiffness properties for aluminum.

Looking at real data, the mostly aluminum (cab and bed are made from aluminum), 2021 Ford F150 achieves less than a 14 percent mass reduction compared to its 2014 all-steel predecessor. This is an especially pertinent comparison because both vehicles have the same footprint within a 2% margin and presumably were engineered to similar duty cycles given that they both came from the same manufacturer. Per our regression analysis, the Ford F-150 achieves MR3. As mentioned in the TSD Chapter 3.4, a body in white structure made almost entirely from aluminum is roughly required to get to MR4. It may be possible to achieve MR5 without the use of carbon fiber, but the resultant vehicle would not achieve performance parity with customer expectations in terms of crash safety, noise and vibration levels, and interior content. The discontinued Lotus Elise is an example of an aluminum and fiberglass car that achieved MR5 but represents an extremely niche vehicle application that is unlikely to translate to mainstream, high-volume models. Therefore, it is entirely reasonable to assume that carbon fiber “hang on” panels and closures would be necessary to achieve MR5 at performance parity.

There were also comments from the NGO coalition regarding the mass reduction section in the NAS study. The commenters noted that the NAS study relies on very little application of carbon fiber technology to achieve their highest level of mass reduction technology. NHTSA would like to note that the NAS study espouses a maximum level of mass reduction of approximately 14.5% using composites ( e.g., fiberglass) and carbon fiber technology only in closures structures ( e.g., doors, hoods, and decklids) and hang-on panels ( e.g., fenders). This is the “alternative scenario 2” in the NAS study. This is similar lightweighting technology application strategy to what our analysis roughly associates with MR5, but MR5 requires a 20% mass reduction. In this scenario, we are allotting more mass reduction potential for the same carbon fiber technology application than the NAS study does.

We assigned MR levels to vehicles in both the LD and HDPUV analysis fleets by using regression analyses that consider a vehicle's body design and body style, in addition to several vehicle design parameters, like footprint, power, bed length (for pickup trucks), and battery pack size (if applicable), among other factors. We have been improving on the LD regression analysis since the 2016 Draft Technical Assessment Report (TAR) and continue to find that it reasonably estimates MR technology levels of vehicles in the analysis fleet. We developed a similar regression for the HDPUV fleet for this analysis using the factors described above and other applicable HDPUV attributes and found that it similarly appropriately assigns initial MR technology levels to analysis fleet vehicles. Chapter 3.4 of the TSD contains a full description of the regression analyses used for each fleet and examples of results of the regression analysis for select vehicles.

NHTSA received comments from a coalition of NGOs that the mass reduction regression curves used in the analysis for quantifying analysis fleet mass reduction overestimates the application mass reduction technology in the fleet. They believe that the mass reduction modeling used by Argonne National Lab for estimating powertrain weight in the Autonomie vehicle simulations more accurately reflects how much mass reduction technology is really in the fleet, and stated that we should be using those regression models for the analysis instead. Although we would like to repeat the NGO's calculations to that led them to this opinion, they did not provide enough detail on its methodology and calculations for NHTSA to confirm its accuracy. Consequently, we are only able to respond with general concepts here.

NHTSA disagrees that the methods used by Autonomie to calculate the MR analysis fleet starting levels would lead to a better analysis than our regression. There are multiple reasons for this. First, Autonomie relies on data collected by the subscription benchmarking database A2Mac1 and other limited sources. As much as NHTSA and Argonne rely on data from A2Mac1 for learning about technical aspects of the fleet, it is not representative data for the entire US fleet. Whereas the CAFE mass reduction regressions use data from all vehicles and multiple trim levels in the US fleet (examples discussed above and further in TSD Chapter 3.4), A2Mac1 is limited in the number of vehicles it can teardown in a given year and thus only makes small samples from the US fleet. Using the entire fleet for the regression analysis provides a more accurate snapshot of how vehicles compare to one another when it comes to assigning MR levels to vehicles in the analysis fleets. Second, the NGOs claim that it is better to arrive at a glider weight by taking the average powertrain weight for a given technology class and subtracting that value from the curb weight of all vehicles in the fleet with that same tech class. We calculate a percentage for the powertrain of the curb weight based on the average powertrain mass for all of the technology classes. We then multiply this same percentage (which for the current fleet is 71%) by the curb weight of each vehicle in the fleet to arrive at the glider share. We did not use bespoke powertrain percentages for each corresponding technology class in the fleet because it will most likely not make a substantial difference in how MR is applied. Third, it must also be noted that Autonomie's glider share percent does not take into account sales weighting because Autonomie simulates every possible combination of vehicles and powertrains. By taking into account sales volumes, our analysis does a better job of representing the actual fleet.

The Joint NGOs also commented that the regression model we used for calculating MR for analysis fleet vehicles is invalid because it was developed using prior model year fleets. We disagree. The regression relies on establishing correlations between various vehicle parameters and the mass of a vehicle. For the most part, these correlations reflect physics and automotive design practices that have not changed substantially since these regressions were developed and updated. For example, one parameter correlated in the regression is rear wheel drive (RWD) vs. front wheel drive (FWD). The regression accurately predicts that going from RWD to FWD will save mass. The mass change associated in going from RWD to FWD arises from the elimination of a drive driveshaft and a discrete differential housing (unless the vehicle is mid or rear engine, which is rare in the fleet). This mass change is expected in the same way today as it would have been when the regression was developed. As a second example, another parameter that we correlate in the regression is convertible vs. non-convertible. Convertibles tend to be heavier than, say, sedans because they do not have the upper load path created by having a sedan's roof rail and C- (or D-) pillars. Consequently, manufacturers must compensate by reinforcing the floor pan to account for the lack of a primary load path. This results in higher mass for convertibles. Between when we developed the regression and today, the physics and fundamentals of this structural dynamic have not changed. Hence the regression we use in this regard is still valid today.

There are several ways we ensure that the CAFE Model considers MR technologies like manufacturers might apply them in the real world. Given the degree of commonality among the vehicle models built on a single platform, manufacturers do not have complete freedom to apply unique technologies to each vehicle that shares the same platform. While some technologies ( e.g., low rolling resistance tires) are very nearly “bolt-on” technologies, others involve substantial changes to the structure and design of the vehicle, and therefore often necessarily affect all vehicle models that share that platform. In most cases, MR technologies are applied to platform level components and therefore the same design and components are used on all vehicle models that share the platform. Each vehicle in the analysis fleet is associated with a specific platform family. A platform “leader” in the analysis fleet is a vehicle variant of a given platform that has the highest level of MR technology in the analysis fleet. As the model applies technologies, it will “level up” all variants on a platform to the highest level of MR technology on the platform. For example, if a platform leader is already at MR3 in MY 2022, and a “follower” starts at MR0 in MY 2022, the follower will get MR3 at its next redesign (unless the leader is redesigned again before that time, and further increases the MR level associated with that platform, then the follower would receive the new MR level).

In addition to leader-follower logic for vehicles that share the same platform, we also restrict MR5 technology to platforms that represent 80,000 vehicles or fewer. The CAFE Model will not apply MR5 technology to platforms representing high volume sales, like a Chevrolet Traverse, for example, where hundreds of thousands of units are sold per year. We use this particular adoption feature and the 80,000-unit threshold in particular, to model several relevant considerations. First, we assume that MR5 would require carbon fiber technology. There is high global demand from a variety of industries for a limited supply of carbon fibers; specifically, aerospace, military/defense, and industrial applications demand most of the carbon fiber currently produced. Today, only about 10 percent of the global dry fiber supply goes to the automotive industry, which translates to the global supply base only being able to support approximately 70,000 cars. In addition, the production process for carbon fiber components is significantly different than for traditional vehicle materials. We use this adoption feature as a proxy for stranded capital ( i.e., when manufacturers amortize research, development, and tooling expenses over many years) from leaving the traditional processes, and to represent the significant paradigm change to tooling and equipment that would be required to support molding carbon fiber panels. There are no other adoption features for MR in the LD analysis, and no adoption features for MR in the HDPUV analysis.

In the Autonomie simulations, MR technology is simulated as a percentage of mass removed from the specific subsystems that make up the glider. The mass of subsystems that make up the vehicle's glider is different for every technology class, based on glider weight data from the A2Mac1 database and two NHTSA-sponsored studies that examined light-weighting a passenger car and light truck. We account for MR from powertrain improvements separately from glider MR. Autonomie considers several components for powertrain MR, including engine downsizing, and, fuel tank, exhaust systems, and cooling system light-weighting. With regard to the LDV fleet, the 2015 NAS report suggested an engine downsizing opportunity exists when the glider mass is light-weighted by at least 10 percent. The 2015 NAS report also suggested that 10 percent light-weighting of the glider mass alone would boost fuel economy by 3 percent and any engine downsizing following the 10 percent glider MR would provide an additional 3 percent increase in fuel economy. The NHTSA light-weighting studies applied engine downsizing (for some vehicle types but not all) when the glider weight was reduced by 10 percent. Accordingly, the analysis limits engine resizing to several specific incremental technology steps; important for this discussion, engines in the analysis are only resized when MR of 10 percent or greater is applied to the glider mass, or when one powertrain architecture replaces another architecture. For the HDPUV analysis, we do not allow engine downsizing at any MR level. This is because HDPUV designs are sized with the maximum GVWR and GCWR in mind, as discussed earlier in this section. We are objectively controlling the vehicles' utility and performance by this method in Autonomie. For example, if more MR technology is applied to a HD van, the payload capacity increases while maintaining the same maximum GVWR and GCWR. The lower laden weight enables these vehicles to improve fuel efficiency by increased capacity. A summary of how the different MR technology levels improve fuel consumption is shown in TSD Chapter 3.4.4.

Our MR costs are based on two NHTSA light-weighting studies—the teardown of a MY 2011 Honda Accord and a MY 2014 Chevrolet Silverado pickup truck —and the 2021 NAS report. The costs for MR1-MR4 rely on the light-weighting studies, while the cost of MR5 references the carbon fiber costs provided in the 2021 NAS report. The same cost curves are used for the HDPUV analysis; however, we used linear interpolation to shift the HDPUV MR2 curve (by roughly a factor of 20) to account for the fact that MR2 in the HDPUV analysis represents a different level than MR2 in the LD analysis. Unlike the other technologies in our analysis that have a fixed technology cost (for example, it costs about $3,000 to add a AT10L3 transmission to a LD SUV or pickup truck in MY 2027), the cost of MR is calculated on a dollar per pound saved basis based on a vehicle's starting weight. Put another way, for a given vehicle platform, an initial mass is assigned using the aforementioned regression model. The amount of mass to reach each of the five levels of MR is calculated by the CAFE Model based on this number and then multiplied by the dollar per pound saved figure for each of the five MR levels. The dollar per pound saved figure increases at a nearly linear rate going from MR0 to M4. However, this figure increases steeply going from MR4 to MR5 because the technology cost to realize the associated mass savings level is an order of magnitude larger. This dramatic increase is reflected by all three studies we relied on for MR costing, and we believe that it reasonably represents what manufacturers would expect to pay for including increasing amounts of carbon fiber on their vehicles. For the HDPUV analysis, there is also a significant cost increase from MR1 to MR2. This is because the MR going from MR1 to MR2 in the HDPUV fleet analysis is a larger step than going from MR1 to MR2 for the LD fleet analysis—5% to 7.5% off the glider compared to 1.4% to 13%.

Like past analyses, we considered several options for MR technology costs. Again, we determined that the NHTSA-sponsored studies accounted for significant factors that we believe are important to include our analysis, including materials considerations (material type and gauge, while considering real-world constraints such as manufacturing and assembly methods and complexity), safety (including the Insurance Institute for Highway Safety's (IIHS) small overlap tests), and functional performance (including towing and payload capacity, noise, vibration, and harshness (NVH)), and gradeability in the pickup truck study.

We received comments that the costs used in the analysis to achieve MR5 are high, both because of the way that we calculated MR5 costs, and how we applied updated costs in the model. Regarding the price of carbon fiber technology, considering a 4-5 year time horizon, we believe that our prices are conservative when taking into account rising energy costs to pyrolyze acrylic fibers to carbon fibers and considering all the costs car manufacturers much shoulder on developing processes to turn the dry fibers into reliable structural components. The recent NAS study confirms our pricing. It explicitly indicates an average price (over the time period of interest, 2027-2030) for carbon fiber materials as approximately $8.25 per pound saved and a manufacturing cost for carbon fiber reinforced polymer components of $13 per pound saved. Multiply the sum of these tow numbers by an RPE of 1.5 (direct and indirect and net income) results in roughly $32 per pound saved which is the figure listed in the Technologies Input File used for the CAFE model for 2027.

Regarding the comment that NHTSA misapplied the MR5 costs in the model, on further review NHTSA agrees that not all MR5 pounds saved will be saved with carbon fiber and that cost should be adjusted to include carbon fiber costs proportional to the materials' use in total pounds saved. We would like to investigate using an incremental or bracketed approach (think US tax structure but with pounds saved and cost) in a future analysis where the costs associated with carbon fiber technology will only be applied to the incremental mass reduction in going from one level of MR to another. We did not make that change for this final rule analysis, however. This is a relatively involved change in the model, which we did not have time to implement and QA/QC in the time available to complete the analysis associated with this final rule. That said, we do not believe that this change would result in a significant change in the analysis for the reasons listed below and are comfortable that the analysis associated with this final rule still reasonably represents manufacturer's decision-making, effectiveness, and cost associated with applying the highest levels of mass reduction technology.

First, we limited application of MR5 in the analysis to represent the limited volume of available dry carbon fiber and the resultant high costs of the raw materials. This constraint is described above and in more detail in TSD Chapter 3. The CAFE Model assumes that there is not enough carbon fiber readily available to support vehicle platforms with more than 80,000 vehicles sold per year. We believe this volume constraint does more to limit the application of MR5 technology in the analysis than does its high price. Even if we used a lower price, this dominant constraint would still be volume. Second, we do not believe that that a lower price would prove to be a competitive pathway to compliance for exotic materials technology compared to other less expensive technologies with higher effectiveness. The MR5 effectiveness as applied to the vehicle in this analysis considers the total effect of reducing that level of mass from the vehicle, from the vehicle's starting MR level. As an example, while the cost of going from MR0 or MR1 to MR5 may be slightly overstated (but still limited in total application by the volume cap), the cost of going from MR4 to MR5 is not. NHTSA will continue to consider the balance of carbon fiber and other advanced materials for mass reduction to meet MR5 levels and update that value in future rules.

b. Aerodynamic Improvements

The energy required for a vehicle to overcome wind resistance, or more formally what is known as aerodynamic drag, ranges from minimal at low speeds to incredibly significant at highway speeds. Reducing a vehicle's aerodynamic drag is, therefore, an effective way to reduce the vehicle's fuel consumption. Aerodynamic drag is characterized as proportional to the frontal area (A) of the vehicle and a factor called the coefficient of drag (C_d). The coefficient of drag (C_d) is a dimensionless value that represents a moving object's resistance against air, which depends on the shape of the object and flow conditions. The frontal area (A) is the cross-sectional area of the vehicle as viewed from the front. Aerodynamic drag of a vehicles is often expressed as the product of the two values, C_d A, which is also known as the drag area of a vehicle. The force imposed by aerodynamic drag increases with the square of vehicle velocity, accounting for the largest contribution to road loads at higher speeds.

Manufacturers can reduce aerodynamic drag either by reducing the drag coefficient or reducing vehicle frontal area, which can be achieved by passive or active aerodynamic technologies. Passive aerodynamics refers to aerodynamic attributes that are inherent to the shape and size of the vehicle. Passive attributes can include the shape of the hood, the angle of the windscreen, or even overall vehicle ride height. Active aerodynamics refers to technologies that variably deploy in response to driving conditions. Example of active aerodynamic technologies are grille shutters, active air dams, and active ride height adjustment. Manufacturers may employ both passive and active aerodynamic technologies to improve aerodynamic drag values.

There are four levels of aerodynamic improvement (over AERO0, the first level) available in the LD analysis (AERO5, AERO10, AERO15, AERO20), and two levels of improvements available for the HDPUV analysis (AERO10, AERO20). There are fewer levels available for the HDPUV analysis because HDPUVs have less diversity in overall vehicle shape; prioritization of vehicle functionality forces a boxy shape and limits incorporation of many of the “shaping”-based aerodynamic technologies, such as smaller side-view mirrors, body air flow, rear diffusers, and so on. Refer back to Figure III-7 and Figure III-8 for a visual of each body style considered in the LD and HDPUV analyses.

Each AERO level associates with 5, 10, 15, or 20 percent aerodynamic drag improvement values over a reference value computed for each vehicle body style. These levels, or bins, respectively correspond to the level of aerodynamic drag reduction over the reference value, e.g., “AERO5” corresponds to the 5 percent aerodynamic drag improvement value over the reference value, and so on. While each level of aerodynamic drag improvement is technology agnostic—that is, manufacturers can ultimately choose how to reach each level by using whatever technologies work for the vehicle—we estimated a pathway to each technology level based on data from an NRC Canada-sponsored wind tunnel testing program. The program included an extensive review of production vehicles utilizing aerodynamic drag improvement technologies, and industry comments. Our example pathways for achieving each level of aerodynamic drag improvements is discussed in Chapter 3.5 of the TSD.

We assigned aerodynamic drag reduction technology levels in the analysis fleets based on vehicle body styles. We computed an average coefficient of drag based on vehicle body styles, using coefficient of drag data from the MY 2015 analysis fleet for the LD analysis, and data from the MY 2019 Chevy Silverado and MY 2020 Ford Transit and the MY 2022 Ford e-Transit for cargo vans for the HDPUV analysis. Different body styles offer different utility and have varying levels of form drag. This analysis considers both frontal area and body style as unchangeable utility factors affecting aerodynamic forces; therefore, the analysis assumes all reduction in aerodynamic drag forces come from improvement in the drag coefficient. Then we used drag coefficients for each vehicle in the analysis fleet to establish an initial aerodynamic technology level for each vehicle. We compared the vehicle's drag coefficient to the calculated drag coefficient by body style mentioned above, to assign initial levels of aerodynamic drag reduction technology to vehicles in the analysis fleets. We were able to find most vehicles' drag coefficients in manufacturer's publicly available specification sheets; however, in cases where we could not find that information, we used engineering judgment to assign the initial technology level.

We also looked at vehicle body style and vehicle horsepower to determine which types of vehicles can adopt different aerodynamic technology levels. For the LD analysis, AERO15 and AERO20 cannot be applied to minivans, and AERO20 cannot be applied to convertibles, pickup trucks, and wagons. We also did not allow application of AERO15 and AERO20 technology to vehicles with more than 780 horsepower. There are two main types of vehicles that inform this threshold: performance ICE vehicles and high-power BEVs. In the case of the former, we recognize that manufacturers tune aerodynamic features on these vehicles to provide desirable downforce at high speeds and to provide sufficient cooling for the powertrain, rather than reducing drag, resulting in middling drag coefficients despite advanced aerodynamic features. Therefore, manufacturers may have limited ability to improve aerodynamic drag coefficients for high performance vehicles with ICEs without reducing horsepower. Only 4,047 units of sales volume in the analysis fleet include limited application of aerodynamic technologies due to ICE vehicle performance.

In the case of high-power BEVs, the 780-horsepower threshold is set above the highest peak system horsepower present on a BEV in the 2020 fleet. We originally set this threshold based on vehicles in the MY 2020 fleet in parallel with the 780-horsepower ICE limitation. For this analysis, the restriction does not have any functional effect because the only BEVs that have above 780-horsepower in the MY 2022 analysis fleet—the Tesla Model S and X Plaid, and variants of the Lucid Air—are already assigned AERO20 as an initial technology state and there are no additional levels of AERO technology left for those vehicles to adopt. Note that these high horsepower BEVs have extremely large battery packs to meet both performance and range requirements. These bigger battery packs make the vehicles heavier, which means they do not have the same downforce requirements as a similarly situated high-horsepower ICE vehicle. Broadly speaking, BEVs have different aerodynamic behavior and considerations than ICE vehicles, allowing for features such as flat underbodies that significantly reduce drag. BEVs are therefore more likely to achieve higher AERO levels, so the horsepower threshold is set high enough that it does not restrict AERO15 and AERO20 application. BEVs that do not currently use high AERO technology levels are generally bulkier ( e.g., SUVs or trucks) or lower budget vehicles.

There are no additional adoption features for aerodynamic improvement technologies in the HDPUV analysis. We limited the range of technology options for reasons discussed above, but both AERO technology levels are available to all HDPUV body styles.

The aerodynamic technology effectiveness values that show the potential fuel consumption improvement from AERO0 technology are found and discussed in Chapter 3.5.4 of the TSD. For example, the AERO20 values shown represent the range of potential fuel consumption improvement values that could be achieved through the replacement of AERO0 technology with AERO20 technology for every technology key that is not restricted from using AERO20. We use the change in fuel consumption values between entire technology keys and not the individual technology effectiveness values. Using the change between whole technology keys captures the complementary or non-complementary interactions among technologies.

We carried forward the established AERO technology costs previously used in the 2020 final rule and again into the MY 2024-2026 standards analysis, and updated those costs to the dollar-year used in this analysis. For LD AERO improvements, the cost to achieve AERO5 is relatively low, as manufacturers can make most of the improvements through body styling changes. The cost to achieve AERO10 is higher than AERO5, due to the addition of several passive aerodynamic technologies, and consecutively the cost to achieve AERO15 and AERO20 are much higher than AERO10 due to use of both passive and active aerodynamic technologies. The two AERO technology levels available for HDPUVs are similar in technology type and application to LDVs in the same technology categories, specifically light trucks. Because of this similarity, and unlike other technology areas that are required to handle higher loads or greater wear, aerodynamics technologies can be almost directly ported between fleets. As a result, there is no difference in technology cost between LD and HDPUV fleets for this analysis. The cost estimates are based on CBI submitted by the automotive industry in advance of the 2018 CAFE NPRM, and on our assessment of manufacturing costs for specific aerodynamic technologies. See the 2018 FRIA for discussion of the cost estimates. We received no additional comments from stakeholders regarding the costs established in the 2018 FRIA during the MY 2024-2026 standards analysis and continued to use the established costs for this analysis. TSD Chapter 3.5 contains additional discussion of aerodynamic improvement technology costs, and costs for all technology classes across all MYs are in the CAFE Model's Technologies Input File. We received no additional comments on aerodynamics technologies and costs and continue to use the established costs for this final rule analysis.

c. Low Rolling Resistance Tires

Tire rolling resistance burns additional fuel when driving. As a car or truck tire rolls, at the point the tread touches the pavement, the tire flattens-out to create what tire engineers call the contact patch. The rubber in the contact patch deforms to mold to the tiny peaks and valleys of the payment. The interlock between the rubber and these tiny peaks and valleys creates grip. Every time the contact patch leaves the road surface as the tire rotates, it must recover to its original shape and then as the tire goes all the way around it must create a new contact patch that molds to a new piece of road surface. However, this molding and repeated re-molding action takes energy. Just like when a person stretches a rubber band it takes work, so does deforming the rubber and the tire to form the contact patch. When thinking about the efficiency of driving a car down the road, this means that not all the energy produced by a vehicle's engine can go into propelling the vehicle forward. Instead, some small, but appreciable, amount goes into deforming the tire and creating the contact patch repeatedly. This also explains why tires with low pressure have higher rolling resistance than properly inflated tires. When the tire pressure is low, the tire deforms more to create the contact patch which is the same as stretching the rubber farther in the analogy above. The larger deformations burn up even more energy and results in worse fuel mileage. Lower-rolling-resistance tires have characteristics that reduce frictional losses associated with the energy dissipated mainly in the deformation of the tires under load, thereby improving fuel economy.

We use three levels of low rolling resistance tire technology for LDVs and two levels for HDPUVs. Each level of low rolling resistance tire technology reduces rolling resistance by 10 percent from an industry-average rolling resistance coefficient (RRC) value of 0.009. While the industry-average RRC is based on information from LDVs, we also determined that value is appropriate for HDPUVs. RRC data from a NHTSA-sponsored study shows that similar vehicles across the LD and HDPUV categories have been able to achieve similar RRC improvements. See Chapter 3.6 of the TSD for more information on this comparison. TSD Chapter 3.6.1 shows the LD and HDPUV low rolling resistance technology options and their associated RRC.

We have been using ROLL10 and ROLL20 in the last several CAFE Model analyses. New for this analysis is ROLL30 for the LD fleet. In past rulemakings, we did not consider ROLL30 due to lack of widespread commercial adoption of ROLL30 tires in the fleet within the rulemaking timeframe, despite commenters' argument on availability of the technology on current vehicle models and possibility that there would be additional tire improvements over the next decade. Comments we received during the comment period for the last CAFE rule also reflected the application of ROLL30 by OEMs, although they discouraged considering the technology due to high cost and possible wet traction reduction. With increasing use of ROLL30 application by OEMs, and material selection making it possible to design low rolling resistance independent of tire wet grip (discussed in detail in Chapter 3.6 of the TSD), we now consider ROLL30 as a viable future technology during this rulemaking period. We believe that the tire industry is in the process of moving automotive manufacturers towards higher levels of rolling resistance technology in the vehicle fleet. We believe that at this time, the emerging tire technologies that would achieve 30 percent improvement in rolling resistance, like changing tire profile, stiffening tire walls, novel synthetic rubber compounds, or adopting improved tires along with active chassis control, among other technologies, will be available for commercial adoption in the fleet during this rulemaking timeframe.

However, we did not consider ROLL30 for the HDPUV fleet, for several reasons. We do not believe that HDPUV manufacturers will use ROLL30 tires because of the significant added cost for the technology while they would see more fuel efficiency benefits from powertrain improvements. As discussed further below, our cost estimates for ROLL30 technology—which incorporate both technology and materials costs—are approximately double the costs of ROLL20. In addition, a significant majority of the HDPUV fleet currently employs no low rolling resistance tire technology. We believe that HDPUV manufacturers will still move through ROLL10 and ROLL20 technology in the rulemaking timeframe. For the final rule, we did not receive feedback from commenters regarding using ROLL30 for HDPUVs. We finalized this rulemaking analysis without including ROLL30 for the HDPUV fleet.

Assigning low rolling resistance tire technology to the analysis fleet is difficult because RRC data is not part of tire manufacturers' publicly released specifications, and because vehicle manufacturers often offer multiple wheel and tire packages for the same nameplate. Consistent with previous rules, we used a combination of CBI data, data from a NHTSA-sponsored ROLL study, and assumptions about parts-sharing to assign tire technology in the analysis fleet. A slight majority of vehicles (52.9%) in the LD analysis fleet do not use any ROLL improvement technology, while 16.2% of vehicles use ROLL10 and 24.9% of vehicles use ROLL20. Only 6% of vehicles in the LD analysis fleet use ROLL30. Most (74.5%) vehicles in the HDPUV analysis fleet do not use any ROLL improvement technology, and 3.0% and 22.5% use ROLL10 and ROLL20, respectively.

The CAFE Model can apply ROLL technology at either a vehicle refresh or redesign. We recognize that some vehicle manufacturers prefer to use higher RRC tires on some performance cars and SUVs. Since most of performance cars have higher torque, to avoid tire slip, OEMs prefer to use higher RRC tires for these vehicles. Like the aerodynamic technology improvements discussed above, we applied ROLL technology adoption features based on vehicle horsepower and body style. All vehicles in the LD and HDPUV fleets that have below 350hp can adopt all levels of ROLL technology.

TSD Chapter 3.6.3 shows that all LDVs under 350 hp can adopt ROLL technology, and as vehicle hp increases, fewer vehicles can adopt the highest levels of ROLL technology. Note that ROLL30 is not available for vehicles in the HDPUV fleet not because of an adoption feature, but because it is not included in the ROLL technology pathway.

TSD Chapter 3.6 shows how effective the different levels of ROLL technology are at improving vehicle fuel consumption.

DMCs and learning rates for ROLL10 and ROLL20 are the same as prior analyses, but are updated to the dollar-year used in this analysis. In the absence of ROLL30 DMCs from tire manufacturers, vehicle manufacturers, or studies, to develop the DMC for ROLL30 we extrapolated the DMCs for ROLL10 and ROLL20. In addition, we used the same DMCs for the LD and HDPUV analyses. This is because the original cost of a potentially heaver or sturdier HDPUV tire is already accounted for in the initial MSRP of a HDPUV in our analysis fleet, and the DMC represents the added cost of the improved tire technology. In addition, as discussed above, LD and HDPUV tires are often interchangeable. We believe that the added cost of each tire technology accurately represents the price difference that would be experienced by the different fleets. ROLL technology costs are discussed in detail in Chapter 3.6 of the TSD, and ROLL technology costs for all vehicle technology classes can be found in the CAFE Model's Technologies Input File. We did not receive comments on this approach used for this analysis and so we finalized the NPRM approach for the final rule.

5. Simulating Air Conditioning Efficiency and Off-Cycle Technologies

Off-cycle and AC efficiency technologies can provide fuel economy benefits in real-world vehicle operation, but the traditional 2-cycle test procedures ( i.e., FTP and HFET) used to measure fuel economy cannot fully capture those benefits. Off-cycle technologies can include, but are not limited to, thermal control technologies, high-efficiency alternators, and high-efficiency exterior lighting. As an example, manufacturers can claim a benefit for thermal control technologies like active seat ventilation and solar reflective surface coating, which help to regulate the temperature within the vehicle's cabin—making it more comfortable for the occupants and reducing the use of low-efficiency heating, ventilation, and air-conditioning (HVAC) systems. AC efficiency technologies are technologies that reduce the operation of or the loads on the compressor, which pressurizes AC refrigerant. The less the compressor operates or the more efficiently it operates, the less load the compressor places on the engine or battery storage system, resulting in better fuel efficiency. AC efficiency technologies can include, but are not limited to, blower motor controls, internal heat exchangers, and improved condensers/evaporators.

Vehicle manufacturers have the option to generate credits for off-cycle technologies and improved AC systems under the EPA's CO₂ program and receive a fuel consumption improvement value (FCIV) equal to the value of the benefit not captured on the 2-cycle test under NHTSA's CAFE program. The FCIV is not a “credit” in the NHTSA CAFE program—unlike, for example, the statutory overcompliance credits prescribed in 49 U.S.C. 32903—but FCIVs increase the reported fuel economy of a manufacturer's fleet, which is used to determine compliance. EPA applies FCIVs during determination of a fleet's final average fuel economy reported to NHTSA. We only calculate and apply FCIVs at a manufacturer's fleet level, and the improvement is based on the volume of the manufacturer's fleet that contains qualifying technologies.

We currently do not model AC efficiency and off-cycle technologies in the CAFE Model like we model other vehicle technologies, for several reasons. Each time we add a technology option to the CAFE Model's technology pathways we increase the number of Autonomie simulations by approximately a hundred thousand. This means that to add just five AC efficiency and five off-cycle technology options would double our Autonomie simulations to around two million total simulations. In addition, 40 CFR 600.512-12 does not require manufacturers to submit information regarding AC efficiency and off-cycle technologies on individual vehicle models in their FMY reports to EPA and NHTSA. In their FMY reports, manufacturers are only required to provide information about AC efficiency and off-cycle technology application at the fleet level. However, starting with MY 2023, manufacturers are required to submit AC efficiency and off-cycle technology data to NHTSA in the new CAFE Projections Reporting Template for PMY, MMY and supplementary reports. Once we begin evaluating manufacturer submissions in the CAFE Projections Reporting Template we may reconsider how off-cycle and AC efficiency technologies are evaluated in future analysis. However, developing a robust methodology for including off-cycle and AC efficiency technologies in the analysis depends on manufacturers giving us robust data.

Instead, the CAFE Model applies predetermined AC efficiency and off-cycle benefits to each manufacturer's fleet after the CAFE Model applies traditional technology pathway options. The CAFE Model attempts to apply pathway technologies and AC efficiency and off-cycle technologies in a way that both minimizes cost and allows the manufacturer to meet a given CAFE standard without over or under complying. The predetermined benefits that the CAFE Model applies for AC efficiency and off-cycle technologies are based on EPA's 2022 Trends Report and CBI compliance data from vehicle manufacturers. We started with each manufacturer's latest reported values and extrapolated the values to the regulatory cap for benefits that manufacturers are allowed to claim, considering each manufacturer's fleet composition ( i.e., passenger cars versus light trucks) and historic AC efficiency and off-cycle technology use. In general, data shows that manufacturers apply less off-cycle technology to passenger cars than pickup trucks, and our input assumptions reflect that. Additional details about how we determined AC efficiency and off-cycle technology application rates are discussed Chapter 3.7 of the TSD.

New for this rulemaking cycle, we also developed a methodology for considering BEV AC efficiency and off-cycle technology application when estimating the maximum achievable credit values for each manufacturer. We did this because the analytical “no-action” reference baseline against which we measure the costs and benefits of our standards includes an appreciable number of BEVs. Because BEVs are not equipped with a traditional engine or transmission, they cannot benefit from off-cycle technologies like engine idle start-stop, active transmission and engine warm-up, and high efficiency alternator technologies. However, BEVs still benefit from technologies like high efficiency lighting, solar panels, active aerodynamic improvement technologies, and thermal control technologies. We calculated the maximum off-cycle benefit that the model could apply for each manufacturer and each MY based on off-cycle technologies that could be applied to BEVs and the percentage of BEVs in each manufacturer's fleet. Note that we do not include PHEVs in this calculation, because they still use a conventional ICE and manufacturers are not required to report UF estimates for individual vehicles, which would have made partial estimation for off-cycle and AC efficiency benefits at the fleet level very difficult. However, we do think that this is reasonable because PHEVs overall constitute less than 2% of the current fleet and the off-cycle and AC efficiency FCIVs for those vehicles only receive a fractional benefit. We discuss additional details and assumptions for this calculation in Chapter 3.7 of the Final TSD.

Note also that we do not model AC efficiency and off-cycle technology benefits for HDPUVs. We have received petitions for off-cycle benefits for HDPUVs from manufacturers, but to date, none have been approved.

Because the CAFE Model applies AC efficiency and off-cycle technology benefits independent of the technology pathways, we must account for the costs of those technologies independently as well. We generated costs for these technologies on a dollars per gram of CO₂ per mile ($ per g/mi) basis, as AC efficiency and off-cycle technology benefits are applied in the CAFE Model on a gram per mile basis (as in the regulations). For this final rule, we updated our AC efficiency and off-cycle technology costs by implementing an updated calculation methodology and converting the DMCs to 2021 dollars. The AC efficiency costs are based on data from EPA's 2010 Final Regulatory Impact Analysis (FRIA) and the 2010 and 2012 Joint NHTSA/EPA TSDs. We used data from EPA's 2016 Proposed Determination TSD to develop the updated off-cycle costs that were used for the 2022 final rule and now this final rule. Additional details and assumptions used for AC efficiency and off-cycle costs are discussed in Chapter 3.7.2 of the Final TSD.

We received limited comments on how we model off-cycle and AC efficiency FCIVs for this rulemaking analysis. Mitsubishi commented that the differences between NHTSA and EPA's proposed rules, “would force manufacturers to choose between applying off-cycle technologies that only apply to the CAFE standard or on-cycle technologies—which are potentially more expensive—that would apply to both the GHG and CAFE standards. NHTSA should model the effects of the EPA GHG proposal on the adoption of off-cycle technology to avoid overestimating the industry's ability to comply, and underestimating the cost of compliance.” The Alliance commented that “for MYs 2023 through 2026 the limit is 15 g/mile on . . . passenger car and trucks fleets. For all other years it is currently 10 g/mile. NHTSA's modeling of off-cycle credits frequently exceeds the 10 g/mile cap in MYs 2027 and later. Assuming NHTSA intends manufacturers to follow the caps defined by EPA, it should correct its modeling so that off-cycle credits are limited to the capped amount.”

We agree with Mitsubishi's comment that differences between the proposed changes to our off-cycle program and EPA's proposed changes to its program could make it difficult for manufacturers to select which off-cycle technologies to place on the vehicles in their compliance fleets. We also agree with the Alliance that, in our modeling for the NPRM, the off-cycle caps exceeded the limits established in the regulation. For this final rule, to align with EPA, NHTSA has changed its proposed limit on the number of off-cycle FCIVs available to manufacturers in MYs 2027 through 2050 in our modeling. For passenger cars powered by an internal combustion engine, we changed the off-cycle FCIV limit from 10.0 g/mi in MYs 2030 through 2050 to 8.0 g/mi in MY 2031, 6.0 g/mi in MY 2032, and 0 g/mi in MYs 2033 through 2050. For light trucks powered by an internal combustion engine, we changed the off-cycle FCIV limit from 15.0 g/mi in MYs 2027 through 2050 to 10.0 g/mi in MYs 2027 through 2030, 8.0 g/mi in MY 2031, 6.0 g/mi in MY 2032, and 0 g/mi in MYs 2033 through 2050. Starting in MY 2027, BEVs will no longer be eligible for off-cycle FCIVs in the CAFE program. To facilitate this, we set the off-cycle FCIV limit for BEVs in both the passenger car and light truck regulatory categories to 0 g/mi for MYs 2027 through 2050.

The Alliance also commented that NHTSA proposed to eliminate AC efficiency FCIVs for BEVs beginning in MY 2027 but allowed the credit caps set prior to MY 2027 to be carried forward through MY 2050. They stated that if NHTSA finalizes its proposal to eliminate AC efficiency FCIVs for BEVs, it should adjust its modeling to reflect that.

We agree with the commenter that, in our proposal, we did not model the elimination of AC efficiency FCIVs for BEVs in MYs 2027 through 2050. However, we have corrected this error in our modeling for the final rule. Starting in MY 2027, BEVs will no longer be eligible for AC efficiency FCIVs in the CAFE program. To facilitate this, we set the AC efficiency credit limit for BEVs in both the passenger car and light truck regulatory categories to 0 g/mi for MYs 2027 through 2050 in our modeling.

E. Consumer Responses to Manufacturer Compliance Strategies

Previous subsections of Section III have so far discussed how manufacturers might respond to changes in the standards. While the technology analysis outlined different compliance strategies available to manufacturers, the tangible costs and benefits that accrue because of the standards also depend on how consumers respond to manufacturers decisions. Some of the benefits and costs resulting from changes to standards are private benefits that accrue to the buyers of new vehicles, produced in the MYs under consideration. These benefits and costs largely flow from changes to vehicle ownership and operating costs that result from improved fuel economy, and the costs of the technologies required to achieve those improvements. The remaining benefits are also derived from how consumers use—or do not use—vehicles, but because these are experienced by the broader public rather than borne directly by consumers who purchase and drive new vehicles, we categorize these as “external” benefits even when they do not meet the formal economic definition of externalities. The next few subsections outline how the agency's analysis models consumers' responses to changes in vehicles implemented by manufacturers to respond to the CAFE and HDPUV standards.

1. Macroeconomic and Consumer Behavior Assumptions

Most economic effects of the new standards this final rule establishes are influenced by macroeconomic conditions that are outside the agency's influence. For example, fuel prices are mainly determined by global petroleum supply and demand, yet they partially determine how much fuel efficiency-improving technology U.S. manufacturers will apply to their vehicles, how much more consumers are willing to pay to purchase models offering higher fuel economy or efficiency, how much buyers decide to drive them, and the value of each gallon of fuel saved from higher standards. Constructing these forecasts requires robust projections of demographic and macroeconomic variables that span the full timeframe of the analysis, including real GDP, consumer confidence, U.S. population, and real disposable personal income.

The analysis presented with this final rule employs fuel price forecasts developed by the U.S. Energy Information Administration (EIA), an agency within the U.S. DOE which collects, analyzes, and disseminates independent and impartial energy information to promote sound policymaking and public understanding of energy and its interaction with the economy and the environment. EIA uses its National Energy Modeling System (NEMS) to produce its Annual Energy Outlook (AEO), which presents forecasts of future fuel prices among many other economic and energy-related variables, and these are the source of some inputs to the agency's analysis. NHTSA noted in its proposal that it was considering updating the inputs used to analyze this final rule to include projections from the 2023 AEO for its final rule, and the California Attorney General and others commented that NHTSA should make this change. The agency's analysis of this final rule uses the 2023 EIA AEO's forecasts of U.S. population, GDP, disposable personal income, GDP deflator, fuel prices and electricity prices.

The analysis also relies on S&P Global's forecasts of total the number of U.S. households, and the University of Michigan's Consumer Confidence Index from its annual Global Economic Outlook, which EIA also uses to develop the projections it reports in its AEO.

While these macroeconomic assumptions are important inputs to the analysis, they are also uncertain, particularly over the long lifetimes of the vehicles affected by this final rule. To reflect the effects of this uncertainty, the agency also uses forecasts of fuel prices from AEO's Low Oil Price and High Oil Price side cases to analyze the sensitivity of its analysis to alternative fuel price projections. The purpose of the sensitivity analyses, discussed in greater detail in Chapter 9 of the FRIA, is to measure the degree to which important outcomes can change under different assumptions about fuel prices. NHTSA similarly uses low and high growth cases from the AEO as bounding cases for the macroeconomic variables in its analysis.

Some commenters argued that electricity prices charged to users of public charging stations are somewhat higher on average than the residential rates in AEO 2023. NHTSA expects that at-home charging will continue to be the primary charging method, and thus residential electricity rates are the most representative electricity prices to use in our analysis, and the CAFE Model as currently constructed cannot differentiate between residential and public charging.

The first year included in this analysis is model year 2022, and data for that year represent actual observations rather than forecasts to the extent possible. The projected macroeconomic inputs used in this analysis as well as the forecasts that depend on them—aggregate demand for driving, new vehicle sales, and used vehicle retirement rates—reflect a continued return to pre-pandemic growth rates under all regulatory alternatives. See Chapter 4.1 of the TSD for a more complete discussion of the macroeconomic forecasts and assumptions used in this analysis.

Another key assumption that permeates the agency's analysis is how much consumers are willing to pay for improved fuel economy. Increased fuel economy offers vehicle owners savings through reduced fuel expenditures throughout the lifetime of a vehicle. If buyers fully value the savings in fuel costs that result from driving (and potentially re-selling) vehicles with higher fuel economy, and manufacturers supply all improvements in fuel economy that buyers demand, then market-determined levels of fuel economy would reflect both the cost of improving it and the private benefits from doing so. In that case, regulations on fuel economy would only be necessary to reflect environmental or other benefits not experienced by buyers themselves. But if consumers instead undervalue future fuel savings or appear unwilling to purchase cost-minimizing levels of fuel economy for other reasons, manufacturers would spend too little on fuel-saving technology (or deploy its energy-saving benefits to improve vehicles' other attributes). In that case, more stringent fuel economy standards could lead manufacturers to make improvements in fuel economy that not only reduce external costs from producing and consuming fuel, but also improve consumer welfare.

Increased fuel economy offers vehicle owners significant potential savings. The analysis shows that the value of prospective fuel savings exceeds manufacturers' technology costs to comply with the preferred alternatives for each regulatory class when discounted at 3 percent. It seems reasonable to assume that well-informed vehicle shoppers who do not face time constraints or other barriers to economically rational decision-making will recognize the full value of fuel savings from purchasing a model that offers higher fuel economy, since they would be compensated with an equivalent increase in their disposable income and the other consumption opportunities it affords them. For commercial operators, higher fuel efficiency and the reduced fuel costs it provides would free up additional capital for either higher profits or additional business ventures. If consumers did value the full amount of fuel savings, more fuel-efficient vehicles would functionally be less costly for consumers to own when considering both their purchase prices and subsequent operating costs, thus making the models that manufacturers are likely to offer under stricter alternatives more attractive than those available under the No-Action Alternative.

