Greenhouse Gas Emissions Standards for Heavy-Duty Vehicles-Phase 3

AGENCY:

Environmental Protection Agency (EPA).

ACTION:

Notice of proposed rulemaking.

SUMMARY:

The Environmental Protection Agency (EPA) is proposing to promulgate new GHG standards for heavy-duty highway vehicles starting in model year (MY) 2028 through MY 2032 and to revise certain GHG standards for MY 2027 that were established previously under EPA's Greenhouse Gas Emissions and Fuel Efficiency Standards for Medium- and Heavy-Duty Engines and Vehicles—Phase 2 rule (“HD GHG Phase 2”). This document proposes updates to discrete elements of the Averaging Banking and Trading program, including a proposal to eliminate the last MY year of the HD GHG Phase 2 advanced technology incentive program for certain types of electric highway heavy-duty vehicles. EPA is proposing to add warranty requirements for batteries and other components of zero-emission vehicles and to require customer-facing battery state-of-health monitors for plug-in hybrid and battery electric vehicles. In this document, we are also proposing additional revisions and clarifying and editorial amendments to certain highway heavy-duty vehicle provisions and certain test procedures for heavy-duty engines. Finally, as part of this action, EPA is proposing to revise its regulations addressing preemption of state regulation of new locomotives and new engines used in locomotives.

DATES:

Comments must be received on or before June 16, 2023. Comments on the information collection provisions submitted to the Office of Management and Budget (OMB) under the Paperwork Reduction Act (PRA) are best assured of consideration by OMB if OMB receives a copy of your comments on or before May 30, 2023. Public hearing: EPA will announce information regarding the public hearing for this proposal in a supplemental Federal Register document. Please refer to the SUPPLEMENTARY INFORMATION section for additional information on the public hearing.

ADDRESSES:

You may send comments, identified by Docket ID No. EPA–HQ–OAR–2022–0985, by any of the following methods:

• Federal eRulemaking Portal: https://www.regulations.gov/ (our preferred method). Follow the online instructions for submitting comments.

• Email: a-and-r-Docket@epa.gov. Include Docket ID No. EPA–HQ–OAR–2022–0985 in the subject line of the message.

• Mail: U.S. Environmental Protection Agency, EPA Docket Center, OAR Docket, Mail Code 28221T, 1200 Pennsylvania Avenue NW, Washington, DC 20460.

• Hand Delivery or Courier: EPA Docket Center, WJC West Building, Room 3334, 1301 Constitution Avenue NW, Washington, DC 20004. The Docket Center's hours of operations are 8:30 a.m.–4:30 p.m., Monday–Friday (except Federal Holidays).

Instructions: All submissions received must include the Docket ID No. for this rulemaking. Comments received may be posted without change to https://www.regulations.gov/, including any personal information provided. For detailed instructions on sending comments and additional information on the rulemaking process, see the “Public Participation” heading of the SUPPLEMENTARY INFORMATION section of this document.

FOR FURTHER INFORMATION CONTACT:

Brian Nelson, Assessment and Standards Division, Office of Transportation and Air Quality, Environmental Protection Agency, 2000 Traverwood Drive, Ann Arbor, MI 48105; telephone number: (734) 214–4278; email address: nelson.brian@epa.gov.

SUPPLEMENTARY INFORMATION:

Public Participation

Written Comments

Submit your comments, identified by Docket ID No. EPA–HQ–OAR–2022–0985, at https://www.regulations.gov (our preferred method), or the other methods identified in the ADDRESSES section. Once submitted, comments cannot be edited or removed from the docket. The EPA may publish any comment received to its public docket. Do not submit to EPA's docket at https://www.regulations.gov any information you consider to be Confidential Business Information (CBI), Proprietary Business Information (PBI), or other information whose disclosure is restricted by statute. If you choose to submit CBI or PBI as a comment to EPA's docket, please send those materials to the person listed in the FOR FURTHER INFORMATION CONTACT section. Multimedia submissions (audio, video, etc.) must be accompanied by a written comment. The written comment is considered the official comment and should include discussion of all points you wish to make. The EPA will generally not consider comments or comment contents located outside of the primary submission ( i.e., on the web, cloud, or other file sharing system). Commenters who would like EPA to further consider in this rulemaking any relevant comments that they provided on the HD2027 NPRM regarding proposed HD vehicle GHG standards for the MYs at issue in this proposal must resubmit those comments to EPA during this proposal's comment period. Please visit https://www.epa.gov/dockets/commenting-epa-dockets for additional submission methods; the full EPA public comment policy; information about CBI, PBI, or multimedia submissions; and general guidance on making effective comments.

Participation in Virtual Public Hearing

EPA will announce information regarding the public hearing for this proposal in a supplemental Federal Register document. The hearing notice, registration information, and any updates to the hearing schedule will also be available at https://www.epa.gov/regulations-emissions-vehicles-and-engines/proposed-rule-greenhouse-gas-emissions-standards-heavy. Please refer to this website for any updates regarding the hearings. EPA does not intend to publish additional documents in the Federal Register announcing updates to the hearing schedule.

Docket: All documents in the docket are listed on the www.regulations.gov website. Although listed in the index, some information is not publicly available, e.g., CBI or other information whose disclosure is restricted by statute. Certain other material, such as copyrighted material, is not placed on the internet and will be publicly available only in hard copy form through the EPA Docket Center at the location listed in the ADDRESSES section of this document.

General Information

Does this action apply to me?

This action relates to companies that manufacture, sell, or import into the United States new heavy-duty highway vehicles and engines. This action also relates to state and local governments. Potentially affected categories and entities include the following:

This table is not intended to be exhaustive, but rather provides a guide for readers regarding entities potentially affected by this action. This table lists the types of entities that EPA is now aware could potentially be affected by this action. Other types of entities not listed in the table could also be affected. To determine whether your entity is regulated by this action, you should carefully examine the applicability criteria found in 40 CFR parts 1036, 1037, 1054, 1065, and 1074. If you have questions regarding the applicability of this action to a particular entity, consult the person listed in the FOR FURTHER INFORMATION CONTACT section.

What action is the Agency taking?

The Environmental Protection Agency (EPA) is proposing to promulgate new GHG standards for heavy-duty highway vehicles starting in model year (MY) 2028 through MY 2032 and to revise certain GHG standards for MY 2027 that were established previously under EPA's Greenhouse Gas Emissions and Fuel Efficiency Standards for Medium- and Heavy-Duty Engines and Vehicles—Phase 2 rule (“HD GHG Phase 2”) that we believe are appropriate and feasible considering lead time, costs, and other factors. EPA also proposes that it is appropriate to eliminate the last model year (MY 2027) of advanced technology incentives for certain electric highway heavy-duty vehicles, initially established under the HD GHG Phase 2 rule. EPA is proposing to add warranty requirements for batteries and other components of zero-emission vehicles and to require customer-facing battery state-of-health monitors for plug-in hybrid and battery electric vehicles. We are also proposing revisions and clarifying and editorial amendments to certain highway heavy-duty vehicle provisions of 40 CFR part 1037 and certain test procedures for heavy-duty engines in 40 CFR parts 1036 and 1065. In addition, in this action EPA is proposing to revise its regulations addressing preemption of state regulation of new locomotives and new engines used in locomotives, to more closely align with language in the Clean Air Act.

What is the Agency's authority for taking this action?

Clean Air Act section 202(a), 42 U.S.C. 7521(a), requires that EPA establish emission standards for air pollutants from new motor vehicles or new motor vehicle engines, which, in the Administrator's judgment, cause or contribute to air pollution that may reasonably be anticipated to endanger public health or welfare. The Administrator has found that GHG emissions from highway heavy-duty vehicles and engines cause or contribute to air pollution that may endanger public health or welfare. Therefore, the Administrator is exercising his authority under CAA section 202(a)(1)–(2) to establish standards for GHG emissions from highway heavy-duty vehicles. In addition, section 209(e)(2)(B) of the CAA, 42 U.S.C. 7543(e)(2)(B), requires EPA to promulgate regulations implementing subsection 209(e) of the Act, which addresses the prohibition of state standards regarding certain classes of new nonroad engines or new nonroad vehicles including new locomotives and new engines used in locomotives, as well as EPA's authorization criteria for certain California standards for other nonroad engines or nonroad vehicles. See Section I.D of this preamble for more information on the agency's authority for this action.

Did EPA conduct a peer review before issuing this action?

This proposed regulatory action is supported by influential scientific information. EPA, therefore, is conducting peer review in accordance with OMB's Final Information Quality Bulletin for Peer Review. Specifically, we conducted the peer review process on two analyses: (1) Emission Adjustments for Onroad Vehicles in MOVES3.R1, and (2) Greenhouse Gas and Energy Consumption Rates for Onroad Vehicles in MOVES3.R1. In addition, we plan to conduct a peer review of inputs to the Heavy-Duty Technology Resource Use Case Scenario (HD TRUCS) tool used to analyze HD vehicle energy usage and associated component costs. All peer review were or will be in the form of letter reviews conducted by a contractor. The peer review reports for each analysis will be posted in the docket for this action and will be posted at EPA's Science Inventory ( https://cfpub.epa.gov/si/).

Table of Contents

Executive Summary

A. Need for Regulatory Action

B. The Opportunity for Clean Air Provided by Zero-Emission Vehicle Technologies

C. Summary of the Major Provisions in the Regulatory Action

D. Impacts of the Proposed Standards

I. Introduction

A. Brief Overview of the Heavy-Duty Industry

B. History of Greenhouse Gas Emission Standards for Heavy-Duty Engines and Vehicles

C. What has changed since we finalized the HD GHG Phase 2 rule?

D. EPA Statutory Authority for the Proposal

E. Coordination With Federal and State Partners

F. Stakeholder Engagement

II. Proposed CO₂ Emission Standards

A. Public Health and Welfare Need for GHG Emission Reductions

B. Summary of Comments Received From HD2027 NPRM

C. Background on the CO₂ Emission Standards in the HD GHG Phase 2 Program

D. Vehicle Technologies

E. Technology, Charging Infrastructure, and Operating Costs

F. Proposed Standards

G. EPA's Basis That the Proposed Standards Are Feasible and Appropriate Under the Clean Air Act

H. Potential Alternatives

I. Small Businesses

III. Compliance Provisions, Flexibilities, and Test Procedures

A. Proposed Revisions to the ABT Program

B. Battery Durability Monitoring and Warranty Requirements

C. Additional Proposed Revisions to the Regulations

IV. Proposed Program Costs

A. IRA Tax Credits

B. Technology Package Costs

C. Manufacturer Costs

D. Purchaser Costs

E. Social Costs

V. Estimated Emission Impacts From the Proposed Program

A. Model Inputs

B. Estimated Emission Impacts From the Proposed Standards

VI. Climate, Health, Air Quality, Environmental Justice, and Economic Impacts

A. Climate Change Impacts

B. Health and Environmental Effects Associated With Exposure to Non-GHG Pollutants

C. Air Quality Impacts of Non-GHG Pollutants

D. Environmental Justice

E. Economic Impacts

F. Oil Imports and Electricity and Hydrogen Consumption

VII. Benefits of the Proposed Program

A. Social Cost of GHGs

B. Criteria Pollutant Health Benefits

C. Energy Security

VIII. Comparison of Benefits and Costs

A. Methods

B. Results

IX. Analysis of Alternative CO₂ Emission Standards

A. Comparison of Proposal and Alternative

B. Emission Inventory Comparison of Proposal and Slower Phase-In Alternative

C. Program Costs Comparison of Proposal and Alternative

D. Benefits

E. How do the proposal and alternative compare in overall benefits and costs?

X. Preemption of State Standards and Requirements for New Locomotives or New Engines Used in Locomotives

A. Overview

B. Background

C. Evaluation of Impact of Regulatory Preemption

D. What is EPA proposing?

XI. Statutory and Executive Order Reviews

A. Executive Order 12866: Regulatory Planning and Review and Executive Order 13563: Improving Regulation and Regulatory Review

B. Paperwork Reduction Act (PRA)

C. Regulatory Flexibility Act (RFA)

D. Unfunded Mandates Reform Act (UMRA)

E. Executive Order 13132: Federalism

F. Executive Order 13175: Consultation and Coordination With Indian Tribal Governments

G. Executive Order 13045: Protection of Children From Environmental Health and Safety Risks

H. Executive Order 13211: Actions Concerning Regulations That Significantly Affect Energy Supply, Distribution, or Use

I. National Technology Transfer and Advancement Act (NTTAA) and 1 CFR Part 51

J. Executive Order 12898: Federal Actions To Address Environmental Justice in Minority Populations and Low-Income Populations.

XII. Statutory Authority and Legal Provisions

List of Subjects

Executive Summary

A. Need for Regulatory Action

The Environmental Protection Agency (EPA) is proposing this action to further reduce GHG air pollution from highway heavy-duty (hereafter referred to as “heavy-duty” or HD) engines and vehicles across the United States. Despite the significant emissions reductions achieved by previous rulemakings, GHG emissions from HD vehicles continue to impact public health, welfare, and the environment. The transportation sector is the largest U.S. source of GHG emissions, representing 27 percent of total GHG emissions. Within the transportation sector, heavy-duty vehicles are the second largest contributor to GHG emissions and are responsible for 25 percent of GHG emissions in the sector. GHG emissions have significant impacts on public health and welfare as evidenced by the well-documented scientific record and as set forth in EPA's Endangerment and Cause or Contribute Findings under Section 202(a) of the CAA. Additionally, major scientific assessments continue to be released that further advance our understanding of the climate system and the impacts that GHGs have on public health and welfare both for current and future generations, as discussed in Section II.A.

The potential for the application of zero-emission vehicle (ZEV) technologies in the heavy-duty sector presents an opportunity for significant reductions in heavy-duty GHG emissions over the long term. Major trucking fleets, HD vehicle and engine manufacturers, and U.S. states have announced plans to increase the use of heavy-duty zero-emissions technologies in the coming years. The 2021 Infrastructure Investment and Jobs Act (commonly referred to as the “Bipartisan Infrastructure Law” or BIL) and the Inflation Reduction Act of 2022 (“Inflation Reduction Act” or IRA) together include many incentives for the development, production, and sale of ZEVs, electric charging infrastructure, and hydrogen, which are expected to spur significant innovation in the heavy-duty sector. In addition, supporting assessments provided by some commenters during the comment period for the EPA's March 2022 Notice of Proposed Rulemaking “Control of Air Pollution from New Motor Vehicles: Heavy-Duty Engine and Vehicle Standards” (hereafter referred to as “HD2027 NPRM”), which proposed strengthening existing MY 2027 GHG standards for heavy-duty vehicles, suggested that significant ZEV adoption rates can be achieved over the next decade. We discuss these developments in more detail in Section I. EPA also projects that improvements in internal combustion engines, powertrains, and vehicle technologies such as those EPA projected would be used to achieve the HD GHG Phase 2 standards will also be needed to continue to reduce GHG emissions from the HD sector, and as described in Section II.D.1, these technology improvements continue to be feasible. With respect to the need for GHG reductions and these heavy-duty sector developments, EPA is proposing in this document more stringent MY 2027 HD vehicle CO₂ emission standards ( i.e., beyond what was finalized in HD GHG Phase 2) and new HD vehicle CO2 emission standards starting in MYs 2028 through 2032 that we believe are appropriate and feasible considering cost, lead time, and other factors, as described throughout this preamble and supporting materials in the docket for this proposed rulemaking.

EPA sets highway heavy-duty vehicle and engine standards for GHG emissions under its authority in CAA section 202(a). Section 202(a)(1) states that “the Administrator shall by regulation prescribe (and from time to time revise) . . . standards applicable to the emission of any air pollutant from any class or classes of new motor vehicles or new motor vehicle engines, . . . which in his judgment cause, or contribute to, air pollution which may reasonably be anticipated to endanger public health or welfare.” Section 202(a)(2) provides that standards under section 202(a) apply to such vehicles and engines “after such period as the Administrator finds necessary to permit the development and application of the requisite technology, giving appropriate consideration to the cost of compliance within such period.” Pursuant to section 202(a)(1), such standards apply to vehicles and engines “for their useful life.” EPA also may consider other factors such as the impacts of potential GHG standards on the industry, fuel savings, oil conservation, energy security, and other relevant considerations. Congress authorized the Administrator to determine the levels of emission reductions achievable for such air pollutants through the application of technologies taking into account cost, lead time, and other factors.

Pursuant to our 202(a) authority, EPA first established standards for the heavy-duty sector in the 1970s. Since then, the Agency has revised the standards multiple times based upon updated data and information, the continued need to mitigate air pollution, and Congressional enactments directing EPA to regulate emissions from the heavy-duty sector more stringently. Since 1985, HD engine and vehicle manufacturers could comply with criteria-pollutant standards using averaging, EPA also introduced banking and trading compliance flexibilities in the HD program in 1990, and EPA's HD GHG standards and regulations have consistently included an averaging, banking, and trading (ABT) program from the start. Since the first standards, subsequent standards have extended to additional pollutants (including GHGs), increased in stringency, and spurred the development and deployment of numerous new vehicle and engine technologies. For example, the most recent GHG standards for HD vehicles will reduce CO₂ emissions by approximately 1.1 billion metric tons over the lifetime of the new vehicles sold under the program (HD GHG Phase 2, 81 FR 73478, October 25, 2016) and the most recent criteria-pollutant standards are projected to reduce NO_X emissions from the in-use HD fleet by almost 50 percent in 2045 (“Control of Air Pollution from New Motor Vehicles: Heavy-Duty Engine and Vehicle Standards” (hereafter referred to as “HD2027 FRM”), 88 FR 4296, January 24, 2023). This proposal builds upon this multi-decadal tradition of regulating heavy-duty vehicles and engines, by applying the Agency's clear and longstanding statutory authority considering new real-world data and information, including recent Congressional action in the Bipartisan Infrastructure Law (BIL) and Inflation Reduction Act (IRA).

This Notice of Proposed Rulemaking is consistent with Executive Order 14037 on Strengthening American Leadership in Clean Cars and Trucks, which directs the Administrator to “consider updating the existing greenhouse gas emissions standards for heavy-duty engines and vehicles beginning with model year 2027 and extending through and including at least model year 2029” and directs EPA to “consider beginning work on a rulemaking under the Clean Air Act to establish new greenhouse gas emissions standards for heavy-duty engines and vehicles to begin as soon as model year 2030.” Consistent with this direction, in the HD2027 NPRM, we proposed building on and improving the existing emission control program for highway heavy-duty vehicles by further strengthening certain MY 2027 GHG standards finalized under the HD GHG Phase 2 rule. However, we did not take final action on the GHG portion of the HD2027 proposal in the final rule (HD2027 FRM). Since that time, EPA has continued its analysis of the heavy-duty vehicle sector including the recent passage of the IRA, which as we discuss further in this preamble provides significant incentives for GHG reductions in the heavy-duty vehicle sector. Based on this updated information and analysis, and consistent with EPA's authority under the Clean Air Act section 202(a), we are issuing this Notice of Proposed Rulemaking (“HD GHG Phase 3 NPRM”) to propose certain revised HD vehicle carbon dioxide (CO₂) standards for MY 2027 and certain new HD vehicle CO₂ standards for MYs 2028, 2029, 2030, 2031, and 2032 that would achieve significant GHG reductions for these and later model years (note the MY 2032 standards would remain in place for MY 2033 and later). We are requesting comment on an alternative set of CO₂ standards that would more gradually increase in stringency than the proposed standards for the same MYs. EPA also requests comment on setting GHG standards starting in MYs 2027 through 2032 that would reflect: values less stringent than the lower stringency alternative for certain market segments, values in between the proposed standards and the alternative standards, values in between the proposed standards and those that would reflect ZEV adoption levels ( i.e., percent of ZEVs in production volumes) used in California's ACT, values that would reflect the level of ZEV adoption in the ACT program, and values beyond those that would reflect ZEV adoption levels in ACT such as the 50- to 60-percent ZEV adoption range represented by the publicly stated goals of several major original equipment manufacturers (OEMs) for 2030. We also request comment on promulgating additional new standards with increasing stringency in MYs 2033 through 2035. EPA anticipates that the appropriate choice of final standards within this range will reflect the Administrator's judgments about the uncertainties in EPA's analyses as well as consideration of public comment and updated information where available.

CAA section 202(a) directs EPA to regulate emissions of air pollutants from new motor vehicles and engines, which in the Administrator's judgment, cause or contribute to air pollution that may reasonably be anticipated to endanger public health or welfare. While standards promulgated pursuant to CAA section 202(a) are based on application of technology, the statute does not specify a particular technology or technologies that must be used to set such standards; rather, Congress has authorized and directed EPA to adapt its standards to emerging technologies. In 2009, the Administrator issued an Endangerment Finding under CAA section 202(a), concluding that GHG emissions from new motor vehicles and engines, including heavy-duty vehicles and engines, cause or contribute to air pollution that may endanger public health or welfare. Pursuant to the 2009 Endangerment and Cause or Contribute Finding, EPA promulgated GHG regulations for heavy-duty vehicles and engines in 2011 and 2016, referred to as the HD GHG Phase 1 and HD GHG Phase 2 programs, respectively. In the HD GHG Phase 1 and Phase 2 programs, EPA set emission standards that the Agency found appropriate and feasible, considering cost, lead time, and other factors.

Over time, manufacturers have not only continued to find ways to further reduce emissions from motor vehicles, including HD vehicles, they have found ways to eliminate tailpipe emissions entirely through the use of zero-emission vehicle technologies. Since the 2009 Endangerment and Cause or Contribute Finding and issuance of the HD GHG Phase 1 and Phase 2 program regulations, there has continued to be significant technological advancement in the vehicle and engine manufacturing sectors, including for such zero-emission vehicle technologies. The HD Phase 3 regulations that we are proposing take into account the ongoing technological innovation in the HD vehicle space and reflect CO₂ emission standards that we consider appropriate and feasible considering cost, lead time, and other factors.

B. The Opportunity for Clean Air Provided by Zero-Emission Vehicle Technologies

When the HD GHG Phase 2 rule was promulgated in 2016, we established CO₂ standards on the premise that ZEV technologies, such as battery electric vehicles (BEVs) and fuel cell electric vehicles (FCEVs), would become more widely available in the heavy-duty market over time, but not in significant volume in the timeframe of the Phase 2 program. We finalized BEV, plug-in hybrid electric vehicle (PHEV), and FCEV advanced technology credit multipliers to encourage the development and sales of these advanced technologies.

Several significant developments have occurred since 2016 that point to ZEV technologies becoming more readily available much sooner than we had previously projected for the HD sector. These developments support the feasibility of ZEV technologies and render adoption of ZEV technologies to reduce GHG emissions more cost-competitive than ever before. First, the HD market has evolved such that early ZEV models are in use today for some applications and are expected to expand to many more applications; costs of ZEV technologies have gone down and are projected to continue to fall; and manufacturers have announced plans to rapidly increase their investments in ZEV technologies over the next decade. In 2022, there were a number of manufacturers producing fully electric HD vehicles for use in a number of applications, and these small volumes are expected to rise (see Section I.C and Draft Regulatory Impact Analysis (DRIA) Chapter 1). The cost to manufacture lithium-ion batteries (the single most expensive component of a BEV) has dropped significantly in the past eight years, and that cost is projected to continue to fall during this decade, all while the performance of the batteries (in terms of energy density) improves. Many of the manufacturers that produce HD vehicles and major firms that purchase HD vehicles have announced billions of dollars' worth of investments in ZEV technologies and significant plans to transition to a zero-carbon fleet over the next ten to fifteen years.

Second, the 2021 BIL and the 2022 IRA laws provide significant and unprecedented monetary incentives for the production and purchase of qualified ZEVs in the HD market. They also provide incentives for qualifying electric charging infrastructure and hydrogen, which will further support a rapid increase in market penetration of HD ZEVs. As a few examples, over the next five years, BIL provisions include $5 billion to fund the replacement of school buses with zero- or low-emission buses and $5.6 billion to support the purchase of zero- or low-emission transit buses and associated infrastructure, with up to $7.5 billion to help build out a national network of EV charging and hydrogen refueling infrastructure, some of which may be used for refueling of heavy duty vehicles. The IRA creates a tax credit of up to $40,000 per vehicle for vehicles over 14,000 pounds (and up to $7,500 per vehicle for vehicles under 14,000 pounds) for the purchase of qualified commercial clean vehicles and provides tax credits for the production and sale of battery cells and modules of up to $45 per kilowatt-hour (kWh). The wide array of incentives in both laws will help to reduce the costs to manufacture, purchase, and operate ZEVs, thereby bolstering their adoption in the market.

Third, there have been multiple actions by states to accelerate the adoption of HD ZEVs. The State of California and other states have adopted the ACT program that includes a manufacturer requirement for zero-emission truck sales. The ACT program would require that “manufacturers who certify Class 2b-8 chassis or complete vehicles with combustion engines would be required to sell zero-emission trucks as an

increasing percentage of their annual [state] sales from 2024 to 2035.” In addition, 17 states and the District of Columbia have signed a Memorandum of Understanding establishing goals to support widespread electrification of the HD vehicle market. We discuss these factors further in Section I.

Recognizing the need for additional GHG reductions from HD vehicles and the growth of ZEV technologies in the HD market, last year we proposed strengthening certain existing MY 2027 HD vehicle CO₂ standards as part of the HD2027 NPRM. We received many comments on the proposed updates to those HD vehicle CO₂ emission standards. Many commenters suggested that EPA should further strengthen HD vehicle CO₂ emission standards in MYs 2027 through 2029 beyond the HD2027 NPRM proposed levels because of the accelerating adoption of HD ZEV technologies, and some commenters provided a number of reports that evaluate the potential of electrification of the HD sector in terms of adoption rates, costs, and other factors. Some commenters raised concerns with the HD2027 NPRM proposed changes to certain HD GHG Phase 2 CO₂ emission standards, asserting the significant investment and lead time required for development and verification of the durability of ZEV technologies, especially given the diverse range of applications in the HD market.

In the HD2027 NPRM, EPA also requested comment on several approaches to modify the existing Advanced Technology Credit Multipliers (“credit multipliers”) under the HD GHG Phase 2 program. Many commenters supported limiting the credits in some fashion, such as eliminating credit multipliers for ZEVs produced due to state requirements or phasing out the credit multipliers earlier than MY 2027, which was the last model year that multipliers could be applied under HD GHG Phase 2. Some of the commenters opposed any changes to the existing credit multipliers, indicating that the multipliers are necessary for the development of these new and higher-cost technologies into existing and new markets. We considered the concerns and information provided in these comments when developing this proposal, as discussed in Sections II and III. Commenters who would like EPA to further consider in this rulemaking any relevant comments that they provided on the HD2027 NPRM regarding proposed HD vehicle GHG standards for the MYs at issue in this proposal must resubmit those comments to EPA during this proposal's comment period.

EPA believes the increased application of ZEV technologies in the HD sector presents an opportunity to strengthen GHG standards, which can result in significant reductions in heavy-duty vehicle emissions. Based on an in-depth analysis of the potential for the development and application of ZEV technologies in the HD sector, we are proposing in this Phase 3 NPRM more stringent GHG standards for MYs 2027 through 2032 and later HD vehicles heavy-duty vehicles that are appropriate and feasible considering lead time, costs, and other factors. These proposed Phase 3 standards include (1) revised GHG standards for many MY 2027 HD vehicles, with a subset of standards that would not change, and (2) new GHG standards starting in MYs 2028 through 2032, of which the MY 2032 standards would remain in place for MY 2033 and later. For the purposes of this preamble, we refer to the Phase 3 NPRM standards generally as applying to MYs 2027 through 2032 and later HD vehicles. In this NPRM, we are also requesting comment on setting additional new, progressively more stringent GHG standards beyond the MYs proposed and starting in MYs 2033 through 2035. In consideration of concerns from manufacturers about lead time needed for technology development and market investments, we request comment in this NPRM on an alternative set of GHG standards starting in MYs 2027 through 2032 that are lower than those proposed yet still more stringent than the Phase 2 standards. We also request comment, including supporting data and analysis, if there are certain market segments, such as heavy-haul vocational trucks or long-haul tractors which may require significant energy content for their intended use, for which it may be appropriate to set standards less stringent than the alternative for the specific corresponding regulatory subcategories in order to provide additional lead time to develop and introduce ZEV or other low emissions technology for those specific vehicle applications. In consideration of the environmental impacts of HD vehicles and the need for significant emission reductions, as well as the views expressed by stakeholders such as environmental justice communities, environmental nonprofit organizations, and state and local organizations for rapid and aggressive reductions in GHG emissions, we are also requesting comment on a more stringent set of GHG standards starting in MYs 2027 through 2032 whose values would go beyond the proposed standards, such as values that would reflect the level of ZEV adoption ( i.e., percent of ZEVs in production volumes) used in California's ACT program, values in between these proposed standards and those that would reflect ZEV adoption levels in ACT, and values beyond those that would reflect ZEV adoption levels in ACT, such as the 50–60 percent ZEV adoption range represented by the publicly stated goals of several major OEMs for 2030.

After considering the state of electrification of the HD market, new incentives, and comments received on the HD2027 NPRM regarding credit multipliers, EPA believes that the HD GHG Phase 2 levels of incentives for electrification are no longer appropriate for certain segments of the HD vehicle market. We are proposing in this document to end credit multipliers for BEVs and PHEVs one year earlier than provided in the existing HD GHG Phase 2 program ( i.e., no credit multipliers for BEVs and PHEVs in MYs 2027 and later).

C. Summary of the Major Provisions in the Regulatory Action

Our proposed program features several key provisions that include, based on consideration of updated data and information, updating the existing MY 2027 GHG emission standards and promulgating new GHG emission standards starting in MYs 2028 through 2032 for HD vehicles. Specifically, we are proposing to set progressively more stringent GHG emission standards that would apply to MYs 2027, 2028, 2029, 2030, 2031, and 2032 and later for numerous vocational vehicle and tractor subcategories. The proposed standards for MY 2032 and later are shown in Table ES–1 and Table ES–2 and are described in detail in Section II, while the proposed standards for MYs 2027 through 2031 are shown in Section II.F. As described in Section II of this preamble, our analysis shows that the proposed revisions to HD GHG Phase 2 CO₂ standards for MY 2027 and the proposed new, progressively lower numeric values of the CO₂ standards starting in MYs 2028 through 2032 are appropriate considering feasibility, lead time, costs, and other factors. We seek comment on these proposed Phase 3 standards starting in MYs 2027 through 2032.

The proposed standards do not mandate the use of a specific technology, and EPA anticipates that a compliant fleet under the proposed standards would include a diverse range of technologies ( e.g., transmission technologies, aerodynamic improvements, engine technologies, battery electric powertrains, hydrogen fuel cell powertrains, etc.). The technologies that have played a fundamental role in meeting the Phase 2 GHG standards will continue to play an important role going forward as they remain key to reducing the GHG emissions of HD vehicles powered by internal combustion engines (referred to in this proposal as ICE vehicles). In developing the proposed standards, EPA has also considered the key issues associated with growth in penetration of zero-emission vehicles, including charging infrastructure and hydrogen production. In our assessment that supports the appropriateness and feasibility of these proposed standards, we developed a technology pathway that could be used to meet each of the standards. The technology package includes a mix of ICE vehicles with CO₂ -reducing technologies and ZEVs. EPA developed an analysis tool to evaluate the design features needed to meet the energy and power demands of various HD vehicle types when using ZEV technologies. The overarching analysis is premised on ensuring each of the ZEVs could perform the same work as its ICE counterpart while oversizing the battery to account for its usable range and that batteries deteriorate over time. The fraction of ZEVs in the technology packages are shown in Table ES–3 and described further in Section II of this preamble.

We are requesting comment on an alternative set of CO₂ standards that would more gradually increase in stringency than the proposed standards starting in MY 2027 through 2032, further described in Section II.H. We developed a technology pathway that could be used to meet the alternatives standards, which projects the aggregated ZEV adoption rates shown in Table ES–4 and described further in Section II of this preamble. As described in more detail in Section II, we also are seeking comment on setting GHG standards starting in MYs 2027 through 2032 that would reflect values less stringent than the lower stringency alternative for certain market segments as well as comment on values in between the proposed standards and the alternative standards. Also described in Section II, we are seeking comment on setting GHG standards starting in MYs 2027 through 2032 that would reflect values above the level of the proposed standards. Some of the HD2027 NPRM commenters provided specific recommendations for ZEV adoption rates to include in our analysis, and these adoption rates are on the order of 40 percent or more electrification by MY 2029. The California Air Resources Board's (CARB's) ACT regulation sets ZEV sales requirements for vocational vehicles at 40 percent and for tractors at 25 percent in MY 2029 (Table ES–4). Announcements by major manufacturers project their HD ZEV sales to be in the 50 percent range for 2030 globally, with one manufacturer projecting sales as high as 60 percent for North America in that year. We request comment and data that would support more stringent GHG standards than we are proposing for MYs 2027 through 2032, including comment and data on different technologies' penetration rates than we included in the technology packages described in Section II of the preamble. Specifically, EPA requests comment on values that would reflect the level of ZEV adoption used in California's ACT program, values in between these proposed standards and those that would reflect ZEV adoption levels in ACT, and values beyond those that would reflect ZEV adoption levels in ACT such as the 50–60 percent ZEV adoption range represented by the publicly stated goals of several major OEMs for 2030. We further request comment on promulgating progressively more stringent standards out through MY 2035.

As discussed in Section II and DRIA Chapters 1 and 2, EPA recognizes that charging and refueling infrastructure for BEVs and FCEVs is critically important for the success in the increasing development and adoption of these vehicle technologies. There are significant efforts already underway to develop and expand heavy-duty electric charging and hydrogen refueling infrastructure. The U.S. government is making large investments through the BIL and the IRA, as discussed in more detail in DRIA Chapter 1.3.2. ( e.g., this includes a tax credit for charging or hydrogen refueling infrastructure) as well as billions of additional dollars for programs that could help fund charging infrastructure if purchased alongside an electric vehicle). However, private investments will also play a critical role in meeting future infrastructure needs. We expect many BEV or fleet owners to invest in charging infrastructure for depot charging. (See DRIA Chapter 2.6 for information on our analysis of depot charging needs and costs associated with this proposal.) Manufacturers, charging network providers, energy companies and others are also investing in high-power public or other stations that could support en-route charging. This includes over a billion dollars for recently announced projects to support electric truck or other commercial vehicle charging in the United States and Europe. For example, Daimler Truck North America is partnering with electric power generation company NextEra Energy Resources and BlackRock Renewable Power to collectively invest $650 million to create a nationwide U.S. charging network for commercial vehicles with a later phase of the project also supporting hydrogen fueling stations. Volvo Group and Pilot recently announced their intent to offer public charging for medium- and heavy-duty BEVs at over 750 Pilot and Flying J North American truck stops and travel plazas. (See DRIA Chapter 1.6.2 for a more detailed discussion of private investments in heavy-duty infrastructure.)

These recent heavy-duty charging announcements come during a period of rapid growth in the broader market for charging infrastructure serving cars or other electric vehicles. BloombergNEF estimates that annual global investment was $62 billion in 2022, nearly twice that of the prior year. Private charging companies have already attracted billions globally in venture capital and mergers and acquisitions. In the United States, there was $200 million or more in mergers and acquisition activity in 2022 according to the capital market data provider Pitchbook, indicating strong interest in the future of the charging industry. Domestic manufacturing capacity is also increasing with over $600 million in announced investments to support the production of charging equipment and components at existing or new U.S. facilities.

These important early actions and market indicators suggest strong growth in charging and refueling ZEV infrastructure in the coming years. Furthermore, as described in Section II of this document, our analysis of charging infrastructure needs and costs supports the feasibility of the future growth of ZEV technology of the magnitude EPA is projecting in this proposal's technology package. EPA has heard from some representatives from the heavy-duty vehicle manufacturing industry both optimism regarding the heavy-duty industry's ability to produce ZEV technologies in future years at high volume, but also concern that a slow growth in ZEV charging and refueling infrastructure can slow the growth of heavy-duty ZEV adoption, and that this may present challenges for vehicle manufacturers ability to comply with future EPA GHG standards. Several heavy-duty vehicle manufacturers have encouraged EPA to consider ways to address this concern both in the development of the Phase 3 program, and in the structure of the Phase 3 program itself. EPA requests comment on this concern, both in the Phase 3 rulemaking process, and in consideration of whether EPA should consider undertaking any future actions related to the Phase 3 standards, if finalized, with respect to the future growth of the charging and refueling infrastructure for ZEVs. EPA has a vested interest in monitoring industry's performance in complying with mobile source emission standards, including the highway heavy-duty industry. EPA monitors industry's performance through a range of approaches, including regular meetings with individual companies and regulatory requirements for data submission as part of the annual certification process. EPA also provides transparency to the public through actions such as publishing industry compliance reports (such as has been done during the heavy-duty GHG Phase 1 program). EPA requests comment on what, if any, additional information and data EPA should consider collecting and monitoring during the implementation of the Phase 3 standards; we also request comment on whether there are additional stakeholders EPA should work with during implementation of the Phase 3 standards, if finalized, and what measures EPA should consider to help ensure success of the Phase 3 program, including with respect to the important issues of refueling and charging infrastructure for ZEVs.

As described in Section III.B of this preamble, we are also proposing updates to the advanced technology incentives in the ABT program for HD GHG Phase 2 for electric vehicles. Given the ZEV-related factors outlined in this section and further described in Sections I and II that have arisen since the adoption of HD GHG Phase 2, EPA believes it is appropriate to limit the availability of credit multipliers, but we also recognize the role these credits play in developing new markets. We are proposing in this action to eliminate the advanced technology vehicle credit multipliers for BEVs and PHEVs for MY 2027, one year before these credit multipliers were set to end under the existing HD GHG Phase 2 program. We propose retaining the existing FCEV credit multipliers, because the HD market for this technology continues to be in the early stage of development. We request comment on this approach. In addition to this preamble, we have also prepared a Draft Regulatory Impact Analysis (DRIA) which is available on our website and in the public docket for this rulemaking. The DRIA provides additional data, analysis, and discussion. We request comment on the analysis and data in the DRIA.

D. Impacts of the Proposed Standards

Our estimated emission reductions, average per-vehicle costs, program costs, and monetized benefits of the proposed program are summarized in this section and detailed in Sections IV through VIII of the preamble and Chapters 3 through 8 of the DRIA. EPA notes that, consistent with CAA section 202, in evaluating potential GHG standards, we carefully weigh the statutory factors, including GHG emissions impacts of the GHG standards, and the feasibility of the standards (including cost of compliance in light of available lead time). We monetize benefits of the proposed GHG standards and evaluate other costs in part to better enable a comparison of costs and benefits pursuant to E.O. 12866, but we recognize that there are benefits that we are currently unable to fully quantify. EPA's consistent practice has been to set standards to achieve improved air quality consistent with CAA section 202, and not to rely on cost-benefit calculations, with their uncertainties and limitations, in identifying the appropriate standards. Nonetheless, our conclusion that the estimated benefits considerably exceed the estimated costs of the proposed program reinforces our view that the proposed GHG standards represent an appropriate weighing of the statutory factors and other relevant considerations.

Our analysis of emissions impacts accounts for downstream emissions, i.e., from emission processes such as engine combustion, engine crankcase exhaust, vehicle evaporative emissions, and vehicle refueling emissions. Vehicle technologies would also affect emissions from upstream sources that occur during, for example, electricity generation and the refining and distribution of fuel. This proposal's analyses include emissions impacts from electrical generating units (EGUs). We also account for refinery emission impacts on non-GHG pollutants in these analyses.

The proposed GHG standards would achieve significant reductions in GHG emissions. As seen in Table ES–5, through 2055 the program would result in significant downstream GHG emission reductions. In addition, considering both downstream and EGU cumulative emissions from calendar years 2027 through 2055, the proposed standards would achieve approximately 1.8 billion metric tons in CO₂ emission reductions (see Section V of the preamble and Chapter 4 of the DRIA for more detail). As discussed in Section VI of this preamble, these GHG emission reductions would make an important contribution to efforts to limit climate change and its anticipated impacts. These GHG reductions would benefit all U.S. residents, including populations such as people of color, low-income populations, indigenous peoples, and/or children that may be especially vulnerable to various forms of damages associated with climate change. We project a cumulative increase from calendar years 2027 through 2055 of approximately 0.4 billion metric tons of CO₂ emissions from EGUs as a result of the increased demand for electricity associated with the proposal, although those projected impacts decrease over time because of projected changes in the future power generation mix, including cleaner combustion technologies and increases in renewables.

We expect the proposed GHG emission standards would lead to an increase in HD ZEVs relative to our reference case without the proposed rule, which would also result in reductions of vehicle emissions of non-GHG pollutants that contribute to ambient concentrations of ozone, particulate matter (PM_2.5), NO₂, CO, and air toxics. Exposure to these non-GHG pollutants is linked to adverse human health impacts such as premature death as well as other adverse public health and environmental effects (see Section VI). As shown in Table ES–6, by 2055, when considering downstream, EGU, and refinery emissions, we estimate a net decrease in emissions from all pollutants modeled ( i.e., NO_X, PM_2.5, VOC, and SO₂). In this year alone, the proposed standards would reduce downstream PM_2.5 by approximately 970 U.S. tons (about 39 percent of heavy-duty sector downstream PM_2.5 emissions) and downstream oxides of nitrogen (NO_X) by over 70,000 U.S. tons (about 28 percent of heavy-duty sector downstream NO_X emissions) (see Section V of the preamble and Chapter 4 of the DRIA for more detail). These reductions in non-GHG emissions from vehicles would reduce air pollution near roads. As described in Section VI of this preamble, there is substantial evidence that people who live or attend school near major roadways are more likely to be of a non-White race, Hispanic ethnicity, and/or low socioeconomic status. In addition, emissions from HD vehicles and engines can significantly affect individuals living near truck freight routes. Based on a study EPA conducted of people living near truck routes, an estimated 72 million people live within 200 meters of a truck freight route. Relative to the rest of the population, people of color and those with lower incomes are more likely to live near truck routes. In addition, children who attend school near major roads are disproportionately represented by children of color and children from low-income households.

Similar to GHG emissions, we project that non-GHG emissions from EGUs would increase as a result of the increased demand for electricity associated with the proposal, and we expect those projected impacts to decrease over time due to EGU regulations and changes in the future power generation mix, including impacts of the IRA. We also project that non-GHG emissions from refineries would decrease as a result of the lower demand for liquid fuel associated with the proposed GHG standards (Section V and DRIA Chapter 4).

We estimate that the present value, at 3 percent, of costs to manufacturers would be $9 billion dollars before considering the IRA battery tax credits. With those battery tax credits, which we estimate to be $3.3 billion, the cost to manufacturers of compliance with the program would be $5.7 billion. The manufacturer cost of compliance with the proposed rule on a per-vehicle basis are shown in Table ES–7. We estimate that the MY 2032 fleet average per-vehicle cost to manufacturers by regulatory group would range between a cost savings for LHD vocational vehicles to $2,300 for HHD vocational vehicles and between $8,000 and $11,400 per tractor. EPA notes the projected costs per vehicle for this proposal are similar to the fleet average per-vehicle costs projected for the HD GHG Phase 2 rule, where the tractor standards were projected to cost between $10,200 and $13,700 per vehicle (81 FR 73621 (October 25, 2016)) and the MY 2027 vocational vehicle standards were projected to cost between $1,486 and $5,670 per vehicle (81 FR 73718 (October 25, 2016)). For this proposal, EPA finds that the expected the additional vehicle costs are reasonable in light of the GHG emissions reductions.

The proposed GHG standards would reduce adverse impacts associated with climate change and exposure to non-GHG pollutants and thus would yield significant benefits, including those we can monetize and those we are unable to quantify. Table ES–8 summarizes EPA's estimates of total monetized discounted costs, operational savings, and benefits. The results presented here project the monetized environmental and economic impacts associated with the proposed program during each calendar year through 2055. EPA estimates that the present value of monetized net benefits to society would be approximately $320 billion through the year 2055 (annualized net benefits of $17 billion through 2055), more than 5 times the cost in vehicle technology and associated electric vehicle supply equipment (EVSE) combined. Regarding social costs, EPA estimates that the cost of vehicle technology (not including the vehicle or battery tax credits) and EVSE would be approximately $9 billion and $47 billion respectively, and that the HD industry would save approximately $250 billion in operating costs ( e.g., savings that come from less liquid fuel used, lower maintenance and repair costs for ZEV technologies as compared to ICE technologies, etc.). The program would result in significant social benefits including $87 billion in climate benefits (with the average SC–GHGs at a 3 percent discount rate). Between $15 and $29 billion of the estimated total benefits through 2055 are attributable to reduced emissions of non-GHG pollutants, primarily those that contribute to ambient concentrations of PM_2.5. Finally, the benefits due to reductions in energy security externalities caused by U.S. petroleum consumption and imports would be approximately $12 billion under the proposed program. A more detailed description and breakdown of these benefits can be found in Section VIII of the preamble and Chapter 7 of the DRIA.

Regarding the costs to purchasers as shown in Table ES–9, for the proposed program we estimated the average upfront incremental cost to purchase a new MY 2032 HD BEV or FCEV relative to an ICE vehicle for a vocational BEV and EVSE, a short-haul tractor BEV and EVSE, a short-haul tractor FCEV, and a long-haul tractor FCEV. These incremental costs account for the IRA tax credits, specifically battery and vehicle tax credits, as discussed in Section II.E.4 and Section IV.C and IV.D. We also estimated the operational savings each year ( i.e., savings that come from the lower costs to operate, maintain, and repair BEV technologies) and payback period ( i.e., the year the initial cost increase would pay back). Table ES–9 shows that for the vocational vehicle ZEVs, short-haul tractor ZEVs, and long-haul tractor FCEVs the incremental upfront costs (after the tax credits) are recovered through operational savings such that pay back occurs after between one and three years on average for vocational vehicles, after three years for short-haul tractors and after seven years on average for long-haul tractors. We discuss this in more detail in Sections II and IV of this preamble and DRIA Chapters 2 and 3.

I. Introduction

A. Brief Overview of the Heavy-Duty Industry

Heavy-duty highway vehicles range from commercial pickup trucks to vocational vehicles that support local and regional transportation, construction, refuse collection, and delivery work, to line-haul tractors (semi trucks) that move freight cross-country. This diverse array of vehicles is categorized into weight classes based on gross vehicle weight ratings (GVWR). These weight classes span Class 2b pickup trucks and vans from 8,500 to 10,000 pounds GVWR through Class 8 line-haul tractors and other commercial vehicles that exceed 33,000 pounds GVWR. While Class 2b and 3 complete pickups and vans are not included in this proposed rulemaking, Class 2b and 3 vocational vehicles are included in this rulemaking (as discussed further in Section III.E.3).

Heavy-duty highway vehicles are powered through an array of different means. Currently, the HD vehicle fleet is primarily powered by diesel-fueled, compression-ignition (CI) engines. However, gasoline-fueled, spark-ignition (SI) engines are common in the lighter weight classes, and smaller numbers of alternative fuel engines ( e.g., liquified petroleum gas, compressed natural gas) are found in the heavy-duty fleet. We refer to the vehicles powered by internal combustion engines (ICE, including SI and CI engines) as ICE vehicles throughout this preamble. An increasing number of HD vehicles are powered by zero emission vehicle (ZEV) technologies such as battery electric vehicle (BEV) technology, e.g., EPA certified 380 HD BEVs in MY 2020 but that number jumped to 1,163 HD BEVs in MY 2021. We use the term ZEV technologies throughout the preamble to refer to technologies that result in zero tailpipe emissions, which in this preamble we refer to collectively as ZEVs. Example ZEV technologies include BEVs and fuel cell vehicles (FCEVs). While hybrid vehicles (including plug-in hybrid electric vehicles) include energy storage features such as batteries, they also include an ICE, which do not result in zero tailpipe emissions.

The industry that designs and manufactures HD vehicles is composed of three primary segments: vehicle manufacturers, engine manufacturers and other major component manufacturers, and secondary manufacturers ( i.e., body builders). Some vehicle manufacturers are vertically integrated—designing, developing, and testing their engines in-house for use in their vehicles; others purchase some or all of their engines from independent engine suppliers. At the time of this proposal, only one major independent engine manufacturer supports the HD industry, though some vehicle manufacturers sell their engines or “incomplete vehicles” ( i.e., chassis that include their engines, the frame, and a transmission) to body builders who design and assemble the final vehicle. Each of these subindustries is often supported by common suppliers for subsystems such as transmissions, axles, engine controls, and emission controls.

In addition to the manufacturers and suppliers responsible for producing HD vehicles, an extended network of dealerships, repair and service facilities, and rebuilding facilities contribute to the sale, maintenance, and extended life of these vehicles and engines. HD vehicle dealerships offer customers a place to order such vehicles from a specific manufacturer and often include service facilities for those vehicles and their engines. Dealership service technicians are generally trained to perform regular maintenance and make repairs, which generally include repairs under warranty and in response to manufacturer recalls. Some trucking fleets, businesses, and large municipalities hire their own technicians to service their vehicles in their own facilities. Many refueling centers along major trucking routes have also expanded their facilities to include roadside assistance and service stations to diagnose and repair common problems.