Recent econometric research is inconclusive. Some studies conclude that consumers value most or all of the potential savings in fuel costs from driving higher-mpg vehicles, and others conclude that consumers significantly undervalue expected fuel savings. More circumstantial evidence appears to show that consumers do not fully value the expected lifetime fuel savings from purchasing higher-mpg models. Although the average fuel economy of new light vehicles reached an all-time high in MY 2021 of 25.4 mpg, this is still significantly below the fuel economy of the fleet's most efficient vehicles that are readily available to consumers. Manufacturers have repeatedly informed the agency that consumers only value between 2 to 3 years of fuel savings when choosing among competing models to purchase.

The potential for buyers to forego improvements in fuel economy that appear to offer future savings exceeding their initial costs is one example of what is often termed the “energy paradox” or “energy-efficiency gap.” This appearance of a gap between the level of energy efficiency that would minimize consumers' overall expenses and the level they choose to purchase is typically based on engineering calculations that compare the initial cost for providing higher energy efficiency to the discounted present value of the resulting savings in future energy costs. There has long been an active debate about whether such a gap actually exists and why it might arise. Economic theory predicts, assuming perfect information and absent market failures, that economically rational individuals will purchase more energy-efficient products only if the savings in future energy costs they offer promise to offset their higher initial purchase cost.

However, the field of behavioral economics has documented situations in which the decision-making of consumers can differ from what the standard model of rational consumer behavior predicts, particularly when the choices facing consumers involve uncertain outcomes. The future value of purchasing a vehicle that offers higher fuel economy is inherently uncertain for many reasons, but particularly because the mileage any particular driver experiences will differ from that shown on fuel economy labels, potential buyers may be uncertain how much they will actually drive a new vehicle, future resale prices may be unpredictable, and future fuel prices are highly uncertain. Recent research indicates that some consumers exhibit several departures from purely rational economic behavior, some of which could account for undervaluation of fuel economy to an extent roughly consistent with the agency's assumed 30-month payback rule. These include valuing potential losses more than potential gains of equal value when faced with an uncertain choice (“loss aversion”), the tendency to apply discount rates that decrease over time (“present bias,” also known as hyperbolic discounting), a preference for choices with certain rather than uncertain outcomes (“certainty bias”), and inattention or “satisficing.”

There are also a variety of more conventional explanations for why consumers might not be willing to pay the cost of improvements in fuel efficiency that deliver net savings, including informational asymmetries among consumers, dealerships, and manufacturers; market power; first-mover disadvantages for both consumers and manufacturers; principal-agent problems that create differences between the incentives of vehicle purchasers and vehicle drivers; and positional externalities.

The proposal assumed that potential buyers value only the undiscounted savings in fuel costs from purchasing a higher-mpg model they expect to realize over the first 30 months ( i.e., 2.5 years) they own it. NHTSA sought comment on the 30-month payback period assumption in its proposal. IPI agreed with NHTSA's choice to include the energy efficiency gap as a potential cause for why consumers may not fully value fuel savings in their purchase decisions. IPI also suggested that NHTSA's discussion of the energy efficiency gap omitted relevant findings from the literature and expressed undue uncertainty regarding the existence of the gap. Consumer Reports suggested that NHTSA should continue to rely on a shorter payback period when modeling how much fuel savings manufacturers believe consumers will value but use a longer payback period to represent consumers preferences.

Valero commented and suggested that NHTSA's 30-month payback assumption is “unsupported,” and that in the proposal's No-Action case a large number of vehicle models were converted to BEVs with payback periods longer than 30 months. The Center for Environmental Accountability suggested that manufacturers have not supported the 30-month payback period and have instead stated that consumers do not display any myopic tendencies. They suggested NHTSA should switch from a 30-month assumption to a more conservative and longer payback period and pointed towards the lower net benefits found in the proposal's 60-month payback period sensitivity case as evidence that this would lower net benefits from the preferred alternative, in some cases causing them to become negative.

Although commenters expressed dissatisfaction with NHTSA's assumption and proposed various alternatives to it, NHTSA ultimately decided to continue using its methodology from the proposal in its final rule analysis. In preparation for the final rule, NHTSA updated its review of research on the energy efficiency gap, concluding that estimates of how consumers value fuel savings reported in recent published literature continue to show a wide range, and updated its discussion of this topic in Chapter 2.4 of the FRIA to reflect this finding. While survey data like the results that Consumer Reports submitted are suggestive of a broad appeal for fuel savings among consumers, they represent the stated preferences of respondents for some increased level of fuel economy and may not accurately describe their actual purchasing behavior when faced with the range of fuel economy levels in today's new vehicle market. In fact, previous surveys performed by Consumer Reports show that a significantly smaller fraction—29%—of those who are willing to pay for increased fuel economy would be willing to pay for improvements that required longer than 3 years to repay the higher costs of purchasing models that offered them, with the average consumer willing to pay only for fuel economy improvements that recouped their upfront costs within 2 to 3 years.

In response to Valero and the Center for Environmental accountability, NHTSA disagrees that its methodology is unsupported. This assumption is based on what manufacturers have told NHTSA they believe to be consumers' willingness to pay, and this belief is ultimately what determines the amount of technology that manufacturers will freely adopt. The Center for Environmental Accountability seems to misconstrue comments submitted by the Alliance to the revised Circular A-4 proposal, which explores the possibility that consumers value most if not all fuel savings at higher personal discount rates. The Alliance's comment to OMB mirrors the language included in the proposal's TSD, and as the agency found in the proposal and again for this final rule, is not incongruent with the 30-month payback assumption, as explained in Chapter 2.4 of the FRIA. The Alliance's comment to OMB also cites a recent paper by Leard (2023) which found higher willingness to pay for fuel economy improvements. NHTSA considered and referenced this same paper alongside other recent research in its own evaluation of the literature in the proposal and in the final rule. Furthermore, the Alliance has traditionally supported a 30-month payback assumption for the central analysis.

NHTSA did not choose to adopt separate assumptions about consumer willingness to pay for fuel savings in its sales and technology modules for the final rule. As profit maximizing firms, manufacturers have a strong interest in producing vehicles with the attributes that consumers will most value. Indeed, the EPA trends report finds that in 2022 the 90th percentile real-world fuel economy for the fleet of new vehicles was over 3 times the median value. If fuel economy was valued by consumers at a significantly higher rate than manufacturers believe that they value it, then presumably these high fuel economy vehicles would have severe excess demand and inventory for them would be incredibly scarce, which NHTSA does not observe in the data. NHTSA would need more compelling evidence about the market failures that would lead manufacturers to consistently incorrectly assess the willingness to pay of consumers for fuel savings. NHTSA believes that without such evidence, the approach from the proposal is a more reasonable method for modeling this variable.

The 30-month payback period assumption also has important implications for other results of our regulatory analysis, including the effect of raising standards on sales and use of new vehicles, the number and use of older vehicles, safety, and emissions of air pollutants. Recognizing the consequences of these effects for our regulatory analysis, NHTSA also includes a handful of sensitivity cases to examine the impacts of longer and shorter payback periods on its outcomes. These concepts are explored more thoroughly in Chapter 4.2.1.1 of the TSD and Chapter 2.4 of the FRIA.

It is possible that buyers of vehicles used in commercial or business enterprises, who presumably act as profit-maximizing entities, could value tradeoffs between long-term fuel savings and initial purchase prices differently than the average non-commercial consumer. However, both commercial and non-commercial consumers face their own sources of uncertainty or other constraints that may prevent them from purchasing levels of fuel efficiency that maximize their private net benefits. Additionally, the CAFE Model is unable to distinguish between these two types of purchasers. Given this constraint, NHTSA believes that using the same payback period for the HDPUV fleet as for the LD fleet continues to make sense. Similar to the light-duty analysis, the agency is including several sensitivity cases testing alternative payback assumptions for HDPUVs. One commenter noted that switching to a 60-month payback period in its sensitivity case caused net benefits to become negative. NHTSA acknowledged the sensitivity of this result in the proposal but believes that for the reasons noted above, that a 30 month payback period is still a better supported choice for modelling HDPUV buyers' payback period within the constraints of the CAFE Model.

2. Fleet Composition

The composition of the on-road fleet—and how it changes in response to establishing higher CAFE and fuel efficiency standards—determines many of the costs and benefits of the final rule. For example, how much fuel the LD fleet consumes depends on the number and efficiency of new vehicles sold, how rapidly older (and less efficient) vehicles are retired, and how much the vehicles of each age that remain in use are driven.

Until the 2020 final rule, previous CAFE rulemaking analyses used static fleet forecasts that were based on a combination of manufacturer compliance data, public data sources, and proprietary forecasts (or product plans submitted by manufacturers). When simulating compliance with regulatory alternatives, those analyses projected identical sales and retirements for each manufacturer and model under every regulatory alternative. Exactly the same number of each model was assumed to be sold in a given MY under both the least stringent alternative (typically the reference baseline) and the most stringent alternative considered (intended to represent “maximum technology” scenarios in some cases).

However, a static fleet forecast is unlikely to be representative of a broad set of regulatory alternatives that feature significant variation in prices and fuel economy levels for new vehicles. Several commenters on previous regulatory actions and peer reviewers of the CAFE Model encouraged NHTSA to consider the potential impact of fuel efficiency standards on new vehicle prices and sales, the changes to compliance strategies that those shifts could necessitate, and the accompanying impact on vehicle retirement rates. In particular, the continued growth of the utility vehicle segment causes changes within some manufacturers' fleets as sales volumes shift from one region of the footprint curve to another, or as mass is added to increase the ride height of a vehicle originally designed on a sedan platform to create a crossover utility vehicle with the same footprint as the sedan on which it is based.

The analysis accompanying this final rule, like the 2020 and 2022 rulemakings, dynamically simulates changes in the vehicle fleet's size, composition, and usage as manufacturers and consumers respond to regulatory alternatives, fuel prices, and macroeconomic conditions. The analysis of fleet composition is comprised of two forces: how sales of new vehicles and their integration into the existing fleet change in response to each regulatory alternative, and the influence of economic and regulatory factors on retirement of used vehicles from the fleet (or scrappage). Below are brief descriptions of how the agency models sales and scrappage; for full explanations, readers should refer to Chapter 4.2 of the TSD.

A number of commenters argued that future demand for BEVs is likely to be weaker than assumed by the agency and that the agency's approach to forecasting sales should account for the possibility of BEV adoption causing the total number of new vehicles sales to drop. These commenters theorize that buyers' skepticism towards new technology, the limited driving range of most current BEVs, lack of charging infrastructure, uncertainty over battery life and resale value, and generally higher purchase prices will combine to hamper BEV sales. Commenters similarly argued that even if consumers do purchase BEVs, they will drive fewer miles because of limited charging infrastructure.

Within the CAFE Model's logic, there is an implicit assumption that new vehicle models within the same vehicle class ( e.g., passenger cars v. light trucks) are close substitutes for one another, including vehicles with differing powertrains. NHTSA recognizes that different vehicle attributes may change a vehicle's utility and NHTSA has implemented several safeguards to prevent the CAFE Model from adopting technologies for fuel economy that could adversely affect the utility of vehicles, such as maintaining performance neutrality, including phase-in caps, and using engineering judgment in defining technology pathways. The agency further considers that even with these safeguards in place, there is a potential that vehicles could have been improved in ways that would have further increased consumer utility in the absence of standards.

This is not the first time the agency has received comments suggesting that other vehicle attributes beyond price and fuel economy affect vehicle sales and usage. Some commenters to past rules have suggested that a more detailed representation of the new vehicle market would enable the agency to incorporate the effect of additional vehicle attributes on buyers' choices among competing models, reflect consumers' differing preferences for specific vehicle attributes, and provide the capability to simulate responses such as strategic pricing strategies by manufacturers intended to alter the mix of models they sell and enable them to comply with new CAFE standards. The agency has previously invested considerable resources in developing such a discrete choice model of the new automobile market, although those investments have not yet produced a satisfactory and operational model.

The agency's experience partly reflects the fact that these models are highly sensitive to their data inputs and estimation procedures, and even versions that fit well when calibrated to data from a single period—usually a cross-section of vehicles and shoppers or actual buyers—often produce unreliable forecasts for future periods, which the agency's regulatory analyses invariably require. This occurs because they are often unresponsive to relevant shifts in economic conditions or consumer preferences, and also because it is difficult to incorporate factors such as the introduction of new model offerings—particularly those utilizing advances in technology or vehicle design—or shifts in manufacturers' pricing strategies into their representations of choices and forecasts of future sales or market shares. For these reasons, most vehicle choice models have been better suited for analysis of the determinants of historical variation in sales patterns than to forecasting future sales volumes and market shares of particular categories.

Commenters' predictions of weak BEV demand demonstrate exactly how formidable these challenges can be. The information commenters used to arrive at their conclusions is largely informed by characteristics from some of the earliest BEVs introduced into the market. Many of the factors that commenters raised as weaknesses such as range, sparse charging infrastructure, and high prices, have already experienced significant improvements since those early models were released, and the agency anticipates that efforts such as funding for charging stations and tax credits from the BIL and the IRA will only serve to further enhance these attributes.

Some commenters also offered subjective opinions of BEVs that they felt the agency should consider in their analysis which NHTSA finds too subjective to include in its primary regulatory analysis. For example, one commenter suggested that consumers will reject BEVs because they are “less fun” to drive than “freedom machines.” However, some consumers find the driving experience of BEVs preferrable to ICE vehicles because of their quietness, quick response, and ability to be charged from nearly anywhere with a working outlet. Moreover, as a larger and more diverse array of vehicle models become available with BEV powertrains consumers will be more likely to find vehicles in this class that satisfy their desire for other attributes. Under these conditions, NHTSA would expect that consumer acceptance for BEVs will normalize and more closely resemble current consumer demand for other new vehicles.

However, commenters are likely to be correct that some demographic segment of consumers will still have reservations about transitioning to BEVs, especially in the near-term. NHTSA's standards are performance-based standards, and the market can dictate which technologies should be applied to meet the standards. While the agency believes there is a strong chance that the number of BEVs that will be voluntarily adopted are underestimated in the agency's CAFE Model simulations due to how the agency incorporates EPCA's statutory constraints, the CAFE Model simulations project that BEVs will represent only a quarter of the fleet by MY 2031—all of which occurs in the reference baseline. While the agency disagrees with these commenters, if commenters are correct in their assertions that BEV demand will be weak, the CAFE Model simulations show that consumers will continue to enjoy a heterogenous marketplace with both BEV and non-BEV options, and those who are strongly averse to purchasing a BEV are represented within the nearly 70 percent of the fleet that remains non-electrified under the reference baseline.

NHTSA also notes that consumer acceptance towards EVs is likely to continue to normalize as a larger and more diverse array of vehicle models become available. The likelihood of weak demand raised by commenters is as likely as the possibility that the agency is understating the demand for BEVs. In FRIA Chapter 9, NHTSA examined sensitivity cases in which it alternately imposed its EPCA standard setting year constraints on BEV adoption in each calendar year of its analysis, and in which it did not force compliance with other ZEV regulatory programs and found positive net benefits from the preferred alternative in each case. For these reasons, NHTSA believes that it is appropriate to continue to assume modeling BEVs and ICE vehicles as substitutes is reasonable.

a. Sales

For the purposes of regulatory evaluation, the relevant metric is the difference in the number of new vehicles sold between the baseline and each alternative rather than the absolute number of sales under any alternative. Recognizing this, the agency's analysis of the response of new vehicle sales to requiring higher fuel economy includes three components: a forecast of sales under the baseline alternative (based exclusively on macroeconomic factors), a price elasticity of new vehicle demand that interacts with estimated price increases under each alternative to create differences in sales relative to the No-Action alternative in each year, and a fleet share model that projects differences in the passenger car and light truck market share under each alternative. For a more detailed description of these three components, see Chapter 4.2 of the TSD.

The agency's baseline sales forecast reflects the idea that total new vehicle sales are primarily driven by conditions in the U.S. economy that are outside the influence of the automobile industry. Over time, new vehicle sales have been cyclical—rising when prevailing economic conditions are positive (periods of growth) and falling during periods of economic contraction. While changes to vehicles' designs and prices that occur as consequences of manufacturers' compliance with earlier standards (and with regulations on vehicles' features other than fuel economy) exert some influence on the volume of new vehicle sales, they are far less influential than macroeconomic conditions. Instead, they produce the marginal differences in sales among regulatory alternatives that the agency's sales module is designed to simulate, with increases in new models' prices reducing their sales, although only modestly.

The first component of the sales response model is the nominal forecast, which is based on a small set of macroeconomic inputs that together determine the size of the new vehicle market in each future year under the baseline alternative. This statistically based model is intended only to project a baseline forecast of LDV sales; it does not incorporate the effect of prices on sales and is not intended to be used for analysis of the response to price changes in the new vehicle market. NHTSA's projection oscillates from model year to model year at the beginning of the analysis, before settling to follow a constant trend in the 2030s. This result seems consistent with the continued response to the pandemic and to supply chain challenges. NHTSA's projections of new light-duty vehicle sales during most future years fall between those reported in AEO 2023, and the 2022 final rule which were used as sensitivity cases. NHTSA will continue to monitor changes in macroeconomic conditions and their effects on new vehicle sales, and to update its baseline forecast as appropriate.

NHTSA received several comments suggesting that EV adoption would weaken demand for new vehicles, leading to a decrease in the total amount of vehicles sold. As noted, NHTSA believes that total vehicle sales are largely driven by exogenous macroeconomic conditions. Some commenters also raised the fact that NHTSA does not account for the effects of higher EV prices in its baseline sales forecast. This is consistent with the agency's treatment of other technologies that it projects will be adopted under the No-Action Alternative, either because they prove to be cost-effective or are compelled by other government standards. In addition, we note that the value of tax credits and additional fuel savings are assumed not to affect new vehicle sales because the forecast of sales generated by the CAFE Model for that alternative does not incorporate a response to changes in their effective price.

The baseline HDPUV fleet is modeled differently. NHTSA considered using a statistical model drawn from the LD specification to project new HDPUV sales but reasoned that the mix of HDPUV buyers and vehicles was sufficiently different that an alternative approach was required. Due to a lack of historical and future data on the changing customer base in the HDPUV market ( e.g., the composition of commercial and personal users) and uncertainty around vehicle classification at the margin between the LDV and HDPUV categories, NHTSA chose to rely on an exogenous forecast of HDPUV sales from the AEO. To align with the technology used to create the model fleet, NHTSA used compliance data from multiple model years to estimate aggregate sales for MY 2022, and then applied year-over-year growth rates implicit in the AEO forecast to project aggregate sales for subsequent MYs. Since the first year of the analysis, MY 2022, was constructed using compliance data spanning nearly a decade, the aggregate number of sales for the simulated fleet in MY 2022 was lower than the MY 2022 AEO forecast. To align with the AEO projections, the agency adjusted the growth rate in HDPUV sales upward by 2 percent for MYs 2023-2025, and 2.5 percent for MYs 2026-2028. Instead of adjusting the fleet size to match AEO's forecast for MY2022, the agency elected to phase-in the increase in growth rates over a span of years to reflect the likelihood that HDPUV production will continue to face supply constraints resulting from the COVID pandemic in the near future but should return to normal levels sometime later in the decade.

TheXXXifferd component of the sales response model captures how price changes affect the number of vehicles sold; NHTSA estimates the change in sales from its baseline forecast during future years under each regulatory alternative by applying an assumed price elasticity of new vehicle demand to the percent difference in average price between that regulatory alternative and the baseline. This price change does not represent an increase or decrease from the previous year, but rather the percent difference in the average price of new vehicles between the baseline and each regulatory alternative for that year. In the baseline, the average new vehicle price is defined as the observed price in 2022 (the last historical year before the simulation begins) plus the average regulatory cost associated with the No-Action Alternative for each future model year. The central analysis in this final rule simulates multiple programs simultaneously (CAFE fuel economy and HDPUV fuel efficiency final standards, EPA's 2021 GHG standards, ZEV, and the California Framework Agreement), and the regulatory cost includes both technology costs and civil penalties paid for non-compliance with CAFE standards in a model year. We also subtract any IRA tax credits that a vehicle may qualify for from those regulatory costs to simulate sales. Because the elasticity assumes no perceived change in the quality of the product, and the vehicles produced under different regulatory scenarios have inherently different operating costs, the price metric must account for this difference. The price to which the elasticity is applied in this analysis represents the residual price difference between the baseline and each regulatory alternative after deducting the value of fuel savings over the first 2.5 years of each model year's lifetime.

The price elasticity is also specified as an input, and for the proposal, the agency assumed an elastic response of −0.4—meaning that a five percent increase in the average price of a new vehicle produces a two percent decrease in total sales. NHTSA sought comment on this assumption. Commenters were split over the magnitude of NHTSA's assumed elasticity value. NRDC suggested that more recent studies support a lower magnitude but agreed that NHTSA's choice was reasonable. NADA argued that NHTSA should consider an elasticity of −1 due to the alternatives available to consumers, like repairing used vehicles, XXXifferc transport, and ridesharing services. After reviewing these and other comments, however, NHTSA does not believe that there is a strong empirical case for changing its assumption. As commenters suggestions reveal, estimates of this parameter reported in published literature vary widely, and NHTSA continues to believe that its choice is a reasonable one within this range, but also includes sensitivity cases that explore higher and lower elasticities. Chapter 4.2.1.2 of the TSD further presents the totality of present evidence that NHTSA believes supports its decision.

NADA also asserted that NHTSA did not release the price data used to conduct its sales adjustment. MSRP data, price increase data, and tax credit value data are all available in NHTSA's vehicles report that accompanied both the proposal and final rule. NADA furthermore suggested that NHTSA did not correctly implement its sales adjustment. NADA submitted a similar comment to the agency's 2024-2026 proposal and like there, NHTSA determined that NADA did not correctly determine the change in effective cost or accurately track the No-Action alternative's average effective cost of vehicles to which the regulatory alternative's average effective cost is compared.

Commenters also offered differing suggestions about whether and how NHTSA should incorporate fuel savings into its sales adjustment. NADA suggested that NHTSA should not include fuel savings in the calculation of sales effects since fuel savings do not affect the ability of consumers to obtain financing for new vehicles and argued that financing would act as a barrier to consumers looking to purchase more expensive vehicles that offer greater fuel savings. In support of their argument, NADA cited informal polls conducted by the American Financial Services Association (AFSA) and Consumer Bankers Association showing that approximately 85% of their surveyed members would not extend additional funds to finance more fuel-efficient vehicles. In contrast, NRDC and others argued that the agency's estimate of sales effects was likely to be too large if, as they suggest, consumers value more than 30 months of fuel savings.

NHTSA continues to believe that its approach is reasonable based on its analysis of consumer valuation of fuel savings. As noted in the FRIA Chapter 2.4, there are recent findings in the literature that show a wide range in the estimates of how consumers value fuel savings.

While fuel savings may not influence the terms of a lease or financing offer, the lack of preferential financing for more fuel-efficient vehicles would only prevent consumers for whom the vehicle's price is nearly prohibitive from purchasing the new vehicle in the event of a price increase ( e.g., only the marginal consumer would be affected). The lack of preferential financing would not affect consumers' willingness to pay for fuel economy or the fuel savings realized by consumers who do purchase more fuel-efficient vehicles. New vehicle prices have grown significantly from 2020, largely due to supply constraints during and immediately following the COVID-19 pandemic, as well as continued growth in demand for more expensive SUVs and trucks, and manufacturers removing some lower priced model lines from their fleets. The NY Federal Reserve's Survey of Consumer Expectations has found that rejection rates for auto loans did increase in 2023 to around 11 percent of auto loans. However, the share of consumers who reported that they are likely to apply for an auto loan in the next year declined only marginally from 2022. Higher rejection rates are in line with other forms of credit like credit cards, and mortgage refinance applications which also increased during this timeframe as interest rates have also increased significantly since 2022. At the same time, new vehicle sales grew sharply from 2022 to 2023. Higher prices and interest rates do not appear to be driving consumers out of the market altogether, but rather leading consumers to pursue longer term loans, as Experian reported that the average auto loan term had grown to 68 months in 2024. The effect of higher new vehicle prices on access to financing does not appear to be significantly driving consumers out of the market altogether. Interest rates are also cyclical and assuming interest rates continue to remain constant over the next decade is unrealistic. Thus, NHTSA believes that the rising prices that consumers would face as a result of higher compliance costs could still be financed by a large share of Americans, allowing them to take advantage of fuel savings. As a result, NHTSA has not chosen to model access to financing as a constraint on sales that would be affected incrementally by changes to fuel economy standards. NHTSA believes that consumers are likely to be willing to pay more in financing costs, if the perceived benefits of the vehicle outweigh these costs. Indeed, Consumer Reports noted in its comments, 70 percent of Americans expressed willingness to pay more to lease or purchase a vehicle if its fuel savings outweighed the added cost.

The third and final component of the sales model, which only applies to the light-duty fleet, is the dynamic fleet share module (DFS). For the 2020 and 2022 rulemakings, NHTSA used a DFS model that combines two functions from an earlier version of NEMS to estimate the sales shares of new passenger cars and light trucks based on their average fuel economy, horsepower, and curb weight, current fuel prices, and their prior year's market shares and attributes. The two independently estimated shares are then normalized to ensure that they sum to one. However, as the agency explained in the 2022 final rulemaking, that approach had several drawbacks including the model showing counterintuitive responses to changes in attributes, its exclusion of a price variable, and the observed tendency of the model to overestimate the share of total sales accounted for by passenger automobiles.

For this final rule, NHTSA has revised the inputs used to develop its DFS. The baseline fleet share projection is derived from the agency's own compliance data for the 2022 fleet, and the 2023 AEO projections for subsequent model years. To reconcile differences in the initial 2022 shares, NHTSA projected the fleet share forward using the annual changes from 2022 predicted by AEO and applied these to the agency's own compliance fleet shares for MY 2022. The fleet is distributed across two different body-types: “cars” and “light trucks.” While there are specific definitions of “passenger cars” and “light trucks” that determine a vehicle's regulatory class, the distinction used in this phase of the analysis is simpler: all body styles that are commonly considered cars, including sedans, coupes, convertibles, hatchbacks, and station wagons, are defined as “cars” for the purpose of determining their fleet share. Everything else—SUVs, smaller SUVs (crossovers), vans, and pickup trucks—are defined as “light trucks,” even though some models included in this category may not be treated as such for compliance purposes.

These shares are applied to the total industry sales derived in the first stage of the total sales model to estimate sales volumes of car and light truck body styles. Individual model sales are then determined using the following sequence: (1) individual manufacturer shares of each body style (either car or light truck) are multiplied by total industry sales of that body style, and then (2) each vehicle within a manufacturer's volume of that body-style is assigned the same percentage share of that manufacturer's sales as in model year 2022. This implicitly assumes that consumer preferences for particular styles of vehicles are determined in the aggregate (at the industry level), but that manufacturers' sales shares of those body styles are consistent with their MY 2022 sales. Within a given body style, a manufacturer's sales shares of individual models are also assumed to be constant over time.

This approach also implicitly assumes that manufacturers are currently pricing individual vehicle models within market segments in a way that maximizes their profit. Without more information about each manufacturer's true cost of production, including its fixed and variable components, and its target profit margins for its individual vehicle models, there is no basis to assume that strategic shifts within a manufacturer's portfolio will occur in response to standards. In its comments, IPI noted that this could lead to overestimates of compliance costs, since manufacturers that can more cost-effectively comply with higher standards will be able to capture a larger market share through lower vehicle prices. IPI's assertion may be correct, however NHTSA believes that within its current model there is not a clear way to incorporate such an adjustment, since it would involve evaluating substitution patterns between individual models over a longtime horizon.

Similar to the second component of the sales module, the DFS then applies an elasticity to the change in price between each regulatory alternative and the No-Action Alternative to determine the change in fleet share from its baseline value. NHTSA uses the net regulatory cost differential (costs minus fuel savings) in a logistic model to capture the changes in fleet share between passenger cars and light trucks, with a relative price coefficient of −0.000042. NHTSA selected this methodology and price coefficient based on a review of academic literature. When the total regulatory costs of meeting new standards for passenger automobiles minus the value of the resulting fuel savings exceeds that of light-trucks, the market share of light-trucks will rise relative to passenger automobiles. For example, a $100 net regulatory cost increase in passenger automobiles relative to light trucks would produce a ~.1% shift in market share towards light trucks, assuming the latter initially represent 60% of the fleet.

The approach for this final rule to modeling changes in fleet share addresses several key concerns raised by NHTSA in its prior rulemaking. The model no longer produces counterintuitive effects, and now directly considers the impacts of changes in price. Because the model applies fuel savings in determining changes in relative prices between passenger cars and light trucks, the current approach does not require it to separately consider the utility of fuel economy when determining the respective market shares of passenger automobiles and light trucks. In prior rules, NHTSA has speculated that the rise in light-truck market share may be attributable to the increased utility that light-trucks provide their operators, and as the fuel economy difference between those two categories diminished, light-trucks have become an even more attractive option. As explained in a docket memo accompanying this final rule, NHTSA has been unable to create a comprehensive model that includes vehicle prices, fuel economy, and other attributes that produces appropriate responses to changes in each of these factors, so the agency is considering applying an elasticity to the changes in fuel economy directly to capture this change in utility. Consumer Reports argued that NHTSA's dynamic fleet share model was too uncertain for use in the CAFE Model. While fleet share's response to changes in the standards is an uncertain factor to project, NHTSA based its model on peer reviewed results and a well-grounded methodology described in a docket memo “Calibrated Estimates for Projecting Light-Duty Fleet Share in the CAFE Model.” Finally, some commenters expressed confusion about NHTSA's approach to modeling fleet share. NHTSA explains its approach using a combination of a fixed fleet share forecast for the No-Action alternative, and a dynamic fleet share model to adjust fleet share projections in the regulatory alternatives in TSD Chapter 4.2.

b. Scrappage

New and used vehicles can substitute for each other within broad limits, and when the prices of substitutes for a good increase or decrease, demand for that good responds by rising or falling, causing its equilibrium price and quantity supplied to also rise or fall. Thus, increasing the quality-adjusted price of new vehicles will increase demand for used vehicles, and by doing so raise their equilibrium market value or price and the number that are kept in service. Because used vehicles are not being produced, their supply can only be increased by keeping more of those that would otherwise be retired in use longer, which corresponds to a reduction in their scrappage or retirement rates.

When new vehicles become more expensive, demand for used vehicles increases, but meeting the increase in demand requires progressively more costly maintenance and repairs to keep more of them in working condition, in turn causing them to become more expensive. Because used vehicles are more valuable in such circumstances, they are scrapped at a lower rate, and just as rising new vehicle prices push some prospective buyers into the used vehicle market, rising prices for used vehicles force some prospective buyers to acquire even older vehicles or models with fewer desired attributes. The effect of fuel economy standards on scrappage is partially dependent on how consumers value future fuel savings and our assumption that consumers value only the first 30 months of fuel savings when making a purchasing decision.

Many competing factors influence the decision to scrap a vehicle, including the cost to maintain and operate it, the household's demand for VMT, the cost of alternative means of transportation, and the value that can be attained through reselling or scrapping the vehicle for parts. In theory, a car owner will decide to scrap a vehicle when the value of the vehicle minus the cost to maintain or repair the vehicle is less than its value as scrap material; in other words, when the owner realizes more value from scrapping the vehicle than from continuing to drive it or from selling it. Typically, the owner that scraps the vehicle is not the original vehicle owner.

While scrappage decisions are made at the household level, NHTSA is unaware of sufficiently detailed household data to sufficiently capture scrappage at that level. Instead, NHTSA uses aggregate data measures that capture broader market trends. Additionally, the aggregate results are consistent with the rest of the CAFE Model, as the model does not attempt to model how manufacturers will price new vehicles; the model instead assumes that all regulatory costs to make a particular vehicle compliant are passed onto the purchaser who buys the vehicle.

The dominant source of vehicles' overall scrappage rates is “engineering scrappage,” which is largely determined by the age of a vehicle and the durability of the specific model year or vintage it represents. NHTSA uses proprietary vehicle registration data from I/Polk to estimate vehicle age and durability. Other factors affecting owners' decisions to retire used vehicles or retain them in service include fuel economy and new vehicle prices; for historical data on new vehicle transaction prices, NHTSA uses National Automobile Dealers Association (NADA) Data. The data consist of the average transaction price of all LDVs; since the transaction prices are not broken-down by body style, the model may miss unique trends within a particular vehicle body style. The transaction prices are the amount consumers paid for new vehicles and exclude any trade-in value credited towards the purchase. This may be particularly relevant for pickup trucks, which have experienced considerable changes in average price as luxury and high-end options entered the market over the past decade. Future versions of the agency's scrappage model may consider incorporating price series that consider the price trends for cars, SUVs and vans, and pickups separately. The final source of vehicle scrappage is from cyclical effects, which the model captures using forecasts of GDP and fuel prices.

Vehicle scrappage follows a roughly logistic function with age—that is, when a vintage is young, few vehicles in the cohort are scrapped; as they age, more and more of the cohort are retired each year and the annual rate at which vehicles are scrapped reaches a peak. Scrappage then declines as vehicles enter their later years as fewer and fewer of the cohort remains on the road. The analysis uses a logistic function to capture this trend of vehicle scrappage with age. The data show that the durability of successive MYs generally increases over time, or put another way, historically newer vehicles last longer than older vintages. However, this trend is not constant across all vehicle ages—the instantaneous scrappage rate of vehicles is generally lower for more recent vintages up to a certain age, but must increase thereafter so that the final share of vehicles remaining converges to a similar share remaining for historically observed vintages. NHTSA's model uses fixed effects to capture potential changes in durability across MYs, and to ensure that vehicles approaching the end of their life are scrapped in the analysis, NHTSA applies a decay function to vehicles after they reach age 30. The macroeconomic conditions variables discussed above are included in the logistic model to capture cyclical effects. Finally, the change in new vehicle prices projected in the model (technology costs minus 30 months of fuel savings and any tax credits passed through to the consumer) is included, and changes in this variable are the source of differing scrappage rates among regulatory alternatives.

For this final rule, NHTSA modeled the retirement of HDPUVs similarly to pick-up trucks. The amount of data for HDPUVs is significantly smaller than for the LD fleet and drawing meaningful conclusions from the small sample size is difficult. Furthermore, the two regulatory classes share similar vehicle characteristics and are likely used in similar fashions, so NHTSA believes that these vehicles will follow similar scrappage schedules. Commercial HDPUVs may endure harsher conditions during their useful life such as more miles in tough operating conditions, which may also affect their retirement schedules. We believe that many light-trucks likely endure the same rigor and are represented in the light-truck segment of the analysis; however, NHTSA recognizes that the intensity or proportionality of heavy use in the HDPUV fleet may exceed that of smaller light trucks.

In addition to the variables included in the scrappage model, NHTSA considered several other variables that likely either directly or indirectly influence scrappage in the real world, including maintenance and repair costs, the value of scrapped metal, vehicle characteristics, the quantity of new vehicles purchased, higher interest rates, and unemployment. These variables were excluded from the model either because of difficulties in obtaining data to measure them accurately or other modeling constraints. Their exclusion from the model is not intended to diminish their importance, but rather highlights the practical constraints of modeling intricate decisions like scrappage.

NHTSA sought comment on its scrappage model, as well as on differences between scrappage for light trucks and HDPUVs. IPI suggested that NHTSA replace its reduced form model for scrappage with a structural model, or that it should incorporate the price of used vehicles and other omitted variables in its model to predict scrappage and change its estimation strategy to avoid threats to identification from endogeneity. NHTSA sees merit in the suggestion of a structural model for scrappage but believes it should be implemented as part of a larger change to the CAFE Model in a future rulemaking, since it would also require NHTSA to incorporate a more complex model of the used vehicle market. AFPM commented that increases in the new vehicle prices of ZEVs will also lead to increases in the prices of new ICE vehicles through cross subsidization. NHTSA notes that its scrappage model determines scrappage rates using the average price of new vehicles in each class. Thus, the manufacturers' pricing strategies assumed in the CAFE Model will not affect predicted scrappage rates, since this would only occur where manufacturers raise prices by more or less than the costs they incur to improve the fuel economy of individual models.

MEMA disagreed with NHTSA's approach of modeling HDPUV and light truck scrappage rates using the same function because of differences between fleetwide average use and the average use of the typical vehicle. MEMA noted that one manufacturer had told them that about one-quarter of its fleet remained active for more than 200 percent of the average vehicle's useful life. The maximum age NHTSA assumes for LDVs (40 years) is more than twice their average or “expected” lifetime (about 15 years), so this experience does not appear to be unusual. Indeed, in NHTSA's No-Action Alternative case, around 21 percent of HDPUVs produced in model years 2030-2035 were still operating 30 years after entering the fleet. NHTSA thus continues to believe that it is properly estimating scrappage rates at the fleet level and using as much available data as possible to estimate its scrappage rates. For additional details on how NHTSA modeled scrappage, see Chapter 4.2.2 of the TSD.

3. Changes in Vehicle Miles Traveled (VMT)

In the CAFE Model, VMT is projected from average use of vehicles with different ages, the total number in use, and the composition of the fleet by age, which itself depends on new vehicle sales during each earlier year and vehicle retirement decisions. These three components—average vehicle usage, new vehicle sales, and older vehicle scrappage—jointly determine total VMT projections for each alternative. VMT directly influences many of the various effects of fuel economy standards that decision-makers consider in determining what levels of standards to set. For example, the value of fuel savings is a function of a vehicle's fuel efficiency, the number of miles it is driven, and fuel price. Similarly, factors like criteria pollutant emissions, congestion, and fatalities are direct functions of VMT. For a more detailed description of how NHTSA models VMT, see Chapter 4.3 of the TSD.

NHTSA's perspective is that the total demand for VMT should not vary excessively across alternatives, because basic travel needs for a typical household are unlikely to be influenced by the stringency of the standards, so the daily need the services of vehicles to transport household members will remain the same. That said, it is reasonable to assume that fleets with differing age distributions and inherent cost of operation will have slightly different annual VMT (even without considering VMT associated with rebound miles). Because of the structure of the CAFE Model, the combined effect of the sales and scrappage responses can produce small differences in total VMT across the range of regulatory alternatives if steps are not taken to constrain VMT. Because VMT is related to many of the costs and benefits of the program, even small differences in VMT among alternatives can have meaningful impacts on their incremental net benefits. Furthermore, since decisions about alternative stringencies look at the incremental costs and benefits across alternatives, it is more important that the analysis capture the variation of VMT across alternatives—mainly how vehicles are distributed across vehicles and how many rebound miles may occur in any given alternative—than to accurately project total VMT for any single scenario.

To ensure that travel demand remains consistent across the different regulatory scenarios for the LD fleet, the agency's analysis relies on a model of aggregate light-duty VMT developed by the Federal Highway Administration (FHWA) to produce that agency's official VMT projections. The annual forecasts of total VMT generated by this model when used in conjunction with the macroeconomic inputs described previously model are used to constrain the forecasts of annual VMT generated internally by the CAFE model to be identical among the regulatory alternatives during each year in the analysis period.

NHTSA considered removing the constraint on VMT for the final rule after seeking comment from the public. IPI supported allowing VMT to vary with fleet size, arguing that if fleet size decreases some travelers would likely choose to use alternative forms of transportation like car-sharing, or mass transit rather than relying on older vehicles. Ultimately NHTSA did not choose to make this change in the absence of a tractable model for how this VMT would be redistributed across alternative forms of transportation (including additional miles driven by the legacy fleet), and the various costs and benefits this change would produce. NHTSA will continue to explore methods for modeling this kind of reallocation for future rulemakings, including estimating the cross price elasticities of demand for these alternative forms of travel as IPI recommended.

Since vehicles of different ages and body styles have different costs to own and operate but also provide different benefits, to account properly for the average value of consumer and societal costs and benefits associated with vehicle usage under various alternatives, it is necessary to partition miles by age and body type. NHTSA created “mileage accumulation schedules” usiIIHS-Polk odometer data to construct mileage accumulation schedules as an initial estimate of how much a vehicle expected to drive at each age throughout its life. NHTSA uses simulated new vehicle sales, annual rates of retirement for used vehicles, and the mileage accumulation schedules to distribute VMT across the age distribution of registered vehicles in each calendar year to preserve the non-rebound VMT constraint.

FHWA does not produce an annual VMT forecast for HDPUVs. Without an annual forecast, NHTSA is unable to constrain VMT for HDPUVs as it does for the LD fleet. Instead, an estimate of total VMT for HDPUVs is developed from the estimates of annual use for vehicles of each age (the “mileage accumulation” schedules) and estimates of the number of HDPUVs of each model year and age that remain in use during each future calendar year. For the reasons described previously, we believe that this method produces reasonable estimates of the differences in total VMT and its distribution among vehicles of different ages that is implied by changes in fleet composition and size between the reference baseline and each regulatory alternative.

The fuel economy rebound effect—a specific example of the well-documented energy efficiency rebound effect for energy-consuming capital goods—refers to motorists who choose to increase vehicle use (as measured by VMT) when their fuel economy is improved and, as a result, the cost per mile (CPM) of driving declines. Establishing more stringent standards than the reference baseline level will lead to comparatively higher fuel economy for new cars and light trucks, and increase fuel efficiency for HDPUVs, thus decreasing the cost of fuel consumed by driving each mile and increasing the amount of travel in new vehicles. NHTSA recognizes that the value selected for the rebound effect influences overall costs and benefits associated with the regulatory alternatives under consideration as well as the estimates of lives saved under various regulatory alternatives, and that the rebound estimate, along with fuel prices, technology costs, and other analytical inputs, is part of the body of information that agency decision-makers have considered in determining the appropriate levels of the standards in this final rule. We also note that larger values for the rebound effect diminishes the economic and environmental benefits associated with increased fuel efficiency.

NHTSA conducted a review of the literature related to the fuel economy rebound effect, which is extensive and covers multiple decades and geographic regions. The totality of evidence, without categorically excluding studies that fail to meet certain criteria and evaluating individual studies based on their particular strengths, suggests that a plausible range for the rebound effect is 10-50 percent. This range implies that, for example, a 10 percent reduction in vehicles' fuel CPM would lead to an increase of 1-5 percent in the number of miles they are driven annually. The central tendency of this range appears to be at or slightly above its midpoint, which is 30 percent. Considering only those studies that NHTSA believes are derived from extremely robust and reliable data, employ identification strategies that are likely to prove effective at isolating the rebound effect, and apply rigorous estimation methods, suggests a range of approximately 10-45 percent, with most of the estimates falling in the 15-30 percent range.

However, published estimates of the rebound effect vary widely, as do the data and methodologies that underpin them. A strong case can also be made to support lower values. Both economic theory and empirical evidence suggest that the rebound effect has been declining over time due to factors such as increasing income (which raises the value of travelers' time), progressive smaller reductions in fuel costs in response to continuing increases in fuel economy, and slower growth in car ownership and the number of license holders. Lower estimates of the rebound effect estimates are associated with recently published studies that rely on U.S. data, measure vehicle use using actual odometer readings, control for the potential endogeneity of fuel economy, and—critically—estimate the response of vehicle use to variation in fuel economy itself rather than to fuel cost per distance driven or fuel prices. According greater weight to these studies suggests that the rebound effect is more likely to be in the 5-15 percent range. For a more complete discussion of the rebound literature, see TSD Chapter 4.3.5.

NHTSA selected a rebound effect of 10% for its analysis of both LD and HDPUV fleets because it was well-supported by the totality of the evidence. It is rarely possible to identify whether estimates of the rebound effect in academic literature apply specifically to household vehicles, LDVs, or another category, and different nations classify trucks included in NHTSA's HDPUV category in varying ways, so NHTSA has assumed the same value for LDVs and HDPUVs.

We also examine the sensitivity of estimated impacts to values of the rebound ranging from 5 percent to 15 percent to account for the uncertainty surrounding its exact value. NHTSA sought comment on the above discussion, and whether to consider a different value for the rebound effect for the final rule analysis for either the LD or HDPUV analyses. IPI agreed with NHTSA's choice, arguing that it was well supported in the literature.

AFPM disagreed with NHTSA's approach to modeling mileage for BEVs, suggesting that some studies find that these vehicles are driven less than ICE vehicles, and so NHTSA's assumption that any decrease in operating costs that these vehicles convey to their owner will not cause them to ultimately be used more overall. In response, NHTSA examined the VMT accumulation for BEVs relative to ICE counterparts. Preliminary results showed lower VMT for these vehicles than ICE vehicles, but the agency notes that given the lack of more recent data, this result is driven mostly by early iterations of mainstream BEVs which had shorter ranges, longer recharging times, and significantly fewer charging stations. NHTSA believes that these factors likely played a bigger role in determining their usage than consumers' innate preferences for EVs vs. ICE vehicles. and concluded that there were significant limitations that prevented the agency from being able to project forward these differences with confidence. First, historically, these vehicles have been limited to only a small subset of manufacturers, and segments of the overall market. According to NHTSA's analysis and publicly announced production plans, this is projected to change in the years prior to NHTSA's standard setting years considered in this rulemaking. This will make the owners of these vehicles, and their use patterns more representative of drivers as a whole. Second, the quality of the vehicle charging network is projected to improve significantly as programs like NEVI funded by the Bipartisan Infrastructure Law continue to be implemented. This will enable drivers in areas without at-home charging to make more use of these vehicles and will enable all drivers to travel longer distances in BEVs. Based on these factors, NHTSA believes that projecting BEV use into the future based on differences in their usage in recent years would introduce more error into the model than maintaining its current assumption. NHTSA is continuing to study this issue and will monitor the evidence to determine if changes need to be made in future rulemakings.

In order to calculate total VMT after allowing for the rebound effect, the CAFE Model applies the price elasticity of VMT (taken from the FHWA forecasting model) to the change in fuel cost per mile resulting from higher fuel economy and uses the result to adjust the initial estimate of each model's annual use accordingly. The CAFE model applies this adjustment after the reallocation step described previously, since that adjustment is intended to ensure that total VMT is identical among alternatives before considering the contribution of increased driving due to the rebound effect. Its contribution differs among regulatory alternatives because those requiring higher fuel economy lead to larger reductions in the fuel cost of driving each mile, and thus to larger increases in vehicle use.

The approach used in NHTSA's CAFE model is thus a combination of “top-down” (relying on the FHWA forecasting model to determine total LD VMT in a given calendar year) and “bottom-up” (where the composition and utilization of the on-road fleet determines a base level of VMT in a calendar year, which is constrained to match the FHWA model) forecasting. See Chapter 4.3 of the TSD for a complete accounting of how NHTSA models VMT.

4. Changes to Fuel Consumption

NHTSA uses the fuel economy and age and body-style VMT estimates to determine changes in fuel consumption. NHTSA divides the expected vehicle use by the anticipated mpg to calculate the gallons consumed by each simulated vehicle, and when aggregated, the total fuel consumed in each alternative.

F. Simulating Emissions Impacts of Regulatory Alternatives

This final rule encourages manufacturers of light-duty vehicles and HDPUVs to employ various fuel-saving technologies to improve the fuel efficiency of some or all the models they produce, and in addition to reducing drivers' outlays for fuel, the resulting reductions in their fuel consumption will produce additional benefits. These benefits include reduced vehicle emissions during their operation, as well as lower “upstream” emissions from extracting petroleum, transporting, and refining it to produce transportation fuels, and finally transporting, storing, and distributing fuel. This section provides a detailed discussion of how the agency estimates the resulting reductions in emissions, particularly for the main standard-setting options, including the development and evolution of parameters to estimate emissions of criteria pollutants, GHGs, and air toxics, and the potential improvements in human health from reducing them.

The rule implements an “emissions inventory” methodology for estimating its emissions impacts. Vehicle emissions inventories are often described as three-legged stools, comprised of vehicle activity ( i.e., miles traveled, hours operated, or gallons of fuel burned), population (or number of vehicles), and emission factors. An emission factor is a representative rate that attempts to relate the quantity of a pollutant released to the atmosphere per unit of activity. For this rulemaking, like past rules, activity levels (both miles traveled and fuel consumption) are generated by the CAFE Model, while emission factors have been adapted from models developed and maintained by other Federal agencies.

The following section briefly discusses the methodology the CAFE Model uses to track vehicle activity and populations, and how we generate the emission factors that relate vehicle activity to emissions of criteria pollutants, GHGs, and air toxics. This section also details how we model the effects of these emissions on human health, especially in regard to criteria pollutants known to cause poor air quality. Further description of how the health impacts of criteria pollutant emissions can vary and how these emission damages have been monetized and incorporated into the rule can be found in Preamble Section III.G, Chapter 6.2.2 of the TSD, and the Final EIS accompanying this analysis.

For transportation applications, emissions are generated at several stages between the initial point of energy feedstock extraction and delivering fuel to vehicles' fuel tanks or energy storage systems; in lifecycle analysis, these are often referred to “upstream” or “well-to-tank” emissions. In contrast, “downstream” or “tank-to-wheel” emissions are primarily comprised of those emitted by vehicles' exhaust systems, but also include other emissions generated during vehicle refueling, use, and inactivity (called `soaking'), including hydrofluorocarbons leaked from vehicles' air conditioning (AC) systems. They also include particulate matter (PM) released into the atmosphere by brake and tire wear (BTW) as well as evaporation of volatile organic compounds (VOCs) from fuel pumps and vehicles' fuel storage systems during refueling and when parked. Cumulative emissions occurring throughout the fuel supply and use cycle are often called “well-to-wheel” emissions in lifecycle analysis.

The CAFE Model tracks vehicle populations and activity levels to produce estimates of the effects of different levels of CAFE standards on emissions and their consequences for human health and the global climate. Tracking vehicle populations begins with the reference baseline or analysis fleet, and estimates of each vehicle's fuel type ( e.g., gasoline, diesel, electricity), fuel economy, and number of units sold in the U.S. As fuel economy-improving technology is added to vehicles in the reference baseline fleet in MYs subject to proposed new standards, the CAFE Model estimates annual rates at which new vehicles are purchased, driven, and subsequently scrapped. The model uses estimates of vehicles remaining in service in each year and the amount those vehicles are driven ( i.e., activity levels) to calculate the quantities of each type of fuel or energy that vehicles in the fleet consume in each year, including gasoline, diesel, and electricity. The quantities of travel and fuel consumption estimated for the cross section of MYs comprising each CYs vehicle fleet represents the “activity levels” the CAFE model uses to calculate emissions. The model does so by multiplying each activity level by the relevant emission factor and summing the results of those calculations.

Emission factors measure the mass of each greenhouse gas or criteria air pollutant emitted per unit of activity, which can be a vehicle-mile of travel, gallon of fuel consumed, or unit of fuel energy content. We generate emission factors for the following regulated criteria pollutants and GHGs: carbon monoxide (CO), VOCs, nitrogen oxides (NO_X), sulfur oxides (SO_X), particulate matter with a diameter of 2.5-micron (μm) or less (PM_2.5); CO₂, methane (CH₄), and nitrous oxide (N₂ O). In this rulemaking, upstream emission factors are based on the volume of each type of fuel supplied, while downstream emission factors are expressed on a distance-traveled (VMT) basis. Simply stated, the rulemaking's upstream emission inventory is the product of the per-gallon emission factor and the corresponding number of gallons of gasoline or diesel, or amount of electricity, produced and distributed. Similarly, the downstream emission inventory is the product of the per-mile emission factor and the appropriate miles traveled estimate. The only exceptions are that tailpipe emissions of SO_X and CO₂ are also calculated on a per-gallon emission basis using appropriate emission factors in the CAFE Model. EVs do not produce combustion-related (tailpipe) emissions, however, EV upstream electricity emissions are also accounted for in the CAFE Model inputs. Upstream and downstream emission factors and subsequent inventories were developed independently from separate data sources, as discussed in detail below.

The analysis for the NPRM used upstream emission factors derived from GREET 2022, which is a lifecycle emissions model developed by the U.S. DOE's Argonne National Laboratory (Argonne). GREET 2022 projected a national mix of fuel sources used for electricity generation (often simply called the grid mix) for transportation from the latest AEO data available, in that case from 2022. For the final rule, we updated upstream petroleum (gasoline and diesel) and electricity emission factors using R&D GREET 2023. Petroleum emission factors are based on R&D GREET 2023 assumptions derived from AEO 2023, while electricity emission factors are derived from an electricity forecast from the National Renewable Energy Laboratory's 2022 Standard Scenarios report. A detailed description of how we used R&D GREET 2023 to generate upstream emission factors appears in Chapter 5 of the TSD, as well as in the Electricity Grid Forecasts docket memo accompanying this rule.

Other grid mixes with higher penetrations of renewables are presented as sensitivity cases in the FRIA and provide some context about how the results of our analysis would differ using a grid mix with a higher penetration of renewable energy sources. We sought comment on these sensitivity cases and which national grid mix forecast best represents the latest market conditions and policies, such as the Inflation Reduction Act. We also sought comments on other forecasts to consider, including EPA's Integrated Planning Model for the post-IRA 2022 reference case for the final rulemaking, and the methodology used to generate alternate forecasts. We received no comments on our grid mix assumptions; however, to be consistent with DOE's projections in their Petroleum Equivalency Factor (PEF) final rule, we chose to use the 2022 Standard Scenarios report projections.

As in past CAFE analyses, we used GREET to derive emission factors for the following four upstream emission processes for gasoline, E85, and diesel: (1) petroleum extraction, (2) petroleum transportation and storage, (3) petroleum refining, and (4) fuel transportation, storage, and distribution (TS&D)). We calculated average emission factors for each fuel and upstream process during five-year intervals over the period from 2022 through 2050. We considered feedstocks including conventional crude oil, oil sands, and shale oils in the gasoline and diesel emission factor calculations and follow assumptions consistent with the GREET Model for ethanol blending.

In the proposal, NHTSA assumed that any reduction in fuel consumption within the United States would lead to an equal increase in gasoline exports. As a consequence, we projected that domestic fuel production and the upstream emissions it generates would not change, although we did acknowledge that emissions from feedstock extraction and fuel production outside the U.S. were likely to be affected. NHTSA also noted that this assumption was strong and that it was considering how to project changes in domestic fuel production that were likely to result from changes in CAFE and fuel efficiency standards over the long run. NHTSA sought comments on how it should model the response of domestic fuel production to changes in fuel consumption. AFPM commented that the scale of reductions in domestic fuel consumption caused by the proposed standards was likely to cause changes in domestic fuel production, and that NHTSA should consider the rule's impact on biofuel production.

NHTSA re-analyzed projections of domestic fuel production from McKinsey & Company (2023), S&P Global (2023), and the 2023 AEO, and concluded that there is a wide range of estimates about how domestic refining is likely to change over the coming decades, even without considering the potential effects of higher standards. Instead of relying on a single set of projections, NHTSA developed a simplified parameterized economic model for estimating the response of domestic fuel production to changes in U.S. fuel consumption. Using this model, for the final rule NHTSA estimates that 20 percent of the reduction in fuel consumption will be translated into reductions in domestic fuel production. See Chapters 5 and 6.2.4 of the TSD for a more detailed discussion of this process.

We estimated non-CO₂ downstream emission factors for gasoline, E85, diesel, and CNG using EPA's Motor Vehicle Emission Simulator (MOVES4) model, a regulatory highway emissions inventory model developed by that agency's National Vehicle and Fuel Emissions Laboratory. We generated downstream CO₂ emission factors based on the carbon content ( i.e., the fraction of each fuel type's mass that is carbon) and mass density per unit of each specific type of fuel, under the assumption that each fuel's entire carbon content is converted to CO₂ emissions during combustion. The CAFE Model calculates CO₂ vehicle-based emissions associated with vehicle operation of the surviving on-road fleet by multiplying the number of gallons of each specific fuel consumed by the CO₂ emission factor for that type of fuel. More specifically, the number of gallons of a particular fuel is multiplied by the carbon content and the mass density per unit of that fuel type, and then the ratio of CO₂ emissions generated per unit of carbon consumed during the combustion process is applied. TSD Chapter 5.3 contains additional detail about how we generated the downstream emission factors used in this analysis.

With stringent LDV standards already in place for PM from vehicle exhaust, particles from brake and tire wear (BTW) are becoming an increasingly important component of PM_2.5 emission inventories. To put the magnitude of future BTW PM_2.5 emissions in perspective, NHTSA conducted MOVES4 analysis using default input values. This analysis indicates that BTW PM_2.5 represent approximately half of gasoline-fueled passenger car and light truck PM_2.5 emissions (from vehicle exhaust, brake wear, and tire wear) after 2020. While previous CAFE rulemakings have not modeled the indirect impacts to BTW emissions due to changes in fuel economy and VMT, this rulemaking considers total PM_2.5 emissions from the vehicle's exhaust, brakes, and tires.

As with downstream emission factors, we generated BTW emission factors using EPA's MOVES4 model. Due to limited BTW measurements, MOVES does not estimate variation in BTW emission factors by vehicle MY, fuel type, or powertrain. Instead, MOVES' estimates of emissions from brake wear are based on weight-based vehicle regulatory classes and operating behavior derived primarily from vehicle speed and acceleration. On the other hand, MOVES' estimates of tire wear emissions depend on the same weight-based regulatory classes, but the effect of operations on emissions is represented only by vehicle speed. Unlike the CAFE Model's downstream emission factors, the BTW estimates were averaged over all vehicle MYs and ages to yield a single grams-per-mile value by regulatory class.

There is some evidence that average vehicle weight will differ by fuel type and powertrain, particularly for longer-range EVs, which are often heavier than a comparable gasoline- or diesel-powered vehicle due to the weight of the battery. This weight increase may result in additional tire wear. While regenerative braking often extends braking systems' useful life and reduces emissions associated with brake wear, the effect of additional mass might be to increase overall BTW emissions. Further BTW field studies are needed to better understand how differences in vehicle fuel and powertrain type are likely to impact PM_2.5 emissions from BTW. The CAFE Model's BTW inputs can be differentiated by fuel type, but for the time being are assumed to have equivalent values for gasoline, diesel, and electricity. Given the degree to which PM_2.5 inventories are expected to shift from vehicle exhaust to BTW in the near future, we believe that it is better to have some BTW estimates—even if imperfect—than not to include them at all, as was the case in prior CAFE rulemakings.

In the NPRM, we sought comment on this updated approach and on additional data sources that could be used to update the BTW estimates. Commenters such as the Alliance for Automotive Innovation and Stellantis recommended that NHTSA refrain from including BTW in the analysis until SAE or another organization publishes a measurement methodology and testing procedures for quantifying BTW. Another commenter, the AFPM, stated that new ZEVs specifically would cause an increase in average vehicle weight in the U.S. fleet, and in turn cause more BTW emissions.

With notable reductions in fine particulate matter (PM_2.5) from tailpipe exhaust due to federal regulation, non-exhaust sources such as brake and tire wear (BTW) constitute a growing proportion of vehicles' PM_2.5 emissions. Although we agree with commenters that EVs could cause disproportionate brake wear compared to internal combustion engine vehicles due to additional battery weight, it is unclear how this might affect LD and HDPUV PM emissions overall. Without any BEV tailpipe exhaust and some evidence to suggest reduced EV brake wear from regenerative braking, NHTSA has not yet been able to determine the relative PM contributions of BEVs, HEVs, and ICE vehicles. In addition, as discussed in more detail in Section III.D, it appears that the trend for manufacturers to produce large EVs may be declining as manufacturers start building smaller and more affordable EVs. While this final rule continues to project differences in BTW emissions among regulatory classes, there has not been enough new BTW data published since the proposal to update non-exhaust PM emission factors by fuel type. That said, we continue to believe that including the best available data on BTW estimates is better than including no estimates. For further reading on BTW assumptions, please refer to TSD Chapter 5.3.3.4.

The CAFE Model computes select health impacts resulting from population exposure to PM_2.5. These health impacts include causing or aggravating several different respiratory conditions and even premature death, each of which is measured by the number of instances predicted to result from exposure to each ton of PM_2.5 -related pollutant emitted (direct PM as well as NO_X and SO₂, both precursors to secondarily-formed PM_2.5). The CAFE Model reports total PM_2.5 -related health impacts by multiplying the estimated emissions of each PM_2.5 -related pollutant (in tons)—generated using the process described above—by the corresponding health incidence per ton value. Broadly speaking, a health incidence per ton value is the morbidity and mortality estimate linked to an additional ton of an emitted pollutant; these can also be referred to as benefit per ton values where monetary measures of adverse health impacts avoided per ton by which emissions are reduced (discussed further in Section III.G).

The American Lung Association commented on the limits of the health impacts analysis, stating that it “does not include monetized health harms of ozone, ambient oxides of nitrogen or air toxics.” We do not include monetized health harms of air toxics as they have not typically been monetized, and as such we currently have no basis for that valuation. The sources used in our health impacts analysis were chosen to best match the pollution source sector categories incorporated in the CAFE Model. For some pollution source sectors, only PM_2.5 BPT values exist, and as such we chose to consistently measure the same damages across all pollution source sectors by focusing on PM_2.5 -related damages. We plan to revisit this portion of analysis when more source sector BPT values become available in the literature. We do note that these benefits (reduced health harms of ozone, ambient oxides of nitrogen, air toxics) are potentially significant despite not being quantified and have added language to our discussion of benefits of the rule to clarify this.

The health incidence per ton values in this analysis reflect the differences in health impacts arising from the five upstream emission source sectors that we use to generate upstream emissions (petroleum extraction, petroleum transportation, refineries, fuel transportation, storage and distribution, and electricity generation). We carefully examined how each upstream source sector is defined in GREET to appropriately map the emissions estimates to data on health incidences from PM_2.5 -related pollutant emissions. As the health incidences for the different source sectors are all based on the emission of one ton of the same pollutants, NO_X, SO_X, and directly-emitted PM_2.5 , differences in the incidence per ton values arise from differences in the geographic distribution of each pollutant's emissions, which in turn affects the number of people exposed to potentially harmful concentrations of each pollutant.

As in past CAFE analyses, we relied on publicly available scientific literature and reports from EPA and EPA-affiliated authors, to estimate per-ton PM_2.5 -related health damage costs for each upstream source of emissions. We used several EPA reports to generate the upstream health incidence per ton values, as different EPA reports provided more up-to-date estimates for different sectors based on newer air quality modeling. These EPA reports use a reduced-form benefit-per-ton (BPT) approach to assess health impacts; PM_2.5 -related BPT values are the total monetized human health benefits (the sum of the economic value of the reduced risk of premature death and illness) that are expected to result from avoiding one ton of directly-emitted PM_2.5 or PM_2.5 precursor such as NO_X or sulfur dioxide (SO₂). We note, however, that the complex, non-linear photochemical processes that govern ozone formation prevent us from developing reduced-form ozone, ambient NO_X, or other air toxic BPT values, an important limitation to recognize when using the BPT approach. We include additional discussion of uncertainties in the BPT approach in Chapter 5.4.3 of the TSD and also conduct full-scale photochemical modeling described in Appendix E of the FEIS. Nevertheless, we believe that the BPT approach provides reasonable estimates of how establishing more stringent CAFE standards is likely to affect public health, and of the value of reducing the health consequences of exposure to air pollution. The BPT methodology and data sources are unchanged from the 2022 CAFE rule, and stakeholders generally agreed that estimates of the benefits of PM_2.5 reductions were improved from prior analyses based on our emissions-related health impacts methodology updated for that rule.

The reports we relied on for health incidences and BPT estimates include EPA's 2018 technical support document titled Estimating the Benefit per Ton of Reducing PM_2.5 Precursors from 17 Sectors (referred to here as the 2018 EPA source apportionment TSD), a 2018 oil and natural gas sector paper (Fann et al.), which estimates health impacts for this sector in the year 2025, and a 2019 paper (Wolfe et al.) that computes monetized per ton damage costs for several categories of mobile sources, based on vehicle type and fuel type.

Some CAFE Model upstream emissions components do not correspond to any single EPA source sector identified in available literature, so we used a weighted average of different source sectors to generate those values. Data we used from each paper for each upstream source sector are discussed in detail in Chapter 5.4 of the TSD.

The CAFE Model follows a similar process for computing health impacts resulting from downstream emissions. We used the Wolfe et al. paper to compute monetized damage costs per ton values for several on-road mobile sources categories based on vehicle type and fuel type. Wolfe et al. did not report incidences per ton, but that information was obtained through communications with the study authors. Additional information about how we generated downstream health estimates is discussed in Chapter 5.4 of the TSD.

We are aware that EPA recently updated its estimated benefits for reducing PM_2.5 from several sources, ⁷⁰⁸ but those do not include mobile sources (which include the vehicles subject to CAFE and HDPUV fuel efficiency standards). After discussion with EPA staff, we retained the PM_2.5 incidence per ton values from the previous CAFE analysis for consistency with the current mobile source emissions estimates.

Although we did not discuss doing a quantitative lifecycle analysis in the preamble of the NRPM, several commenters stressed the importance of lifecycle analysis, identified suitable methods for conducting such an analysis, and suggested how the results of such an analysis should factor into the finding that final standards indeed meet the “maximum feasible” test. The Agency understands the concern that many commenters have with the potential environmental impacts of vehicle production, including battery material extraction, manufacturing, and end-vehicle and battery disposal. With rapidly expanding EV production, this is a fast-evolving area of research and not one that can be fully addressed in this rule. While some evidence suggests that emissions from vehicle production would likely be greater for EVs than conventionally fueled vehicles, there is also evidence that ICEs continue to have greater total lifecycle emissions than EVs, depending on where the EV is charged. NHTSA is not yet prepared to quantify these relative vehicle cycle impacts. Further investigation across different fuels and vehicle powertrains is warranted and is currently underway with Argonne National Laboratory. For a review of relevant research and additional qualitative discussion on the vehicle cycle and its impacts, readers should refer to FEIS Chapter 6 (Lifecycle Analysis).

G. Simulating Economic Impacts of Regulatory Alternatives

The following sections describe NHTSA's approach for measuring the economic costs and benefits that would result from establishing alternative standards for future MYs. The measures that NHTSA uses are important considerations, because as OMB Circular A-4 states, benefits and costs reported in regulatory analyses must be defined and measured consistently with economic theory and should also reflect how alternative regulations are anticipated to change the behavior of producers and consumers from a baseline scenario. For both the fuel economy and fuel efficiency standards, those include vehicle manufacturers, buyers of new vehicles, owners of used vehicles, and suppliers of fuel, all of whose behavior is likely to respond in complex ways to the level of standards that DOT establishes for future MYs.

A number of commenters asked the agency to more explicitly account for effects that occur in the analytical baseline in the agency's incremental cost-benefit analysis. The agency responds substantively to those comments below. The typical approach to quantifying the impacts of regulations implies that these costs and benefits should be excluded from the incremental cost-benefit analysis given these effects are assumed to occur absent the regulation. Thus, quantifying them in the incremental cost-benefit analysis would obscure the effects the agency needs to isolate in order to analyze the effects of the regulation. For these reasons, the agency does not explicitly account for some of the costs and benefits requested by commenters that accrue in the baseline, and instead focuses on the costs and benefits that may change in response to the final rule.

It is also important to report the benefits and costs of this final rule in a format that conveys useful information about how those impacts are generated, while also distinguishing the economic consequences for private businesses and households from the action's effects on the remainder of the U.S. economy. A reporting format will accomplish this objective to the extent that it clarifies who incurs the benefits and costs of the final rule, while also showing how the economy-wide or “social” benefits and costs of the final rule are composed of direct effects on vehicle producers, buyers, and users, plus the indirect or “external” benefits and costs it creates for the general public. NHTSA does not attempt to distinguish benefits and costs into co-benefits or secondary costs.

Table III-7 lists the economic benefits and costs analyzed in conjunction with this final rule, and where to find explanations for what we measure, why we include it, how we estimate it, and the estimated value for that specific line item. The table also shows how the different elements of the analysis piece together to inform NHTSA's estimates of private and external costs and benefits.

NHTSA reports the costs and benefits of standards for LDVs and HDPUVs separately. While the effects are largely the same for the two fleets, our fuel economy and fuel efficiency programs are separate, and NHTSA makes independent determinations of the maximum feasible standards for each fleet.

A standard function of regulatory analysis is to evaluate tradeoffs between impacts that occur at different points in time. Many Federal regulations involve costly upfront investments that generate future benefits in the form of reductions in health, safety, or environmental damages. To evaluate these tradeoffs, the analysis must account for the social rate of time preference—the broadly observed social preference for benefits that occur sooner versus those that occur further in the future. This is accomplished by discounting impacts that occur further in the future more than impacts that occur sooner.

OMB Circular A-4 (2003) affirms the appropriateness of accounting for the social rate of time preference in regulatory analyses and recommends discount rates of 3 and 7 percent for doing so. The recommended 3 percent discount rate was chosen to represent the “consumption rate of interest” approach, which discounts future costs and benefits to their present values using the rate at which consumers appear to make tradeoffs between current consumption and equal consumption opportunities when deferred to the future. OMB Circular A-4 (2003) reports an inflation-adjusted or “real” rate of return on 10-year Treasury notes of 3.1 percent between 1973 and its 2003 publication date and interprets this as approximating the rate at which society is indifferent between consumption today and in the future. The 7 percent rate reflects the opportunity cost of capital approach to discounting, where the discount rate approximates the forgone return on private investment if the regulation were to divert resources from capital formation. Fuel savings and most other benefits from tightening standards will be experienced directly by owners of vehicles that offer higher fuel economy and thus affect their future consumption opportunities, while benefits or costs that are experienced more widely throughout the economy will also primarily affect future consumption. Circular A-4 indicates that discounting at the consumption rate of interest is the “analytically preferred method” when effects are presented in consumption-equivalent units. Thus, applying OMB's guidance to NHTSA's final rule suggests the 3 percent rate is the appropriate rate. However, NHTSA reports both the 3 and 7 percent rates for transparency and completeness. It should be noted that the OMB finalized a revision to Circular A-4 on November 9th, 2023. The 2023 Circular A-4 is effective for NPRMs, IFRs, and direct final rules submitted to OMB on or after March 1st, 2024, while the effective date for other final rules is January 1st, 2025. Thus, while NHTSA has considered the guidance in the revised circular for the final rule, as this final rule will be published before January 1, 2025, the agency will continue to use the discount rates in the prior version for the primary analysis. The agency performed a sensitivity case using a 2 percent social discount rate consisted with the guidance of revised Circular A-4 (2023) which can be found in Chapter 9 of the RIA.

A key exception to Circular A-4's guidance on social discounting implicates the case of discounting climate related impacts. Because some GHGs emitted today can remain in the atmosphere for hundreds of years, burning fossil fuels today not only imposes uncompensated costs on others around the globe today, but also imposes uncompensated damages on future generations. As OMB Circular A-4 (2003) indicates “special ethical considerations arise when comparing benefits and costs across generations” and that future citizens impacted by a regulatory choice “cannot take part in making them, and today's society must act with some consideration of their interest.” Thus, NHTSA has elected to discount these effects from the year of abatement back to the present value with lower rates. For further discussion, see Section III.G.2.b(1) of the Preamble.

For a complete discussion of the methodology employed and the results, see Chapter 6 of the TSD and Chapter 8 of the RIA, respectively. The safety implications of the final rule—including the monetary impacts—are reserved for Section III.H.

1. Private Costs and Benefits

a. Costs to Consumers

(1) Technology Costs

The technology applied to meet the standards would increase the cost to produce new cars, light trucks and HDPUVs. Within this analysis, manufacturers are assumed to transfer these costs to the consumers who purchase vehicles offering higher fuel economy. While NHTSA recognizes that some manufacturers may defray their regulatory costs for meeting increased fuel economy and fuel efficiency standards through more complex pricing strategies or by accepting lower profits, NHTSA lacks sufficient insight into manufacturers' pricing strategies to confidently model alternative approaches. Thus, we simply assume that manufacturers raise the prices of models whose fuel economy they elect to improve sufficiently to recover their increased costs for doing so. The technology costs are incurred by manufacturers and then passed onto consumers. While we include the effects of IRA tax credits in our modeling of consumer responses to the standards, the effect of the tax credit is an economic transfer where the costs to one party are exactly offset by benefits to another and have no impact on the net benefits of the final rule. While NHTSA could include IRA tax credits as a reduction in the technology costs for manufacturers and purchasing prices in our cost-benefit accounting, tax credits are a transfer from the government to private parties, and as such have no net effect on the benefits or costs of the final rule. As such, the line item included in the tables summarizing the cost of technology throughout this final rule should be considered pre-tax unless otherwise noted.

NHTSA did not receive comments pertaining to this topic. See Section III.C.6 of this preamble and Chapter 2.5 of the TSD for more details.

(2) Consumer Sales Surplus

Consumers who forgo purchasing a new vehicle because of the increase in the price of new vehicles' prices caused by more stringent standards will experience a decrease in welfare. The collective welfare loss to these “potential” new vehicle buyers is measured by their foregone consumer surplus.

Consumer surplus is a fundamental economic concept and represents the net value (or net benefit) a good or service provides to consumers. It is measured as the difference between what a consumer is willing to pay for a good or service and its market price. OMB Circular A-4 explicitly identifies consumer surplus as a benefit that should be accounted for in cost-benefit analysis. For instance, OMB Circular A-4 states the “net reduction in total surplus (consumer plus producer) is a real cost to society,” and elsewhere recommends that consumer surplus values be monetized “when they are significant.”

Accounting for the limited portion of lifetime fuel savings that the average new vehicle buyer values, and holding all else equal, higher average prices should depress new vehicle sales and by extension reduce consumer surplus. The inclusion of the effects on the final rule on consumer surplus is not only consistent with OMB guidance, but with other parts of this regulatory analysis. For instance, we calculate the increase in consumer surplus associated with increased driving that results from the lower CPM of driving under more stringent regulatory alternatives, as discussed in Section II.G.1.b(3). The surpluses associated with sales and additional mobility are inextricably linked, as they capture the direct costs and benefits to purchasers of new vehicles. The sales surplus captures the welfare loss to consumers when they forego purchasing new vehicles because of higher prices, while the consumer surplus associated with additional driving measures the benefit of the increased mobility it provides.

NHTSA estimates the loss of sales surplus based on the change in quantity of vehicles projected to be sold, after adjusting for quality improvements attributable to higher fuel economy or fuel efficiency. Several commenters mention that there may be distributional impacts in terms of the less financially privileged not being able to afford higher priced vehicles. Consumers in rural areas are specifically mentioned as being adversely affected due to the higher cost of charging an EV in rural areas which would presumably act as a barrier to purchasing one of these vehicles.

While these commenters allege that consumers will be harmed by the inability to purchase new vehicles because of the regulations, commenters did not provide any evidence to support that these effects will, or even likely to occur, and seemingly ignored how these communities may value and benefit from reduced operational costs. Regardless, NHTSA accounted for the possibility that there would be a change in welfare associated with decreased sales, but NHTSA did not receive any comments suggesting that its estimation of the consumer sales surplus was inadequate. Nor did any commenters suggest changes to the agency's methodology. As such, the agency has elected to use the same methodology as the proposal and feels that the lost welfare from the consumer sales surplus adequately captures the effects raised by commenters. Furthermore, the IRA provides a 30% tax credit for qualified alternative fuel vehicle refueling property supporting the installation of charging infrastructure in low-income and non-urban areas. For additional information about consumer sales surplus, see Chapter 6.1.2 of the TSD.

(3) Ancillary Costs of Higher Vehicle Prices

Some costs of purchasing and owning a new or used vehicle increase in proportion to its purchase price or market value. At the time of purchase, the price of the vehicle combined with the state-specific tax rate determine the sales tax paid. Throughout the lifetime of the vehicle, the residual value of the vehicle—which is determined by its initial purchase price, age, and accumulated usage—determine value-related registration fees and insurance premiums. The analysis assumes that the transaction price is a fixed share of the MSRP, which allows calculation of these factors as shares of MSRP. As the standards influence the price of vehicles, these ancillary costs will also increase. For a detailed explanation of how NHTSA estimates these costs, see Chapter 6.1.1 of the TSD. These costs are included in the consumer per-vehicle cost-benefit analysis but not in the societal cost-benefit analysis, because they are assumed to be transfers from consumers to government agencies or to reflect actuarially “fair” insurance premiums. NHTSA did not receive any comments about its treatment of state sales taxes or changes to insurance premiums.

In previous proposals and final rules, NHTSA also included the costs of financing vehicle purchases as an ancillary cost to consumers. However, as we noted in the 2022 final rule, the availability of vehicle financing offers a benefit to consumers by spreading out the costs of additional fuel economy technology over time. Thus, we no longer include financing as a cost to consumers. Lucid supports NHTSA's decision to exclude financing as an ancillary cost, recognizing the benefit of smoothing out consumer costs over time. NADA and MEMA have mentioned that the majority of prospective new vehicle purchasers finance their transactions, and expressed concern that higher interest rates may be impacting the affordability of financing and that consumer credit may not reach to meet changing vehicle prices. NHTSA has determined it is appropriate to continue to exclude these costs from the analysis for the following reasons. With regards to the impact of increasing vehicle purchasing costs, as previously mentioned, NHTSA calculates and includes the change in consumer surplus of those who choose not to purchase a new vehicle as a result of higher vehicle prices due to the stringency of the standards. In addition, explicitly modeling future long-run changes in financing costs due to changes in interest rates is a technically uncertain undertaking and outside the current bounds of this work. Forecasting long-run interest rates includes making a variety of assumptions on the structure that these rates might take, such as a random walk or equivalence to a forward rate and are subject to numerous exogenous macroeconomic factors and uncertainties. Commenters did not identify any long-run projections that supported their conclusions pertaining to this aspect of consumer costs. Therefore, it is inaccurate to assume that high interest rates at one point in time will lead to higher rates (and therefore higher costs) for all consumers during the regulatory period.

b. Benefits to Consumers

(1) Fuel Savings

The primary benefit to consumers of increasing standards is the savings in future fuel costs that accrue to buyers and subsequent owners of new vehicles. The value of fuel savings is calculated by multiplying avoided fuel consumption by retail fuel prices. Each vehicle of a given body style is assumed to be driven the same amount in each year of its lifetime as all those of comparable age and body style. The ratio of that cohort's annual VMT to its fuel efficiency produces an estimate of its yearly fuel consumption. The difference between fuel consumption in the No-Action Alternative, and in each regulatory alternative, represents the gallons (or energy content) of fuel saved.

Under this assumption, our estimates of fuel consumption from increasing the fuel economy or fuel efficiency of each individual model depend only on how much its fuel economy or efficiency is increased, and do not reflect whether its actual use differs from other models of the same body type. Neither do our estimates of fuel consumption account for variation in how much vehicles of the same body type and age are driven each year, which appears to be significant (see Chapter 4.3.1.2 of the TSD). Consumers save money on fuel expenditures at the average retail fuel price (fuel price assumptions are discussed in detail in Chapter 4.1.2 of the TSD), which includes all taxes and represents an average across octane blends. For gasoline and diesel, the included taxes reflect both the Federal tax and a calculated average state fuel tax. Expenditures on alternative fuels (E85 and electricity, primarily) are also included in the calculation of fuel expenditures, on which fuel savings are based. However, since alternative fuel technology is not applied to meet the standards, the majority of the costs associated with operating alternative fuels net to zero between the reference baseline and action alternatives. And while the included taxes net out of the social benefit cost analysis (as they are a transfer), consumers value each gallon saved at retail fuel prices including any additional fees or taxes they pay.

Chapter 6.1.3 of the TSD provides additional details. As explained in the TSD, NHTSA considers the possibility that several of the assumptions made about vehicle use could lead to misstating the benefits of fuel savings. NHTSA notes that these assumptions are necessary to model fuel savings and likely have minimal impact to the accuracy of the analysis for this final rule.

A variety of commenters discussed how fuel savings are valued by both manufacturers and consumers, with some discussion on whether NHTSA has under or over-valued the benefits to consumers, the appropriate use of discount rate to apply to fuel savings, and the source of data used to project fuel savings. AEI commented that the “inclusion of fuel savings is illegitimate as a component of the `benefits' the [rule] because the economic benefits of fuel savings are captured fully by consumers of the fuel.” Conversely, IPI commented that including all fuel savings as a benefit of the rule is appropriate because the rule is addressing the energy efficiency gap.

NHTSA agrees with IPI that fuel savings should be accounted for within the rule. AEI's comment is premised on the theory that the vehicle market is efficient and therefore consumers must not value fuel savings, and NHTSA's regulations may only address market failures that address externalities. As discussed in III.E, the energy efficiency gap has long been recognized as a market failure that may impact the ability of consumers to realize fuel savings. Furthermore, the notion that only externalities may be counted as a benefit is unfounded. Executive Order 12866 and Circular A-4 (2003) have long required agencies to attempt to quantify as many benefits as possible and costs that can reasonably be ascertained and quantified into its analysis, and courts have frowned upon federal agencies ignoring known and quantifiable costs or benefits. In addition, how the agency quantifies and monetizes this benefit is not the same as how the agency considers it in making its determination of what standards are “maximum feasible,” and thus the extent to which the agency should consider consumer fuel savings is addressed in that discussion.

NADA commented that “NHTSA correctly noted that EV owners will save refueling time by charging at home, but the analysis is flawed in that it does not account for the impact of increased electricity consumption and related expenditures for those who charge at home.” NADA is incorrect in their assertion that NHTSA ignores the cost of recharging at home. The fuel savings benefit is derived from all fuel sources consumed—including electricity—and is intended to capture the total cost spent to refuel and recharge in each alternative.

Some commenters argued that NHTSA's use of static electricity price projections could lead to an underestimate of the operating costs of BEVs. The Heritage Foundation and NADA both argued that increased demand for electricity induced by BEV adoption—which happens solely in the analytical reference baseline through the end of the standard setting years—would necessitate increased investment in the electricity grid and thus lead to higher electricity prices to recover the costs of these investments. The Heritage Foundation also suggested that NHTSA's cost-benefit analysis should account for incremental infrastructure costs required to comply with changes to the standards. NHTSA believes it is properly accounting for the impact of greater penetration of BEVs on electricity prices in its regulatory analysis. The electricity prices used in its analysis are taken from AEO 2023 and represent EIA's best projection of how greater electrification in the automobile market will impact electricity prices. Due to its statutory constraints under EPCA, NHTSA does not permit production of BEVs as a compliance strategy during model years for which it is establishing standards, which restricts BEV adoption to the reference baseline. NHTSA believes that the modest difference in projected adoption of BEVs between even the most stringent alternatives and the reference baseline is unlikely to necessitate significant additional investment in the electricity generation and distribution grid beyond the No-Action Alternative, and thus will have only minimal effects on electricity prices. NHTSA's choice not to account for potential effects of its standards on future electricity prices in its analysis of costs and benefits is consistent with the agency's treatment of fuel prices, which is discussed in TSD Chapter 6.2.4.

Some commenters, such as the Center for Environmental Accountability, argued that electricity prices charged to users of public charging stations are somewhat higher on average than those of at home charging. NHTSA believes that at-home charging will continue to be the primary charging method during the time period relevant to this rulemaking, and thus residential electricity rates are the most representative electricity prices to use in our analysis. However, the agency notes again that electrification is restricted to the reference baseline through the standard setting years, accounting for the price difference between at-home versus public charging would result in minor differences between the alternatives that would have little impact in changing the net benefits of any of the scenarios.

Finally, there is some discussion among the commenters related to the appropriate choice of discount rate to apply to fuel savings. Valero suggests that valuing medium-term impacts at a discount rate of 3 percent is inappropriate due to the consumer's investment perspective, while CEA suggests that a 7 percent discount rate is a more appropriate choice over 3 percent due to differences paid for risk-free versus risky assets. Consumer Reports supports the use of a 3 percent discount rate in its calculation of discounted net savings for the consumer in the medium term.

NHTSA believes that is appropriate to account for fuel savings with the same 3 and 7 percent discount rates used for other costs and benefits, such as technology costs which are also accrued by consumers. This approach, as explained in Circular A-4, captures discount rates that reflect different preferences, and looking at both rates provides policy makers a more well-informed perspective. It is important to note that NHTSA's assumptions regarding how consumers value fuel savings at the time of new vehicle purchase do not apply to how NHTSA values fuel savings in its benefit-cost analysis. The prior discussion of the energy efficiency gap and consumer's undervaluation of lifetime fuel savings relates to the consumer decision in the vehicle market. NHTSA's societal-level benefit cost analysis includes the full lifetime fuel savings discounted using both 3 and 7 percent discount rates. Additional detail can be found in Chapter 4.2.1.1 of the TSD.

(2) Refueling Benefit

Increasing standards affects the amount of time drivers spend refueling their vehicles in several ways. First, higher standards increase the fuel efficiency of ICE vehicles produced in the future, which may increase their driving range and decrease the number of refueling events. Conversely, to the extent that more stringent standards increase the purchase price of new vehicles, they may reduce sales of new vehicles and scrappage of existing ones, causing more VMT to be driven by older and less efficient vehicles that require more refueling events for the same amount of driving. Finally, as the number of EVs in the fleet increases, some of the time spent previously refueling ICE vehicles at the pump will be replaced with recharging EVs at public charging stations. While the analysis does not allow electrification to be chosen as a compliance pathway with the standards for LDVs, it is still important to model recharging since excluding these costs would underestimate scenarios with additional BEVs, such as our sensitivity cases that examine lower battery costs.

NHTSA estimates these savings by calculating the amount of refueling time avoided—including the time it takes to locate a retail outlet, refuel one's vehicle, and pay—and multiplying it by DOT's estimated value of travel time. For a full description of the methodology, refer to Chapter 6.1.4 of the TSD. An alternative hypothesis NHTSA is still considering, but not adopting for the final rule, is whether manufacturers maintain vehicle range by lowering tank size as vehicle efficiency improves without, therefore, reducing refueling time.

NADA commented that the agency's assumption that EVs will only be recharged when necessary mid-trip is inaccurate. NADA noted that “many BEV owners and operators, particularly those living in urban areas, will not charge at home.” As noted earlier, NHTSA believes that most charging will occur in the home during time period relevant to this rulemaking, but NHTSA agrees with NADA that not all EV owners may have access to home charging. Commenters did not come forward with any specifics of how to best quantify these costs, but we may revisit these assumptions in the future when more information is available. For the time being, the agency believes that, even if it were to quantify the recharging time of EVs for non-mid-trip refuelings, the differences between the alternatives would be negligible given most of those costs would be incurred in the reference baseline.