The end-users for HD vehicles are as diverse as the applications for which these vehicles are purchased. Smaller weight class HD vehicles are commonly purchased by delivery services, contractors, and municipalities. The middle weight class vehicles tend to be used as commercial vehicles for business purposes and municipal work that transport people and goods locally and regionally or provide services such as utilities. Vehicles in the heaviest weight classes are generally purchased by businesses with high load demands, such as construction, towing or refuse collection, or freight delivery fleets and owner-operators for regional and long-haul goods movement. The competitive nature of the businesses and owner-operators that purchase and operate HD vehicles means that any time at which the vehicle is unable to operate due to maintenance or repair ( i.e., downtime) can lead to a loss in income. The customers' need for reliability drives much of the vehicle manufacturers innovation and research efforts.

B. History of Greenhouse Gas Emission Standards for Heavy-Duty Engines and Vehicles

EPA has a longstanding practice of regulating GHG emissions from the HD sector. In 2009, EPA and the U.S. Department of Transportation's (DOT's) National Highway Traffic Safety Administration (NHTSA) began working on a joint regulatory program to reduce GHG emissions and fuel consumption from HD vehicles and engines. The first phase of the HD GHG and fuel efficiency program was finalized in 2011 (76 FR 57106, September 15, 2011) (“HD GHG Phase 1”). The HD GHG Phase 1 program largely adopted approaches consistent with recommendations from the National Academy of Sciences. The HD GHG Phase 1 program, which began in MY 2014 and phased in through MY 2018, included separate standards for HD vehicles and HD engines. The program offered flexibility allowing manufacturers to attain these standards through a mix of technologies and the option to participate in an emissions credit ABT program.

In 2016, EPA and NHTSA finalized the HD GHG Phase 2 program. The HD GHG Phase 2 program included technology-advancing, performance-based emission standards for HD vehicles and HD engines that phase in over the long term, with initial standards for most vehicles and engines commencing in MY 2021, increasing in stringency in MY 2024, and culminating in even more stringent MY 2027 standards. HD GHG Phase 2 built upon the Phase 1 program and set standards based not only on then-currently available technologies, but also on technologies that were either still under development or not yet widely deployed at the time of the HD GHG Phase 2 final rule. To ensure adequate time for technology development, HD GHG Phase 2 provided up to 10 years lead time to allow for the development and phase-in of these control technologies. EPA recently finalized technical amendments to the HD GHG Phase 2 rulemaking (“HD Technical Amendments”) that included changes to the test procedures for heavy-duty engines and vehicles to improve accuracy and reduce testing burden.

As with the previous HD GHG Phase 1 and Phase 2 rules and light-duty GHG rules, EPA has coordinated with the DOT and NHTSA during the development of this proposed rule. This included coordination prior to and during the interagency review conducted under E.O. 12866. EPA has also consulted with CARB during the development of this proposal, as EPA also did during the development of the HD GHG Phase 1 and 2 and light-duty rules. See Section I.E for additional detail on EPA's coordination with DOT/NHTSA, CARB, and additional Federal Agencies.

C. What has changed since we finalized the HD GHG Phase 2 rule?

In 2016, we established the HD GHG Phase 2 CO₂ standards on the premise that zero-emission technologies would not be available and cost-competitive in significant volumes in the timeframe of the HD GHG Phase 2 program but would become more widely available in the HD market over time. To encourage that availability at faster pace, we finalized BEV, PHEV, and FCEV advanced technology credit multipliers for HD vehicles. As described in the Executive Summary and Section II of this preamble, we have considered new data and recent policy changes and we are now projecting that ZEV technologies will be readily available and technologically feasible much sooner than we had projected. We list the developments pointing to this increased application of ZEV technologies again in the following paragraphs (and we discuss their impacts on the HD market in more detail in the Sections I.C.1 through I.C.3):

First, the HD market has evolved such that early ZEV models are in use today for some applications and are expected to expand to many more applications, ZEV technologies costs have gone down and are projected to continue to fall, and manufacturers have announced plans to rapidly increase their investments in ZEV technologies over the next decade. For example, in 2022, several manufacturers are producing fully electric HD vehicles in several applications, and these applications are expected to expand (see Section I.C.1 and DRIA Chapter 1). Furthermore, several HD manufacturers have announced their ZEV projections that signify a rapid increase in BEVs over the next decade. This increase in HD ZEVs is in part due to the significant decrease in cost to manufacture lithium-ion batteries, the single most expensive component of a BEV, in the past decade; those costs are projected to continue to fall during this decade, all while the performance of these batteries in terms of energy density has improved and is projected to continue to improve. Many of the manufacturers who produce HD vehicles and firms that purchase HD vehicles have announced billions of dollars' worth of investments in ZEV technologies and significant plans to transition to a zero-carbon fleet over the next ten to fifteen years.

Second, the 2021 BIL and the 2022 IRA laws have been enacted, and together these two laws provide significant and unprecedented monetary incentives for the production and purchase of ZEVs in the HD market, as well as incentives for electric vehicle charging and hydrogen, which will further support a rapid increase in market penetration of ZEVs.

Third, there have been multiple actions by states to accelerate the adoption of HD ZEVs. The State of California and other states have adopted the ACT program that includes a manufacturer requirement for zero-emission truck sales. The ACT program provides that “manufacturers who certify Class 2b-8 chassis or complete vehicles with combustion engines would be required to sell zero-emission trucks as an increasing percentage of their annual [state] sales from 2024 to 2035.” In addition, 17 states and the District of Columbia have signed a Memorandum of Understanding establishing goals to support widespread electrification of the HD vehicle market.

We note that the improvements in internal combustion engine technologies that began under the HD GHG Phase 1 program and are being advanced under the HD GHG Phase 2 standards are still necessary for reducing GHG emissions from the HD sector. As we discuss in Section II.D.1, these technology improvements exist today and we believe they will continue to be feasible during the timeframe at issue in this proposed rulemaking.

1. The HD Zero-Emission Vehicle Market

Since 2012, manufacturers have developed a number of prototype and demonstration HD BEV projects, particularly in the State of California, establishing technological feasibility and durability of BEV technology for specific applications used for specific services, as well as building out necessary infrastructure. In 2019, approximately 60 makes and models of HD BEVs were available for purchase, with additional product lines in prototype or other early development stages. According to the Global Commercial Vehicle Drive to Zero Zero-Emission Technology Inventory (ZETI), 160 BEV models were commercially available on the market in the United States and Canada region in 2021, and around 200 BEV models are projected to be available by 2024. DRIA Chapter 1 provides a snapshot of BEV models in the HD vehicle market.

Current production volumes of HD BEVs originally started increasing in the transit bus market, where electric bus sales grew from 300 to 650 in the United States between 2018 to 2019. In 2020, the market continued to expand beyond transit, with approximately 900 HD BEVs sold in the United States and Canada combined, consisting of transit buses (54 percent), school buses (33 percent), and straight trucks (13 percent). By 2021, M.J. Bradley's analysis of the HD BEV market found that 30 manufacturers had at least one BEV model for sale and an additional nine companies had made announcements to begin BEV production by 2025. In April 2022, the Environmental Defense Fund (EDF) projected deployments and major orders of electric trucks and buses in the United States to rise to 54,000 by 2025 based on an analysis of formal statements and announcements by auto manufacturers, as well as analysis of the automotive press and data from financial and market analysis firms that regularly cover the auto industry. Given the dynamic nature of the BEV market, the number and types of vehicles available are increasing fairly rapidly.

The current market for HD FCEVs is not as developed as the market for HD BEVs, but models are being designed, tested, and readied for purchase in the coming years. According to ZETI, at least 16 HD FCEV models are expected to become commercially available for production in the United States and Canada region by 2024, as listed in DRIA Chapter 1. The Hydrogen Fuel Cell Partnership reports that fuel cell electric buses have been in commercial development for 20 years and, as of May 2020, over 100 buses are in operation or in planning in the United States. Foothill Transit in Los Angeles County ordered 33 transit buses that they expect to be operating in early 2023. Ten Toyota-Kenworth Class 8 fuel cell tractors were successfully tested in the Port of Los Angeles and surrounding area through 2022. Hyundai is scheduled to test 30 Class 8 tractors in the Port of Oakland in 2023. Nikola has agreements with fleets to purchase or lease over 200 Class 8 trucks upon satisfactory completion of demonstrations and is building a manufacturing facility in Coolidge, Arizona, with an expected production capacity of up to 20,000 BEV and FCEV trucks by the end of 2023.

For this proposed rulemaking, EPA conducted an analysis of manufacturer-supplied end-of-year production reports provided to us as a requirement of the process to certify HD vehicles to our GHG emission standards. Based on the end-of-year production reports for MY 2019, manufacturers produced approximately 350 certified HD BEVs. This is out of nearly 615,000 HD diesel ICE vehicles produced in MY 2019 and represents approximately 0.06 percent of the HD vehicles market. In MY 2020, 380 HD BEVs were certified, an increase of 30 BEVs from 2019. The BEVs were certified in a variety of the Phase 1 vehicle subcategories, including light, medium, and heavy heavy-duty vocational vehicles and vocational tractors. Out of the 380 HD BEVs certified in MY 2020, a total of 177 unique makes and models were available for purchase by 52 manufacturers in Classes 3–8. In MY 2021, EPA certified 1,163 heavy-duty BEVs, representing 0.2 percent of the HD vehicles. There were no HD FCEVs certified through MY 2021. We note that these HD BEV certifications preceded implementation of incentives in the 2022 IRA, which we expect to increase adoption (and certification) of BEV and FCEV technology in the heavy-duty sector.

Based on current trends, manufacturer announcements, the 2021 BIL and 2022 IRA, and state-level actions, electrification of the HD market is expected to substantially increase over the next decade from current levels. The projected rate of growth in electrification of the HD vehicle sector currently varies widely. After passage of the IRA, EDF's September 2022 report update projected deployments and major orders of electric trucks and buses to rise to 166,000 by the end of 2022. ERM updated an analysis for EDF that projected five scenarios that span a range of between 13 and 48 percent Class 4–8 ZEV sales in 2029, with an average of 29 percent. The International Council for Clean Transportation (ICCT) and Energy Innovation conducted an analysis of the impact of the IRA on electric vehicle uptake, projecting between 39 and 48 percent Class 4–8 ZEV sales in 2030 across three scenarios and between 47 and 56 percent in 2035.

One of the most important factors influencing the extent to which BEVs are available for purchase and able to enter the market is the cost of lithium-ion batteries, the single most expensive component of a BEV. According to Bloomberg New Energy Finance, average lithium-ion battery costs have decreased by more than 85 percent since 2010, primarily due to global investments in battery production and ongoing improvements in battery technology. A number of studies, including the Sharpe and Basma meta-study of direct manufacturing costs from a variety of papers, show that battery pack costs are projected to continue to fall during this decade. Cost reductions in battery packs for electric trucks are anticipated due to continued improvement of cell and battery pack performance and advancements in technology associated with energy density, materials for cells, and battery packaging and integration.

Currently, the fuel cell stack is the most expensive component of a HD FCEV, due primarily to the technological requirements of manufacturing rather than raw material costs. Projected costs are expected to decrease as manufacturing matures and materials improve. Larger production volumes are anticipated as global demand increases for fuel cell systems for HD vehicles, which would improve economies of scale. Costs of the onboard hydrogen storage tank, another component unique to a FCEV, are also projected to drop due to lighter weight and lower cost carbon fiber-reinforced materials, technology improvements, and economies of scale.

As the cost of components has come down, manufacturers have increasingly announced their projections for zero-emission HD vehicles, and these projections signify a rapid increase in BEVs and FCEVs over the next decade. For example, Volvo Trucks and Scania announced a global electrification target of 50 percent of trucks sold being electric by 2030. Daimler Trucks North America has committed to offering only what they refer to as “carbon-neutral” trucks in the United States. by 2039 and expects that by 2030 as much as 60 percent of its sales will be ZEVs. Navistar has a goal of having 50 percent of its sales volume be ZEVs by 2030, and it has committed to achieve 100 percent zero emissions by 2040. Cummins targets net-zero carbon emissions by 2050.

On a parallel path, large private HD fleet owners are also increasingly committing to expanding their electric fleets. A report by the International Energy Agency (IEA) provides a comprehensive accounting of recent announcements made by UPS, FedEx, DHL, Walmart, Anheuser-Busch, Amazon, and PepsiCo for fleet electrification. Amazon and UPS, for example, placed orders in 2020 for 10,000 BEV delivery vans from EV start-ups Rivian and Arrival, respectively, and Amazon has plans to scale up to 100,000 BEV vans by 2030.^{122 123} Likewise, in December 2022, PepsiCo added the first of 100 planned Tesla Semis to its fleet. These announcements include not only orders for electric delivery vans and semi-trucks, but more specific targets and dates to full electrification or net-zero emissions. Amazon, FedEx, DHL, and Walmart have set a commitment to fleet electrification and/or achieving net-zero emissions by 2040. We recognize that certain delivery vans will likely fall into the Class 2b and 3 regulatory category, the vast majority of which are not covered in this rule's proposed updates; we intend to address this category in a separate light and medium-duty vehicle rulemaking.

Amazon and Walmart are among fleets owners and operators that are also considering hydrogen. Amazon signed an agreement with Plug Power, a company building an end-to-end hydrogen ecosystem, to supply hydrogen for up to 800 HD long-haul trucks or 30,000 forklifts (which are commonly powered using hydrogen) starting in 2025 through 2040. Walmart is purchasing hydrogen from Plug Power and plans to expand pilots of fuel cell forklifts, yard trucks, and possibly HD long-haul trucks by 2040. Plug Power has agreed to purchase up to 75 Nikola Class 8 fuel cell trucks over the next three years in exchange for supplying the company with hydrogen fuel.

The lifetime total cost of ownership (TCO), which includes maintenance and fuel costs, is likely a primary factor for HD vehicle and fleet owners considering BEV and FCEV purchases. In fact, a 2018 survey of fleet owners showed “lower cost of ownership” as the second most important motivator for electrifying their fleet. An ICCT analysis from 2019 suggests that TCO for light and medium heavy-duty BEVs could reach cost parity with comparable diesel ICE vehicles in the early 2020s, while heavy HD BEVs and FCEVs are likely to reach cost parity with comparable diesel ICE vehicles closer to the 2030 timeframe. Recent findings from Phadke et al. suggest that BEV TCO could be 13 percent less than that of a comparable diesel ICE vehicle if electricity pricing is optimized. These studies do not consider the IRA. The Rocky Mountain Institute found that because of the IRA, the TCO of electric trucks will be lower than the TCO of comparable diesel trucks about five years faster than without the IRA. They expect cost parity as soon as 2023 for urban and regional duty cycles that travel up to 250 miles and 2027 for long-hauls that travel over 250 miles.

As the ICCT and Phadke et al. studies suggest, fuel costs are an important part of TCO. While assumptions about vehicle weight and size can make direct comparisons between HD ZEVs and ICE vehicles challenging, data show greater energy efficiency of battery-electric and fuel cell technology relative to ICE technologies. Better energy efficiency leads to lower electricity or hydrogen fuel costs for ZEVs relative to ICE fuel costs. Maintenance and service costs are also an important component within TCO; although there is limited data available on actual maintenance costs for HD ZEVs, early experience with BEV medium HD vehicles and transit buses suggests the potential for lower maintenance costs after an initial period of learning to refine both component durability and maintenance procedures. We expect similar trends for FCEVs, as discussed in Chapter 2 of the DRIA. To facilitate HD fleets transitioning to ZEVs, some manufacturers are currently including maintenance in leasing agreements with fleets; it is unclear the extent to which a full-service leasing model will persist or will be transitioned to a more traditional purchase model after an initial period of learning.

The growth in incentive programs will continue to play an important role in the HD ZEV market. For example, as discussed in more detail in this section, FHWA-approved plans providing $1.5 billion in funding for expanding charging on over 75,000 miles of highway encourages states to consider station designs and power levels that could support heavy-duty vehicles. In a 2017 survey of fleet managers, upfront purchase price was listed as the primary barrier to HD fleet electrification. This suggests that federal incentive programs like those in the BIL and IRA (discussed in Section I.C.2) to offset ZEV purchase costs, as well as state and local incentives and investments, can be influential in the near term, with improvements in BEV and FCEV component costs playing an increasing role in reducing costs in the longer term. For example, BEV incentive programs for transit and school buses have experienced growth and are projected to continue to influence BEV markets. The Los Angeles Department of Transportation (LADOT) is one of the first transit organizations in the country to develop a program committed to transitioning its transit fleets to ZEVs by 2030—a target that is 10 years sooner than CARB's Innovative Clean Transportation (ICT) regulation requiring all public transit to be electric by 2040. Since these announcements, LADOT has purchased 27 BEV transit and school buses from BYD and Proterra; by 2030, the number of BEV buses in the LADOT fleet is expected to grow to 492 buses. Outside of California, major metropolitan areas including Chicago, Seattle, New York City, and Washington, DC, have zero-emissions transit programs with 100 percent ZEV target dates ranging from 2040 to 2045. EV school bus programs, frequently in partnership with local utilities, are also being piloted across the country and are expanding under EPA's Clean School Bus Program (CSB). These programs initially included school districts in, but not limited to, California, Virginia, Massachusetts, Michigan, Maryland, Illinois, New York, and Pennsylvania. Going forward, they will continue to expand with BIL funding of over $5 billion over the next five years (FY 2022–2026) to replace existing school buses with zero-emission and low-emission models, as discussed more in Section I.C.2.

In summary, the HD ZEV market is growing rapidly, and ZEV technologies are expected to expand to many applications across the HD sector. As the industry is dynamic and changing rapidly, the examples presented here represent only a sampling of the ZEV HD investment policies and markets. DRIA Chapter 1 provides a more detailed characterization of the HD ZEV technologies in the current and projected ZEV market. We request comment on our assessment of the HD ZEV market and any additional data sources we should consider.

2. Bipartisan Infrastructure Law and Inflation Reduction Act

i. BIL

The BIL was enacted on November 15, 2021, and contains provisions to support the deployment of low- and zero-emission transit buses, school buses, and trucks that service ports, as well as electric vehicle charging infrastructure and hydrogen. These provisions include Section 71101 funding for EPA's Clean School Bus Program, with $5 billion to fund the replacement of ICE school buses with clean and zero-emission buses over the next five years. In its first phase of funding for the Clean School Bus Program, EPA is issuing nearly $1 billion in rebates (up to a maximum of $375,000 per bus, depending on the bus fuel type, bus size, and school district prioritization status) for replacement clean and zero-emission buses and associated infrastructure costs. The BIL also includes funding for DOT's Federal Transit Administration (FTA) Low- or No-Emission Grant Program, with over $5.6 billion over the next five years to support the purchase of zero- or low-emission transit buses and associated infrastructure.

The BIL includes up to $7.5 billion to help build out a national network of EV charging and hydrogen fueling through DOT's Federal Highway Administration (FHWA). This includes $2.5 billion in discretionary grant programs for charging and fueling infrastructure along designated alternative fuel corridors and in communities (Section 11401) and $5 billion for the National Electric Vehicle Infrastructure (NEVI) Formula Program (under Division J, Title VIII). In September 2022, the FHWA approved the first set of plans for the NEVI program covering all 50 states, Washington, DC, and Puerto Rico. The approved plans provide $1.5 billion in funding for fiscal years (FY) 2022 and 2023 to expand charging on over 75,000 miles of highway. While jurisdictions are not required to build stations specifically for heavy-duty vehicles, FHWA's guidance encourages states to consider station designs and power levels that could support heavy-duty vehicles.

The BIL funds other programs that could support HD vehicle electrification. For example, there is continued funding of the Congestion Mitigation and Air Quality (CMAQ) Improvement Program, with more than $2.5 billion authorized for FY 2022 through FY 2026. The BIL (Section 11115) amended the CMAQ Improvement Program to add, among other things, “the purchase of medium- or heavy-duty zero emission vehicles and related charging equipment” to the list of activities eligible for funding. The BIL establishes a program under Section 11402 “Reduction of Truck Emissions at Port Facilities” that includes grants to be administered through FHWA aimed at reducing port emissions, including through electrification. In addition, the BIL includes funding for DOT's Maritime Administration (MARAD) Port Infrastructure Development Program; and DOT's Federal Highway Administration (FHWA) Carbon Reduction Program.

The BIL also targets batteries used for electric vehicles. It funds DOE's Battery Materials Processing and Battery Manufacturing program, which grants funds to promote U.S. processing and manufacturing of batteries for automotive and electric grid use through demonstration projects, the construction of new facilities, and the retooling, retrofitting, and expansion of existing facilities. This includes a total of $3 billion for battery material processing and $3 billion for battery manufacturing and recycling, with additional funding for a lithium-ion battery recycling prize competition, research and development activities in battery recycling, state and local programs, and the development of a collection system for used batteries. In addition, the BIL includes $200 million for the Electric Drive Vehicle Battery Recycling and Second-Life Application Program for research, development, and demonstration of battery recycling and second-life applications.

Hydrogen provisions of the BIL include funding for several programs to accelerate progress towards the Hydrogen Shot goal, launched on June 7, 2021, to reduce the cost of clean hydrogen production by 80 percent to $1 for 1 kg in 1 decade and jumpstart the hydrogen market in the United States. This includes $8 billion for the Department of Energy's Regional Clean Hydrogen Hubs Program to establish networks of clean hydrogen producers, potential consumers, and connective infrastructure in close proximity; $1 billion for a Clean Hydrogen Electrolysis Program; and $500 million for Clean Hydrogen Manufacturing and Recycling Initiatives. The BIL also called for development of a Clean Hydrogen Production Standard to guide DOE hub and Research, Development, Deployment, and Diffusion (RDD&D) actions; and a National Clean Hydrogen Strategy and Roadmap to facilitate widescale production, processing, delivery, storage, and use of clean hydrogen. These BIL programs are currently under development, and further details are expected over the course of calendar year (CY) 2023.

ii. IRA Sections 13502 and 13403

The IRA, which was enacted on August 16, 2022, contains several provisions relevant to vehicle electrification and the associated infrastructure via tax credits, grants, rebates, and loans through CY 2032, including two key provisions that provide a tax credit to reduce the cost of producing qualified batteries (battery tax credit) and to reduce the cost of purchasing qualified ZEVs (vehicle tax credit). The battery tax credit in “Advanced Manufacturing Production Credit” in IRA section 13502 and the “Qualified Commercial Clean Vehicles” vehicle tax credit in IRA section 13403 are included quantitatively in our analysis.

IRA section 13502, “Advanced Manufacturing Production Credit,” provides tax credits for the production and sale of battery cells and modules of up to $45 per kilowatt-hour (kWh), and for 10 percent of the cost of producing applicable critical minerals (including those found in batteries and fuel cells, provided that the minerals meet certain specifications), when such components or minerals are produced in the United States. These credits begin in CY 2023 and phase down starting in CY 2030, ending after CY 2032. With projected direct manufacturing costs for heavy- duty vehicle batteries on the order of $65 to $275/kWh in the 2025–2030 timeframe, this tax credit has the potential to noticeably reduce the cost of qualifying batteries and, by extension, the cost of BEVs and FCEVs with qualifying batteries. We did not include a detailed cost breakdown of fuel cells quantitatively in our analysis, but the potential impact on fuel cells may also be significant because platinum (an applicable critical mineral commonly used in fuel cells) is a major contributor to the cost of fuel cells.

We limited our assessment of this tax credit in our DRIA Chapter 2 analysis to the tax credits for battery cells and modules. Pursuant to the IRA, qualifying battery cells must have an energy density of not less than 100 watt-hours per liter, and we expect that batteries for heavy-duty BEVs and FCEVs will exceed this requirement as described in DRIA Chapter 2.4.2.2. Qualifying battery cells must be capable of storing at least 12 watt-hours of energy and qualifying battery modules must have an aggregate capacity of not less than 7 kWh (or, for FCEVs, not less than 1 kWh); typical battery cells and modules for motor vehicles also exceed these requirements. Additionally, the ratio of the capacity of qualifying cells and modules to their maximum discharge amount shall not exceed 100:1. We expect that battery cells and modules in heavy-duty BEVs and FCEVs will also meet this requirement because the high costs and weight of the batteries and the competitiveness of the heavy-duty industry will pressure manufacturers to allow as much of their batteries to be useable as possible. We did not consider the tax credits for critical minerals quantitatively in our analysis. However, we note that any applicability of the critical mineral tax credit may further reduce the costs of batteries.

We included this battery tax credit by reducing the direct manufacturing costs of batteries in BEVs and FCEVs, but not the associated indirect costs. At present, there are few manufacturing plants for HD vehicle batteries in the United States, which means that few batteries would qualify for the tax credit now. We expect that the industry will respond to this tax credit incentive by building more domestic battery manufacturing capacity in the coming years, but this will take several years to come to fruition. Thus, we have chosen to model this tax credit by assuming that HD BEV and FCEV manufacturers fully utilize the module tax credit (which provides $10 per kWh) and gradually increase their utilization of the cell tax credit (which provides $35 per kWh) for MY 2027–2029 until MY 2030 and beyond, when they earn 100 percent of the available cell and module tax credits. Further discussion of this battery tax credit and our battery costs can be found in DRIA Chapter 2.4.3.1.

IRA section 13403, “Qualified Commercial Clean Vehicles,” creates a tax credit of up to $40,000 per Class 4 through 8 HD vehicle (up to $7,500 per Class 2b or 3 vehicle) for the purchase or lease of a qualified commercial clean vehicle. This tax credit is available from CY 2023 through CY 2032 and is based on the lesser of the incremental cost of the clean vehicle over a comparable ICE vehicle or the specified percentage of the basis of the clean vehicle, up to the maximum applicable limitation. By effectively reducing the price a vehicle owner must pay for a HD ZEV and the incremental difference in cost between it and a comparable ICE vehicle—by $40,000 in many cases—more vehicle purchasers will be poised to take advantage of the cost savings anticipated from total cost of ownership, including operational cost savings from fuel and maintenance and repair compared with ICE vehicles. Among other specifications, these vehicles must be on-road vehicles (or mobile machinery) that are propelled to a significant extent by a battery-powered electric motor or are qualified fuel cell motor vehicles (also known as fuel cell electric vehicles, FCEVs). For the former, the battery must have a capacity of at least 15 kWh (or 7 kWh if it has a gross vehicle weight rating of less than 14,000 pounds (Class 3 or below)) and must be rechargeable from an external source of electricity. This limits the qualified vehicles to BEVs and plug-in hybrid electric vehicles (PHEVs), in addition to FCEVs. Since this tax credit overlaps with the model years for which we are proposing standards (MYs 2027 through 2032), we included it in our calculations for each of those years in our feasibility analysis for our proposed standards (see DRIA Chapter 2).

For BEVs and FCEVs, the per-vehicle tax credit is equal to the lesser of the following, up to the cap limitation: (A) 30 percent of the BEV or FCEV cost, or (B) the incremental cost of the BEV or FCEV when compared to a comparable (in size and use) ICE vehicle. The limitation on this tax credit is $40,000 for vehicles with a gross vehicle weight rating of equal to or greater than 14,000 pounds (Class 4–8 commercial vehicles) and $7,500 for vehicles with a gross vehicle weight rating of less than 14,000 pounds (commercial vehicles Class 3 and below). For example, if a BEV with a gross vehicle weight rating of equal to or greater than 14,000 pounds costs $350,000 and a comparable ICE vehicle costs $150,000, the tax credit would be the lesser of the following, subject to the limitation: (A) 30 percent × $350,000 = $105,000 or (B) $350,000−$150,000 = $200,000. (A) is less than (B), but (A) exceeds the limit of $40,000, so the tax credit would be $40,000. For PHEVs, the per-vehicle tax credit follows the same calculation and cap limitation as for BEVs and FCEVs except that (A) is 15 percent of the PHEV cost.

In order to estimate the impact of this tax credit in our feasibility analysis for BEVs and FCEVs, we first applied a retail price equivalent to our direct manufacturing costs for BEVs, FCEVs, and ICE vehicles. Note that the direct manufacturing costs of BEVs and FCEVs were reduced by the amount of the battery tax credit in IRA section 13502, as described in DRIA Chapter 2.4.3.1. We calculated the purchaser's incremental cost of BEVs and FCEVs compared to ICE vehicles and not the full cost of vehicles in our analysis. We based our calculation of the tax credit on this incremental cost. When the incremental cost exceeded the tax credit limitation (determined by gross vehicle weight rating as described in the previous paragraph), we decreased the incremental cost by the tax credit limitation. When the incremental cost was between $0 and the tax credit limitation, we reduced the incremental cost to $0 ( i.e., the tax credit received by the purchaser was equal to the incremental cost). When the incremental cost was negative ( i.e., the BEV or FCEV was cheaper to purchase than the ICE vehicle), no tax credit was given. In order for this calculation to be appropriate, we determined that all Class 4–8 BEVs and FCEVs must cost more than $133,333 such that 30 percent of the cost is at least $40,000 (or $25,000 and $7,500, respectively, for BEVs and FCEVs Class 3 and below), which is reasonable based on our review of the literature on the costs of BEVs and FCEVs. The tax credit amounts for each vehicle type included in our analysis in MYs 2027 and 2032 are shown in DRIA Chapter 2.8.2.

We project that the impact of the IRA vehicle tax credit will be significant, as shown in DRIA Chapter 2.8.2. In many cases, the incremental cost (with the tax credit) of a BEV compared to an ICE vehicle is eliminated, leaving only the cost of the electric vehicle supply equipment (EVSE) as an added upfront cost to the BEV owner. Similarly, in some cases, the tax credit eliminates the upfront cost of a FCEV compared to an ICE vehicle.

iii. Other IRA Provisions

There are many other provisions of the IRA that we expect will support electrification of the heavy-duty fleet. Importantly, these other provisions do not serve to reduce ZEV adoption rates from our current projections. Due to the complexity of analyzing the combined potential impact of these provisions, we did not quantify their potential impact in our assessment of costs and feasibility, but we note that they may help to reduce many obstacles to electrification of HDVs and may further support or even increase ZEV adoption rates beyond the levels we currently project. Our assessment of the impacts of these provisions of the IRA on ZEV adoption rates are, therefore, somewhat conservative.

Section 13404, “Alternative Fuel Refueling Property Credit,” modifies an existing tax credit that applies to alternative fuel refueling property ( e.g., electric vehicle chargers and hydrogen fueling stations) and extends the tax credit through CY 2032. The credit also applies to refueling property that stores or dispenses specified clean-burning fuels, including at least 85 percent hydrogen, into the fuel tank of a motor vehicle. Starting in CY 2023, this provision provides a tax credit of up to 30 percent of the cost of the qualified alternative fuel refueling property ( e.g., HD BEV charger), and up to $100,000 when located in low-income or non-urban area census tracts and certain other requirements are met. We expect that many HD BEV owners will need chargers installed in their depots for overnight charging, and this tax credit will effectively reduce the costs of installing charging infrastructure and, in turn, further effectively reduce the total costs associated with owning a BEV for many HD vehicle owners. Additionally, this tax credit may offset some of the costs of installing very high-powered public and private chargers that are necessary to recharge HD BEVs with minimal downtime during the day. Similarly, we expect that this tax credit will reduce the costs associated with refueling heavy-duty FCEVs, whose owners may rely on public hydrogen refueling stations or those installed in their depots. We expect that this tax credit will help incentivize the build out of the charging and hydrogen refueling infrastructure necessary for high BEV and FCEV adoption, which may further support increased BEV and FCEV uptake.

Section 60101, “Clean Heavy-duty Vehicles,” amends the CAA to add new section 132 (42 U.S.C. 7432) and appropriates $1 billion to the Administrator, including $600 million generally for carrying out CAA section 132 (3 percent of which must be reserved for administrative costs necessary to carry out the section's provisions) and $400 million to make awards under CAA section 132 to eligible recipients/contractors that propose to replace eligible vehicles to serve one or more communities located in an air quality area designated pursuant to CAA section 107 as nonattainment for any air pollutant, in FY 2022 and available through FY 2031. CAA section 132 requires the Administrator to implement a program to make awards of grants and rebates to eligible recipients (defined as States, municipalities, Indian tribes, and nonprofit school transportation associations), and to make awards of contracts to eligible contractors for providing rebates, for up to 100 percent of costs for: (1) the incremental costs of replacing a Class 6 or Class 7 heavy-duty vehicle that is not a zero-emission vehicle with a zero-emission vehicle (as determined by the Administrator based on the market value of the vehicles); (2) purchasing, installing, operating, and maintaining infrastructure needed to charge, fuel, or maintain zero-emission vehicles; (3) workforce development and training to support the maintenance, charging, fueling, and operation of zero-emission vehicles; and (4) planning and technical activities to support the adoption and deployment of zero-emission vehicles.

Section 60102, “Grants to Reduce Air Pollution at Ports,” amends the CAA to add a new section 133 (42 U.S.C. 7433) and appropriates $3 billion (2 percent of which must be reserved for administrative costs necessary to carry out the section's provisions), $750 million of which is for projects located in areas of nonattainment for any air pollutant, in FY 2022 and available through FY 2027, to reduce air pollution at ports. Competitive rebates or grants are to be awarded for the purchase or installation of zero-emission port equipment or technology for use at, or to directly serve, one or more ports; to conduct any relevant planning or permitting in connection with the purchase or permitting of zero-emission port equipment or technology; and to develop qualified climate action plans. The zero-emission equipment or technology either (1) produces zero emissions of GHGs, listed criteria pollutants, and hazardous air pollutants or (2) it captures 100 percent of the emissions produced by an ocean-going vessel at berth.

Section 60103, “Greenhouse Gas Reduction Fund,” amends the CAA to add a new section 134 (42 U.S.C. 7434) and appropriates $27 billion, $15 billion of which is for low-income and disadvantaged communities, in FY 2022 and available through FY 2024, for a GHG reduction grant program. The program supports direct investments in qualified projects at the national, regional, State, and local levels, and indirect investments to establish new or support existing public, quasi-public, not-for-profit, or nonprofit entities that provide financial assistance to qualified projects. The program focuses on the rapid deployment of low- and zero-emission products, technologies, and services to reduce or avoid GHG emissions and other forms of air pollution.

Section 60104, “Diesel Emissions Reductions,” appropriates $60 million (2 percent of which must be reserved for administrative costs necessary to carry out the section's provisions), in FY 2022 and available through FY 2031, for grants, rebates, and loans under section 792 of the Energy Policy Act of 2005 (42 U.S.C. 16132) to identify and reduce diesel emissions resulting from goods movement facilities and vehicles servicing goods movement facilities in low-income and disadvantaged communities to address the health impacts of such emissions on such communities.

Section 70002 appropriates $3 billion in FY 2022 and available through FY 2031 for the U.S. Postal Service to purchase ZEVs ($1.29 billion) and to purchase, design, and install infrastructure to support zero-emission delivery vehicles at facilities that the U.S. Postal Service owns or leases from non-Federal entities ($1.71 billion).

Section 13501, “Extension of the Advanced Energy Project Credit,” allocates $10 billion in tax credits for facilities to domestically manufacture advanced energy technologies, subject to certain application and other requirements and limitations. Qualifying properties now include light-, medium-, or heavy-duty electric or fuel cell vehicles along with the technologies, components, or materials for such vehicles and the associated charging or refueling infrastructure. They also include hybrid vehicles with a gross vehicle weight rating of not less than 14,000 pounds along with the technologies, components, or materials for them.

Sections 50142, 50143, 50144, 50145, 50151, 50152, and 50153 collectively appropriate nearly $13 billion to support low- and zero-emission vehicle manufacturing and energy infrastructure. These provisions are intended to help accelerate the ability for industry to meet the demands spurred by the previously mentioned IRA sections, both for manufacturing vehicles, including BEVs and FCEVs, and for energy infrastructure.

Section 13204, “Clean Hydrogen,” amends section 45V of the Internal Revenue Code ( i.e., Title 26) to offer a tax credit to produce hydrogen for qualified clean production facilities that use a process that results in a lifecycle GHG emissions rate of not greater than 4 kg of CO₂ e per kg of hydrogen. This credit is eligible for qualified clean hydrogen production facilities whose construction begins before January 1, 2033, and is available during the 10-year period beginning on the date such facility was originally placed in service. The credit increases to a maximum of $3 per kilogram produced as the lifecycle GHG emissions rate is reduced to less than 0.45 kg of CO₂ e per kg of hydrogen. Facilities that received credit for the construction of carbon capture and direct air capture equipment or facilities ( i.e., under 45Q) do not qualify, and prevailing wage and apprenticeship requirements apply. Section 60113, “Methane Emissions Reduction Program,” amends the CAA by adding Section 136 and appropriates $850 million to EPA to support methane mitigation and monitoring, plus authorizes a new fee of $900 per ton on “waste” methane emissions that escalates after two years to $1,500 per ton. These combined incentives promote the production of hydrogen in a manner that minimizes its potential greenhouse gas impact.

While there are challenges facing greater adoption of heavy-duty ZEV technologies, the IRA provides many financial incentives to overcome these challenges and thus would also support our proposed rulemaking. We expect IRA sections 13502 and 13403 to support the adoption of HD ZEV technologies in the market, as detailed in our assessment of the appropriate GHG standards we are proposing. Additionally, we expect IRA sections 13404, 60101–60104, 70002, 13501, 50142–50145, 50151–50153, and 13204 to further accelerate ZEV adoption, but we are not including them quantitatively in our analyses.

As described in Section II of the proposed rule, EPA has considered the potential impacts of the BIL and the IRA in our assessment of the appropriate proposed GHG standards both quantitatively and qualitatively, and we request comment on our approach.

3. States' Efforts To Increase Adoption of HD ZEVs

HD vehicle sales and on-road vehicle populations are significant in the state of California. Approximately ten percent of U.S. HD ICE vehicles in 2016 were registered in California. California adopted the ACT program in 2020, which will also influence the market trajectory for BEV and FCEV technologies. The ACT program requires manufacturers who certify HD vehicles for sale in California to sell a certain percentage of zero-emission HD vehicles (BEVs or FCEVs) in California for each model year, beginning with MY 2024. As shown in Table I–1, the sales requirements vary by vehicle class, starting at 5 to 9 percent of total MY 2024 HD vehicle sales in California and increasing to 40 to 75 percent of a total MY's HD vehicle sales in California in MYs 2035 and later.

Outside of California, a number of states have signaled interest in greater adoption of HD ZEV technologies and/or establishing specific goals to increase the HD electric vehicle market. As one example, the Memorandum of Understanding (MOU), “Multi-State Medium- and Heavy-Duty Zero Emission Vehicle,” (Multi-State MOU) organized by Northeast States for Coordinated Air Use Management (NESCAUM), sets targets “to make all sales of new medium- and heavy-duty vehicles [in the jurisdictions of the signatory states and the District of Columbia] zero emission vehicles by no later than 2050” with an interim goal of 30 percent of all sales of new medium- and heavy-duty vehicles being zero emission vehicles no later than 2030. The Multi-State MOU was signed by the governors of 17 states including California, Colorado, Connecticut, Hawaii, Maine, Maryland, Massachusetts, New Jersey, New York, North Carolina, Nevada, Oregon, Pennsylvania, Rhode Island, Vermont, Virginia, and Washington, as well as the mayor of the District of Columbia. The Multi-State MOU outlines these jurisdictions' more specific commitments to move toward ZEVs through the Multi-State ZEV Task Force and provides an action plan for zero-emission medium- and heavy-duty vehicles with measurable sales targets and a focus on overburdened and underserved communities. Several states that signed the Multi-State MOU have since adopted California's ACT program, pursuant to CAA section 177, and we anticipate more jurisdictions will follow with similar proposals.

D. EPA Statutory Authority for the Proposal

This section briefly summarizes the statutory authority for the proposed rule. Statutory authority for the GHG standards EPA is proposing is found in CAA section 202(a)(1) (2), 42 U.S.C. 7521(a)(1)–(2), which requires EPA to establish standards applicable to emissions of air pollutants from new motor vehicles and engines which cause or contribute to air pollution which may reasonably be anticipated to endanger public health or welfare. Additional statutory authority for the proposed action is found in CAA sections 202–209, 216, and 301, 42 U.S.C. 7521–7543, 7550, and 7601. We discuss some key aspects of these sections in relation to this proposed action immediately below.

Title II of the Clean Air Act provides for comprehensive regulation of mobile sources, authorizing EPA to regulate emissions of air pollutants from all mobile source categories, including motor vehicles under CAA section 202(a). In turn, CAA section 216(2) defines “motor vehicle” as “any self-propelled vehicle designed for transporting persons or property on a street or highway.” Congress has intentionally and consistently used the broad term “any self-propelled vehicle” since the Motor Vehicle Air Pollution Control Act of 1965 so as not to limit standards adopted under CAA section 202 to vehicles running on a particular fuel, power source, or system of propulsion. Congress's focus was on emissions from classes of motor vehicles and the “requisite technologies” that could feasibly reduce those emissions giving appropriate consideration to cost of compliance and lead time, as opposed to being limited to any particular type of vehicle.

Section 202(a)(1) of the CAA states that “the Administrator shall by regulation prescribe (and from time to time revise) . . . standards applicable to the emission of any air pollutant from any class or classes of new motor vehicles . . . which in his judgment cause, or contribute to, air pollution which may reasonably be anticipated to endanger public health or welfare.” CAA section 202(a)(1) also requires that any standards promulgated thereunder “shall be applicable to such vehicles and engines for their useful life (as determined under [CAA section 202(d)], relating to useful life of vehicles for purposes of certification), whether such vehicle and engines are designed as complete systems or incorporate devices to prevent or control such pollution.” CAA section 202(d) directs EPA to prescribe regulations under which the “useful life” of vehicles and engines shall be determined for the purpose of setting standards under CAA section 202(a)(1). For HD highway vehicles and engines, CAA section 202(d) establishes “useful life” minimum values of 10 years or 100,000 miles, whichever occurs first, unless EPA determines that greater values are appropriate.

While emission standards set by the EPA under CAA section 202(a)(1) generally do not mandate use of particular technologies, they are technology-based, as the levels chosen must be premised on a finding of technological feasibility. Thus, standards promulgated under CAA section 202(a) are to take effect only “after such period as the Administrator finds necessary to permit the development and application of the requisite technology, giving appropriate consideration to the cost of compliance within such period.” CAA section 202(a)(2); see also NRDC v. EPA, 655 F. 2d 318, 322 (D.C. Cir. 1981). EPA must consider costs to those entities which are directly subject to the standards. Motor & Equipment Mfrs. Ass'n Inc. v. EPA, 627 F. 2d 1095, 1118 (D.C. Cir. 1979). Thus, “the [s]ection 202(a)(2) reference to compliance costs encompasses only the cost to the motor-vehicle industry to come into compliance with the new emission standards, and does not mandate consideration of costs to other entities not directly subject to the proposed standards.” Coalition for Responsible Regulation v. EPA, 684 F.3d 120, 128 (D.C. Cir. 2012). EPA is afforded considerable discretion under section 202(a) when assessing issues of technical feasibility and availability of lead time to implement new technology. Such determinations are “subject to the restraints of reasonableness,” which “does not open the door to `crystal ball' inquiry.” NRDC, 655 F. 2d at 328, quoting International Harvester Co. v. Ruckelshaus, 478 F. 2d 615, 629 (D.C. Cir. 1973); see also Growth Energy v. EPA, 5 F.4th 1, 15 (D.C. Cir. 2021) (“The court is `particularly deferential' to agencies' predictive judgments, requiring only that `the agency acknowledge factual uncertainties and identify the considerations it found persuasive.' EPA cleared that modest bar.”) (internal citations omitted). Moreover, “EPA is not obliged to provide detailed solutions to every engineering problem posed in the perfection of [a particular device]. In the absence of theoretical objections to the technology, the agency need only identify the major steps necessary for development of the device, and give plausible reasons for its belief that the industry will be able to solve those problems in the time remaining. The EPA is not required to rebut all speculation that unspecified factors may hinder `real world' emission control.” NRDC, 655 F. 2d at 333–34. In developing such technology-based standards, EPA has the discretion to consider different standards for appropriate groupings of vehicles (“class or classes of new motor vehicles”), or a single standard for a larger grouping of motor vehicles. NRDC, 655 F.2d at 338.

Although standards under CAA section 202(a)(1) are technology-based, they are not based exclusively on technological capability. Pursuant to the broad grant of authority in section 202, when setting GHG emission standards for HD vehicles, EPA must consider certain factors and may also consider other factors and has done so previously when setting such standards. For instance, in HD GHG Phase 1 and Phase 2, EPA explained that when acting under this authority EPA has considered such issues as technology effectiveness, its cost (including per vehicle, per manufacturer, and per purchaser), the lead time necessary to implement the technology, and based on this the feasibility and practicability of potential standards; the impacts of potential standards on emissions reductions; the impacts of standards on oil conservation and energy security; the impacts of standards on fuel savings by vehicle operators; the impacts of standards on the heavy-duty vehicle industry; as well as other relevant factors such as impacts on safety.

In addition, EPA has clear authority to set standards under CAA section 202(a)(1)–(2) that are technology forcing when EPA considers that to be appropriate, but is not required to do so (as compared to standards under provisions such as section 202(a)(3), which require the greatest degree of emissions reduction achievable, giving appropriate consideration to cost, energy and safety factors). CAA section 202(a) does not specify the degree of weight to apply to each factor, and EPA accordingly has discretion in choosing an appropriate balance among factors. See Sierra Club v. EPA, 325 F.3d 374, 378 (D.C. Cir. 2003) (even where a provision is technology-forcing, the provision “does not resolve how the Administrator should weigh all [the statutory] factors in the process of finding the 'greatest emission reduction achievable'”); National Petrochemical and Refiners Ass'n v. EPA, 287 F.3d 1130, 1135 (D.C. Cir. 2002) (EPA decisions, under CAA provision authorizing technology-forcing standards, based on complex scientific or technical analysis are accorded particularly great deference); see also Husqvarna AB v. EPA, 254 F. 3d 195, 200 (D.C. Cir. 2001) (great discretion to balance statutory factors in considering level of technology-based standard, and statutory requirement “to [give appropriate] consideration to the cost of applying . . . technology” does not mandate a specific method of cost analysis); Hercules Inc. v. EPA, 598 F. 2d 91, 106 (D.C. Cir. 1978) (“In reviewing a numerical standard we must ask whether the agency's numbers are within a zone of reasonableness, not whether its numbers are precisely right.”).

As noted previously in this section, there are also other provisions of the CAA that provide authority for EPA's proposed action, including CAA sections 203, 206, and 207. Under section 203 of the CAA, sales of vehicles are prohibited unless the vehicle is covered by a certificate of conformity, and EPA issues certificates of conformity pursuant to section 206 of the CAA. Certificates of conformity are based on (necessarily) pre-sale testing conducted either by EPA or by the manufacturer. Compliance with standards is required not only at certification but throughout a vehicle's useful life, so that testing requirements may continue post-certification. To assure each engine and vehicle complies during its useful life, EPA may apply an adjustment factor to account for vehicle emission control deterioration or variability in use (section 206(a)). EPA establishes the test procedures under which compliance with the CAA emissions standards is measured. EPA's testing authority under the CAA is broad and flexible.

Under CAA section 207, manufacturers are required to provide emission-related warranties. The emission-related warranty period for HD engines and vehicles under CAA section 207(i) is “the period established by the Administrator by regulation (promulgated prior to November 15, 1990) for such purposes unless the Administrator subsequently modifies such regulation.” For HD vehicles, part 1037 currently specifies that the emission-related warranty for Light HD vehicles is 5 years or 50,000 miles and for Medium HD and Heavy HD vehicles is 5 years or 100,000 miles, and specifies the components covered for such vehicles. Section 207 of the CAA also grants EPA broad authority to require manufacturers to remedy nonconformity if EPA determines there are a substantial number of noncomplying vehicles. Additional aspects of EPA's legal authority are more fully discussed in the HD GHG Phase 1 final rule. Further discussion of EPA's authority under CAA section 202(a)(1)–(2) may also be found in the HD GHG Phase 1 final rule.

With regard to the specific technologies that could be used to meet the emission standards promulgated under the statutory authorities discussed in this Section I.D, EPA's rules have historically not required the use of any particular technology, but rather have allowed manufacturers to use any technology that demonstrates the engine or vehicle meets the standards over the applicable test procedures. Similarly, in determining the standards, EPA appropriately considers updated data and analysis on pollution control technologies, without a priori limiting its consideration to a particular set of technologies. Given the continuous development of pollution control technologies since the early days of the CAA, this approach means that EPA routinely considers novel and projected technologies developed or refined since the time of the CAA's enactment, including for instance, electric vehicle technologies. In requiring EPA to consider lead time that takes into consideration development and application of technology when setting standards before such standards may take effect, Congress directed EPA to consider future technological advancements and innovation rather than limiting the Agency to setting standards that reflect only technologies in place at the time the standards are developed. This forward-looking regulatory approach keeps pace with real-world technological developments that have the potential to reduce emissions and comports with Congressional intent.