(3) Additional Mobility

Any increase in travel demand provides benefits that reflect the value to drivers and passengers of the added—or more desirable—social and economic opportunities that additional travel makes available. Under each of the alternatives considered in this analysis, the fuel CPM of driving would decrease as a consequence of higher fuel economy and efficiency levels, thus increasing the number of miles that buyers of new cars, light trucks, and HDPUVs would drive as a consequence of the well-documented fuel economy rebound effect.

In theory, the decision by drivers and their passengers to make more frequent or longer trips when the cost of driving declines demonstrates that the benefits that they gain by doing so must exceed the costs they incur. At a minimum, one would expect the benefits of additional travel to equal the cost of the fuel consumed to travel additional miles (or they would not have occurred). Because the cost of that additional fuel is reflected in the simulated fuel expenditures, it is also necessary to account for the benefits associated with those extra miles traveled. But those benefits arguably should also offset the economic value of their (and their passengers') travel time, other vehicle operating costs, and the economic cost of safety risks due to the increase in exposure to crash risks that occurs with additional travel. The amount by which the benefit of this additional travel exceeds its economic costs measures the net benefits drivers and their passengers experience, usually referred to as increased consumer surplus.

Chapter 6.1.5 of the TSD explains NHTSA's methodology for calculating benefits from additional mobility. The benefit of additional mobility over and above its costs is measured by the change in consumers' surplus, which NHTSA approximates as one-half of the change in fuel CPM times the increase in VMT due to the rebound effect. In the proposal, NHTSA sought comments on the assumptions and methods used to calculate benefits derived from additional mobility. NHTSA received several comments addressing its approach for estimating the total change in VMT caused by changes in the standard. These comments are addressed in section III.E. However, NHTSA did not receive comments on its methodology for quantifying the related change in benefits from additional mobility.

When the size of the vehicle stock decreases in the LD alternative cases, VMT and fuel cost per-vehicle increase. Because maintaining constant non-rebound VMT assumes consumers are willing to pay the full cost of the reallocated vehicle miles, we offset the increase in fuel cost per-vehicle in the LD analysis by adding the product of the reallocated VMT and fuel CPM to the mobility value in the per-vehicle consumer analysis. Because we do not estimate other changes in cost per-vehicle that could result from the reallocated miles ( e.g., maintenance, depreciation, etc.) we do not estimate the portion of the transferred mobility benefits that would correspond to con'umers' willingness to pay for those costs. We do not estimate the con'umers' surplus associated with the reallocated miles because there is no change in total non-rebound VMT and thus no change in con'umers' surplus per consumer. Chapter 6.1.5 of the TSD explains NHTSA's methodology for calculating the benefits of reallocated miles. NHTSA sought comment in the proposal on its methodology for calculating the benefits from reallocated milage. NHTSA did not receive comments on this subject.

2. External Costs and Benefits

a. Costs

(1) Congestion and Noise

Increased vehicle use associated with the rebound effect also contributes to increased traffic congestion and highway noise. Although drivers obviously experience these impacts, they do not fully value their effects on other travelers or bystanders, just as they do not fully value the emissions impacts of their own driving. Congestion and noise costs are thus “external” to the vehicle owners whose decisions about how much, where, and when to drive more in response to changes in fuel economy result in these costs. Thus, unlike changes in the costs incurred by drivers for fuel consumption or safety risks they willingly assume, changes in congestion and noise costs are not offset by corresponding changes in the travel benefits drivers experience.

Congestion costs are limited to road users; however, since road users include a significant fraction of the U.S. population, changes in congestion costs are treated as part of the final rule's external economic impact on society as a whole instead of as a cost to private parties. Costs resulting from road and highway noise are even more widely dispersed because they are borne partly by surrounding residents, pedestrians, and other non-road users, and for this reason are also considered as costs that drivers impose on society as a whole.

To estimate the economic costs associated with changes in congestion and noise caused by increases in driving, NHTSA updated the estimates of per-mile congestion and noise costs from increased automobile and light truck use reported in FHWA's 1997 Highway Cost Allocation Study to account for changes in travel activity and economic conditions since they were originally developed, as well as to express them in 2021 dollars for consistency with other economic inputs. NHTSA employed a similar approach for the 2022 final rule. Because HDPUVs and light-trucks share similar operating characteristics, we also apply the noise and congestion cost estimates for light-trucks to HDPUVs.

See Chapter 6.2 of the TSD for details on how NHTSA calculated estimates of the economic costs associated with changes in congestion and noise caused by differences in miles driven. In the NPRM, NHTSA requested comment on the congestion costs employed in this analysis, but we did not receive any and have not changed our methodology from the NPRM for this final rule.

(2) Fuel Tax Revenue

As mentioned in Section II.G.1.b(1), a portion of the fuel savings experienced by consumers includes avoided fuel taxes. While fuel taxes are a transfer and do not affect net benefits, NHTSA reports an estimate of changes in fuel tax revenues together with external costs to show the potential impact on state and local government finances.

Several commenters, including AHUA and the ID, MT, ND, SD, and WY DOTs, discussed changes in the Highway Trust Fund as a result of changes in gasoline tax payment by consumers, and mentioned concern in funding for highway infrastructure, a potential cost that was not incorporated or accounted for in the rule. NHTSA reports changes in gasoline tax payments by consumers and in revenues to government agencies, and NHTSA's proposal explained in multiple places that gasoline taxes are considered a transfer—a cost to governments and an identical benefit to consumers that has already been accounted for in reported fuel savings—and have no impact on net benefits. As indicated above, any reduction in tax revenue received by governments that levy taxes on fuel is exactly offset by lower fuel tax payments by consumers, so from an economy-wide standpoint reductions in gasoline tax revenues are simply a transfer of economic resources and has no effect on net benefits. The agency notes that a decrease in revenue from gasoline taxes does not preclude alternative methods from funding the Highway Trust Fund or infrastructure, and—while fiscal policy is outside the scope of this rulemaking—some of the more hyperbolic claims that less fuel taxes “would threaten the viability of the national highway system” are clearly unfounded.

b. Benefits

(1) Climate Benefits

The combustion of petroleum-based fuels to power cars, light trucks, and HDPUVs generates emissions of various GHGs, which contribute to changes in the global climate and resulting economic damages. Extracting and transporting crude petroleum, refining it to produce transportation fuels, and distributing fuel all generate additional emissions of GHGs and criteria air pollutants beyond those from vehicle usage. By reducing the volume of petroleum-based fuel produced and consumed, adopting standards will thus mitigate global climate-related economic damages caused by accumulation of GHGs in the atmosphere, as well as the more immediate and localized health damages caused by exposure to criteria pollutants. Because they fall broadly on the U.S. population, and on the global population as a whole in the case of climate damages, reducing GHG emissions and criteria pollutants represents an external benefit from requiring higher fuel economy.

(a) Social Cost of Greenhouse Gases Estimates

NHTSA estimated the climate benefits of CO₂, CH₄, and N₂ O emission reductions expected from the proposed rule using the Interagency Working Group's (IWG) interim SC-GHG estimates presented in the Technical Support Document: SC of Carbon (SCC), Methane, and Nitrous Oxide Interim Estimates (“February 2021 TSD”). NHTSA noted in the proposal that E.O. 13990 envisioned these estimates to act as a temporary surrogate until the IWG could finalize new estimates. NHTSA acknowledged in the proposal that our understanding of the SC-GHG is still evolving and that the agency would continue to track developments in the economic and environmental sciences literature regarding the SC of GHG emissions, including research from Federal sources like the EPA. NHTSA sought comment on whether an alternative approach should be considered for the final rule.

On December 22, 2023, the IWG issued a memorandum to Federal agencies, directing them to “use their professional judgment to determine which estimates of the SC-GHG reflect the best available evidence, are most appropriate for particular analytical contexts, and best facilitate sound decision-making.” NHTSA determined that the 2023 EPA SC-GHG Report for the final rule would be the most appropriate estimate to use for the final rule.

NHTSA arrived at this decision for several reasons. E.O. 13990 tasked the IWG with devising long-term recommendations to update the methodologies used in calculating these SC-GHG values, based on “the best available economics and science,” and incorporating principles of “climate risk, environmental justice (EJ), and intergenerational equity.” The E.O. also instructed the IWG to take into account recommendations from the National Academies of the Sciences (NAS) committee convened on this topic, which were published in 2017. Specifically, the National Academies recommended that the SC-GHG should be developed using a modular approach, where the separate modules address socioeconomic projections, climate science, economic damages, and discounting. The NAS recommended that the methodology underlying each of the four modules be updated by drawing on the latest research and expertise from the scientific disciplines relevant to that module.

The 2023 EPA SC-GHG Report presents a set of SC-GHG estimates that incorporate the National Academies' near-term recommendations and reflects the most recent scientific evidence. The report was also subject to notice, comment, and a peer review to ensure the quality and integrity of the information it contains and concluded after NHTSA issued its proposal. NHTSA specifically cited EPA's proposed estimates and final external peer review report on EPA's draft methodology in its proposal, as that was the most up-to-date version of the estimates available as of the date of NHTSA's proposal. Several commenters, including IPI, suggested that the agency use EPA's estimates for the final rule. This is further discussed in subsection (c) of this Climate Benefits section. NHTSA believes the 2023 EPA SC-GHG Report represent the most comprehensive SC-GHGs estimates currently available. For additional details, see Chapter 6.2.1.1 of the TSD.

(b) Discount Rates for Climate Related Benefits

As mentioned earlier, NHTSA discounts non-climate benefits and costs at both the 3% consumption rate of interest and the 7% opportunity cost of capital, in accordance with OMB Circular A-4 (2003). Because GHGs degrade slowly and accumulate in the earth's atmosphere, the economic damages they cause increase as their atmospheric concentration accumulates. Some GHGs emitted today will remain in the atmosphere for hundreds of years, therefore, burning fossil fuels today not only imposes uncompensated costs on others around the globe today, but also imposes uncompensated damages on future generations. As OMB Circular A-4 (2003) indicates “special ethical considerations arise when comparing benefits and costs across generations” and that future citizens impacted by a regulatory choice “cannot take part in making them, and today's society must act with some consideration of their interest.” As the EPA's report states, “GHG emissions are stock pollutants, in which damages result from the accumulation of the pollutants in the atmosphere over time. Because GHGs are long-lived, subsequent damages resulting from emissions today occur over many decades or centuries, depending on the specific GHG under consideration.” NHTSA's analysis is consistent with the notion that intergenerational considerations merit lower discount rates for rules such as CAFE with impacts over very long-time horizons.

In addition to the ethical considerations, Circular A-4 also identifies uncertainty in long-run interest rates as another reason why it is appropriate to use lower rates to discount intergenerational impacts, since recognizing such uncertainty causes the appropriate discount rate to decline gradually over progressively longer time horizons. The social costs of distant future climate damages—and by implication, the value of reducing them by lowering emissions of GHGs—are highly sensitive to the discount rate, and the present value of reducing future climate damages grows at an increasing rate as the discount rate used in the analysis declines. This “non-linearity” means that even if uncertainty about the exact value of the long-run interest rate is equally distributed between values above and below the 3 percent consumption rate of interest, the probability-weighted (or “expected”) present value of a unit reduction in climate damages will be higher than the value calculated using a 3 percent discount rate. The effect of such uncertainty about the correct discount rate can be accounted for by using a lower “certainty-equivalent” rate to discount distant future damages, defined as the rate that produces the same expected present value of a reduction in future damages implied by the distribution of possible discount rates around what is believed to be the most likely single value.

For the final rule, NHTSA is updating its discount rates from the IWG recommendations to those found in the 2023 EPA SC-GHG Report. The EPA's discounting module represents an advancement on the work of the IWG in a number of ways. First, the EPA report uses the most recent evidence on the “consumption rate of interest”—the rate at which we observe consumers trading off consumption today for consumption in the future. Second, EPA's approach incorporates the uncertainty in the consumption rate of interest over time, specifically by using certainty-equivalent discount factors which effectively reduce the discount rate progressively over time, so that the rate applied to near-term avoided climate damages will be higher than the rate applied to damages anticipated to occur further in the future. Finally, EPA's revised approach incorporates risk aversion into its modeling framework,, to recognize that individuals are likely to be willing to pay some additional amount to avoid the risk that the actual damages they experience might exceed their expected level. This gives some consideration to the insurance against low-probability but high-consequence climate damages that interventions to reduce GHG emissions offer. For more detail, see the 2023 EPA SC-GHG Report.

When the streams of future emissions reductions being evaluated are moderate in terms of time (30 years or less), the EPA suggests to discount from the year of abatement to the present using the corresponding constant near-term target rates of 2.5, 2.0, and 1.5 percent. NHTSA's calendar year analysis includes fewer than 30 years of impacts (the calendar year captures emissions of all model years on the road through 2050), and the majority of emissions impacts considered in NHTSA's model year analysis also occur within this timeframe (vehicles in the MY analysis will continue to be on the road past 30 years, however nearly 97 percent of their lifetime emissions will occur during the first 30 years of their service given vehicles are used less as they age on average and a majority of the vehicles in this cohort will have already been retired completely from the fleet). Thus, NHTSA has elected to discount from the year of abatement back to the present value using constant near-term discount rates of 2.5, 2.0, and 1.5 percent. The 2023 EPA SC-GHG Report's central SC-GHG values are based on a 2 percent discount rate, and for this reason NHTSA presents SC-GHG estimates discounted at 2 percent alongside its primary estimates of other costs and benefits wherever NHTSA does not report the full range of SC-GHG estimates. The agency's analysis showing our primary non-GHG impacts at 3 and 7 percent alongside climate-related benefits may be found in Chapter 8 of the FRIA for both LDVs and HDPUVs. We believe that this approach provides policymakers with a range of costs and benefits associated with the rule using a reasonable range of discounting approaches and associated climate benefits.

NHTSA has also produced sensitivity analyses that vary the SC-GHG values, as discussed in Section V.D, by applying the IWG SC-GHG values. NHTSA finds net benefits in each of these sensitivity cases. Accordingly, NHTSA's conclusion that this rule produces net benefits is consistent across a range of SC-GHG choices.

For additional details, see Chapter 6.2.1.2 of the TSD. For costs and benefits calculated with SC-GHG values and corresponding discount rates of 2.5 percent and 1.5 percent, see Chapter 9 of tIRIA.

(c) Comments and Responses About the Agency's Choice of Social Cost of Carbon Estimates and Discount Rates

A wide variety of comments were received regarding the social cost of greenhouse gas emissions. The first category pertains to the inclusion of a SC-GHG value in cost-benefit analysis calculations. Commenters including IPI and NRDC proposed that NHTSA incorporates the updated SC-GHG values from EPA's 2023 Report in the final rule. Valero and others suggested that climate benefits, should they be included, be valued at discount rate above 7 percent. Other commenters mention that research in this area is ongoing, has a degree of uncertainty regarding the choice of underlying parameters and models, and that a global consensus value has not been reached, therefore such a measure should not be incorporated in the analysis.

Estimating the social costs of future climate damages caused by emissions of greenhouse gases, or SC-GHG, requires analysts to make a number of projections that necessarily involve uncertainty—for example, about the likely future pattern of global emissions of GHGs—and to model multifaceted scientific phenomena, including the effect of cumulative emissions and atmospheric concentrations of GHGs on climate measures including global surface temperatures and precipitation patterns. Each of these entail critical judgements about complex scientific and modeling questions. Doing so requires specialized technical expertise, accumulated experience, and expert judgment, and highly trained, experienced, and informed analysts can reasonably differ in their judgements. Further, in CBD v. ^NH TSA, the 9th Circuit concluded that uncertainty in SC-GHG estimates is not a reasonable excuse for excluding any estimate of the SC-GHG in the analysis of CAFE standards.

Commenters raise questions about the specific assumptions and parameter values used to produce the estimates of the social costs of various GHGs that NHTSA relied upon in the proposed regulatory analysis and contend that using alternative assumptions and values would reduce the recommended values significantly. The agency notes EPA's analysis, like the IWG's, includes experts in climate science, estimation of climate-related damages, and economic valuation of those impacts, and that these individuals applied their collective expertise to review and evaluate available empirical evidence and alternative projections of important measures affecting the magnitude and cost of such damages. We believe that EPA's update, which builds on the IWG's work, represents the best current culmination in the field and has been vetted by both the public and experts in the field during the peer review. As such, we believe that EPA's estimates best represent the culminative impact of GHGs analyzed by this rule.

DOT uses its own judgment in applying the estimates in this analysis. As a consequence, NHTSA views the chosen SC-GHG values as the most reliable among those that were available for it to use in its analysis. We feel that commenters did not address the inherent uncertainty in estimating the SC-GHG. Specifically, we note that any alternative model that attempts to project the costs of GHGs over the coming decades—and centuries—will be subject to the same uncertainty and criticisms raised by commenters.

A greater number of commenters mention the global scope involved in the calculation of the social cost of greenhouse gas emissions. Some contend that NHTSA should not consider any valuation which includes global benefits of reduced emissions, as the costs are incurred by manufacturers and consumers within the United States. In contrast, the Center for Biological Diversity, Environmental Defense Fund, and others comment that,

NHTSA appropriately focuses on a global estimate of climate benefits . . . While NHTSA offers persuasive justifications for this decision, many additional justifications further support this approach . . . The Energy Policy and Conservation Act (“EPCA”), National Environmental Policy Act, Administrative Procedure Act, and other key sources of law permit, if not require, NHTSA to consider the effects of U.S. pollution on foreign nations . . . Executive Order 13,990 instructs agencies to “tak[e] global damages into account” when assessing climate impacts because “[d]oing so facilitates sound decision-making, recognizes the breadth of climate impacts, and support the international leadership of the United States on climate issues.

NHTSA agrees that climate change is a global problem and that the global SC-GHG values are appropriate for this analysis. Emitting greenhouse gases creates a global externality, in that GHG emitted in one country mix uniformly with other gases in the atmosphere and the consequences of the resulting increased concentration of GHG are felt all over the world. The IWG concluded that a global analysis is essential for SC-GHG estimates because climate impacts directly and indirectly affect the welfare of U.S. citizens and residents through complex pathways that spill across national borders. These include direct effects on U.S. citizens and assets, investments located abroad, international trade, and tourism, and spillover pathways such as economic and political destabilization and global migration that can lead to adverse impacts on U.S. national security, public health, and humanitarian concerns. Those impacts are more fully captured within global measures of the social cost of greenhouse gases.

In addition, assessing the benefits of U.S. GHG mitigation activities requires consideration of how those actions may affect mitigation activities by other countries, as those international actions will provide a benefit to U.S. citizens and residents. A wide range of scientific and economic experts have emphasized the issue of reciprocity as support for considering global damages of GHG emissions. Using a global estimate of damages in U.S. analyses of regulatory actions allows the U.S. to continue to actively encourage other nations, including emerging major economies, to take significant steps to reduce emissions. The only way to achieve an efficient allocation of resources for emissions reduction on a global basis—and so benefit the U.S. and its citizens—is for all countries to base their policies on global estimates of damages.

The SC-GHG values reported in EPA's 2023 Report provide a global measure of monetized damages from GHG reductions. EPA's report explains that “The US economy is . . . inextricably linked to the rest of the world” and that “over 20% of American firms' profits are earned on activities outside of the country.” On this basis EPA concludes “Climate impacts that occur outside U.S. borders will impact the welfare of individuals and the profits of firms that reside in the US because of the connection to the global economy . . . through international markets, trade, tourism, and other activities.” Like the IWG, EPA also concluded that climate damages that originate in other nations can produce “economic and political destabilization, and global migration that can lead to adverse impacts on U.S. national security, public health, and humanitarian concerns.” NHTSA is aligned with EPA that climate damages to the rest of the world will result in damages that will be felt domestically, and thus concludes that SC-GHG values that incorporate both domestic and international damages are appropriate for its analyses.

While global estimates of the SC-GHG are the most appropriate values to use for the above stated reasons, new modeling efforts suggest that U.S.-specific damages are very likely higher than previously estimated. For instance, the EPA's Framework for Evaluating Damages and Impacts (FrEDI) is a “reduced complexity model that projects impacts of climate change within the United States through the 21st century” that offers insights on some omitted impacts that are not yet captured in global models. Results from FrEDI suggest that damages due to climate change within the contiguous United States are expected to be substantial. EPA's recent tailpipe emissions standards cite a FrEDI-produced partial SC-CO₂ estimate of $41 per metric ton. This U.S.-specific value is comparable to SC-CO₂ estimates NHTSA has used for prior rulemakings and used in sensitivity analyses for this rulemaking. NHTSA notes both that the FrEDI estimates do not include many climate impacts and thus are underestimates of harm, and that the FrEDI estimates include impact categories that are not available for the rest of the world. and thus, are missing from the global estimates used here. The damage models applied to generate EPA's estimates of the global SC-CO₂ estimates used in this final rule (the Data-driven Spatial Climate Impact Model (DSCIM) and the Greenhouse Gas Impact Value Estimator (GIVE)), which as noted do not reflect many important climate impacts, provide estimates of climate change impacts physically occurring within the United States of $16-$18 per metric ton for 2030 emissions. EPA notes that “[w]hile the FrEDI results help to illustrate how monetized damages physically occurring within the [continental US] increase as more impacts are reflected in the modeling framework, they are still subject to many of the same limitations associated with the DSCIM and GIVE damaIules, including the omission or partial modeling of important damage categories.” EPA also notes that the DSCIM and GIVE estimates of climate change impacts physically occurring within the United States are, like FrEDI, “not equivalent to an estimate of the benefits of marginal GHG mitigation accruing to U.S. citizens and residents” in part because they “exclude the myriad of pathways through which global climate impacts directly and indirectly affect the interests of U.S. citizens and residents.”

Taken together, applying the U.S.-specific partial SC-GHG estimates derived from the multiple lines of evidence described above to the GHG emissions reduction expected under the final rule would yield substantial benefits. For example, the present value of the climate benefits as measured by FrEDI (under a 2 percent near-term Ramsey discount rate) from climate change impacts in the contiguous United States for the preferred alternative for passenger cars and light trucks (CY perspective), for passenger cars and light trucks (MY perspective), and for HDPUVs, are estimated to be $19.6 billion, $4.7 billion, and $1.5 billion, respectively. However, the numerous explicitly omitted damage categories and other modeling limitations discussed above and throughout the EPA's 2023 Report make it likely that these estimates significantly underestimate the benefits to U.S. citizens and residents of the GHG reductions from the final rule; the limitations in developing a U.S.-specific estimate that accurately captures direct and spillover effects on U.S. citizens and residents further demonstrates that it is more appropriate to use a global measure of climate benefits from GHG reductions.

Finally, the last major category of comments pertained to the choice of discount rate applied to climate-related benefits and costs. Valero contends that the appropriate choice of discount rate in this case is an unsettled issue and that if global climate benefits are considered, a global discount rate above 8 percent should be used. Our Children's Trust commented that NHTSA should consider intergenerational equity and calculate climate benefits using negative, zero, or near-zero percent discount rates. Several commenters, including CBD and IPI, support the usage of the discount rates included in the EPA's SC-GHG update, mention that Executive Order 13990 instructs agencies to ensure that the social cost of greenhouse gas values adequately account for intergenerational equity, and argue that a capital-based discount rate is inappropriate for these multigenerational climate effects.

As previously noted, NHTSA presents and considers a range of discount rates for climate-related benefits and costs, including 2.5, 2.0, and 1.5 percent. Contrary to the position put forward by Children's Trust that it is unlawful to discount the estimated costs of SC-GHG, we also believe that discounting the stream of climate benefits from reduced emissions from the rule in order to develop a present value of the benefits of reducing GHG emissions is consistent with the law, and that the discounting approach used by the EPA is reasonable. Courts have previously reviewed and affirmed rules that discount climate-related costs. Courts have likewise advised agencies to approach cost-benefit analyses with impartiality, to ensure that important factors are captured in the analysis, including climate benefits, and to ensure that the decision rests “on a consideration of the relevant factors.” NHTSA has followed these principles here. In addition, NHTSA believes that discount rates at or above the opportunity cost of capital (7 percent) are inappropriate to use for GHG emissions that have intergenerational impacts. As discussed at length above, the consumption rate of interest is a more appropriate choice as it is the rate at which we observe consumers trading off consumption today for consumption in the future. Circular A-4 also identifies uncertainty in long-run interest rates as another reason why it is appropriate to use lower rates to discount intergenerational impacts, since recognizing such uncertainty causes the appropriate discount rate to decline gradually over progressively longer time horizons. In addition, the approach used incorporates rIrsion into its the modeling framework, which recognizes that individuals are likely willing to pay some additional amount to avoid the risk that the actual damages they experience might exceed their expected level. This gives some consideration to the insurance against low-probability but high-consequence climate damages that interventions to reduce GHG emissions offer. The impacts on future generations, uncertainty, and risk aversion are reflected in the estimates used in this analysis. The 2023 EPA SC-GHG Report's central SC-GHG values are based on a 2 percent discount rate, and for this reason NHTSA presents in its analysis of this Final Rule SC-GHG estimates discounted at 2 percent together with its primary estimates of other costs and benefits wherever NHTSA does not report the full range of SC-GHG estimates. For additional details regarding the choice of discount rates for climate related benefits, see Chapter 6.2.1.2 of the TSD.

(2) Reduced Health Damages

The CAFE Model estimates monetized health effects associated with emissions from directly emitted particulate matter 2.5 microns or less in diameter (PM_2.5) and two precursors to PM_2.5 (NO_X and SO₂). As discussed in Section III.F above, although other criteria pollutants are currently regulated, only impacts from these three pollutants are calculated since they are known to be emitted regularly from mobile sources, have the most adverse effects on human health, and have been the subject of extensive research by EPA to estimate the benefits of reducing these pollutants. The CAFE Model computes the monetized PM_2.5 -related health damages from each of the three pollutants by multiplying the monetized health impact per ton by the total tons of each pollutant emitted, including from both upstream and downstream sources. Reductions in these costs from their level under the reference baseline alternative that are projected to result from adopting alternative standards are treated as external benefits of those alternatives. Chapter 5 of the TSD accompanying this final rule includes a detailed description of the emission factors that inform the CAFE Model's calculation of the total tons of each pollutant associated with upstream and downstream emissions.

These monetized health benefit per ton values are closely related to the health incidence per ton values described above in Section III.F and in detail in Chapter 5.4 of the TSD. We use the same EPA sources that provided health incidence values to determine which monetized health impacts per ton values to use as inputs in the CAFE Model. Like the estimates associated with health incidences per ton of criteria pollutant emissions, we used an EPA TSD, multiple papers written by EPA staff and conversations with EPA staff to appropriately account for monetized damages for each pollutant associated with the source sectors included in the CAFE Model. The various emission source sectors included in the EPA papers do not always correspond exactly to the emission source categories used in the CAFE Model. In those cases, we mapped multiple EPA sectors to a single source category and computed a weighted average of the health impact per ton values.

The EPA uses the value of a statistical life (VSL) to estimate premature mortality impacts, and a combination of willingness to pay estimates and costs of treating the health impact for estimating the morbidity impacts. EPA's 2018 technical support document, “Estimating the Benefit per Ton of Reducing PM_2.5 Precursors from 17 Sectors,” (referred to here as the 2018 EPA source apportionment TSD) contains a more detailed account of how health incidences are monetized. It is important to note that the EPA sources cited frequently refer to these monetized health impacts per ton as “benefits per ton,” since they describe these estimates in terms of emissions avoided. In the CAFE Model input structure, these are generally referred to as monetized health impacts or damage costs associated with pollutants emitted (rather than avoided), unless the context states otherwise.

The CAFE Model health impacts inputs are based partially on the structure of the 2018 EPA source apportionment TSD, which reported benefits per ton values for the years 2020, 2025, and 2030. For the years in between the source years used in the input structure, the CAFE Model applies values from the closest source year. For example, the model applies 2020 monetized health impact per ton values for calendar years 2020-2022 and applies 2025 values for calendar years 2023-2027. In order for some of the monetized health damage values to match the structure of other impacts costs, DOT staff developed proxies for 7% discounted values for specific source sectors by using the ratio between a comparable sector's 3% and 7% discounted values. In addition, we used implicit price deflators from the Bureau of Economic Analysis (BEA) to convert different monetized estimates to 2021 dollars, in order to be consistent with the rest of the CAFE Model inputs.

This process is described in more detail in Chapter 6.2.2 of the TSD accompanying this final rule. In addition, the CAFE Model documentation contains more details of the model's computation of monetized health impacts. All resulting emission damage costs for PM_2.5 -related pollutants are located in the Criteria Emissions Cost worksheet of the Parameters file. The States and Cities commented that NHTSA should emphasize that although only NO_X, SO_X, and PM_2.5 reductions are monetized (in terms of their contribution to ambient PM_2.5 formation), total benefits of reduced pollution are larger although they do not appear in the benefit-cost-analysis. NHTSA agrees, and notes that although we do not have a basis for valuing other pollutants, we acknowledge that they form part of the unquantified benefits that likely arise from this rule.

One specific category of benefits that is not monetized in our analysis is the health harms of air toxics and ozone. ALA brought forward the absence of the health harms of air toxics in their comments on the NPRM, stating that the missing health harms of air toxics are a limit of the health impacts analysis. Historically, these pollutants have not typically been monetized, and as such we currently have no basis for that valuation. In the case of ozone, monetized BPT values that exist in the literature do not correspond to the source sectors we need for our analysis (namely NHTSA notes that these benefits are important although they have not been quantified.

(3) Reduction in Petroleum Market Externalities

The standards would decrease domestic consumption of gasoline, producing a corresponding decrease in the Nation's demand for crude petroleum, a commodity that is traded actively in a worldwide market. Because the U.S. accounts for a significant share of global oil consumption, the resulting decrease in global petroleum demand will exert some downward pressure on worldwide prices.

U.S. consumption and imports of petroleum products have three potential effects on the domestic economy that are often referred to collectively as “energy security externalities,” and increases in their magnitude are sometimes cited as possible social costs of increased U.S. demand for petroleum. Symmetrically, reducing U.S. petroleum consumption and imports can reduce these costs, and by doing so provide additional external benefits from establishing higher CAFE and fuel efficiency standards.

First, any increase in global petroleum prices that results from higher U.S. gasoline demand will cause a transfer of revenue to oil producers worldwide from consumers of petroleum, because consumers throughout the world are ultimately subject to the higher global price that results. Under competitive market assumptions, this transfer is simply a shift of resources that produces no change in global economic output or welfare. Since the financial drain it produces on the U.S. economy may not be considered by individual consumers of petroleum products, it is sometimes cited as an external cost of increased U.S. petroleum consumption.

As the U.S. has transitioned towards self-sufficiency in petroleum production (the nation became a net exporter of petroleum in 2020), this transfer is increasingly from U.S. consumers of refined petroleum products to U.S. petroleum producers, so it not only leaves welfare unaffected but even ceases to be a financial burden on the U.S. economy. In fact, to the extent that the U.S. becomes a larger net petroleum exporter, any transfer from global consumers to petroleum producers becomes a financial benefit to the U.S. economy. Nevertheless, uncertainty in the nation's long-term import-export balance makes it difficult to project precisely how these effects might change in response to increased consumption.

The loss of potential GDP from this externality will depend on the degree that global petroleum suppliers like the Organization of Petroleum Exporting Countries (OPEC) and Russia exercise market power which raise oil market prices above competitive market levels. In that situation, increases in U.S. gasoline demand will drive petroleum prices further above competitive levels, thus exacerbating this deadweight loss. More stringent standards lower gasoline demand and hence reduce these losses.

Over most of the period spanned by NHTSA's analysis, any decrease in domestic spending for petroleum caused by the effect of lower U.S. fuel consumption and petroleum demand on world oil prices is expected to remain entirely a transfer within the U.S. economy. In the case in which large producers are able to exercise market power to keep global prices for petroleum above competitive levels, this reduction in price should also increase potential GDP in the U.S. However, the degree to which OPEC and other producers like Russia are able to act as a cartel depends on a variety of economic and political factors and has varied widely over recent history, so there is significant uncertainty over how this will evolve over the horizon that NHTSA models. For these reasons, lower U.S. spending on petroleum products that results from raising standards, reducing U.S. gasoline demand, and the downward pressure it places on global petroleum prices is not included among the economic benefits accounted for in the agency's evaluation of this final rule.

Second, higher U.S. petroleum consumption can also increase domestic consumers' exposure to oil price shocks and thus increase potential costs to all U.S. petroleum users from possible interruptions in the global supply of petroleum or rapid increases in global oil prices. Because users of petroleum products are unlikely to consider the effect of their increased purchases on these risks, their economic value is often cited as an external cost of increased U.S. consumption. Decreased consumption, which we expect as a result of the standards, decreases this cost. We include an estimate of this impact of the standards, and an explanation of our methodology can be found in Chapter 6.2.4.4 of the TSD.

Finally, some analysts argue that domestic demand for imported petroleum may also influence U.S. military spending; because the increased cost of military activities would not be reflected in the price paid at the gas pump, this is often suggested as a third category of external costs from increased U.S. petroleum consumption. For example, NHTSA has received extensive comments to past rulemakings about exactly this effect on its past actions from the group Securing America's Energy Future. Most recent studies of military-related costs to protect U.S. oil imports conclude that significant savings in military spending are unlikely to result from incremental reductions in U.S. consumption of petroleum products on the scale that would result from adopting higher standards. While the cumulative effects of increasing fuel economy over the long-term likely have reduced the amount the U.S. has to spend to protect its interest in energy sources globally—avoid being beholden to geo-political forces that could disrupt oil supplies—it is extremely difficult to quantify the impacts and even further to identify how much a single fuel economy rule contributes. As such NHTSA does not estimate the impact of the standards on military spending. See Chapter 6.2.4.5 of the TSD for additional details.

Each of these three factors would be expected to decrease incrementally as a consequence of a decrease in U.S. petroleum consumption resulting from the standards. Chapter 6.2.4 of the TSD provides a comprehensive explanation of NHTSA's analysis of these three impacts.

NHTSA sought comment on its accounting of energy security in the proposal. The Institute for Energy Research and AFPM both noted that the United States is now a net-exporter of crude oil, and that a significant share of imported crude oil is sourced from other North American countries. The American Enterprise Institute suggested that the macroeconomic risks associated with oil supply shocks like those described by NHTSA in its proposal are reflected in the price of oil since it is a globally traded commodity. As a result, they argue that since all countries face common international prices for these products (outside of transportation costs and other second order differences), the energy security of countries does not depend on its overall level of imports. Several commenters also argued that increasing reliance on domestically produced ethanol rather than battery electric vehicles represents a superior method for improving energy security.

NHTSA noted in its proposal the importance of the United States' role as a net exporter in its quantification of energy security related benefits. For example, NHTSA discussed the so-called “monopsony effect” or the effect of reduced consumption on global oil prices. NHTSA noted that this represents a transfer between oil producers and consumers, rather than a real change in domestic welfare, and since the United States is no longer a net importer the monopsony effect on global prices no longer represents a transfer from producers in other countries. However, NHTSA disagrees with the suggestion that this status eliminates the energy security externalities that NHTSA quantified in its analysis. As described in TSD Chapter 6, NHTSA considered the effect of reductions in domestic consumption on the expected value of U.S. macroeconomic losses due to foreign oil supply shocks in future years. The expected magnitude of the effect of these shocks on overall domestic economic activity is determined by the probability of these shocks, the overall exposure of the global oil supply to these shocks, (which depends upon the size of U.S. gross oil imports), the short run elasticities of supply and demand for oil, and the sensitivity of the U.S. economy to changes in oil prices.

NHTSA analyzed these drivers of energy security costs in its proposal and concluded that there were still strong reasons to believe that changes in fuel economy standards could produce economic benefits by reducing them. As can be seen through the events NHTSA listed in its discussion of energy security in Chapter 6 of the TSD, foreign oil shocks like the one caused by Russia's invasion of Ukraine remain a risk that can at least in the short-term influence global oil supply and prices, which adversely affect consumers and disrupt economic growth, although no recent example of oil supply shocks has reached the magnitude of the OPEC oil embargo or Iranian Revolution during the 1970s. NHTSA will continue to monitor the literature for updated estimates of the probability and size foreign oil shocks and update its estimates accordingly. As noted in the TSD, the U.S. has in recent years become a net exporter of oil. However, the U.S. still only accounts for about 14.7 percent of global oil production, and the U.S., Canada, and Mexico together account for less than a quarter of global oil production according to the U.S. EIA. By contrast, seven countries in the Persian Gulf region account for about one-third of production and held about half of the world's proven reserves. Russia alone accounted for 12.7 percent of production in 2022, and the global supply shock caused by Russia's invasion of Ukraine was followed by a surge of more than 20 percent in crude oil prices. Clearly substantial shares of the global oil supply remain in regions that have proven vulnerable to the exact supply shocks described by NHTSA in its rulemaking documents. Furthermore, the U.S., while on balance a net-exporter, continues to import substantial quantities of oil from countries at risk of shocks. In 2022, Iraq, Saudi Arabia, and Colombia accounted for 14 percent of oil imports in the U.S., or about 1.1 million barrels per day. On net, the U.S. still imports just under 3 million barrels of crude oil per day. Due to refinery configurations, many refiners in the U.S., especially in the Midwest and Gulf Coast still most profitably refine heavy, sour crude oil from abroad. Indeed, in its 2023 AEO the EIA still projects that the U.S. will import 6.65 million barrels per day of oil in 2050. Moreover, U.S. consumers are also exposed to foreign oil shocks through other imported goods that use petroleum as an input. Thus, NHTSA still believes that it is correct to assume that changes in domestic consumption are likely to affect demand for foreign oil.

NHTSA also disagrees with the conclusion that these energy security risks are efficiently priced by global markets. Traded oil prices represent equilibrium outcomes determined by global supply and demand for oil. Global demand is determined by the aggregation of global consumers' willingness to pay for oil and the products it produces. This willingness to pay depends on the private benefits derived from oil products. The macroeconomic disruption costs described by NHTSA are borne across the economy, meaning that they are unlikely to be considered by individual consumers in their decision-making calculus. For this reason, economists have classified them as externalities, and thus a potential source of socially inefficient outcomes. The magnitude of these macroeconomic disruptions from oil supply shocks depends directly on the overall oil intensity of the economy. A more fuel-efficient fleet of vehicles is expected to lower the economy's oil intensity. Furthermore, EPCA, the statute that confers the agency with the authority to set standards, was enacted with the stated purpose to increase energy independence and security, and set out to accomplish these goals through increasing the efficiency of energy consuming goods such as automobiles. Congress explicitly directed the agency to consider the need of the United States to conserve energy when setting maximum feasible standards. The suggestion that NHTSA should forgo the potential impacts to energy security of setting standards cuts against the very fabric of public policy underlying EPCA.

NHTSA is also monitoring the availability of critical minerals used in electrified powertrains and whether any shortage of such materials could emerge as an additional energy security concern. While nearly all electricity in the United States is generated through the conversion of domestic energy sources and thus its supply does not raise security concerns, EVs also require batteries to store and deliver that electricity. Currently, the most commonly used electric vehicle battery chemistries include relatively scarce materials (compared to other automotive parts) which are sourced, in part, from potentially insecure or unstable overseas sites and like all mined materials (including those in internal combustion engine vehicles) can pose environmental challenges during extraction and conversion to usable material. Known supplies of some of these critical minerals are also highly concentrated in a few countries and therefore face similar market power concerns to petroleum products.

NHTSA is restricted from considering the fuel economy of alternative fuel sources in determining CAFE standards, and as such, the CAFE Model restricts the application of BEV pathways and PHEV electric efficiency in simulating compliance with fuel economy regulatory alternatives. While the cost of critical minerals may affect the cost to supply both plug-in and non-plug-in hybrids that require larger batteries, this would apply primarily to manufacturers whose voluntary compliance strategy includes electrification given the greater mineral requirements of battery electric vehicles and plug-in hybrid-electric vehicles compared with non-plug-in hybrids. NHTSA did not include costs or benefits related to these emerging energy security considerations in its analysis for its proposal and sought comment on whether it is appropriate to include an estimate in the analysis and, if so, which data sources and methodologies it should employ.

NHTSA received a number of comments suggesting that it should include costs and benefits related to these emerging energy security considerations. Several commenters noted that politically unstable countries or countries with which the U.S. does not have friendly trade relations, including China, mine or process a significant share of the minerals used in battery production, including lithium, cobalt, graphite and nickel. AFPM also argued that the penetration rate of BEVs in NHTSA's No-Action alternative would require supply chain improvements that they contend are highly uncertain to occur, or that the battery chemistry technologies necessary to alleviate these concerns were not likely to be available in the timeframe suggested by NHTSA's analysis. Some of these commenters suggested that mineral security should be included in NHTSA's analysis as a cost associated with adoption of technologies that require these minerals, and that the failure to include this as a cost was arbitrary and capricious. ZETA on the other hand suggested that the demands for critical minerals could be met through reserves in friendly countries, and noted the steps taken by both the public and private sector to expand domestic critical mineral production. The National Association of Manufacturers and the U.S. Chamber of Commerce both suggested that expanding domestic supply of critical minerals required the Administration and Congress to expedite permitting.

NHTSA agrees with commenters that the increase in battery demand likely will require significant expansion of production of certain critical minerals, although critical minerals have long been a component of vehicles and many other goods consumed in the United States. NHTSA also notes the concerted efforts across the federal government to shift supply chains to ensure that a larger share of critical mineral production comes from politically stable sources. Between the publication of NHTSA's proposal and the final rule, ANL produced a study of the prospective supply of upstream critical materials used to meet the U.S.'s EV and Energy Storage System deployment targets for 2035. According to ANL, the U.S. is positioned to meet lithium demand through a combination of domestic production as well as imports from FTA countries. The U.S. will need to source graphite, nickel, and cobalt from partner countries (including those with and without FTAs) in the near and medium term. Thus, NHTSA believes that there is strong evidence that the U.S. has significant opportunities to diversify supply chains away from current suppliers like China.

Further, NHTSA notes that considering mineral security in its analysis of incremental societal costs and benefits would be unlikely to materially impact the ranking of its regulatory alternatives. EPCA constrains NHTSA from considering BEV adoption as a compliance strategy during standard setting years in its light duty analysis. As a result, there will be minimal incremental demand for batteries and critical minerals in regulatory alternatives, and thus minimal incremental societal costs related to mineral security. While BEV adoption—including compliance with ZEV regulatory programs—is considered in the No-Action Alternative, mineral security costs associated with the adoption of BEVs in these cases are (1) not incremental costs associated with changes in CAFE standards, and (2) not considered by consumers and manufacturers outside of how they impact technology costs and vehicle prices, both of which are considered in NHTSA's analysis. In the HDPUV fleet, a similar pattern emerges even in the absence of similar constraints; the overwhelming majority of electrification takes place in the reference baseline. Further, given the relatively small volume of HDPUVs, the incremental demand for any critical minerals is minimal compared to the total global supply.

Finally, NHTSA notes that while commenters suggested that NHTSA include mineral security in its analysis, they did not recommend a specific methodology for how to do so. During its analysis NHTSA surveyed the economics literature and did not find a comparable existing set of methods for analyzing mineral security as it did for petroleum market externalities. This is largely due to the relatively recent emergence of this topic. Several of the inputs used in NHTSA's energy security analysis (distributions of estimates of its elasticity parameters, supply shock probability distributions, long term projections of supply and demand for petroleum) rely on decades of research which do not exist for the emerging topic of mineral security. NHTSA is continuing to monitor research in this field and is considering implementing estimates of these costs in future rulemakings but did not include them in this final rule.

(4) Changes in Labor Use and Employment

As vehicle prices rise, we expect consumers to purchase fewer vehicles than they would have at lower prices. If manufacturers produce fewer vehicles as a consequence of lower demand, they may need less labor to produce and assemble vehicles, while dealers may need less labor to sell the vehicles. Conversely, as manufacturers add equipment to each new vehicle, the industry will require labor resources to develop, sell, and produce additional fuel-saving technologies. We also account for the possibility that new standards could shift the relative shares of passenger cars and light trucks in the overall fleet. Since the production of different vehicles involves different amounts of labor, this shift affects the required quantity of labor.

The analysis considers the direct labor effects that the standards have across the automotive sector. The effects include (1) dealership labor related to new light-duty and HDPUV unit sales; (2) assembly labor for vehicles, engines, and transmissions related to new vehicle unit sales; and (3) labor related to mandated additional fuel savings technologies, accounting for new vehicle unit sales. NHTSA has now used this methodology across several rulemakings but has generally not emphasized its results, largely because NHTSA found that attempting to quantify the overall labor or economic effects was too uncertain and difficult. We have also excluded any analysis of how changes in direct labor requirements could change employment in adjacent industries.

NHTSA still believes that such an expanded analysis may be outside the effects that are reasonably traceable to the final rule; however, NHTSA has identified an exogenous model that can capture both the labor impacts contained in the CAFE Model and the secondary macroeconomic impacts due to changes in sales, vehicle prices, and fuel savings. Accompanying this final rule is a docket memo explaining how the CAFE Model's outputs may be used within Regional Economic Models, Inc. (REMI)'s PI + employment model to quantify the impacts of this final rule. We received comment from the Joint NGOs regarding the proposal for additional analysis in the docket memo stating that NHTSA should not include this additional analysis since the public was not given the opportunity to comment on results. Although we were unable to fully implement the side analysis with finalized results for this rule, we are continuing to explore the possibility of including these impacts in future analyses.

The United Auto Workers (UAW) commented that NHTSA should perform additional analysis of the impacts of the standards on employment, with a particular focus on union jobs and new EV jobs. Although we do not currently look at labor impacts by specific technologies, we may consider including it in future analyses. All labor effects are estimated and reported at a national aggregate level, in person-years, assuming 2,000 hours of labor per person-year. These labor hours are not converted to monetized values because we assume that the labor costs are included into a new vehicle's purchasing price. The analysis estimates labor effects from the forecasted CAFE Model technology costs and from review of automotive labor for the MY 2022 fleet. NHTSA uses information about the locations of vehicle assembly, engine assembly, and transmission assembly, and the percent of U.S. content of vehicles collected from American Automotive Labeling Act (AALA) submissions for each vehicle in the reference fleet. The analysis assumes that the fractions of parts that are currently made in the U.S. will remain constant for each vehicle as manufacturers add fuel-savings technologies. This should not be construed as a prediction that the percentage of U.S.-made parts—and by extension U.S. labor— will remain constant, but rather as an acknowledgement that NHTSA does not have a clear basis to project where future production may shift. The analysis also uses data from the NADA annual report to derive dealership labor estimates.

While the IRA tax credit eligibility is not dependent on our labor assumptions here, if NHTSA were able to dynamically model changes in parts content with enough confidence in its precision, NHTSA could potentially employ those results to dynamically model a portion of tax credit eligibility.

Some commenters argued that culmination of the standards and the further adoption of BEVs would significantly impair the automotive industry through dramatically reduced sales, leading to a substantial number of layoffs, and accused the agency of improperly ignoring this unintended consequence. The agency disagrees. First, the agency notes that the premise in these comments is unsupported. As noted in sales, we believe that sales are largely determined by exogenous market factors, and our standards will have a marginal impact. Second, electrification is not a compliance pathway for CAFE, so any impacts would be contained to the reference baseline fleet through standard setting years. Finally, commenters did not provide any evidence that BEV adoption would harm domestic jobs and sales and relied solely on speculation.

In sum, the analysis shows that the increased labor from producing additional technology necessary to meet the preferred alternative will outweigh any decreases attributable to the change in new vehicle sales. For a full description of the process NHTSA uses to estimate labor impacts, see Chapter 6.2.5 of the TSD.

3. Costs and Benefits Not Quantified

In addition to the costs and benefits described above, Table III-7 includes two-line items without values. The first is maintenance and repair costs. Many of the technologies manufacturers apply to vehicles to meet the standards are sophisticated and costly. The technology costs capture only the initial or “upfront” costs to incorporate this equipment into new vehicles; however, if the equipment is costlier to maintain or repair—as seems likely for at least more conventional technology because the materials used to produce the equipment are more expensive and the equipment itself is significantly more complex and requires more time and labor to maintain or repair—, then consumers will also experience increased costs throughout the lifetime of the vehicle to keep it operational. Conversely, electrification technologies offer the potential to lower repair and maintenance costs. For example, BEVs do not have engines that are costly to maintain, and all electric pathways with regenerative braking may reduce the strain on braking equipment and consequentially extend the useful life of braking equipment. We received several comments concerned with electric vehicle battery replacement costs and maintenance/repair cost differences between EVs and ICEs. The Heritage Foundation and the American Consumer Institute noted that EV battery replacement costs are expensive, and AFPM commented that these battery replacement costs will impact lower-income households.

The West Virginia Attorney General's Office commented that NHTSA should include a life-cycle analysis, emphasizing that EVs' complicated powertrains could lead to higher maintenance and repair costs. We do not currently include a life-cycle analysis as part of the CAFE Model but may consider incorporating some aspects of this into future rules. For a literature review and additional qualitative discussion on the vehicle cycle and its impacts, readers should refer to FEIS Chapter 6 (Lifecycle Analysis) (See III.F as well). Other commenters have been just as adamant that BEVs offer lifetime maintenance and repair benefits.

NHTSA notes that due to statutory constraints on considering the fuel economy of BEVs and the full fuel economy of PHEVs in determining maximum feasible CAFE standards, any change in maintenance and repair costs due to electrification would have a limited impact on NHTSA's analysis comparing alternatives. Given that this topic is still emerging, and that the results would not affect the agency's decision given the statutory constraint on consideration of BEV fuel economy in determining maximum feasible CAFE standards, the agency believes it is reasonable not to attempt to model these benefits or costs in this final rule. See Section VI.A on economic practicability for discussion on affordability impacts more generally.

Consumer Reports commented that hybrid-cost effectiveness is, on average, better than that of non-hybrids due to maintenance and repair cost savings over time, citing their 2023 analysis focusing on ten bestselling hybrids and their ICE counterparts. NHTSA is continuing to study the relative maintenance and repair costs associated with adopting fuel saving technologies. In order to conduct this analysis properly NHTSA would require more granular data on a larger set of technologies than what is included in Consumer Reports' study and would also need to estimate the effects of changes in vehicle usage on these costs. NHTSA will continue to consider these costs in the future as more information becomes available.

The second empty line item in the table is the value of potential sacrifices in other vehicle attributes. Some technologies that are used to improve fuel economy could have also been used to increase other vehicle attributes, especially performance, carrying capacity, comfort, and energy-using accessories, though some technologies can also increase both fuel economy and performance simultaneously. While this is most obvious for technologies that improve the efficiency of engines and transmissions, it may also be true of technologies that reduce mass, aerodynamic drag, rolling resistance or any road or accessory load. The exact nature of the potential to trade-off attributes for fuel economy varies with specific technologies, but at a minimum, increasing vehicle efficiency or reducing loads allows a more powerful engine to be used while achieving the same level of fuel economy. Performance is held constant in our analysis. However, if a consumer values a performance attribute that cannot be added to a vehicle because fuel economy improvements have “used up” the relevant technologies, or if vehicle prices become too high wherein either a consumer cannot obtain additional financing or afford to pay more for a vehicle within their household budget that consumers may opt to purchase vehicles that are smaller or lack features such as heated seats, advanced entertainment or convenience systems, advance safety systems, or panoramic sunroofs, that the consumer values but are unrelated to the performance of the drivetrain. Alternatively, manufacturers may voluntarily preclude these features from certain models or limit the development of other new features in anticipation that new vehicle price affordability will limit the amount they may be able to charge for these new features. How consumers value increased fuel economy and how fuel economy regulations affect manufacturers' decisions about using efficiency-improving technologies can have important effects on the estimated costs, benefits, and indirect impacts of fuel economy standards. Nevertheless, any sacrifice in potential improvements to vehicles' other attributes could represent a net opportunity cost to their buyers (though performance-efficiency tradeoffs could also lower compliance costs, and some additional attributes, like acceleration, could come with their own countervailing social costs).

NHTSA has previously attempted to model the potential sacrifice in other vehicle attributes in sensitivity analyses by assuming the opportunity cost must be greater than some percentage of the fuel savings they seemingly voluntarily forego. In those previous rulemakings, NHTSA acknowledged that it is extremely difficult to quantify the potential loss of other vehicle attributes, and therefore included the value of other vehicle attributes only in sensitivity analyses. This approach is used as a sensitivity analysis for the final rule and is discussed in RIA 9.2.3. This approach is only relevant if the foregone fuel savings cannot be explained by the energy paradox.

The results of NHTSA's analysis of the HDPUV standards suggest that buyer's perceived reluctance to purchasing higher-mpg models is due to undervaluation of the expected fuel savings due to market failures, including short-termism, principal-agent split incentives, uncertainty about the performance and service needs of new technologies and first-mover disadvantages for consumers, uncertainty about the resale market, and market power and first-mover disadvantages among manufacturers. This result is the same for vehicles purchased by individual consumers and those bought for commercial purposes. NHTSA tested the sensitivity of the analysis to the potential that the market failures listed do not apply to the commercial side of the HDPUV market. In this sensitivity analysis, commercial operators are modeled as profit maximizers who would not be made more or less profitable by more stringent standards by offsetting the estimated net private benefit to commercial operators. NHTSA decided against including this alternative in the primary analysis to align with its approach to market failures in the light-duty analysis. Furthermore, there is insufficient data on the size and composition of the commercial share of the HDPUV market to develop a precise estimate of a commercial operator opportunity cost. For additional details, see Chapter 9.2.3.10 of the FRIA.

Several commenters argued that NHTSA's assumption that increases in fuel economy to meet the new standards are not accompanied by foregone vehicle performance leads to an overestimate of net-benefits from increasing standards. For example Valero commented that “NHTSA offer[ed] no convincing rationale for omitting foregone performance gains from the central-case analysis” and claimed “NHTSA does its best to completely avoid the performance issue.” IPI shared a similar belief and commented that “NHTSA should further highlight [the implicit opportunity cost] sensitivity results.” NHTSA agrees with IPI that it could do a better job highlighting the results of sensitivities that stakeholders considered, especially ones like the implicit opportunity cost which some commenters felt were either missing or underrepresented.

More specifically, Landmark argued that improvements in fuel economy necessitate performance tradeoffs to reduce the weight of vehicles. Other commenters argued that there is evidence that in the absence of changes to standards manufacturers have chosen to make further improvements to performance features of vehicles, and that similar future improvements to performance would be sacrificed by manufacturers in order to comply with the standards NHTSA proposed, and thus should be counted as incremental consumer costs.

Valero, CEA, and NADA referenced a recent paper from Leard, Linn, and Zhou (2023), who estimate that this opportunity cost of fuel economy improvement could offset much of the private fuel cost savings benefits that consumers receive from the increase in stringency of standards. The authors of this paper estimate that consumers value improvements in acceleration much more highly than the fuel economy improvements that manufacturers trade them off for in an effort to comply with higher standards. However, the authors of this paper note that their study does not account for the potential induced innovations from tightened standards, or market failures associated with imperfect competition in the new vehicle market. NHTSA discussed this paper in its proposal, but recognized the limitations that the authors noted, as well as the degree of uncertainty in the literature regarding the implicit opportunity cost of fuel economy standards.

Valero suggests that in the absence of higher standards, manufacturers would channel investment into improvements in vehicle performance, which is foregone when standards are raised. As a result, Valero commented that fuel economy standards cause performance to increase less than it would in the absence of standards and referenced the findings of Klier and Linn (2016). NHTSA also discussed this paper in its proposal (see PRIA Chapter 9). The authors of the paper note that during the period they examined, for passenger cars in the United States there was no statistically significant evidence that stringency affected the direction of technology adoption between fuel efficiency and either horsepower or weight (the two attributes considered). While the authors do find evidence of an effect on this tradeoff for light trucks, they admit that there is significant uncertainty over the consumer's willingness to pay for this foregone performance (indeed they do not quantify the dollar value of the effect on vehicle weight due to this uncertainty). Recent data also casts doubt on Valero's deterministic understanding of the relationship between tightening standards and vehicle performance. Between 2000 and 2010 CAFE standards for passenger cars were unchanged. According to the 2023 EPA Automotive Trends report, real world fuel economy for vehicles rose at a rate of about 1.3 percent per year during this period, while horsepower rose at a rate of 1.2 percent, weight increased at a rate of 0.4 percent, and acceleration as measured by 0 to 60 miles per hour time declined at an average rate of 0.8 percent. Between 2010 and 2023, standards increased substantially and the fuel economy of these vehicles has improved at a rate of around 2.4 percent per year over this period. However, this has not caused improvements in other attributes to slow down. Instead, weight (0.5 percent), horsepower (1.7 percent), and 0 to 60 time (−1.4 percent) all improved at faster rates than the previous period. While these attributes could have potentially improved at still greater rates in the absence of standards, these headline values suggest that standards have at least not caused a significant slow-down relative to prior trends. Also, as noted in FRIA Chapter 9, other research suggests that consumers have not had to tradeoff performance for fuel economy improvements, and should not be expected to in the future, due to fuel saving technologies whose adoption does not lead to adverse effects on the performance of vehicles (Huang, Helfand, et al. 2018; Watten, Helfand and Anderson 2021; Helfand and Dorsey-Palmateer 2015). Indeed, there are technologies that exist that provide improved fuel economy without hindering performance, and in some cases, also improve performance (such as high-strength aluminum alloy bodies, turbocharging, and increasing the number of gear ratios in new transmissions). Even as the availability of more fuel-efficient vehicles has increased steadily over time, research has shown that the attitudes of drivers towards those vehicles with improved fuel economy has not been affected negatively. To the extent some performance-efficiency tradeoffs may have occurred in the past, such tradeoffs may decline over time, with technological advancements and manufacturer learning over longer vehicle design periods (Bento 2018; Helfand & Wolverton 2011).

NHTSA thus maintains that there is significant uncertainty in the literature over the degree to which changes in fuel economy standards will cause manufacturers to lower the performance of vehicles, and how much this will be valued by consumers. Indeed, the possibility that there are ancillary benefits to adopting fuel saving technology means that the directionality of the effect of excluding these additional attributes from the central analysis is unknown. In its analysis, NHTSA assumes that the performance features listed by commenters remain fixed across alternatives, and that manufacturers instead adopt fuel economy improving technology in order to comply with standards without reducing the quality of those features. NHTSA assumes that manufacturers are aware of consumers' willingness to pay for performance features like those noted by the commenters and would be reluctant to make sacrifices to them as part of their compliance strategies. This, of course, is not the only path to compliance for manufacturers. However, given uncertainty over consumer willingness to pay for the full set of potentially affected attributes, the long-term pricing strategies of firms, and firm specific costs, it is a reasonable approach for NHTSA to use when modeling the behavior of all manufacturers in the market. Modeling the decisions of all manufacturers over the complete set of attributes and technologies available would lead to a computationally infeasible model of compliance. Moreover, without highly detailed data about the manufacturing process of each manufacturer and vehicle model, it could introduce significant opportunities for errors in the agency's measurements of compliance costs. Omitting ancillary benefits and only including the attributes that could be traded off for fuel savings improvements by firms could bias the agency's analysis. Absent a better understanding of consumer willingness to pay for these other attributes, including them would create a misleading model of how firms would choose to comply with the standards as well as how consumer welfare would be affected. While commenters suggested that the performance neutrality assumption in NHTSA's analysis is unrealistic, they did not propose an alternative methodology for modeling how manufacturers would adjust performance attributes in response to changes in CAFE Standards. This performance neutrality assumption is intended to isolate the impacts of the standards and is necessary with or without a separate estimation of a potential implicit opportunity cost. Since NHTSA believes that its assumption of performance neutrality is a reasonable approach to modeling compliance, and since alternative approaches would introduce highly uncertain effects (with unknown directionality) and are currently infeasible, NHTSA has chosen to maintain its assumption of performance neutrality.

NHTSA does take seriously the possibility of opportunity costs as described by these commenters. For this reason, the agency included sensitivity cases in its analysis for both light duty and HDPUV in Chapter 9 of the PRIA and FRIA. In this sensitivity case, the opportunity cost of fuel economy for light duty vehicles is assumed to be equal to the discounted fuel cost savings for a vehicle over its first 72 months of use (roughly how long they are held, on average, by their first owner), less the undiscounted fuel cost savings over the first 30 months of use. NHTSA believes that this is a reasonable approach, since this value is equivalent to the value of fuel savings that new vehicle owners are assumed to not value in their purchase decision. If consumers are not myopic and value fuel savings fully, and assuming perfect information and no market distortions, then offsetting losses in performance would be at least this high. For HDPUVs, NHTSA also considered two additional sensitivity cases in which it assumed that this opportunity cost fully offset any net private benefits of fuel economy improvements for commercial buyers. This higher value for opportunity cost for commercial buyers was based on the assumption that commercial buyers are more likely to fully value the lifetime fuel savings of their fleet vehicles, since these buyers are profit maximizing businesses. As noted by IPI in its comments, NHTSA found in the proposal that while net social benefits under the preferred alternative are lower under these alternative assumptions, under 3 percent discounting they remain positive in all cases. This is caused by reductions in emissions externalities offsetting increases in safety externalities. NHTSA conducted similar sensitivity exercises in its final rule and found that societal net benefits remained positive in the preferred alternative regardless of discount rate. Since neither of these cases include the potential ancillary benefits of fuel saving technology adoption, and do not take into account the full set of compliance methods that manufacturers could employ to meet the standards in a cost effective way, NHTSA views these cases as bounding exercises that allow the agency to see whether a relatively high estimate of the potential opportunity costs of the standards outweigh the other net societal benefits included in NHTSA's analysis. Valero suggested that the agency's analysis of the implicit opportunity cost should equal to all private fuel savings. We disagree for several reasons. First, the average consumer will not hold onto new vehicles for a vehicle's entire lifetime, and even if the first owner valued all of the forgone attributes at the price of fuel savings, the second or third owner would have her own set of preferences that likely do not overlap the first owner's perfectly. Second, assigning a specific dollar value on vehicle luxuries is likely difficult for consumers, and there is a tendency for vehicle buyers to splurge at the dealership only to regret overspending when the monthly payments become due. For example, a Lending Tree survey found that 14 percent of car buyers wish ex post that they had chosen a different make or model, 10 percent bought too expensive of a car, 4 percent bought a more expensive car than they planned, and 3 percent noted they regretted buying features they did not need. Similarly, not all vehicle attributes are offered à la carte (some vehicle attributes are sometimes only available in packages with other additions or require consumers to purchase higher trims) and consumers may only value one or two items in a larger package and are stuck buying as a bundle.

H. Simulating Safety Effects of Regulatory Alternatives

The primary objective of the standards is to achieve maximum feasible fuel economy and fuel efficiency, thereby reducing fuel consumption. In setting standards to achieve this intended effect, the potential of the standards to affect vehicle safety is also considered. As a safety agency, NHTSA has long considered the potential for adverse or positive safety consequences when establishing fuel economy and fuel efficiency standards.

This safety analysis includes the comprehensive measure of safety impacts of the light-duty and HDPUV standards from three sources:

• Changes in Vehicle Mass

Similar to previous analyses, NHTSA calculates the safety impact of changes in vehicle mass made to reduce fuel consumption to comply with the standards. Statistical analysis of historical crash data indicates reducing mass in heavier vehicles generally improves safety for occupants in lighter vehicles and other road users like pedestrians and cyclists, while reducing mass in lighter vehicles generally reduces safety. NHTSA's crash simulation modeling of vehicle design concepts for reducing mass revealed similar effects. These observations align with the role of mass disparity in crashes; when vehicles of different masses collide, the smaller vehicle will experience a larger change in velocity (and, by extension, force), which increases the risk to its occupants. NHTSA believes the most recent analysis represents the best estimate of the impacts of mass reduction (MR) on crash fatalities attributable to changes in mass disparities., One caveat to note is that the best estimates are not significantly different from zero and are not statistically significant at the 95th confidence level. In other words, the effects of changes in mass due to this rule cannot be distinguished from zero.

Two individuals, Mario Loyola and Steven G. Bradbury, submitted a joint comment (referred to herein as “Loyola and Bradbury”), speculating that the agency is “downplay[ing] and minimize[ing] the loss of lives and serious injuries [the] standards [caused] by attributing many of these deaths and injuries to other regulators.” The commentors would have the agency include fatalities that are projected to occur in the reference baseline as attributable t' this rule. While NHTSA's analysis includes the impacts of other regulations in the reference baseline, it does not separate the safety impacts attributable to individual regulations. Instead, the analysis considers the aggregate impact of these other regulations for comparison with the impacts of CAFE standards. NHTSA does not have information, nor do the commenters provide any specific information, indicating that the inclusion of the impacts of these other regulations results in undercounting of safety impacts attributable to the Preferred Alternative. The purpose of calculating a reference baseline is to show the world in the absence of further government action. If NHTSA chose not to finalize the standards, the agency believes that the reference baseline fatalities would still occur. As such, we disagree with the authors' proposed suggestion.

• Impacts of Vehicle Prices on Fleet Turnover

Vehicles have become safer over time through a combination of new safety regulations and voluntary safety improvements. NHTSA expects this trend to continue as emerging technologies, such as advanced driver assistance systems, are incorporated into new vehicles. Safety improvements will likely continue regardless of changes in the standards.

As discussed in Section III.E.2, technologies added to comply with fuel economy and efficiency standards have an impact on vehicle prices, therefore slowing the acquisition of newer vehicles and retirement of older ones. The delay in fleet turnover caused by the effect of new vehicle prices affect safety by slowing the penetration of new safety technologies into the fleet.

The standards also influence the composition of the light-duty fleet. As the safety provided by light trucks, SUVs and passenger cars responds differently to technology that manufacturers employ to meet the standards—particularly mass reduction—fleets with different compositions of body styles will have varying numbers of fatalities, so changing the share of each type of light-duty vehicles in the projected future fleet impacts safety outcomes.

• Increased Driving Because of Better Fuel Economy

The “rebound effect” predicts consumers will drive more when the cost of driving declines. More stringent standards reduce vehicle operating costs, and in response, some consumers may choose to drive more. Additional driving increases exposure to risks associated with motor vehicle travel, and this added exposure translates into higher fatalities and injuries. However, most fatalities associated with rebound driving are the result of consumers choosing to drive more. Therefore, most of the societal safety costs of rebound vehicle travel are offset in our net benefits analysis.

The contributions of the three factors described above generate the differences in safety outcomes among regulatory alternatives. NHTSA's analysis makes extensive efforts to allocate the differences in safety outcomes between the three factors. Fatalities expected during future years under each alternative are projected by deriving a fleet-wide fatality rate (fatalities per vehicle mile of travel) that incorporates the effects of differences in each of the three factors from reference baseline conditions and multiplying it by that alternative's expected VMT. Fatalities are converted into a societal cost by multiplying fatalities with the DOT-recommended value of a statistical life (VSL) supplemented by economic impacts that are external to VSL measurements. Traffic injuries and property damage are also modeled directly using the same process and valued using costs that are specific to each injury severity level.

All three factors influence predicted fatalities, but only two of them—changes in vehicle mass and in the composition of the light-duty fleet in response to changes in vehicle prices—impose increased risks on drivers and passengers that are not compensated for by accompanying benefits. In contrast, increased driving associated with the rebound effect is a consumer choice that reveals the benefits of additional travel. Consumers who choose to drive more have apparently concluded that the utility of additional driving exceeds the additional costs for doing so, including the crash risk that they perceive additional driving involves. As discussed in Chapter 7 of the final TSD, the benefits of rebound driving are accounted for by offsetting a portion of the added safety costs.

For the safety component of the analysis for this final rule, NHTSA assumed that HDPUVs have the same risk exposure as light trucks. Given that the HDPUV fleet is significantly smaller than the light-duty fleet, the sample size to derive safety coefficients separately for HDPUVs is challenging. We believe that HDPUVs share many physical commonalities with light trucks and the incidence and crash severity are likely to be similar. As such, we concluded it was appropriate to use the light truck safety coefficients for HDPUVs.

NHTSA is continuing to use the proposal's approach of including non-occupants in the analysis. The agency categorizes safety outcome through three measures of light-duty and HDPUV vehicle safety: fatalities occurring in crashes, serious injuries, and the amount of property damage incurred in crashes with no injuries. Counts of fatalities to occupants of automobiles and non-occupants are obtained from NHTSA's Fatal Accident Reporting System. Estimates of the number of serious injuries to drivers and passengers of light-duty and HDPUV vehicles are tabulated from NHTSA's General Estimates System (GES) for 1990-2015, and from its Crash Report Sampling System (CRSS) for 2016-2019. Both GES and CRSS include annual samples of motor vehicle crashes occurring throughout the United States. Weights for different types of crashes were used to expand the samples of each type to estimates of the total number of crashes occurring during each year. Finally, estimates of the number of automobiles involved in property damage-only crashes each year were also developed using GES.

NHTSA sought comment on its safety assumptions and methodology in the proposal.

1. Mass Reduction Impacts

Vehicle mass reduction can be one of the more cost-effective means of improving efficiency, particularly for makes and models not already built with much high-strength steel or aluminum closures or low-mass components. Manufacturers have stated that they will continue to reduce mass of some of their models to meet more stringent standards, and therefore, this expectation is incorporated into the modeling analysis supporting the standards. Safety trade-offs associated with mass-reduction have occurred in the past, particularly before standards were attribute-based because manufacturers chose, in response to standards, to build smaller and lighter vehicles; these smaller, lighter vehicles did not fare as well in crashes as larger, heavier vehicles, on average. Although NHTSA now uses attribute-based standards, in part to reduce or eliminate the incentive to downsize vehicles to comply with the standards, NHTSA must be mindful of the possibility of related safety trade-offs. For this reason, NHTSA accounts for how the application of MR to meet standards affects the safety of a specific vehicle given changes in GVWR.

For this final rule, the agency employed the modeling technique, which was developed in the 2016 Puckett and Kindelberger report and used in the proposal, to analyze the updated crash and exposure data by examining the cross sections of the societal fatality rate per billion vehicle miles of travel (VMT) by mass and footprint, while controlling for driver age, gender, and other factors, in separate logistic regressions for five vehicle groups and nine crash types. NHTSA utilized the relationships between weight and safety from this analysis, expressed as percentage increases in fatalities per 100-pound weight reduction (which is how MR is applied in the technology analysis; see Section III.D.4), to examine the weight impacts applied in this analysis. The effects of MR on safety were estimated relative to (incremental to) the regulatory reference baseline in the analysis, across all vehicles for MY 2021 and beyond. The analysis of MR includes two opposing impacts. Research has consistently shown that MR affects “lighter” and “heavier” vehicles differently across crash types. The 2016 Puckett and Kindelberger report found MR concentrated among the heaviest vehicles is likely to have a beneficial effect on overall societal fatalities, while MR concentrated among the lightest vehicles is likely to have a detrimental effect on occupant fatalities but a slight benefit to pedestrians and cyclists. This represents a relationship between the dispersion of mass across vehicles in the fleet and societal fatalities: decreasing dispersion is associated with a decrease in fatalities. MR in heavier vehicles is more beneficial to the occupants of lighter vehicles than it is harmful to the occupants of the heavier vehicles. MR in lighter vehicles is more harmful to the occupants of lighter vehicles than it is beneficial to the occupants of the heavier vehicles.

To accurately capture the differing effect on lighter and heavier vehicles, NHTSA splits vehicles into lighter and heavier vehicle classifications in the analysis. However, this poses a challenge of creating statistically meaningful results. There is limited relevant crash data to use for the analysis. Each partition of the data reduces the number of observations per vehicle classification and crash type, and thus reduces the statistical robustness of the results. The methodology employed by NHTSA was designed to balance these competing forces as an optimal trade-off to accurately capture the impact of mass-reduction across vehicle curb weights and crash types while preserving the potential to identify robust estimates.

Loyola and Bradbury commented that smaller and lighter vehicles built in response to the standards will increase the number of fatalities but did not note any deficiencies in the agency's analysis or consideration of mass-safety impacts. ACC and the Joint NGOs commented that changes in vehicle design and materials technology may lead to changes in relationships among vehicle mass and safety outcomes. NHTSA has acknowledged this potential outcome across multiple rulemakings and has continued to keep abreast of any new developments; however, for the time being, NHTSA feels there is insufficient data to support alternative estimates. NRDC further commented that manufacturers are capable of applying MR to a greater degree in heavier vehicles, yielding a net safety benefit to society. The CAFE Model incorporates the relationship raised by NRDC and the mass-size-safety coefficients applied in the model yield results consistent with this relationship when MR is applied to heavier vehicles more than lighter vehicles.

Multiple stakeholders commented that NHTSA failed to adequately account for changes in vehicle mass associated with changing from ICE to BEV platforms for a given vehicle model in the analysis of the reference baseline. In related comments, ACC and the Aluminum Association noted that BEVs are likely to have different safety profiles than ICE vehicles. We note, however, that there are no safety impacts resulting from a shift from ICE to BEV platforms in NHTSA's central analysis of the impact of CAFE standards because NHTSA's model is constrained such that no BEVs are added to the fleet during standard-setting years as a result of an increase in the stringency of CAFE standards. That is, any shift from ICE vehicles to BEVs in the standard setting years is limited to actions occurring in the reference baseline. In our analysis of the reference baseline, we account for an expected increase in BEVs as a result of market forces (like manufacturers' expected deployment of electric vehicles consistent with levels required by California's ACC II program) and regulatory requirements. However, while we acknowledge that, all else equal, vehicle masses likely increase when shifting from ICE to BEV platforms and BEVs may have distinct safety characteristics relative to ICE vehicles across crash types, we have insufficient data to account for how safety outcomes would be affected by shifting from ICE to BEV platforms in the analysis of the reference baseline, including insufficient information to justify an assumption that changes in mass associated with BEV structural differences are equivalent to changes in mass within ICE platforms. The CAFE Model is not currently designed to account for differences in vehicle mass associated with changes from ICE to BEV platforms. We are conducting research to address this lack of data in future rulemakings, but for this rule in the absence of sufficient data we have chosen to assume a neutral net safety effect for mass (and center of gravity) changes associated with shifts from ICE to BEV platforms for a given vehicle model in the baseline analysis. We acknowledge that ICE and BEV platforms for otherwise equivalent vehicles may differ in center of gravity, frontal crush characteristics, and acceleration. This creates uncertainty as to the validity of extrapolating observed mass-safety relationships from ICE vehicles to BEVs, however, until there is sufficient data and research to uncover an alternative relationship for BEVs, we believe that our current approach is reasonable.

The Joint NGOs and Consumer Reports also commented that the estimated mass-size-safety coefficients are statistically insignificant. We have acknowledged this relationship in this rulemaking along with previous rulemakings where the estimated coefficients are not statistically significant at the 95 percent confidence level. In this rulemaking, the distinction between using insignificant estimates and zeroes is functionally moot because the estimated societal safety impacts associated with changes in vehicle mass associated with the rule are estimated to be zero in the Preferred Alternative. Furthermore, courts have discouraged agencies from excluding specific costs or benefits because the magnitude is uncertain. Given the agency believes that the point estimates still represent the best available data, NHTSA continues to include a measurement of mass-safety impacts in its analysis.

A more detailed description of the mass-safety analysis can be found in Chapter 7.2 of the Final TSD.

2. Sales/Scrappage Impacts

The sales and scrappage responses to higher vehicle prices discussed in Section III.E.2 have important safety consequences and influence safety through the same basic mechanism, fleet turnover. In the case of the scrappage response, delaying fleet turnover keeps drivers in older vehicles which tend to be less safe than newer vehicles. Similarly, the sales response slows the rate at which newer vehicles, and their associated safety improvements, enter the on-road population. The sales response also influences the mix of vehicles on the road-with more stringent CAFE standards leading to a higher share of light trucks sold in the new vehicle market, assuming all else is equal. Light trucks have higher rates of fatal crashes when interacting with passenger cars and as earlier discussed, different directional responses to MR technology based on the existing mass and body style of the vehicle.

Any effect on fleet turnover (either from delayed vehicle retirement or deferred sales of new vehicles) will affect the distribution of both ages and MYs present in the on-road light duty and HDPUV fleets. Because each of these vintages carries with it inherent rates of fatal crashes, and newer vintages are generally safer than older ones, changing that distribution will change the total number of on-road fatalities under each regulatory alternative. Similarly, the Dynamic Fleet Share (DFS) model captures the changes in the light-duty fleet's composition of cars and trucks. As cars and trucks have different fatality rates, differences in fleet composition across the alternatives will affect fatalities.

At the highest level, NHTSA calculates the impact of the sales and scrappage effects by multiplying the VMT of a vehicle by the fatality risk of that vehicle. For this analysis, calculating VMT is rather simple: NHTSA uses the distribution of miles calculated in Chapter 4.3 of the Final TSD. The trickier aspect of the analysis is creating fatality rate coefficients. The fatality risk measures the likelihood that a vehicle will be involved in a fatal accident per mile driven. NHTSA calculates the fatality risk of a vehicle based on the vehicle's MY, age, and style, while controlling for factors that are independent of the intrinsic nature of the vehicle, such as behavioral characteristics. Using this same approach, NHTSA designed separate models for fatalities, non-fatal injuries, and property damaged vehicles.

The vehicle fatality risk described above captures the historical evolution of safety. Given that modern technologies are proliferating faster than ever and offer greater safety benefits than traditional safety improvements, NHTSA augmented the fatality risk projections with knowledge about forthcoming safety improvements. NHTSA applied estimates of the market uptake and improving effectiveness of crash avoidance technologies to estimate their effect on the fleet-wide fatality rate, including explicitly incorporating both the direct effect of those technologies on the crash involvement rates of new vehicles equipped with them, as well as the “spillover” effect of those technologies on improving the safety of occupants of vehicles that are not equipped with these technologies.

NHTSA's approach to measuring these impacts is to derive effectiveness rates for these advanced crash-avoidance technologies from safety technology literature. NHTSA then applies these effectiveness rates to specific crash target populations for which the crash avoidance technology is designed to mitigate, which are then adjusted to reflect the current pace of adoption of the technology, including any public commitment by manufacturers to install these technologies. These technologies include Forward Collision Warning, Automatic Emergency Braking, Lane Departure Warning, Lane Keep Assist, Blind Spot Detection, Lane Change Assist, and Pedestrian Automatic Emergency Braking. The products of these factors, combined across all 7 advanced technologies, produce a fatality rate reduction percentage that is applied to the fatality rate trend model discussed above, which projects both vehicle and non-vehicle safety trends. The combined model produces a projection of impacts of changes in vehicle safety technology as well as behavioral and infrastructural trends. A much more detailed discussion of the methods and inputs used to make these projections of safety impacts from advanced technologies is included in Chapter 7.1 of the Final TSD.

Loyola and Bradbury commented that the slowing of fleet turnover in response to the standards will increase fatalities but did not note any deficiencies in the agency's analysis or consideration of fleet turnover impacts. As such, the agency believes it has appropriately considered the issue the commenters raised.

Consumer Reports cited the sensitivity and uncertainty of NHTSA's sales module, including the dynamic fleet share component and scrappage model, and questioned the astuteness of including the safety impacts from these effects. Consumer Reports also noted that they have not observed these effects in practice. NHTSA thanks Consumer Reports for providing their research in their comments. While the agency believes their research is valuable, we were unable to arrive at the same conclusions.

3. Rebound Effect Impacts

The additional VMT demanded due to the rebound effect is accompanied by more exposure to risk, however, rebound miles are not imposed on consumers by regulation. They are a freely chosen activity resulting from reduced vehicle operational costs. As such, NHTSA believes a large portion of the safety risks associated with additional driving are offset by the benefits drivers gain from added driving. The level of risk internalized by drivers is uncertain. This analysis assumes that drivers of both HDPUV and light duty vehicles internalize 90 percent of this risk, which mostly offsets the societal impact of any added fatalities from this voluntary consumer choice. Additional discussion of internalized risk is contained in Chapter 7.5 of the TSD.

Consumer Reports commented that there is “no evidence whatsoever to support NHTSA's assumption that consumers internalize only 90% of the safety risk” and asks the agency to offset the entirety of rebound fatalities. Alternatively, Consumer Reports suggests that even though the agency's logic is sound for offsetting externality risks, if the risk were not internalized, because rebound driving is voluntary, it is still inappropriate to account for the increased fatality risks. Consumer Reports also expressed concern about the precedent of accounting for additional driving when consumers save money. The agency appreciates Consumer Reports comment but has chosen not to adjust its approach to offsetting rebound safety for the final rule. We agree with Consumer Reports that there is a dearth of evidence to support a 90 percent offset, but the agency also notes that there is no evidence to support a higher offset either. Accounting for rebound effects does not set a broader precedent beyond fuel efficiency rules. The rebound effect is generally recognized to be the phenomena of using more of an energy consuming product when its operating costs decline rather than how consumers will use energy consuming products as their income increases.

4. Value of Safety Impacts

Fatalities, nonfatal injuries, and property damage crashes are valued as a societal cost within the CAFE Model's cost and benefit accounting. Their value is based on the comprehensive value of a fatality, which includes lost quality of life and is quantified in the VSL as well as economic consequences such as medical and emergency care, insurance administrative costs, legal costs, and other economic impacts not captured in the VSL alone. These values were first derived from data in Blincoe et al. (2015), updated in Blincoe et al. (2023), and adjusted to 2021 dollars, and updated to reflect the official DOT guidance on the VSL.

Nonfatal injury costs, which differ by severity, were weighted according to the relative incidence of injuries across the Abbreviated Injury Scale (AIS). To determine this incidence, NHTSA applied a KABCO/MAIS translator to CRSS KABCO based injury counts from 2017 through 2019. This produced the MAIS-based injury profile. This profile was used to weight nonfatal injury unit costs derived from Blincoe et al. (2023), adjusted to 2021 economics and updated to reflect the official DOT guidance on the VSL. Property-damaged vehicle costs were also taken from Blincoe et al (2023). and adjusted to 2021 economics.

For the analysis, NHTSA assigns a societal value of $12.2 million for each fatality, $181,000 for each nonfatal injury, and $8,400 for each property damaged vehicle. As discussed in the previous section, NHTSA discounts 90% of the safety costs associated with the rebound effect. The remaining 10% of those safety costs are not considered to be internalized by drivers and appear as a cost of the standards that influence net benefits. Similarly, the effects on safety attributable to changes in mass and fleet turnover are not considered costs internalized by drivers since manufacturers are responsible for deciding how to design and price vehicles. The costs not internalized by drivers is therefore the summation of the mass-safety effects, fleet turnover effects, and the remaining 10% of rebound-related safety effects.

IV. Regulatory Alternatives Considered in This Final Rule

A. General Basis for Alternatives Considered

Agencies typically consider regulatory alternatives in order to evaluate the comparative effects of different potential ways of implementing their statutory authority to achieve their intended policy goals. NEPA requires agencies to compare the potential environmental impacts of their actions to a reasonable range of alternatives. E.O. 12866 and E.O. 13563, as well as OMB Circular A-4, also request that agencies evaluate regulatory alternatives in their rulemaking analyses.

Alternatives analysis begins with a “No-Action” Alternative, typically described as what would occur in the absence of any further regulatory action by the agency. OMB Circular A-4 states that “the choice of an appropriate baseline may require consideration of a wide range of potential factors, including:

• evolution of markets;
• changes in regulations promulgated by the agency or other government entities;
• other external factors affecting markets;
• the degree of compliance by regulated entities with other regulations; and
• the scale and number of entities or individuals that will be subject to, or experience the benefits or costs of, the regulation.”

This final rule includes a No-Action Alternative for passenger cars and light trucks and a No-Action alternative for HDPUVs, both described below; five “action alternatives” for passenger cars and light trucks; and four action alternatives for HDPUVs. Within both the set of alternatives that apply to passenger cars and light trucks and the set of alternatives that apply to HDPUVs, one alternative is identified as the “Preferred Alternative,” which is NEPA parlance. In some places the Preferred Alternative may also be referred to as the “standards” or “final standards,” but NHTSA intends “standards” and “Preferred Alternative” to be used interchangeably for purposes of this final rule. NHTSA believes the range of No-Action and action alternatives for each set of standards appropriately comports with CEQ's directive that “agencies shall . . . limit their consideration to a reasonable number of alternatives.”

The different regulatory alternatives for passenger cars and light trucks are defined in terms of percent-changes in CAFE stringency from year to year. Readers should recognize that those year-over-year changes in stringency are not measured in terms of mile per gallon differences (as in, 1 percent more stringent than 30 mpg in one year equals 30.3 mpg in the following year), but rather in terms of shifts in the footprint functions that form the basis for the actual CAFE standards (as in, on a gallon per mile basis, the CAFE standards change by a given percentage from one model year to the next).

For PCs, consistent with prior rulemakings, NHTSA is defining final fuel economy targets as shown in Equation IV-1.

Where:

TARGET_FE is the fuel economy target (in mpg) applicable to a specific vehicle model type with a unique footprint combination,

a is a minimum fuel economy target (in mpg),

b is a maximum fuel economy target (in mpg),

c is the slope (in gallons per mile per square foot, or gpm per square foot), of a line relating fuel consumption (the inverse of fuel economy) to footprint, and

d is an intercept (in gpm) of the same line.

Here, MIN and MAX are functions that take the minimum and maximum values, respectively, of the set of included values. For example, MIN[40, 35] = 35 and MAX(40, 25) = 40, such that MIN[MAX(40, 25), 35] = 35.

The resultant functional form is reflected in graphs displaying the passenger car target function in each model year for each regulatory alternative in Sections IV.B.1 and IV.B.3.

For LTs, also consistent with prior rulemakings, NHTSA is defining fuel economy targets as shown in Equation IV-2.

Where:

TARGET_FE is the fuel economy target (in mpg) applicable to a specific vehicle model type with a unique footprint combination,

a, b, c, and d are as for PCs, but taking values specific to LTs,

e is a second minimum fuel economy target (in mpg),

f is a second maximum fuel economy target (in mpg),

g is the slope (in gpm per square foot) of a second line relating fuel consumption (the inverse of fuel economy) to footprint), and

h is an intercept (in gpm) of the same second line.