Section 202 does not specify or expect any particular type of motor vehicle propulsion system to remain prevalent, and it was clear as early as the 1960s that ICE vehicles might be inadequate to achieve the country's air quality goals. In 1967, the Senate Committees on Commerce and Public Works held five days of hearings on “electric vehicles and other alternatives to the internal combustion engine,” which Chairman Magnuson opened by saying “The electric will help alleviate air pollution. . . . The electric car does not mean a new way of life, but rather it is a new technology to help solve the new problems of our age.” In a 1970 message to Congress seeking a stronger CAA, President Nixon stated he was initiating a program to develop “an unconventionally powered, virtually pollution free automobile” because of the possibility that “the sheer number of cars in densely populated areas will begin outrunning the technological limits of our capacity to reduce pollution from the internal combustion engine.”

Since the earliest days of the CAA, Congress has emphasized that the goal of section 202 is to address air quality hazards from motor vehicles, not to simply reduce emissions from internal combustion engines to the extent feasible. In the Senate Report accompanying the 1970 CAA Amendments, Congress made clear the EPA “is expected to press for the development and application of improved technology rather than be limited by that which exists” and identified several “unconventional” technologies that could successfully meet air quality-based emissions targets for motor vehicles. In the 1970 amendments Congress further demonstrated its recognition that developing new technology to ensure that pollution control keeps pace with economic development is not merely a matter of refining the ICE, but requires considering new types of motor vehicle propulsion. Congress provided EPA with authority to fund the development of “low emission alternatives to the present internal combustion engine” as well as a program to encourage Federal purchases of “low-emission vehicles.” See CAA section 104(a)(2) (previously codified as CAA section 212). Congress also adopted section 202(e) expressly to grant the Administrator discretion regarding the certification of vehicles and engines based on “new power sources or propulsion system[s],” that is to say, power sources and propulsion systems beyond the existing internal combustion engine and fuels available at the time of the statute's enactment, if those vehicles emitted pollutants which the Administrator judged contributed to dangerous air pollution but had not yet established standards for under section 202(a). As the D.C. Circuit stated in 1975, “We may also note that it is the belief of many experts—both in and out of the automobile industry—that air pollution cannot be effectively checked until the industry finds a substitute for the conventional automotive power plant–the reciprocating internal combustion ( i.e., “piston”) engine. . . . It is clear from the legislative history that Congress expected the Clean Air Amendments to force the industry to broaden the scope of its research—to study new types of engines and new control systems.” International Harvester Co. v. Ruckelshaus, 478 F.2d 615, 634–35 (D.C. Cir. 1975).

Since that time, Congress has continued to emphasize the importance of technology development to achieving the goals of the CAA. In the 1990 amendments, Congress instituted a clean fuel vehicles program to promote further progress in emissions reductions, which also applied to motor vehicles as defined under section 216, see CAA section 241(1), and explicitly defined motor vehicles qualifying under the program as including vehicles running on an alternative fuel or “power source (including electricity),” CAA section 241(2). Congress also directed EPA to phase-in certain section 202(a) standards, see CAA section 202(g)–(j), which confirms EPA's authority to promulgate standards, such as fleet averages, phase-ins, and averaging, banking, and trading programs, that are fulfilled through compliance over an entire fleet, or a portion thereof, rather than through compliance by individual vehicles. As previously noted in the Executive Summary of this preamble, EPA has long included averaging provisions for complying with emission standards in the HD program and in upholding the first HD final rule that included such a provision the D.C. Circuit rejected petitioner's challenge in the absence of any clear evidence that Congress meant to prohibit averaging. NRDC v. Thomas, 805 F.2d 410, 425 (D.C. Cir. 1986). In the subsequent 1990 amendments, Congress, noting NRDC v. Thomas, opted to let the existing law “remain in effect,” reflecting that “[t]he intention was to retain the status quo,” i.e., EPA's existing authority to allow averaging. Averaging, banking, and trading is discussed further in Sections II and III of this preamble; additional history of ABT is discussed in EPA's Answering Brief in Texas v. EPA (D.C. Cir., 22–1031, at § IV.A–B).

The recently-enacted IRA “reinforces the longstanding authority and responsibility of [EPA] to regulate GHGs as air pollutants under the Clean Air Act,” and “the IRA clearly and deliberately instructs EPA to use” this authority by “combin[ing] economic incentives to reduce climate pollution with regulatory drivers to spur greater reductions under EPA's CAA authorities.” To assist with this, as described in Section I.C.2, the IRA provided a number of economic incentives for HD ZEVs and the infrastructure necessary to support them, and specifically affirms Congress's previously articulated statements that non-ICE technologies will be a key component of achieving emissions reductions from the mobile source sector, including the HD industry sector. The Congressional Record reflects that “Congress recognizes EPA's longstanding authority under CAA Section 202 to adopt standards that rely on zero emission technologies, and Congress expects that future EPA regulations will increasingly rely on and incentivize zero-emission vehicles as appropriate.”

Consistent with Congress's intent, EPA's CAA Title II emission standards have been based on and stimulated the development of a broad set of advanced technologies, such as electronic fuel injection systems, gasoline catalytic convertors, diesel particulate filters, diesel NO_X reduction catalysts, gasoline direct injection fuel systems, active aerodynamic grill shutters, and advanced transmission technologies, which have been the building blocks of heavy-duty vehicle designs and have yielded not only lower pollutant emissions, but improved vehicle performance, reliability, and durability. As previously discussed, beginning in 2011, EPA has set HD vehicle and engine standards under section 202(a)(1)–(2) for GHGs. Manufacturers have responded to standards over the past decade by continuing to develop and deploy a wide range of technologies, including more efficient engine designs, transmissions, aerodynamics, and tires, air conditioning systems that contribute to lower GHG emissions, as well as vehicles based on methods of propulsion beyond diesel- and gasoline-fueled ICE vehicles, including ICE running on alternative fuels (such as natural gas, biodiesel, renewable diesel, methanol, and other fuels), as well as various levels of electrified vehicle technologies from mild hybrids, to strong hybrids, and up through battery electric vehicles and fuel cell electric vehicles. In addition, the continued application of performance-based standards take into consideration averaging provisions that provide an opportunity for all technology improvements and innovation to be reflected in a vehicle manufacturers' compliance results.

With regard to EPA's proposed revised preemption regulations regarding locomotives described in Section X of the preamble, statutory authority is found in CAA section 209. CAA section 209(e)(1)(B), 42 U.S.C. 7543(e)(1)(B), prohibits states and political subdivisions thereof from adopting or attempting to enforce any standard or other requirement relating to the control of emissions from new locomotives or new engines used in locomotives. However, CAA section 209(e)(2)(A)–(B), 42 U.S.C. 7543(e)(2)(A)–(B), requires EPA to authorize, after notice and an opportunity for public hearing, California to adopt and enforce standards and other requirements relating to control of emissions from other nonroad vehicles or engines provided certain criteria are met, and allows states other than California to adopt and enforce, after notice to EPA, such standards provided they are equivalent to California's authorized standards. CAA section 209(e)(2)(B) then requires EPA to issue regulations to implement subsection 209(e).

E. Coordination With Federal and State Partners

Executive Order 14037 directs EPA and DOT to coordinate, as appropriate and consistent with applicable law, during consideration of this rulemaking. EPA has coordinated and consulted with DOT/NHTSA, both on a bilateral level during the development of the proposed program as well as through the interagency review of the EPA proposal led by the Office of Management and Budget. EPA has set some previous heavy-duty vehicle GHG emission standards in joint rulemakings where NHTSA also established heavy-duty fuel efficiency standards. In the light-duty GHG emission rulemaking establishing standards for model years 2023 through 2026, EPA and NHTSA concluded that it was appropriate to coordinate and consult but not to engage in joint rulemaking. EPA has similarly concluded that it is not necessary for this EPA proposal to be issued in a joint action with NHTSA. In reaching this conclusion, EPA notes there is no statutory requirement for joint rulemaking and that the agencies have different statutory mandates and their respective programs have always reflected those differences. As the Supreme Court has noted, “EPA has been charged with protecting the public's 'health' and 'welfare,' a statutory obligation wholly independent of DOT's mandate to promote energy efficiency.” Although there is no statutory requirement for EPA to consult with NHTSA, EPA has consulted with NHTSA in the development of this proposal. For example, staff of the two agencies met frequently to discuss various technical issues and to share technical information.

EPA also has consulted with other federal agencies in developing this proposal, including the Federal Energy Regulatory Commission, the Department of Energy and several national labs. EPA collaborates with DOE and Argonne National Laboratory on battery cost analyses and critical materials forecasting. EPA also coordinates with the Joint Office of Energy and Transportation on charging infrastructure. EPA and the Oak Ridge National Laboratory collaborate on energy security issues. EPA also participates in the Federal Consortium for Advanced Batteries led by DOE and the Joint Office of Energy and Transportation. EPA and DOE also have entered into a Joint Memorandum of Understanding to provide a framework for interagency cooperation and consultation on electric sector resource adequacy and operational reliability.

E.O. 14037 also directs EPA to coordinate with California and other states that are leading the way in reducing vehicle emissions, as appropriate and consistent with applicable law, during consideration of this rulemaking. EPA has engaged with the California Air Resources Board on technical issues in developing this proposal. EPA has considered certain aspects of the CARB Advanced Clean Trucks Rule, as discussed elsewhere in this document. We also have engaged with other states, including members of the National Association of Clean Air Agencies, the Association of Air Pollution Control Agencies, the Northeast States for Coordinated Air Use Management, and the Ozone Transport Commission.

F. Stakeholder Engagement

EPA has conducted extensive engagement with a diverse range of interested stakeholders in developing this proposal. We have engaged with those groups with whom E.O. 14037 specifically directs EPA to engage, including labor unions, states, industry, environmental justice organizations and public health experts. In addition, we have engaged with environmental NGOs, vehicle manufacturers, technology suppliers, dealers, utilities, charging providers, Tribal governments, and other organizations. For example, in April–May 2022, EPA held a series of engagement sessions with organizations representing all of these stakeholder groups so that EPA could hear early input in developing its proposal. EPA has continued engagement with many of these stakeholders throughout the development of this proposal. EPA looks forward to hearing from all stakeholders through comments on this proposal and during the public hearing.

II. Proposed CO ₂ Emission Standards

Under our CAA section 202(a)(1)–(2) authority, and consistent with E.O. 14037, we are proposing new GHG standards for MYs 2027 through 2032 and later HD vehicles. We are retaining and not reopening the nitrous oxide (N₂ O), methane (CH₄), and CO₂ emission standards that apply to heavy-duty engines, the HFC emission standards that apply to heavy-duty vehicles, and the general compliance structure of existing 40 CFR part 1037 except for some proposed revisions described in Section III. In this Section II, we describe our assessment that these stringent standards are appropriate and feasible considering lead time, costs, and other factors. These proposed Phase 3 standards include (1) revised GHG standards for many MY 2027 HD vehicles, and (2) new GHG standards starting in MYs 2028 through 2032. The proposed standards do not mandate the use of a specific technology, and EPA anticipates that a compliant fleet under the proposed standards would include a diverse range of technologies, including ZEV and ICE vehicle technologies. In developing the proposed standards, EPA has considered the key issues associated with growth in penetration of zero-emission vehicles, including charging infrastructure and hydrogen production. In this section, we describe our assessment of the appropriateness and feasibility of these proposed standards and present a technology pathway for achieving each of those standards through increased ZEV adoption. In this section, we also present and request comment on an alternative that would provide a more gradual phase-in of the standards. As described in Section II.H., EPA also requests comment on setting GHG standards starting in MYs 2027 through 2032 that would reflect: values less stringent than the lower stringency alternative for certain market segments, values in between the proposed standards and the alternative standards, values in between the proposed standards and those that would reflect ZEV adoption levels ( i.e., percent of ZEVs in production volumes) used in California's ACT, values that would reflect the level of ZEV adoption in the ACT program, and values beyond those that would reflect ZEV adoption levels in ACT such as the 50- to 60-percent ZEV adoption range.

In the beginning of this section, we first describe the public health and welfare need for GHG emission reductions (Section II.A). In Section II.B, we provide an overview of the comments the Agency received in response to the GHG standards previously proposed as part of the HD2027 NPRM. In Section II.C, we provide a brief overview of the existing CO₂ emission standards that we promulgated in HD GHG Phase 2. Section II.D contains our technology assessment and Section II.E includes our assessment of technology costs, EVSE costs, operating costs, and payback. Section II.F includes the proposed standards and the analysis demonstrating the feasibility and Section II.G discusses the feasibility and appropriateness of the proposed emission standards under the Clean Air Act. Section II.H presents potential alternatives to the proposed standards, including requests for comment on standards other than those proposed. Finally, Section II.I summarizes our consideration of small businesses.

A. Public Health and Welfare Need for GHG Emission Reductions

The transportation sector is the largest U.S. source of GHG emissions, representing 27 percent of total GHG emissions. Within the transportation sector, heavy-duty vehicles are the second largest contributor, at 25 percent. GHG emissions have significant impacts on public health and welfare as set forth in EPA's 2009 Endangerment and Cause or Contribute Findings under CAA section 202(a) and as evidenced by the well-documented scientific record.

Elevated concentrations of GHGs have been warming the planet, leading to changes in the Earth's climate including changes in the frequency and intensity of heat waves, precipitation, and extreme weather events; rising seas; and retreating snow and ice. The changes taking place in the atmosphere as a result of the well-documented buildup of GHGs due to human activities are altering the climate at a pace and in a way that threatens human health, society, and the natural environment. While EPA is not making any new scientific or factual findings with regard to the well-documented impact of GHG emissions on public health and welfare in support of this rule, EPA is providing some scientific background on climate change to offer additional context for this rulemaking and to increase the public's understanding of the environmental impacts of GHGs.

Extensive additional information on climate change is available in the scientific assessments and the EPA documents that are briefly described in this section, as well as in the technical and scientific information supporting them. One of those documents is EPA's 2009 Endangerment and Cause or Contribute Findings for Greenhouse Gases Under section 202(a) of the CAA (74 FR 66496, December 15, 2009). In the 2009 Endangerment Finding, the Administrator found under section 202(a) of the CAA that elevated atmospheric concentrations of six key well-mixed GHGs—CO₂, methane (CH₄), nitrous oxide (N₂ O), hydrofluorocarbons (HFCs), perfluorocarbons (PFCs), and sulfur hexafluoride (SF₆)—“may reasonably be anticipated to endanger the public health and welfare of current and future generations” (74 FR 66523). The 2009 Endangerment Finding, together with the extensive scientific and technical evidence in the supporting record, documented that climate change caused by human emissions of GHGs (including HFCs) threatens the public health of the U.S. population. It explained that by raising average temperatures, climate change increases the likelihood of heat waves, which are associated with increased deaths and illnesses (74 FR 66497). While climate change also increases the likelihood of reductions in cold-related mortality, evidence indicates that the increases in heat mortality will be larger than the decreases in cold mortality in the United States (74 FR 66525). The 2009 Endangerment Finding further explained that compared with a future without climate change, climate change is expected to increase tropospheric ozone pollution over broad areas of the United States., including in the largest metropolitan areas with the worst tropospheric ozone problems, and thereby increase the risk of adverse effects on public health (74 FR 66525). Climate change is also expected to cause more intense hurricanes and more frequent and intense storms of other types and heavy precipitation, with impacts on other areas of public health, such as the potential for increased deaths, injuries, infectious and waterborne diseases, and stress-related disorders (74 FR 66525). Children, the elderly, and the poor are among the most vulnerable to these climate-related health effects (74 FR 66498).

The 2009 Endangerment Finding also documented, together with the extensive scientific and technical evidence in the supporting record, that climate change touches nearly every aspect of public welfare in the United States., including: changes in water supply and quality due to changes in drought and extreme rainfall events; increased risk of storm surge and flooding in coastal areas and land loss due to inundation; increases in peak electricity demand and risks to electricity infrastructure; and the potential for significant agricultural disruptions and crop failures (though offset to a lesser extent by carbon fertilization). These impacts are also global and may exacerbate problems outside the United States. that raise humanitarian, trade, and national security issues for the U.S. (74 FR 66530).

The most recent information demonstrates that the climate is continuing to change in response to the human-induced buildup of GHGs in the atmosphere. Recent scientific assessments show that atmospheric concentrations of GHGs have risen to a level that has no precedent in human history and that they continue to climb, primarily because of both historic and current anthropogenic emissions, and that these elevated concentrations endanger our health by affecting our food and water sources, the air we breathe, the weather we experience, and our interactions with the natural and built environments.

Global average temperature has increased by about 1.1 degrees Celsius (°C) (2.0 degrees Fahrenheit (°F)) in the 2011–2020 decade relative to 1850–1900. The IPCC determined with medium confidence that this past decade was warmer than any multi-century period in at least the past 100,000 years. Global average sea level has risen by about 8 inches (about 21 centimeters (cm)) from 1901 to 2018, with the rate from 2006 to 2018 (0.15 inches/year or 3.7 millimeters (mm)/year) almost twice the rate over the 1971 to 2006 period, and three times the rate of the 1901 to 2018 period. The rate of sea level rise during the 20th Century was higher than in any other century in at least the last 2,800 years. The CO₂ being absorbed by the ocean has resulted in changes in ocean chemistry due to acidification of a magnitude not seen in 65 million years putting many marine species—particularly calcifying species—at risk. Human-induced climate change has led to heatwaves and heavy precipitation becoming more frequent and more intense, along with increases in agricultural and ecological droughts in many regions. The NCA4 found that it is very likely (greater than 90 percent likelihood) that by mid-century, the Arctic Ocean will be almost entirely free of sea ice by late summer for the first time in about 2 million years. Coral reefs will be at risk for almost complete (99 percent) losses with 1 °C (1.8 °F) of additional warming from today (2 °C or 3.6 °F since preindustrial). At this temperature, between 8 and 18 percent of animal, plant, and insect species could lose over half of the geographic area with suitable climate for their survival, and 7 to 10 percent of rangeland livestock would be projected to be lost. The IPCC similarly found that climate change has caused substantial damages and increasingly irreversible losses in terrestrial, freshwater, and coastal and open ocean marine ecosystems.

In 2016, the Administrator issued a similar finding for GHG emissions from aircraft under section 231(a)(2)(A) of the CAA. In the 2016 Endangerment Finding, the Administrator found that the body of scientific evidence amassed in the record for the 2009 Endangerment Finding compellingly supported a similar endangerment finding under CAA section 231(a)(2)(A), and also found that the science assessments released between the 2009 and the 2016 Findings “strengthen and further support the judgment that GHGs in the atmosphere may reasonably be anticipated to endanger the public health and welfare of current and future generations” (81 FR 54424). Pursuant to the 2009 Endangerment and Cause or Contribute Findings, CAA section 202(a) requires EPA to issue standards applicable to emissions of those pollutants from new motor vehicles. See Coalition for Responsible Regulation, 684 F.3d at 116–125, 126–27; Massachusetts, 549 U.S. at 533. See also Coalition for Responsible Regulation, 684 F.3d at 127–29 (upholding EPA's light-duty GHG emission standards for MYs 2012–2016 in their entirety). Since the 2016 Endangerment Finding, the climate has continued to change, with new observational records being set for several climate indicators such as global average surface temperatures, GHG concentrations, and sea level rise. Additionally, major scientific assessments continue to be released that further advance our understanding of the climate system and the impacts that GHGs have on public health and welfare both for current and future generations. These updated observations and projections document the rapid rate of current and future climate change both globally and in the United States.

B. Summary of Comments Received From HD2027 NPRM

We received a significant number of comments to the proposed updates to the HD GHG emission standards proposed as part of the HD2027 NPRM. A number of commenters provided support and reasoning for revising the HD CO₂ standards while a number of other commenters expressed concerns about reopening the HD GHG Phase 2 program. This Section II.B includes a summary of the comments received. Commenters who would like EPA to further consider in this rulemaking any relevant comments that they provided on the HD2027 NPRM regarding proposed HD vehicle GHG standards for the MYs at issue in this proposal must resubmit those comments to EPA during this proposal's comment period. EPA considered the comments received in response to the HD2027 NPRM when developing this Phase 3 proposal. The proposed standards were developed based on a more in-depth analysis of the potential for electrification of the heavy-duty sector and attendant emissions reductions than was used in the HD2027 NPRM analysis and is described in Sections II.D through II.F. This analysis addresses many of the concerns raised in comments summarized in the following subsections, such as the need to consider a wide range of HD applications, technology and operating costs of BEVs, the impact of heating and cooling on the energy demands of electric vehicles, infrastructure concerns, and the potential impact of weight and space for packaging of batteries. This analysis also includes consideration of the IRA provisions that provide significant financial incentives for the heavy-duty ZEV market and reduce or eliminate the cost difference between ICE vehicles and ZEVs. In consideration of some commenters' concerns about the time needed for research plans, product development, manufacturing investment, and charging infrastructure, we discuss these topics in our technical analysis supporting this NPRM. As described in Section II.H., EPA also requests comment on setting GHG standards starting in MYs 2027 through 2032 that would reflect: values less stringent than the lower stringency alternative for certain market segments, values in between the proposed standards and the alternative standards, values in between the proposed standards and those that would reflect ZEV adoption levels ( i.e., percent of ZEVs in production volumes) used in California's ACT, values that would reflect the level of ZEV adoption in the ACT program, and values beyond those that would reflect ZEV adoption levels in ACT such as the 50- to 60-percent ZEV adoption range.

1. Summary of Comments in Support of Revising the Phase 2 GHG Emission Standards for MY 2027

Many commenters, including non-governmental organizations, states, and mass comment campaigns, provided support for revising the targeted HD vehicle MY 2027 CO₂ emission standards to reflect the increase in electrification of the HD market and attendant potential for additional emission reductions. Additionally, many commenters suggested that EPA should further reduce the emission standards in MYs 2027 through 2029 beyond the levels proposed because of the accelerating adoption of HD ZEVs. Many commenters also highlighted that five additional states besides California adopted the California ACT program in late 2021 and noted that this would also drive additional electrification in the HD segment of the transportation sector. Finally, some commenters pointed to the “Multi-State Medium and Heavy-Duty Zero Emission Vehicle Memorandum of Understanding” (Multi-State MOU) signed by 17 states and the District of Columbia establishing goals to increase HD electric vehicle sales in those jurisdictions to 30 percent by 2030 and 100 percent by 2050. Commenters also provided a number of reports that evaluate the potential of electrification of the HD sector in terms of adoption rates, costs, and other factors.

Some of the commenters provided specific recommendations for HD ZEV adoption rates in the MYs 2027 through 2029 timeframe. For example, the American Council for an Energy-Efficient Economy (ACEEE) suggested that, based on a recent NREL study, EPA could set standards that reflect 20 percent electrification in MY 2027 and up to 40 percent in MY 2029. The Environmental Defense Fund (EDF) suggested standards to achieve 80 percent sales of ZEVs for new school and transit buses and 40 percent of new Class 4–7 vehicles and Class 8 short-haul vehicles by MY 2029. EDF also referenced an analysis from Environmental Resources Management (ERM) that included a range of scenarios, with midpoint scenarios projecting HD ZEV deployment in excess of 20 percent in MY 2029 and more optimistic scenarios projecting HD ZEV sales of over 33 percent of all Class 4–8 single unit trucks, short-haul tractors, and school and transit buses in MY 2029. The ICCT suggested HD ZEV ranges of 15 to 40 percent depending on the vehicle segment in MY 2027, increasing up to 40 to 80 percent in MY 2029. Moving Forward Network suggested that ZEVs could comprise 20 percent of new sales in MY 2027 and increase 10 percent each year, with a goal of 100 percent by MY 2035. Tesla referenced a NREL study, a forecast from Americas Commercial Transportation Research Co. (ACT Research) that projected a 26 percent sales share of HD ZEVs nationwide in 2030, and another study that projected 25 percent of the global HD fleet will be electric by 2030. Other commenters, such as AMPLY Power (rebranded to bp plus), suggest that the federal CO₂ emission standards should achieve ZEV deployments on par with California's ACT program.

Some commenters also referred to manufacturer statements regarding such manufacturers' projections for HD electrification. For example, ACEEE pointed to Volvo's and Scania's announcements for global electrification targets of 50 percent by 2030. EDF pointed to several manufacturer's statements. First, EDF noted Daimler Trucks North America has committed to offering only carbon-neutral trucks in the United States by 2039 and expects that by 2030, as much as 60 percent of its sales will be ZEVs. Second, EDF noted Navistar has a goal of having 50 percent of its sales volume be ZEVs by 2030, and its commitment to achieve 100 percent zero emissions by 2040 across all operations and carbon-neutrality by 2050.

Finally, some commenters discussed hydrogen-powered ICEs and asserted that there are benefits associated with that technology as a potential CO₂ -reducing technology for the HD segment of the transportation sector.

2. Summary of Comments Expressing Concern With Revising the Phase 2 GHG Emission Standards for MY 2027

Some commenters raised concerns with the HD2027 NPRM proposed changes to certain HD GHG Phase 2 CO₂ emission standards. Some highlighted the significant investment and lead time required for development and verification of durability of ZEVs and stated EPA should not adopt standards that project broad adoption of heavy-duty ZEVs.

Some commenters stated that EPA should not reopen the HD GHG Phase 2 emission standards. Several manufacturers and suppliers pointed to the need for regulatory certainty and stability, stating that reopening the Phase 2 standards would threaten their long-term investments and production planning. Some commenters went further and stated that certain technologies that EPA projected for use to meet the existing Phase 2 emission standards are seeing lower-than-expected penetration rates in MY 2021; these commenters suggested that EPA relax the Phase 2 standards. The technologies highlighted by the commenters suggesting that EPA relax Phase 2 standards include tamper-resistant automatic shutdown systems, neutral idle, low rolling resistance tires, stop-start, and advanced transmission shift strategies.

Commenters also stated that it takes time to develop ZEV technologies for the wide range of HD applications. They also raised concerns regarding asserted high costs and long lead times associated with the necessary charging infrastructure, the weight impact of batteries, the impact of battery degradation and ambient temperatures on the range of electric vehicles, and the impact on operations due to the time required to charge. Commenters also raised issues regarding the upstream and lifecycle emissions impact of ZEVs, including minerals and battery manufacturing, battery disposal and recycling, potential higher tire and brake wear from electric vehicles, and the availability of minerals and other supply chain issues.

Some commenters raised concerns about the approach used in the HD2027 NPRM to project ZEV sales in MY 2027. Concerns raised by commenters include the uncertainty of the actual production levels needed to meet California ACT program requirements; that EPA has not approved a waiver for the California ACT program and, therefore, should not consider full implementation of that program; and that the current HD ZEVs are expensive.

One commenter raised concerns related to small businesses. The commenter stated that its less diverse product mix and low sales volume present challenges in meeting the proposed GHG standards in the HD2027 NPRM.

C. Background on the CO₂ Emission Standards in the HD GHG Phase 2 Program

In the Phase 2 Heavy-Duty GHG rule, we finalized GHG emission standards tailored to three regulatory categories of HD vehicles—heavy-duty pickups and vans, vocational vehicles, and combination tractors. In addition, we set separate standards for the engines that power combination tractors and for the engines that power vocational vehicles. The heavy-duty vehicle CO₂ emission standards are in grams per ton-mile, which represents the grams of CO₂ emitted to move one ton of payload a distance of one mile. In promulgating the Phase 2 standards, we explained that the stringency of the Phase 2 standards were derived on a fleet average technology mix basis and that the emission averaging provisions of ABT meant that the regulations did not require all vehicles to meet the standards (contrasted with the banking and trading provisions of the HD GHG Phase 2 ABT program which were not relied upon in selecting the stringency the HD GHG Phase 2 standards). For example, we projected that diversified manufacturers would continue to use the averaging provisions in the ABT program to meet the standards on average for each of their vehicle families. In addition, the Phase 2 program established subcategories of vehicles ( i.e., custom chassis vocational vehicles and heavy-haul tractors) that were specifically designed to recognize the limitations of certain vehicle applications to adopt some technologies due to specialized operating characteristics or generally low sales volumes with prohibitively long payback periods. The vehicles certified to the custom chassis vocational vehicle standards are not permitted to bank or trade credits and some have limited averaging provisions under the HD GHG Phase 2 ABT program.

In this proposal, we continue to expect averaging would play an important role in manufacturer strategies to meet the proposed standards. In Section II.F, we are proposing new standards for vocational vehicles and combination tractors, which we project are feasible to meet through a technology pathway where vehicle manufacturers would adopt ZEV technologies for a portion of their product lines. This Section II.C includes additional background information on these two vehicle categories. At this time, we are not proposing to update engine standards in 40 CFR 1036.108. Additionally, we intend to separately pursue a combined light-duty and medium-duty rulemaking to propose more stringent standards for complete and incomplete vehicles at or below 14,000 pounds. GVWR that are certified under 40 CFR part 86, subpart S. Manufacturers of incomplete vehicles at or below 14,000 pounds GVWR would continue to have the option of either meeting the greenhouse gas standards under 40 CFR parts 1036 and 1037, or instead meeting the greenhouse gas standards with chassis-based measurement procedures under 40 CFR part 86, subpart S.

We are continuing and are not reopening the existing approach taken in both HD GHG Phase 1 and Phase 2, that compliance with the vehicle exhaust CO₂ emission standards is based on CO₂ emissions from the vehicle. See 76 FR 57123 (September 15, 2011); see also 77 FR 51705 (August 24, 2012), 77 FR 51500 (August 27, 2012), and 81 FR 75300 (October 25, 2016). EPA's heavy-duty standards have been in place as engine- and vehicle-based standards for decades, for all engine and vehicle technologies. We estimated the upstream emission impact of the proposed standards for heavy-duty vehicles on both the refinery and electricity generation sectors, as shown in Section V, and those analyses also support the proposed CO₂ emission standards.

1. Vocational Vehicles

Vocational vehicles include a wide variety of vehicle types, spanning Class 2b-8, and serve a wide range of functions. We define vocational vehicles as all heavy-duty vehicles greater than 8,500 lb GVWR that are not certified under 40 CFR part 86, subpart S, or a combination tractor under 40 CFR 1037.106. Some examples of vocational vehicles include urban delivery trucks, refuse haulers, utility service trucks, dump trucks, concrete mixers, transit buses, shuttle buses, school buses, emergency vehicles, motor homes, and tow trucks. The HD GHG Phase 2 vocational vehicle program also includes a special regulatory subcategory called vocational tractors, which covers vehicles that are technically tractors but generally operate more like vocational vehicles than line-haul tractors. These vocational tractors include those designed to operate off-road and in certain intra-city delivery routes.

The existing HD GHG Phase 2 CO₂ standards for vocational vehicles are based on the performance of a wide array of control technologies. In particular, the HD GHG Phase 2 vocational vehicle standards recognize detailed characteristics of vehicle powertrains and drivelines. Driveline improvements present a significant opportunity for reducing fuel consumption and CO₂ emissions from vocational vehicles. However, there is no single package of driveline technologies that will be equally suitable for all vocational vehicles, because there is an extremely broad range of driveline configurations available in the market. This is due in part to the variety of final vehicle build configurations, ranging from a purpose-built custom chassis to a commercial chassis that may be intended as a multi-purpose stock vehicle. Furthermore, the wide range of applications and driving patterns of these vocational vehicles leads manufacturers to offer a variety of drivelines, as each performs differently in use.

In the final HD GHG Phase 2 rule, we recognized the diversity of vocational vehicle applications by setting unique CO₂ emission standards evaluated over composite drive cycles for 23 different regulatory subcategories. The program includes vocational vehicle standards that allow the technologies that perform best at highway speeds and those that perform best in urban driving to each be properly recognized over appropriate drive cycles, while avoiding potential unintended results of forcing vocational vehicles that are designed to serve in different applications to be measured against a single drive cycle. The vehicle CO₂ emissions are evaluated using EPA's Greenhouse Gas Emissions Model (GEM) over three drive cycles, where the composite weightings vary by subcategory, with the intent of balancing the competing pressures to recognize the varying performance of technologies, serve the wide range of customer needs, and maintain a workable regulatory program. The HD GHG Phase 2 primary vocational standards, therefore, contain subcategories for Regional, Multi-purpose, and Urban drive cycles in each of the three weight classes (Light Heavy-Duty (Class 2b-5), Medium Heavy-Duty (Class 6–7) and Heavy Heavy-Duty (Class 8)), for a total of nine unique subcategories. These nine subcategories apply for compression-ignition (CI) vehicles. We separately, but similarly, established six subcategories of spark-ignition (SI) vehicles. In other words, there are 15 separate numerical performance-based emission standards for each model year.

EPA also established optional custom chassis categories in the Phase 2 rule in recognition of the unique technical characteristics of these applications. These categories also recognize that many manufacturers of these custom chassis are not full-line heavy-duty vehicle companies and thus do not have the same flexibilities as other firms in the use of the Phase 2 program emissions averaging program which could lead to challenges in meeting the standards EPA established for the overall vocational vehicle and combination tractor program. We therefore established optional custom chassis CO₂ emission standards for Motorhomes, Refuse Haulers, Coach Buses, School Buses, Transit Buses, Concrete Mixers, Mixed Use Vehicles, and Emergency Vehicles. In total, EPA set CO₂ emission standards for 15 subcategories of vocational vehicles and eight subcategories of specialty vehicle types for a total of 23 vocational vehicle subcategories.

The HD GHG Phase 2 standards phase in over a period of seven years, beginning with MY 2021. The HD GHG Phase 2 program progresses in three-year stages with an intermediate set of standards in MY 2024 and final standards in MY 2027 and later. In the HD GHG Phase 2 final rule, we identified a potential technology path for complying with each of the three increasingly stringent stages of the HD GHG Phase 2 program standards. These standards are based on the performance of more efficient engines, workday idle reduction technologies, improved transmissions including mild hybrid powertrains, axle technologies, weight reduction, electrified accessories, tire pressure systems, and tire rolling resistance improvements. We developed the Phase 2 vocational vehicle standards using the methodology where we applied fleet average technology mixes to fleet average baseline vehicle configurations, and each average baseline and technology mix was unique for each vehicle subcategory. When the HD GHG Phase 2 final rule was promulgated in 2016, we established CO₂ standards on the premise that electrification of the heavy-duty market would occur in the future but was unlikely to occur at significant sales volumes in the timeframe of the program. As a result, the Phase 2 vocational vehicle CO₂ standards were not in any way premised on the application of ZEV technologies. Instead, we finalized BEV, PHEV, and FCEV advanced technology credit multipliers within the HD GHG ABT program to incentivize a transition to these technologies (see Section III of this preamble for further discussion on this program and proposed changes). Details regarding the HD GHG Phase 2 standards can be found in the HD GHG Phase 2 final rule preamble, and the HD GHG Phase 2 vocational vehicle standards are codified at 40 CFR part 1037.

2. Combination Tractors

The tractor regulatory structure is attribute-based in terms of dividing the tractor category into ten subcategories based on the tractor's weight rating, cab configuration, and roof height. The tractors are subdivided into three weight ratings—Class 7 with a gross vehicle weight rating (GVWR) of 26,001 to 35,000 pounds; Class 8 with a GVWR over 33,000 pounds; and Heavy-haul with a gross combined weight rating of greater than or equal to 120,000 pounds. The Class 7 and 8 tractor cab configurations are either day cab or sleeper cab. Day cab tractors are typically used for shorter haul operations, whereas sleeper cabs are often used in long haul operations. EPA set CO₂ emission standards for 10 tractor subcategories.

Similar to the vocational program, implementation of the HD GHG Phase 2 tractor standards began in MY 2021 and will be fully phased in for MY 2027. In the HD GHG Phase 2 final rule, EPA analyzed the feasibility of achieving the CO₂ standards and identified technology pathways for achieving the standards. The existing HD GHG Phase 2 CO₂ emission standards for combination tractors reflect reductions that can be achieved through improvements in the tractor's powertrain, aerodynamics, tires, idle reduction, and other vehicle systems as demonstrated using GEM. As we did for vocational vehicles, we developed a potential technology package for each of the tractor subcategories that represented a fleet average application of a mix of technologies to demonstrate the feasibility of the standard for each MY. EPA did not premise the HD GHG Phase 2 CO₂ tractor emission standards on application of hybrid powertrains or ZEV technologies. However, we predicted some limited use of these technologies in MY 2021 and beyond and we finalized BEV, PHEV, and FCEV advanced technology credit multipliers within the HD GHG ABT program to incentivize a transition to these technologies (see Section III of this preamble for further discussion on this program and proposed changes). More details can be found in the HD GHG Phase 2 final rule preamble, and the HD GHG Phase 2 tractor standards are codified at 40 CFR part 1037.

3. Heavy-Duty Engines

In HD GHG Phase 1, we developed a regulatory structure for CO₂, nitrous oxide (N2O), and methane (CH4) emission standards that apply to the engine, separate from the HD vocational vehicle and tractor. The regulatory structure includes separate standards for spark-ignition engines (such as gasoline engines) and compression-ignition engines (such as diesel engines), and for heavy heavy-duty (HHD), medium heavy-duty (MHD) and light heavy-duty (LHD) engines, that also apply to alternative fuel engines. We also used this regulatory structure for HD engines in HD GHG Phase 2. More details can be found in the HD GHG Phase 2 final rule preamble, and the HD GHG Phase 2 engine standards are codified at 40 CFR part 1036.

4. Heavy-Duty Vehicle Average, Banking, and Trading Program

Beginning in HD GHG Phase 1, EPA adopted an averaging, banking, and trading (ABT) program for CO₂ emission credits that allows ABT within a vehicle weight class. For the HD GHG Phase 2 ABT program, the three credit averaging sets for HD vehicles are Light Heavy-Duty Vehicles, Medium Heavy-Duty Vehicles, and Heavy Heavy-Duty Vehicles. This approach allows ABT between CI-powered vehicles, SI-powered vehicles, BEVs, FCEVs, and hybrid vehicles in the same weight class, which have the same regulatory useful life. Although the vocational vehicle emission standards are subdivided by Urban, Multi-purpose, and Regional regulatory subcategories, credit exchanges are currently allowed between them within the same weight class. However, these averaging sets currently exclude vehicles certified to the separate optional custom chassis standards. Finally, the ABT program currently allows credits to exchange between vocational vehicles and tractors within a weight class.

ABT is commonly used by vehicle manufacturers for the HD GHG Phase 2 program. In MY 2022, 93 percent of the vehicle families (256 out of 276 families) certified used ABT. Similarly, 29 out of 40 manufacturers in MY 2022 used ABT to certify some or all of their vehicle families. Most of the manufacturers that did not use ABT produced vehicles that were certified to the optional custom chassis standards where the banking and trading components of ABT are not allowed, and averaging is limited.

D. Vehicle Technologies

As explained in Section ES.B, EPA is both proposing to revise the MY 2027 HD vehicle CO₂ emission standards and proposing new CO₂ emission standards that phase in annually from MY 2028 through 2032 for HD vocational vehicles and tractors. We are proposing that these Phase 3 vehicle standards are appropriate and feasible, including consideration of cost of compliance and other factors, for their respective MYs and vehicle subcategories through technology improvements in several areas. To support the feasibility and appropriateness of the proposed standards, we evaluated each technology and estimated a potential technology adoption rate in each vehicle subcategory per MY (our technology packages) that EPA projects is achievable based on nationwide production volumes, considering lead time, technical feasibility, cost, and other factors. At the same time, the proposed standards are performance-based and do not mandate any specific technology for any manufacturer or any vehicle subcategory. The following subsections describe the GHG emission-reducing technologies for HD vehicles considered in the proposal, including those for HD vehicles with ICE (Section II.D.1), BEVs (Section II.D.2), and FCEVs (Section II.D.3), as well as a summary of the technology assessment that supports the feasibility of the proposed Phase 3 standards (Section II.D.4) and the primary inputs we used in our new technology assessment tool, Heavy-Duty Technology Resource Use Case Scenario (HD TRUCS), that we developed to evaluate the design features needed to meet the power and energy demands of various HD vehicles when using ZEV technologies, as well as costs related to manufacturing, purchasing and operating ICE and ZEV technologies (Section II.D.5).

We are not proposing changes to the existing Phase 2 GHG emission standards for HD engines and are not reopening those standards in this rulemaking. As noted in the following section and DRIA Chapter 1.4, there are technologies available that can reduce GHG emissions from HD engines, and we anticipate that many of them will be used to meet the MY 2024 and MY 2027 CO₂ emission standards, while development is underway to meet the new low NO_X standards for MY 2027. At this time, we believe that additional GHG reductions would be best driven through more stringent vehicle-level CO₂ emission standards as we are proposing in this rulemaking, which also account for the engine's CO₂ emissions, instead of also proposing new CO₂ emission standards that apply to heavy-duty engines.

1. Technologies To Reduce GHG Emissions From HD Vehicles With ICEs

The CO₂ emissions of HD vehicles vary depending on the configuration of the vehicle. Many aspects of the vehicle impact its emissions performance, including the engine, transmission, drive axle, aerodynamics, and rolling resistance. For this proposed rule, as we did for HD Phase 1 and Phase 2, we are proposing more stringent CO₂ emissions standards for each of the regulatory subcategories based on the performance of a package of technologies that reduce CO₂ emissions. And in this rule, we developed technology packages that include both ICE vehicle and ZEV technologies.

For each regulatory subcategory, we selected a theoretical ICE vehicle with CO₂ -reducing technologies to represent the average MY 2027 vehicle that meets the existing MY 2027 Phase 2 standards. These vehicles are used as baselines from which to evaluate costs and effectiveness of additional technologies and more stringent standards on a per-vehicle basis. The MY 2027 technology package for tractors include technologies such as improved aerodynamics; low rolling resistance tires; tire inflation systems; efficient engines, transmissions, and drivetrains, and accessories; and extended idle reduction for sleeper cabs, The GEM inputs for the individual technologies that make up the fleet average technology package that meets the existing MY 2027 CO₂ tractor emission standards are shown in Table II–1. The comparable table for vocational vehicles is shown in Table II–2. The technology package for vocational vehicles include technologies such as low rolling resistance tires; tire inflation systems; efficient engines, transmissions, and drivetrains; weight reduction; and idle reduction technologies. Note that the HD GHG Phase 2 standards are performance-based; EPA does not require this specific technology mix, rather the technologies shown in Table II–1 and II–2 are potential pathways for compliance.

Technologies exist today and continue to evolve to improve the efficiency of the engine, transmission, drivetrain, aerodynamics, and tire rolling resistance in HD vehicles and therefore reduce their CO₂ emissions. As discussed in the preamble to the HD GHG Phase 2 program and shown in Table II–1 and Table II–2, there are a variety of such technologies. In developing the Phase 2 CO₂ emission standards, we developed technology packages that were premised on technology adoption rates of less than 100 percent. There may be an opportunity for further improvements and increased adoption through MY 2032 for many of these technologies included in the HD GHG Phase 2 technology package used to set the existing MY 2027 standards. For example, DRIA Chapter 1.4 provides an update to tractor aerodynamic designs developed by several of the manufacturers as part of the DOE SuperTruck program that demonstrate aerodynamics that are better than those used in the existing MY 2027 standards' HD GHG Phase 2 technology package for high roof sleeper cab tractors in MYs beyond 2027.

The heavy-duty industry has also been developing hybrid powertrains, as described in DRIA Chapter 1.4.1.1. Hybrid powertrains consist of an ICE as well as an electric drivetrain and some designs also incorporate plug-in capability. Hybrid powered vehicles may provide CO₂ emission reductions through the use of downsized engines, recover energy through regenerative braking system that is normally lost while braking, and provide additional engine-off operation during idling and coasting. Hybrid powertrains are available today in a number of heavy-duty vocational vehicles including passenger van/shuttle bus, transit bus, street sweeper, refuse hauler, and delivery truck applications. Heavy-duty hybrid vehicles may include a power takeoff (PTO) system that is used to operate auxiliary equipment, such as the boom/bucket on a utility truck or the water pump on a fire truck.

Furthermore, manufacturers may develop new ICE vehicle technologies through the MY 2032 timeframe. An example of a new technology under development that would reduce GHG emissions from HD vehicles with ICEs is hydrogen-fueled internal combustion engines (H2–ICE). These engines are currently in the prototype stage of innovation for HD vehicles, but have also been demonstrated as technically feasible in the past in the LD fleet. H2–ICE is a technology that produces zero hydrocarbon (HC), carbon monoxide (CO), and CO₂ engine-out emissions.

H2–ICE are similar to existing internal combustion engines and could leverage the technical expertise manufacturers have developed with existing products. H2–ICEs use many of the same components as existing internal combustion engines for many key systems. Similarly, H2–ICE vehicles could be built on the same assembly lines as existing ICE vehicles, by the same workers and with many of the same suppliers.

Though many engine components would be similar between H2–ICE and, for example, a comparable existing diesel-fueled ICE, components such as the cylinder head, piston and piston rings would be unique to H2–ICE as well as intake and exhaust valves and seats to control H2 leakage during combustion. Fuel systems would require changes to fuel injectors and the fuel delivery system. The H2–ICE aftertreatment systems may be simpler than today's comparable diesel-fueled ICEs. They likely would not require the use of a diesel oxidation catalyst (DOC) or a diesel particulate filter (DPF) system. NO_X emissions are still present in the H2–ICE exhaust and therefore a selective catalyst reduction (SCR) system would likely still be required, though smaller in size than an existing comparable diesel-fueled ICE aftertreatment system. The use of lean air-fuel ratios, not exhaust gas recirculation (EGR), would be the most effective way to control NO_X in H2 combustion engines. EGR is less effective with H2 due to the absence of CO₂ in the exhaust gas. Additional information regarding H2–ICE can be found in the DRIA Chapter 1.4.2.

One key significant difference between an existing comparable diesel-fueled ICE and a H2–ICE is the fuel storage tanks. The hydrogen storage tanks that would replace existing ICE fuel tanks are significantly more expensive. The fuel tanks used by H2–ICE would be the same as those used by a FCEV and may be either compressed storage (350 or 700 Bar) or cryogenic (storage temperatures reaching −253 degrees Celsius). Please refer to Section II.D.3 for the discussion regarding H2 fuel storage tanks. Furthermore, like FCEVs, H2 refueling infrastructure would be required for H2–ICE vehicles.

We request comment on whether we should include additional GHG-reducing technologies and/or higher levels of adoption rates of existing technologies for ICE vehicles in our technology assessment for the final rule.

2. HD Battery Electric Vehicle Technology

The HD BEV market has been growing significantly since MY 2018. DRIA Chapter 1.5 includes BEV vehicle information on over 170 models produced by over 60 manufacturers that cover a broad range of applications, including school buses, transit buses, straight trucks, refuse haulers, vans, tractors, utility trucks, and others, available to the public through MY 2024.

The battery electric propulsion system includes a battery pack that provides the energy to the motor that moves the vehicle. In this section, and in DRIA Chapter 1.5.1 and 2.4, we discuss battery technology that can be found in both BEVs and FCEVs. We request comment on our assessment of heavy-duty battery designs, critical materials, and battery manufacturing.

i. Batteries Design Parameters

Battery design involves considerations related to cost and performance including specific energy and power, energy density, temperature impact, durability, and safety. These parameters typically vary based on the cathode and anode materials, and the conductive electrolyte medium at the cell level. Different battery chemistries have different intrinsic values. Here we provide a brief overview of the different energy and power parameters of batteries and battery chemistries.

a. Battery Energy and Power Parameters

Specific energy and power and energy density are a function of how much energy or power can be stored per unit mass (in Watt-hour per kilogram (Wh/kg) or watt per kilogram (W/kg)) or volume (in Watt-hour per liter (Wh/L)). Therefore, for a given battery weight or mass, the energy (in kilowatt-hour or kWh) can be calculated. For example, a battery with high specific energy and a lower weight may yield the same amount of energy as a chemistry with a lower specific energy and more weight.

Battery packs have a “nested” design where a group of cells are combined to make a battery module and a group of modules are combined to make a battery pack. Therefore, the battery systems can be described on the pack, module, and cell levels. Design choices about the different energy and power capacities to prioritize in a battery can depend on its battery chemistry. Common battery chemistries today include nickel-manganese-cobalt (NMC), nickel-cobalt-aluminum (NCA), and iron-phosphate (LFP) based-chemistries. Nickel-based chemistries typically have higher gravimetric and volumetric energy densities than iron phosphate-based chemistries. Since energy or power is only housed at the chemistry level, any additional mass such as the cell, module, and pack casings will only add to the weight of the battery without increasing the energy of the overall system. Therefore, some pack producers have eliminated the module in favor of a “cell-to-pack” design in recent years.

External factors, especially temperature, can have a strong influence on the performance of the battery. Heavy-duty BEVs today include thermal management systems to keep the battery operating within a desired temperature range, which is commonly referred to as conditioning of the battery. Therefore, while operating a vehicle in cold temperatures, some of the battery energy is used to heat both the battery packs and the vehicle interior. Cold temperatures, in particular, can result in reduced mobility of the lithium ions in the liquid electrolyte inside the battery; for the driver, this may mean lower range. Battery thermal management is also used during hot ambient temperatures to keep the battery from overheating. We consider and account for the energy required for battery thermal management in our analysis, as discussed in Section II.D.5.ii.b.

b. Battery Durability

Another important battery design consideration is the durability of the battery. Durability is frequently associated with cycle life, where cycle life is the number of times a battery can fully charge and discharge before the battery is no longer used for its original purpose. In 2015 the United Nations Economic Commission for Europe (UN ECE) began studying the need for a Global Technical Regulation (GTR) governing battery durability in light-duty vehicles. In 2021 it finalized United Nations Global Technical Regulation No. 22, “In-Vehicle Battery Durability for Electrified Vehicles,” or GTR No. 22, which provides a regulatory structure for contracting parties to set standards for battery durability in light-duty BEVs and PHEVs. Likewise, although not finalized, the UN ECE GTR working group began drafting language for HD BEVs and hybrid electric vehicles. Loss of electric range could lead to a loss of utility, meaning electric vehicles could be driven less and therefore displace less distance travelled than might otherwise be driven in conventional vehicles. Furthermore, a loss in utility could also dampen purchaser sentiment.

For batteries that are used in HD BEVs, the state-of-health (SOH) is an important design factor. The environmental performance of electrified vehicles may be affected by excess degradation of the battery system over time. However, the durability of a battery is not limited to the cycling of a battery, there are many phenomena that can impact the duration of usability of a battery. As a battery goes through charge and discharge cycles, the SOH of the battery decreases. Capacity fade, increase in internal resistance, and voltage loss, for example, are other common metrics to measure the SOH of a battery. These parameters together help better understand and define the longevity or durability of the battery. The SOH and, in turn, the cycle life of the battery is determined by both the chemistry of the battery as well as external factors including temperature. The rate at which the battery is discharged as well as the rate at which it is charged will also impact the SOH of the battery. Lastly, calendar aging, or degradation of the battery while not in use, can also contribute to the deterioration of the battery.

There are a number of ways to improve and prolong the battery life in a vehicle. We took considerations on maintaining the battery temperature while driving by applying additional energy required for conditioning the battery. Furthermore, battery size is increased by 20 percent to accommodate additional energy that may be required resulting from loss of capacity over time.

c. HD BEV Safety Assessment

HD BEV systems must be designed to always maintain safe operation. As with any onroad vehicle, BEVs must be robust while operating in temperature extremes as well as rain and snow. The BEV systems must be designed for reasonable levels of immersion, including immersion in salt water or brackish water. BEV systems must also be designed to be crashworthy and limit damage that compromises safety. If the structure is compromised by a severe impact, the systems must provide first responders with a way to safely conduct their work at an accident scene. The HD BEV systems must be designed to ensure the safety of users, occupants, and the general public in their vicinity.

In DRIA Chapter 1.5.4, we discuss the industry codes and standards used by manufacturers that guide safe design and development of heavy-duty BEVs, including those for developing battery systems and charging systems that protect people and the equipment. These standards have already been developed by the industry and are in place for manufacturers to use today to develop current and future products. The standards guide the design of BEV batteries to allow them to safely accept and deliver power for the life of the vehicle. The standards provide guidance to design batteries that also handle vibration, temperature extremes, temperature cycling, water, and mechanical impact from items such as road debris. For HD BEVs to uphold battery/electrical safety during and after a crash, they are designed to maintain high voltage isolation, prevent leakage of electrolyte and volatile gases, maintain internal battery integrity, and withstand external fire that could come from the BEV or other vehicle(s) involved in a crash. NHTSA continues work on battery safety requirements and extend the applicability of FMVSS No. 305 to HD vehicles and would align with the existing Global Technical Regulation (GTR) No. 20 to include safety requirements during normal operation, charging, and post-crash. We request comment on our assessment that HD BEVs can be designed to maintain safety.

ii. Assessment of Battery Materials and Production

Although the market share of light-duty and heavy-duty ZEVs in the United States is already growing, EPA recognizes that the proposed standards may accelerate this trend. Assessing the feasibility of incremental penetrations of ZEVs that may result from the proposed standards includes consideration of the readiness of the supply chain to provide the required quantities of critical minerals, components, and battery manufacturing capacity. This section provides a general review of how we considered supply chain and manufacturing in this analysis, the sources we considered, and how we used this information in the analysis. It also provides a high-level discussion of the security implications of increased demand for minerals and other commodities used to manufacture ZEVs.

In developing these standards, we considered the ability for global and domestic manufacturing and critical mineral capacity to respond to the projected demand for ZEVs that manufacturers may choose to produce to comply with the proposed standards. As described in this section, we consulted with industry and government agency sources (including DOE, U.S. Geological Survey (USGS), and several analysis firms) to collect information on production capacity, price forecasts, global mineral markets, and related topics, and have considered this information to inform our assumptions about future manufacturing capabilities and costs. We have included consideration of the influence of critical minerals and materials availability as well as vehicle and battery manufacturing capacities on the production of ZEVs.

We believe that the proposed rate of stringency is appropriate in light of this assessment. It is also our assessment that increased vehicle electrification in the United States will not lead to a critical long term dependence on foreign imports of minerals or components, nor that increased demand for these products will become a vulnerability to national security. First, in many cases the reason that these products are often sourced from outside of the United States is not because the products cannot be produced in the U.S., but because other countries have already invested in developing a supply chain for their production. Moreover, the United States will likely develop a domestic supply chain for these products because U.S. manufacturers will need to remain competitive in a global market where electrification is already proceeding rapidly. Second, many vehicle manufacturers, suppliers, startups, and related industries have already recognized the need for increased domestic production capacity as a business opportunity, and are basing business models on building out various aspects of the supply chain. Third, Congress and the Administration have taken significant steps to accelerate this activity by funding, facilitating, and otherwise promoting the rapid growth of U.S. supply chains for these products through the Inflation Reduction Act, the Bipartisan Infrastructure Law, and numerous Executive Branch initiatives. EPA has confidence that these efforts are effectively addressing supply chain concerns. Finally, utilization of critical minerals is different from the utilization of foreign oil, in that oil is consumed as a fuel while minerals become a constituent of manufactured vehicles. Minerals that are imported for vehicle production remain in the vehicle, and can be reclaimed through recycling. Each of these points will be expanded in more detail in the sections below.

We request comment on our assessment and data to support our assessment of battery critical raw materials and battery production for the final rule.

a. Battery Critical Raw Materials

Critical minerals are generally considered to include a large diversity of products, ranging from relatively plentiful materials that are constrained primarily by production capacity and refining, such as aluminum, to those that are both relatively rare and costly to process, such as the rare-earth metals that are used in magnets for permanent-magnet synchronous motors (PMSMs) that are used as the electric motors to power heavy-duty ZEVs and some semiconductor products. Extraction, processing, and recycling of certain critical minerals (such as lithium, cobalt, nickel, magnesium, graphite and rare earth metals) are also an important part of the supply chain supporting the production of battery components.

These minerals are also experiencing increasing demand across many other sectors of the global economy, not just the transportation industry, as the world seeks to reduce carbon emissions. As with any emerging technology, a transition period must take place in which a robust supply chain develops to support production of these products. At the present time, they are commonly sourced from global suppliers and do not yet benefit from a fully developed domestic supply chain. As demand for these materials increases due to increasing production of ZEVs, current mining and processing capacity will expand.

The U.S. Geological Survey lists 50 minerals as “critical to the U.S. economy and national security.” The Energy Act of 2020 defines a “critical mineral” as a non-fuel mineral or mineral material essential to the economic or national security of the United States and which has a supply chain vulnerable to disruption. Critical minerals are not necessarily short in supply, but are seen as essential to the manufacture of products that are important to the economy or national security. The risk to their availability may stem from geological scarcity, geopolitics, trade policy, or similar factors.

Emission control catalysts for ICE vehicles utilize critical minerals including cerium, palladium, platinum, and rhodium. These are also required for hybrid vehicles due to the presence of the ICE. Critical minerals most relevant to lithium-ion battery production include cobalt, graphite, lithium, manganese, and nickel, which are important constituents of electrode active materials, their presence and relative amounts depending on the chemistry formulation. Aluminum is also used for cathode foils and in some cell chemistries. Rare-earth metals are used in permanent-magnet electric machines, and include several elements such as dysprosium, neodymium, and samarium.

Some of the electrification technologies that use critical minerals have alternatives that use other minerals or eliminate them entirely. For these, vehicle manufacturers in some cases have some flexibility to modify their designs to reduce or avoid use of minerals that are difficult or expensive to procure. For example, in some ZEV battery applications it is feasible and increasingly common to employ an iron phosphate cathode which has lower energy density but does not require cobalt, nickel, or manganese. Similarly, rare earths used in permanent-magnet electric machines have potential alternatives in the form of ferrite or other advanced magnets, or the use of induction machines or advanced externally excited motors, which do not use permanent magnets.

This discussion therefore focuses on minerals that are most critical for battery production, including nickel, cobalt, graphite, and lithium.

Availability of critical minerals for use in battery production depends on two primary considerations: production of raw minerals from mining (or recycling) operations and refining operations that produce purified and processed substances (precursors, electrolyte solutions, and finished electrode powders) made from the raw minerals, that can then be made into battery cells.

As shown in Figure II–1, in 2019 about 50 percent of global nickel production occurred in Indonesia, Philippines, and Russia, with the rest distributed around the world. Nearly 70 percent of cobalt originated from the Democratic Republic of Congo, with some significant production in Russia and Australia, and about 20 percent in the rest of the world. More than 60 percent of graphite production occurred in China, with significant contribution from Mozambique and Brazil for another 20 percent. About half of lithium was mined in Australia, with Chile accounting for another 20 percent and China about 10 percent.

According to the 100-day review under E.O. on America's Supply Chains (E.O. 14017), of the major actors in mineral refining, 60 percent of lithium refining occurred in China, with 30 percent in Chile and 10 percent in Argentina. 72 percent of cobalt refining occurred in China, with another 17 percent distributed among Finland, Canada, and Norway. 21 percent of Class 1 nickel refining occurred in Russia, with 16 percent in China, 15 percent in Japan and 13 percent in Canada. Similar conclusions were reached in an analysis by the International Energy Agency, shown in Figure II–2.

Currently, the United States is lagging behind much of the rest of the world in critical mineral production. Although the United States has nickel reserves, and opportunity also exists to recover significant nickel from mine waste remediation and similar activities, it is more convenient for U.S. nickel to be imported from other countries, with 68 percent coming from Canada, Norway, Australia, and Finland, countries with which the United States has good trade relations. According to the USGS, ample reserves of nickel exist in the United States and globally, potentially constrained only by processing capacity. The United States has numerous cobalt deposits but few are developed while some have produced cobalt only in the past; about 72 percent of U.S. consumption is imported. Similar observations may be made about graphite and lithium. Significant lithium deposits do exist in the United States in Nevada and California as well as several other locations, and are currently the target of development by suppliers and vehicle manufacturers. U.S. deposits of natural graphite deposits also exist but graphite has not been produced in the United States since the 1950s and significant known resources are largely undeveloped.

Although predicting mineral supply and demand into the future is challenging, it is possible to identify general trends likely to occur in the future. As seen in Figure II–3 and Figure II–4, preliminary projections prepared by Li-Bridge for DOE, and presented to the Federal Consortium for Advanced Batteries (FCAB) in November 2022, indicate that global supplies of cathode active material (CAM) used as a part of the cathode manufacturing process and lithium chemical product are expected to be sufficient through 2035.

The most recent information indicates that the market is responding robustly to demand and lithium supplies are expanding as new resources are characterized, projects continue through engineering economic assessments, and others begin permitting or construction. For example, in October 2022, the IEA projected that global Lithium Carbonate Equivalent (LCE) production from operating mines and those under construction may sufficiently meet primary demand until 2028 under the Stated Policies Scenario. In December 2022, BNEF projected lithium mine production can meet end-use demand until 2028. Notably, the BNEF data is not exhaustive and includes only three U.S. projects: Silver Peak (phase I and II), Rhyolite Ridge (phase I), and Carolina Lithium (phase I). Additionally, in March 2023 DOE communicated to EPA that DOE and ANL have identified 21 additional lithium production projects in the United States in addition to the three identified in the December 2022 BNEF data. Were they to achieve commercial operations, the 24 U.S. projects would produce an additional 1,000 kilotons per year LCE not accounted for in the December BNEF analysis, and suggests that lithium supplies would meet the BNEF Net-Zero demand projection.

In addition, the European Union is seeking to promote rapid development of Europe's battery supply chains by considering targeted measures such as accelerating permitting processes and encouraging private investment. To these ends the European Parliament proposed a Critical Raw Materials Act on March 16, 2023, which includes these and other measures to encourage the development of new supplies of critical minerals not currently anticipated in market projections. In DRIA 1.5.1.3 we detail these and many other examples that demonstrate how momentum has picked up in the lithium market since IEA's May 2022 report. For more discussion, please see DRIA 1.5.1.3.

Despite recent short-term fluctuations in price, the price of lithium is expected to stabilize at or near its historical levels by the mid- to late-2020s. This perspective is also supported by proprietary battery price forecasts by Wood Mackenzie that include the predicted effect of temporarily elevated mineral prices. This is consistent with the BNEF battery price outlook 2022 which expects battery prices to start dropping again in 2024, and BNEF's 2022 Battery Price Survey which predicts that average pack prices should fall below $100/kWh by 2026. Taken together these outlooks support the perspective that lithium is not likely to encounter a critical shortage as supply responds to meet growing demand.

As described in the following section, the development of mining and processing capacity in the United States is a primary focus of efforts on the part of both industry and the Administration toward building a robust domestic supply chain for electrified vehicle production, and will be greatly facilitated by the provisions of the BIL and the IRA as well as large private business investments that are already underway and continuing.

b. Battery Market and Manufacturing Capacity

Battery systems can be described on the pack, module, and cell levels. A pack typically consists of a group of modules, a module consists of a group of cells, and cells consist of the half-cell electrodes. Cells can be directly supplied to the manufacturer to be assembled into modules and packs; alternatively, cell producers may assemble cells into modules before sending the modules to another supplier to be assembled into a pack, before then sending it to the OEM for final assembly. While there are hundreds of reported automotive battery cell producers, major LD automakers use batteries produced by a handful of battery cell manufacturers. These suppliers include LG Chem, Samsung SDI, SK Innovation, Panasonic/Tesla, Contemporary Amperex Technology Co., Limited (CATL) and BYD. A 2021 report developed by DOE's Argonne National Lab (ANL) found significant growth in the annual battery supply between 2010 and 2020.

In both the LD and HD industry sectors, there is a meaningful distinction between 1) battery cell suppliers, and 2) battery pack assemblers who refer to themselves as battery producers while using cells produced by a different cell supplier, in understanding how impacts from the increased production volumes of cells and costs of cells in both industries flow to these different types of suppliers. The cost of cells occupies a significant percent of the final pack cost, and cell costs are inversely proportional to cell production volume. In other words, increased cell production volume lowers the cost of battery cells, which in turn lowers the overall pack cost. Thus, though the LD sector demand for automotive batteries is significantly outpacing the demand for vehicle batteries in the HD sector, the battery cell industry for both sectors will likely be significantly influenced by the demand in the LD industry.

Although most global battery manufacturing capacity is currently located outside the U.S., most of the batteries and cells present today in the domestic EV fleet were manufactured in the United States We expect domestic manufacturing of batteries and cells to increase considerably over the coming decade. According to the Department of Energy, at least 13 new battery plants are expected to become operational in the United States within the next four years. Among these 13 new battery plants include the following activities by battery suppliers and vehicle manufacturers. In partnership with SK Innovation, Ford is building three large new battery plants in Kentucky and Tennessee. General Motors is partnering with LG Energy Solutions to build another three battery cell manufacturing plants in Tennessee, Michigan, and Ohio, and there are discussions about another plant in Indiana. LG Chem has also announced plans for a cathode material production facility in Tennessee, said to be sufficient to supply 1.2 million high-performance electric vehicles per year by 2027. CATL is considering construction of plants in Arizona, Kentucky, and South Carolina. In addition, CATL is partnering with Daimler Truck to expand their global partnership to producm ion batteries for their all electric long haul heavy duty trucks starting 2024 to 2030. Panasonic, already partnering with Tesla for its factories in Texas and Nevada, is planning two new factories in Oklahoma and Kansas. Furthermore, Tesla is also planning a $3.6 billion expansion to their Nevada Gigafactory to mass produce all electric semi trucks. Toyota plans to be operational with a plant in Greensboro, North Carolina in 2025, and Volkswagen in Chattanooga, Tennessee at about the same time. According to S&P Global, announcements such as these could result in a U.S. manufacturing capacity of 382 GWh by 2025, and 580 GWh by 2027, up from roughly 60 GWh today. More recently, the Department of Energy estimates that recent plant announcements for North America to date could enable an estimated 838 GWh of capacity by 2025, 896 GWh by 2027, and 998 GWh by 2030, the vast majority of which is cell manufacturing capacity.

The expected HD battery capacity demand based on this proposed rule would be 17 GWh in MY 2027 and grow to 36 GWh by MY 2032 (as described in DRIA 2.8.3.1), which is well below the expected manufacturing capacity for this time frame. It should be noted that the projected U.S. HD battery demand would be only a fraction of total U.S. battery demand. In comparison, we project in the Light- and Medium-Duty Multipollutant Emissions Standards Proposed Rule that the annual battery production required for the light-duty fleet would be slightly less than 900 GWh in MY 2030, and stabilize at around 1,000 GWh per year for MY 2031 and beyond. Therefore, between the two proposed highway motor vehicle rules, the U.S. market could require 940 GWh of battery capacity by 2030 and 1050 GWh of battery capacity by 2032. DOE estimates plant announcements of ~1,000 GWh by 2030; furthermore, the battery market is an international market where IEA projects 3.7 terrawatt hours (TWh) of battery globally by 2030 in their “Sustainable Development Scenario”

In addition, the IRA and the BIL are providing significant support to accelerate these efforts to build out a U.S. supply chain for mineral, cell, and battery production. The IRA offers sizeable incentives and other support for further development of domestic and North American manufacture of these components. According to the Congressional Budget Office, an estimated $30.6 billion will be realized by manufacturers through the Advanced Manufacturing Production Credit, which includes a tax credit to manufacturers for battery production in the United States. According to one third-party estimate based on information from Benchmark Mineral Intelligence, the recent increase in U.S. battery manufacturing plant announcements could increase this figure to $136 billion or more. Another $6.2 billion or more may be realized through expansion of the Advanced Energy Project Credit, a 30 percent tax credit for investments in projects that reequip, expand, or establish certain energy manufacturing facilities. Together, these provisions create a strong motivation for manufacturers to support the continued development of a North American supply chain and already appear to be proving influential on the plans of manufacturers to procure domestic or North American mineral and component sources and to construct domestic manufacturing facilities to claim the benefits of the act.

In addition, the BIL provides $7.9 billion to support development of the domestic supply chain for battery manufacturing, recycling, and critical minerals. Notably, it supports the development and implementation of a $675 million Critical Materials Research, Development, Demonstration, and Commercialization Program administered by the Department of Energy (DOE), and has created numerous other programs in related areas, such as for example, critical minerals data collection by the USGS. Provisions extend across several areas including critical minerals mining and recycling research, USGS energy and minerals research, rare earth elements extraction and separation research and demonstration, and expansion of DOE loan programs in critical minerals and zero-carbon technologies. The Department of Energy is working to facilitate and support further development of the supply chain, by identifying weaknesses for prioritization and rapidly funding those areas through numerous programs and funding opportunities. According to a final report from the Department of Energy's Li-Bridge alliance, “the U.S. industry can double its value-added share by 2030 (capturing an additional $17 billion in direct value-add annually and 40,000 jobs in 2030 from mining to cell manufacturing), dramatically increase U.S. national and economic security, and position itself on the path to a near-circular economy by 2050.” The $7.9 billion provided by the BIL for U.S. battery supply chain projects represents a total of about $14 billion when industry cost matching is considered. Other recently announced projects will utilize another $40 billion in private funding. According to DOE's Li-Bridge alliance, the total of these commitments already represents more than half of the capital investment that Li-Bridge considers necessary for supply chain investment to 2030.

Further, the DOE Loan Programs Office is administering a major loans program focusing on extraction, processing and recycling of lithium and other critical minerals that will support continued market growth, through the Advanced Technology Vehicles Manufacturing (ATVM) Loan Program and Title 17 Innovative Energy Loan Guarantee Program. This program includes over $20 billion of available loans and loan guarantees to finance critical materials projects.

c. Mineral Security

As stated at the beginning of this Section II.D, it is our assessment that increased electrification in the U.S. transportation sector does not constitute a vulnerability to national security, for several reasons supported by the discussion in this preamble and in DRIA 1.5.1.2.

A domestic supply chain for battery and cell manufacturing is rapidly forming by the actions of stakeholders including vehicle manufacturers and suppliers who wish to take advantage of the business opportunities that this need presents, and by vehicle manufacturers who recognize the need to remain competitive in a global market that is shifting to electrification. It is therefore already a goal of the U.S. manufacturing industry to create a robust supply chain for these products, in order to supply not only the domestic vehicle market, but also all of the other applications for these products in global markets as the world decarbonizes.

Further, the IRA and BIL are proving to be a highly effective means by which Congress and the Administration have provided support for the building of a robust supply chain, and to accelerate this activity to ensure that it forms as rapidly as possible. An example is the work of Li Bridge, a public-private alliance committed to accelerating the development of a robust and secure domestic supply chain for lithium-based batteries. It has set forth a goal that by 2030 the United States should capture 60 percent of the economic value associated with the U.S. domestic demand for lithium batteries. Achieving this target would double the economic value expected in the United States under “business as usual” growth. More evidence of recent growth in the supply chain is found in a February 2023 report by Pacific Northwest National Laboratory (PNNL), which documents robust growth in the North American lithium battery industry.

Finally, it is important to note that utilization of critical minerals is different from the utilization of foreign oil, in that oil is consumed as a fuel while minerals become a constituent of manufactured vehicles. That is, mineral security is not a perfect analogy to energy security. Supply disruptions and fluctuating prices are relevant to critical minerals as well, but the impacts of such disruptions are felt differently and by different parties. Disruptions in oil supply or gasoline price has an immediate impact on consumers through higher fuel prices, and thus constrains the ability to travel. In contrast, supply disruptions or price fluctuations of minerals affect only the production and price of new vehicles. In practice, short-term price fluctuations do not always translate to higher production cost as most manufacturers purchase minerals via long-term contracts that insulate them to a degree from changes in spot prices. Moreover, critical minerals are not a single commodity but a number of distinct commodities, each having its own supply and demand dynamics, and some being capable of substitution by other minerals. Importantly, while oil is consumed as a fuel and thus requires continuous supply, minerals become part of the vehicle and have the potential to be recovered and recycled. Thus even when minerals are imported from other countries, their acquisition adds to the domestic mineral stock that is available for domestic recycling in the future.

Over the long term, battery recycling will be a critical component of the BEV supply chain and will contribute to mineral security and sustainability, effectively acting as a domestically produced mineral source that reduces overall reliance on foreign-sourced products. While the number of end-of-life BEV batteries available for recycling will lag the market penetration of BEVs, it is important to consider the projected growth in development of a battery recycling supply chain during the time frame of the rule and beyond.

By 2050, battery recycling could be capable of meeting 25 to 50 percent of total lithium demand for battery production. To this end, battery recycling is avery active area of research. The Department of Energy coordinates much research in this area through the ReCell Center, described as “a national collaboration of industry, academia and national laboratories working together to advance recycling technologies along the entire battery life-cycle for current and future battery chemistries.” Funding is also being disbursed as directed by the BIL, as discussed in Chapter 1.3.2 of the DRIA. A growing number of private companies are entering the battery recycling market as the rate of recyclable material becoming available from battery production facilities and salvaged vehicles has grown, and manufacturers are already reaching agreements to use these recycled materials for domestic battery manufacturing. For example, Panasonic has contracted with Redwood Materials Inc. to supply domestically processed cathode material, much of which will be sourced from recycled batteries.

Recycling infrastructure is one of the targets of several provisions of the BIL. It includes a Battery Processing and Manufacturing program, which grants significant funds to promote U.S. processing and manufacturing of batteries for automotive and electric grid use, by awarding grants for demonstration projects, new construction, retooling and retrofitting, and facility expansion. It will provide a total of $3 billion for battery material processing, $3 billion for battery manufacturing and recycling, $10 million for a lithium-ion battery recycling prize competition, $60 million for research and development activities in battery recycling, an additional $50 million for state and local programs, and $15 million to develop a collection system for used batteries. In addition, the Electric Drive Vehicle Battery Recycling and Second-Life Application Program will provide $200 million in funds for research, development, and demonstration of battery recycling and second-life applications.

The efforts to fund and build a mid-chain processing supply chain for active materials and related products will also be important to reclaiming minerals through domestic recycling. While domestic recycling can recover minerals and other materials needed for battery cell production, they commonly are recovered in elemental forms that require further midstream processing into precursor substances and active material powders that can be used in cell production. The DOE ReCell Center coordinates extensive research on development of a domestic lithium-ion recycling supply chain, including direct recycling, in which materials can be recycled for direct use in cell production without destroying their chemical structure, and advanced resource recovery which uses chemical conversion to recover raw minerals for processing into new constituents.

Currently, pilot-scale battery recycling research projects and private recycling startups have access to only limited amounts of recycling stock that originate from sources such as manufacturer waste, crashed vehicles, and occasional manufacturer recall/repair events. As ZEVs are currently only a small portion of the U.S. vehicle stock, some time will pass before vehicle scrappage can provide a steady supply of end-of-life batteries to support large-scale battery recycling. During this time, we expect that the midchain processing portion of the supply chain will continue to develop and will be able to capture much of the resources made available by the recycling of used batteries coming in from the fleet.

3. HD Fuel Cell Electric Vehicle Technology

Fuel cell technologies that run on hydrogen have been in existence for decades, though they are just starting to enter the heavy-duty transportation market. Hydrogen FCEVs are similar to BEVs in that they have batteries and use an electric motor instead of an internal combustion engine to power the wheels. Unlike BEVs that need to be plugged in to recharge, FCEVs have fuel cell stacks that use a chemical reaction involving hydrogen to generate electricity. Fuel cells with electric motors are two-to-three times more efficient than ICEs that run on gasoline or diesel, requiring less energy to fuel.

Hydrogen FCEVs do not emit air pollution at the tailpipe—only heat and pure water. With current and near-future technologies, energy can be stored more densely onboard a vehicle as gaseous or liquid hydrogen than it can as electrons in a battery. This allows FCEVs to perform periods of service between fueling events that batteries currently cannot achieve without affecting vehicle weight and limiting payload capacity. Thus, fuel cells are of interest for their potential use in heavy-duty sectors that are difficult to electrify using batteries due to range or weight limitations.

In the following sections, and in DRIA Chapter 1.7, we discuss key technology components unique to HD FCEVs. We request comment on our assessment and data to support our assessment of FCEV technology for the final rule.

i. Fuel Cell Stack

A fuel cell system is composed of a fuel cell stack and “balance of plant” (BOP) components that support the fuel cell stack ( e.g., pumps, sensors, compressors, humidifiers). A fuel cell stack is a module that may contain hundreds of fuel cell units, typically combined in series. A heavy-duty FCEV may have several fuel cell stacks to meet the power needs of a comparable ICE vehicle.

Though there are many types of fuel cell technologies, polymer electrolyte membrane (PEM) fuel cells are typically used in transportation applications because they offer high power density, therefore have low weight and volume, and can operate at relatively low temperatures. PEM fuel cells are built using membrane electrode assemblies (MEA) and supportive hardware. The MEA includes the PEM electrolyte material, catalyst layers (anode and cathode), and gas diffusion layers. Hydrogen fuel and oxygen enter the MEA and chemically react to generate electricity, which is either used to propel the vehicle or is stored in a battery to meet future power needs. The process creates excess water vapor and heat.

Key BOP components include the air supply system that provides oxygen, the hydrogen supply system, and the thermal management system. With the help of compressors and sensors, these components monitor and regulate the pressure and flow of the gases supplied to the fuel cell along with relative humidity and temperature. Similar to ICEs and batteries, PEM fuel cells require thermal management systems to control the operating temperatures. It is necessary to control operating temperatures to maintain stack voltage and the efficiency and performance of the system. There are different strategies to mitigate excess heat that comes from operating a fuel cell. For example, a HD vehicle may include a cooling system the circulates cooling fluid through the stack. Waste heat recovery solutions are also emerging. The excess heat also can be in turn used to heat the cabin, similar to ICE vehicles. Power consumed to operate BOP components can also impact the stack's efficiency.

To improve fuel cell performance, the air and hydrogen fuel that enter the system may be compressed, humidified, and/or filtered. A fuel cell operates best when the air and the hydrogen are free of contaminants, since contaminants can poison and damage the catalyst. PEM fuel cells require hydrogen that is over 99 percent pure, which can add to the fuel production cost. Hydrogen produced from natural gas tends to initially have more impurities ( e.g., carbon monoxide and ammonia, associated with the reforming of hydrocarbons) than hydrogen produced from water through electrolysis. There are standards such as ISO 14687 that include hydrogen fuel quality specifications for use in vehicles to minimize impurities.

Fuel cell durability is important in heavy-duty applications, given that vehicle owners and operators often have high expectations for drivetrain lifetimes in terms of years, hours, and miles. Fuel cells can be designed to meet durability needs, or the ability of the stack to maintain its performance over time. Considerations must be included in the design to accommodate operations in less-than-optimized conditions. For example, prolonged operation at high voltage (low power) or when there are multiple transitions between high and low voltage can stress the system. As a fuel cell system ages, a fuel cell's MEA materials can degrade, and performance and maximum power output can decline. The fuel cell can become less efficient, which can cause it to generate more excess heat and consume more fuel. DOE's ultimate long-term technology target for Class 8 HD trucks is a fuel cell lifetime of 30,000 hours, corresponding to an expected vehicle lifetime of 1.2 million miles. A voltage degradation of 10 percent at rated power ( i.e., the power level the cell is designed for) by end-of-life is considered by DOE when evaluating targets.

Currently, the fuel cell stack is the most expensive component of a heavy-duty FCEV, primarily due to the technological requirements of manufacturing rather than raw material costs. Larger production volumes are anticipated as global demand increases for fuel cell systems for HD vehicles, which could improve economies of scale. Costs are also anticipated to decline as durability improves, which could extend the life of fuel cells and reduce the need for parts replacement. Fuel cells contain PEM catalysts that typically are made using precious metals from the platinum group, which are expensive but efficient and can withstand conditions in a cell. With today's technology, roughly 50 grams of platinum may be required for a 160-kW fuel cell in a vehicle. Platinum group metals are classified as critical minerals in the DOE Critical Minerals and Materials Strategy. Efforts are underway to minimize or eliminate the use of platinum in catalysts.

ii. Fuel Cell and Battery Interaction

The instantaneous power required to move a FCEV can come from either the fuel cell stack, the battery, or a combination of both. Interactions between the fuel cell stacks and batteries of a FCEV can be complex and may vary based on application. Each manufacturer likely would employ a unique strategy to optimize the durability of these components and manage costs. The strategy selected would impact the size of the fuel cell stack and the size of the battery.

The fuel cell stack can be used to charge the battery that in turn powers the wheels ( i.e., series hybrid or range-extending), or it can work with the battery to provide power ( i.e., parallel hybrid or primary power) to the wheels. In the emerging HD FCEV market, when used to extend range, the fuel cell tends to have a lower peak power potential and may be sized to match the average power needed during a typical use cycle, including steady highway driving. At idle, the fuel cell may run at minimal power or turn off based on state of charge of the battery. The battery is used during prolonged high-power operations such as grade climbing and is typically in charge-sustaining mode, which means the average state of charge is maintained above a certain level while driving. When providing primary power, the fuel cell tends to have a larger peak power potential, sized to match all power needs of a typical duty cycle and to meet instantaneous power needs. The battery is mainly used to capture energy from regenerative braking and to help with acceleration and other transient power demands.

Based on how the fuel cell stacks and batteries are managed, manufacturers may use different types of batteries in HD FCEVs. Energy battery cells are typically used to store energy for applications with distance needs, so may be used more with range-extending fuel cells in vehicles with a relatively large battery. Power battery cells are typically used to provide additional high power for applications with high power needs in primary power fuel cell-dominant vehicles.

iii. Onboard Hydrogen Storage Tanks

Fuel cell vehicles carry hydrogen fuel onboard using large tanks. Hydrogen has extremely low density, so it must be compressed or liquified for use. There are various techniques for storing hydrogen onboard a vehicle, depending on how much fuel is needed to meet range requirements. Most transportation applications today use Type IV tanks, which typically include a plastic liner wrapped with a composite material such as carbon fiber that can withstand high pressures with minimal weight. High-strength carbon fiber is expensive, accounting for over 50 percent of the cost of onboard storage at production volumes of over 100,000 tanks per year.

Some existing fuel cell buses use compressed hydrogen gas at 350 bars (~5,000 pounds per square inch, or psi) of pressure, but other applications are using tanks with increased compressed hydrogen gas pressure at 700 bar (~10,000 psi) for extended driving range. A Heavy-Duty Vehicle Industry Group was formed in 2019 to standardize 700 bar high-flow fueling hardware components globally that meet fueling speed requirements ( i.e., so that fill times are similar to comparable HD ICE vehicles, as identified in DOE technical targets for Class 8 long-haul tractor-trailers). High-flow refueling rates for heavy-duty vehicles of 60–80 kg hydrogen in under 10 minutes were recently demonstrated in a DOE lab setting.

Based on our review of the literature, we believe that most HD vehicles likely have sufficient physical space to package hydrogen storage tanks onboard. Geometry and packing challenges may constrain the amount of gaseous hydrogen that can be stored onboard and, thus, the maximum range of trucks that travel longer distances without a stop for fuel. Liquid hydrogen is emerging as a cost-effective onboard storage option for long-haul operations; however, the technology readiness of liquid storage and refueling technologies is relatively low compared to compressed gas technologies. Nonetheless, companies like Daimler and Hyzon are pursuing onboard liquid hydrogen to minimize potential payload impacts and maintain the flexibility to drive up to 1,000 miles between refueling, comparable to today's diesel ICE vehicle refueling ranges. Therefore given our assessment of technology readiness, liquid storage tanks were not included as part of the technology packages that support the feasibility and appropriateness of our proposed standards. We request comment and data related to packaging space availability associated with FCEVs and projections for the development and application of liquid hydrogen in the HD transportation sector over the next decade.

iv. HD FCEV Safety Assessment

FCEVs have two potential risk factors that can be mitigated through proper design, process, and training: hydrogen and electricity. Electricity risks are identical to those of BEVs and, thus, are discussed in Section II.D.2 and DRIA Chapter 1.5.2. Hydrogen risks can occur throughout the process of fueling a vehicle. FCEVs must be designed such that hydrogen can safely be delivered to a vehicle and then transferred into a vehicle's onboard storage tanks and fuel cell stacks. Hydrogen has been handled, used, stored, and moved in industrial settings for more than 50 years, and there are many established methods for doing so safely. There is also federal oversight and regulation throughout the hydrogen supply chain system. Safety training and education are key for maintaining reasonable risk while handling and using hydrogen. For example, hydrogen-related fuel cell vehicle risks can be mitigated by following various SAE and OSHA standards, as discussed in DRIA Chapter 1.7.4. We request comment on our assessment that HD FCEVs can be designed to maintain safety.

4. Summary of Technology Assessment

In prior HD GHG rulemakings, EPA promulgated standards that could feasibly be met through technological improvements in many areas of the vehicle. For example, the HD GHG Phase 2 CO₂ emission standards were premised on technologies such as engine waste heat recovery, advanced aerodynamics (like those developed for DOE's SuperTruck programs), and, in some cases, hybrid powertrains. We evaluated each technology's effectiveness as demonstrated over the regulatory duty cycles using EPA's GEM and estimated the appropriate adoption rate of each technology. We then developed a technology package for each of the regulatory subcategories. We are following a similar approach in this Phase 3 NPRM.

In the HD GHG Phase 2 final rule, we included ZEV technologies in our assessment of the suite of technologies for HD vocational vehicles and tractors. However, in 2016, when the HD GHG Phase 2 rule was being developed, we stated that “adoption rates for these advanced technologies in heavy-duty vehicles are essentially non-existent today and seem unlikely to grow significantly within the next decade without additional incentives.” Thus, at that time, instead of including ZEV technologies in the technology packages for setting the Phase 2 standards, we provided advanced technology credit multipliers to help incentivize the development of ZEV technologies.

Since the 2016 promulgation of the HD GHG Phase 2 final rule, as discussed in Section I.C, a number of important factors have contributed to changes in the HD landscape. Therefore, as detailed in this Section II and DRIA Chapter 2, we now are proposing that BEV technologies and FCEV technologies will be technically feasible for HD vehicles and suitable for most applications, as assessed by vehicle type and each Phase 3 MY. As further detailed in this Section II and DRIA Chapter 2, we are also proposing that BEV and FCEV technologies are feasible at the adoption rates included in the technology packages, which vary depending on the respective vehicle type and Phase 3 MY, and thus that the proposed revised standards for MY 2027 and proposed new standards for MYs 2028 through 2032 are feasible and appropriate. Similar to Phase 1 and Phase 2, the technology packages used to support the standards in this proposal include a mix of technologies applied to HD vehicles, and development of those technology packages included an assessment of the projected feasibility of the development and application of BEV, FCEV, and other technologies that reduce GHG emissions from HD vehicles. While our analysis in this Section II.D focuses on certain technologies in the technology packages to demonstrate the feasibility of the proposed HD vehicle GHG emission standards, there are other technologies as described in DRIA Chapter 1 that can reduce CO2 emissions. Under the proposed rule, manufacturers may choose to produce the technologies that work best for their business case and the operator's needs in meeting the proposed standards, as the proposed standards are performance-based and do not mandate any specific technology for any manufacturer or any vehicle subcategory.

EPA developed a bottom-up approach to estimate the operational characteristics and costs of ZEV technologies for this proposal. This approach takes into consideration concerns received on the HD2027 NPRM concerning the proposed revised MY 2027 GHG vehicle standards' analysis presented in the HD2027 NPRM. We developed a new technology assessment tool, Heavy-Duty Technology Resource Use Case Scenario (HD TRUCS), to evaluate the design features needed to meet the power and energy demands of various HD vehicles when using ZEV technologies, as well as costs related to manufacturing, purchasing and operating ICE and ZEV technologies. HD TRUCS is described in more detail in Section II.D.5 and DRIA Chapter 2 but we briefly summarize the approach here.

To build technology packages using HD TRUCS, we created 101 representative HD vehicles that cover the full range of weight classes within the scope of this rulemaking (Class 2b through 8 vocational vehicles and tractors). The representative vehicles cover many aspects of work performed by the industry. This work was translated into energy and power demands per vehicle type based on everyday use of HD vehicles, ranging from moving goods and people to mixing cement. We then identified the technical properties required for a BEV or FCEV to meet the operational needs of a comparable ICE HD vehicle.

Since batteries can add weight and volume to a vehicle, we evaluated battery mass and physical volume required to package a battery pack. If the performance needs of a BEV resulted in a battery that was too large or heavy, then we did not consider the BEV for that application in our technology package because of, for example, the impact on payload and, thus, potential work accomplished relative to a comparable ICE vehicle.

To evaluate costs, including costs of compliance for manufacturers as well as user costs related to purchasing and operating ZEVs, we sized vehicle components that are unique to ZEVs to meet the work demands of each representative vehicle. We applied cost estimates to each vehicle component based on sizing to assess the difference in total powertrain costs between the ICE and ZEV powertrains. We accounted for the IRA battery tax credit and vehicle tax credit, as discussed in Section II.E.4. We also compared operating costs due to fuel consumption as well as vehicle maintenance and repair, and we included the cost to procure and install depot charging infrastructure for BEVs. For FCEVs, similar to ICE vehicles' infrastructure and fuel costs, we assumed hydrogen infrastructure costs were embedded in the cost of hydrogen fuel.

We relied on research and findings discussed in DRIA Chapters 1 and 2 to conduct this analysis. For MYs 2027 through 2029, we focused primarily on BEV technology. Consistent with our analysis, research shows that BEV technologies can become cost-competitive in terms of total cost of ownership for many HD vehicles by the late 2020s, but it would take longer for FCEVs. Given that there are more BEV models available today compared to FCEV models (see, e.g., DRIA Chapters 1.7.5 and 1.7.6), we inferred that BEV adoption is likely to happen sooner than the adoption of FCEV technology.

Starting in MY 2030, we also considered FCEV technology for select applications. BEV technology is more energy efficient than FCEV technology but may not be suitable for all applications, such as when the performance needs result in additional battery mass that affects payload. FCEVs are more energy efficient than diesel vehicles and can have shorter refueling times than BEVs with large batteries. We considered FCEVs in the technology packages for applications that travel longer distances and/or carry heavier loads ( i.e., for those that may be sensitive to refueling times or payload impacts). This included coach buses, heavy-haul tractors, sleeper cab tractors, and day cab tractors.

Though fuel cell technology is still emerging in HD vehicle applications, FCEVs are a viable ZEV technology for heavy-duty transportation and will be available in the 2030 timeframe (see DRIA Chapter 1.7.5). Inclusion of FCEVs in the technology packages starting in MY 2030 takes into consideration additional lead time to allow manufacturers to design, develop, and manufacture HD FCEV models. Fuel cell technology in other sectors has been in existence for decades and has been demonstrated to be technically feasible in heavy-duty transportation. Interim research and development (R&D) technical targets and projects (see DRIA Chapter 1.7.7) are in place to facilitate necessary improvements in the performance, durability, and costs of hydrogen-fueled long-haul HD tractors in 2030. With substantial federal investment in low-GHG hydrogen production (see DRIA Chapter 1.3.2), we project that the price of hydrogen fuel will drop enough by 2030 to make HD FCEVs cost-competitive with comparable ICE vehicles for some duty cycles. Hydrogen infrastructure is expected to need the additional time prior to MY 2030 to further develop, as discussed in greater detail in DRIA Chapter 1.8, but we expect the refueling needs can be met by MY 2030. We also recognize that regulations, like this proposed rule, can further incentivize technology and refueling infrastructure development and deployment. Therefore, we included FCEVs in our technology assessment beginning in MY 2030, which is our best projection after considering the IRA incentives related to hydrogen as a transportation fuel and FCEVs, DOE's hydrogen assessments, and other information discussed here in Section II and in DRIA Chapter 1.

After considering operational characteristics and costs in 2021 dollars, we determined the payback period, which is the number of years it would take to offset any incremental cost increase of a ZEV over a comparable ICE vehicle. Lastly, technology adoption rates for BEVs or FCEVs for the technology packages were selected based on the payback period. We request comment on this approach and any supporting data on the potential for these and additional technologies to be available in the HD market in the MY 2027 through MY 2032 timeframe.

5. EPA's HD TRUCS Analysis Tool

For this proposal, EPA developed an analysis tool, HD TRUCS, to evaluate the design features needed to meet the energy and power demands of various HD vehicle types when using ZEV technologies. The overarching design and functionality of HD TRUCS is premised on ensuring each of the 101 ZEV types could perform the same work as its ICE counterpart. We did this by sizing the BEV and FCEV components such that they could meet the driving demands based on the 90th percentile daily VMT for each application, while also accounting for the HVAC and battery thermal conditioning load requirements in hot and cold weather and any PTO demands for the vehicle. Furthermore, we accounted for the fact that the usable battery capacity is less than 100 percent and that batteries deteriorate over time. We also sized the ZEV powertrains to ensure that the vehicles would meet an acceptable level of acceleration from a stop and be able to maintain a cruise speed while going up a hill at six-percent grade. In this subsection, we discuss the primary inputs used in HD TRUCS. Additional details on HD TRUCS can be found in DRIA Chapter 2. We welcome comment on all aspects of HD TRUCS.

i. Vehicles Analyzed

We developed inputs for 101 different vehicle types for our assessment in HD TRUCS. This encompasses 22 different applications in the HD vehicle market, as shown in Table II–3. These vehicles applications are further differentiated by weight class, duty cycle, and daily vehicle miles traveled (VMT) for each of these vehicle applications into 101 vehicle types. These 101 vehicle types cover all 33 of the heavy-duty regulatory subcategories, as shown in DRIA Chapter 2.8.3.1. The initial list of HD TRUCS vehicles contained 87 vehicle types and was based on work the Truck and Engine Manufacturers Association (EMA) and California Air Resources Board (CARB) conducted for CARB's ACT rule. We consolidated the list; eliminated some of the more unique vehicles with small populations like mobile laboratories; and assigned operational characteristics that correspond to the Urban, Multi-Purpose, and Regional duty cycles used in GEM. We also added additional vehicle types to reflect vehicle applications that were represented in EPA's certification data. Chapter 2.1 of the DRIA summarizes the 101 unique vehicle types represented in HD TRUCS and how they are categorized, each with a vehicle identifier, vehicle application, vehicle weight class, MOtor Vehicle Emission Simulator (MOVES) SourceTypeID and RegClassID, and GEM duty cycle category. We request comment on our approach, including our categorization of vehicle types and applications in the data, and whether there are additional specific vehicle types we should include in our assessment.

Heavy-duty vehicles are typically powered by a diesel-fueled compression-ignition (CI) engine, though the heavy-duty market also includes vehicles powered by gasoline-fueled spark-ignition (SI) engines and alternative-fueled ICE. We selected diesel-powered ICE vehicles as the baseline vehicle for the assessment in HD TRUCS in our analysis because a diesel-fueled CI engine is broadly available for all of the 101 vehicle types and are more efficient than SI engines. Chapter 2.2 of the DRIA includes the details we developed for each of the baseline vehicles, including the size of the engine and the transmission type. This information was used to determine the weight and the cost of the ICE powertrains.

In the analysis, for MYs 2027 through 2029, we focused primarily on BEV technology. Starting in MY 2030, we also considered FCEV technology for select applications that travel longer distances and/or carry heavier loads. This included coach buses, heavy-haul tractors, sleeper cab tractors, and day cab tractors that are designed to travel longer distances. We request comment on our approach that focuses primarily on BEVs, which currently are more prevalent in the HD vehicle market, and whether there are additional vehicle types that should be evaluated as FCEVs along with BEVs.

ii. Vehicle Energy Demand

Vehicles require energy to perform the work required of the vehicle. This work includes driving, idling, and providing heating and cooling; in addition, some vehicles require energy to operate equipment. Vehicles with regenerative braking systems have the opportunity to recover some of the kinetic energy that would otherwise be lost during braking. There are a wide variety of energy demands across the heavy-duty sector, depending on the vehicle's application. For example, some vehicles, such as long-haul tractors, spend the vast majority of the time driving, a fraction of the time idling, and require heating and cooling of the cabin, but do not require operation of additional equipment. A transit bus typically operates at low speeds, so it requires less energy for driving than a long-haul tractor, but requires more energy for heating or cooling due to its large amount of interior cabin volume. Unlike ICE vehicles where the cabin heating is often provided by excess heat from the main ICE, BEVs do not have excess heat from an ICE to utilize in this manner and thus require more energy than ICE vehicles to heat the cabin and additional energy to manage the temperature of the batteries. As another example of the wide variety of energy demands for HD vehicles, a utility truck, also known as a bucket truck, may only drive a few miles to a worksite while idling for the majority of the day and using energy to move the bucket up and down. The power to run the separate equipment on ICE vehicles is typically provided by a PTO from the main engine. In HD TRUCS, we determined the daily energy demand for each of the 101 vehicle types by estimating both the baseline energy demands that are similar regardless of the powertrain configuration and the energy demands that vary by powertrain. The baseline energy includes energy at the axle to move the vehicle, energy recovered from regenerative braking energy, and PTO energy. Powertrain-specific energy includes energy required to condition the battery and heat or cool the cabin using a heating, ventilation, and air conditioning (HVAC) system. We discuss each of these in the following subsections.

a. Baseline Energy

The amount of energy needed at the axle to move the vehicle down the road is determined by a combination of the type of drive cycle (such as urban or freeway driving) and the number of miles traveled over a period of time. For each HD TRUCS vehicle type, we determined the baseline energy consumption requirement that would be needed for each of the ZEV applications. To do this, we used the drive cycles and cycle weightings adopted for HD GHG Phase 2 for our assessment of the energy required per mile for each vehicle type. EPA's GEM model simulates road load power requirements for various duty cycles to estimate the energy required per mile for HD vehicles. To understand the existing heavy-duty industry, we performed an analysis on current heavy-duty vehicles in the market in order to determine typical power requirements and rates of energy consumption at the axle. These values represent the energy required to propel a vehicle of a given weight, frontal area, and tire rolling resistance to complete the specified duty cycle on a per-mile basis, independent of the powertrain. In DRIA Chapter 2.2.2, we describe the GEM inputs and results used to estimate the propulsion energy and power requirements at the axle for ICE vehicles on a per-mile basis. We also used these inputs, along with some simple electric vehicle assumptions, to develop a model for electric vehicles to calculate weighted percent of energy recovery due to regenerative braking. Additional detail can be found in DRIA Chapter 2.2.2.1.3. We request comment on our approach, including other data we should consider in our assessment of energy consumption.