NHTSA is defining HDPUV fuel efficiency targets as shown in Equation IV-3:

Where:

c is the slope of the gasoline, CNG, Strong Hybrid, and PHEV work factor target curve in gal/100 mile per WF

For diesel engines, BEVs and FCEVs, c will be replaced with e

d is the gasoline CNG, Strong Hybrid, and PHEV minimum fuel consumption work factor target curve value in gal/100 mile

For diesel engines, BEVs and FCEVs, d will be replaced with f

WF = Work Factor = [0.75 × (Payload Capacity + Xwd)] + [0.25 × Towing Capacity]

Where:

Xwd = 4wd adjustment = 500 lbs. if the vehicle group is equipped with 4wd and all-wheel drive (AWD), otherwise equals 0 lbs. for 2wd

Payload Capacity = GVWR (lbs.)−Curb Weight (lbs.) (for each vehicle group)

Towing Capacity = GCWR (lbs.)−GVWR (lbs.) (for each vehicle group)

In a departure from recent CAFE rulemaking trends, for this final rule, we have applied different rates of increase to the passenger car and the light truck fleets in different model years. For the Preferred Alternative, rather than have both fleets increase their respective standards at the same rate, passenger car standards will increase at a steady rate year over year, while light truck standards will not increase for a few years before beginning to rise again at the passenger car rate. Several action alternatives evaluated for this final rule have passenger car fleet rates-of-increase of fuel economy that are different from the rates-of-increase of fuel economy for the light truck fleet, while the Preferred Alternative has the same rate of increase for passenger cars and light trucks for three out of the five model years. NHTSA has discretion, by law, to set CAFE standards that increase at different rates for cars and trucks, because NHTSA must set maximum feasible CAFE standards separately for cars and trucks.

For HDPUVs, the different regulatory alternatives are also defined in terms of percent-increases in stringency from year to year, but in terms of fuel consumption reductions rather than fuel economy increases, so that increasing stringency appears to result in standards going down (representing a direct reduction in fuel consumed) over time rather than up. Also, unlike for the passenger car and light truck standards, because HDPUV standards are in fuel consumption space, year-over-year percent changes actually do represent gallon/mile differences across the work-factor range. For the Preferred Alternative, the stringency increases at one fixed percentage rate in each the first three model years, and a different fixed percentage rate in each of the remaining three model years in the rulemaking time frame. Under the other action alternatives, the stringency changes at the same percentage rate in each model year in the rulemaking time frame. One action alternative is less stringent than the Preferred Alternative for HDPUVs, and two action alternatives are more stringent.

B. Regulatory Alternatives Considered

The regulatory alternatives considered by the agency in this final rule are presented here as the percent-changes-per-year that they represent. The sections that follow will present the alternatives as the literal coefficients that define standards curves increasing at the given percentage rates.

A variety of factors will be at play simultaneously as manufacturers seek to comply with the final standards that NHTSA is promulgating. NHTSA, EPA, and CARB will all be regulating simultaneously; manufacturers will be responding to those regulations as well as to foreseeable shifts in market demand during the rulemaking time frame (both due to cost/price changes for different types of vehicles over time, fuel price changes, and the recently-passed tax credits for BEVs and PHEVs). Many costs and benefits that will accrue as a result of manufacturer actions during the rulemaking time frame will be occurring for reasons other than CAFE standards, and NHTSA believes it is important to try to reflect many of those factors in order to present a more accurate picture of the effects of different potential CAFE and HDPUV standards to decision-makers and to the public. Because the EPA and NHTSA programs were developed in coordination jointly, and stringency decisions were made in coordination, NHTSA did not incorporate EPA's only recently-finalized CO₂ standards as part of the analytical reference baseline for the main analysis. The fact that EPA finalized its rule before NHTSA is an artifact of circumstance only.

The following sections define each regulatory alternative, including the No-Action Alternative, for each program, and explain their derivation.

1. Reference Baseline/No-Action Alternative

As with the 2022 final rule, our No-Action Alternative (also referred to as the reference baseline) is fairly nuanced. In this analysis, the reference No-Action Alternative assumes:

• The existing (through model year 2026) national CAFE and GHG standards are met, and that the CAFE and GHG standards for model year 2026 finalized in 2022 continue in perpetuity.

• Manufacturers who committed to the California Framework Agreements met their contractual obligations for model year 2022.
• The HDPUV model year 2027 standards finalized in the NHTSA/EPA Phase 2 program continue in perpetuity.
• Manufacturers will comply with the Advanced Clean Trucks (ACT) program that California and other states intend to implement through 2035.
• Manufacturers will, regardless of the existence or non-existence of a legal requirement, produce additional electric vehicles consistent with the levels that would be required under the ZEV/Advanced Clean Cars II program, if it were to be granted a Clean Air Act preemption waiver.
• Manufacturers will make production decisions in response to estimated market demand for fuel economy or fuel efficiency, considering estimated fuel prices, estimated product development cadence, the estimated availability, applicability, cost, and effectiveness of fuel-saving technologies, and available tax credits.

NHTSA continues to believe that to properly estimate fuel economies/efficiencies (and achieved CO₂ emissions) in the No-Action Alternative, it is necessary to simulate all of these legal requirements, additional deployment plans of automakers, and other influences affecting automakers and vehicle design simultaneously. Consequently, the CAFE Model evaluates each requirement in each model year, for each manufacturer/fleet. Differences among fleets and compliance provisions often create over-compliance in one program, even if a manufacturer is able to exactly comply (or under-comply) in another program. This is similar to how manufacturers approach the question of concurrent compliance in the real world—when faced with multiple regulatory programs, the most cost-effective path may be to focus efforts on meeting one or two sets of requirements, even if that results in “more effort” than would be necessary for another set of requirements, in order to ensure that all regulatory obligations are met. We elaborate on those model capabilities below. Generally speaking, the model treats each manufacturer as applying the following logic when making technology decisions, both for simulating passenger car and light truck compliance, and HDPUV compliance, with a given regulatory alternative:

1. What do I need to carry over from last year?

2. What should I apply more widely in order to continue sharing (of, e.g., engines) across different vehicle models?

3. What new BEVs do I need to build in order to satisfy the various state ZEV programs and voluntary deployment of electric vehicles consistent with ACC II?

4. What further technology, if any, could I apply that would enable buyers to recoup additional costs within 30 months after buying new vehicles?

5. What additional technology, if any, should I apply to respond to potential new CAFE and CO₂ standards for PCs and LTs, or to potential new HDPUV standards?

Additionally, within the context of 4 and 5, the CAFE Model may consider, as appropriate and allowed by statutory restrictions on technology application for a given model year, the applicability of recently-passed tax credits for battery-based vehicle technologies, which improve the attractiveness of those technologies to consumers and thus the model's likelihood of choosing them as part of a compliance solution. The model can also apply over-compliance credits if applicable and not legally prohibited. The CAFE Model simulates all of these simultaneously. As mentioned above, this means that when manufacturers make production decisions in response to actions or influences other than CAFE or HDPUV standards, those costs and benefits are not attributable to possible future CAFE or HDPUV standards. This approach allows the analysis to isolate the effects of the decision being made on the appropriate CAFE standards, as opposed to the effects of many things that will be occurring simultaneously.

To account for the existing CAFE standards finalized in model year 2026 for passenger cars and light trucks, the No-Action Alternative includes the following coefficients defining those standards, which (for purposes of this analysis) are assumed to persist without change in subsequent model years:

These coefficients are used to create the graphic below, where the x-axis represents vehicle footprint and the y-axis represents fuel economy, showing that in “CAFE space,” targets are higher in fuel economy for smaller footprint vehicles and lower for larger footprint vehicles.

Additionally, EPCA, as amended by EISA, requires that any manufacturer's domestically-manufactured passenger car fleet must meet the greater of either 27.5 mpg on average, or 92 percent of the average fuel economy projected by the Secretary for the combined domestic and non-domestic passenger automobile fleets manufactured for sale in the United States by all manufacturers in the model year. NHTSA retains the 1.9 percent offset to the Minimum Domestic Passenger Car Standard (MDPCS), first used in the 2020 final rule, to account for recent projection errors as part of estimating the total passenger car fleet fuel economy, and used in rulemakings since. The projection shall be published in the Federal Register when the standard for that model year is promulgated in accordance with 49 U.S.C. 32902(b). For purposes of the No-Action Alternative, the MDPCS is as it was established in the 2022 final rule for model year 2026, as shown in Table IV-5 below:

To account for the HDPUV standards finalized in the Phase 2 rule, the No-Action Alternative for HDPUVs includes the following coefficients defining those standards, which (for purposes of this analysis) are assumed to persist without change in subsequent model years:

These equations are represented graphically below:

As the reference baseline scenario, the No-Action Alternative also includes the following additional actions that NHTSA believes will occur in the absence of further regulatory action by NHTSA:

To account for the existing national GHG emissions standards, the No-Action Alternative for passenger cars and light trucks includes the following coefficients defining the GHG standards set by EPA in 2022 for model year 2026, which (for purposes of this analysis) are assumed to persist without change in subsequent model years:

Coefficients a, b, c, d, e, and f define the model year 2026 Federal CO₂ standards for passenger cars and light trucks, respectively, in Table IV-8 and Table IV-9 above. Analogous to coefficients defining CAFE standards, coefficients a and b specify minimum and maximum CO₂ targets in each model year. Coefficients c and d specify the slope and intercept of the linear portion of the CO₂ target function, and coefficients e and f bound the region within which CO₂ targets are defined by this linear form.

To account for the NHTSA/EPA Phase 2 national GHG emission standards, the No-Action Alternative for HDPUVs includes the following coefficients defining the WF based standards set by EPA for model year 2027 and beyond. The four-wheel drive coefficient is maintained at 500 (coefficient `a') and the weighting multiplier coefficient is maintained at 0.75 (coefficient `b'). The CI and SI coefficients are in the tables below:

Coefficients c, d, e, and f define the existing model year 2027 and beyond CO₂ standards from Phase 2 rule for HDPUVs, in Table III-10 and Table III-11 above. The coefficients are linear work-factor based function with c and d representing gasoline, CNG vehicles, SHEVs and PHEVS and e and f representing diesels, BEVS and FCEVs. For this rulemaking, this is identical to the NHTSA's fuel efficiency standards No Action alternative.

The reference baseline No-Action Alternative also includes NHTSA's estimates of ways that each manufacturer could introduce new PHEVs and BEVs in response to state ZEV programs and additional production of PHEVs and BEVs that manufacturers have indicated they will undertake consistent with ACC II, regardless of whether it becomes a legal requirement. To account for manufacturers' expected compliance with the ACC I and ACT programs and additional deployment of electric vehicles consistent with ACC II, NHTSA has included the main provisions of the ACC, ACC II, (as currently submitted to EPA), and ACT programs in the CAFE Model's analysis. Incorporating these programs into the model includes converting vehicles that have been identified as potential ZEV candidates into battery-electric vehicles (BEVs) and taking into account PHEVs that meet the ZEV PHEV credit requirements so that a manufacturer's fleet meets the calculated ZEV credit requirements or anticipated voluntary compliance. The CAFE Model makes manufacturer fleets consistent with ACC I, ACC II (as currently submitted to EPA), and ACT first in the reference baseline, then solves for the technology pathway used to meet increasing ZEV penetration levels described by the state programs. Chapter 2.3 of the Final TSD discusses, in detail, how NHTSA developed these estimates.

Several stakeholders commented in support of NHTSA's inclusion of state ZEV programs and assumptions regarding other electric vehicles that will be deployed in the absence of legal requirements in the reference baseline. The States and Cities, for example, commented that “[g]iven NHTSA's duty to project a No-Action baseline that accounts for sharply growing zero emission vehicle sales, modeling compliance with California's Advanced Clean Cars I (“ACCI”), Advanced Clean Cars II (“ACCII”), and Advanced Clean Trucks (“ACT”) regulations is a reasonable methodology to do so, at least in the event that California is granted its requested waiver for ACCII and ACCII thus becomes enforceable.” Similarly, the Joint NGOs commented that “consistent with EPCA's language, history, and legislative intent, NHTSA models an accurate, real-world `no action' baseline for the rulemaking, a task that requires a rational accounting of the real-world BEVs and PHEVs projected to exist in the absence of the CAFE standards NHTSA is considering. . . . NHTSA has done so here.”

Some stakeholders commented about uncertainties that they believe could impact the reference baseline. For example, Kia commented that “[w]hile automakers will plan to comply with the regulations, there is great uncertainty as to whether automakers have the capacity to do so, whether the California ZEV mandate will remain as currently written through 2035, whether states that have adopted it will remain in the program, and whether California will be granted a waiver.”

Other stakeholders commented in explicit opposition to modeling state ZEV programs in the reference baseline. Stakeholders asserted that NHTSA could not account for state ZEV programs in the light-duty standards reference baseline because of EPCA/EISA's statutory prohibition on considering electric vehicle fuel economy in 49 U.S.C. 32902(h). Several of these commenters objected in particular to NHTSA's use of OMB Circular A-4 to guide the development of the light-duty regulatory reference baseline, as they believe that Circular A-4 cannot “trump a clear statutory requirement,” referring to 49 U.S.C. 32902(h). Stakeholders also commented that state ZEV programs should not be included in the reference baseline because they are preempted by various federal laws, and/or because EPA has not yet granted a waiver of preemption to California for the ACC II program. Commenters opposing the inclusion of state ZEV programs in the reference baseline also alleged that it was a backdoor way to establish an EV mandate when setting CAFE standards.

Toyota did not explicitly object to NHTSA's consideration of state ZEV regulatory programs in the reference baseline but stated that “NHTSA should consider the impact of the EVs stemming from both the ZEV Mandate and the GHG Program, but then use that knowledge to establish economically practicable CAFE standards for the remining ICEs in the U.S. fleet, thereby simultaneous[sic] satisfying 49 U.S.C. 32902(h). For example, if 45 percent of a projected 17 million vehicle fleet in 2030 model year will be electrified due to other government programs, CAFE standards would be set for the remaining 9.4 million ICE and hybrid vehicles.”

Several stakeholders also commented about specific assumptions used in the ZEV modeling such as the number of states signed on to the program, how some compliance obligations should be assumed to be met through credits, and assumptions around PHEV credit values; those comments are addressed in Section III.C.5, above.

NHTSA agrees with commenters that the agency has a duty to model a reference baseline that includes increasing zero emission vehicle sales in response to state standards, and that the agency's methodology for doing so is consistent with EPCA's language, history, and legislative intent. NHTSA continues to believe that it is appropriate for the reference baseline to reflect legal obligations other than CAFE standards that automakers will be meeting and additional non-regulatory deployment of electric vehicles during this time period so that the regulatory analysis can identify the distinct effects of the CAFE standards. Information provided by California continues to show there has been industry compliance with the ZEV standards, which provides further confirmation that manufacturers will meet legally-binding state standards. This is also confirmed by manufacturers' stated intent to deploy electric vehicles consistent with what would be required under ACC II, regardless of whether it becomes a binding legal obligation, as discussed in more detail below.

In response to comments opposing the inclusion of state ZEV programs in the reference baseline because doing so conflicts with 49 U.S.C. 32902(h), NHTSA maintains that it is perfectly possible to give meaningful effect to the 49 U.S.C. 32902(h) prohibition by not allowing the CAFE Model to rely on ZEV (or other dedicated alternative fuel) technology during the rulemaking time frame, while still acknowledging the clear reality that the state ZEV programs exist, and manufacturers are complying with them, just like the agency acknowledges that electric vehicles exist in the fleet independent of the ZEV program. Comments regarding whether including state ZEV programs in the reference baseline is consistent with 49 U.S.C. 32902(h) are discussed in more detail below in Section VI.A.5.a.(5), and in the final rule for model years 2024-2026 CAFE standards. Regarding commenters' views that state ZEV programs are preempted, NHTSA addressed preemption in the agency's 2021 rulemaking, and further discussion is located in the NPRM and final rule for that rulemaking. In that rulemaking, the agency expressed “significant doubts as to the validity” of preemption positions similar to those raised by commenters here.

NHTSA also disagrees that including state ZEV programs in the reference baseline is a way to, according to commenters, “bypass” limitations in 49 U.S.C. 32902(h). ACC I is a relevant legal requirement that manufacturers must meet, and as mentioned above, manufacturers are not just meeting those standards, they are exceeding them. Further, manufacturers have indicated their intent to deploy electric vehicles consistent with what would be required under ACC II, regardless of whether it becomes a binding legal obligation. Vehicle manufacturers told NHTSA, in CBI conversations regarding planned vehicle product and technology investments, that they are complying with and plan to comply in the future with ZEV programs. These conversations were later confirmed by manufacturers' subsequent public announcements, confirming both their support for California's programs and for meeting their own stated electrification goals, which are discussed in extensive detail below.

Kia, stating in their comments that “automakers will plan to comply with the regulations,” joins a list of OEMs that have established that they are planning technology decisions to comply with state ZEV program deployment levels: Stellantis in a recent agreement with California confirmed that they will explicitly comply with the ACC programs through 2030; General Motors sent a letter to California Governor Gavin Newsom both recognizing California's authority under the Clean Air Act to set vehicle emissions standards and expressing its commitment to “emissions reductions that are aligned with the California Air Resources Board's targets and . . . complying with California's regulations”, and Ford, Volkswagen, BMW, Honda, and Volvo formed a group of five manufacturers that committed in 2020 to comply with ZEV program requirements and have since reiterated their support for California's programs in a lengthy declaration to the D.C. Circuit Court of Appeals. Not only have all three domestic automakers expressed support for California's standards, several other automakers have followed suit in explicitly expressing support for California's programs, as shown above.

Further, automakers have publicly signaled their commitment to the EV transition at levels that well exceed the 28 percent BEV market share in MY 2031 reflected in the baseline reference case. In August 2021, major automakers including GM, Ford, Stellantis, BMW, Honda, Volkswagen, and Volvo pledged their support to achieve 40 to 50 percent sales of electric vehicles by 2030. These announcements are consistent with previous and ongoing corporate statements. Several manufacturers have announced plans to fully transition to electric vehicles, such as General Motors ambition to shift its light-duty vehicles entirely to zero-emissions by 2035, Volvo's plans to make only electric cars by 2030, Mercedes plans to become ready to go all-electric by 2030 where possible, and Honda's full electrification plan by 2040. Other car makers have chosen incremental commitments to electrification that are still exceed the equivalent national EV market share reflected in the reference baseline, such as Ford's announcement that the company expects 40 percent of its global sales will be all-electric by 2030, Volkswagen's expectation that half of its U.S. sales will be all-electric by 2030, Subaru's global target to achieve 50 percent BEVs by 2030, and Toyota's plans to introduce 30 BEV models by 2030. In addition to Honda's fully-electric target in 2040, the company also expects 40 percent of North American sales to be fully electric by 2030, and 80 percent by 2035.

The transition to electric vehicles is also taking place among heavy-duty pick-up trucks and vans, with much of the initial focus on last mile delivery vans. Several models of parcel delivery vans have already entered the market including GM's BrightDrop Zevo 400 and Zevo 600; and the Rivian EDV 500 and EDV 700. Commercial fleets have announced commitments to purchase zero emission delivery trucks and vans, including FedEx, Amazon, and Walmart. Amazon reached 10,000 electric delivery vans operating in over 18,000 U.S. cities.

These commitments provide further confirmation that automakers plan to deploy electric vehicles at the levels indicated in the reference baseline. They also provide further evidence that NHTSA's modeled reference baseline is a reasonable—yet, as discussed further below, likely conservative—representation of manufacturers' future product offerings. Nevertheless, NHTSA developed an alternative baseline that does not include ACC I or manufacturer deployment of electric vehicles that would be consistent with ACC II—and as discussed below, NHTSA determined that its final standards are reasonable as compared against this alternative baseline.

In response to Toyota's alternative approach to considering state ZEV programs in the analysis, not only does NHTSA not believe this approach would allow the agency to set maximum feasible standards, but NHTSA believes that the agency functionally already does what Toyota is describing. In addition, by converting vehicles to BEVs to comply with the ZEV program first, and then applying technology to the rest of the remaining fleet, NHTSA is setting a standard based only on the capability of the rest of the fleet to apply non-BEV technology.

Finally, in regards to including BEVs in the light-duty reference baseline, while NHTSA agrees that OMB Circular A-4 cannot trump a clear statutory requirement, NHTSA disagrees the agency's reference baseline does or attempts to do so. Nowhere does EPCA/EISA say that NHTSA should not consider the best available evidence in establishing the regulatory reference baseline for its CAFE rulemakings. As explained in Circular A-4, “the benefits and costs of a regulation are generally measured against a no-action baseline: an analytically reasonable forecast of the way the world would look absent the regulatory action being assessed, including any expected changes to current conditions over time.” NHTSA makes clear that its interpretation of 49 U.S.C. 32902(h) restricts the agency's analytical options when analyzing what standards are maximum feasible, while being consistent with A-4's guidance about how best to construct the reference baseline. Thus, absent a clear indication to blind itself to important facts, NHTSA continues to believe that the best way to implement its duty to establish maximum feasible CAFE standards is to establish as realistic a reference baseline as possible, including, among other factors, the most likely composition of the fleet. This concept is discussed in more detail in Section VI.A.

In addition to their comments opposing the inclusion of ACC I and ACC II in the light duty reference baseline, Valero also commented opposing NHTSA's inclusion of the ACT program in the HDPUV reference baseline, for several reasons. Regarding Valero's statutory arguments, we direct Valero to EPA's grant of the waiver of preemption for California's ACT program. EPA made requisite findings under the Clean Air Act that the waiver should be granted and also grappled with several issues that commenters raised about the program. NHTSA defers to EPA's judgment there. Valero also took issue with the fact that all states that have adopted California's ACT program standards have adopted them on a different timeline than California, for example Massachusetts' program beings with model year 2025 and Vermont's program begins in model year 2026. NHTSA defers to EPA on what is an appropriate interpretation of 42 U.S.C. 7507 but believes the agency has appropriately modeled a most likely future scenario as a reference baseline for future years.

Separately, NHTSA can include a legal obligation in the reference baseline that “has not yet begun implementation or demonstrated feasibility,” contrary to Valero's assertions. First, regarding the program having “not yet begun implementation”: a reference baseline is an “analytically reasonable forecast of the way the world would look absent the regulatory action being assessed” (emphasis added), and the nature of the Clean Air Act waiver process is that EPA grants waivers for programs that will affect future model years.

Regarding the argument that the ACT program has not demonstrated feasibility, Chapter 2.5.1 of the TSD shows the ZEV sales percentage requirements for Class 2b and 3 trucks (the vehicles covered by the HDPUV standards included in this final rule) and in the near-term, model years 2024-2026, the requirements increase by just 3% per year, and then only by 5% per year in the model years after that. The HDPUV segment is also a fraction of the size of the light-duty segment, as discussed elsewhere in this preamble, but stakeholders have already identified portions of the HDPUV segment that are candidates for electrification. For example, a North American Council for Freight Efficiency (NACFE) study of electrification for vans and step vans found that “fleets are aggressively expanding their purchases of electric vans and step vans after successful pilot programs.” Delivery vans are especially suited for electrification because range is typically not a major factor in urban delivery/e-commerce solutions, which in particular are spurring a rapid growth in the van and step van market segment. In other words, the market seems to be heading in a direction to meet state HDPUV ZEV programs not solely because of the requirements, but also because the segment is ready for it. Valero's characterization of state ACT programs as “the transition of a large and complex transportation system” and a “massive undertaking,” is an inaccurate dramatization of the scale of the ACT program in relation to NHTSA's current analysis.

Like for the NPRM, NHTSA additionally ran the CAFE Model for the HDPUV analysis assuming the ACT program was not included in the reference baseline. In the RIA, Table 9-8 highlights the changes in technology penetration for the HDPUV No ZEV sensitivity. We see that by model year 2038, BEV penetration decreases by just 0.2% and mild hybrid penetration increases by 4.9% when compared to the reference baseline. Between 2022-2050 we also see net social benefits increase by $1.81b, gasoline consumption is reduced by 1 billion gallons, and regulatory costs per vehicle increase by $41. This happens for two reasons: BEVs are still a relatively cost-effective technology for compliance with increasing levels of standards, and all of the benefits captured by the ACT program in the reference baseline are now attributable to our HDPUV program in the alternative case. Removing the ACT program from the HDPUV reference baseline has little impact on the analysis and it alone does not lead us to change our preferred alternative.

The No-Action Alternative also includes NHTSA estimates of ways that manufacturers could take advantage of recently-passed tax credits for battery-based vehicle technologies. NHTSA explicitly models portions of three provisions of the IRA when simulating the behavior of manufacturers and consumers. The first is the Advanced Manufacturing Production Tax Credit (AMPC). The AMPC also includes a credit for the production of applicable minerals. This provision of the IRA provides a $35 per kWh tax credit for manufacturers of battery cells and an additional $10 per kWh for manufacturers of battery modules (all applicable to manufacture in the United States). These credits, with the exception of the critical minerals credit, phase out 2030 to 2032. The agency also jointly modeled the Clean vehicle credit and the Credit for qualified commercial clean vehicles (CVCs), which provides up to $7,500 toward the purchase of clean vehicles covered by this regulation. The AMPC and CVCs provide tax credits for light-duty and HDPUV PHEVs, BEVs, and FCVs. Chapter 2.3 in the TSD discusses, in detail, how NHTSA has modeled these tax credits.

Stakeholders commented that NHTSA both underestimated and overestimated the effect of tax credits on reference baseline EV adoption for both the light-duty and HDPUV analyses. For example, IPI commented that “[a]lthough NHTSA's baseline modeling includes many commendable elements . . . NHTSA appears to underestimate the baseline share of BEVs resulting from the IRA during the Proposed Rule's compliance period. This, in turn, likely produces an underestimate of baseline average fuel economy and a corresponding overestimate of compliance cost.” On the other hand, the Corn Growers Associations commented that NHTSA overestimated the CVC, and did not support its assumptions surrounding its credit estimates. In regards to the HDPUV analysis, ACEEE commented that “[b]y excluding the Commercial Credit from its baseline analysis, NHTSA risks underestimating the additional positive impact that the IRA is projected to have on market penetration of BEVs in its no-action scenarios for passenger cars and HDPUVs.” Rivian similarly commented that they strongly supported NHTSA's stated intention to consult with EPA to implement the Commercial CVC in the final rule. NHTSA did not receive any comments recommending the agency not include tax credits in the final rule.

NHTSA believes that its approach to modeling available tax credits reasonably represents the ways that tax credits could be applied to vehicles in the reference baseline during the years covered by the standards. NHTSA disagrees that its assumptions were not well supported and notes that the agency included a significant and transparent discussion of the modeling assumptions the agency used in the NPRM and associated technical documents. However, for this final rule, NHTSA has refined important aspects of its tax credit modeling and presents additional supporting documentation about those assumptions in Section III.C.5, above, and in Chapter 2 of the Final TSD. In particular, for the final rule analysis in response to comments and in light of further guidance from the Department of Treasury, NHTSA modeled the § 45W tax credit jointly with § 30D. NHTSA believes that these additional updates ensure the agency's handling of tax credits does not over or underestimate their effect in the reference baseline.

The No-Action Alternative for the passenger car, light truck, and HDPUV fleets also includes NHTSA's assumption, for purposes of compliance simulations, that manufacturers will add fuel economy- or fuel efficiency-improving technology voluntarily, if the value of future undiscounted fuel savings fully offsets the cost of the technology within 30 months. This assumption is often called the “30-month payback” assumption, and NHTSA has used it for many years and in many CAFE rulemakings. It is used to represent consumer demand for fuel economy. It can be a source of apparent “over-compliance” in the No-Action Alternative, especially when technology is estimated to be extremely cost-effective, as occurs later in the analysis time frame when learning has significant effects on some technology costs.

NHTSA has determined that manufacturers do at times improve fuel economy even in the absence of new standards, for several reasons. First, overcompliance is not uncommon in the historical data, both in the absence of new standards, and with new standards—NHTSA's analysis in the 2022 TSD included CAFE compliance data showing that from 2004-2017, while not all manufacturers consistently over-complied, a number did. Of the manufacturers who did over-comply, some did so by 20 percent or more, in some fleets, over multiple model years. Ordinary market forces can produce significant increases in fuel economy, either because of consumer demand or because of technological advances.

Second, manufacturers have consistently told NHTSA that they do make fuel economy improvements where the cost can be fully recovered in the first 2-3 years of ownership. The 2015 NAS report discussed this assumption explicitly, stating: “There is also empirical evidence supporting loss aversion as a possible cause of the energy paradox. Greene (2011) showed that if consumers accurately perceived the upfront cost of fuel economy improvements and the uncertainty of fuel economy estimates, the future price of fuel, and other factors affecting the present value of fuel savings, the loss-averse consumers among them would appear to act as if they had very high discount rates or required payback periods of about 3 years.” Furthermore, the 2020 NAS HD report states: ”The committee has heard from manufacturers and purchasers that they look for 1.5- to 2-year paybacks or, in other cases, for a payback period that is half the expected ownership period of the first owner of the vehicle.” Naturally, there are heterogenous preferences for vehicle attributes in the marketplace: at the same time that we are observing record sales of electrified vehicles, we are also seeing sustained demand for pickup trucks with higher payloads and towing capacity and hence lower fuel economy. This analysis, like all the CAFE analyses preceding it, uses an average value to represent these preferences for the CAFE fleet and the HDPUV fleet. The analysis balances the risks of estimating too low of a payback period, which would preclude most technologies from consideration regardless of potential cost reductions due to learning, against the risk of allowing too high of a payback period, which would allow an unrealistic cost increase from technology addition in the reference baseline fleet.

Third, as in previous CAFE analyses, our fuel price projections assume sustained increases in real fuel prices over the course of the rule (and beyond). As readers are certainly aware, fuel prices have changed over time—sometimes quickly, sometimes slowly, generally upward. See further details of this in TSD Chapter 3.2.

In the 1990s, when fuel prices were historically low, manufacturers did not tend to improve their fuel economy, likely in part because there simply was very little consumer demand for improved fuel economy and CAFE standards remained flat due to appropriations riders from Congress preventing their increase. In subsequent decades, when fuel prices were higher, many of them have exceeded their standards in multiple fleets, and for multiple years. Our current fuel price projections look more like the last two decades, where prices have been more volatile, but also closer to $3/gallon on average. In recent years, when fuel prices have generally declined on average and CAFE standards have continued to increase, fewer manufacturers have exceeded their standards. However, our compliance data show that at least some manufacturers do improve their fuel economy if fuel prices are high enough, even if they are not able to respond perfectly to fluctuations precisely when they happen. This highlights the importance of fuel price assumptions both in the analysis and in the real world on the future of fuel economy improvements.

Stakeholders commented that the 30-month/2.5-year payback assumption should be shorter (or nonexistent) or significantly longer and specifically mentioned the effects of that assumption and alternative assumptions on the reference baseline. Consumer Reports reiterated their opposition to NHTSA's inclusion of the 2.5-year payback assumption, citing previous comments they had submitted to past CAFE rules and discussing additional historical data and arguments. The Joint NGOs also re-submitted comments to prior rules opposing the 30-month payback assumptions.

On the other hand, CEA commented in opposition to the use of a 30-month payback period and stated that it should be significantly longer, and pointed to NHTSA's 60-month sensitivity case as an example of how that assumption was important enough to be included in the main analysis. Valero also commented in opposition to the 30-month payback assumption specifically in the HDPUV analysis, calling it “unsupported” and identified a situation where “between model year 2029 and 2030, the CAFE Model projects that 168 models of Conventional, MHEV, or SHEV HDPUVs will be converted to BEVs in the No Action scenario—only 40 of those powertrain conversions have a modeled “Payback” of less than 30 months, and none have a “Payback TCO” of less than 30 months.” CEA similarly commented in opposition to the use of a 30-month payback period in the HDPUV analysis.

In preparation for this final rule, NHTSA updated its review of research supporting the 30-month payback assumption and continued to use that value for this final rule. Additional details on this research survey are discussed in Section III.E, above, and in detail in FRIA Chapter 2.1.4. NHTSA also performed a range of sensitivity cases using different payback assumptions, and those cases are discussed in detail in FRIA Chapter 9. While NHTSA modeled those cases to determine the effect of different payback assumptions on the levels of standards, NHTSA still believes that 30 months is the most appropriate value to use for the central analysis. Regarding Valero's comment about cost-effective technology application in the HDPUV analysis, NHTSA believes that Valero is missing the effect of tax credits in the effective cost calculation. When the CAFE Model determines if a technology is cost effective, it assesses the total cost of applying that technology and subtracts any available tax credits, fuel savings, and reduction in fines (if applicable for the analysis). The columns in the output file that Valero references in their comments is what the CAFE Model computes internally for only fuel savings for each vehicle and does not include tax credits or fines (if applicable). Additional details on the effective cost calculation are included in Section III.C.6 above and in the FRM CAFE Model Documentation.

NHTSA also received several general comments that reiterated the need for the agency to accurately consider EVs in the reference baseline, unrelated to state ZEV programs, tax credits, or consumer willingness to pay for increased fuel economy. Rivian commented that “ignoring [EVs] in determining how automakers can and should improve fuel economy in their fleets is nonsensical.” As discussed above, the Joint NGOs commented that “consistent with EPCA's language, history, and legislative intent, NHTSA models an accurate, real-world `no action' baseline as a starting point for the rulemaking, a task that requires a rational accounting of the real-world BEVs and PHEVs projected to exist in the absence of the CAFE standards NHTSA is considering setting.” However, the Joint NGOs stated that “in an abundance of caution” in light of the ongoing litigation in NRDC v. NHTSA, No. 221080 (D.C. Cir.), NHTSA should “model and evaluate the effect of alternative ways in which it could account for the real-world existence of BEVs/PHEVs in regulatory no-action alternatives,” like changing its assumptions surrounding compliance with state ZEV programs.

NHTSA also received several requests for the agency to account for manufacturer EV announcements in the reference baseline, or general comments that because manufacturer EV announcements were not included in the reference baseline, NHTSA's reference baseline underrepresented future EV penetration rates. Consumer Reports commented that “[i]n order to finalize a rule that achieves its statutory requirements to set maximum feasible standards that continue to reduce fuel consumption from gasoline-powered vehicles, NHTSA must appropriately consider the market share of electric vehicles that will exist in the fleet in the absence of the CAFE rule. Failure to consider the significant and rapidly growing sales of electric vehicles will result in a rule that serves no useful purpose, because the stringency will be too low to affect automakers' decisions to deploy fuel saving technology.” However, Consumer Reports also stated that they found the percentage of EVs in NHTSA's modeled reference baseline to be “extremely conservative” based on projections of future EV market share: “even some of the most cautious estimates are significantly greater than NHTSA's constrained baseline, indicating that it is an extremely conservative approach” Similarly, the States and Cities commented that “[b]ecause NHTSA's modeling does not account for significant zero-emission vehicle sales outside of the States adopting ACCI/II and ACT, its No-Action scenario likely significantly underestimates the zero emission vehicles in the baseline fleet. Because this underestimation may result in less stringent standards than are truly the “maximum feasible” standards, 49 U.S.C. 32902(a), NHTSA should consider modeling zero-emission vehicle adoption in States not adopting ACCI/II and ACT.” Tesla likewise commented that “NHTSA's baseline suggests BEV technology market penetration rates that are low,” and that NHTSA “must ensure it utilize[s public commitments from manufacturers] in its analysis of the industry and recognize shifts towards BEV technology in the marketplace is occurring for reasons outside of the CAFE standards setting process.”

NHTSA agrees that having an accurate reference baseline results in a more accurate analysis. However, in practice, it can be difficult to model manufacturer deployment plans without the structure that a regulatory program provides. NHTSA believes that the agency's modeling methodology, which incorporates state ZEV requirements that are legally binding and manufacturer commitments to deploy electric vehicles that would be consistent with the targets of California's ACC II program, regardless of whether it receives a waiver of Clean Air Act preemption, is the most reasonable approach available to the agency at present. Per the nature of NHTSA's standard-setting modeling, the agency recognizes that the reference baseline will necessarily reflect fewer EVs than will likely exist in the future fleet. However, the approach used to construct the reference baseline necessarily reflects the data constraints under which NHTSA was operating regarding manufacturer plans outside of voluntary alignment with ACC II. Regarding NRDC's comment, NHTSA did model several alternative ways that manufacturers could comply with the agency's standards, including as assessed against an alternative baseline that does not include state ZEV programs or voluntary deployment consistent with ACC II. The alternative baseline and range of sensitivity cases that NHTSA modeled, and results are discussed in more detail in Chapters 3 and 9 of the FRIA, and the No ZEV alternative baseline is discussed further below.

Lastly, regarding the reference baseline, the Joint NGOs commented that the methodology of holding the reference baseline constant for years prior to the start of the analysis year unrealistically restricted automakers from adopting fuel economy improving technologies they might otherwise adopt in response to increasingly stringent standards. The Joint NGOs stated that this modeling decision had a significant effect on the reference baseline, “particularly for the standard-setting runs where additional, economically efficient electric vehicle technologies cannot be deployed in the model year 2027-2032 period.” The Joint NGOs also stated that NHTSA did not explain this methodology or decision in any of the agency's rulemaking documents.

By way of additional background on this modeling approach: any fleet improvements obtained when evaluating the No-Action Alternative during model years 2022-2026 for the passenger car and light truck fleets, and during model years 2022-2029 for the HDPUV fleet will be carried over into the Action Alternatives for the same range of model years. Additionally, during those “reference baseline” set of years, any further fleet upgrades will not be performed under the Action Alternatives. For the Action Alternatives, technology evaluation and fleet improvements will then begin starting with the first standard-setting year, which is model year 2027 for passenger cars and light trucks, and model year 2030 for HDPUV. Doing so prevents the reference baseline years from being affected by standards defined under the Action Alternatives and ensures that the reference baseline years remain constant irrespective of the alternative being evaluated.

NHTSA believes that this approach captures the impact of new regulations more accurately, as compared to the previously established standards defined under the No-Action Alternative. More specifically, this better allows the agency to capture the costs and benefits of the range of standards being considered. If NHTSA allowed manufacturers to apply technology in advance of increasing standards in later model years, the costs and benefits of those improvements would be attributable to the reference baseline and not NHTSA's action. Moreover, this approach provides an additional level of certainty that the model is not selecting BEV technology in the reference baseline before the operative standards begin to take effect. Put another way, this requirement was intended to ensure that the model does not simulate manufacturers creating new BEVs prior to the standard-setting years in anticipation of the need to comply with the CAFE standards during those standard-setting years. It is exactly the situation that the Joint NGOs describe—that the model might apply BEV technology in the reference baseline but in response to the standards—that NHTSA seeks to avoid in order to fully comply with 49 U.S.C. 32902(h). In sum, not only does this approach allow NHTSA to better capture the costs and benefits of different levels of standards under consideration, but it ensures the modeling comports with all relevant statutory constraints.

2. Alternative Baseline/No-Action Alternative

In addition to the reference baseline for the passenger car and light truck fleet analysis, NHTSA considered an alternative baseline analysis. This alternative baseline analysis for the passenger car and light truck fleets was performed to provide a greater level of insight into the possibilities of a changing baseline landscape. The alternative baseline analysis is not meant to be a replacement for the reference analysis, but a secondary review of the NHTSA analysis with all of the assumptions from the reference baseline held (see Section IV.B.1 above), except for the assumption of compliance with CARB ZEV programs, and the voluntary deployment of electric vehicles consistent with ACC II. The alternative baseline does not assume manufacturers will comply with any of the California light duty ZEV programs or voluntarily deploy electric vehicles consistent with ACC II during any of the model years simulated in the analysis. Results for this alternative baseline are shown in Chapter 8.2.7 of the FRIA and discussed in more detail in Section VI.

3. Action Alternatives for Model Years 2027-2032 Passenger Cars and Light Trucks

In addition to the No-Action Alternatives, NHTSA has considered five “action” alternatives for passenger cars and light trucks, each of which is more stringent than the No-Action Alternative during the rulemaking time frame. These action alternatives are specified below and demonstrate different possible approaches to balancing the statutory factors applicable for passenger cars and light trucks. Section VI discusses in more detail how the different alternatives reflect different possible balancing approaches.

a. Alternative PC1LT3

Alternative PC1LT3 would increase CAFE stringency by 1 percent per year, year over year, for model years 2027-2032 passenger cars, and by 3 percent per year, year over year, for model years 2027-2032 light trucks.

These equations are represented graphically below:

Under this alternative, the MDPCS is as follows:

b. Alternative PC2LT002—Final Standards

Alternative PC2LT002 would increase CAFE stringency by 2 percent per year, year over year for model years 2027-2032 for passenger cars, and by 0 percent per year, year over year for model years 2027-2028 light trucks and then 2 percent per year, year over year for model years 2029-2032 for light trucks.

These equations are represented graphically below:

c. Alternative PC2LT4

Alternative PC2LT4 would increase CAFE stringency by 2 percent per year, year over year, for model years 2027-2032 for passenger cars, and by 4 percent per year, year over year, for model years 2027-2032 for light trucks.

These equations are represented graphically below:

Under this alternative, the MDPCS is as follows:

d. Alternative PC3LT5

Alternative PC3LT5 would increase CAFE stringency by 3 percent per year, year over year, for model years 2027-2032 for passenger cars, and by 5 percent per year, year over year, for model years 2027-2032 for light trucks.

These equations are represented graphically below:

Under this alternative, the MDPCS is as follows:

e. Alternative PC6LT8

Alternative PC6LT8 would increase CAFE stringency by 6 percent per year, year over year, for model years 2027-2032 for passenger cars, and by 8 percent per year, year over year, for model years 2027-2032 for light trucks.

These equations are represented graphically below:

Under this alternative, the MDPCS is as follows:

f. Other Alternatives Suggested by Commenters for Passenger Car and LT CAFE Standards

Commenters also suggested a variety of other regulatory alternatives for NHTSA to analyze for the final rule.

Rivian commented that NHTSA should increase stringency for light trucks relative to passenger cars by an even greater degree than the proposal, such as “stringency combinations in which standards would increase by 2 percent annually for passenger cars but 5 to 8 percent annually for light trucks.” Rivian argued that this was appropriate given “that more stringent light truck targets perform well from a cost-benefit perspective.” Rivian also suggested that NHTSA evaluate an alternative in which only light truck standards were increased.

IPI commented that NHTSA should (1) evaluate an alternative which expressly maximizes net benefits (suggesting PC2LT8, specifically), and (2) “assess a broader range of alternatives that decouple increases from light trucks from those for passenger cars and that impose non-linear increases, which could further maximize net benefits.”

NHTSA appreciates Rivian's comment; however, we have an obligation to set maximum feasible CAFE standards separately for passenger cars and light trucks (see 49 U.S.C. 32902). We would not be in compliance with our statutory authority if we failed to increase passenger car standards despite concluding that Alternative PC2LT002 is feasible for the industry. Establishing maximum feasible standards involves balancing several factors, which means that some factors, like net benefits, may not reach their maximum level. As previously mentioned, NHTSA is statutorily required to set independent standards for passenger cars and light trucks. As such, NHTSA's preferred alternative contains passenger car and light truck standards that are already “decoupled.” Also, the stringency for the light truck fleet is non-linear where it increases by 0 percent per year, year over year for MYs 2027-2028 light trucks and then 2 percent per year, year over year for model years 2029-2031.