Some vocational vehicles have attachments that perform work, typically by powering a hydraulic pump, which are powered by PTOs. Information on in-use PTO energy demand cycles is limited. NREL published two papers describing investigative work into PTO usage and fuel consumption. These studies, however, were limited to electric utility vehicles, such as bucket trucks and material handlers. To account for PTO usage in HD TRUCS, we chose to rely on a table described in California's Diesel Tax Fuel Regulations, specifically in Regulation 1432, “Other Nontaxable Uses of Diesel Fuel in a Motor Vehicle,” that covers a wider range of vehicles beyond the electric utility vehicles in the referenced NREL studies. This table contains “safe-harbor” percentages that are presumed amounts of diesel fuel used for “auxiliary equipment” operated from the same fuel tank as the motor vehicle. We used this source to estimate PTO energy use as a function of total fuel consumed by vehicle type, as discussed in DRIA Chapter 2.2.2.1.4. We request additional data that could be considered in our assessment of PTO loads in our final rulemaking assessment.

Within HD TRUCS, we calculated the total energy needed daily based on a daily VMT for each vehicle type. We used multiple sources to develop the VMT for each vehicle. Daily VMT for each vehicle came from one of five Sources: the NREL FleetDNA database, a University of California-Riverside (UCR) database, the 2002 Vehicle Inventory and Use Survey (VIUS), the CARB Large Entity Report, or an independent source specific to an application, as discussed in DRIA Chapter 2.2.1.2. Each vehicle type was assigned a 50th percentile or average daily VMT that was used to estimate operational costs, such as average annual fuel, hydrogen, or electricity costs, and maintenance and repair costs (see DRIA Chapters 2.3.4, 2.4.4, and 2.5.3). We also account for the change in use of the vehicle over the course of its ownership and operation in HD TRUCS by applying a MOVES-based VMT ratio based on vehicle age to the 50th percentile VMT to arrive at a 10 year average VMT, as described in more detail in DRIA Chapter 2.2.1.2.2. We also developed a 90th percentile daily VMT and used it in HD TRUCS to size ZEV components, such as batteries, and estimate the size requirements for EVSE. We selected the 90th percentile daily VMT data because we project that manufacturers will design their BEVs to meet most daily VMT needs, but not the most extreme operations. BEVs designed for all daily VMT needs would be unnecessarily heavy and expensive for most operations, which would limit their appeal in the broad market. Please see DRIA Chapter 2.2.1.2 for the complete list of VMT for each of the 101 vehicle types. We request comment, including comment with data, on our VMT assessments.

b. Powertrain-Specific Energy

Heating, ventilation, and air conditioning (HVAC) requirements vary by vehicle type, location, and duty cycle. The HVAC energy required to heat and cool interior cabins is considered separately from the baseline energy in HD TRUCS, since these energy loads are not required year-round or in all regions of the country. Nearly all commercial vehicles are equipped with heat and basic ventilation and most vehicles are equipped with air conditioning (A/C). In ICE vehicles, traditional cabin heating uses excess thermal energy produced by the main ICE. This is the only source of cabin heating for many vehicle types. Additionally, on ICE vehicles, cabin A/C uses a mechanical refrigerant compressor that is engine belt-driven.

For BEVs, the energy required for thermal management is different than for ICE vehicles. First, the loads for HVAC are different because the vehicle is not able to be heated from excess heat from the engine. In this analysis, we project HD BEVs would be equipped with either a positive temperature coefficient (PTC) electric resistance heater with traditional A/C, or a full heat pump system, as described in DRIA Chapter 1. The vehicle's battery is used to power either system, but heat pumps are many times more efficient than PTC heaters. Given the success and increasing adoption of heat pumps in light-duty EVs, we believe that heat pumps will be the more commonly used technology and thus assume the use of heat pumps in HD TRUCS.

To estimate HVAC energy consumption of BEVs in HD TRUCS, we performed a literature and market review. Even though there are limited real-world studies, we agreed with the HVAC modeling-based approach described in Basma et. al. This physics-based cabin thermal model considers four vehicle characteristics: the cabin interior, walls, materials, and number of passengers. The authors modeled a Class 8 electric transit bus with an HVAC system consisting of two 20-kW reversible heat pumps, an air circulation system, and a battery thermal management system. We used their estimated HVAC power demand values as a function of temperature, resembling a parabolic curve, where hotter and colder temperatures require more power with the lowest power demand between 59 to 77 °F.

The power required for HVAC in HD TRUCS is based on a Basma et. al study that determined the HVAC power demand across a range of ambient temperatures. We created three separate ambient temperature bins: one for heating (less than 55 °F), one for cooling (greater than 80 °F), and one for a temperature range that requires only ventilation (55–80 °F). In HD TRUCS, we already accounted for the energy loads due to ventilation in the axle loads, so no additional energy consumption is applied here for the ventilation-only operation. We then weighted the power demands by the percent HD VMT traveled at a specific temperature range. The results of the VMT-weighted HVAC power demand for a Class 8 Transit Bus are shown in Table II–4. We request comment on and data to support other approaches to quantify the HVAC energy demand in BEVs, including the ambient temperature ranges where heating and cooling are utilized.

Lastly, HVAC load is dependent on cabin size—the larger the size of the cabin, the greater the HVAC demand. The values for HVAC power demand shown in Table II–4 represent the power demand to heat or cool the interior of a Class 8 Transit bus. However, HD vehicles have a range of cabin sizes; therefore, we developed scaling ratios relative to the cabin size of a Class 8 bus. Each vehicle's scaling factor is based on the surface area of the vehicle compared to the surface area of the Class 8 bus. For example, a Class 4–5 shuttle bus has a cabin size ratio of 0.6, in this case, the heating demand for the vehicle will be 3.04 kW and the cooling demand would be 1.99 kW. The adjustment ratio for buses and ambulances are between 0.3–0.6, while the cabin size for remaining HDVs have a similar cabin to a mid-size light duty vehicle and therefore, a single average scaling factor of 0.2 was applied to all remaining vehicle types. We welcome data to support these or other cabin size scaling factors.

Fuel cell stacks produce excess heat during the conversion of hydrogen to electricity, similar to an ICE during combustion. This excess heat can be used to heat the interior cabin of the vehicle. In HD TRUCS, we already accounted for the energy loads due to ventilation in the axle loads, so no additional energy consumption is applied to FCEV for heating operation. Therefore, for FCEV energy consumption in HD TRUCS, we only include additional energy requirements for air conditioning ( i.e. not for heating). As described in DRIA Chapter 2.4.1.1.1, we assigned a power demand of 3.32 kW for powering the air conditioner on a Class 8 bus. The A/C loads are then scaled by the cabin volume for other vehicle applications in HD TRUCS and applied to the VMT fraction that requires cooling, just as we did for BEVs.

BEVs have thermal management systems to maintain battery core temperatures within an optimal range of approximately 68 to 95 degrees Fahrenheit (F). In HD TRUCS, we accounted for the battery thermal management energy demands as a function of ambient temperature based on a Basma et. al study. As described in DRIA Chapter 2.4.1.1.3, we determined the amount of energy consumed to heat the battery with cabin air when it is cold outside (less than 55 °F) and energy consumed to cool the battery when it is hot outside (greater than 80 °F) with refrigerant cooling. For the ambient temperatures between these two regimes, we agreed with Basma, et. al that only ambient air cooling is required for the batteries, which requires no additional load. We first determined a single VMT-weighted power consumption value for battery heating and a value for battery cooling based on the MOVES HD VMT distribution, based on the same method used for HVAC. Then, we determined the energy required for battery conditioning required for eight hours of daily operation and expressed it in terms of percent of total battery size. Table II–5 shows the energy consumption for battery conditioning for both hot and cold ambient temperatures, expressed as a percentage of battery capacity, used in HD TRUCS. We request additional data on the battery thermal management loads for HD BEVs.

iii. BEV Component Sizing and Weight

We used HD TRUCS to determine the size of two of the major components in a BEV—the battery and the motor. The size of these components is determined by the energy needs of the specific vehicle to meet its daily operating requirements. In this subsection, we also discuss our method to evaluate the payload and packaging impact of the battery.

a. Battery

First, in HD TRUCS, we based the size of the battery on the daily demands on the vehicle to perform a day's work, based on the 90th percentile VMT (sizing VMT). As described in the Vehicle Energy Demand subsection, this daily energy consumption is a function of miles the vehicle is driven and the energy it consumes because of: (1) moving the vehicle per unit mile, including the impact of regenerative braking, and PTO energy requirements and (2) battery conditioning and HVAC energy requirements. Then we also accounted for the battery efficiency, depth of discharge, and deterioration in sizing of the batteries for BEVs in HD TRUCS.

The daily energy consumption of each BEV in HD TRUCS is determined by applying efficiency losses to energy consumption at the axle, as described in DRIA Chapter 2.4.1.1.3. We have accounted for these losses in the battery, inverter, and e-motor before the remaining energy arrives at the axle, as shown in Table II–6. We request comment, including data, on our approach and the results for our assessment of system efficiencies for HD BEV components.

Next, we oversized the battery to account separately for the typical usable amount of battery and for battery deterioration over time. We sized the battery limiting the battery to a maximum depth of discharge of 80 percent, recognizing that manufacturers and users likely would not allow the battery capacity to be depleted beyond 80 percent of original capacity. We also accounted for deterioration of the battery capacity over time by oversizing the battery by 20 percent, assuming only 80 percent of the battery storage is available throughout its life. Therefore, the battery sizes we used in our assessment are conservative because they could meet 100 percent of the daily operating requirement using the 90th percentile VMT at the battery end of life. This is described in greater detail in DRIA Chapter 2.4.1.1 and 2.7.5.4. We request comment on approach and results for the useable battery range and battery deterioration for HD BEVs that we could consider for our final rule analysis.

b. Motor

We determined the size of the motor for each BEV based on the peak power of the transient cycle and highway cruise cycles, the vehicle's ability to meet minimum performance targets in terms of acceleration rate of the vehicle, and the ability of the vehicle to maintain speed going up a hill. As described in DRIA Chapter 2.4.1.2, we estimated a BEV motor's peak power needs to size the e-motor, after considering the peak power required during the ARB transient cycle and performance targets included in ANL's Autonomie model and in Islam et al., as indicated in Table II–7. We assigned the target maximum time to accelerate a vehicle from stop to 30 mph and 60 mph based on weight class of each vehicle. We also used the criteria that the vehicle must be able to maintain a specified cruise speed while traveling up a road with a 6 percent grade, as shown in Table II–7. In the case of cruising at 6 percent grade, the road load calculation is set at a constant speed for each weight class bin on a hill with a 6 percent incline. We determined the required power rating of the motor as the greatest power required to drive the vehicle over the ARB transient test cycle, at 55 mph and 65 mph constant cruise speeds, or at constant speed at 6 percent grade, and then applied losses from the e-motor. We request comment on our approach using these performance targets.

c. Battery Weight and Volume

Performance needs of a BEV can result in a battery that is so large or heavy that it impacts payload and, thus, potential work accomplished relative to a comparable ICE vehicle. We determined the battery weight and physical volume for each vehicle application in HD TRUCS using the specific energy and energy density of the battery for each battery capacity. As described in DRIA Chapter 2.4.2, to determine the weight impact, we used battery specific energy, which measures battery energy per unit of mass. While battery technologies have made tremendous advancements in recent years, it is well known that current automotive batteries add mass to the vehicle. Our values for the specific energy of battery packs with lithium-ion cell chemistries are based on Autonomie. The values we used in HD TRUCS are shown in Table II–8.

To evaluate battery volume and determine the packaging space required for each HD vehicle type, we used battery energy density. We also estimated the battery's width using the wheelbase and frame depths.

Battery energy density (also referred to as volumetric energy density) measures battery energy per unit of volume. This value was not available as a part of the Autonomie; however, the overall trend of energy density shows a linear correlation with specific energy. In this analysis, we determined the energy density is 2.5 times that of specific energy, as shown in Table II–9.

We request comment on our approach and results as well as comment and data on current and projected levels of battery-specific energy and battery-specific density values for HD vehicles.

Heavy-duty vehicles are used to perform work, such as moving cargo or carrying passengers. Consequently, heavy-duty vehicles are sensitive to increases in vehicle weight and carrying volume. To take this into account, we also evaluated BEVs in terms of the overall impact on payload-carrying ability and battery packaging space. The results of this analysis can be found in DRIA Chapters 2.4.2 and 2.8.1. We found that the extra weight of the batteries for applications such as coach buses and tractors that travel long distances could have an impact on operations of these vehicles as BEVs. Therefore, for applications where our analysis showed that BEVs impacted the payload capacity by over 30 percent, we assessed fuel cell technology. In this proposal we are using a single technology package that supports the feasibility of the proposed standards, but we recognize the potential of BEVs in the applications where we evaluate FCEVs, as demonstrated by the development of a long-haul battery electric tractor by Tesla.

iv. Charging Infrastructure for BEVs

Charging infrastructure represents a key element required for HD BEV operation. More charging infrastructure will be needed to support the growing fleet of HD BEVs. This will likely consist of a combination of (1) depot charging—with infrastructure installed in parking depots, warehouses, and other private locations where vehicles are parked off-shift (when not in use), and (2) en-route charging, which provides additional electricity for vehicles during their operating hours.

In draft RIA Chapters 2.6 and 2.7.7 we describe how we accounted for charging infrastructure in our analysis of HD BEV technology feasibility and adoption rates for MYs 2027–2032. For this analysis, we estimate infrastructure costs associated with depot charging to fulfill each BEV's daily charging needs off-shift with the appropriately sized electrical vehicle supply equipment. This approach reflects our expectation that many heavy-duty BEV owners will opt to purchase and install EVSE at depots; accordingly, we explicitly account for all of these upfront costs in our analysis. By contrast, we do not estimate upfront hardware and installation costs for public and other en-route electric charging infrastructure because the BEV charging needs are met with depot charging in our analysis. Discussion of private sector infrastructure investments and charging deployment projects is included in DRIA Chapter 1.6.2. We request comment on this analytical approach.

Vehicle owners with return-to-base operations who choose to install depot charging equipment have many options from which to select. This includes AC or DC charging, power level, as well as the number of ports and connectors per charging unit, connector type(s), communications protocols, and additional features such as vehicle-to-grid capability (which allows the vehicle to supply energy back to the grid). Many of these selections will impact EVSE hardware and installation costs, with power level as one of the most significant drivers of cost. While specific cost estimates vary across the literature, higher-power charging equipment is typically more expensive than lower-power units. For this reason, we have chosen to evaluate infrastructure costs separately for four different, common charging types in our depot charging analysis: AC Level 2 (19.2 kW) and 50 kW, 150 kW, and 350 kW DC fast charging (DCFC).

How long a vehicle is off-shift and parked at a depot, warehouse, or other home base each day is a key factor for determining which charging type(s) could meet its needs. The amount of time available at the depot for charging (dwell time) will depend on a vehicle's duty cycle. For example, a school bus or refuse truck may be parked at a depot in the afternoon and remain there until the following morning whereas a transit bus may continue to operate throughout the evening. Even for a specific vehicle, off-shift dwell times may vary between weekends and weekdays, by season, or due to other factors that impact its operation. The 101 vehicle types in our analysis span a wide range of vehicle applications and duty cycles, and we expect their off-shift dwell times at depots to vary accordingly. As described in DRIA Chapter 2.6.4.1, in order to better understand what an average depot dwell time might look like, we examined a dataset with engine start and off times for 564 commercial vehicles. We used the longest time the vehicle engine was off each day as a rough proxy for depot dwell time, finding the average across all 564 vehicles to be over 14 hours, with proxy dwell times for most of the seven vehicle categories examined rounding to 12 hours or longer. However, assigning specific dwell times for each of the 101 vehicle types in our analysis is challenging due a lack of comprehensive datasets on parking times and locations, and, as further detailed in DRIA Chapter 2.6.4.1, we acknowledge limitations in the approach and dataset we examined. Given these uncertainties, we used an off-shift dwell time for all vehicle types of 12 hours for the purpose of selecting charging equipment at depots in our analysis.

v. FCEV Component Sizing

To compare diesel-fueled HD ICE vehicles and HD FCEV technology costs and performance in HD TRUCS, this section explains how we define HD FCEVs based on the performance and use criteria in DRIA Chapter 2.2 (that we also used for HD BEVs, as explained in Section D.5.ii). We determined the e-motor, fuel cell stack, and battery pack sizes to meet the power requirements for each of the eight FCEVs represented in HD TRUCS. We also estimated the size of the onboard fuel tank needed to store the energy, in the form of hydrogen, required to meet typical range and duty cycle needs. See DRIA Chapter 2.5 for further details. We request comment, including data, on our approach and results from our assessment of HD FCEV component sizing.

a. E-Motor

As discussed in DRIA Chapter 2.4.1.2, the electric motor (e-motor) is part of the electric drive system that converts the electric power from the battery or fuel cell into mechanical power to move the wheels of the vehicle. In HD TRUCS, the e-motor was sized for a FCEV like it was sized for a BEV—to meet peak power needs of a vehicle, which is the maximum power to drive the ARB transient cycle, meet the maximum time to accelerate from 0 to 30 mph, meet the maximum time to accelerate from 0 to 60 mph, and maintain a set speed up a six-percent grade. Additional power was added to account for e-motor efficiency losses using the same e-motor efficiency losses calculated and applied for BEVs, as discussed in DRIA Chapter 2.4.1.1.3.

b. Fuel Cell Stack

Vehicle power in a FCEV comes from a combination of the fuel cell (FC) stack and the battery pack. The FC stack behaves like the internal combustion engine of a hybrid vehicle, converting chemical energy stored in the hydrogen fuel into useful work. The battery is charged by power derived from regenerative braking, as well as excess power from the FC stack. Some FCEVs are designed to primarily rely on the fuel cell stack to produce the necessary power, with the battery exclusively used to capture energy from regenerative braking. Other FCEVs are designed to store more energy in a battery to meet demand during situations of high-power need.

While much of FCEV design is dependent on the use case of the vehicle, manufacturers also balance the cost of components such as the FC stack, the battery, and the hydrogen fuel storage tanks. For the purposes of this HD TRUCS analysis, we focused on proton-exchange membrane (PEM) fuel cells that use energy battery cells, where the fuel cell and the battery were sized based on the demands of the vehicle. In HD TRUCS, the fuel cell stack was sized either to reach the 90th percentile of power required for driving the ARB transient cycle or to maintain a constant highway speed of 75 mph. The 90th percentile power requirement was used to size the fuel cells of vocational vehicles. For sleeper and day cabs, the fuel cell was sized using the power required to drive at 75 mph with 80,000-pound gross combined vehicle weight (GCVW).

To avoid undersizing the fuel cell stack, we applied efficiency values to account for losses that take place before the remaining energy arrives at the axle. The same battery and inverter efficiencies from Table II–10 were used for the FCEV calculations. Fuel cell stack efficiency losses are due to the conversion of onboard hydrogen to electricity. The DOE technical targets for Class 8 long-haul tractor-trailers are to reach 68 percent peak efficiency by around 2030 (this is the interim target; the ultimate target is to reach 72 percent efficiency). Table II–10 shows the fuel cell efficiency values that we used for MYs 2027–2032 in HD TRUCS, which are slightly more conservative yet include expected improvements over time. We averaged the high-tech peak efficiency estimates with low-tech peak efficiency estimates from ANL's 2022 Autonomie for 2025, 2030, and 2035 for available vehicle types. We then linearly interpolated these averaged values to calculate values for each year.

c. Battery Pack

As described in DRIA Chapter 2.5.1.1.3, in HD TRUCS, the battery power accounts for the difference between the power demand of the e-motor at any moment and the maximum power output of the fuel cell stack. We sized the battery to meet these power needs in excess of the fuel cell stack's capability only when the fuel cell cannot provide sufficient power. In our analysis, the remaining power needs are sustained for a duration of 10 minutes ( e.g., to assist with a climb up a steep hill).

d. Onboard Hydrogen Storage Tank

A FCEV is re-fueled like a gasoline or diesel-fueled vehicle. We determined the capacity of the onboard hydrogen energy storage system using an approach like the BEV methodology for battery pack sizing in DRIA Chapter 2.4.1.1, but we based the amount of hydrogen needed on the daily energy consumption needs of a FCEV.

As described in DRIA Chapter 2.5.1.2, we converted FCEV energy consumption (kWh) into hydrogen weight using an energy content of 33.33 kWh per kg of hydrogen. In our analysis, 95 percent of the hydrogen in the tank (“usable H2”) can be accessed. This is based on targets for light-duty vehicles, where a 700-bar hydrogen fuel tank with a capacity of 5.9 kg has 5.6 kg of usable hydrogen. Furthermore, we added an additional 10 percent to the tank size in HD TRUCS to avoid complete depletion of hydrogen from the tank.

E. Technology, Charging Infrastructure, and Operating Costs

In the following subsections, we first discuss BEV technology (Section II.E.1) and associated EVSE technology costs (Section II.E.2) and FCEV technology costs (Section II.E.3). DRIA Chapter 2.4.3. (for BEVs) and DRIA Chapter 2.5.2 (for FCEVs) includes the cost estimates for each of the 101 applications. We then discuss the Inflation Reduction Act tax credits we quantified in our analysis in Section II.E.4. Our assessment of operating costs including the fuel or electricity costs, along with the maintenance and repair costs, are presented in Section II.E.5. This subsection concludes with the overall payback analysis in Section II.E.6. DRIA Chapter 2.8.2 includes the vehicle technology costs, EVSE costs, operating costs, and payback results for each of the 101 HD applications. The technology costs aggregated into MOVES categories are also described in detail in DRIA Chapter 3.1.

1. BEV Technology Costs

The incremental cost of a BEV powertrain system is calculated as the cost difference from the comparable vehicle powertrain with an ICE. The ICE vehicle powertrain cost is a sum of the costs of the engine (including the projected cost of the HD2027 standards), alternator, gearbox (transmission), starter, torque converter, and final drive system.

Heavy-duty BEV powertrain costs consist of the battery, electric motor, inverter, converter, onboard charger, power electronics controller, transmission or gearbox, final drive, and any electrical accessories. DRIA Chapter 2.4.3 contains additional detail on our cost projections for each of these components. We request comment, including additional data, on our analysis for consideration in the final rule regarding current and projected BEV component costs.

Battery costs are widely discussed in the literature because they are a key driver of the cost of a HD electric vehicle. The per unit cost of the battery, in terms of $/kWh, is the most common metric in determining the cost of the battery as the final size of the battery may vary significantly between different applications. The total battery pack cost is a function of the per unit kWh cost and the size (in terms of kWh) of the pack.

There are numerous projections for battery costs and battery pricing in the literature that cover a range of estimates. Sources do not always clearly define what is included in their cost or price projections, nor whether the projections reflect direct manufacturing costs incurred by the manufacturer or the prices seen by the end-consumer. Except as noted, the values in the literature we used were developed prior to enactment of the Inflation Reduction Act. For example, BloombergNEF presents battery prices that would reach $100 per kWh in 2026. In 2021, ANL developed cost projections for heavy-duty vehicle battery packs in their benefit analysis (BEAN) model, that ranged from $225 per kWh to $175 per kWh in 2027 and drop to $150 per kWh to $115 per kWh in 2035. In a recent update to BEAN, released after the IRA was passed, ANL now projects heavy-duty battery pack costs in the range of $95 per kWh to $128 per kWh in 2025 and a drop to between $70 per kWh and $90 per kWh in 2035. The direct manufacturing battery cost for MY 2027 used in HD TRUCS is based on a literature review of costs of zero-emission truck components conducted by the International Council on Clean Transportation (ICCT). As described in detail in DRIA Chapter 2.4.3.1, we considered this source to be a comprehensive review of the literature at the time of the HD TRUCS analysis for the cost of battery packs in the absence of the IRA, which may mean that it presents higher costs than will be realized with the incentives in the IRA, even when accounting for the battery tax credit described in Section II.E.4. In 2025, the average cost is estimated to be $163.50/kWh (2019$) and, in 2030, the average cost is projected to fall to $100 (2019$). We applied a linear interpolation of these values that yields an estimated cost of $138/kWh (2019$) for MY 2027. We then projected the costs to MY 2032 by using an EPA estimate of market learning related to battery production and the respective reduction in battery costs over this period of time, as shown in Table II–11. We request comment, including data, on our approach and projections for battery pack costs for the heavy-duty sector, including values that specifically incorporate the potential impacts of the IRA.

Batteries are the most significant cost component for BEVs and the IRA section 13502, “Advanced Manufacturing Production Credit,” has the potential to significantly reduce the cost of BEVs whose batteries are produced in the United States. As discussed in Section II.E.4, we thus then also accounted for the IRA Advanced Manufacturing Production Credit, which provides up to $45 per kWh tax credits (with specified phase-out in calendar years (CYs) 2030–2033) for the production and sale of battery cells and modules, and additional tax credits for producing critical minerals such as those found in batteries, when such components or minerals are produced in the United States and other criteria are met.

An electric drive (e-drive)—another major component of an electric vehicle—includes the electric motor, an inverter, a converter, and optionally, a transmission system or gearbox. The electric energy in the form of direct current (DC) is provided from the battery; an inverter is used to change the DC into alternating current (AC) for use by the motor. The motor then converts the electric power into mechanical or motive power to move the vehicle. Conversely, the motor also receives AC from the regenerative braking, whereby the converter changes it to DC to be stored in the battery. The transmission reduces the speed of the motor through a set of gears to an appropriate speed at the axle. An emerging trend is to replace the transmission and driveline with an e-axle, which is an electric motor integrated into the axle, e-axles are not explicitly covered in our cost analysis. We request data on e-axle costs that we could consider for the final rule.

Similar to the battery cost, there is a range of electric drive cost projections available in the literature. One reason for the disparity is differences across the literature is what is included in each for the “electric drive”; some cost estimates include only the electric motor and others present a more integrated model of e-motor/inverter/gearbox combination. As described in detail in DRIA Chapter 2.4.3.2.1, EPA's MY 2027 e-drive cost, shown in Table II–12, comes from ANL's 2022 BEAN model and is a linear interpolation of the average of the high- and low-tech scenarios for 2025 and 2030, adjusted to 2021$. We then calculated MY 2028–2032 values, also shown in Table II–12, using an EPA estimate of market learning shown in DRIA Chapter 3.2.1. We welcome comment, including data, on our assessment of e-drive costs.

Gearbox and final drive units are used to reduce the speed of the motor and transmit torque to the axle of the vehicle. In HD TRUCS, the final drive unit direct manufacturing cost is $1,500 per unit, based on the “Power Converter” average cost in ANL's BEAN model. The cost of the gearbox varies depending on the vehicle weight class and duty cycle. In our assessment, all light heavy-duty BEVs would be direct drive and have no transmission and therefore no cost, consistent with ANL's BEAN model. We then mapped BEAN gearbox costs for BEVs to the appropriate medium heavy-duty and heavy heavy-duty vehicles in HD TRUCS. Gearbox and final drive costs for BEVs are in DRIA Chapter 2.4.3.2.

Power electronics are another electrification component (along with batteries and motors) where a DC–DC converter transitions high battery voltage to a common 12V level for auxiliary uses. EPA's power electronics and electric accessories costs of $6,000 per unit came from ANL's BEAN model. See DRIA Chapter 2.4.3.2.2 for further details.

When using a Level 2 charging plug, an on-board charger converts AC power from the grid to usable DC power via an AC–DC converter. When using a DC fast charger (DCFC), any AC–DC converter is bypassed, and the high-voltage battery is charged directly. As further discussed in DRIA Chapter 2.4.3.3, EPA's on-board charger costs, as shown in Table II–13, come from ANL's BEAN model and we averaged the low-tech and high-tech values for 2025 and 2030, and then MY 2027 was linearly interpolated and adjusted to 2021$. We then calculated the MY 2028–2032 costs using the learning curve shown in DRIA Chapter 3.2.1.

The total upfront BEV direct manufacturing cost is the summation of the per-unit cost of the battery, motor, power electronics, on-board charger, gearbox, final drive, and accessories. The total direct manufacturing technology costs for BEVs for each of the 101 vehicle types in HD TRUCS can be found in DRIA Chapter 2.4.3.5 for MY 2027 and MY 2032.

2. Charging Infrastructure Costs

In our analysis of depot charging infrastructure costs, we account for the cost to purchasers to procure both EVSE (which we refer to as the hardware costs) as well as costs to install the equipment. These installation costs typically include labor and supplies, permitting, taxes, and any upgrades or modifications to the on-site electrical service. We developed our EVSE cost estimates from the available literature, as discussed in DRIA Chapter 2.6.

Both hardware and installation costs could vary over time. For example, hardware costs could decrease due to manufacturing learning and economies of scale. Recent studies by ICCT assumed a 3 percent reduction in hardware costs for EVSE per year to 2030. By contrast, installation costs could increase due to growth in labor or material costs. Installation costs are also highly dependent on the specifics of the site including whether sufficient electric capacity exists to add charging infrastructure and how much trenching or other construction is required. If fleet owners choose to install charging stations at easier, and therefore, lower cost sites first, then installation costs could rise over time as stations are developed at more challenging sites. One of the ICCT studies found that these and other countervailing factors could result in the average cost of a 150 kW EVSE port in 2030 being similar (~3 percent lower) to that in 2021. After considering the uncertainty on how costs may change over time, we keep the combined hardware and installation costs per EVSE port constant. We request comment on this approach.

Our infrastructure analysis centered around four charging types for heavy-duty depot charging. As shown in Table II–14, the EVSE costs we used in our analysis range from about $10,000 for a Level 2 port to over $160,000 for a 350 kW DCFC port. As described in Chapter 2.6, in our analysis, we allow up to two vehicles to share one DCFC port if there is sufficient depot dwell time for both vehicles to meet their daily charging needs. In those cases, the EVSE costs per vehicle are halved. We request comment, including data, on our approach and assessment of current and future costs for charging equipment and installation.

EPA acknowledges that there may be additional infrastructure needs and costs beyond those associated with charging equipment itself. While planning for additional electricity demand is a standard practice for utilities and not specific to BEV charging, the buildout of public and private charging stations (particularly those with multiple high-powered DC fast charging units) could in some cases require upgrades to local distribution systems. For example, a recent study found power needs as low as 200 kW could trigger a requirement to install a distribution transformer. The use of onsite battery storage and renewables may be able to reduce the need for some distribution upgrades; station operators may also opt to install these to mitigate demand charges associated with peak power. However, there is considerable uncertainty associated with future distribution upgrade needs, and in many cases, some costs may be borne by utilities rather than directly incurred by BEV or fleet owners. Therefore, we do not model them directly as part of our infrastructure cost analysis. We welcome comments on this and other aspects of our cost analysis.

As discussed in Section V, we model changes to power generation due to the increased electricity demand anticipated in the proposal as part of our upstream analysis. We project the additional generation needed to meet the demand of the heavy-duty BEVs in the proposal to be relatively modest (as shown in DRIA Chapter 6.5). As the proposal is estimated to increase electric power end use by heavy-duty electric vehicles by 0.1 percent in 2027 and increasing to 2.8 percent in 2055. The U.S. electricity end use between the years 1992 and 2021, a similar number of years included in our proposal analysis, increased by around 25 percent without any adverse effects on electric grid reliability or electricity generation capacity shortages. Grid reliability is not expected to be adversely affected by the modest increase in electricity demand associated with HD BEV charging.

A GAO report noted that the private sector and the government share responsibility for the reliability of the U.S. electric power grid. The report stated, “Most of the electricity grid—the commercial electric power transmission and distribution system comprising power lines and other infrastructure—is owned and operated by private industry. However, Federal, state, local, Tribal, and territorial governments also have significant roles in enhancing the resilience of the electricity grid.” For instance, at the Federal level, the Department of Homeland Security (DHS) coordinates Federal efforts to promote the security and reliability of the nation's energy sector; the Department of Energy (DOE) leads Federal efforts including research and technology development; and the Federal Energy Regulatory Commission (FERC) regulates the interstate electricity transmission and is responsible for reviewing and approving mandatory electric Reliability Standards, which are developed by the North American Electric Reliability Corporation (NERC). NERC is the federally designated U.S. electric reliability organization which “develops and enforces Reliability Standards; annually assesses seasonal and long‐term reliability; monitors the bulk power system through system awareness; and educates, trains, and certifies industry personnel.” These efforts help to keep the U.S. electric power grid is reliable. We also consulted with FERC and EPRI staff on bulk power system reliability and related issues.

U.S. electric power utilities routinely upgrade the nation's electric power system to improve grid reliability and to meet new electric power demands. For example, when confronted with rapid adoption of air conditioners in the 1960s and 1970s, U.S. electric power utilities successfully met the new demand for electricity by planning and building upgrades to the electric power distribution system. Likewise, U.S. electric power utilities planned and built distribution system upgrades required to service the rapid growth of power-intensive data centers and server farms over the past two decades. U.S. electric power utilities have already successfully designed and built the distribution system infrastructure required for 1.4 million battery electric vehicles. Utilities have also successfully integrated 46.1 GW of new utility-scale electric generating capacity into the grid.

When taking into consideration ongoing upgrades to the U.S. electric power grid, and that the U.S. electric power utilities generally have more capacity to produce electricity than is consumed, the expected increase in electric power demand attributable to vehicle electrification is not expected to adversely affect grid reliability due to the modest increase in electricity demand associated with electric vehicle charging. The additional electricity demand from HD BEVs will depend on the time of day that charging occurs, the type or power level of charging, and the use of onsite storage and vehicle-to-grid (V2G) or other vehicle-grid-integration (VGI) technology, among other considerations, as discussed in DRIA Chapter 1.6.4. As noted by Lipman et al., a wide variety of organizations are engaged in VGI research, including the California Energy Commission, California Public Utilities Commission, California Independent System Operator, the Electric Power Research Institute, as well as charging providers, utilities ( e.g., SCE, PG&E, SDG&E), and automakers. Electric Island, a truck charging station deployed by Daimler Trucks North America and Portland General Electric which is planned to eventually include megawatt-level charging, will offer an opportunity to test energy management and VGI with heavy-duty BEVs. Future plans for Electric Island also include the use of onsite solar generation and battery storage.

Finally, we note that DOE is engaged in multiple efforts to modernize the grid and improve resilience and reliability. For example, in November 2022, DOE announced $13 billion in funding opportunities under the BIL to support transmission and distribution infrastructure. This includes $3 billion for smart grid grants with a focus on PEV integration among other topics.

3. FCEV Technology Costs

FCEVs and BEVs include many of the same components such as a battery pack, e-motor, power electronics, gearbox unit, final drive, and electrical accessories. Therefore we used the same costs for these components across vehicles used for the same applications; for detailed descriptions of these components, see DRIA Chapter 2.4.3. In this subsection and DRIA Chapter 2.5.2, we present the costs for components for FCEVs that are different from a BEV. These components include the fuel cell stack and hydrogen fuel tank. The same energy cell battery costs used for BEVs are used for fuel cell vehicles, but the battery size of a comparable FCEV is smaller. We request comment, including data, on our approach and cost projections for FCEV components.

i. Fuel Cell Stack Costs

The fuel cell stack is the most expensive component of a heavy-duty FCEV. Fuel cells for the heavy-duty sector are expected to be more expensive than fuel cells for the light-duty sector because they operate at higher average continuous power over their lifespan, which requires a larger fuel cell stack size, and because they have longer durability needs ( i.e., technology targets are for 25,000 to 30,000 hours for a truck versus 8,000 hours for cars).

Projected costs vary widely in the literature. They are expected to decrease as manufacturing matures. Larger production volumes are anticipated as global demand increases for fuel cell systems for HD vehicles, which could improve economies of scale. Costs are also anticipated to decline as durability improves, which could extend the life of fuel cells and reduce the need for parts replacement. Burke et al. compared estimates from the literature and chose values of $240 per kW in 2025 for a high case in their analysis, based on 1,000 heavy-duty fuel cell units produced per year, and $145 per kW for both a low case in 2025 and a high case in 2030, based on 3,000 units produced per year.

The interim DOE cost target for Class 8 tractor-trailer fuel stacks is $80 per kW by 2030. Their ultimate target is $60 per kW in 2050, set to ensure that costs are comparable to those of advanced diesel engines and other factors. These targets are based on 100,000 units per year production volume. They pointed to analysis that suggests that 2019 costs at a manufacturing volume of 1,000 units per year were around $190 per kW. In BEAN model updates, ANL estimated a range based on vehicle type of between $156 per kW and $174 per kW in 2025, and from $65 per kW to $99 per kW by 2035.

A Sharpe and Basma meta-study of other reports found 2025 costs ranging from $750 per kW to $50 per kW. The authors stated that they expect fuel cell costs to drop by about 30 percent between 2025 and 2030 due to manufacturer learning, improved materials and performance, and economies of scale. Like the approach we took for BEV battery costs, we averaged the 2025 cost values from the Sharpe and Basma meta-study, averaged the 2030 values, and then linearly interpolated to get MY 2027 values and adjusted to 2021$; we then applied the learning curve shown in DRIA Chapter 3.2.1 to calculate MY 2028–2032 values. The resulting fuel cell stack direct manufacturing costs are shown in Table II–15.

ii. Hydrogen Fuel Tank Costs

Hydrogen storage cost projections also vary widely in the literature. Sharpe and Basma reported costs ranging from as high as $1,289 per kg to $375 per kg of usable hydrogen in 2025. They expect hydrogen tank costs to drop by 21 percent between 2025 and 2030 due to lighter weight and lower cost carbon fiber-reinforced materials, technology improvements, and economies of scale.

The interim DOE target for Class 8 tractor-trailers is $300 per kg of hydrogen by 2030. Their ultimate target is $266 per kg (2016$) by 2050. They include all components necessary to support the tank and are based on a production volume of 100,000 tanks per year. They point to analysis that suggests that 2019 costs for 700-bar tanks at a manufacturing volume of 1,000 tanks per year were roughly $1,200 per kg. For reference, the Kenworth “beta” fuel cell truck holds six 10-kg hydrogen storage tanks at 700 bar.

Like the approach we took for battery and fuel cell stack costs, we averaged all of the 2025 cost values in the Sharpe and Basma meta-study, averaged all of the 2030 values, and then linearly interpolated to determine the MY 2027 value, adjusted to 2021 dollars. We applied the learning curve shown in DRIA Chapter 3.2.1 to calculate MY 2028–2032 values. The hydrogen fuel tank direct manufacturing costs are shown in Table II–16.

4. Inflation Reduction Act Tax Credits

The IRA, which was signed into law on August 16, 2022, includes a number of provisions relevant to vehicle electrification. There are two provisions of the IRA we included within our quantitative analysis in HD TRUCS. First, Section 13502, “Advanced Manufacturing Production Credit,” provides up to $45 per kWh tax credits for the production and sale of battery cells and modules when such components are produced in the United States and other qualifications are met. Second, Section 13403, “Qualified Commercial Clean Vehicles,” provides for a vehicle tax credit applicable to HD vehicles if certain qualifications are met. Beyond these two tax credits described in sections 13403 and 13502 of the IRA, there are numerous provisions in the IRA and the BIL that may impact HD vehicles and increase adoption of HD ZEV technologies. These range from tax credits across the supply chain, to grants which may help direct ZEVs to communities most burdened by air pollution, to funding for programs to build out electric vehicle charging infrastructure, as described in Section I of this preamble and DRIA Chapter 1.3.2. We welcome comment on our assessment of how the IRA will impact the heavy-duty industry, and how EPA could consider reflecting those impacts in our assessment for establishing the HD GHG standards under this proposal, including comment on methods to appropriately account for these provisions in our assessment.

Regarding the first of the two provisions, IRA section 13502, “Advanced Manufacturing Production Credit,” provides up to $45 per kWh tax credits for the production and sale of battery cells (up to $35 per kWh) and modules (up to $10 per kWh) and 10 percent of the cost of producing critical minerals such as those found in batteries, when such components or minerals are produced in the United States and other qualifications are met. These credits begin in CY 2023 and phase down starting in CY 2030, ending after CY 2032. As further discussed in DRIA Chapter 2.4.3.1, we recognize that there are currently few manufacturing plants for HD vehicle batteries in the United States. We expect that the industry will respond to this tax credit incentive by building more domestic battery manufacturing capacity in the coming years, in part due to the BIL and IRA. For example, Proterra recently announced its first heavy-duty battery manufacturing plant in the United States, Tesla is expanding its facilities in Nevada to produce its Semi BEV tractor and battery cells, and Cummins has entered into an agreement with Arizona-based Sion Power to design and supply battery cells for commercial electric vehicle applications. In addition, DOE is funding through the BIL battery materials processing and manufacturing projects to “support new and expanded commercial-scale domestic facilities to process lithium, graphite and other battery materials, manufacture components, and demonstrate new approaches, including manufacturing components from recycled materials.” Thus, we model this tax credit in HD TRUCS such that HD BEV and FCEV manufacturers fully utilize the battery module tax credit and gradually increase their utilization of the cell tax credit for MY 2027–2029 until MY 2030 and beyond, when they earn 100 percent of the available cell and module tax credits. The battery pack costs and battery tax credits used in our analysis are shown in Table II–17. We request comment on our approach to modeling this tax credit, including our projection that the full value of the tax credit earned by the manufacturer is passed through to the purchaser because market competition would drive manufacturers to minimize their prices.

Regarding the second of the two provisions, IRA section 13403 creates a tax credit applicable to each purchase of a qualified commercial clean vehicle. These vehicles must be on-road vehicles (or mobile machinery) that are propelled to a significant extent by a battery-powered electric motor. The battery must have a capacity of at least 15 kWh (or 7 kWh if it is Class 3 or below) and must be rechargeable from an external source of electricity. This limits the qualified vehicles to BEVs and plug-in hybrid electric vehicles (PHEVs). Additionally, fuel cell electric vehicles (FCEVs) are eligible. The credit is available from calendar year (CY) 2023 through 2032, which overlaps with the model years for which we are proposing standards (MYs 2027 through 2032), so we included the tax credit in our calculations for each of those years in HD TRUCS.

For BEVs and FCEVs, the tax credit is equal to the lesser of: (A) 30 percent of the BEV or FCEV cost, or (B) the incremental cost of a BEV or FCEV when compared to a comparable ICE vehicle. The limit of this tax credit is $40,000 for Class 4–8 commercial vehicles and $7,500 for commercial vehicles Class 3 and below. For example, if a BEV costs $350,000 and a comparable ICE vehicle costs $150,000, the tax credit would be the lesser of: (A) 0.30 × $350,000 = $105,000 or (B) $350,000 − $150,000 = $200,000. In this example, (A) is less than (B), but (A) exceeds the limit of $40,000, so the tax credit would be $40,000.

We included this tax credit in HD TRUCS by decreasing the incremental upfront cost a vehicle purchaser must pay for a ZEV compared to a comparable ICE vehicle following the process explained in the previous paragraph. The calculation for this tax credit was done after applying a retail price equivalent to our direct manufacturing costs. We did not calculate the full cost of vehicles in our analysis, instead we determined that all Class 4–8 ZEVs could be eligible for the full $40,000 (or $7,500 for ZEVs Class 3 and below) if the incremental cost calculated compared to a comparable ICE vehicle was greater than that amount. In order for this determination to be true, all Class 4–8 ZEVs must cost more than $133,333 such that 30 percent of the cost is at least $40,000 (or $25,000 and $7,500, respectively, for ZEVs Class 3 and below), which seems reasonable based on our assessment of the literature. As in the calculation described in the previous paragraph, both (A) and (B) are greater than the tax credit limit and the vehicle purchaser may receive the full tax credit. The incremental cost of a ZEV taking into account the tax credits for each vehicle segment in MY 2027 and MY 2032 are included in DRIA Chapter 2.8.2. We welcome comment on how we included the IRA tax credits for HD vehicles in our assessment.

5. Operating Costs

Operating costs for HD vehicles encompass a variety of costs, such as labor, insurance, registration fees, fueling, maintenance and repair (M&R), and other costs. For this analysis, we are primarily interested in costs that would differ for a comparable diesel-powered ICE vehicle and a ZEV. These operational cost differences are used to calculate an estimated payback period in HD TRUCS. We expect fueling costs and M&R costs to be different for ZEVs than for comparable diesel-fueled ICE vehicles, but we do not anticipate other operating costs, such as labor and insurance, to differ significantly, so the following subsections focus on M&R and fueling costs. Operating costs are averaged over a 10-year time period of the annual M&R cost and annual fuel cost.

i. Maintenance and Repair Costs

M&R costs contribute to the overall operating costs for HD vehicles. To establish a baseline cost for maintenance and repair of diesel-fueled ICE vehicles, we relied on the research compiled by Burnham et al. and used equations found in the ANL's BEAN model. Burnham et al. used data from Utilimarc and the American Transportation Research Institute (ATRI) to estimate maintenance and repair costs per mile for multiple heavy-duty vehicle categories over time. We selected the box truck curve to represent vocational vehicles and short-haul tractors, and the semi-tractor curve to represent long-haul tractors. Additional details regarding this analysis can be found in DRIA Chapter 2.3.4.2. Averaging the M&R costs for years 0–9 yields about 67 cents per mile for vocational vehicles and short-haul tractors and about 25 cents per mile for long-haul tractors, after adjusting to 2021$. We welcome comment, including additional data, on our approach and assessment of HD ICE vehicle M&R costs.

Data on real-world M&R costs for HD ZEVs is limited due to limited HD ZEV technology adoption today. We expect the overall maintenance costs to be lower for ZEVs compared to a comparable ICE vehicles for several reasons. First, an electric powertrain has fewer moving parts that accrue wear or need regular adjustments. Second, ZEVs do not require fluids such as engine oil or diesel exhaust fluid (DEF), nor do they require exhaust filters to reduce particulate matter or other pollutants. Third, the per-mile rate of brake wear is expected to be lower for ZEVs due to regenerative braking systems. Several literature sources propose applying a scaling factor to diesel vehicle maintenance costs to estimate ZEV maintenance costs. We followed this approach and applied a maintenance and repair cost scaling factor of 0.71 for BEVs and 0.75 for FCEVs to the maintenance and repair costs of diesel-fueled ICE vehicles. The scaling factors are based on an analysis from Wang et al. that estimates a future BEV heavy-duty truck would have a 29 percent reduction, and a future FCEV heavy-duty vehicle would have a 25 percent reduction, compared to a diesel-powered heavy-duty vehicle. We welcome comment on our approach and these projections.

In our payback analysis in HD TRUCS, we did not account for potential diesel engine rebuild costs for ICE vehicles, potential replacement battery costs for BEVs, or potential replacement fuel cell stack costs for FCEVs because our payback analysis typically covers a shorter period of time than the expected life of these components. Typical battery warranties being offered by HD BEV manufacturers range between 8 and 15 years today. A BEV battery replacement may be practically necessary over the life of a vehicle if the battery deteriorates to a point where the vehicle range no longer meets the vehicle's operational needs. We believe that proper vehicle and battery maintenance and management can extend battery life. For example, manufacturers will utilize battery management system to maintain the temperature of the battery as well active battery balancing to extend the life of the battery. Likewise, pre-conditioning has also shown to extend the life of the battery as well. Furthermore, research suggests that battery life is expected to improve with new batteries over time as battery chemistry and battery charging strategies improve, such that newer MY BEVs will have longer battery life. We request comment on this approach for both ICE vehicles and ZEVs, in addition to data on battery and fuel stack replacement costs, engine rebuild costs, and expected component lifetime periods.

ii. Fuel, Electricity, and Hydrogen Costs

The annual fuel cost for operating a diesel-fueled ICE vehicle is a function of its yearly fuel consumption and the cost of diesel fuel. The yearly fuel consumption is described in DRIA Chapter 2.3.4.3. We used the DOE Energy Information Administration's (EIA) Annual Energy Outlook (AEO) 2022 transportation sector reference case projection for diesel fuel for on-road use for diesel prices. This value includes Federal and State taxes but excludes county and local taxes. The average annual fuel cost is averaged over a 10-year period.

The annual electricity cost for operating a HD electric vehicle is a function of the electricity price, daily energy consumption of the vehicle, and number of operating days in a year. In HD TRUCS, we used the DOE EIA AEO 2022 reference case commercial electricity end-use rate projection. We selected this value instead of the transportation end use prices in AEO because those are similar to the prices for the residential sector, which implies they may be more relevant to light-duty vehicle charging than commercial truck charging.

For the purposes of the HD TRUCS analysis, rather than focusing on depot hydrogen fueling infrastructure costs that would be incurred upfront, we included infrastructure costs in our per-kilogram retail price of hydrogen. The retail price of hydrogen is the total price of hydrogen when it becomes available to the end user, including the costs of production, distribution, storage, and dispensing at a fueling station. This price per kilogram of hydrogen includes the amortization of the station capital costs. This approach is consistent with the method we use in HD TRUCS for ICE vehicles, where the equivalent diesel fuel costs are included in the diesel fuel price instead of accounting for the costs of fuel stations separately.

We acknowledge that this market is still emerging and that hydrogen fuel providers will likely pursue a diverse range of business models. For example, some businesses may sell hydrogen to fleets through a negotiated contract rather than at a flat market rate on a given day. Others may offer to absorb the infrastructure development risk for the consumer, in exchange for the ability to sell excess hydrogen to other customers and more quickly amortize the cost of building a fueling station. FCEV manufacturers may offer a “turnkey” solution to fleets, where they provide a vehicle with fuel as a package deal. These uncertainties are not reflected in our hydrogen price estimates presented in the DRIA.