4. Action Alternatives for Model Years 2030-2035 Heavy-Duty Pickups and Vans

In addition to the No-Action Alternative, NHTSA has considered four action alternatives for HDPUVs. Each of the Action Alternatives, described below, would establish increases in stringency over the No-Action Alternative from model year 2030 through model year 2035. In the NPRM, NHTSA also sought comment on a scenario in which the Action Alternatives would extend only through model year 2032. Ford supported NHTSA ending its HDPUV standards in model year 2032 as more harmonized with EPA's proposed standards, and as aligning “better . . . with the Inflation Reduction Act's ZEV credits, scheduled to end by 2032.” Ford suggested re-evaluating the standards for model years 2033-2035 at a later time. Wisconsin DNR, in contrast, stated that “given the different statutory authorities under which EPA and NHTSA promulgate vehicle standards, it is appropriate for NHTSA to set standards for the model year ranges it has proposed, rather than extending these standards only through 2032 (which would align with the final model year of EPA's proposed multipollutant standards).”

We believe that setting HDPUV standards through model year 2035 is appropriate based on our review of the baseline fleet and its capability, in addition to the range of technologies that are available for adoption in the rulemaking timeframe. In addition to the advanced credit multiplier that is available for manufacturers until model year 2027, the current standards do not require significant improvements from model year 2027 through model year 2029. Accordingly, our analysis for model years 2030-2035 shows the potential for high technology uptake; this can be seen in detail in RIA Chapter 8. We proposed 10 percent year over year increases and now we are finalizing 8 percent year over year increases. This means that over the six-year period where these standards are in effect, the stringency of our standards almost matches the stringency of the EPA standards in model year 2032. Our regulatory model years are different due to our statutory requirements, however, as our statutory lead time requirements prevented us from harmonizing with EPA directly on the model year 2027-2029 standards. For a more detailed discussion on the lead time for HDPUVs, see Section VI.A.1.b. Section VI also discusses in more detail how the different alternatives reflect different possible balancing approaches for setting HDPUV standards. HDPUV action alternatives are specified below.

a. Alternative HDPUV4

Alternative HDPUV4 would increase HDPUV standard stringency by 4 percent per year for model years 2030-2035 for HDPUVs. NHTSA included this alternative in order to evaluate a possible balancing of statutory factors in which cost-effectiveness outweighed all other factors. The four-wheel drive coefficient is maintained at 500 (coefficient `a') and the weighting multiplier coefficient is maintained at 0.75 (coefficient `b').

These equations are represented graphically below:

b. Alternative HDPUV108—Final Standards

Alternative HDPUV108 would increase HDPUV standard stringency by 10 percent per year, year over year for model years 2030-2032, and by 8 percent per year, year over year for model years 2033-2035 for HDPUVs. The four-wheel drive coefficient is maintained at 500 (coefficient `a') and the weighting multiplier coefficient is maintained at 0.75 (coefficient `b').

These equations are represented graphically below:

c. Alternative HDPUV10

Alternative HDPUV10 would increase HDPUV standard stringency by 10 percent per year for model years 2030-2035 for HDPUVs. The four-wheel drive coefficient is maintained at 500 (coefficient `a') and the weighting multiplier coefficient is maintained at 0.75 (coefficient `b').

These equations are represented graphically below:

d. Alternative HDPUV14

Alternative HDPUV14 would increase HDPUV standard stringency by 14 percent per year for model years 2030-2035 for HDPUVs. The four-wheel drive coefficient is maintained at 500 (coefficient `a') and the weighting multiplier coefficient is maintained at 0.75 (coefficient `b').

These equations are represented graphically below:

V. Effects of the Regulatory Alternatives

A. Effects on Vehicle Manufacturers

1. Passenger Cars and Light Trucks

Each regulatory alternative considered in this final rule, aside from the No-Action Alternative, would increase the stringency of both passenger car and light truck CAFE standards during model years 2027-2031 (with model year 2032 being an augural standard). To estimate the potential effects of each of these alternatives, NHTSA has, as with all recent rulemakings, assumed that standards would continue unchanged after the last model year to be covered by CAFE targets (in this case model year 2031 for the primary analysis and 2032 for the augural standards). NHTSA recognizes that it is possible that the size and composition of the fleet ( i.e., in terms of distribution across the range of vehicle footprints) could change over time, affecting the average fuel economy requirements under both the passenger car and light truck standards, and for the overall fleet. If fleet changes ultimately differ from NHTSA's projections, average requirements would differ from NHTSA's projections.

Following are the estimated required average fuel economy values for the passenger car, light truck, and total fleets for each action alternative that NHTSA considered alongside values for the No-Action Alternative. (As a reminder, all projected effects presented use the reference baseline unless otherwise stated.)

Manufacturers do not always comply exactly with each CAFE standard in each model year. To date, some manufacturers have tended to exceed at least one requirement. Many manufacturers in practice make use of EPCA's provisions allowing CAFE compliance credits to be applied when a fleet's CAFE level falls short of the corresponding requirement in a given model year. Some manufacturers have paid civil penalties ( i.e., fines) required under EPCA when a fleet falls short of a standard in a given model year and the manufacturer lacks compliance credits sufficient to address the compliance shortfall. As discussed in the accompanying FRIA and TSD, NHTSA simulates manufacturers' responses to each alternative given a wide range of input estimates ( e.g., technology cost and efficacy, fuel prices), and, per EPCA requirements, setting aside the potential that any manufacturer would respond to CAFE standards in model years 2027-2031 by applying CAFE compliance credits or considering the fuel economy attributable to alternative fuel sources. Many of these inputs are subject to uncertainty, and, in any event, as in all CAFE rulemakings, NHTSA's analysis simply illustrates one set of ways manufacturers could potentially respond to each regulatory alternative. The tables below show the estimated achieved fuel economy produced by the CAFE Model for each regulatory alternative.

While these increases in estimated fuel economy levels are partially attributable to changes in the composition of the fleet as simulated by the CAFE Model ( i.e., the relative shares of passenger cars and light trucks), they result almost entirely from the projected application of fuel-saving technology. Manufacturers' actual responses will almost assuredly differ from NHTSA's simulations, and therefore the achieved compliance levels will differ from these tables.

The SHEV share of the light-duty fleet initially ( i.e., in model year 2022) is relatively low, but increases to approximately 23 to 27 percent by the beginning of the final rule's regulatory period (MY2027). Across action alternatives, SHEV penetration rates increase as alternatives become more stringent, in both the passenger car and light truck fleets. SHEVs are estimated to make up a larger portion of light truck fleet than passenger car fleet across model years 2027-2031. While their market shares do not increase to the levels of SHEVs, PHEVs make up between 7 to 8 percent of the estimated light truck fleet across the alternatives by the end of the regulatory period. In the passenger car fleet, PHEV penetration stays under 2 percent for all alternatives and all model years. Variation in penetration rates across alternatives generally results from how many vehicles or models require additional technology to become compliant, e.g. one technology pathway is the most cost-effective pathway if a manufacturer is just shy of their fuel economy target, but becomes ineffective if there's a larger gap which may necessitate pursuing broader changes in powertrain across the manufacturers' fleet. For example, Honda is projected to redesign several of its models from MHEV to PHEV in 2027. This accounts for the slightly increased PHEV

penetration rate in PC2LT002. For more detail on the technology application by regulatory fleet, see FRIA Chapter 8.2.2.1.

Due to the statutory constraints imposed on the analysis by EPCA that exclude consideration of AFVs, BEVs are not a compliance option through model year 2031. Similarly, PHEVs can be introduced by the CAFE Model, but only their charge-sustaining fuel economy value is considered during standard setting years (as opposed to their charge-depleting fuel economy value, which is used in all other years). As seen in Table V-9 and Table V-10, BEV penetration increases across model years in the No-Action Alternative. During the standard setting years, BEVs are only added to account for manufacturers' expected response to state ZEV programs and additional electric vehicles that manufacturers have committed to deploy consistent with ACC II, regardless of whether it becomes legally binding. In model years outside of the standard setting restrictions, BEVs may be added if they are cost-effective to produce for reasons other than the CAFE standards The action alternatives show nearly the same BEV penetration rates as the No-Action Alternative during the standard setting years, although in some cases there is a slight deviation despite no new BEV models entering the fleet, due to rounding in some model years where fewer vehicles are being sold in response to the standards and altering fleet shares.

The FRIA provides a longer summary of NHTSA's estimates of manufacturers' potential application of fuel-saving technologies (including other types of technologies, such as advanced transmissions, aerodynamic improvements, and reduced vehicle mass) in response to each regulatory alternative. Appendices I and II of the accompanying FRIA provide more detailed and comprehensive results, and the underlying CAFE Model output files provide all the information used to construct these estimates, including the specific combination of technologies estimated to be applied to every vehicle model/configuration in each of model years 2022-2050.

NHTSA's analysis shows manufacturers' regulatory costs for compliance with the CAFE standards, combined with existing EPA GHG standards, state ZEV programs, and voluntary deployment of electric vehicles consistent with ACC II unsurprisingly increasing more under the more stringent alternatives as more fuel-saving technologies would be required. As summarized in Table V-11, NHTSA estimates manufacturers' cumulative regulatory costs across model years 2027-2031 could total $148b under the No-Action Alternative, and an additional $18b, $21.8b, $33b, $41.4b, and $55.5b under alternatives PC2LT002, PC1LT3, PC2LT4, PC3LT5, and PC6LT8, respectively, when accounting for fuel-saving technologies added under the simulation for each regulatory alternative (including AC improvements and other off-cycle technologies), and also accounting for CAFE civil penalties that NHTSA estimates some manufacturers could elect to pay rather than achieving full compliance with the CAFE targets in some model years in some fleets. The table below shows how these costs are estimated to vary among manufacturers, accounting for differences in the quantities of vehicles produced for sale in the U.S. Differences in technology application and compliance pathways play a significant role in determining variation across aggregate manufacturer costs, and technology costs for each model year are defined on an incremental basis, with costs equal to the relevant technology applied minus the costs of the initial technology state in a reference fleet. Appendices I and II of the accompanying FRIA present results separately for each manufacturer's passenger car and light truck fleets in each model year under each regulatory alternative, and the underlying CAFE Model output files also show results specific to manufacturers' domestic and imported car fleets.

As discussed in the TSD, these estimates reflect technology cost inputs that, in turn, reflect a “markup” factor that includes manufacturers' profits. In other words, if costs to manufacturers are reflected in vehicle price increases, NHTSA estimates that the average costs to new vehicle purchasers could increase through model year 2031 as summarized in Table V-12 and Table V-13. Table V-14 shows how these costs could vary among manufacturers, suggesting that price differences between manufacturers could increase as the stringency of standards increases. See Chapter 8.2.2 of the FRIA for more details of the effects on vehicle manufacturers, including compliance and regulatory costs.

Fuel savings and regulatory costs act as competing forces on new vehicle sales. All else being equal, as fuel savings increase, the CAFE Model projects higher new vehicle sales, but as regulatory costs increase, the CAFE Model projects lower new vehicle sales. Both fuel savings and regulatory costs increase with stringency. NHTSA observed that on net that regulatory costs were increasing faster than the first 30 months of fuel savings in the CAFE Model projections and as such, sales decreased in higher stringency alternatives. The magnitude of these fuel savings and vehicle price increases depends on manufacturer compliance decisions, especially technology application. In the event that manufacturers select technologies with lower prices and/or higher fuel economy improvements, vehicle sales effects could differ. TSD Chapter 4.2.1.2 discusses NHTSA's approach to estimating new vehicle sales, including NHTSA's estimate that new vehicle sales could recover from 2020's aberrantly low levels. Figure V-1 shows the estimated annual light-duty industry sales by regulatory alternative. For all scenarios, sales stay constant relative to the No-Action scenario through model year 2026, after which the model begins applying technology in response to the action alternatives. Excluding the most stringent case, light-duty vehicle sales differ from the No-Action Alternative by approximately 1 percent or less through model year 2050, and PC6LT8 sales differ from the No-Action Alternative by less than 2.5 percent through model year 2050.

These slight reductions in new vehicle sales tend to reduce projected automobile industry labor projections by small margins. NHTSA estimates that the cost increases could reflect an underlying increase in employment to produce additional fuel-saving technology, such that automobile industry labor could remain relatively similar under each of the five regulatory alternatives.

The accompanying TSD Chapter 6.2.5 discusses NHTSA's approach to estimating automobile industry employment, and the accompanying FRIA Chapter 8.2 (and its Appendices I and II) and CAFE Model output files provide more detailed results of NHTSA's light-duty analysis.

We also include in the analysis a No ZEV alternative baseline, wherein some sales volumes do not in MYs 2023 and beyond turn into ZEVs in accordance with OEM commitments to deploy additional electric vehicles consistent with ACC II, regardless of whether it becomes legally binding. The No ZEV alternative baseline still includes BEVs and PHEVs, but they are those that were already observed in the MY 2022 analysis fleet, as well as any made by the model outside of standard setting years for LD BEVs (or in all years, in the case of PHEVs and HDPUV BEVs). Across the entire light-duty fleet, the technology penetration rates differ mainly from 2027 onwards. In the reference baseline, BEVs make up approximately 28 percent of the total light-duty fleet by model year 2031; they make up only 19 percent of the total light-duty fleet by 2031 in the No ZEV alternative baseline.

PHEVs have virtually the same tech penetration in the reference baseline as in the no ZEV alternative baseline, as the CAFE Model does not build PHEVs for ZEV program compliance (only counts PHEVs built for other reasons towards ZEV program compliance) or deploy them based on OEM commitments to deploy electric vehicles consistent with ACC II. PHEVs increase only from 2 percent in the reference case to 3 percent in the No ZEV alternative baseline by model year 2031. Strong hybrids have a slightly higher tech penetration rate under the reference baseline than in the No ZEV case in model years between 2027 and 2031 at 27 percent compared to 23 percent in the reference baseline in model year 2031.

2. Heavy-Duty Pickups and Vans

Each of the regulatory alternatives considered represents an increase in HDPUV fuel efficiency standards for model years 2030-2035 relative to the existing standards set in 2016, with increases in efficiency each year through model year 2035. Unlike the light-duty CAFE program, NHTSA may consider AFVs when setting maximum feasible average standards for HDPUVs. Additionally, for purposes of calculating average fuel efficiency for HDPUVs, NHTSA considers EVs, fuel cell vehicles, and the proportion of electric operation of EVs and PHEVs that is derived from electricity that is generated from sources that are not onboard the vehicle to have a fuel efficiency value of 0 gallons/mile.

NHTSA recognizes that it is possible that the size and composition of the fleet ( i.e., in terms of vehicle attributes that impact calculation of standards for averaging sets) could change over time, which would affect the currently-estimated average fuel efficiency requirements. If fleet changes ultimately differ from NHTSA's projections, average requirements could, therefore, also differ from NHTSA's projections. The table below includes the estimated required average fuel efficiency values for the HDPUV fleet in each of the regulatory alternatives considered in this final rule.

As with the light-duty program, manufacturers do not always comply exactly with each fuel efficiency standard in each model year. Manufacturers may bank credits from overcompliance in one year that may be used to cover shortfalls in up to five future model years. Manufacturers may also carry forward credit deficits for up to three model years. If a manufacturer is still unable to address the shortfall, NHTSA may assess civil penalties. As discussed in the accompanying FRIA and TSD, NHTSA simulates manufacturers' responses to each alternative given a wide range of input estimates ( e.g., technology cost and effectiveness, fuel prices, electrification technologies). For this final rule, NHTSA estimates that manufacturers' responses to standards defined in each alternative could lead average fuel efficiency levels to improve through model year 2035, as shown in the following tables.

Table V-16 displays the projected achieved FE levels for the HDPUV fleet through model year 2035. Estimates of achieved levels are very similar between the No-Action Alternative and the least stringent action alternative, with even the most stringent action alternative differing by less than 0.8 gallons/100 miles from the No-Action Alternative. The narrow band of estimated average achieved levels in Table V-16 is primarily due to several factors. Relative to the LD fleet, the HDPUV fleet (i) represents a smaller number of vehicles, (ii) includes fewer manufacturers, and (iii) is composed of a smaller number of manufacturer product lines. Technology choices for an individual manufacturer or individual product line can therefore have a large effect on fleet-wide average fuel efficiency. Second, Table V-17 shows that in the No-Action Alternative a substantial portion of the fleet converts to an electrified powertrain ( e.g., SHEV, PHEV, BEV) between model year 2022 and model year 2030. This reduces the availability of, and need for, additional fuel efficiency improvement to meet more stringent standards.

In line with the technology application trends above, regulatory costs do not differ by large amounts between the No-Action Alternative and the action alternatives. The largest differences in regulatory costs occur in the HDPUV14 alternative and are also concentrated in a few manufacturers ( e.g., Ford, GM), where the compliance modeling projects increases in PHEV and advanced engine technologies. For example, GM is projected to increase its turbo parallel engine technology penetration by 2038, which is modeled as a lower cost than the superseded advanced diesel engine technology in the reference baseline, contributing to the negative cost in the No-Action Alternative. See RIA Chapter 8.3.2 for more detail on the manufacturer regulatory cost by action alternative.

On a per-vehicle basis, costs by 2033 increase progressively with stringency. Average per-vehicle costs are estimated to decrease slightly for alternatives HDPUV108 and HPUV10 relative to the No-Action Alternative for model year 2030-2032. Cost reductions of technology applied in these years, combined with shifts altering the combination of technologies to comply with different stringencies, result in negative regulatory costs relative to the No-Action Alternative. Specifically, differences in the quantity and type of technology applications in the compliance pathways contribute to the cost variation across regulatory alternatives. Overall, the two least stringent alternatives represent less than a 12 percent difference in average per-vehicle cost compared to the No-Action Alternative. FRIA Chapter 8.3.2.1 provides more information about the technology penetration changes and the subsequent costs.

The sales and labor markets are estimated to have relatively little variation in impacts across the No-Action Alternative and action alternatives. The increase in sales in the No-Action Alternative carries over to each of the action alternatives as well. The vehicle-level cost increases noted above in Table V-19 produce very small declines in overall sales. With the exception of HDPUV14, the change in sales across alternatives stays within about a 0.21 percent change relative to the No-Action Alternative, and HDPUV14 stays within a 0.6 percent change relative to the No-Action Alternative.

These minimal sales declines and limited additional technology application produce small decreases in labor utilization, as the sales effect ultimately outweighs job gains due to development and application of advanced technology. In aggregate, the alternatives represent less than half of a percentage point deviation from the No-Action Alternative.

The accompanying TSD Chapter 6.2.5 discusses NHTSA's approach to estimating automobile industry employment, and the accompanying FRIA Chapter 8.3.2.3 (and its Appendix III) and CAFE Model output files provide more detailed results of NHTSA's HDPUV analysis.

B. Effects on Society

NHTSA accounts for the effects of the standards on society using a benefit-cost framework. The categories considered include private costs borne by manufacturers and passed on to consumers, social costs, which include Government costs and externalities pertaining to emissions, congestion, noise, energy security, and safety, and all the benefits resulting from related categories in the form of savings, however they may occur across the presented alternatives. In this accounting framework, the CAFE Model records costs and benefits for vehicles in the fleet throughout the lifetime of a particular model year and also allows for the accounting of costs and benefits by calendar years. Examining program effects through this lens illustrates the temporal differences in major cost and benefit components and allows us to examine costs and benefits for only those vehicles that are directly regulated by the standards. In the HDPUV FE analysis, where the standard would continue until otherwise amended, we report only the costs and benefits across calendar years.

1. Passenger Cars and Light Trucks

We split effects on society into private costs, social costs, private benefits, and external benefits. Table V-21 and Table V-22 describe the costs and benefits of increasing CAFE standards in each alternative, as well as the party to which they accrue. Manufacturers are directly regulated under the program and incur additional production costs when they apply technology to their vehicle offerings in order to improve their fuel economy. We assume that those costs are fully passed through to new car and truck buyers in the form of higher prices. We also assume that any civil penalties paid by manufacturers for failing to comply with their CAFE standards are passed through to new car and truck buyers and are included in the sales price. However, those civil penalties are paid to the U.S. Treasury, where they currently fund the general business of government. As such, they are a transfer from new vehicle buyers to all U.S. citizens, who then benefit from the additional Federal revenue. While they are calculated in the analysis, and do influence consumer decisions in the marketplace, they do not directly contribute to the calculation of net benefits (and are omitted from the tables below).

While incremental maintenance and repair costs and benefits would accrue to buyers of new cars and trucks affected by more stringent CAFE standards, we do not carry these impacts in the analysis. They are difficult to estimate but represent real costs (and potential benefits in the case of AFVs that require less frequent maintenance events). They may be included in future analyses as data become available to evaluate lifetime maintenance impacts. This analysis assumes that drivers of new vehicles internalize 90 percent of the risk associated with increased exposure to crashes when they engage in additional travel (as a consequence of the rebound effect).

Private benefits are dominated by the value of fuel savings, which accrue to new car and truck buyers at retail fuel prices (inclusive of Federal and state taxes). In addition to saving money on fuel purchases, new vehicle buyers also benefit from the increased mobility that results from a lower cost of driving their vehicle (higher fuel economy reduces the per-mile cost of travel) and fewer refueling events. The additional travel occurs as drivers take advantage of lower operating costs to increase mobility, and this generates benefits to those drivers—equivalent to the cost of operating their vehicles to travel those miles, the consumer surplus, and the offsetting benefit that represents 90 percent of the additional safety risk from travel.

In addition to private benefits and costs—those borne by manufacturers, buyers, and owners of cars and light trucks—there are other benefits and costs from increasing CAFE standards that are borne more broadly throughout the economy or society, which NHTSA refers to as social costs. The additional driving that occurs as new vehicle buyers take advantage of lower per-mile fuel costs is a benefit to those drivers, but the congestion (and road noise) created by the additional travel also imposes a small additional social cost to all road users. We also include transfers from one party to another other than those directly incurred by manufacturers or new vehicle buyers with social costs, the largest of which is the loss in fuel tax revenue that occurs as a result of falling fuel consumption. Buyers of new cars and light trucks produced in model years subject to increasing CAFE standards save on fuel purchases that include Federal, state, and sometimes local taxes, so revenues from these taxes decline; because that revenue funds maintenance of roads and bridges as well as other government activities, the loss in fuel tax revenue represents a social cost, but is offset by the benefits gained by drivers who spend less at the pump.

Among the purely external benefits created when CAFE standards are increased, the largest is the reduction in damages resulting from GHG emissions. Table V-20 shows the different social cost results that correspond to each GHG discount rate. The associated benefits related to reduced health damages from criteria pollutants and the benefit of improved energy security are both significantly smaller than the associated change in GHG damages across alternatives. As the tables also illustrate, the majority of costs are private costs that accrue to buyers of new cars and trucks, but the plurality of benefits stem from external welfare changes that affect society more generally. These external benefits are driven mainly by the benefits from reducing GHGs.

The tables show that the social and SC-GHG discount rates have a significant impact on the estimated benefits in terms of magnitudes. Net social benefits are positive for all alternatives at both the 3 percent and 7 percent social discount rates but have higher magnitudes under the lower SC-GHG discount rates. Net benefits are higher when assessed at a 3 percent social discount rate since the largest benefit—fuel savings—accrues over a prolonged period, while the largest cost—technology costs—accrue predominantly in earlier years. Totals in the following table may not sum perfectly due to rounding.

Our analysis also includes a No ZEV alternative baseline for light-duty, and the CAFE Model outputs results for all scenarios relative to that baseline as well. Net benefits in the preferred alternative increase when viewing the analysis from the perspective of the No ZEV alternative baseline. Using the model year perspective, the SC-GHG DR of 2% and a social discount rate of 3%, net benefits in the preferred alternative of the No ZEV alternative baseline are 44.9 billion, compared to the preferred alternative's net benefits relative to the reference baseline (35.2 billion).

2. Heavy-Duty Pickups and Vans

Our categorizations of benefits and costs in the HDPUV space mirrors the approach taken above for light-duty passenger trucks and vans. Table V-22 describes the costs and benefits of increasing standards in each alternative, as well as the party to which they accrue. Manufacturers are directly regulated under the program and incur additional production costs when they apply technology to their vehicle offerings in order to improve their fuel efficiency. We assume that those costs are fully passed through to new HDPUV buyers, in the form of higher prices.

One key difference between the light-duty and HDPUV analysis is how the agency approaches VMT. As explained in more detail in III.E.3 and TSD Chapter 4.3, the agency does not constrain non-rebound VMT. As a result, decreasing sales in the HDPUV fleet will lower the amount of total VMT, while the rebound effect will cause those vehicles that are improved and sold, to be driven more. On net, the CAFE Model shows that the amount of VMT forgone from lower sales slightly outweighs the amount of VMT gained through rebound driving, and as a result some of the externalities from driving, such as safety costs and congestion, appear as a cost reduction relative to the No-Action Alternative.

The choice of GHG discount rate also affects the resulting benefits and costs. As the tables show, net social benefits are positive for all alternatives, and are greatest when the SC-GHG discount rate of 1.5 percent is used. Totals in the following table may not sum perfectly due to rounding.

C. Physical and Environmental Effects

1. Passenger Cars and Light Trucks

NHTSA estimates various physical and environmental effects associated with the standards. These include quantities of fuel and electricity consumed, GHGs and criteria pollutants reduced, and health and safety impacts. Table V-23 shows the cumulative impacts grouped by decade, including the on-road fleet sizes, VMT, fuel consumption, and CO₂ emissions, across alternatives. The size of the on-road fleet increases in later decades regardless of alternative, but the greatest on-road fleet size projection is seen in the reference baseline, with fleet sizes declining as the alternatives become increasingly more stringent. This is attributable to the reduction in sales caused by increased regulatory costs, which overtime decreases the existing vehicle stock, and therefore the size of the overall fleet.

VMT increases occur in the two later decades, with the highest miles occurring from 2041-2050. Fuel consumption (measured in gallons or gasoline gallon equivalents) declines across both decades and alternatives as the alternatives become more stringent, as do GHG emissions.

From a calendar year perspective, NHTSA's analysis estimates total annual consumption of fuel by the entire on-road fleet from calendar year 2022 through calendar year 2050. On this basis, gasoline and electricity consumption by the U.S. light-duty fleet evolves as shown in Figure IV-5 and Figure IV-6, each of which shows projections for the No-Action Alternative (No-Action Alternative, i.e., the reference baseline), Alternative PC2LT002, Alternative PC1LT3, Alternative PC2LT4, Alternative PC3LT5, and Alternative PC6LT8. Gasoline consumption decreases over time, with the largest decreases occurring in more stringent alternatives. Electricity consumption increases over time, with the same pattern of Alternative PC6LT8 experiencing the highest magnitude of change.

NHTSA estimates the GHGs attributable to the light-duty on-road fleet, from both vehicles and upstream energy sector processes ( e.g., petroleum refining, fuel transportation and distribution, electricity generation). Figure IV-7, Figure IV-8, and Figure IV-9 present NHTSA's estimate of how emissions from these three GHGs across all fuel types could evolve over the years. Note that these graphs include emissions from both downstream (powertrain and BTW) and upstream processes. All three GHG emissions follow similar trends of decline in the years between 2022-2050. Note that CO₂ emissions are expressed in units of million metric tons (mmt) while emissions from other pollutants are expressed in metric tons.

The figures presented here are not the only estimates NHTSA calculates regarding projected GHG emissions in future years. The accompanying EIS uses an “unconstrained” analysis as opposed to the “standard setting” analysis presented in this final rule. For more information regarding projected GHG emissions, as well as model-based estimates of corresponding impacts on several measures of global climate change, see the EIS.

NHTSA also estimates criteria pollutant emissions resulting from downstream (powertrain and BTW) and upstream processes attributable to the light-duty on-road fleet. Since the NPRM, NHTSA has adopted the NREL 2022 grid mix forecast which projects significant reductions in criteria emission rates from upstream electricity production. This results in further emission reductions across alternatives as EVs in the reference baseline induce marginally less emissions relative to the NPRM. This decrease in criteria pollutant emissions in turn leads to a decrease in adverse health outcomes described in later sections. Under each regulatory alternative, NHTSA projects a dramatic decline in annual emissions of NO_X, and PM_2.5 attributable to the light-duty on-road fleet between 2022 and 2050. As exemplified in Figure V-10, NOx emissions in any given year could be very nearly the same under each regulatory alternative.

On the other hand, as discussed in the FRIA Chapter 8.2 and Chapter 4 of the EIS accompanying this document, NHTSA projects that annual SO₂ emissions attributable to the LD on-road fleet could increase by 2050, after significant fluctuation, in all of the alternatives, including the reference baseline, due to greater use of electricity for PHEVs and BEVs (See Figure IV-6). Differences between the action alternatives are modest.

Health impacts quantified by the CAFE Model include various instances of hospital visits due to respiratory problems, minor restricted activity days, non-fatal heart attacks, acute bronchitis, premature mortality, and other effects of criteria pollutant emissions on health. Table V-24 shows the split in select health impacts relative to the No-Action Alternative, across all action alternatives. The magnitude of the differences relates directly to the changes in tons of criteria pollutants emitted. Magnitudes differ across health impact types because of variation in the reference baseline totals; for example, the total Minor Restricted Activity Days are much higher than the Respiratory Hospital Admissions. See Chapter 5.4 of the TSD for information regarding how the CAFE Model calculates these health impacts.

Lastly, NHTSA also quantifies safety impacts in its analysis. These include estimated counts of fatalities, non-fatal injuries, and property damage crashes occurring over the lifetimes of the LD on-road vehicles considered in the analysis. The following table shows the changes in these counts projected in action alternatives relative to the reference baseline.

Generally, increasing fuel economy stringency leads to more adverse safety outcomes from increased rebound VMT (motorists choosing to drive more as driving becomes cheaper), and the reduction in scrappage causing older vehicles with less safety features to remain in the fleet longer. The impacts of mass reduction are nonlinear and depend on the specific fleet receiving those reductions, with mass reduction to PCs generally causing an increase in adverse safety outcomes and mass reductions for LTs generally causing a decrease in adverse safety outcomes; this explains the difference in the impacts of mass reduction for Alternative PC6LT8, as this alternative sees the largest transition from LTs to PCs and has PCs receiving the most mass reductions. NHTSA notes that none of these safety outcomes due to mass reduction can be statistically distinguished from zero. Chapter 7.1.5 of the FRIA accompanying this document contains an in-depth discussion on the effects of the various alternatives on these safety measures, and Chapter 7 of the TSD contains information regarding the construction of the safety estimates.

We also analyze physical and environmental effects relative to the No ZEV alternative baseline. In the model year perspective (model years through 2031), in the preferred alternative (PC2LT002) relative to the No ZEV alternative baseline, CO₂ emission reductions are 1,207 MMT, compared to the reduction in CO₂ emissions in the preferred alternative relative to the reference baseline (659 MMT).

2. Heavy-Duty Pickups and Vans

NHTSA estimates the same physical and environmental effects for HDPUVs as it does for LDVs, including: quantities of fuel and electricity consumption; tons of GHG emissions and criteria pollutants reduced; and health and safety impacts. Table V-26 shows the cumulative impacts grouped by decade, including the on-road fleet sizes, VMT, fuel consumption, and CO₂ emissions, across alternatives. The size of the on-road fleet increases in later decades regardless of the alternative, but the greatest on-road fleet size projection is seen in the reference baseline. Most differences between the alternatives are not visible in the Table V-26 due to rounding.

VMT increases occur in the later two decades, with the highest numbers occurring from 2041-2050. Across alternatives, the VMT increases remain around approximately the same magnitude. Fuel consumption (measured in gallons or gasoline gallon equivalents) declines across decades, as do GHG emissions. Differences between the alternatives are minor but fuel consumption and GHG emissions also decrease as alternatives become more stringent. As discussed in the previous section, since the agency does not constrain VMT for HDPUVs, alternatives with fewer vehicles see a corresponding decrease in VMT.

Figure V-13 and Figure V-14 show the estimates of gasoline and electricity consumption of the on-road HDPUV fleet for all fuel types over time on a calendar year basis, from 2022-2050. The four action alternatives, HDPUV4, HDPUV108, HDPUV10, and HDPUV14, are compared to the reference baseline changes over time.

Gasoline consumption decreases over time, with the largest decreases occurring in more stringent alternatives. Electricity consumption increases over time, with the same pattern of Alternative HDPUV14 experiencing the highest magnitude of change. In both charts, the differences in magnitudes across alternatives do not vary drastically.

NHTSA estimates the GHGs attributable to the HDPUV on-road fleet, from both downstream and upstream energy sector processes ( e.g., petroleum refining, fuel transportation and distribution, electricity generation). These estimates mirror those discussed in the light-duty section above. Figure IV15, Figure IV16, and Figure IV17 present NHTSA's estimate of how emissions from these three GHGs could evolve over the years (CY 2022-2050). Emissions from all three GHG types tracked follow similar trends of decline in the years between 2022-2050. Note that these graphs include emissions from both vehicle and upstream processes and scales vary by figure (CO₂ emissions are expressed in units of million metric tons (mmt) while emissions from other pollutants are expressed in metric tons). NHTSA's calculation of N₂ O emissions has changed since the NPRM resulting in increased emission rates for diesel vehicles, which comprise a significant portion of the HDPUV fleet.

For more information regarding projected GHG emissions, as well as model-based estimates of corresponding impacts on several measures of global climate change, see the EIS.

NHTSA also estimates criteria pollutant emissions resulting from vehicle and upstream processes attributable to the HDPUV on-road fleet. Under each regulatory alternative, NHTSA projects a significant decline in annual emissions of NO_X, and PM_2.5 attributable to the HDPUV on-road fleet between 2022 and 2050. As exemplified in Figure IV-18, the magnitude of emissions in any given year could be very similar under each regulatory alternative.

On the other hand, as discussed in the FRIA Chapter 8.3 and the EIS, NHTSA projects that annual SO₂ emissions attributable to the HDPUV on-road fleet could increase modestly under the action alternatives, because, as discussed above, NHTSA projects that each of the action alternatives could lead to greater use of electricity (for PHEVs and BEVs) in later calendar years.

Health impacts quantified by the CAFE Model include various instances of hospital visits due to respiratory problems, minor restricted activity days, non-fatal heart attacks, acute bronchitis, premature mortality, and other effects of criteria pollutant emissions on health. Table V-27 shows select health impacts relative to the baseline, across all action alternatives. The magnitude of the differences relates directly to the changes in tons of criteria pollutants emitted. The magnitudes differ across health impact types because of variation in the totals; for example, the total Minor Restricted Activity Days are much higher than the Respiratory Hospital Admissions. See Chapter 5.4 of the TSD for information regarding how the CAFE Model calculates these health impacts.

Lastly, NHTSA also quantifies safety impacts in its analysis. These include estimated counts of fatalities, non-fatal injuries, and property damage crashes occurring over the lifetimes of the HD on-road vehicles considered in the analysis. The following table shows projections of these counts in action alternatives relative to the baseline. As noted earlier, the safety impacts for HDPUV are a result of changes in aggregate VMT.

Chapter 7.1.5 of the FRIA accompanying this document contains an in-depth discussion on the effects of the various alternatives on these safety measures, and TSD Chapter 7 contains information regarding the construction of the safety estimates.

D. Sensitivity Analysis, Including Alternative Baseline

The analysis conducted to support this rulemaking consists of data, estimates, and assumptions, all applied within an analytical framework, the CAFE Model. Just as with all past CAFE and HDPUV rulemakings, NHTSA recognizes that many analytical inputs are uncertain, and some inputs are very uncertain. Of those uncertain inputs, some are likely to exert considerable influence over specific types of estimated impacts, and some are likely to do so for the bulk of the analysis. Yet making assumptions in the face of that uncertainty is necessary when analyzing possible future events ( e.g., consumer and industry responses to fuel economy/efficiency regulation). In other cases, we made assumptions in how we modeled the effects of other existing regulations that affected the costs and benefits of the action alternatives ( e.g., state ZEV programs were included in the No-Action Alternative). To better understand the effect that these assumptions have on the analytical findings, we conducted additional model runs with alternative assumptions. These additional runs were specified in an effort to explore a range of potential inputs and the sensitivity of estimated impacts to changes in these model inputs. Sensitivity cases and the alternative baseline in this analysis span assumptions related to technology applicability and cost, economic conditions, consumer preferences, externality values, and safety assumptions, among others. A sensitivity analysis can identify two critical pieces of information: how big of an influence does each parameter exert on the analysis, and how sensitive are the model results to that assumption?

That said, influence is different from likelihood. NHTSA does not mean to suggest that any one of the sensitivity cases presented here is inherently more likely than the collection of assumptions that represent the reference baseline in the figures and tables that follow. Nor is this sensitivity analysis intended to suggest that only one of the many assumptions made is likely to prove off-base with the passage of time or new observations. It is more likely that, when assumptions are eventually contradicted by future observation ( e.g., deviations in observed and predicted fuel prices are nearly a given), there will be collections of assumptions, rather than individual parameters, that simultaneously require updating. For this reason, we do not interpret the sensitivity analysis as necessarily providing justification for alternative regulatory scenarios to be preferred. Rather, the analysis simply provides an indication of which assumptions are most critical, and the extent to which future deviations from central analysis assumptions could affect costs and benefits of the rule. For a full discussion of how this information relates to NHTSA's determination of which regulatory alternatives are maximum feasible, please see Section VI.D].

Table V-29 lists and briefly describes the cases and alternative baseline that we examined in the sensitivity analysis. Note that some cases only apply to the LD fleet ( e.g., scenarios altering assumptions about fleet share modeling) and others only affect the HDPUV analysis ( e.g., initial PHEV availability).

Chapters 3 and 9 of the accompanying FRIA summarize results for the alternative baseline and sensitivity cases, and detailed model inputs and outputs for curious readers are available on NHTSA's website. For purposes of this preamble, the figures in Section V.D.1 illustrate the relative change of the sensitivity effect of selected inputs on the costs and benefits estimated for this rule for LDVs, while the figures in Section V.D.2 present the same data for the HDPUV analysis. Each collection of figures groups sensitivity cases by the category of input assumption ( e.g., macroeconomic assumptions, technology assumptions, and so on).

While the figures in this section do not show precise values, they give us a sense of which inputs are ones for which a different assumption would have a much different effect on analytical findings, and which ones would not have much effect. For example, assuming a different oil price trajectory would have a relatively large effect, as would doubling, or eliminating the assumed “payback period.” Sensitivity analyses also allow us to examine the impact of specific changes from the proposal on our findings. For example, in the final rule analysis, NHTSA used estimates of the social costs of greenhouse gases produced by the EPA, whereas these inputs were taken from the IWG in the proposal. This has a significant impact on net benefits, though they would remain strongly positive regardless of which set of estimates was used. The relative magnitude of these effects also varies by fleet. Making alternative assumptions about the future costs of battery technology has a larger effect on the HDPUV results. Adjusting assumptions related to the tax credits included in the IRA has a significant impact on results for both LDVs and HDPUVs. On the other hand, assumptions about which there has been significant disagreement in the past, like the rebound effect or the sales-scrappage response to changes in vehicle price, appear to cause only relatively small changes in net benefits across the range of analyzed input values. Chapter 9 of the FRIA provides an extended discussion of these findings, and presents net benefits estimated under each of the cases included in the sensitivity analysis.

The results presented in the earlier subsections of Section V and discussed in Section VI reflect NHTSA's best judgments regarding many different factors, and the sensitivity analysis discussed here is simply to illustrate the obvious, that differences in assumptions can lead to differences in analytical outcomes, some of which can be large and some of which may be smaller than expected. Policymaking in the face of future uncertainty is inherently complex. Section VI explains how NHTSA balances the statutory factors in light of the analytical findings, the uncertainty that we know exists, and our nation's policy goals, to set CAFE standards for model years 2027-2031, and HDPUV fuel efficiency standards for model year 2030 and beyond that NHTSA concludes are maximum feasible.

1. Passenger Cars and Light Trucks

Overall, NHTSA finds that for light duty vehicles, the preferred alternative PC2LT002 produces positive societal net benefits for each sensitivity and alternative baseline at both 3 and 7 percent discount rates. Societal net benefits are highest in the “No payback period” case ($33 billion) and lowest in the “Standard-setting conditions for MY 2023-2050” case ($7.7 billion) at a 3 percent social discount rate and 2 percent SC-GHG discount rate.

2. Heavy-Duty Pickups and Vans

In our HDPUV analysis the preferred alternative HDPUV108 produces positive net benefits for all but a handful of cases. In these cases, the alternative assumptions lead to greater technology adoption in the No-Action Alternative and lead to net benefits that are just below 0.

VI. Basis for NHTSA's Conclusion That the Standards Are Maximum Feasible

NHTSA's purpose in setting CAFE standards is to conserve energy, as directed by EPCA/EISA. Energy conservation provides many benefits to the American public, including better protection for consumers against changes in fuel prices, significant fuel savings and reduced impacts from harmful pollution. NHTSA continues to believe that fuel economy standards can function as an important insurance policy against oil price volatility, particularly to protect consumers even as the U.S. has improved its energy independence over time. Although NHTSA proposed PC2LT4 as the preferred alternative for CAFE standards for model years 2027-2031, NHTSA is finalizing PC2LT002 for those model years. Based on comments received and a closer look at the model results under the statutorily-constrained analysis, NHTSA now concludes that “shortfalls” and civil penalties must be managed in order to conserve manufacturer capital and resources for making the technological transition that NHTSA is prohibited from considering directly.

Similarly, for HDPUV, while NHTSA proposed HDPUV10 for model years 2030-2035, NHTSA is finalizing HDPUV108 for those model years. Based on comments received and a closer look at the model results—and specifically, as in the NPRM, the sensitivity analyses, as well as the apparent effects on certain manufacturers—NHTSA recognizes that uncertainty, particularly in the later model years of the rulemaking, means that a slower rate of increase is maximum feasible for those years. These conclusions, for both passenger cars and light trucks and for HDPUVs, will be discussed in more detail below.

A. EPCA, as Amended by EISA

EPCA, as amended by EISA, contains provisions establishing how NHTSA must set CAFE standards and fuel efficiency standards for HDPUVs. DOT (by delegation, NHTSA) must establish separate CAFE standards for passenger cars and light trucks for each model year, and each standard must be the maximum feasible that the Secretary (again, by delegation, NHTSA) determines manufacturers can achieve in that model year. In determining the maximum feasible levels of CAFE standards, EPCA requires that NHTSA consider four statutory factors: technological feasibility, economic practicability, the effect of other motor vehicle standards of the Government on fuel economy, and the need of the United States to conserve energy. NHTSA must also set separate standards for HDPUVs, and while those standards must also “achieve the maximum feasible improvement,” they must be “appropriate, cost-effective, and technologically feasible” —factors slightly different from those required to be considered for passenger car and light truck standards. NHTSA has broad discretion to balance the statutory factors in developing fuel consumption standards to achieve the maximum feasible improvement. In addition, NHTSA has the authority to consider (and typically does consider) other relevant factors, such as the effect of CAFE standards on motor vehicle safety.

The ultimate determination of what standards can be considered maximum feasible involves a weighing and balancing of factors, and the balance may shift depending on the information NHTSA has available about the expected circumstances in the model years covered by the rulemaking. NHTSA's decision must also be guided by the overarching purpose of EPCA, energy conservation, while balancing these factors.

EPCA/EISA also contain several other requirements, as follows.

1. Lead Time

a. Passenger Cars and Light Trucks

EPCA requires that NHTSA prescribe new CAFE standards at least 18 months before the beginning of each model year. Thus, if the first year for which NHTSA is establishing new CAFE standards is model year 2027, NHTSA interprets this provision as requiring us to issue a final rule covering model year 2027 standards no later than April 2025. No specific comments were received regarding the 18-month lead time requirement for CAFE standards, although ZETA and Hyundai commented that NHTSA should wait to finalize the CAFE standards until after DOE finalized the PEF revision, out of concern that failing to do so would “increase administrative burden for both” agencies, and that NHTSA's final rule would not otherwise “accurately reflect the final PEF.” Because NHTSA coordinated with DOE as both agencies worked to finalize their respective rules, this final rule reflects DOE's final PEF. Given that the Deputy Administrator of NHTSA signed this final rule in June 2024, the statutory lead time requirement is met.

b. Heavy-Duty Pickups and Vans

EISA requires that standards for commercial medium- and HD on-highway vehicles and work trucks (of which HDPUVs are part) provide not less than four full model years of regulatory lead time. Thus, if the first year for which NHTSA is establishing new fuel efficiency standards for HDPUVs is model year 2030, NHTSA interprets this provision as requiring us to issue a final rule covering model year 2030 standards no later than October 2025. Stellantis commented that it agreed with the proposal, that in order to provide four full model years of regulatory lead time, the earliest model year for which NHTSA could establish new standards was model year 2030. NHTSA agrees and is establishing new standards for HDPUVs beginning in model year 2030. This means that the applicable model years of NHTSA's final rule do not align perfectly with EPA's recent final rule establishing multipollutant (including GHG) standards for the same vehicles, but this is a direct consequence of the statutory lead time requirement in EISA. The Alliance and GM also agreed in their comments that model year 2030 was an appropriate start year for new HDPUV standards. GM stated that that timeframe “would provide manufacturers sufficient lead time to adjust product plans to standards.” Given that the Deputy Administrator of NHTSA signed this final rule in June, 2024, this lead time requirement is met.

EISA contains a related requirement for HDPUVs that the standards provide not only four full model years of regulatory lead time, but also three full model years of regulatory stability. As discussed in the Phase 2 final rule, Congress has not spoken directly to the meaning of the words “regulatory stability.” NHTSA interprets the “regulatory stability” requirement as ensuring that manufacturers will not be subject to new standards in repeated rulemakings too rapidly, given that Congress did not include a minimum duration period for the MD/HD standards. NHTSA further interprets the statutory meaning as reasonably encompassing standards which provide for increasing stringency during the rulemaking time frame to be the maximum feasible. In this statutory context, NHTSA thus interprets the phrase “regulatory stability” in section 32902(k)(3)(B) as requiring that the standards remain in effect for three years before they may be increased by amendment. It does not prohibit standards that contain predetermined stringency increases.

CEA commented that this interpretation was inconsistent with the law. It stated that a standard could not be “stable” if it “continually ratchets up each year,” and argued that HDPUV redesign cycles are longer than light truck redesign cycles and that “manufacturers would therefore have difficulty meeting standards that ratchet up every year.” In response, NHTSA continues to believe that “stable” can reasonably be interpreted as “known in advance” and “remaining in effect for three years,” in part because the dictionary provides definitions for “stable” that include “firmly established; fixed; steadfast; enduring.” While some definitions of “stable” mention “not changing or fluctuating; unvarying,” NHTSA believes that standards that are known in advance and established in three-year tranches can reasonably fit these definitions—the standards will not change or vary from what is established here, except by rulemaking as necessary (and as permissible given lead time requirements). EISA does not suggest that NHTSA interpret “unvarying” as exclusively suggesting that “standards may only increase once every three years and then must be held at that level,” and could also be reasonably read to suggest that “standards should not change from established levels, once established.” NHTSA is accordingly establishing new HDPUV standards in two tranches: standards that increase 10 percent per year for model years 2030-2031-2032, and standards that increase at 8 percent per year for model years 2033-2034-2035.

NHTSA also believes, based on comments, that redesign cycles should not be a problem for the HDPUV standards. NHTSA notes the comment from GM, mentioned above, that NHTSA beginning new standards in model year 2030 will provide sufficient lead time for manufacturers to adjust their product plans as needed, even while GM also noted that redesign cycles were longer for HDPUVs than for LTs. GM further stated that the lead time provided “lowers the likelihood of product disruptions in the market.” NHTSA agrees that HDPUV redesign cycles are longer than light truck redesign cycles and reflects this in our analysis, which shows the final standards (and indeed, all of the alternatives) as being achievable for the entirety of the HDPUV fleet, with no shortfalls under any regulatory alternative:

This approach is consistent with our understanding of regulatory stability. Manufacturers appear likely to have little to zero difficulty in meeting the final standards. Setting HDPUV standards that did not increase for three years instead would make little functional difference to compliance, given the availability of credit banking.

2. Separate Standards for Passenger Cars, Light Trucks, and Heavy-Duty Pickups and Vans, and Minimum Standards for Domestic Passenger Cars

EPCA requires NHTSA to set separate standards for passenger cars and light trucks for each model year. Based on the plain language of the statute, NHTSA has long interpreted this requirement as preventing NHTSA from setting a single combined CAFE standard for cars and trucks together. Congress originally required separate CAFE standards for cars and trucks to reflect the different fuel economy capabilities of those different types of vehicles, and over the history of the CAFE program, has never revised this requirement. Even as many cars and trucks have come to resemble each other more closely over time—many crossover and sport-utility models, for example, come in versions today that may be subject to either the car standards or the truck standards depending on their characteristics—it is still accurate to say that vehicles with truck-like characteristics such as 4-wheel drive, cargo-carrying capability, etc., currently consume more fuel per mile than vehicles without these components. While there have been instances in recent rulemakings where NHTSA raised passenger car and light truck standard stringency at the same numerical rate year over year, NHTSA also has precedent for setting passenger car and light truck standards that increase at different numerical rates year over year, as in the 2012 final rule. This underscores that NHTSA's obligation is to set maximum feasible standards separately for each fleet, based on our assessment of each fleet's circumstances as seen through the lens of the four statutory factors that NHTSA must consider. Regarding the applicability of the CAFE standards, individual citizens commenting via Climate Hawks Civic Action asked whether U.S. Postal Service vehicles, airplanes, and non-road engines (such as for lawn equipment) could also be subject to CAFE standards. Postal Service vehicles are generally HDPUVs, and thus subject to those standards rather than to CAFE standards. Airplanes and non-road engines are not automobiles under 49 U.S.C. 32901, so they cannot be subject to CAFE standards. An individual citizen with Climate Hawks Civic Action also requested that NHTSA not set separate standards for light trucks, on the basis that doing so would be detrimental to energy conservation. As explained above, NHTSA interprets 49 U.S.C. 32902 as requiring NHTSA to set separate standards for passenger cars and light trucks. Again, NHTSA does not believe that it has statutory authority to set a single standard for both passenger cars and light trucks.

EPCA, as amended by EISA, also requires another separate standard to be set for domestically manufactured passenger cars. Unlike the generally applicable standards for passenger cars and light trucks described above, the compliance obligation of the minimum domestic passenger car standard (MDPCS) is identical for all manufacturers. The statute clearly states that any manufacturer's domestically manufactured passenger car fleet must meet the greater of either 27.5 mpg on average, or “92 percent of the average fuel economy projected by the Secretary for the combined domestic and non-domestic passenger automobile fleets manufactured for sale in the United States by all manufacturers in the model year, which projection shall be published in the Federal Register when the standard for that model year is promulgated in accordance with [49 U.S.C. 32902(b)].” Since that statutory requirement was established, the “92 percent” has always been greater than 27.5 mpg, and foreseeably will continue to be so in the future. As in the 2020 and 2022 final rules, NHTSA continues to recognize industry concerns that actual total passenger car fleet standards have differed significantly from past projections, perhaps more so when NHTSA has projected significantly into the future. In the 2020 final rule, the compliance data showed that standards projected in the 2012 final rule were consistently more stringent than the actual standards as calculated at the end of the model year, by an average of 1.9 percent. NHTSA has stated that this difference indicates that in rulemakings conducted in 2009 through 2012, NHTSA's and EPA's projections of passenger car vehicle footprints and production volumes, in retrospect, underestimated the production of larger passenger cars over the model years 2011 to 2018 period.

Unlike the passenger car standards and light truck standards which are vehicle-attribute-based and automatically adjust with changes in consumer demand, the MDPCS are not attribute-based, and therefore do not adjust with changes in consumer demand and production. They are, instead, fixed standards that are established at the time of the rulemaking. As a result, by assuming a smaller-footprint fleet, on average, than what ended up being produced, the model year 2011-2018 MDPCS ended up being more stringent and placing a greater burden on manufacturers of domestic passenger cars than was projected and expected at the time of the rulemakings that established those standards. In the 2020 final rule, therefore, NHTSA agreed with industry concerns over the impact of changes in consumer demand (as compared to what was assumed in 2012 about future consumer demand for greater fuel economy) on manufacturers' ability to comply with the MDPCS and in particular, manufacturers that produce larger passenger cars domestically. Some of the largest civil penalties for noncompliance in the history of the CAFE program have been paid for noncompliance with the MDPCS. NHTSA also expressed concern at that time that consumer demand may shift even more in the direction of larger passenger cars if fuel prices continue to remain low. Sustained low oil prices can be expected to have real effects on consumer demand for additional fuel economy, and if that occurs, consumers may foreseeably be even more interested in 2WD crossovers and passenger-car-fleet SUVs (and less interested in smaller passenger cars) than they are at present.

Therefore, in the 2020 final rule, to help avoid similar outcomes in the 2021 to 2026 time frame to what had happened with the MDPCS over the preceding model years, NHTSA determined that it was reasonable and appropriate to consider the recent projection errors as part of estimating the total passenger car fleet fuel economy for model years 2021-2026. NHTSA therefore projected the total passenger car fleet fuel economy using the central analysis value in each model year, and applied an offset based on the historical 1.9 percent difference identified for model years 2011-2018.

For the 2022 final rule, NHTSA retained the 1.9 percent offset, concluding that it is difficult to predict passenger car footprint trends in advance, which means that, as various stakeholders have consistently noted, the MDPCS may turn out quite different from 92 percent of the ultimate average passenger car standard once a model year is complete. NHTSA also expressed concern, as suggested by the United Automobile, Aerospace, and Agricultural Implement Workers of America (UAW), that automakers struggling to meet the unadjusted MDPCS may choose to import their passenger cars rather than producing them domestically.

In the NPRM, NHTSA proposed to continue employing the 1.9 percent offset for model years 2027-2032, stating that NHTSA continued to believe that the reasons presented previously for the offset still apply, and that therefore the offset is appropriate, reasonable, and consistent with Congress' intent.

The Alliance, Ford, Nissan, and Kia commented that retaining the MDPCS offset was appropriate. Kia, for example, stated that it helped manufacturers avoid civil penalty payments, but expressed concern that the stringency of the proposed passenger car standards was so high that “even strong hybrids may not achieve the proposed MDPCS in the outer years.” Despite the offset, Kia suggested that this overall passenger car stringency could “complicate” Kia's continued ability to produce passenger cars in the United States.

The States and Cities commented that while the offset to the MDPCS was not “inherently unreasonable,” they disagreed with NHTSA's interpretation of 32902(b)(4). Specifically, they argued that “the average fuel economy projected by the Secretary for the combined domestic and non-domestic passenger car fleets . . .” should be interpreted to refer to the estimated achieved value rather than (as NHTSA has long interpreted it) to the estimated required value. The States and Cities commented that this reading was closer to the plain language of the statute, and asked NHTSA to clarify in the final rule that the offset was a “ proxy for the required projected average, [rather than] an interpretation away from the plain statutory text.” The States and Cities further requested that the offset, if any, be calculated as “the difference between the previous model years' central analysis value and average fuel economies achieved, rather than the difference between the projected and actual fleet-average standard.”

NHTSA has interpreted “projected” as referring to estimated required levels rather than estimated achieved levels since at least 2010. In the final rule establishing CAFE standards for model years 2012-2016, NHTSA noted that the Alliance had requested in its comments that the MDPCS be based on estimated achieved values. NHTSA responded that because Congress referred in the second clause of 32904(b)(4)(B) to the standard promulgated for that model year, therefore NHTSA interpreted the “projection” as needing to be based on the estimated required value ( i.e., the projection of the standard). The estimated achieved value represents manufacturers' assumed performance against the standard, not the standard itself. NHTSA believes that this logic continues to hold, and thus continues to determine the MDPCS based on the estimated required mpg levels projected for the model years covered by the rulemaking, and to determine the offset based on the estimated required levels rather than on the estimated achieved levels.

That said, NHTSA agrees that the offset is in some ways a proxy for 92 percent of the projected standard, insofar as the future is inherently uncertain and many different factors may combine to result in actual final passenger car mpg values that differ from those estimated as part of this final rule. Vehicle manufacturers may face even more uncertainty in the time frame of this rulemaking than they have faced since the MDPCS offset was first implemented. While NHTSA believes that the overall passenger car standards are maximum feasible based on the discussion in Section VI.D below, in response to Kia's comment that passenger car standard stringency may cause Kia to move its car production offshore, NHTSA continues to believe that the MDPCS offset helps to mitigate that uncertainty and perhaps to ease the major transition through which the industry is passing.

For HDPUVs, Congress gave DOT (by delegation, NHTSA) broad discretion to “prescribe separate standards for different classes of vehicles” under 49 U.S.C. 32902(k). HDPUVs are defined by regulation as “pickup trucks and vans with a gross vehicle weight rating between 8,501 pounds and 14,000 pounds (Class 2b through 3 vehicles) manufactured as complete vehicles by a single or final stage manufacturer or manufactured as incomplete vehicles as designated by a manufacturer.” NHTSA also allows HD vehicles above 14,000 pounds GVWR to be optionally certified as HDPUVs and comply with HDPUV standards “if properly included in a test group with similar vehicles at or below 14,000 pounds GVWR,” and “The work factor for these vehicles may not be greater than the largest work factor that applies for vehicles in the test group that are at or below 14,000 pounds GVWR.” Incomplete HD vehicles at or below 14,000 pounds GVWR may also be optionally certified as HDPUVs and comply with the HDPUV standards.

GM commented that it was appropriate for NHTSA to set HDPUV standards and passenger car/light truck CAFE standards in the same rulemaking, because electrifying certain light trucks could increase their weight to the point where they become HDPUVs, and “Conducting these rulemakings together is an important first step to considering this possibility when setting standards.” In response, NHTSA does track the classification of vehicles in order to ensure that its consideration of potential future CAFE and HDPUV stringencies is appropriately informed, and NHTSA did reassign vehicles from the light truck fleet to the HDPUV fleet (and vice versa) in response to stakeholder feedback to the NPRM. RVIA commented that the NPRM neither considered nor specifically mentioned motorhomes weighing less than 14,000 pounds GVWR, and expressed concern that the new standards would apply to these vehicles and “require [them] to be electrified.” In response, the Phase 2 MD/HD final rule explains that these vehicles are properly classified under EISA's definitions as Class 2b-8 vocational vehicles and not as HDPUVs. NHTSA is not setting new standards for vocational vehicles as part of this action. Moreover, as discussed elsewhere in this document, the HDPUV standards are performance-based standards and not electric-vehicle mandates.

AFPM commented that NHTSA “failed to address any of the unique statutory factors for HDPUVs,” pointing to 49 U.S.C. 32902(k)(1) and suggesting that NHTSA had not followed that section in developing its proposal. NHTSA agrees that it did not follow 32902(k)(1) in developing its proposal, because NHTSA executed the requirements of that section as part of the Phase 1 MD/HD fuel efficiency rulemaking, completed in 2011. NHTSA's website contains a link to the independent study that NHTSA performed, as directed by 32902(k)(1), following the publication of the NAS report. Because that statutory requirement has been executed, NHTSA did not undertake it again as part of this rulemaking.

NHTSA is establishing separate standards for “spark ignition” (SI, or gasoline-fueled) and “compression ignition” (CI, or diesel-fueled) HDPUVs, consistent with the existing Phase 2 standards. Each class of vehicles has its own work-factor based target curve; alternative fueled vehicles (such as BEVs) are subject to the standard for CI vehicles and HEVs and PHEVs are subject to the standard for SI vehicles. We understand that EPA has recently finalized a single curve for all HDPUVs regardless of fuel type. ACEEE commented that NHTSA should follow suit and raise the stringency of the gasoline standards to match that of the diesel standards, arguing that it would improve consistency with EPA's program and be consistent with NHTSA's acknowledgement of the emergence of van electrification. NHTSA is not taking this approach, for several reasons. First, EPA is modifying the model year 2027 standards set in the 2016 “Phase 2” rulemaking, and NHTSA cannot follow suit due to statutory lead time requirements. Second, EPA's single curve standard developed in GHG gas units (g CO₂ /mile) will still result in two separate curves when converted to the units used by NHTSA to set standards for fuel efficiency (gal/100 miles). This is a result of the differing amount of CO₂ released by each fuel type represented by each standard curve. Gasoline releases about 8,887g of CO₂ per gallon burned and diesel fuel releases about 10,180g of CO₂ per gallon burned. As an example, a model year 2030 HDPUV with a WF of 4500 would be required to produce less than 346 gCO2/mile according to the current EPA single curve standards; due to the difference in carbon content for fuels this translates to either a required gasoline consumption of less than 3.89 gal/100miles or a required diesel consumption of less than 3.4 gal/ 100miles. Considering difference in carbon content between gasoline and diesel, NHTSA chose to continue to use two separate curves based on combustion (and fuel) type because the agency believes it results in a closer harmonization between the NHTSA and EPA's standards when compared in fuel efficiency space. By retaining separate CI and SI curves NHTSA's standards will not only align closer with EPA's standards, but also better balance to the agency's statutory factors for HDPUVs: cost-effectiveness and technological feasibility.

3. Attribute-Based and Defined by a Mathematical Function

For passenger cars and light trucks, EISA requires NHTSA to set CAFE standards that are “based on 1 or more attributes related to fuel economy and express[ed]. in the form of a mathematical function.” Historically, NHTSA has based standards on vehicle footprint, and will continue to do so for model years 2027-2031. As in previous rulemakings, NHTSA defines the standards in the form of a constrained linear function that generally sets higher (more stringent) targets for smaller-footprint vehicles and lower (less stringent) targets for larger-footprint vehicles. Comments received on these aspects of the final rule are summarized and addressed in Section III.B of this preamble.

For HDPUVs, NHTSA also sets attribute-based standards defined by a mathematical function. HDPUV standards have historically been set in units of gallons per 100 miles, rather than in mpg, and the attribute for HDPUVs has historically been “work factor,” which is a function of a vehicle's payload capacity and towing capacity. Valero argued that setting HDPUV standards in units of gallons per 100 miles was inconsistent with the statutory text, and referred to 49 U.S.C. 32902(b)(1), which states that “average fuel economy standards” shall be prescribed for, among other things, “work trucks and commercial medium- and heavy-duty on-highway vehicles in accordance with subsection (k).” Valero argued that therefore the HDPUV standards are “fuel economy standards” and subject to the 32902(h) prohibitions. In response, NHTSA has long interpreted “fuel economy standards” in the context of 49 U.S.C. 32902(k) as referring not specifically to mpg, as in the passenger car/light truck context, but instead more broadly to account as accurately as possible for MD/HD fuel efficiency. In the Phase 1 MD/HD rulemaking, NHTSA considered setting standards for HDPUVs (and other MD/HD vehicles) in mpg, but concluded that that would not be an appropriate metric given the work that MD/HD vehicles are manufactured to do. NHTSA has thus set fuel efficiency standards for HDPUVs in this manner since 2011, and further notes that 32902(h) applies by its terms to subsections (c), (f), and (g), but not (b) or (k).

While NHTSA does not interpret EISA as requiring NHTSA to set attribute-based standards defined by a mathematical function for HDPUVs, given that 49 U.S.C. 32902(b)(3)(A) refers specifically to fuel economy standards for passenger and non-passenger automobiles, NHTSA has still previously concluded that following that approach for HDPUVs is reasonable and appropriate, as long as the work performed by HDPUVs is accounted for. NHTSA therefore continues to set work-factor based gallons-per-100-miles standards for HDPUVs for model years 2030-2035.

4. Number of Model Years for Which Standards May Be Set at a Time

For passenger cars and light trucks, EISA also states that NHTSA shall “issue regulations under this title prescribing average fuel economy standards for at least 1, but not more than 5, model years.” For this final rule, NHTSA is establishing new CAFE standards for passenger cars and light trucks for model years 2027-2031, and to facilitate longer-term product planning by industry and in the interest of harmonization with EPA, NHTSA is also presenting augural standards for model year 2032 as representative of what levels of stringency NHTSA currently believes could be appropriate in that model year, based on the information before us today. Hyundai commented that it supported the inclusion of the augural standards for model year 2032 to the extent that they were coordinated with EPA's final GHG standards for model year 2032, and were “representative of the actual starting point for the standards commencing in model year 2032.” The Alliance, in contrast, argued that presenting augural standards was “unnecessary and generally inconsistent with Congressional intent,” and that therefore NHTSA should defer any further mention of model year 2032 standards until a future rulemaking. In response, NHTSA has coordinated with EPA to the extent possible given our statutory restrictions and we continue to emphasize that the augural standards are informational only. As explained in the NPRM, a future rulemaking consistent with all applicable law will be necessary for NHTSA to establish final CAFE standards for model year 2032 passenger cars and light trucks. While the NPRM provided information about the impacts of the standards throughout the documents without distinguishing between the standards and the augural standards in the interest of brevity, the final rule and associated documents divorced the results for the augural model year 2032 standards (including the net benefits) to be abundantly clear that they are neither final nor included as part of the agency's decision on the model year 2027-2031 standards.

The five-year statutory limit on average fuel economy standards that applies to passenger cars and light trucks does not apply to the HD pickup and van standards. NHTSA has previously stated that “it is reasonable to assume that if Congress intended for the [MD/HD] regulatory program to be limited by the timeline prescribed in [49 U.S.C. 32902(b)(3)(B)], it would have either mentioned [MD/HD] vehicles in that subsection or included the same timeline in [49 U.S.C. 32902(k)].” Additionally, “in order for [49 U.S.C. 32902(b)(3)(B) to be interpreted to apply to [49 U.S.C. 32902(k)], the agency would need to give less than full weight to the . . . phrase in [49 U.S.C. 32902(b)(1)(C)] directing the Secretary to prescribe standards for `work trucks and commercial MD or HD on-highway vehicles in accordance with Subsection (k).' Instead, this direction would need to be read to mean `in accordance with Subsection (k) and the remainder of Subsection (b).' NHTSA believes this interpretation would be inappropriate. Interpreting `in accordance with Subsection (k)' to mean something indistinct from `in accordance of this Subsection' goes against the canon that statutes should not be interpreted in a way that `render[s] language superfluous.' Dobrova v. Holder, 607 F.3d 297, 302 (2d Cir. 2010), quoting Mendez v. Holder, 566 F.3d 316, 321-22 (2d Cir. 2009).” As a result, the standards previously set remain in effect indefinitely at the levels required in the last model year, until amended by a future rulemaking action.

5. Maximum Feasible Standards

As discussed above, EPCA requires NHTSA to consider four factors in determining what levels of CAFE standards (for passenger cars and light trucks) would be maximum feasible—technological feasibility, economic practicability, the effect of other motor vehicle standards of the Government on fuel economy, and the need of the United States to conserve energy. For determining what levels of fuel efficiency standards (for HDPUVs) would be maximum feasible, EISA requires NHTSA to consider three factors—whether a given fuel efficiency standard would be appropriate, cost-effective, and technologically feasible. NHTSA presents in the sections below its understanding of the meanings of all those factors in their respective decision-making contexts.

a. Passenger Cars and Light Trucks

(1) Technological Feasibility

“Technological feasibility” refers to whether a particular method of improving fuel economy is available for deployment in commercial application in the model year for which a standard is being established. Thus, NHTSA is not limited in determining the level of new standards to technology that is already being applied commercially at the time of the rulemaking. For this final rule, NHTSA has considered a wide range of technologies that improve fuel economy, while considering the need to account for which technologies have already been applied to which vehicle mode/configuration, as well as the need to estimate, realistically, the cost and fuel economy impacts of each technology as applied to different vehicle models/configurations. MEMA commented that it “appreciated NHTSA's openness to using different constellations of powertrains (BEV, PHEV, mild hybrid, ICE, FCEV, etc.) to comply with the standards.” NHTSA thanks MEMA, and continues to believe that the range of technologies considered, as well as how the technologies are defined for purposes of the analysis, is reasonable, based on our technical expertise, our independent research, and our interactions with stakeholders. NHTSA has not, however, attempted to account for every technology that might conceivably be applied to improve fuel economy, nor does NHTSA believe it is necessary to do so, given that many technologies address fuel economy in similar ways.

Several commenters focused on the technological feasibility of electrifying vehicle fleets. Jaguar commented that “At present, there are increasingly limited opportunities with regards to technologies that will meet the incredibly challenging standards set. Soon, it will only be possible to meet these targets with increased BEV sales.” Volkswagen commented that there may not be enough American-sourced batteries to meet both Inflation Reduction Act requirements and the proposed standards, that those limitations would prevent industry from manufacturing more than a certain number of BEVs per year, and that therefore the proposed standards were beyond technologically feasible and civil penalty payment would be unavoidable. AVE expressed concern about whether supply chains would be fully developed to support compliance. CFDC et al., a group of corn-based ethanol producers' organizations, argued that “shockingly high numbers” of electric vehicles would be required by the proposed standards, and that therefore the proposed standards were infeasible and unlawful because they could not be met without electric vehicles. The commenter further argued that “the proposal systematically neglects the fact that there are simply not enough minerals, particularly lithium, available to sustain global electric vehicle growth,” and that “this is an insuperable obstacle [that makes] NHTSA's proposal not technologically feasible.” RFA et al., another group of corn-based ethanol producers' organizations, commented that NHTSA had not adequately considered the technological feasibility of the regulatory reference baseline ( i.e., the amount of electrification assumed in response to State ZEV programs and assumed market demand), and that NHTSA's analysis of technological feasibility now needed to include consideration of critical mineral availability and BEV charging infrastructure. The Alliance commented that when it ran the CAFE model with BEVs removed from the analysis entirely and with no option for paying civil penalties, many fleets appeared unable to meet the proposed standards, which meant that the proposed standards were not technologically feasible. AFPM offered similar comments.

In response, NHTSA clarifies, again, that CAFE standards are performance-based standards, not technology mandates, and that NHTSA cannot set standards that require BEVs because NHTSA is statutorily prohibited from considering BEV fuel economy in determining maximum feasible CAFE standards. Commenters objecting to electrification shown in NHTSA's analysis are looking at what is assumed in the reference baseline levels, not what is required to meet NHTSA's final standards being promulgated in this rulemaking. As Table VI1 shows, the technology penetration rates for the various alternatives do not result in further penetration of BEVs in response to the action alternatives, although they do illustrate a potential compliance path for industry that would rely on somewhat higher numbers of SHEVs.

As to whether NHTSA is required to prove that the reference baseline as well as the CAFE standards are technologically feasible—a point also inherent in the Alliance comments, because the BEVs that the Alliance removed from its analysis were nearly all in the reference baseline—NHTSA disagrees that this is the agency's obligation under EPCA/EISA. Section IV above discusses the various considerations that inform the reference baselines. NHTSA has determined it is reasonable to assume that certain technologies will appear in the reference baseline, regardless of any action by NHTSA, in response to cost-effectiveness/market demand (as would occur if battery prices fall as currently assumed in our analysis, for example). Similarly, if certain technologies appear in the reference baseline because manufacturers have said they would plan to meet State regulations, then either the manufacturers have concluded that doing so is feasible (else they would not plan to do so), and/or the State(s) involved have made and are responsible for any determinations about feasibility. Nothing in EPCA/EISA compels NHTSA to be responsible for proving the feasibility of things which are beyond our authority, like State regulations or development of charging infrastructure or permitting of critical minerals production sites, and which involve consideration of technologies which NHTSA itself is prohibited from considering. Just as it is not NHTSA's authority or responsibility to determine whether State programs are feasible, so it is not NHTSA's responsibility to determine whether State programs are not feasible. State programs are developed under State legal authority, and their feasibility is a matter for the State(s) and vehicle manufacturers (and other interested parties) to discuss. Nonetheless, NHTSA continues to believe that it is reasonably foreseeable that manufacturers will at least plan to meet legally binding State regulations, and thus to reflect that intent in our regulatory reference baseline so that we may best reflect the world as it would look in the absence of further regulatory action by NHTSA.

Reviewing Table VI-1 above, our analysis of the final rule illustrates a technology path in which manufacturers might modestly increase strong hybrid-based technologies beyond reference baseline levels. CTLCV commented that the technology exists to meet the standards, but that the auto industry “must be required to provide the most efficient versions of gas-powered vehicles possible and not stand in the way of our transition to zero-emission vehicles.” The Joint NGOs commented that NHTSA's proposed standards were below maximum feasible levels because they do not represent future possible improvements that manufacturers could conceivably make to ICE vehicles. The Joint NGOs cited the 2022 EPA Trends Report as indicating that various manufacturers had “underutilized” technologies “such as turbocharged engines, continuously variable transmission and cylinder deactivation.” The Joint NGOs next cited an ICCT study suggesting that further “continual” improvements to cylinder deactivation, high compression Atkinson cycle engines, light weighting, and mild hybridization” could increase the fuel economy benefits of those technologies. The Joint NGOs then suggested that manufacturers could change the mix of vehicles they produced in a given model year so that only the “cleanest powertrain” was sold for each vehicle model. The Joint NGOs later stated that NHTSA's analysis was based on “what manufacturers `will,' `would,' or are `likely to' do—rather than what manufacturers `can' or `could' do.” The Joint NGOs argued that “many of these assumptions about what `would' happen are also based on a review of historical practice, rather than a forward-looking assessment of possibility.” The States and Cities also argued that all of the alternatives in the proposal were technologically feasible because they could be met with varying amounts of mass reduction and strong hybrids, technologies that certainly exist and are available for deployment. This commenter further argued that mass reduction was highly effective and that NHTSA should use its authority to encourage more mass reduction. Nissan, in contrast, expressed concern that the proposal would “divert significant resources towards further technological development of ICE vehicles, rather than allowing manufacturers to focus on fleet electrification goals.”

In response, while NHTSA sets performance-based standards rather than specifying which technologies should be used, NHTSA is mindful that industry is in the early to mid-stages of a major technological transition. NHTSA may not consider the fuel economy of BEVs when setting standards, but industry has made it extremely clear that it is committed to the transition to electric vehicles. The contrast between the comments from NRDC and the States and Cities, calling on NHTSA to somehow specifically require ongoing ICE vehicle improvements, and from Nissan, arguing that NHTSA must not require further ICE vehicle improvements, highlights this issue. NHTSA agrees that the technological feasibility factor allows NHTSA to set standards that force the development and application of new fuel-efficient technologies but notes this factor does not require NHTSA to do so. In the 2012 final rule, NHTSA stated that “[i]t is important to remember that technological feasibility must also be balanced with the other of the four statutory factors. Thus, while `technology feasibility' can drive standards higher by assuming the use of technologies that are not yet commercial, `maximum feasible' is also defined in terms of economic practicability, for example, which might caution the agency against basing standards (even fairly distant standards) entirely on such technologies.” NHTSA further stated that “as the `maximum feasible' balancing may vary depending on the circumstances at hand for the model year in which the standards are set, the extent to which technological feasibility is simply met or plays a more dynamic role may also shift.” With performance-based standards, NHTSA cannot mandate the mix of technologies that manufacturers will use to achieve compliance, so it is not within NHTSA's power to specifically require any particular type of ICE vehicle improvements, as NRDC and the States and Cities suggest and as Nissan fears. In determining maximum feasible CAFE standards, however, NHTSA can do its best to balance the concerns raised by all parties, as they fall under the various statutory factors committed to NHTSA's discretion. Whether these concerns are properly understood as ones of “technological feasibility” is increasingly murky as the technology transition (that NHTSA cannot consider directly) proceeds. NHTSA has also grappled with whether the “available for deployment in commercial application” language of our historical interpretation of technological feasibility is appropriately read as “available for deployment in the world” or “available for deployment given the restrictions of 32902(h).” The Heritage Foundation commented that “There is no doubt that EPCA is referring to” ICE vehicles in describing technological feasibility, because EPCA defines “fuel” as referring to gasoline or diesel fuels and electricity as an “alternative fuel,” and NHTSA is prohibited from considering alternative fueled vehicles in determining maximum feasible CAFE standards. Hyundai argued that the proposed PC2LT4 standards were not technologically feasible, because (1) the regulatory reference baseline included BEVs, and (2) DOE's changes to the PEF value and NHTSA's proposal to reduce available AC/OC flexibilities made any standards harder to meet. NHTSA agrees that it cannot consider BEV fuel economy in determining maximum feasible standards, but NHTSA reiterates that the technological transition that NHTSA is prohibited from considering in setting standards complicates the historical approach to the statutory factors. It may well be that in light of this transition, a better interpretation is for technological feasibility to be specifically limited to the technologies that NHTSA is permitted to consider.

Nevertheless, in the overall balancing of factors for determining maximum feasible, the above interpretive question may not matter, because it is clear that the very high cost of the most stringent alternatives likely puts them out of range of economic practicability, especially if manufacturers appear in NHTSA's analysis to be broadly resorting to payment of civil penalties rather than complying through technology application. Although some companies historically have chosen to pay civil penalties as a more cost-effective option than compliance, which NHTSA has not seen as an indication of infeasibility previously, the levels of widespread penalty payment rather than compliance projected in this analysis is novel. Further, penalty payment could detract from fuel economy during these particular model years, where manufacturers are devoting significant resources to a broader transition to electrification. Effectively, given the statutory constraints under which NHTSA must operate, and constraining technology deployment to what is feasible under expected redesign cycles, NHTSA does not see a technology path to reach the higher fuel economy levels that would be required by the more stringent alternatives, in the time frame of the rulemaking. Moreover, even if technological feasibility were not a barrier, that does not mean that requiring that technology to be added would be economically practicable under these specific circumstances.

IPI commented that NHTSA's inclusion in the NPRM of tables showing technology penetration rates under the “standard setting” analysis belied NHTSA's suggestion in the NPRM that there did not appear to be a technology path to reach the higher fuel economy levels that would be required by the more stringent alternatives. IPI suggested that either NHTSA must believe the more stringent alternatives to be impossible to meet in the rulemaking time frame, or that NHTSA was “collapsing” the technological feasibility factor into the economic practicability factor by considering cost under the heading of technological feasibility.

In response, within the context of the constrained analysis which NHTSA must consider by statute, NHTSA does find that there is no technology path for the majority of manufacturers to meet the most stringent CAFE alternatives, considering expected redesign cycles, without shortfalling and resorting to penalties. Even setting aside that some manufacturers have historically chosen to pay penalties rather than applying technology as an economic decision, NHTSA's final rule (constrained) analysis illustrates that a number of manufacturers do not have enough opportunities to redesign enough vehicles during the rulemaking time frame in order to achieve the levels estimated to be required by the more stringent alternatives.

Figure VI-2 through Figure VI-4 present several manufacturer-fleet combinations that clearly illustrate these limits in NHTSA's statutorily constrained analysis. The figures present fleet powertrain distribution along with vehicle redesign cycles. Each bar in the figure represents total manufacturer-fleet sales in a given model year, and bars are shaded to indicate the composition of sales by powertrain. Any portion of the bar with overlayed hashed lines denotes the portion of the manufacturer's fleet that is not eligible for redesign ( i.e., cannot change powertrain) in that model year, often due to recent redesigns and the need to adhere to the redesign cycle to avoid imposing costs for which NHTSA does not currently account. The left and right panels of the figure present results for the least and most stringent action alternatives, respectively, for comparison.

Figure VI-2 displays these results for Ford's light truck fleet. Under PC2LT002 (left panel), Ford's fleet complies with the standards in all model years, as shown in the row “Achieved FE relative to standard,” which has all results either positive or zero (meaning that the fleet exactly complies with Ford's estimated applicable standard). This occurs because the model converts a large part of the Ford light trucks eligible for redesign to SHEVs in model year 2027, represented by the large, un-hashed dark gray segment in the center of the model year 2027 bar. It continues to convert eligible MHEVs to SHEVs in model year 2028 and model year 2029. Under PC6LT8 (right panel), Ford converts all eligible vehicles to SHEVs in model years 2027, 2028, and 2029. Even with this technology application, Ford's achieved fuel economy levels do not meet the alternative's estimated standards (note the negative values in the row “Achieved FE relative to standard,”) and Ford is therefore assumed to pay civil penalties for model years 2028 and beyond. Under all alternatives, Ford has no light trucks eligible for redesign in model year 2030, and the only vehicles whose redesign schedule makes them eligible in model year 2031 are BEVs, which represent the end of the powertrain pathway and have no other technology that may be applied. According to the statutorily-constrained analysis that NHTSA considers for determining maximum feasible standards, Ford simply cannot comply with the PC6LT8 light truck standards beginning in model year 2028, because it has redesigned all the light trucks that it can (consistent with its redesign schedule) and is out of technology moves.

Other manufacturers encounter similar constraints at higher stringency levels and across fleets. As shown in Figure VI-3, the model converts nearly all eligible portions of GM's light truck fleet to PHEVs, but GM still encounters compliance constraints. These constraints are marginal under PC2LT002, but under PC6LT8, GM is unable to comply beginning in model year 2027, with shortfalls exceeding 3 MPG.

Figure VI-4 and Figure VI-5 show Toyota's import and domestic passenger car fleet, respectively. Under PC2LT002, Toyota's import passenger car fleet exceeds the applicable standard for all years, but in contrast Toyota's domestic passenger car fleet falls slightly short during model years 2027-2029. As in the other examples, this occurs due to the lack of powertrains eligible for redesign during those years. This phenomenon is even more pronounced and affects both Toyota's import and domestic passenger car fleets, under PC6LT8. Both of Toyota's passenger car fleets develop shortfalls but only the domestic fleet is able to eliminate the shortfall in the rulemaking time frame when redesigns are available in model year 2030.

Figure VI-6 shows Honda's domestic passenger car fleet CAFE performance. Under PC2LT002, the passenger car fleet complies with the standard across all years, achieving a 0.1 mpg overcompliance in model year 2027 and slowly increasing to a 2.2 mpg overcompliance by the end of the standard setting years. Under PC6LT8, Honda is unable to meet the standard for model year 2027 but reaches compliance by model year 2028 and maintains it through the standard-setting years. However, it is worth noting that the fleet drops from a 6.6 mpg overcompliance in model year 2029 to zero overcompliance in model year 2031, after converting over 75 percent of their fleet to advanced powertrain technologies, and Honda is the only non-BEV manufacturer to achieve consistent compliance under the highest stringency.

NHTSA conducted similar analysis for every manufacturer-fleet combination and found similar patterns and constraints on compliance. Results for manufacturers that make up the top 80 percent of fleet sales in model year 2031 are included in Table VI-2 and Table VI-3.

In the light truck fleet, nearly all vehicles are either ineligible for redesign or reach the end of their powertrain compliance pathways under PC6LT8, with the majority of manufacturers not achieving compliance, some falling short by as much as 18.7 mpg. Under PC2LT002, most manufacturers achieve the standard and overcomply somewhat, with only two manufacturers showing any shortfalls. And in all cases shown, representing 80 percent of all light truck sales volume, shortfalls are 1.8 mpg or less under PC2LT002.

All manufacturers shown, representing 80 percent of all passenger car sales volume, generally comply with fleet fuel economy levels in the passenger car fleet for the preferred alternative. Some manufacturers do show one or two years of shortfalls in the rulemaking time frame, resulting from redesign rate constraints, indicated by a lack of share eligibility. At high stringency levels, such as PC6LT8, the rate of stringency increase coupled with limited share eligibility makes compliance for the majority of the fleet untenable in NHTSA's statutorily constrained analysis.

The compliance illustrations in the figures and tables above demonstrate the challenge that higher stringencies pose, especially within the constrained modeling framework required by statute. Historically, in the constrained analysis, the higher levels of electrification that could be considered under the statute (SHEV and PHEV in charge sustaining mode) in addition to advanced engine modifications (turbocharging and HCR) easily provided the effectiveness levels needed to raise the manufacturers' fleet fuel economy when applied at the rates governed by refresh and redesign schedules. In past analyses, the cost of converting the vehicles to the new technologies was the limiting factor. However, the remaining percentages of fleets that can be modified consistent with redesign and refresh cycles, coupled with the limits of total fuel efficiency improvement possible (considering only statutorily-allowed technologies), now limits what is achievable by the manufacturers in the time frame of the rule. Regardless of the technology cost, or application of penalties, higher levels of fuel economy improvement are simply not achieved under the higher stringency alternatives, often because manufacturers have no opportunity to make the improvement and the statutorily-available technologies will not get them to where they would need to be.

For purposes of model years 2027-2031, NHTSA concludes that sufficient technology and timely opportunities to apply that technology exist to meet the final standards. Moreover, as Table VI-1 above shows, NHTSA's analysis demonstrates a technology path to meet the standards that does not involve application of BEVs, FCEVs, or other prohibited technologies. NHTSA therefore believes that the final standards are technologically feasible.

As discussed above, NHTSA also conducted a “No ZEV alternative baseline” analysis. Technology penetration rates and manufacturer compliance status results are somewhat different under that analysis, as might be foreseeable.

Comparing to the reference case baseline analysis results in Table VI-1, under the No ZEV alternative baseline analysis, BEV rates in the baseline go down in every model year (and remain at 0 percent for all action alternatives due to statutory constraints implemented in the model); SHEV rates increase by several percentage points; PHEV rates go up by about 1 percent; and advanced gasoline engine rates remain roughly the same in the baseline but drop several percentage points in the action alternatives. These trends hold across action alternatives.

In terms of manufacturers' ability to comply with different regulatory alternatives given existing redesign schedules, results for the light truck fleet under the No ZEV alternative baseline did not vary significantly from the results presented in Table VI-2 for the reference case baseline analysis. Manufacturer light truck shortfalls under PC6LT8 were still nearly universal, with maximum shortfalls reaching more than 19 mpg, higher than the shortfalls under the reference case baseline. Ford, GM, and Nissan light truck penalties are almost identical under both baselines. Under the No ZEV alternative baseline analysis, Toyota still pays no light truck penalties under PC2LT002, and generally lower penalties under PC6LT8. Stellantis pays slightly higher penalties under PC2LT002, and generally lower penalties under PC6LT8. Honda and Subaru still pay no penalties under PC2LT002 and pay somewhat higher penalties under PC6LT8.

For passenger car shortfalls, the use of the No ZEV alternative baseline does not change much for Hyundai, Kia, VWA, Tesla, or GM (which in GM's case, illustrates that most of GM's compliance difficulty is in its light truck fleet), when comparing the results of the above table with Table VI-3. Toyota and Honda see higher passenger car penalties under the No ZEV alternative baseline for both PC2LT002 and PC6LT8, with fewer opportunities for redesigns. Nissan sees higher penalties under the No ZEV alternative baseline even though redesign opportunities are nearly identical.

Based on these results, which are generally quite similar to those under the reference case baseline, NHTSA finds that using the No ZEV alternative baseline would not change our conclusions regarding the technological feasibility of the various action alternatives.

(2) Economic Practicability