As discussed in DRIA Chapter 1.3.2 and 1.8, large incentives are in place to reduce the price of hydrogen production, particularly from electrolytic sources. In June 2021, DOE launched a Hydrogen Shot goal to reduce the cost of renewable hydrogen production by 80 percent to $1 per kilogram in one decade. The BIL and IRA included funding for several hydrogen programs to accelerate progress towards the Hydrogen Shot and jumpstart the hydrogen market in the U.S.

For example, the BIL requires development of a National Clean Hydrogen Strategy and Roadmap. In September 2022, DOE released a draft of a holistic plan that shows how low-GHG hydrogen can help reduce emissions throughout the country by about 10 percent by 2050 relative to 2005 levels. DRIA Chapter 2.5.3.1 further discusses DOE's National Clean Hydrogen Strategy and Roadmap.

Recent analysis from ANL using BEAN includes a hydrogen price of $4.37 per gallon diesel equivalent (gde) in 2030, which equates to roughly $3.92 per kg hydrogen. This analysis was published after the IRA was passed, and reflects a lower H2 price in 2030 than was in the previous year's analysis. This price is at the low end of the range published in DOE's “Pathways to Commercial Liftoff” report on Clean Hydrogen (“Liftoff Report”), which projects that heavy-duty road transport can expect to pay a retail price of between $4 and $5 per kg of hydrogen in 2030 if advances in distribution and storage are commercialized. This price incorporates BIL and IRA incentives for hydrogen. Other DOE estimates prior to the IRA ranged from $6-$7 per kg in 2030, inclusive of production, delivery, and dispensing, with the range representing uncertainty in the assumed rate of technological progress.

Other available estimates explore clean hydrogen production costs alone. For example, Rhodium Group found a hydrogen producer price of $0.39–1.92 per kg, including the IRA hydrogen production tax credit and assuming the use of utility-scale solar to produce hydrogen. McKinsey projected green hydrogen costs of roughly $1.30–2.30 per kg in 2030, produced using alkaline electrolyzers. Their analysis did not mention the IRA. It showed lower costs for blue and grey hydrogen in 2030 before green hydrogen out-competes both by around 2040. An ICCT estimate of average hydrogen production costs in 2030 is closer to $3.10 per kg, but their analysis did not consider IRA impacts.

According to the Hydrogen Council, increasing the scale and rate of use of hydrogen across sectors could substantially reduce the costs of local distribution. As trucking capacity increases and the use, size, and density of refueling stations increases, equipment manufacturing costs could decline. For example, they suggest that 2020 distribution costs of about $5–6 per kg could decline by approximately 80 percent to get to $1–1.50 per kg in 2030. A 2018 DOE document details similar opportunities to reach $2 per kg in distribution and dispensing costs. In addition to learning and economies of scale associated with scaled use, they suggest that potential research and development advancements related to the efficiency and reliability of components could help meet related DOE price targets.

As further explained in DRIA Chapter 2.5.3.1, for use in HD TRUCS, we projected the future hydrogen prices shown in Table II–18 for 2027–2030 and beyond. These values are based on ANL BEAN values and are in line with price projections in DOE's Liftoff Report that consider the impacts of BIL and IRA. We converted the $/kg estimates for 2025 and 2030 included in BEAN to dollar per kg by using the conversion factor of 1 gallon of diesel is equivalent to 1.116 kg of hydrogen, based on its lower heating value. We rounded up to the nearest $0.50 increment given the uncertainty of projections, and then interpolated for 2027 to 2029. Prices for 2030 and beyond are held constant in BEAN and in HD TRUCS.

We request comment on our approach and assessment of future fuel, electricity, and hydrogen prices for the transportation sector.

6. Payback

After assessing the suitability of the technology and costs associated with ZEVs, a payback calculation was performed on each of the 101 HD TRUCS vehicles for the BEV technology and FCEV technology that we were considering for the technology packages for each use case for each MY in the MY 2027–2032 timeframe. The payback period was calculated by determining the number of years that it would take for the annual operational savings of a ZEV to offset the incremental upfront purchase price of a BEV or FCEV (after accounting for the IRA section 13502 battery tax credit and IRA section 13403 vehicle tax credit as described in DRIA Chapters 2.4.3.1 and 2.4.3.5, respectively) and charging infrastructure costs (for BEVs) when compared to purchasing a comparable ICE vehicle. The ICE vehicle and ZEV costs calculated include the retail price equivalent (RPE) multiplier of 1.42 to include both direct and indirect manufacturing costs, as discussed further in DRIA Chapter 3. The operating costs include the diesel, hydrogen or electricity costs, DEF costs, and the maintenance and repair costs. The payback results are shown in Table 2–75 and Table 2–76 for BEVs for MY 2027 and MY 2032, and in Table 2–77 for FCEVs for MY 2032 of DRIA Chapter 2.

F. Proposed Standards

Similar to the approach we used to support the feasibility of the HD GHG Phase 2 vehicle CO₂ emission standards, we developed technology packages that, on average, would meet each of the proposed standards for each regulatory subcategory of vocational vehicles and tractors after considering the various factors described in this section, including technology costs for manufacturers and costs to purchasers. We applied these technology packages to nationwide production volumes to support the proposed Phase 3 GHG vehicle standards. The technology packages utilize the averaging portion of the longstanding ABT program, and we project manufacturers would produce a mix of HD vehicles that utilize ICE-powered vehicle technologies and ZEV technologies, with specific adoption rates for each regulatory subcategory of vocational vehicles and tractors for each MY. Note that we have analyzed a technology pathway to support the feasibility and appropriateness of each proposed level of stringency for each proposed standard, but manufacturers would be able to use a combination of HD engine or vehicle GHG-reducing technologies, including zero-emission and ICE technologies, to meet the standards.

The proposed standards are shown in Table II–19 and Table II–20 for vocational vehicles and Table II–21and Table II–22 for tractors. We request comment and data on our proposal as well as comment and data supporting more or less stringent HD vehicle GHG standards than those proposed, as specified in Section II.H. We also request comment on setting additional new HD vehicle GHG standards in MYs 2033 through 2035 that are more progressively stringent than the MY 2032 standards and that either continue the approach and trajectory of the proposed standards or utilize a different approach and trajectory that we solicited comment on in this proposal.

The approach we used to select the proposed standards, described in this Section II, does not specifically include accounting for ZEV adoption rates that would result from compliance with the California ACT program. The approach we used developed ZEV technology adoption rates on a nationwide basis. EPA granted the California ACT waiver request on March 30, 2023, which did not allow sufficient time for us to consider an alternative approach for this proposal. With the granting of the California ACT waiver, we intend to consider for the final rule how vehicles sold to meet the ACT requirement in California and other states that may adopt it under CAA section 177 would impact or be accounted for in the standard setting approach described in this Section II. For example, we may adjust our reference case to reflect the ZEV levels projected from ACT in California and other states. We also may consider increasing the technology adoption rates in the technology packages and correspondingly increase the stringency of the proposed Phase 3 emission standards to account for the incremental difference in the projected ZEV adoption levels from the proposed Phase 3 emission standards and the adoption levels projected from ACT in those states. We welcome comment on how to consider this ACT in our proposed approach or in other approaches.

We are proposing new CO₂ emission standards using the regulatory subcategories we adopted in HD GHG Phase 2, as discussed in Section II.C. As we discuss later in this subsection, the fraction of ZEVs and fraction of ICE vehicles in the technology packages varies across the 101 HD TRUCS vehicle types and thus in the regulatory subcategories. We recognize there may be different regulatory structures that could be used to reduce GHG emissions from the HD vehicles.

During the development of this proposed action, EPA has heard requests from several stakeholders that EPA consider establishing CO₂ standards for specific vehicle applications ( e.g., school buses, urban buses, pick-up and delivery vehicles, drayage trucks, etc.), as a complement to CO₂ emission standards that utilize the existing HD GHG Phase 2 program structure. There are several reasons stakeholders have explained for asking EPA to consider this approach. One reason is to target specific applications which may be the most suited for more stringent CO₂ standards at a more rapid pace than a broader regulatory subcategory. For example, a pick-up and delivery application may be more suitable for faster adoption of BEV technology than the broader subcategory of medium heavy-duty vocational vehicles. This approach could further support the industry and marketplace focusing resources on specific applications in the near term in response to more stringent EPA standards, rather than potentially spreading those resources across a broader range of products. Another reason some stakeholders suggested EPA consider an application-specific approach would be to accelerate the deployment of ZEVs that are concentrated in frontline communities to reduce air pollution more quickly in those communities.

We note the current HD GHG Phase 2 program structure includes standards at broad vehicle subcategory levels ( e.g., light heavy-duty vocational vehicles, medium heavy-duty vocational vehicles, etc.) as well as optional CO₂ emission standards for seven specific custom chassis applications ( e.g., emergency vehicles, motor homes, cement mixers, school buses). It is important to note the suggestions from stakeholders for EPA to establish application-specific standards for some heavy-duty vehicles to accelerate emission reductions in the Phase 3 program are much different than the reasons EPA established the HD GHG Phase 2 optional custom chassis standards. EPA established the optional custom chassis program for a number of reasons, including: a recognition there are manufacturers who produce specialized heavy-duty vocational vehicles where some of the technologies EPA used for the primary program standards would be unsuited for use, concern that the primary program drive cycles are either unrepresentative or unsuitable for certain specialized heavy-duty vocational vehicles, concern that some manufacturers of these specialized vocational vehicles have limited product offerings such that the primary program's emissions averaging is not of practical value as a compliance flexibility, and also concern regarding the appropriateness of the primary program's vocational vehicle standards as applied to certain specialized/custom vocational vehicles (See 81 FR 73531 and 81 FR 73686, October 25, 2016).

Potential challenges EPA recognizes with an application-specific, more stringent CO₂ standard approach include determining what criteria EPA would use to establish application-specific standards, how such standards would fit in the overall Phase 3 program structure, and the difficulty in defining some applications. For example, a drayage truck in general can be any Class 8 tractor (both sleeper cab and day cab) that is used to move shipping containers to and from ports from other locations, including rail yards, shipping terminals, or other destinations. A drayage tractor is not a unique application nor do these tractors contain unique design features to differentiate them from other tractors—nearly any tractor can be used for drayage operation. Nevertheless, in consideration of potentially targeting specific applications most suited for more stringent CO₂ standards at a more rapid pace than a broader regulatory subcategory, EPA requests comment on a standards structure for Phase 3 which would establish unique, mandatory, application-specific standards for some subset of heavy-duty vehicle applications. EPA requests comment on what data, what program structure, what applications, and what criteria EPA should consider for designing application-specific standards. EPA also requests comment on how the application-specific CO₂ standards would interact with the broader Phase 3 program structure EPA has included in this proposal, including the CO₂ emissions averaging, banking, and trading program. For example, if EPA were to separate these applications and apply more stringent standards, EPA requests comment on whether emission credits should be allowed to be averaged across the primary Phase 3 program and the application specific standards, and if yes, what limits if any should apply to those standards. Under this example, EPA may consider that allowing credits to flow into an application-specific category could undermine the reasons for establishing such a category (to accelerate the application of technology and accelerate emission reductions), while allowing credits generated within an application specific category to flow into the primary program may provide incentive for even greater reductions from the application-specific category.

To support that the proposed standards are achievable through the technology pathway projected in the technology packages, the proposed CO₂ standards for each subcategory were determined in two steps giving consideration to costs, lead time, and other factors, as described in this section and Section II.G. First, we determined the technology packages that include ZEVs and ICE vehicles with GHG-reducing technologies for each of the vocational vehicle and tractor subcategories as discussed in Section II.F.1. Then we determined the numeric level of the proposed standards as discussed in Sections II.F.2 and II.F.3.

1. Technology Adoption Rates in the Technology Packages

We based the proposed standards on technology packages that include both ICE vehicle and ZEV technologies. In our analysis, the ICE vehicles include a suite of technologies that represent a vehicle that meets the existing MY 2027 Phase 2 CO₂ emission standards. These technologies exist today and continue to evolve to improve the efficiency of the engine, transmission, drivetrain, aerodynamics, and tire rolling resistance in HD vehicles and therefore reduce their CO₂ emissions. There also may be opportunity for further adoption of these Phase 2 ICE technologies beyond the adoption rates used in the HD GHG Phase 2 rule. In addition, the heavy-duty industry continues to develop CO₂ -reducing technologies such as hybrid powertrains and H2–ICE powered vehicles.

In the transportation sector, new technology adoption rates often follow an S-shape. As discussed in the preamble to the HD GHG Phase 2 final rule, the adoption rates for a specific technology are initially slow, followed by a rapid adoption period, then leveling off as the market saturates, and not always at 100 percent. For this proposal, we developed a method to project adoption rates of BEVs and FCEVs in the HD vehicle market after considering methods in the literature. Our adoption function, and methods considered and explored in the formulation of the method used in this proposal, are described in DRIA Chapter 2.7.9. As stated there, given information currently available and our experience with the HD vehicle industry, when purchasing a new vehicle, we believe that the payback period is the most relevant metric to determine adoption rates in the HD vehicle industry.

The ZEV adoption rate schedule, shown in Table II–23, shows that when the payback is immediate, we project up to 80 percent of a manufacturer's fleet to be ZEV, with diminishing adoption as the payback period increases. The schedule was used to assign ZEV adoption rates to each of the 101 HD TRUCS vehicle types based on its payback period for MYs 2027 and 2032.

We phased in the proposed standards gradually between MYs 2027 and 2032 to address potential lead time concerns associated with feasibility for manufacturers to deploy ZEV technologies that include consideration of time necessary to ramp up battery production, including the need to increase the availability of critical raw materials and expand battery production facilities, as discussed in Section II.D.2.ii. We also phased in the proposed standards recognizing that it will take time for installation of EVSE by the BEV purchasers. We project BEV adoption as early as MY 2027, and we project adoption of FCEVs in the technology packages starting in MY 2030 for select applications that travel longer distances and/or carry heavier loads ( i.e., coach buses, heavy-haul tractors, sleeper cab tractors, and day cab tractors). There has been only limited development of FCEVs for the HD market to date, therefore our assessment is that it would be appropriate to provide manufacturers with additional lead time to design, develop, and manufacture FCEV models, but that it would be feasible by MY 2030. With substantial Federal investment in low-GHG hydrogen production (see DRIA Chapter 1.8.2), we anticipate that the price of hydrogen fuel could drop enough by 2030 to make HD FCEVs cost-competitive with comparable ICE vehicles for some duty cycles. We also note that the hydrogen infrastructure is expected to need additional time to further develop, as discussed in greater detail in DRIA Chapter 1.8, but we expect the refueling needs can be met by MY 2030. We also recognize the impact regulations can have on technology and recharging/refueling infrastructure development and deployment. Thus we request comment and data on our proposed adoption rate, including schedule and methods. We also request comment and data to support other adoption rate schedules; see also Section II.H.

We applied an additional constraint within HD TRUCS that limited the maximum penetration rate to 80 percent for any given vehicle type. This conservative limit was developed after consideration of the actual needs of the purchasers related to two primary areas of our analysis. First, this 80 percent volume limit takes into account that we sized the batteries, power electronics, e-motors, and infrastructure for each vehicle type based on the 90th percentile of the average VMT. We utilize this technical assessment approach because we do not expect heavy-duty OEMs to design ZEV models for the 100th percentile VMT daily use case for vehicle applications, as this could significantly increase the ZEV powertrain size, weight, and costs for a ZEV application for all users, when only a relatively small part of the market would need such capabilities. Therefore, the ZEVs we analyzed and have used for the feasibility and cost projections for this proposal are likely not appropriate for 100 percent of the vehicle applications in the real-world. Our second consideration for including an 80 percent volume limit for ZEVs is that we recognize there is a wide variety of real-world operation even for the same type of vehicle. For example, some owners may not have the ability to install charging infrastructure at their facility, or some vehicles may need to be operational 24 hours a day. Under our proposed standards, ICE vehicles would continue to be available to address these specific vehicle applications. We request comment, data, and analysis on both of these considerations and our use of an 80 percent volume limit. Our request for comment includes a request for data to inform an assessment of the distribution of daily miles traveled and the distribution of the number of hours available daily to charge for each of the vehicle types that we could use to update a constraint like this in the final rulemaking analysis.

After the technology assessment, as described in Section II.D.4 and DRIA Chapter 2, and payback analysis, as described in Section II.E.6 and DRIA Chapter 2.8.2, EPA determined the ICE vehicle and ZEV adoption rates for each regulatory subcategory. We first determined the ZEV adoption rates projected for each of the 101 vehicle types for MYs 2027 and 2032, which can be found in DRIA Chapter 2.8.3.1. We then aggregated the projected ZEV adoption rates for the specific vehicle types into their respective regulatory subcategories relative to the vehicle's sales weighting, as described in DRIA Chapter 2.9.1. The resulting projected ZEV adoption rates (shown in Table II–24) and projected ICE vehicle adoption rates that achieve a level of CO₂ emissions performance equal to the existing MY 2027 emission standards (shown in Table II–21) were built into our technology packages. We request comment and data on our projected adoption rates in the technology packages as well as data supporting higher or lower adoption rates than the projected levels. We also request comment on projecting adoption rates out through MY 2035.

2. Calculation of the Proposed Tractor Standards

The proposed CO₂ emission standards for the tractor subcategories are calculated by determining the CO₂ emissions from a technology package that consists of both ICE-powered vehicles and ZEVs. The projected fraction of ZEVs that emit zero grams CO₂ /ton-mile at the tailpipe are shown in Table II–24. The remaining fraction of vehicles in the technology package are ICE-powered vehicles that include the technologies listed in Table II–1 (reflecting the GEM inputs for the individual technologies that make up the technology packages that meets the existing MY 2027 CO₂ tractor emission standards). Thus, in the technology packages, the ICE-powered vehicles emit at the applicable existing MY 2027 CO₂ emission standards, as shown in Table II–26. We request comment on ICE vehicle technologies that could support more stringent standards than those proposed.

The proposed CO₂ emission standards for each model year are calculated by multiplying the fraction of ICE-powered vehicles in each technology package by the applicable existing MY 2027 CO₂ emission standards. The proposed standards are presented in Section II.F.

3. Calculation of the Proposed Standards for Vocational Vehicles

The proposed CO₂ emission standards for the vocational vehicles regulatory subcategories are calculated by determining the CO₂ emissions from a technology package that consists of both ICE-powered vehicles and ZEVs. The projected fraction of ZEVs that emit zero grams CO₂ /ton-mile at the tailpipe are shown in Table II–24. The remaining fraction of vehicles in the technology package are ICE-powered vehicles that include the technologies listed in Table II–2 (reflecting the GEM inputs for the individual technologies that make up the technology packages that meets the existing MY 2027 CO₂ vocational vehicles emission standards). We request comment on ICE vehicle technologies that could support more stringent standards than those proposed.

As discussed in Section II.C, vocational vehicle CO₂ emission standards are subdivided by weight class, SI-powered or CI-powered vehicles, and by operation. There are a total of 15 different vocational vehicle standards in the primary program for each model year, in addition to the optional custom chassis standards. The existing MY 2027 vocational vehicle emission standards are shown in Table II–27 (which, like tractors, are what the ICE-powered vehicles emit at in the proposed technology packages). The HD GHG Phase 2 structure enables the technologies that perform best during urban driving or the technologies that perform best at highway driving to each be properly recognized over the appropriate drive cycles. The HD GHG Phase 2 structure was developed recognizing that there is not a single package of engine, transmission, and driveline technologies that is suitable for all ICE-powered vocational vehicle applications. In this proposal, we are continuing the current approach of deeming tailpipe emissions of regulated GHG pollutants (including CO₂ ) to be zero from electric vehicles and hydrogen fuel cell vehicles. Therefore, the need to recognize the variety in vocational vehicle CO₂ emissions may no longer be necessary for ZEVs because ZEVs are deemed to have zero CO₂ emissions. Similarly, the existing SI and CI distinction within vocational vehicle regulatory subcategory structure is not optimal for vocational ZEVs because they cannot be technically described as either SI-powered or CI-powered.

Also discussed in Section II.C, the vehicle ABT program allows credits to exchange with all vehicles within a weight class. ABT CO₂ emission credits are determined using the equation in 40 CFR 1037.705. The credits are calculated based on the difference between the applicable standard for the vehicle and the vehicle's family emission limit multiplied by the vehicle's regulatory payload and useful life. For example, as shown in Table II–28, using the existing light heavy-duty vocational vehicle MY 2027 CO₂ emission standards, the amount of credit a ZEV would earn varies between 124 Mg and 177 Mg, depending on the regulatory subcategory it would be certified to. We recognize that in many cases it may not be clear to the manufacturer whether to certify the vocational ZEV to a SI or CI regulatory subcategory, i.e. for the manufacturer to know whether the ZEV was purchased in lieu of a comparable CI-powered or SI-powered vehicle. Furthermore, as just discussed, because ZEVs have zero CO₂ vehicle exhaust emissions the programmatic basis for requiring the manufacturer to differentiate the ZEVs by operation to appropriately account for the variety of driveline configurations would not exist, though the amount of credit the ZEV would earn would depend on the regulatory subcategory selected for certification. In short, we recognize the difficulties in, and consequences of, determining which of the regulatory subcategories to which a ZEV should be certified under the existing HD GHG Phase 2 emission standards' structure for vocational vehicles. To address this concern, we are proposing a two-step approach. First, we propose to revise the ABT credit calculation regulations; this change would begin in MY 2027. Second, we derived the proposed MY 2027 and later standards accounting for the proposed changes to the ABT credit calculations. Note that BEVs, FCEVs, and H2–ICE vehicles would still be able to be certified to the vocational vehicle urban, multi-purpose, or regional standards or to the applicable optional custom chassis standards.

EPA proposes to revise the definition of the variable “Std” in 40 CFR 1037.705 to establish a common reference emission standard for vocational vehicles with tailpipe CO₂ emissions deemed to be zero ( i.e., BEVs, FCEVs, and vehicles with engines fueled with pure hydrogen). Beginning in MY 2027, manufacturers would use the applicable Compression-Ignition Multi-Purpose (CI MP) standard for their vehicle's corresponding weight class when calculating ABT emission credits for vocational vehicles with tailpipe CO₂ emissions deemed to be zero. We selected the CI MP standard because it is the regulatory subcategory with the highest production volume in MY 2021. We also recognize a need to balance two different timing considerations concerning the potential impacts of this proposed change. First, prior to the effective date of this proposed change, there is a potential for manufacturers producing BEVs, FCEVs, and certain H2–ICE vehicles to generate larger credits than they would after this change, depending on the vocational vehicle subcategory to which a vehicle is certified. Second, we recognize that manufacturers develop their emissions compliance plans several years in advance to manage their R&D and manufacturing investments. After taking these into account, we propose that this regulation revision become effective beginning in MY 2027 to provide manufacturers with sufficient time to adjust their production plans, if necessary. We request comment on this proposed revision.

Taking the proposed change to the ZEV ABT credit calculation into account, if we calculated the proposed standards by multiplying the fraction of ICE-powered vehicles in the technology package (by model year) by the applicable existing MY 2027 CO₂ emission standards, like we did for tractors, then this would lead to a scenario where it would take different levels of ZEV adoption rates to meet the proposed standards in each regulatory subcategory than we included in our assessment. Therefore, we used an alternate approach that maintains the same level of ZEV adoption rates in each regulatory subcategory within a weight class, taking the proposed change to the ZEV ABT credit calculation into account. The equation for calculating the proposed MY 2032 vocational vehicle standards is shown in Equation II–1. This equation is used to calculate the proposed standards for each vocational vehicle regulatory subcategory, using the existing MY 2027 CI MP standard for each corresponding weight class (LH, MH, HH). Equation II–2 through Equation II–4 show how the proposed Equation II–1 would be used for each regulatory subcategory for an example model year (MY 2032). The existing MY 2027 standards can be found in Table II–27, and the projected ZEV adoption rates by model year are in Table II–24. The same equations were used for the proposed MY 2027 through 2031 standards but replacing the MY 2032 Standards and ZEV adoption rates with values for the specific model year. The results of the calculations for the MY 2032 LHD vocational vehicles are shown in Table II–29. The calculations for the other model years and vocational vehicle subcategories are shown in DRIA Chapter 2.9.

Equation II–1: Proposed Vocational Vehicle Standard Calculation

MY 2032 Std_RegSubcat = Existing 2027 Std_RegSubcat −(MY 2027 Existing CI MP Std_RegSubcat * MY 2032 ZEV%)

Equation II–2: Proposed Vocational Vehicle Standard Calculation Light Heavy-Duty Regulatory Subcategories for MY 2032

MY 2032 Std_RegSubcat = Existing 2027 Std_RegSubcat −(330 g/mi * 57%)

Equation II–3: Proposed Vocational Vehicle Standard Calculation Medium Heavy-Duty Regulatory Subcategories for MY 2032

MY 2032 Std_RegSubcat = Existing 2027 Std_RegSubcat −(235 g/mi * 35%)

Equation II–4: Proposed Vocational Vehicle Standard Calculation Heavy Heavy-Duty Regulatory Subcategories for MY 2032

MY 2032 Std_RegSubcat = Existing 2027 Std_RegSubcat − (230 g/mi * 40%)

The calculations for the other model years and vocational vehicle subcategories are shown in DRIA Chapter 2.9. We welcome comment on this approach to taking the proposed change to the ZEV ABT credit calculation into account in setting vocational vehicle standards. We also request comment alternatively on using the same approach for vocational vehicles as we are proposing for tractors (see Section II.F.2).

After considering the potential concerns with ZEVs fitting into the existing HD GHG Phase 2 vocational vehicle regulatory subcategories structure, we are proposing to maintain the existing HD GHG Phase 2 vocational vehicle regulatory subcategories with the proposed changes noted in this section. We request comment on possible alternative vocational vehicle regulatory subcategory structures, such as reducing the number of vocational vehicle subcategories to only include the Multi-Purpose standards in each weight class, and/or maintaining Urban, Multipurpose, and Regional but combining SI and CI into a standard for each weight class.

The HD GHG Phase 2 program includes optional custom chassis emission standards for eight specific vehicle types. Those vehicle types may either meet the primary vocational vehicle program standards or, at the vehicle manufacturer's option, they may comply with these optional standards. The existing optional custom chassis standards are numerically less stringent than the primary HD GHG Phase 2 vocational vehicle standards, but the ABT program is more restrictive for vehicles certified to these optional standards. Banking and trading of credits is not permitted, with the exception that small businesses that may use traded credits to comply. Averaging is only allowed within each subcategory for vehicles certified to these optional standards. If a manufacturer wishes to generate tradeable credits from the production of these vehicles, they may certify them to the primary vocational vehicle standards.

In this action, we are proposing to establish more stringent standards for several, but not all, of these optional custom chassis subcategories. We are proposing revised MY 2027 emission standards and new MY 2028 through MY 2032 and later emission standards for the school bus, other bus, coach bus, refuse hauler, and concrete mixer optional custom chassis regulatory subcategories. We are not proposing any changes to the existing ABT program restrictions for the optional custom chassis regulatory subcategories. Because vehicles certified to the optional custom chassis standards would continue to have restricted credit use and can only be used for averaging within a specific custom chassis regulatory subcategory, we do not have the same potential credit concern as we do for the primary vocational vehicle standards. Therefore, we determined the proposed optional custom chassis emission standards by multiplying the fraction of ICE-powered vehicles in the technology package (by model year) by the applicable existing MY 2027 CO₂ emission standards, like we did for determining the proposed tractor emission standards.

We are not proposing new standards for motor homes certified to the optional custom chassis regulatory subcategory because of the projected impact of the weight of batteries in BEVs in the MYs 2027–2032, as described in DRIA Chapter 2.8.1. Furthermore, we also are not proposing new standards for emergency vehicles certified to the optional custom chassis regulatory subcategory due to our assessment that these vehicles have unpredictable operational requirements and may have limited access to recharging facilities while handling emergency situations in the MYs 2027–2032 timeframe. Finally, we are not proposing new standards for mixed-use vehicle optional custom chassis regulatory subcategory because these vehicles are designed to work inherently in an off-road environment (such as hazardous material equipment or off-road drill equipment) or be designed to operate at low speeds such that it is unsuitable for normal highway operation and therefore may have limited access to on-site depot or public charging facilities in the MYs 2027–2032 timeframe. We do not have concerns that manufacturers could inappropriately circumvent the proposed vocational vehicle standards or proposed optional custom chassis standards because vocational vehicles are built to serve a purpose. For example, a manufacturer cannot certify a box truck to the emergency vehicle custom chassis standards. We request comment on specific considerations and impacts the proposed standards would have on vehicles certified to these optional custom chassis standards. We also request comment and data regarding the potential for more stringent GHG standards for the motor homes, emergency vehicles, or mixed-use vehicles optional custom chassis regulatory subcategories in this time frame.

4. Summary of Costs To Meet the Proposed Emission Standards

We based the proposed standards on technology packages that include both ICE vehicle and ZEV technologies. In our analysis, the ICE vehicles include a suite of technologies that represent a vehicle that meets the existing MY 2027 Phase 2 CO₂ emission standards. We accounted for these technology costs as part of the HD GHG Phase 2 final rule. Therefore, our technology costs for the ICE vehicles are considered to be $0 because we did not add additional CO₂ -reducing technologies to the ICE vehicles beyond those in the baseline vehicle. The incremental cost of a heavy-duty ZEV is the marginal cost of ZEV powertrain components compared to ICE powertrain components on a comparable ICE vehicle. This includes the removal of the associated costs of ICE-specific components from the baseline vehicle and the addition of the ZEV components and associated costs. DRIA Chapter 2.3.2 and 2.4.3 includes the ICE powertrain and BEV powertrain cost estimates for each of the 101 HD vehicle types. DRIA Chapter 2.5.2 includes the FCEV powertrain cost projections for the coach buses, heavy-haul tractors, sleeper cab tractors, and day cab tractors.

i. Manufacturer Costs

Table II–30 and Table II–31 show the ZEV technology costs for manufacturers relative to the reference case described in Section V.A.1, including the direct manufacturing costs that reflect learning effects, the indirect costs, and the IRA section 13502 Advanced Manufacturing Production Credit, on average aggregated by regulatory group for MYs 2027 and 2032, respectively. The incremental ZEV adoption rate reflects the difference between the ZEV adoption rates in the technology packages that support our proposed standards and the reference case. As shown in Table II–30 and Table II–31, we project that some vocational vehicle types will achieve technology cost parity between comparable ICE vehicles and ZEVs for manufacturers by MY 2032. These vehicles in our analysis include school buses and single unit trucks (which include vehicles such as delivery trucks). Our analysis is consistent with other studies. For example, an EDF/Roush study found that by MY 2027, BEV transit buses, school buses, delivery vans, and refuse haulers would each cost less upfront than a comparable ICE vehicle. ICCT similarly found that “although zero-emission trucks are more expensive in the near-term than their diesel equivalents, electric trucks will be less expensive than diesel in the 2025–2030 time frame, due to declining costs of batteries and electric motors as well as increasing diesel truck costs due to emission standards compliance.” These studies were developed prior to passage of the IRA, and therefore we would expect the cost comparisons to be even more favorable after considering the IRA provisions. For example, the Rocky Mountain Institute found that because of the IRA, the TCO of electric trucks will be lower than the TCO of comparable diesel trucks about five years faster than without the IRA. They expect cost parity as soon as 2023 for urban and regional duty cycles that travel up to 250 miles and 2027 for long-hauls that travel over 250 miles.

i. Purchaser Costs

We also evaluated the costs of the proposed standards for purchasers on average by regulatory group, as shown in Table II–32 and Table II–33. Our assessment of the upfront purchaser costs include the incremental cost of a ZEV relative to a comparable ICE vehicle after accounting for the two IRA tax credits (IRA section 13502, “Advanced Manufacturing Production Credit,” and IRA section 13403, “Qualified Commercial Clean Vehicles”) and the associated EVSE costs, if applicable. We also assessed the incremental annual operating savings of a ZEV relative to a comparable ICE vehicle. The payback periods shown reflect the number of years it would take for the annual operating savings to offset the increase in total upfront costs for the purchaser.

As shown in Table II–33, under our proposal we estimate that the average upfront cost per vehicle to purchase a new MY 2032 vocational ZEV and associated EVSE compared to a comparable ICE vehicle (after accounting for two IRA tax credits, IRA section 13502, “Advanced Manufacturing Production Credit,” and IRA section 13403, “Qualified Commercial Clean Vehicles”), would be offset by operational costs ( i.e., savings that come from the lower costs to operate, maintain, and repair ZEV technologies), such that we expect the upfront cost increase would be recouped due to operating savings in one to three years, on average for vocational vehicles. For a new MY 2032 day cab tractor ZEV and associated EVSE, under our proposal we estimate the average incremental upfront cost per vehicle would be recovered in three years, on average. Similarly, for sleeper cab tractors, we estimate that the initial cost increase would be recouped in seven years. We discuss this in more detail in DRIA Chapter 2.

The average per-vehicle purchaser costs shown in Table II–32 for MY 2027 are higher than the MY 2032 per-vehicle costs. The reduction in costs over time are reflective of technology learning, as discussed in Section IV.B. It is worth noting that though the upfront costs of a BEV day cab tractor, for example, are higher when one considers both the vehicle and the EVSE, purchasers would still recoup these upfront costs within eight years of ownership on average. Also of note, our proposed standards in MY 2027 have a lower adoption rate of 10 percent for these day cab tractors, in recognition of the higher cost in MY 2027 than in MY 2032. The upfront vehicle cost increase projected at $24,000 represents a less than a 25 percent increase when compared to the average price of $100,000 for a new day cab tractor. Purchasers also would have the option to consider alternatives to purchasing an EVSE at the time of purchasing a vehicle. For example, depending on the location of the vehicle, heavy-duty public charging may be a better solution than depot charging. The purchaser could instead of spending over $37,000 upfront on average for EVSE, they could instead spread the cost over time through public charging where the EVSE costs would be built into the electricity cost.

5. Lead Time Assessment

Two of the significant aspects of the IRA are the tax credit available for the manufacturing of batteries and the tax credit available for the purchase of HD zero-emission vehicles, where the IRA provisions' qualifications are met. The tax credits significantly reduce, and in many cases erase, the incremental cost of purchasing a HD ZEV when compared to the cost of purchasing a comparable ICE vehicle. Therefore, as explained in our payback analysis, we expect the IRA will incentivize the demand and purchaser acceptance for HD ZEVs. However, demand and purchaser acceptance are only two of the factors we consider when evaluating the feasibility of HD ZEV technologies in the MY 2027 through MY 2032 timeframe. As we propose standards for MYs 2027 through 2032, which are between four and nine years from now, we considered the lead time required for manufacturers to design, develop, and produce the ZEV and ICE vehicle technologies in the technology packages, in addition to lead time considerations for the charging and hydrogen refueling infrastructure. We welcome comment on our assessment of lead time in these areas.

Manufacturers require time to design, develop, and build new vehicles. Based on discussions with heavy-duty manufacturers, depending on the amount of content that is new on a vehicle, it could take two to four years or more years to design, develop and prove the safety and reliability of a new HD vehicle. A typical design process includes the design and building of prototype or demonstration vehicles that are evaluated over several months or years in real world operation. The manufacturers need to accumulate miles and experience a wide variety of environmental conditions on these prototype vehicles to demonstrate the product's durability and reliability. Then manufacturers would work to commercialize the vehicle and in turn build it in mass production. We also considered that manufacturers are likely limited in terms of the financial resources, human resources, and testing facilities to redesign all of their vehicles at the same time. Typically, manufacturers would focus on the applications with the best business case because these would be where the customers would be most willing to purchase, therefore the proposed standards phase in over a period of time starting in MY 2027 through MY 2032. For HD BEVs, we have considered that BEV technology has been demonstrated to be technically feasible in heavy-duty transportation and that manufacturers will learn from the research and development work that has gone into developing the significant number of LD and HD electric vehicle models that are on the road today, as noted in Section II.D.2 and DRIA Chapter 1.5.5, and our proposed standards are supported by technology packages with increasing BEV adoption rates beginning in MY 2027 (see also our discussion in this subsection regarding our consideration of adequate time for infrastructure development for HD BEVs). For HD FCEVS, as discussed in Section II.D.3 and II.D.4, along with DRIA Chapter 1.7.5, fuel cell technology in other sectors has been in existence for decades, has been demonstrated to be technically feasible in heavy-duty transportation, and there are a number of HD FCEV models that are commercially available today with more expected to become available by 2024. However, we included this technology for our proposed standards starting in MY 2030 in part to take into consideration additional lead time to allow manufacturers to design, develop, and manufacture HD FCEV models (see also our discussion in this subsection regarding our consideration of adequate time for infrastructure development for HD FCEVs).

We discuss in Sections II.D.1 and II.F.1 the need for ICE vehicles to continue to install CO₂ -reducing technologies, such as advanced aerodynamics, efficient powertrains, and lower rolling resistance tires. In our technology assessment for this proposal, we included the technology packages we considered in setting the existing Phase 2 MY 2027 CO₂ emission standards. Each of these technologies exists today and continues to be developed by manufacturers. As noted in 2016 when we issued the HD GHG Phase 2 final rule, at that time we provided over ten years of lead time to the manufacturers to continue the development and deployment of these technologies. Our current assessment is that these ICE vehicle technologies continue to be feasible in the MY 2027 and later timeframe.

As a new vehicle is being designed and developed, we considered that manufacturers will also need time to significantly increase HD ZEV production volumes from today's volumes. In particular, manufacturers will need to build new or modify existing manufacturing production lines to assemble the new products that include ZEV powertrains. We also considered that manufacturers will require time to source new components, such as heavy-duty battery packs, motors, fuel cell stacks, and other ZEV components, including the sourcing of the critical materials, as discussed in Section II.D.2.ii. As described in Section II.D.5, we anticipate that manufacturers will not develop vehicles to cover all types of HD vehicles at once but will focus on those with the most favorable business case first, increase the adoption of those vehicles over time, and then develop other applications. We believe our approach described in Section II.D.5 shows the adoption rates for the applications we have considered would be achievable in the MY 2027 and later timeframe. We welcome comment on the manufacturer lead time requirements for HD ZEVs.

Purchasers of BEVs will also need to consider how they will charge their vehicles. Our assessment of the availability of public charging infrastructure, EVSE technology, and costs associated with depot charging are included in Section II.E.2 of this preamble, DRIA Chapter 1 and DRIA Chapter 2. As noted in DRIA Chapter 2, we anticipate that many first-time BEV owners may opt to purchase and install EVSE at or near the time of vehicle purchase and we therefore account for these capital costs upfront. In terms of EVSE for HD BEVs, this equipment is available today for purchase. However, it takes time for individual or fleet owners to develop charging site plans for their facility, obtain permits, purchase the EVSE, and have it installed. For the depots that may be charging a greater number of vehicles or with high-power DCFC ports, an upgrade to the electricity distribution system may be required. As noted in DRIA Chapter 1, we expect significant increases in HD charging infrastructure due to a combination of public and private investments. This includes Federal funding available through the BIL and the IRA. As discussed in DRIA Chapter 1.6.2.2, states, OEMs, utilities, EVSE providers and others are also investing in and supporting the deployment of charging infrastructure. For example, Daimler Trucks North America, Volvo Trucks, Navistar, and PACCAR are a few of the HD manufacturers investing in EVSE, sometimes packaging the sale of EVSE with the vehicle. Because of these projected increases and the funding available through the BIL and IRA, and as we are proposing more stringent standards that begin in MY 2027, our assessment supports that there is sufficient time for the infrastructure, especially for depot charging, to gradually increase over the remainder of this decade to levels that support the stringency of the proposed standards for the timeframe they would apply. We request comment on time considerations for all levels of HD charging infrastructure, including Level 2 up to 350 kW DCFC systems.

Purchasers of FCEVs will need to consider how they will obtain hydrogen to refuel the vehicles. As discussed in DRIA Chapter 1.8, there are currently 54 public retail hydrogen fueling stations in the United States, primarily for light-duty vehicles in California according to DOE's Alternative Fuels Data Center. When including private and planned stations in a search, there are over 130 refueling station locations nationwide. There are also numerous nationally designated hydrogen-ready or hydrogen-pending Alternative Fueling Corridors. Corridor-ready designations have public hydrogen stations no greater than 100 miles apart and no greater than five miles off the highway. Corridor-pending designations have public hydrogen stations separated by more than 100 miles but no greater than five miles off the highway. In addition, DOE's draft Clean Hydrogen Strategy and Roadmap suggests a regional “clean hydrogen hub” approach to infrastructure. Under provisions of the BIL, DOE is investing $8 billion through 2026 to support the development of at least four hubs that can demonstrate the production, processing, delivery, storage, and end use of clean hydrogen.

DOE released a Liftoff Report on clean hydrogen to establish a common fact base moving forward for dialogue and coordinated action across the full technology value chain ( e.g., from upstream production to downstream end uses). The report considers the impact of hub funding and tax credits under BIL and IRA, including the hydrogen production tax credit (PTC). It identifies three phases of rapid market growth: near-term expansion (~2023–2026), industrial scaling (~2027–2034), and long-term growth (~2035+). The report acknowledges that there are both opportunities and challenges for sectors with few decarbonization alternatives like heavy-duty transportation end uses, including long-haul trucks. During the timeframe of this rule ( i.e., through 2032), the Liftoff Report supports a scenario where low-GHG hydrogen will be emerging for long-haul trucks. We project that hydrogen consumption from FCEVs in this proposal would be a small proportion of total low-GHG hydrogen expected to be produced through 2030 in the United States.

To meet more immediate needs, end users may expect to rely on hydrogen deliveries from central production facilities. After evaluating the existing and future hydrogen refueling infrastructure, we considered FCEVs only in the MY 2030 and later timeframe to better ensure we have provided adequate time for infrastructure development and because we expect that refueling needs can be met by MY 2030, as discussed in Section II.D.4 and in DRIA Chapter 2.1. We request comment on lead time considerations related to the development of HD hydrogen fueling infrastructure.

Giving consideration to these factors, our analysis supports that there is sufficient lead time to meet the proposed standards, which manufacturers may comply with through application of BEV technologies, FCEV technologies, or further improvements to ICE vehicles, including H2–ICE powered vehicles. However, we also considered and are requesting comment on an alternative standards reflecting a slower phase-in of HD ZEV adoption rates, and are also seeking comment on more stringent standards reflecting a more aggressive phase-in of HD ZEV adoption rates, as described in Section II.H.

Additionally, while we believe there is sufficient time for the charging and refueling infrastructure to develop for the reasons explained in this section, EPA recognizes that such infrastructure for BEVs and FCEVs is important for the success of the increasing development and adoption of these vehicle technologies. EPA carefully considered that there are significant efforts already underway to develop and expand heavy-duty electric charging and hydrogen refueling infrastructure both at the local, State and Federal government level as well as from private industry, as discussed in DRIA Chapters 1 and 2 and this section. Those are important early actions that, as we just explained, will support the increase in ZEV charging and refueling infrastructure needed for the future growth of ZEV technology of the magnitude EPA is projecting in this proposal's technology packages. EPA has heard from some representatives from the heavy-duty vehicle manufacturing industry both optimism regarding the heavy-duty industry's ability to produce ZEV technologies in future years at high volume, but also concern that a slow growth in ZEV refueling infrastructure can slow the growth of heavy-duty ZEV adoption, and that this may present challenges for vehicle manufacturers' ability to comply with future EPA GHG standards. EPA has a vested interest in monitoring industry's performance in complying with mobile source emission standards, including the highway heavy-duty industry. EPA monitors industry's performance through a range of approaches, including regular meetings with individual companies and regulatory requirements for data submission as part of the annual certification process. EPA also provides transparency to the public through actions such as publishing industry compliance reports (such as has been done during the heavy-duty GHG Phase 1 program). EPA requests comment on what, if any, additional information and data EPA should consider collecting and monitoring during the implementation of the Phase 3 standards; we also request comment on whether there are additional stakeholders EPA should work with during implementation of the Phase 3 standards and what measures EPA should include to help ensure success of the Phase 3 program, including with respect to the important issues of refueling and charging infrastructure for ZEVs.

G. EPA's Basis That the Proposed Standards Are Feasible and Appropriate Under the Clean Air Act

1. Overview

As discussed in Section II.A of this preamble, there is a critical need for further GHG reductions to address the adverse impacts of air pollution from HD vehicles on public health and welfare. With continued advances in internal combustion emissions controls and vehicle zero emission technologies coming into the mainstream as key vehicle emissions controls, EPA believes substantial further emissions reductions are feasible and appropriate under the Clean Air Act.

The Clean Air Act authorizes EPA to establish emissions standards for motor vehicles to regulate emissions of air pollutants that contribute to air pollution which, in the Administrator's judgment, may reasonably be anticipated to endanger public health or welfare. Heavy-duty vehicles are significant contributors to the U.S. GHG emissions inventories, and additional reductions in GHGs from vehicles are needed to avoid the worst consequences of climate change as discussed in Section II.A.

This proposed rule also considers the large potential impact that the Inflation Reduction Act (IRA) will have on facilitating production and adoption of ZEV technologies. The IRA provides powerful incentives in reducing the cost to manufacture and purchase ZEVs, as well as reducing the cost of charging infrastructure, that will help facilitate increased market penetration of ZEV technology in the time frame considered in this rulemaking. Thus, it is an important element of EPA's cost and feasibility assessment, and EPA has considered the impacts of the IRA in our assessment of the appropriate proposed standards.

As we did in HD GHG Phase 1 and Phase 2 rulemakings, in this Phase 3 proposal we considered the following factors: the impacts of potential standards on emissions reductions of GHG emissions; technical feasibility and technology effectiveness; the lead time necessary to implement the technologies; costs to manufacturers; costs to purchasers including operating savings; reduction of non-GHG emissions; the impacts of standards on oil conservation and energy security; impacts of standards on the truck industry; other energy impacts; as well as other relevant factors such as impacts on safety. See Section II.G.5 for further discussion of how we balanced the factors we considered for the proposed Phase 3 standards.

2. Consideration of Technological Feasibility, Compliance Costs and Lead Time

The technological readiness of the heavy-duty industry to meet the proposed standards for model years 2027–2032 and beyond is best understood in the context of over a decade of heavy-duty vehicle emissions reduction programs in which the HD industry has introduced emissions reducing technologies in a wide lineup of ever more efficient and cost-competitive vehicle applications. Electrification technologies have seen particularly rapid development over the last several years such that early HD ZEV models are in use today for some applications and and are expected to expand to many more applications, as discussed DRIA Chapters 1.5 and 2, and as a result the number of ZEVs projected in the proposal and across all the alternatives considered here is much higher than in any of EPA's prior rulemaking analyses.

As discussed in DRIA Chapter 1.5.5 and Section I, the ZEV technology necessary to achieve significantly more stringent standards has already been developed and deployed. Additionally, manufacturers have announced plans to rapidly increase their investments in ZEV technologies over the next decade. In addition, the IRA and the BIL provide many monetary incentives for the production and purchase of ZEVs in the heavy-duty market, as well as incentives for electric vehicle charging infrastructure. Furthermore, there have been multiple actions by states to accelerate the adoption of heavy-duty ZEVs, such as (1) a multi-state Memorandum of Understanding for the support of heavy-duty ZEV adoption; and (2) the State of California's ACT program, which has also been adopted by other states and includes a manufacturer requirement for zero-emission truck sales. Together with the range of ICE technologies that have been already demonstrated over the past decade, BEVs and FCEVs with no tailpipe emissions (and 0 g CO₂ /ton-mile certification values) are capable of supporting rates of annual stringency increases that are much greater than were typical in earlier GHG rulemakings.

In setting standards for a future model year, EPA considers the extent deployment of advanced technologies would be available and warranted in light of the benefits to public health and welfare in GHG emission reductions, and potential constraints, such as cost of compliance, lead time, raw material availability, component supplies, redesign cycles, charging and refueling infrastructure, and purchasers' willingness to purchase (including payback). The extent of these potential constraints has diminished significantly in light of increased and further projected investment by manufacturers, increased and further projected acceptance by purchasers, and significant support from Congress to address such areas as upfront purchase price, charging infrastructure, critical mineral supplies, and domestic supply chain manufacturing. In response to the increased stringency of the proposed standards, manufacturers would be expected to adopt advanced technologies, such as increased electrification, at an increasing pace across more of their vehicles. To evaluate the feasibility of BEVs and FCEVs in our technology packages that support the proposed standards, EPA developed a tool called HD TRUCS, to evaluate the design features needed to meet the energy and power demands of various HD vehicle types when using ZEV technologies. The overarching design and functionality of HD TRUCS is premised on ensuring each of the 101 ZEV types could perform the same work as a comparable ICE vehicle counterpart. Within the HD TRUCS modeling that EPA conducted to support this proposal, we have imposed constraints to reflect the rate at which a manufacturer can deploy ZEV technologies that include consideration of time necessary to ramp up battery production, including the need to increase the availability of critical raw materials and expand battery production facilities, as discussed in Section II.D.2.ii.

Constraints on the technology adoption limits in our compliance modeling as well as other aspects of our lead time assessment are described in Section II.F. Overall, given the number and breadth of current low or zero emission vehicles and the constraints we have made to limit the rate of development for new HD vehicles, our assessment shows that there is sufficient lead time for the industry to more broadly deploy existing technologies and successfully comply with the proposed standards.

Our analysis projects that for the industry overall, nearly 50 percent of new vocational vehicles and 25 to 35 percent of new tractors in MY 2032 would be ZEVs. EPA believes that this is an achievable level based on our technical assessment for this proposal that includes consideration of the feasibility and lead time required for ZEVs and appropriate consideration of the cost of compliance for manufacturers. Our assessment of the appropriateness of the level of ZEVs in our analysis is also informed by public announcements by manufacturers about their plans to transition fleets to electrified vehicles, as described in Section I.A.2 of this preamble. More detail about our technical assessment, and our assessment of the production feasibility of ZEVs is provided in Section II.D and II.E of this Preamble and Chapters 1 and 2 of the DRIA. At the same time, we note that the proposed standards are performance-based and do not mandate any specific technology for any manufacturer or any vehicles. Moreover, the overall industry does not necessarily need to reach this level of ZEVs in order to comply—this is one of many possible compliance pathways that manufacturers could choose to take under the performance-based standards. For example, manufacturers that choose to increase their sales of hybrid vehicle technologies or apply more advanced technology to non-hybrid ICE vehicles would require a smaller number of ZEVs than we have projected in our assessment to comply with the proposed standards.

In considering feasibility of the proposed standards, EPA also considers the impact of available compliance flexibilities on manufacturers' compliance options. Manufacturers widely utilize the program's established averaging, banking and trading (ABT) provisions which provide a variety of flexible paths to plan compliance. We have discussed this dynamic in past rules, and we anticipate that this same dynamic will support compliance with this rulemaking. The GHG credit program was designed to recognize that manufacturers typically have a multi-year redesign cycle and not every vehicle will be redesigned every year to add emissions-reducing technology. Moreover, when technology is added, it will generally not achieve emissions reductions corresponding exactly to a single year-over-year change in stringency of the standards. Instead, in any given model year, some vehicles will be “credit generators,” over-performing compared to their respective CO₂ emission standards in that model year, while other vehicles will be “debit generators” and under-performing against their standards. As the proposed standards reach increasingly lower numerical levels, some vehicle designs that had generated credits against their CO₂ emission standard in earlier model years may instead generate debits in later model years. In MY 2032 when the proposed standards reach the lowest level, it is possible that only BEVs, FCEVs, and H2–ICE vehicles are generating positive credits, and all ICE vehicles generate varying levels of deficits. Even in this case, the application of ICE technologies can remain an important part of a manufacturer's compliance strategy by reducing the amount of debits generated by these vehicles. A greater application of ICE technologies ( e.g., hybrids) can enable compliance with fewer ZEVs than if less ICE technology was adopted, and therefore enable the tailoring of a compliance strategy to the manufacturer's specific market and product offerings. Together, a manufacturer's mix of credit-generating and debit-generating vehicles contribute to its sales-weighted average performance, compared to its standard, for that year.

Just as the averaging approach in the HD vehicle GHG program allows manufacturers to design a compliance strategy relying on the sale of both credit-generating vehicles and debit-generating vehicles in a single year, the credit banking and trading provisions of the program allow manufacturers to design a compliance strategy relying on overcompliance and undercompliance in different years, or even by different manufacturers. Credit banking allows credits to carry-over for up to five years and allows manufacturers up to three years to address any credit deficits. Credit trading is a compliance flexibility provision that allows one vehicle manufacturer to purchase credits from another, though trading of GHG credits has not occurred with HD GHG credits.

The proposed performance-based standards with ABT provisions give manufacturers a degree of flexibility in the design of specific vehicles and their fleet offerings, while allowing industry overall to meet the standards and thus achieve the health and environmental benefits projected for this rulemaking. EPA has considered the averaging portion of the ABT program in the feasibility assessments for previous rulemakings and continues that practice here. We also continue to acknowledge that the other provisions in ABT that provide manufacturers additional flexibility also support the feasibility of the proposed standards. By averaging across vehicles in the vehicle averaging sets and by allowing for credit banking across years, manufacturers have the flexibility to adopt emissions-reducing technologies in the manner that best suits their particular market and business circumstances. EPA's annual Heavy-Duty Vehicle and Engine Greenhouse Gas Emissions Compliance Report illustrates how different manufacturers have chosen to make use of the GHG program's various credit features. It is clear that manufacturers are widely utilizing several of the credit programs available, and we expect that manufacturers will continue to take advantage of the compliance flexibilities and crediting programs to their fullest extent, thereby providing them with additional tools in finding the lowest cost compliance solutions in light of the proposed standards.

In addition to technological feasibility and lead time, EPA has considered the cost for the heavy-duty industry to comply with the proposed standards. See Section II.F.4 and Chapter 2 of the DRIA for our analysis of compliance costs for manufacturers. We estimate that the MY 2032 fleet average per-vehicle cost to manufacturers by regulatory group would range between a cost savings for LHD vocational vehicles to $2,300 for HHD vocational vehicles and between $8,000 and $11,400 per tractor. EPA notes the projected costs per vehicle for this proposal are similar to the fleet average per-vehicle costs projected for the HD GHG Phase 2 rule that we considered to be reasonable. The Phase 2 tractor standards were projected to cost between $10,200 and $13,700 per vehicle (81 FR 73621). The Phase 2 vocational vehicle standards were projected to cost between $1,486 and $5,670 per vehicle (81 FR 73718). Furthermore, the estimated MY 2032 costs to manufacturers represent less than about ten percent of the average price of a new heavy-duty tractor today (conservatively estimated at $100,000 in 2022). For this proposal, EPA finds that the expected vehicle compliance costs are reasonable in light of the emissions reductions in air pollutants and the resulting benefits for public health and welfare.

3. Consideration of Emissions of GHGs

An essential factor that EPA considered in determining the appropriate level of the proposed standards is the reductions in GHG emissions and associated public health and welfare impacts.

The proposed GHG standards would achieve significant reductions in GHG emissions. The proposed standards would achieve approximately 1.8 billion metric tons in net CO₂ cumulative emission reductions from calendar years 2027 through 2055 (see Section V of the preamble and Chapter 4 of the DRIA). As discussed in Section VI of this preamble, these GHG emission reductions would make an important contribution to efforts to limit climate change and its anticipated impacts.

The proposed CO₂ emission standards would reduce adverse impacts associated with climate change and would yield significant benefits, including those we can monetize and those we are unable to fully monetize due to data and modeling limitations. The program would result in significant social benefits including $87 billion in climate benefits (with the average SC–GHGs at a 3 percent discount rate). A more detailed description and breakdown of these benefits can be found in Section VII of the preamble and Chapter 7 of the DRIA.

As discussed in Section VII, we monetize benefits of the proposed CO₂ standards and evaluate other costs in part to better enable a comparison of costs and benefits pursuant to E.O. 12866, but we recognize that there are benefits we are unable to fully quantify. EPA's consistent practice has been to set standards to achieve improved air quality consistent with CAA section 202, and not to rely on cost-benefit calculations, with their uncertainties and limitations, in identifying the appropriate standards. Nonetheless, our conclusion that the estimated benefits considerably exceed the estimated costs of the proposed program reinforces our view that the proposed standards represent an appropriate weighing of the statutory factors and other relevant considerations.

4. Consideration of Impacts on Purchasers, Non-GHG Emissions, Energy, Safety and Other Factors

Another factor that EPA considered in determining the proposed standards is the impact of the proposed HD CO₂ standards on purchasers, consistent with the approach we used in HD GHG Phase 1 and Phase 2. In this proposal, we considered willingness to purchase (such as practicability, payback, and costs for vehicle purchasers including EVSE) in determining the appropriate level of the proposed standards. Businesses that operate HD vehicles are under competitive pressure to reduce operating costs, which should encourage purchasers to identify and rapidly adopt vehicle technologies that provide a positive total cost of ownership. Outlays for labor and fuel generally constitute the two largest shares of HD vehicle operating costs, depending on the price of fuel, distance traveled, type of HD vehicle, and commodity transported (if any), so businesses that operate HDVs face strong incentives to reduce these costs. However, as noted in DRIA Chapter 6.2, there are a number of other considerations that may impact a purchaser's willingness to adopt new technologies. Within HD TRUCS, we considered the impact on purchasers through our evaluation of payback periods. The payback period is the number of years that it would take for the annual operational savings of a ZEV to offset the incremental upfront purchase price of a BEV or FCEV (after accounting for the IRA section 13502 battery tax credit and IRA section 13403 vehicle tax credit) and charging infrastructure costs (for BEVs) when compared to purchasing a comparable ICE vehicle. The average per-vehicle costs to a purchaser by regulatory group for a MY 2032 heavy-duty vehicle, including associated EVSE and after considering the IRA battery-manufacturer and vehicle-purchaser tax credits, are projected to range between $900 and $11,000 for vocational vehicles and $14,700 and $17,300 for tractors. As noted in Section II.F.4.ii, EPA concludes that the proposed standards would be beneficial for purchasers because the lower operating costs during the operational life of the vehicle would offset the increase in vehicle technology costs. For example, purchasers of MY 2032 vocational vehicles and day cab tractors on average by regulatory group would recoup the upfront costs through operating savings within the first three years of ownership. Furthermore, the purchasers would benefit from annual operating cost savings for each year after the payback occurs. EPA finds that these average costs to purchasers are reasonable considering the operating savings which more than offsets these costs, as was also the case with the HD GHG Phase 2 rule. See 81 FR 73482.

We also considered the practicability and suitability of the proposed standards as we applied an additional constraint within HD TRUCS that limited the maximum ZEV adoption rate to 80 percent for any given vehicle type. This conservative limit was developed after consideration of the actual needs of the purchasers, as discussed in Section II.F.1.

Within our analysis, to support the practicability and suitability of the proposed standards we also considered the lead time necessary for purchasers to install depot charging and the lead time necessary for development of hydrogen infrastructure that would be required for the use of these technologies. As further explained in DRIA Chapter 1.6 and Sections II.E.2 and II.F.5, our assessment supports that depot charging can be installed in time for the purchase and use of the volume of MY 2027 BEVs we project could be used to comply with the proposed standards. With respect to hydrogen infrastructure, as further explained in DRIA Chapter 1.8 and Section II.F.5, we recognize that this may take longer to develop, and therefore included a constraint for FCEVs such that we did not propose new standards for long-haul vehicles until MY 2030, when we expect refueling needs can be met for the volume of FCEVs we project could be used to comply with the proposed standards. Furthermore, we also assessed the impact of future HD BEVs on the grid, as discussed in Section II.E.2. Our assessment is that grid reliability is not expected to be adversely affected by the modest increase in electricity demand associated with HD BEV charging and thus was not considered to be a constraining consideration.

EPA considers our analysis of the impact of the proposed CO₂ emission standards on vehicle and upstream emissions for non-GHG pollutants as supportive of the proposed standards. The proposed standards would decrease vehicle emissions of non-GHG pollutants that contribute to ambient concentrations of ozone, particulate matter (PM_2.5), NO₂, CO, and air toxics. By 2055, when considering downstream vehicle, EGU, and refinery emissions, we estimate a net decrease in emissions from all pollutants modeled ( i.e., NO_X, PM_2.5, VOC, and SO₂) (see Section V of the preamble and Chapter 4 of the DRIA for more detail).

As also explained in Section II.G.3, and as discussed in Section VII, we monetize benefits of the proposed standards and evaluate other costs in part to better enable a comparison of costs and benefits pursuant to E.O. 12866, but we recognize that there are benefits we are unable to fully quantify. EPA's consistent practice has been to set standards to achieve improved air quality consistent with CAA section 202, and not to rely on cost-benefit calculations, with their uncertainties and limitations, in identifying the appropriate standards.

EPA also evaluated the impacts of the proposed HD GHG standards on energy, in terms of oil conservation and energy security through reductions in fuel consumption. This proposal is projected to reduce U.S. oil imports 4.3 billion gallons through 2055 (see Section VI.F). We estimate the benefits due to reductions in energy security externalities caused by U.S. petroleum consumption and imports would be approximately $12 billion under the proposed program. EPA considers this proposal to be beneficial from an energy security perspective and thus this factor was considered to be a supportive and not constraining consideration.

EPA estimates that the present value of monetized net benefits to society would be approximately $320 billion through the year 2055 (annualized net benefits of $17 billion through 2055), more than 5 times the cost in vehicle technology and associated electric vehicle supply equipment (EVSE) combined. Regarding social costs, EPA estimates that the cost of vehicle technology (not including the vehicle or battery tax credits) and EVSE would be approximately $9 billion and $47 billion respectively, and that the HD industry would save approximately $250 billion in operating costs ( e.g., savings that come from less liquid fuel used, lower maintenance and repair costs for ZEV technologies as compared to ICE technologies, etc.). The program would result in significant social benefits including $87 billion in climate benefits (with the average SC–GHGs at a 3 percent discount rate). Between $15 and $29 billion of the estimated total benefits through 2055 are attributable to reduced emissions of non-GHG pollutants, primarily those that contribute to ambient concentrations of PM_2.5. Finally, the benefits due to reductions in energy security externalities caused by U.S. petroleum consumption and imports would be approximately $12 billion under the proposed program. A more detailed description and breakdown of these benefits can be found in Section VIII of the preamble and Chapter 7 of the DRIA. Our conclusion that the estimated benefits considerably exceed the estimated costs of the proposed program reinforces our view that the proposed standards represent an appropriate weighing of the statutory factors and other relevant considerations.

Section 202(a)(4)(A) of the CAA specifically prohibits the use of an emission control device, system or element of design that will cause or contribute to an unreasonable risk to public health, welfare, or safety. EPA has a history of considering the safety implications of its emission standards, including the HD Phase 1 and Phase 2 rule. We highlight the numerous industry standards and safety protocols that exist today for heavy-duty BEVs and FCEVs that provide guidance on the safe design of these vehicles in Section II.D and DRIA Chapter 1 and thus this factor was considered to be a supportive and not constraining consideration.

5. Selection of Proposed Standards Under CAA 202(a)

Under section 202(a), EPA has a statutory obligation to set standards to reduce emissions of air pollutants from classes of motor vehicles that the Administrator has found contribute to air pollution that may be expected to endanger public health and welfare. In setting such standards, the Administrator must provide adequate lead time for the development and application of technology to meet the standards, taking into consideration the cost of compliance. EPA's proposed standards properly implement this statutory provision, as discussed in this Section II.G. In setting standards for a future model year, EPA considers the extent deployment of advanced technologies would be available and warranted in light of the benefits to public health and welfare in GHG emission reductions, and potential constraints, such as cost of compliance, lead time, raw material availability, component supplies, redesign cycles, charging and refueling infrastructure, and purchasers' willingness to purchase (including payback). The extent of these potential constraints has diminished significantly in light of increased and further projected investment by manufacturers, increased and further projected acceptance by purchasers, and significant support from Congress to address such areas as upfront purchase price, charging infrastructure, critical mineral supplies, and domestic supply chain manufacturing. The proposed standards would achieve significant and important reductions in GHG emissions that endanger public health and welfare. Furthermore, as discussed throughout this preamble, the emission reduction technologies needed to meet the proposed standards have already been developed and are feasible and available for manufacturers to utilize in their fleets at reasonable cost in the timeframe of these proposed standards, even after considering key elements including battery manufacturing capacity and critical materials availability.

As discussed throughout this preamble, the emission reduction technologies needed to meet the proposed standards are feasible and available for manufacturers to utilize in HD vehicles in the timeframe of these proposed standards. The proposed emission standards are based on one potential technology path (represented in multiple technology packages for the various HD vehicle regulatory subcategories per MY) that includes adoption rates for both ICE vehicle technologies and zero-emission vehicle technologies that EPA regards as feasible and appropriate under CAA section 202(a) for the reasons given in this Section II.G, and as further discussed throughout Section II and DRIA Chapter 2. For the reasons described in that analysis, EPA believes these technologies can be developed and applied in HD vehicles and adopted at the projected rates for these proposed standards within the lead time provided, as discussed in Section II.F.6 and in DRIA Chapter 2.

EPA also gave appropriate consideration of cost of compliance in the selection of the proposed standards as described in this Section II.G, and as further discussed in Section II.F and DRIA Chapter 2. The MY 2027 through MY 2031 emission standards were developed using less aggressive application rates and, therefore, are projected to have lower technology package costs than the proposed MY 2032 standards. Additionally, as described in this Section II.G and as further discussed in Section II.F and DRIA Chapter 2, we considered impacts on vehicle purchasers and willingness to purchase (including practicability, payback, and costs to vehicle purchasers) in applying constraints in our analysis and selecting the proposed standards. For example, in MY 2032, we estimated that the incremental cost to purchase a ZEV would be recovered in the form of operational savings during the first one to three years of ownership, on average by regulatory group, for the vocational vehicles; approximately three years, on average by regulatory group, for short-haul tractors; and seven years, on average by regulatory group, for long-haul tractors, as shown in the payback analysis included in Section II.F.4. The length of ownership of new tractors varies. One study found that first ownership is customarily four to seven years for For-Hire companies and seven to 12 years for Private fleets. Another survey found that the average trade-in cycle for tractors was 8.7 years. Therefore, we find that these tractor technologies on average by regulatory group pay for themselves within the customary ownership timeframe for the initial owner. As we discussed in the HD GHG Phase 2 rulemaking, vocational vehicles generally accumulate far fewer annual miles than tractors and would lead owners of these vehicles to keep them for longer periods of time. To the extent vocational vehicle owners may be similar to owners of tractors in terms of business profiles, they are more likely to resemble private fleets or owner-operators than for-hire fleets. Regardless, the technologies would also pay for themselves on average by regulatory group within the ownership timeframe for vocational vehicles as well.

Moreover, the additional flexibilities beyond averaging already available under EPA's existing regulations, including banking and trading provisions in the ABT program—which, for example, in effect enable manufacturers to spread the compliance requirement for any particular model year across multiple model years—further support EPA's conclusion that the proposed standards provide sufficient time for the development and application of technology, giving appropriate consideration to cost.

The Administrator has significant discretion to weigh various factors under CAA section 202, and, as with the HD GHG Phase 1 and Phase 2 rules, the Administrator notes that the purpose of adopting standards under that provision of the Clean Air Act is to address air pollution that may reasonably be anticipated to endanger public health and welfare and that reducing air pollution has traditionally been the focus of such standards. Taking into consideration the importance of reducing GHG emissions and the primary purpose of CAA section 202 to reduce the threat posed to human health and the environment by air pollution, the Administrator finds it is appropriate to propose standards that, when implemented, would result in meaningful reductions of HD vehicle GHG emissions both near term and over the longer term, and to select such standards taking into consideration the enumerated statutory factors of technological feasibility and cost of compliance within the available lead time, as well as the discretionary factor of impacts on purchasers and willingness to purchase. In identifying the proposed standards, EPA's goal was to maximize emissions reductions given our assessment of technological feasibility and accounting for cost of compliance, lead time, and impacts on purchasers and willingness to purchase. The Administrator concludes that this approach is consistent with the text and purpose of CAA section 202.

There have been very significant developments in the adoption of ZEVs since EPA promulgated the HD GHG Phase 2 rule. One of the most significant developments for U.S. heavy-duty manufacturers and purchasers is the adoption of the IRA, which takes a comprehensive approach to addressing many of the potential barriers to wider adoption of heavy-duty ZEVs in the United States. As noted in Section I, the IRA provides tens of billions of dollars in tax credits and direct Federal funding to reduce the upfront cost of purchasing ZEVs, to increase the number of charging stations across the country, to reduce the cost of manufacturing batteries, and to promote domestic source of critical minerals and other important elements of the ZEV supply chain. By addressing all of these potential obstacles to wider ZEV adoption in a coordinated, well-financed, strategy, Congress significantly advanced the potential for ZEV adoption in the near term.

In developing this estimate, EPA considered a variety of constraints which have to date limited ZEV adoption and/or could limit it in the future, including: cost to manufacturers and purchasers; availability of raw materials, batteries, and other necessary supply chain elements; adequate electricity supply and distribution; and availability of hydrogen. EPA has consulted with analysts from other agencies, including the Federal Energy Regulatory Commission, DOE, DOT, and the Joint Office for Energy and Transportation, extensively reviewed published literature and other data, and, as discussed thoroughly in this preamble and the accompanying DRIA, has incorporated limitations into our modeling to address these potential constraints, as appropriate.

As discussed in Section II.G.4, there are additional considerations that support, but were not used to select, the proposed standards. These include the non-GHG emission and energy impacts, energy security, safety, and net benefits. EPA estimates that the present value of monetized net benefits to society would be approximately $320 billion through the year 2055 (annualized net benefits of $17 billion through 2055), more than five times the cost in vehicle technology and associated electric vehicle supply equipment (EVSE) combined (see preamble Section VII and Chapter 8 of the DRIA). We recognize the these estimates do not reflect unquantified benefits, and the Administrator has not relied on these estimates in identifying the appropriate standards under CAA section 202. Nonetheless, our conclusion that the estimated benefits considerably exceed the estimated costs of the proposed program reinforces our view that the proposed standards represent an appropriate weighing of the statutory factors and other relevant considerations.

In addition to our proposed standards, we also considered and are seeking comment on a range of alternatives above and below the proposed standards, as specified and discussed in Section II.H and Section IX. Our approach and goal in selecting standards were generally the same for the range of alternative standards as they were for the proposed standards, while also recognizing that there are uncertainties in our projections and aiming to identify where additional information that may become available during the course of the rulemaking may support standards within that range as feasible and reasonable. EPA anticipates that the appropriate choice of final standards within this range will reflect the Administrator's judgments about the uncertainties in EPA's analyses as well as consideration of public comment and updated information where available. We considered an alternative with a slower phase-in with less stringent CO₂ emission standards; however, we did not select this level for the proposed standards because our assessment in this proposal is that feasible and appropriate standards are available that provide for greater GHG emission reductions than would be provided by this slower phase-in alternative. We also considered a more stringent alternative with emission standards similar to those required by the CA ACT program. At this time, we consider the proposed standards as the appropriate balancing of the factors. However, if our analysis for the final rule of relevant existing information, public comments, or new information that becomes available between the proposal and the final rule supports a set of standards within the range of alternatives we are requesting comment on, we may promulgate final CO₂ emission standards different from those proposed if we determine that those emission standards are feasible and appropriate. For example, we could finalize different standards based on different ZEV adoption rates than described for the proposed standards based on different considerations within the inputs of HD TRUCS or other approaches that we have requested comment on in this proposal ( e.g. payback schedules, consideration of technology development lead time, ZEV refueling infrastructure growth, consideration of the need for and level of emissions reductions which can be achieved through the standards to protect public health, etc.).

In summary, after consideration of the very significant reductions in GHG emissions, given the technical feasibility of the proposed standards and the moderate costs per vehicle in the available lead time, and taking into account a number of other factors such as the savings to purchasers in operating costs over the lifetime of the vehicle, safety, the benefits for energy security, and the significantly greater quantified benefits compared to quantified costs, EPA believes that the proposed standards are appropriate under EPA's section 202(a) authority.

H. Potential Alternatives

EPA developed and considered an alternative level of proposed stringency for this rule which we are seeking comment on. The results of the analysis of this alternative are included in Section IX of the preamble. We also request comment, including supporting data and analysis, if there are certain market segments, such as heavy-haul vocational trucks or long-haul tractors which may require significant energy content for their intended use, that it may be appropriate to set standards less stringent than the alternative for the specific corresponding regulatory subcategories in order to provide additional lead time to develop and introduce ZEV or other low emissions technology for those specific vehicle applications. As described in more detail throughout this preamble, we also are seeking comment on setting GHG standards that would reflect values less stringent than the lower stringency alternative for certain market segments, values in between the proposed standards and the alternative standards, values in between the proposed standards and those that would reflect ZEV adoption levels used in California's ACT, values that would reflect the level of ZEV adoption in the ACT program, and values beyond those that would reflect ZEV adoption levels in ACT such as the 50- to 60-percent ZEV adoption range represented by the publicly stated goals of several major OEMs for 2030. For all of these scenarios we are requesting comment on, EPA anticipates that the same approach explained in Section II and DRIA Chapter 2 would generally be followed, including for estimating costs, though the rationale for the different ZEV adoption rates may be based on different considerations within the inputs of HD TRUCS or other approaches that we have requested comment on in this proposal ( e.g. payback schedules, consideration of technology development lead time, ZEV refueling infrastructure growth, etc.). As explained in this Section I.D of the preamble, EPA has significant discretion in choosing an appropriate balance among factors in setting standards under CAA section 202(a)(1)–(2). If our analysis for the final rule of relevant existing information, public comments, or new information that becomes available between the proposal and final rule supports a slower or a more accelerated implementation of the proposed standards, we may promulgate final CO₂ emission standards different from those proposed (within the range between the less stringent alternative and the most stringent standards we request comment on in this section) if we determine that those emission standards are feasible and appropriate.

While our assessment in this proposal is that the proposed standards provide adequate lead time, in order to ensure fulsome comment on all of dynamics involved in the market responding to the proposed standards, we also considered an alternative with less stringent standards and a more gradual phase-in. As discussed in Section II.F.6, we considered while developing the proposed standards that manufacturers would need time to ramp up ZEV production from the numbers of ZEVs produced today to the higher adoption rates we project in the proposed standards that begin between four and eight years from now. Manufacturers would need to conduct research and develop electrified configurations for a diverse set of applications. They would also need time to conduct durability assessments because downtime is very critical in the heavy-duty market. Furthermore, manufacturers would require time to make new capital investments for the manufacturing of heavy-duty battery cells and packs, motors, and other EV components, along with changing over the vehicle assembly lines to incorporate an electrified powertrain. In addition, the purchasers of HD BEVs would need time to design and install charging infrastructure at their facilities or determine their hydrogen refueling logistics for FCEVs. Therefore, we developed and considered an alternative that reflects a more gradual phase-in of ZEV adoption rates to account for this uncertainty. The ZEV adoption rates associated with level of stringency of the proposed CO₂ emission standards shown in Section II.F.4 and the alternative CO₂ emission standards shown in Section IX.A.1 are shown in Table II–34. We are not proposing this alternative set of standards because, as already described, our assessment is that feasible and appropriate standards are available that provide for greater emission reductions than provided under this alternative. We request comment on whether our assessment that there is adequate lead time provided in the proposed standards is correct or if a more gradual phase in like the one described in this alternative would be more appropriate.

In consideration of the environmental impacts of HD vehicles and the need for significant emission reductions, as well as the views expressed by stakeholders in comments on the HD2027 NPRM such as environmental justice communities, environmental nonprofit organizations, and state and local organizations for rapid and aggressive reductions in GHG emissions, we are also requesting comment on a more stringent set of GHG standards starting in MYs 2027 through 2032 than the proposed standards and requesting that commenters provide supporting information regarding whether such standards are feasible, appropriate, and consistent with our CAA section 202 authority for a national program. We specifically are seeking comment on values that would reflect the level of ZEV adoption used in California's ACT program (as shown in Table II–35), values in between the proposed standards and those that would reflect ZEV adoption levels in ACT, and values beyond those that would reflect ZEV adoption levels in ACT, such as the 50–60 percent ZEV adoption range represented by the publicly stated goals of several major OEMs for 2030. Under any of these more stringent set of standards that we are requesting comment on, we estimate that the individual per-vehicle ZEV technology and operating costs reflecting these higher level of ZEV technology adoption rates would be the same as the individual per-vehicle ZEV costs of the proposed standards, as described in DRIA Chapter 2.8.2 because the costs were calculated as the incremental cost between a ZEV and a comparable ICE vehicle. Also under a scenario with more stringent standards, the total costs across the fleet would be higher but the total emission reductions would be greater. The MYs 2027 through 2032 and beyond emission standards reflecting the ZEV adoptions levels in California's ACT that we are requesting comment on can be found in a memo to the docket.

I. Small Businesses

EPA is proposing to make no changes to ( i.e., maintain the existing) MY 2027 and later GHG vehicle emission standards for any heavy-duty manufacturers that meet the “small business” size criteria set by the Small Business Administration. In other words, these manufacturers would not be subject to the proposed revised MY 2027 and new MYs 2028 through 2032 and later HD vehicle CO₂ emission standards but would remain subject to the HD vehicle CO₂ emission standards previous set in HD GHG Phase 2. Additionally, we are proposing that qualifying small business manufacturers could continue to average within their averaging sets for each 2027 and later model year to achieve the applicable standards; however, we are proposing to restrict banking, trading, and the use of advanced technology credit multipliers for credits generated against the Phase 2 standards for qualifying manufacturers that utilize this small business interim provision.

We are also proposing that vehicle manufacturers that qualify as a small business may choose not to utilized the proposed interim provision and voluntarily certify their vehicles to the Phase 3 standards without ABT participation restrictions if they certify all their vehicle families within a given averaging set to the Phase 3 standards for the given MY. In other words, small businesses that opt into the Phase 3 program for a given MY for all their vehicle families within a given averaging set would be eligible for the full ABT program for those vehicle families for that MY, including advanced technology credit multipliers. While we are proposing not to apply the proposed new standards for vehicles produced by small businesses, we propose that some small business manufacturers would be subject to some other new requirements we are proposing in this rule related to ZEVs, such as the battery durability monitor and warranty provisions proposed in 40 CFR 1037.115(f) and described in Section III.B.

EPA may consider new GHG emission standards to apply for vehicles produced by small business vehicle manufacturers as part of a future regulatory action. At this time, we believe the proposed new standards, which were developed based on technology packages using increasing adoption of ZEVs, may create a disproportionate burden on small business vehicle manufacturers. As described in DRIA Chapter 9, we have identified a small number of manufacturers that would appear to qualify as small businesses under the heavy-duty vehicle manufacturer category. The majority of these small businesses currently only produce ZEVs, while one company currently produces ICE vehicles.

Since there would only be a small emissions benefit from applying the proposed standards to the relatively low production volume of ICE vehicles produced by small businesses, we believe that maintaining the existing HD vehicle CO₂ standards for these companies at this time would have a negligible impact on the overall GHG emission reductions that the program would otherwise achieve. We request comment on our assessment that the emission impact of this approach for small businesses would be small considering the number and type of vehicle manufacturers described in DRIA Chapter 9.

III. Compliance Provisions, Flexibilities, and Test Procedures

In this proposed rule, we are retaining the general compliance structure of existing 40 CFR part 1037 with some revisions described in this section. Vehicle manufacturers would continue to demonstrate that they meet emission standards using emission modeling and EPA's Greenhouse gas Emissions Model (GEM) and would use fuel-mapping or powertrain test information from procedures established and revised in previous rulemakings.

The existing HD GHG Phase 2 program provides flexibilities, primarily through the HD GHG ABT program, that facilitate compliance with the emission standards. In addition to the general ABT provisions, the current HD GHG Phase 2 program also includes advanced technology credit (including for BEVs and FCEVs) and innovative technology credit provisions. As described in Section II of this preamble, the proposed revisions to the existing MY 2027 Phase 2 GHG emission standards and new proposed standards for MYs 2028 through 2032 are premised on utilization of a variety of technologies, including technologies that are considered advanced technologies in the existing HD GHG Phase 2 ABT program. As also explained in Section II, we consider averaging in supporting the feasibility of the proposed Phase 3 GHG standards in this rule. Averaging and other aspects of the ABT program would also continue to help provide additional flexibility for manufacturers to make necessary technological improvements and reduce the overall cost of the program, without compromising overall environmental objectives.

We are not proposing any changes to and are not reopening the use of credits from MY 2027 and earlier in MY 2027 and later. In other words, credits earned in HD GHG Phase 2 would be allowed to carry over into Phase 3, subject to the existing credit life limitation of five years, as described in 40 CFR 1037.740(c). Similarly, we are not proposing any revisions to and are not reopening the allowance that provides manufacturers three years to resolve credit deficits, as detailed in 40 CFR 1037.745.

In Section III.A, we describe the general ABT program and how we expect manufacturers to apply ABT to meet the proposed standards. In Section III.A, we propose a revision to the definition of “U.S.-directed production volume” to clarify consideration in this rulemaking of nationwide production volumes, including those that may in the future be certified to different state emission standards. This proposed revision is intended to address a potential interaction between the existing definition of U.S.-directed production volume and the ACT regulation for HD vehicles. Section III.A.2 includes proposed updates to advanced technology credit provisions after considering comments received on the HD2027 NPRM (87 FR 17592, March 28, 2022). In Section III.A.3, we request comment on other flexibilities, including how credits could be used across averaging sets. In Section III.B, we propose durability monitoring requirements for BEVs and PHEVs, clarify existing warranty requirements for PHEVs, and propose warranty requirements for BEVs and FCEVs. Finally, in Section III.C, we propose additional clarifying and editorial amendments to the HD highway engine provisions of 40 CFR part 1036, the HD vehicle provisions of 40 CFR part 1037 and the test procedures for HD engines in 40 CFR part 1065.

A. Proposed Revisions to the ABT Program

As noted in the introduction to this section, we are generally retaining the HD GHG Phase 2 ABT program that allows for emission credits to be averaged, banked, or traded within each of the averaging sets specified in 40 CFR 1037.740(a). To generate credits, a vehicle manufacturer must reduce CO₂ emission levels below the level of the standard for one or more vehicle families. The manufacturer can use those credits to offset higher emission levels from vehicles in the same averaging set such that the averaging set meets the standards on “average”, “bank” the credits for later use, or “trade” the credits to another manufacturer. The credits are calculated based on the production volume of the vehicles in the averaging set and their respective emission levels relative to the standard. To incentivize the research and development of the new technologies, the current HD vehicle ABT program also includes credit multipliers for certain advanced technologies. In this Section III.A, we describe proposed changes to two aspects of the ABT program: the applicable production volume for use in calculating ABT credits and credit multipliers for advanced technologies. We also request comment on other potential flexibilities we could consider adopting in this rule.

1. U.S-Directed Production Volume

As described in Section II, the proposed Phase 3 GHG vehicle standards include consideration of nationwide production volumes. Correspondingly, we are proposing that the GHG ABT program for compliance with those standards would be applicable to the same production volumes considered in setting the standards. In Section II, we also request comment on how to account for ZEV adoption rates that would result from compliance with the California ACT program in setting the proposed GHG standards. The existing HD GHG Phase 2 vehicle program has certain provisions (based off the regulatory definition of “U.S.-directed production volume”) that would exclude production volumes that are certified to different state emission standards, including exclusion from participation in ABT. To address this potential interaction between the existing definition of U.S.-directed production volume and the ACT regulation for HD vehicles, we propose a revision to the definition of “U.S.-directed production volume.” The proposed revision would clarify that in this rulemaking we consider nationwide production volumes, including those that may in the future be certified to different state emission standards, within the proposed Phase 3 standards described in Section II and within the ABT GHG vehicle program.

The exclusion of engines and vehicles certified to different state standards in the existing definitions have not impacted the HD GHG program under parts 1036 and 1037 to-date because California has adopted GHG emission standards for HD engines and vehicles that align with the Federal HD GHG Phase 1 and Phase 2 standards. As discussed in Section I, the ACT regulation requires manufacturers to produce and sell increasing numbers of zero-emission medium- and heavy-duty highway vehicles. Given the distinct difference between what is required under the ACT compared to the existing Phase 2 vehicle program and the HD vehicle GHG standards proposed under this rulemaking, we are considering the impact of the ACT on the HD GHG vehicle program. To that end, we are proposing that the revision to this definition revision apply starting with MY 2024 to provide consistent treatment of any production volumes certified to ACT. We request comment on the MY 2024 start and whether other options should be considered for transitioning to this new definition.

The existing definition of “U.S.-directed production volume” for HD vehicles explicitly does not include vehicles certified to state emission standards that are different than the emission standards in 40 CFR part 1037. The term U.S.-directed production volume is key in how the existing regulations direct manufacturers to calculate credits in the HD vehicle ABT GHG program, in 40 CFR part 1037, subpart H. In the existing regulations, vehicle production volumes that are excluded from that term's definition cannot generate credits. EPA first excluded such production volumes from participation in HD ABT in a 1990 rulemaking on NO_X emissions from HD engines. In the preamble to that rulemaking, which established NO_X and PM banking and trading and expanded the averaging program for HD engines, EPA explained that HDEs certified under the California emission control program are excluded from this program. We further explained that HDEs certified under the California emission control program may not generate credits for use by Federal engines (49-state) or use credits generated by Federal engines. In addition, we explained that while fifty-state engines participating in the Federal banking, trading or averaging programs may be sold in California if their FELs are lower than the applicable emission standard, California engines may not generate credits for the Federal program.

In 2001, in a rulemaking that established criteria pollutant emission standards phasing in to MY 2010 and later for HD engines and vehicles, EPA adopted a definition for “U.S.-directed production.” The adopted definition included similar regulatory language to our existing part 1037 definition. Regarding compliance with the MY 2007–2009 emission standards phase-in requirements, which were based on percentage of production volumes meeting the MY 2010 and later standards, EPA again noted our intent to exclude production volumes certified to different state standards. We explained that we were clarifying that this phase-in excludes California complete heavy- duty vehicles, which are already required to be certified to the California emission standards. We further explained that the phase-in also excludes vehicles sold in any state that has adopted California emission standards for complete heavy-duty vehicles. We also explained that it would be inappropriate to allow manufacturers to “double-count” the vehicles by allowing them to count those vehicles both as part of their compliance with this phase-in and for compliance with California requirements. In addition, we noted that we would handle HD engines similarly if California were to adopt different emission standards than those being established by this rule.

In the HD GHG Phase 1 rule, EPA adopted the existing definitions of U.S.-directed production volume in 40 CFR 1036.801 and 1037.801, which were unchanged in HD GHG Phase 2 and currently apply for HD engines and vehicles.

We are proposing a revision to the definition of “U.S.-directed production volume” in 40 CFR 1037.801 such that it represents the total nationwide production volumes, including vehicles certified to state emission standards that are different than the emission standards of 40 CFR part 1037. As described in Section II, the proposed standards are feasible and appropriate based on nationwide adoption rates of technology packages that include adoption of ZEV technologies. Manufacturers may be motivated to produce ZEVs by this rule and in response to other initiatives and we want to support any U.S. adoption of these technologies by allowing manufacturers to account for their nationwide production volumes to comply with the proposed standards. We recognize that the existing definition of “U.S.-directed production volume” may cause challenges to manufacturer plans, including long-term compliance planning, due to the uncertainty surrounding whether additional states may adopt more stringent standards in the future.

Given that EPA granted the ACT rule waiver requested by California under CAA section 209(b) on March 30, 2023, the existing definition of U.S.-directed production volume excludes all vehicles (ICE vehicles and ZEVs) certified to meet the ACT program in California and any other states that adopt the ACT. In this scenario, the ZEV production volumes destined for California and other states would correspond to a large portion of the nationwide production on which the proposed EPA standards are based, and it would be challenging for vehicle manufacturers to comply with the proposed standards if they could not account for those ZEVs. As described in Section II, we request comment on how to account for ZEV adoption rates that would result from compliance with the California ACT program in setting the proposed GHG standards. If we were to finalize standards that account for the ACT program, we expect to similarly base the final standards on nationwide production volumes that would continue to rely on our proposal to revise the current definition of U.S.-directed production volume to include nationwide production.

We are proposing this revision consistent with our intended approach of considering such production volumes in setting the stringency of the Phase 3 standards in this rulemaking, as well as allowing inclusion of such production volumes in demonstrating compliance with the standards through participation in the HD vehicle ABT GHG program. We believe this approach would address both the potential “double counting” issue EPA previously articulated in past HD rulemakings and the potential difficulties surrounding manufacturers' long-term compliance planning (due to the uncertainty surrounding whether additional states may adopt the California ACT program in the future) we recognize in the context of this rulemaking. Our proposed revision would also align with the approach in the LD GHG program.

In addition to this proposed revision to the definition of “U.S.-directed production volume”, we are proposing additional conforming amendments throughout 40 CFR part 1037 to streamline references to the revised definition; see Section III.E.3 for further discussion on one of those proposed revisions.

2. Advanced Technology Credits for CO₂ Emissions

In HD GHG Phase 1, we provided advanced technology credits for hybrid powertrains, Rankine cycle waste heat recovery systems on engines, all-electric vehicles, and fuel cell electric vehicles to promote the implementation of advanced technologies that were not included in our technical basis of the feasibility of the Phase 1 emission standards (see 40 CFR 86.1819–14(k)(7), 1036.150(h), and 1037.150(p)). The HD GHG Phase 2 CO₂ emission standards that followed Phase 1 were premised on the use of mild hybrid powertrains in vocational vehicles and waste heat recovery systems in a subset of the engines and tractors, and we removed mild hybrid powertrains and waste heat recovery systems as options for advanced technology credits. At the time of the HD GHG Phase 2 final rule, we believed the HD GHG Phase 2 standards themselves provided sufficient incentive to develop those specific technologies. However, none of the HD GHG Phase 2 standards were based on projected utilization of the other even more-advanced Phase 1 advanced credit technologies ( e.g., plug-in hybrid electric vehicles, all-electric vehicles, and fuel cell electric vehicles). For HD GHG Phase 2, EPA promulgated advanced technology credit multipliers through MY 2027, as shown in Table III–1 (see also 40 CFR 1037.150(p)).

As stated in the HD GHG Phase 2 rulemaking, our intention with these multipliers was to create a meaningful incentive for those manufacturers considering developing and applying these qualifying advanced technologies into their vehicles. The multipliers under the existing program are consistent with values recommended by CARB in their HD GHG Phase 2 comments. CARB's values were based on a cost analysis that compared the costs of these advanced technologies to costs of other GHG-reducing technologies. CARB's cost analysis showed that multipliers in the range we ultimately promulgated as part of the HD GHG Phase 2 final rule would make these advanced technologies more competitive with the other GHG-reducing technologies and could allow manufacturers to more easily generate a viable business case to develop these advanced technologies for HD vehicles and bring them to market at a competitive price.

In establishing the multipliers in the HD GHG Phase 2 final rule, we also considered the tendency of the HD sector to lag behind the light-duty sector in the adoption of a number of advanced technologies. There are many possible reasons for this, such as:

• HD vehicles are more expensive than light-duty vehicles, which makes it a greater monetary risk for purchasers to invest in new technologies.
• These vehicles are primarily work vehicles, which makes predictable reliability and versatility important.
• Sales volumes are much lower for HD vehicles, especially for specialized vehicles.

At the time of the HD GHG Phase 2 rulemaking, we concluded that as a result of factors such as these, and the fact that adoption rates for the aforementioned advanced technologies in HD vehicles were essentially non-existent in 2016, it seemed unlikely that market adoption of these advanced technologies would grow significantly within the next decade without additional incentives.

As we stated in the HD GHG Phase 2 final rule preamble, our determination that it was appropriate to provide large multipliers for these advanced technologies, at least in the short term, was because these advanced technologies have the potential to lead to very large reductions in GHG emissions and fuel consumption, and advance technology development substantially in the long term. However, because the credit multipliers are so large, we also stated that they should not necessarily be made available indefinitely. Therefore, they were included in the HD GHG Phase 2 final rule as an interim program continuing only through MY 2027.

The HD GHG Phase 2 CO₂ emission credits for HD vehicles are calculated according to the existing regulations at 40 CFR 1037.705. For BEVs and FCEVs, the family emission level (FEL) value for CO₂ emissions is deemed to be 0 grams per ton-mile. Under those existing regulations, the CO₂ emission credits for HD BEVs built between MY 2021 and MY 2027 would be multiplied by 4.5 (or the values shown in Table III–1 for the other technologies) and, for discussion purposes, can be visualized as split into two shares. The first share of credits would come from the reduction in CO₂ emissions realized by the environment from a BEV that is not emitting from the tailpipe, represented by the first 1.0 portion of the multiplier. Therefore, each BEV or FCEV produced receives emission credits equivalent to the level of the standard, even before taking into account the effect of a multiplier. The second share of credits does not represent CO₂ emission reductions realized in the real world but rather, as just explained, was established by EPA to help incentivize a nascent market: in this example, the emission credits for BEVs built between MY 2021 and 2027 receive an advanced technology credit multiplier of 4.5, i.e., an additional 3.5 multiple of the standard.

The HD GHG Phase 2 advanced technology credit multipliers represent a tradeoff between incentivizing new advanced technologies that could have significant benefits well beyond what is required under the standards and providing credits that do not reflect real world reductions in emissions, which could allow higher emissions from credit-using engines and vehicles. At low adoption levels, we believe the balance between the benefits of encouraging additional electrification as compared to any negative emissions impacts of multipliers would be appropriate and would justify maintaining the current advanced technology multipliers. At the time we finalized the HD GHG Phase 2 program in 2016, we balanced these factors based on our estimate that there would be very little market penetration of ZEVs in the heavy-duty market in the MY 2021 to MY 2027 timeframe, during which the advanced technology credit multipliers would be in effect. Additionally, the primary technology packages in our technical basis of the feasibility of the HD GHG Phase 2 standards did not include any ZEVs.

In our assessment conducted during the development of HD GHG Phase 2, we found only one manufacturer had certified HD BEVs through MY 2016, and we projected “limited adoption of all-electric vehicles into the market” for MYs 2021 through 2027. However, as discussed in Section II, we are now in a transitional period where manufacturers are actively increasing their PHEV, BEV, and FCEV HD vehicle offerings and are being further supported through the IRA tax credits, and we expect this growth to continue through the remaining timeframe for the HD GHG Phase 2 program and into the proposed Phase 3 program timeframe.

i. Advanced Technology Credits in the HD2027 NPRM

We requested comment in the HD2027 NPRM on three approaches that would reduce the number of incentive credits produced by battery electric vehicles in the MY 2024 through MY 2027 timeframe. The three approaches considered in the HD2027 NPRM (87 FR 17605–17606) are summarized as follows:

• Approach 1: The MY 2024 through MY 2027 ZEVs certified in California to meet the ACT program would not receive the advanced technology credit multipliers that currently exist.

• Approach 2: The advanced technology credits generated by a manufacturer would be capped on an annual basis. Advanced technology credits generated for EVs on an annual basis that are under a cap would remain unchanged. Above the cap, the multipliers would effectively be a value of 1.0; in other words, after a manufacturer reaches their cap in any model year, the multipliers would no longer be available and would have no additional effect on credit calculations. This advanced technology credit cap approach would limit the credits generated by a manufacturer's use of the advanced technology credit multipliers for battery electric vehicles to the following levels of CO₂ per manufacturer per model year beginning in MY 2024 and extending through MY 2027:

○ Light Heavy-Duty Vehicle Averaging Set: 42,000 Mg CO₂.

○ Medium Heavy-Duty Vehicle Averaging Set: 75,000 Mg CO₂.

○ Heavy Heavy-Duty Vehicle Averaging Set: 325,000 Mg CO₂.

• Approach 3: Phase-out the magnitude of the credit multipliers from MY 2024 through MY 2027.

EPA received a number of comments on the HD2027 NPRM in response to our request for comment on potential approaches to modify the existing Advanced Technology Credit multipliers. The entire set of comments may be found in Section 28 of EPA's Response to Comments Document for the HD2027 final rule.

Several commenters supported Approach 1, sometimes along with Approach 3. A common theme in these comments was that the incentive provided by the credit multipliers is not warranted for ZEVs that will already be produced due to state requirements. Some commenters also stated that the credit multipliers should not apply to any state that adopts ACT and should not be limited to California. Another commenter suggested an alternate approach whereby credit multipliers would not be provided for the vehicle segments targeted in the HD2027 NPRM for early adoption, such as some vocational vehicles and short-haul tractors, but remain available for other vehicle segments.

Other commenters raised concerns with Approach 1. For example, some commenters stated that the states' adoption of the ACT rule is unpredictable and may have a negative impact on manufacturer and supplier development plans. Another commenter raised a concern that eliminating the credit multipliers for ZEVs sold in California could impact manufacturers unequally and have a greater negative impact on manufacturers with more ZEV sales in California. One commenter suggested that this approach would create a disincentive for additional states to adopt ACT. Another commenter recommended that if EPA selects this approach, then EPA should consider allowing credit multipliers for ZEVs sold in California that exceed the ACT sales requirements. Finally, another commenter raised concerns about the implementation of this approach because it is difficult for manufacturers to account for sales by state in the heavy-duty market.

No commenters expressed support for Approach 2, and some commenters raised potential concerns with this approach. For example, a commenter stated this approach creates a disincentive to produce ZEVs above the annual cap and would negatively impact manufacturers that sell a greater number of ZEVs by making a smaller percentage of their fleet eligible for the credit multipliers. One commenter questioned whether a cap approach, while an incentive to small manufacturers and low volume ZEV producers, would incentivize additional sales beyond what is required by the states that adopt ACT under CAA section 177.

Many commenters supported a phase out or elimination of the credit multipliers, similar to Approach 3. A theme among many of the commenters was to phase out the credit multiplier as soon as practicable, with some commenters suggesting the phase out begin as early as MY 2024. On the other hand, two commenters suggested an annual decrease in the value of the credit multipliers to prevent a potential pre-buy situation. Common themes expressed by the commenters supporting an elimination of phase-out of the credit multipliers included stating that the credit multipliers are no longer necessary because of state requirements and that the credit multipliers reduce the overall effectiveness of the HD GHG regulatory program. One concern raised by a commenter is that the existing credit multipliers would slow the progression of CO₂ -reducing technologies for HD vehicles that are powered by ICE. Some commenters suggested removing the credit multipliers for all of the existing technologies qualifying for advanced technology credits, including PHEVs, BEVs, and FCEVs.

Some of the commenters opposed any changes to the existing credit multipliers. Some commenters indicated that the credit multipliers are necessary to justify the research and development of these new and higher-cost technologies into new markets. They also noted that the credit multipliers provide a role in the overall suite of incentives for ZEVs and infrastructure in the HD market. Two commenters suggested extending the credit multipliers beyond MY 2027 to allow the HD ZEV market to further mature.

ii. Proposed Changes to the Advanced Technology Credit Multipliers

While we did anticipate some growth in electrification would occur due to the credit incentives in the HD GHG Phase 2 final rule when we finalized the rule, we did not expect the level of innovation since observed, the IRA or BIL incentives, or that California would adopt the ACT rule at the same time these advanced technology multipliers were in effect. Based on this new information, we believe the existing advanced technology multiplier credit levels may no longer be appropriate for maintaining the balance between encouraging manufacturers to continue to invest in new advanced technologies over the long term and potential emissions increases in the short term. We believe that, if left as is, the multiplier credits could allow for backsliding of emission reductions expected from ICE vehicles for some manufacturers in the near term ( i.e., the generation of excess credits which could delay the introduction of technology in the near or mid-term) as sales of advanced technology vehicles which can generate the incentive credit continue to increase.

After considering the comments received on the HD2027 NPRM and the proposed HD vehicle Phase 3 GHG standards and program described in Section II and this Section III, we propose to phase-out the advanced technology credit multipliers for HD plug-in hybrid and battery electric vehicles after MY 2026, one year earlier than what is currently in the regulations. We weighed several considerations in proposing this one year earlier phase-out. We do not foresee a need for any advanced technology credits for these technologies to extend past MY 2026. We recognize the need to continue to incentivize the development of BEVs in the near-term model years, prior to MY 2027. However, our analysis of the feasibility of PHEVs and BEVs described in Section II indicates there is sufficient incentive for those technologies for the model years we are proposing HD vehicle Phase 3 GHG emission standards (MYs 2027 through 2032). We note that we did not rely on credits generated from credit multipliers in developing the proposed HD vehicle Phase 3 emission standards, however this flexibility further supports the feasibility of the proposed Phase 3 emission standards.

As explained earlier in this subsection, we recognize that a portion of the credits that result from an advanced technology multiplier do not represent CO₂ emission reductions realized in the real world and thus should be carefully balanced amongst the other considerations. We considered that we are proposing to revise the existing regulatory definition of “U.S.-directed production volume,” as discussed in Section II, such that vehicle production volumes sold in California or Section 177 states that adopt ACT would be included in the ABT credit calculations and continuing to allow these multipliers could create a large bank of credits with the potential to delay the real world benefits of the proposed program. We also took into consideration that the IRA and other new incentives are available that could help reduce the role of the multipliers. Finally, we recognize that some manufacturers' long-term product plans for PHEV or BEV technologies may have extended to model years closer to MY 2027 when the HD GHG Phase 2 standards were at their most stringent levels. We are proposing a MY 2026 phase-out for PHEV and BEV credit multipliers, in part, because it is expected to have a lesser impact on current manufacturer product plans. We request comment on our proposed MY 2026 phase-out date or whether we should consider other approaches to account for ACT or incentive programs.

We propose to revise existing 40 CFR 1037.150(p) to reflect the proposed phase-out of advanced technology credit multipliers for BEVs and PHEVs and clarify the applicable standards for calculating credits. We propose parallel edits to existing 40 CFR 1037.615(a) to clarify when the advanced technology credit calculations described in that section would apply. We are not proposing any changes to the existing advanced technology multipliers for fuel cell electric vehicles, which continue to apply through MY 2027. We believe it is still appropriate to incentivize fuel cell technology, because it has been slower to develop in the HD market, as discussed in Section II.D, but request comment on this approach for FCEVs. Additionally, we are retaining and are not reopening the existing off-cycle provisions of 40 CFR 1037.610 that allow manufacturers to request approval for other “innovative” technologies not reflected in GEM.

3. Other Potential HD CO₂ Emission Credit Flexibilities

We recognize that the proposed HD GHG Phase 3 standards would require significant investments from manufacturers to reduce GHG emissions from HD vehicles. We request comment on the potential need for additional flexibilities to assist manufacturers in the implementation of Phase 3.

Specifically, we request comment on providing the flexibility for manufacturers to use advanced technology credits across averaging sets, subject to a cap. In HD GHG Phase 1, the advanced technology credits earned a multiplier of 1.5 and they could be applied to any heavy-duty engine or vehicle averaging set. To prevent market distortions, we capped the amount of advanced credits that could be brought into any service class in any model year of the Phase 1 program at 60,000 Mg. In HD GHG Phase 2, we adopted larger advanced technology multipliers, and we discontinued the allowance for advanced technology credits to be used across averaging sets. The primary reason for the averaging set restriction was to reduce the risk of market distortions if we allowed the use of the credits across averaging sets combined with the larger credit multipliers. As discussed in Section III.A.2, we are proposing to phase-out the advanced technology credit multipliers for HD plug-in hybrid and battery electric vehicles after MY 2026, one year earlier than what is currently in the regulations, and under the existing regulations the fuel cell electric vehicle advanced technology multipliers end after MY 2027.

We recognize the proposed Phase 3 standards would require the increasing use of CO₂ emission reducing technologies. During this proposed Phase 3 standards transition, we are considering whether additional flexibilities in the Phase 3 program emissions credit ABT program design may be warranted, similar to the Phase 1 provision which allowed credits generated from advanced technologies to be transferred across averaging sets. We request comment on including a similar flexibility for the Phase 3 program. For example, we may consider an interim provision that would allow vehicle CO₂ credits generated by PHEVs, BEVs, and FCEVs to be used across vehicle averaging sets or possibly across engine averaging sets as specified in 40 CFR part 1036. If we were to adopt such an allowance, we would expect this flexibility to begin with MY 2027 and end after the last year the new Phase 3 standards phase-in, which as proposed is after MY 2032. We also would expect to restrict the number of credits ( i.e., the quantity of CO₂ megagrams) that could be transferred from one averaging set to another in a given model year, considering the level of the standards and our goal to prevent market distortions, and we request comment on what an appropriate restriction should be. We also may set different credits transfer cap values per averaging set that vary across the various averaging sets. We request comment on the model years and credit volume limitations we should consider for such an allowance for PHEV, BEV, and FCEV generated CO₂ credits. We also request comment on extending this flexibility with some restrictions to the PHEV, BEV, and FCEV generated CO₂ credits from chassis-certified Class 2b and Class 3 vehicles. More specifically, we request comment on allowing PHEV, BEV, and FCEV generated CO₂ credits in the chassis-certified Class 2b and Class 3 vehicle category (under the part 86, subpart S ABT program for MYs 2027–2032) to be used in the HD Phase 3 light heavy-duty and medium heavy-duty vehicle averaging sets (under the part 1037 ABT program for MYs 2027–2032) in a single direction of movement ( i.e., not into the heavy heavy-duty averaging set, and not allowing HD Phase 3 credits from light heavy-duty and medium heavy-duty averaging sets to be transferred into the chassis-certified Class 2b and Class 3 vehicle category), and similarly request comment on what appropriate restrictions to MYs and credit volume limitations should be included if adopted.

We also request comment on considerations of a program similar to CARB's credit program included in their ACT rule. As briefly described in DRIA Chapter 1.3.3, CARB would apply vehicle class-specific “weight class modifiers” ( i.e., credit multipliers) for credits generated by ZEVs and near zero-emission vehicles to further incentivize adoption electrification of the larger vehicle classes.

B. Battery Durability Monitoring and Warranty Requirements

This section describes our proposal to adopt battery durability monitoring requirements for BEVs and PHEVs and to clarify how warranty applies for several advanced technologies. Our proposal is motivated by three factors: BEV, PHEV, and FCEV are playing an increasing role in vehicle manufacturers' compliance strategies to control GHG emissions from HD vehicles; BEV, PHEV, and FCEV durability and reliability are important to achieving the GHG emissions reductions projected by this proposed program; and that GHG emissions credit calculations are based on mileage over a vehicle's full useful life.

1. Battery and Plug-In Hybrid Electric Vehicle Durability Monitoring Requirements

EPA's HD vehicle GHG emission standards apply for the regulatory useful life of the HD vehicle, consistent with CAA section 202(a)(1) (“Such standards shall be applicable to such vehicles and engines for their useful life”). Accordingly, EPA has historically required manufacturers to demonstrate the durability of their emission control systems on vehicles, including under our CAA section 206 authority. Without durability demonstration requirements, EPA would not be able to assess whether vehicles originally manufactured in compliance with relevant emissions standards would remain compliant over the course of their useful life. Recognizing that BEVs, PHEVs, and FCEVs are playing an increasing role in manufacturers' compliance strategies, and that emission credit calculations are based on mileage over a vehicle's useful life, the same logic applies to BEV, PHEV, and FCEV durability. Under 40 CFR part 1037, subpart H, credits are calculated by determining the family emission limit (FEL) each vehicle achieves beyond the standard and multiplying that by the production volume and a useful life mileage attributed to each vehicle subfamily. Having a useful life mileage figure for each vehicle subfamily is integral to calculating the credits attributable to that vehicle, whether those credits are used for calculating compliance through averaging, or for banking or trading. Compliance with standards through averaging depends on all vehicles in the regulatory subcategory, or averaging set, achieving their certified level of emission performance throughout their useful life. As explained in Section II and this Section III, EPA also anticipates most if not all manufacturers would include the averaging of credits generated by BEVs and FCEVs as part of their compliance strategies for the proposed standards, thus this is a particular concern given that the calculation of credits for averaging (as well as banking and trading) depend on the battery and emission performance being maintained for the full useful life of the vehicle. Thus, without durability requirements applicable to such vehicles guaranteeing certain performance over the entire useful life of the vehicles, EPA is mindful that there would not be a guarantee that a manufacturer's overall compliance with emission standards would continue throughout that useful life. Similarly, EPA is concerned that we would not have assurance that the proposed standards would achieve the emission reductions projected by this proposed program. Therefore, EPA is proposing new battery durability monitoring for HD BEVs and PHEVs as a first key step towards this end, beginning with MY 2027.

As implemented by light-duty vehicle manufacturers in current BEVs and PHEVs, lithium-ion battery technology has been shown to be effective and durable for use and we expect that this will also be the case for HD BEVs and PHEVs. It is also well known that the energy capacity of a battery will naturally degrade to some degree with time and usage, resulting in a reduction in driving range as the vehicle ages. The degree of this energy capacity and range reduction effectively becomes an issue of durability if it negatively affects how the vehicle can be used, or how many miles it is likely to be driven during its useful life.

Vehicle and engine manufacturers are currently required to account for potential battery degradation in both hybrid and plug-in hybrid vehicles that could result in an increase in CO₂ emissions (see, e.g., existing 40 CFR 1037.241(c) and 1036.241(c)). In addition, engine manufacturers are required to demonstrate compliance with criteria pollutant standards using fully aged emission control components that represent expected degradation during useful life (see, e.g.,40 CFR 1036.235(a)(2) and 1036.240). EPA is applying this well-established approach to the durability of BEV and PHEV batteries by proposing to require battery durability monitoring.

The proposed requirements are similar to the battery durability monitor regulation framework developed by the United Nations Economic Commission for Europe (UN ECE) and adopted in 2022 as Global Technical Regulation (GTR) No. 22. The proposed durability monitoring regulations would require manufacturers of BEVs and PHEVs to develop and implement an on-board state-of-certified-energy (SOCE) monitor that can be read by the vehicle user. We are not proposing durability monitoring requirements for FCEVs at this time because the technology is currently still emerging in heavy-duty vehicle applications and we are still learning what the appropriate metric might be for quantifying FCEV performance.

The importance of battery durability in the context of zero-emission and hybrid vehicles, such as BEVs and PHEVs, is well documented and has been cited by several authorities in recent years. In their 2021 report, the National Academies of Science (NAS) identified battery durability as an important issue with the rise of electrification. Among the findings outlined in that report, NAS noted that: “battery capacity degradation is considered a barrier for market penetration of BEVs,” and that “[knowledge of] real-world battery lifetime could have implications on R&D priorities, warranty provision, consumer confidence and acceptance, and role of electrification in fuel economy policy.” NAS also noted that “life prediction guides battery sizing, warranty, and resale value [and repurposing and recycling]”, and discussed at length the complexities of state of health (SOH) estimation, life-cycle prediction, and testing for battery degradation.

Several rulemaking bodies have also recognized the importance of battery durability in a world with rapidly increasing numbers of zero-emission vehicles. In 2015, the United Nations Economic Commission for Europe began studying the need for a GTR governing battery durability in light-duty vehicles. In 2021, it finalized United Nations GTR No. 22, “In-Vehicle Battery Durability for Electrified Vehicles,” which provides a regulatory structure for contracting parties to set standards for battery durability in light-duty BEVs and PHEVs. In 2022, the United Nations Economic Commission for Europe began studying the need for a GTR governing battery durability in heavy-duty vehicles. EPA representatives chaired the informal working group that developed the GTR and worked closely with global regulatory agencies and industry partners to complete its development in a form that could be adopted in various regions of the world, including potentially the United States. The European Commission and other contracting parties have also recognized the importance of durability provisions and are working to adopt the GTR standards in their local regulatory structures. In addition, the California Air Resources Board, as part of the Zero-Emission Powertrains (ZEP) Certification program, has also included battery durability and warranty requirements as part of a suite of customer assurance provisions designed to ensure that zero-emission vehicles maintain similar standards for usability, useful life, and maintenance as for ICE vehicles.

EPA concurs with the emerging consensus that battery durability is an important issue. The ability of a zero-emission vehicle to achieve the expected emission reductions during its lifetime depends in part on the ability of the battery to maintain sufficient driving range, capacity, power, and general operability for a period of use comparable to that expected of a comparable ICE vehicle. Durable and reliable electrified vehicles are therefore critical to ensuring that projected emissions reductions are achieved by this proposed program.

Because vehicle manufacturers can use electrification as an emissions control technology to comply with EPA standards as well as generate credits for use in averaging, and also banking and trading, EPA believes that it is appropriate to set requirements to ensure that electrified vehicles certifying to EPA standards are durable and capable of providing the anticipated emissions reductions, including those that they are credited under our provisions. For example, in order for the environmental emission reductions that are credited to BEVs and PHEVs to be fully realized under this proposed rule's structure, it is important that their potential to achieve a similar mileage during their lifetime be comparable to that assumed for ICE vehicles in the same vehicle service class. In addition, under the EPA GHG program, BEVs and PHEVs generate credits that can be traded among manufacturers and used to offset debits generated by vehicles using other technologies that do not themselves meet the proposed standards. In either case, if credits generated by zero-emission vehicles are to offset debits created by other vehicles on an equivalent basis, it is thus important that they should be capable of achieving a similar mileage, and this depends, in part, on the life of the battery. Further, if BEVs and PHEVs were less durable than comparable ICE vehicles, this could result in increased use of ICE vehicles. In particular, and especially for vehicles with shorter driving ranges, loss of a large portion of the original driving range capability as the vehicle ages could reduce the ability for zero-emission miles to displace greater-than-zero-emission miles traveled, as well as undermine purchaser confidence in this emerging but highly effective technology.

We proposed a specific durability testing requirement in the HD2027 NPRM and received comment on that proposal, including comment stating that the requirements could result in increases in the battery capacity beyond what was needed to meet the job of the customer. Due to these concerns and because we are still evaluating the range of durability metrics that could be used for quantifying HD BEV performance, EPA is not proposing specific durability testing requirements in this rule. However, EPA is including in this proposal a requirement for a battery durability monitor that would be applicable to HD BEVs and PHEVs. The battery durability monitor proposal would require manufacturers to provide a customer-facing battery state-of-health (SOH) monitor for all heavy-duty BEVs and PHEVs. We are proposing a new 40 CFR 1037.115(f) that would require manufacturers to install a customer-accessible SOH monitor which estimates, monitors, and communicates the vehicle's state of certified energy (SOCE) as it is defined in GTR No. 22. Specifically, manufacturers would implement onboard algorithms to estimate the current state of health of the battery, in terms of the state of its usable battery energy (UBE) expressed as a percentage of the original UBE when the vehicle was new.

For HD PHEVs, we are proposing that manufacturers would use the existing powertrain test procedures defined in 40 CFR 1036.545 to determine UBE. The powertrain test procedures requires that PHEVs be tested in charge depleting and charge sustaining modes using a range of vehicle configurations. For the determination of UBE, we are proposing that the PHEV manufacturer would select the most representative vehicle configuration.

For HD BEVs, we are proposing that manufacturers develop their own test procedures for determining UBE. This is due to the range of HD BEV architectures, and the limited test facilities for conducting powertrain testing of BEVs with e-axles. With the SOCE being a relative measure of battery health and not absolute UBE, we believe that leaving the test procedure up to the manufacturer will still provide a meaningful measure of the health of the battery. We also believe that requiring the SOH to be customer-accessible will provide assurance that the SOH monitor is relatively accurate while also providing more time for EPA to work with manufacturers to develop a standardized test procedure for determining UBE for HD BEVs.

We proposed a specified test procedure to determine UBE in the HD2027 NPRM and received comment on that proposal, including comment requesting changes to the proposed test procedure, which EPA considered in developing this proposal's approach. EPA requests comment both on this rule's proposed approach and on an alternative approach of EPA defining a test procedure to determine UBE, such as the test procedure EPA proposed in the HD2027 NPRM, CARB zero-emission powertrain certification, and the test procedures being considered by the UN ECE EVE IWG. Regarding our request for comment on the HD2027 NPRM test procedure, we note that one of the main concerns with the test procedure in submitted comments on the HD2027 NPRM was that commenters stated the powertrain test cell required for powertrains with e-axles were not widely available, and we believe there has been some indication that this is changing; we request comment on this issue. Regarding our request for comment on the test procedures being considered by the UN ECE EVE IWG, we note that some of these test procedures don't rely on chassis or powertrain dynamometers, like the charge-discharge test procedure, and request comment on this issue.

Many of the organizations and authorities that have examined the issue of battery durability, including the UN Economic Commission for Europe, the European Commission, and the California Air Resources Board, have recognized that monitoring driving range as an indicator of battery durability performance (instead of or in addition to UBE) may be an attractive option because driving range is a metric that is more directly experienced and understood by the consumer. While we are not proposing to require that heavy-duty BEVs and PHEVs implement a state-of-certified-range (SOCR) monitor, we are requesting comment on whether we should require the SOCR monitor defined in GTR No. 22.

2. Battery and Fuel Cell Electric Vehicle Component Warranty

EPA is proposing new warranty requirements for BEV and FCEV batteries and associated emission-related electric powertrain components ( e.g., fuel-cell stack, electric motors, and inverters) and is proposing to clarify how existing warranty requirements apply for PHEVs. The proposed warranty requirements build on existing emissions control warranty provisions by establishing specific new requirements tailored to the emission control-related role of the high-voltage battery and fuel-cell stack in durability and performance of BEVs and FCEVs.

As described in the previous section, the National Academies of Science (NAS) in their 2021 report identified battery warranty along with battery durability as an important issue with the rise of electrification. The proposed vehicle warranty requirements for battery and other emission-related electric powertrain components of HD BEVs and FCEVs would be similar to those that EPA has the authority to require and has historically applied to emission control-related components for HD vehicles, including HD ICE vehicles, under EPA's HD vehicle regulations, and would similarly implement and be under the authority of CAA section 207. EPA believes that this practice of ensuring a minimum level of warranty protection should be extended to the high-voltage battery and other emission-related electric powertrain components of HD BEV, PHEV, and FCEV for multiple reasons. Recognizing that BEVs, PHEVs, and FCEVs are playing an increasing role in manufacturers' compliance strategies, the high-voltage battery and the powertrain components that depend on it are emission control devices critical to the operation and emission performance of HD vehicles, as they play a critical role in reducing the vehicles' emissions and allowing BEVs and FCEVs to have zero tailpipe emissions. As explained in Section II and this Section III, EPA also anticipates most if not all manufacturers would include the averaging of credits generated by BEVs and FCEVs as part of their compliance strategies for the proposed standards, thus this is a particular concern given that the calculation of credits for averaging (as well as banking and trading) depend on the battery and emission performance being maintained for the full useful life of the vehicle. Additionally, warranty provisions are a strong complement to the proposed battery durability monitoring requirements. We believe a component under warranty is more likely to be properly maintained and repaired or replaced if it fails, which could help ensure that credits granted for BEV and FCEV production volumes represent real emission reductions achieved over the life of the vehicle. Finally, we expect manufacturers provide warranties at the existing 40 CFR 1037.120 levels for the BEVs they currently produce, and the proposed requirements to certify to offering those warranty periods and document them in the owner's manual would provide additional assurance for owners that all BEVs have the same minimum warranty period.

For heavy-duty vehicles, EPA is proposing that manufacturers identify BEV and FCEV batteries and associated electric powertrain components as component(s) covered under emission-related warranty in the vehicle's application for certification. We propose those components would be covered by the existing regulations' emissions warranty periods of 5 years or 50,000 miles for Light HDV and 5 years or 100,000 miles for Medium HDV and Heavy HDV (see proposed revisions to 40 CFR 1037.120).

We are not proposing new battery warranty requirements for PHEVs as “hybrid system components” are covered under the existing regulations in 40 CFR part 1036 and 40 CFR part 1037. In the HD2027 rule, we finalized as proposed that when a manufacturer's certified configuration includes hybrid system components ( e.g., batteries, electric motors, and inverters), those components are considered emission-related components, which would be covered under the warranty requirements (see, e.g.,88 FR 4363, January 24, 2023). We are proposing revisions to 40 CFR 1036.120(c) to clarify that the warranty requirements of 40 CFR part 1036 apply to hybrid system components for any hybrid manufacturers certifying to the part 1036 engine standards. In 40 CFR 1037.120(c), we are also proposing a clarifying revision to remove the sentence stating that the emission-related warranty does not need to cover components whose failure would not increase a vehicle's emissions of any regulated pollutant while extending the existing statement that warranty covers other emission-related components in a manufacturer's application for certification to specifically include any other components whose failure would increase a vehicle's CO₂ emissions.

C. Additional Proposed Revisions to the Regulations

In this subsection, we discuss proposed revisions to 40 CFR parts 1036, 1037, 1065.

1. Updates for Cross-Sector Issues

This section includes proposed updates that would make the same or similar changes in related portions of the CFR or across multiple standard-setting parts for individual industry sectors.

i. LLC Cycle Smoothing and Accessory Load

EPA finalized a new LLC duty-cycle in the HD2027 rule that included a procedure for smoothing the nonidle nonmotoring points immediately before and after idle segments within the duty-cycle. It was brought to our attention that the smoothing procedure in 40 CFR 1036.514(c)(3) allows smoothing based on the idle accessory torque but says nothing about how to address the contribution of curb idle transmission torque (CITT), while 40 CFR 1065.610(d)(3)(v) through (viii) requires smoothing based on CITT and says nothing about how to address idle accessory torque. This could create confusion and difficulties for common cases where CITT is required in addition to the 40 CFR 1036.514 idle accessory torques. 40 CFR 1036.514(c)(3), as currently written, would only apply if the transmission was in neutral, because it only allows you to account for the accessory load and not CITT, which was not EPA's intent. To illustrate the concern, for example, a MHD engine could have an LLC idle accessory load of 23.5 foot-pounds, which is 19 percent of a typical automatic transmission CITT of 124 foot-pounds. To resolve this potential issue, we are proposing to remove the smoothing instructions in 40 CFR 1036.514 and incorporate them into 40 CFR 1065.610.

The original intent of the 40 CFR 1065.610 duty-cycle generation procedure was to avoid discontinuities in the reference torque values. It was written with the assumption that idle load in neutral was zero, meaning the vehicle or machine idle accessory load was zero. When we introduced the required LLC idle accessory load in 40 CFR 1036.514, we failed to realize that amendments would be needed to 40 CFR 1065.610(d)(3) to clarify how to handle the accessory load in the denormalization process. The engine mapping section 40 CFR 1065.510 is another area of concern as it does not address the possibility of droop in the idle governor, which would result in different idle speeds when the transmission is in drive versus neutral. This results in an additional complication as the required idle accessory torque will be different in drive versus neutral to keep the accessory power at the level specified in Table 1 to 40 CFR 1036.514(c)(4).

40 CFR 1065.610(d)(4) is a related paragraph that allows a different deviation for an optional declared minimum torque that applies to variable- and constant-speed engines and both idle and nonidle nonmotoring points in the cycle. Its scope of application is wider than 40 CFR 1065.610(d)(3). 40 CFR 1065.610(d)(4) applies to all nonidle nonmotoring points in the cycle, not just the ones immediately preceding or following an idle segment and using it instead of (d)(3) would not get the intended constant idle accessory power loads or the intended smoothing.

There is also an existing historical conflict between 40 CFR 1065.510(f)(4) and 1065.610(d)(4). 40 CFR 1065.510(f)(4) requires that manufacturers declare non-zero idle, or minimum torques, but 40 CFR 1065.610(d)(4), permissible deviations, make their use in cycle generation optional. This results in an inconsistency between the two sections as 40 CFR 1065.510(f)(4) requires these parameters to be declared, but 40 CFR 1065.610(d)(4) does not require them to be used.

Additionally, there is a historical conflict in 40 CFR 1065.610(d)(3)(v). This paragraph, as written, includes zero percent speed and, if the paragraph is executed in the order listed, it would include idle points that were changed to neutral in the previous step for neutral while stationary transmissions. This conflict would change the torque values of those idle-in-neutral points back to the warm-idle-in-drive torque and the speed would be left unaltered at the idle-in-neutral speed. This was clearly not the intent of this paragraph, yet we note that this conflict spans back all the way to when these procedures were located in 40 CFR 86.1333–90.

The smoothing of idle points also raises the need for smoothing of the few occurances of non-idle points in the duty-cycles where the vehicle may be moving, the torque converter may not be stalled, and the warm-idle-in-drive torque may not be appropriate. This would result in the smoothing of consecutive points around nonidle nonmotoring points with normalized speed at or below zero percent and reference torque from zero to the warm-idle-in-drive torque value where the reference torque is set to the warm-idle-in-drive torque value.

To address all of these concerns, we are proposing to make changes to 40 CFR 1065.510, 1065.512, and 1065.610. Note, other proposed changes to these subsections not specifically mentioned here are edits to fix citations to relocated or new paragraphs and to improve the clarity of the test procedures. The proposed changes to 40 CFR 1065.610 include basing the smoothing of points preceding an idle segment and following an idle segment on the warm-idle-in-drive torque value (sum of CITT and idle accessory torque). Exceptions to this are for manual transmissions and for the first 24 seconds of initial idle segments for automatic transmissions. Here the warm-idle-in-neutral torque value (idle accessory torque) is used. We are proposing to include manual transmissions in the required deviations for reference torque determination for variable-speed engines in 40 CFR 1065.610(d)(3) for completeness. The proposed amendments to 40 CFR 1065.610(d)(3) include the option to skip these deviations for a manual transmission where optional declared idle torque and the optional declared power are not declared (idle torque is zero). This provides labs that have not yet implemented these required deviations the option to not implement them if they only need to run tests with manual transmissions with zero idle torque. We also proposed the addition of manual transmissions to 40 CFR 1065.512(b)(2) where these required deviations in 40 CFR 1065.610 are cited.

We are also proposing changes to 40 CFR 1065.510(b) and (f) to address the effect of droop in the idle governor and how to determine idle speed when idle torque is a function of idle speed (where a component is specified as power or CITT is specified as a function of speed and the idle speeds need to be determined for each setpoint of the idle governor). We are also proposing the addition of an option to declare the warm idle speed(s) equal to the idle speed setpoint for electronically governed variable-speed engines with an isochronous low-speed governor. Recent updates to the mapping test procedure in 40 CFR 1065.510 regarding running the map at the minimum user-adjustable idle speed setpoint and using the map for any test assumed that one could declare the warm idle speed(s) equal to the idle speed setpoint for electronically governed variable-speed engines. We are proposing changes to make it clear that this option is allowed, which would help simplify the mapping process.

To resolve the conflict between 40 CFR 1065.510(f)(4) and 1065.610(d)(4), we are proposing to move the requirement to declare torques to 40 CFR 1065.510(f)(5), which would make it optional and consistent with 40 CFR 1065.610(d)(4).

To resolve the conflict in 40 CFR 1065.610(c)(3)(v), which we are proposing to reorganize as 40 CFR 1065.610(c)(3)(vii), we are proposing to restrict the applicability of the paragraph from “all points” to “all nonidle nonmotoring points.” To address the smoothing of consecutive nonidle nonmotoring points that immediately follow and precede any smoothed idle points we are proposing to change their reference torques to the warm-idle-in-drive torque value by adding a new 40 CFR 1065.610(c)(3)(xi).

We are also proposing revisions to 40 CFR 1036.514 to reorganize and clarify the process for cycle denormalization of speed and torque where accessory load is included and to add more specific transmission shift points for greater than 200 seconds idle segments for LLC engine and hybrid powertrain testing. Shifting the transmission to neutral during very long idle segments is more representative of in-use operation than leaving it in drive, so we are proposing more specific shift points instead of a range to reduce lab-to-lab variability. The proposal would require setting the reference speed and torque values to the warm-idle-in-drive values for the first three seconds and the last three seconds of the idle segment for an engine test, requiring keeping the transmission in drive for the first 3 seconds of the idle segment, shifting the transmission from drive to park or neutral immediately after the third second in the idle segment, and shifting the transmission into drive again three seconds before the end of the idle segment.

ii. Calculating Greenhouse Gas Emission Rates

We are proposing to revise 40 CFR 1036.550(b)(2) and 40 CFR 1054.501(b)(7) to clarify that when determining the test fuel's carbon mass fraction, W_C, the fuel properties that must be measured are α (hydrogen) and β (oxygen). These paragraphs, as currently written, imply that you cannot use the default fuel properties in 40 CFR 1065.655 for α, β, γ (sulfur), and δ (nitrogen). The fuel property determination in 40 CFR 1065.655(e) makes it clear that if you measure fuel properties and the default γ and δ values for your fuel type are zero in Table 2 to 40 CFR 1065.655, you do not need to measure those properties. The sulfur (γ) and nitrogen (δ) content of these highly refined gasoline and diesel fuels are not enough to affect the W_C determination and the original intent was to not require their measurement. We are proposing this change to ensure there is no confusion on the requirement. We are also proposing to update 40 CFR 1036.550(b)(2) and 40 CFR 1054.501(b)(7) so that they reference 40 CFR 1065.655(e), which includes the default fuel property table whose number had been previously changed and we did not make the corresponding update in 40 CFR 1036.550(b)(2) and 40 CFR 1054.501(b)(7).

iii. ABT Reporting

We are proposing to allow manufacturers to correct previously submitted vehicle and engine GHG ABT reports, where a mathematical or other error in the GEM-based or fleet calculations used for compliance was discovered after the 270-day final report submission deadline. In the Phase 1 program, EPA chose the 270-day deadline for submitting a final GHG ABT report to coincide with existing criteria pollutant report requirements that manufacturers follow for heavy-duty engines. The 270-day deadline was based on our interest in manufacturers maintaining good quality assurance/quality control (QA/QC) processes in generating ABT reports. We continue to believe that aligning the ABT report deadlines for criteria and GHG pollutants can provide consistency within a manufacturer's certification and compliance processes, but further consideration of the inherent differences and complexities in how credits are calculated and accounted for in the two programs led us to consider a time window beyond 270 days for allowing corrections to the GHG report. Certifying an engine or vehicle fleet with attribute-based features (Phase 1) or GEM (Phase 2) involves a greater risk of error compared to EPA's engine or vehicle test-based programs for criteria pollutants, where direct measurement of criteria pollutant emissions at time of certification is well established. Whether an indirect, physics-based model for quantifying GHG emissions such as GEM, or a unique technology-, attribute-, or engine production volume-based credit accounting system, unintentional errors, if not detected prior to submitting the final GHG ABT report and not realized until the accounting process for the following model year was initiated, could negatively affect a manufacturer's credit balance. For example, the loss of these credits could result in a manufacturer purchasing credits or making unplanned investments in additional technologies to make up for the credits lost due to the report error.

Under the proposed revisions to 40 CFR 1036.730(f) and 1037.730(f), EPA would consider requests to correct previously submitted MY 2021 or later ABT reports only when notified of the error within a time period of 24 months from the September 30 final report deadline. For requests to correct reports for MY 2020 or earlier, we are proposing an interim deadline of October 1, 2024 (see proposed new 40 CFR 1036.150(aa) and 1037.150(y)). We believe that corrections to ABT reports, where justified, will have no impact on emissions compliance as the actual performance of a manufacturer's fleet was better than what was reported in error, and correcting the report simply adjusts the credit balance for the model year in question to the appropriate value, such that those credits can then be used in future model years.

This proposed narrowly focused allowance for correcting accounting, typographical, or GEM-based errors after a manufacturer submits the 270-day final report (see proposed revisions in 40 CFR 1037.730) is intended to address the disproportionate financial impact of an unintentional error in the complex modeling and accounting processes that manufacturers use to determine compliance and credit balances for a given model year. We are proposing a 10 percent discount to these credit corrections to the final report, which will reduce the value of the credits that are restored upon approval of the request. The 10 percent discount is intended to balance the goal of encouraging accuracy in ABT reports and use of robust QA/QC processes against the considerations for allowing manufacturers the ability to correct unforeseen errors.

iv. Migration of 40 CFR 1037.550 to 40 CFR 1036.545

We are proposing to migrate the powertrain test procedure in 40 CFR 1037.550 to 40 CFR 1036.545. Over the course of the development of this test procedure, its use expanded to include certification of engines to the criteria pollutant standards in 40 CFR part 1036 (including test procedures in 40 CFR 1036.510, 1036.512, and 1036.514) and the procedure can be used in place of the engine GHG testing procedures (40 CFR 1036.535 and 1036.540) for hybrid engines and hybrid powertrains. We are proposing to migrate the test procedure to 40 CFR 1036.545 as-is, with the following exceptions. We are proposing to add a new figure that provides an overview of the steps involved in carrying out testing under this section. We are proposing to clarify that if the test setup has multiple locations where torque is measured and speed is controlled, the manufacturer would be required to sum the measured torque and validate that the speed control meets the requirements defined in the proposed 40 CFR 1036.545(m). Positive cycle work, W_[cycle], would then be determined by integrating the sum of the power measured at each location in the proposed 40 CFR 1036.545(o)(7). We are also proposing to clarify that manufacturers may test the powertrain with a chassis dynamometer as long as they measure speed and torque at the powertrain's output shaft or wheel hubs. We are also proposing to replace all references to 40 CFR 1037.550 throughout 40 CFR part 1036 and part 1037 with new references to 40 CFR 1036.545. For test setups where speed and torque are measured at multiple locations, determine W[cycle] by integrating the sum of the power measured at each location.

v. Median Calculation for Test Fuel Properties in 40 CFR 1036.550

40 CFR 1036.550 currently requires the use of the median value of measurements from multiple labs for the emission test fuel's carbon-mass-specific net energy content and carbon mass fraction for manufacturers to determine the corrected CO₂ emission rate using equation 40 CFR 1036.550–1. The current procedure does not provide a method for determining the median value. We are proposing to add a new calculation for the median value in the statistics calculation procedures of 40 CFR 1065.602 as a new paragraph (m). We also propose to reference the new paragraph (m) in 40 CFR 1036.550(a)(1)(i) and (a)(2)(i) for carbon-mass-specific net energy content and carbon mass fraction, respectively. This proposed new calculation procedure would ensure that labs are using the same method to calculate the median value. This proposed calculation is a standard statistical method for determining median and it would require order ranking the data in increasing order from smallest value to largest.

Determining the median from data sets containing an even number of data points would require dividing the number of data points by two to determine the rank of one of the data points whose value would be used to determine the median. This data point would then be added to the next highest ranked data point and the sum would be divided by two to determine the median.

Determining the median from data sets containing an odd number of data points would be determined by adding one to the number of data points and dividing the sum by two to determine the rank of the data point whose value would be the median.

2. Updates to 40 CFR Part 1036 Heavy-Duty Highway Engine Provisions

i. Manufacturer Run Heavy Duty In-Use Testing

We are proposing a clarification to 40 CFR 1036.405(d) regarding the starting point for the 18-month window manufacturers have to complete an in-use test order. Under the current provision, the clock for the 18-month window starts after EPA has received the manufacturer's proposed plan for recruiting, screening, and selecting vehicles. There is concern that manufacturers could delay testing by unnecessarily prolonging the selection process. To alleviate this concern and keep the testing timeline within the originally intended 18-month window, we are proposing to start the clock on the 18-month window when EPA issues the order for the manufacturer to test a particular engine family.

In the HD2027 final rule, we adopted a new 40 CFR 1036.420 that includes the pass criteria for individual engines tested under the manufacturer run in-use testing program. Table 1 to 40 CFR 1036.420 contains the accuracy margins for each criteria pollutant. We are proposing to correct an inadvertent error in the final rule's amendatory text for the regulations that effects the accuracy margin for carbon monoxide (CO), which is listed in Table 1 as 0.025 g/hp-hr. The HD2027 preamble is clear that the CO accuracy margin that we finalized was intended to be 0.25 g/hp-hr and we are proposing to correct Table 1 to reflect the value in the preamble.

ii. Low Load Cycle (LLC)—Cycle Statistics

We are proposing to update 40 CFR 1036.514 to address the ability of gaseous fueled non-hybrid engines with single point fuel injection to pass cycle statistics to validate the LLC duty cycle. We referenced, in error in 40 CFR 1036.514(e), the alternate cycle statistics for gaseous fueled engines with single point fuel injection in the cycle average fuel map section in 40 CFR 1036.540(d)(3) instead of adding LLC specific cycle statistics in 40 CFR 1036.514(e). We are proposing the addition of a new Table 1 in 40 CFR 1036.514(b) to provide cycle statistics that are identical to those used by the California Air Resources Board for the LLC and to remove the reference to 40 CFR 1036.540(d)(3) in 40 CFR 1036.514(e).

iii. Low Load Cycle (LLC)—Background Sampling

We are proposing to remove the provision in 40 CFR 1036.514(d) that allows periodic background sampling into the bag over the course of multiple test intervals during the LLC because the allowance to do this is convered in 40 CFR 1065.140(b)(2). The LLC consists of a very long test interval and the intent of the provision was to address emission bag sampling systems that do not have enough dynamic range to sample background constantly over the entire duration of the LLC. 40 CFR 1065.140(b)(2) affords many flexibilities regarding the measurement of background concentrations, including sampling over multiple test intervals as long as it does not affect your ability to demonstrate compliance with the applicable emission standards.

iv. U.S.-Directed Production Volume

In the recent HD2027 rule, we amended the heavy-duty highway engine provision in 40 CFR 1036.205 and several other sections to replace “U.S.-directed production volume” with the more general term “nationwide” where we intended manufacturers report total nationwide production volumes, including production volumes that meet different state standards.

In this rule, for the reasons explained in Section III.A.1, we are proposing a broader change to the definition of “U.S.-directed production volume” for vehicles in 40 CFR 1037.801 to include production volumes for vehicles certified to different standards. We are proposing to adopt the same updated definition of “U.S.-directed production volume” in 40 CFR 1036.801 to maintain consistency between the engine and vehicle regulations' definitions, and are proposing to reinstate the term “U.S.-directed production volume” where we currently use “nationwide” in 40 CFR part 1036 to avoid having two terms with the same meaning.

Since certain existing part 1036 requirements use the existing term and definition to exclude production volumes certified to different state standards ( i.e., the NO_X ABT program for HD engines), we are proposing corresponding clarifying updates throughout 40 CFR part 1036 to ensure no change to those existing exclusions in tandem with the proposed change to the definition of the term “U.S.-directed production volume.” For example, we are also proposing to update 40 CFR 1036.705(c) to establish this paragraph as the reference for specifying the engines that are excluded from the production volume used to calculate emission credits for HD highway engines, and we propose that a new 40 CFR 1036.705(c)(4) be the location where we exclude engines certified to different state emission standards for the HD engine program. The proposed changes also include replacing several instances of “U.S.-directed production volume” with a more general “production volume” where the text clearly is connected to ABT or a more specific reference to the production volume specified in 40 CFR 1036.705(c).

v. Correction to NO_X ABT FEL Cap

We are proposing to amend 40 CFR 1036.104(c)(2) to remove paragraph (iii) which corresponds to a FEL cap of 70 mg/hp-hr for MY 2031 and later Heavy HDE that we proposed in HD2027 but did not intend to include in the final amendatory text. In the final rule for the HD2027 rule, we did not intend to include in the final amendatory text paragraph (iii) alongside the final FEL cap of 50 mg/hp-hr for MY 2031 and later which applies to all HD engine service classes including Heavy HDE in paragraph (ii) described by EPA in the preamble and supporting rule record. We are proposing to correct this error and remove paragraph (iii). This correction will not impact the stringency of the final NO_X standards because even without correction paragraph (ii) controls.

vi. Rated Power and Continuous Rated Power Coefficient of Variance in 40 CFR 1036.520

We are proposing to correct an error and include a revision to a provision we intended to include in HD2027, regarding determining power and vehicle speed values for powertrain testing. In 40 CFR 1036.520, paragraphs (h) and (i) describe how to determine rated power and continuous rated power, respectively, from the 5 Hz data in paragraph (g) averaged from the 100 Hz data collected during the test. We inadvertently left out the coefficient of variance (COV) limits of 2 percent that are needed for making the rated and continuous rated power determinations in the HD2027 final 40 CFR 1036.520(h) and (i), which were intended to be based on the COVs calculated in 40 CFR 1036.520(g) and we correctly included in the HD2027 final 40 CFR 1036.520(g). We are proposing to add the 2 percent COV limit into 40 CFR 1036.520(h) and (i). We are also proposing to correct a paragraph reference error in 40 CFR 1036.520(h). The paragraph references the data collected in paragraph (f)(2) of the section. The data collection takes place in paragraph (d)(2) of the section.

vii. Selection of Drive Axle Ratio and Tire Radius for Hybrid Engine and Hybrid Powertrain Testing

We are proposing to combine and modify the drive axle ratio and tire radius selection paragraphs in 40 CFR 1036.510(b)(2)(vii) and (viii). When testing hybrid engines and hybrid powertrains a series of vehicle parameters must be selected. The paragraphs for selecting drive axle ratio and tire radius are separate from each other, however the selection of the drive axle ratio must be done in conjunction with the tire radius as not all tire sizes are offered with a given drive axle ratio. We are proposing to combine these paragraphs into one to eliminate any possible confusion on the selection of these two parameters.

The maximum vehicle speed for SET testing of hybrid engines and powertrains is determined based on the vehicle parameters and maximum achievable speed for the configuration in 40 CFR 1036.510. This is not the case for the FTP vehicle speed which reaches a maximum of 60 miles per hour. It has been brought to our attention that there are some vehicle configurations that cannot achieve the FTP maximum speed of 60 mile per hour. To resolve this, we are proposing changes to 40 CFR 1036.510(b)(2)(vii) instructing the manufacturer to select a representative combination of drive axle ratio and tire size that ensure a vehicle speed of no less than 60 miles per hour. We are also proposing to include, as a reminder, that manufacturers may request approval for selected drive axle ratio and tire radius consistent with the provisions of 40 CFR 1036.210. We are also proposing to add a provision for manufacturers to follow the provisions of 40 CFR 1066.425(b)(5) if the hybrid powertrain or hybrid engine is used exclusively in vehicles which are not capable of reaching 60 mi/hr. This would allow the manufacturer to seek approval of an alternate test cycle and cycle-validation criteria for powertrains where the representative tire radius and axle ratio do not allow the vehicle to achieve the maximum speeds of the specified test cycle.

viii. Determining Power and Vehicle Speed Values for Powertrain Testing

We are proposing to revise 40 CFR 1036.520(d)(2) to address the possibility of clutch slip when performing the full load acceleration with maximum driver demand at 6.0 percent road grade where the initial vehicle speed is 0 mi/hr. The proposed revision would allow hybrid engines and hybrid powertrains to modify the road grade in the first 30 seconds or increase the initial speed from 0 miles per hour to 5 miles per hour to mitigate clutch slip. This road grade alteration or change in initial speed should reduce the extreme force on the clutch when accelerating at 6.0 percent grade.

We are proposing to revise 40 CFR 1036.520(d)(3) to address situations where the powertrain does not reach maximum power in the highest gear 30 seconds after the grade setpoint has reached 0.0 percent. To address this we are proposing to replace the 30 second time limit with a speed change stability limit of 0.02 m/s² which would trigger the end of the test.

ix. Request for Comment on Determining Vehicle Mass in 40 CFR 1036.510

As engines and powertrains evolve with time, changes to vehicle mass may be needed to maintain equivalent cycle work between the powertrain and engine test procedures. We request comment on updating equation 40 CFR 1036.510–1 to better reflect the relationship of vehicle mass and rated power. With the increase in rated power of heavy-duty engines, at least one manufacturer has raised to EPA that there is some concern that equation 40 CFR 1036.510–1 might need updating to better reflect the relationship of vehicle mass and rated power. If you provide comment that the equation should be updated, we request that you provide data to justify the change and show that the change would provide comparable values of cycle work and power versus time, for both the engine and powertrain versions of the duty cycles. For the engine duty cycles ( e.g., FTP and SET), the cycle work of the duty cycle is a function of the engine torque curve. For the powertrain duty cycles ( e.g., vehicle FTP and vehicle SET), the cycle work of the duty cycle is a function of the rated power of the powertrain.

x. Test Procedure for Engines Recovering Kinetic Energy for Electric Heaters

We are proposing a clarification in the existing definition for hybrid in 40 CFR 1036.801 to add a sentence stating that systems recovering kinetic energy to power an electric heater for the aftertreatment would not qualify as a hybrid engine or hybrid powertrain. Under the existing hybrid definition, systems that recover kinetic energy, such as regenerative braking, would be considered “hybrid components” and manufacturers would be required to use the powertrain test procedures to account for the electric heater or use the engine test procedures and forfeit the emission reductions from heating the aftertreatment system. With the proposed clarification to the hybrid definition, engines that use regenerative braking only to power an electric heater for aftertreatment devices would not be considered hybrid engines and, therefore, would not be required to use the powertrain test procedures; instead, those engines could use the test procedures for engines without hybrid components.

We are proposing to supplement the new definitions with direction for testing these systems in 40 CFR 1036.501. In a proposed new 40 CFR 1036.501(g), we would clarify that an electric heater for aftertreatment can be installed and functioning when creating fuel maps using 40 CFR 1036.505(b), and measuring emissions over the duty cycles specified in 40 CFR 1036.510(b), 40 CFR 1036.512(b), and 40 CFR 1036.514(b). This proposed allowance would be limited to hybrid engines where the system recovers less than 10 percent of the total positive work over each applicable transient cycle and the recovered energy is exclusively used to power an electric heater in the aftertreatment. Since the small amount of recovered energy is stored thermally and can't be used to move the vehicle, we believe that the engine test procedures are just as representative of real-world operation as the powertrain test procedures. We request comment on using a different limit than 10 percent of the total positive work over the transient cycle for this flexibility. The proposed limit of 10 percent is based on the amount of negative work versus positive work typical of conventional engines over the transient cycle. After evaluating a range of HDE, we have observed that the negative work from the transient FTP cycle during engine motoring is less than 10 percent of the positive work of the transient FTP cycle. In the same paragraph (g), we also propose that manufacturers have the option to use the powertrain test procedures for these systems, which would not have the same restrictions we are proposing for the amount of recovered energy.

xi. Updates to 40 CFR Part 1036 Definitions

We propose new and updated definitions in 40 CFR 1036.801 in support of several proposed requirements in Section II or this Section III. We propose to add a reference to two new definitions proposed in 40 CFR part 1065: Carbon-containing fuel and “neat”. The proposed definition of carbon-containing fuel will help identify the applicable test procedures for engines using fuels that do not contain carbon and would not produce CO₂. The proposed definition of “neat” would indicate that a fuel is not mixed or diluted with other fuels, which would help distinguish between fuels that contain no carbon, such as hydrogen, and fuels that that contain carbon through mixing, such as hydrogen where a diesel pilot is used for combustion. We also propose to update the definition for U.S.-directed production volume to be equivalent to nationwide production.

We propose to consolidate the definitions of hybrid, hybrid engine, and hybrid powertrain into a single definition of “hybrid” with subparagraphs distinguishing hybrid engines and powertrains. The proposed definition of hybrid retains most of the existing definition, except that we propose to remove the unnecessary “electrical” qualifier from batteries and propose to add a statement relating to recovering energy to power an electric heater in the aftertreatment (see Section III.C.2.x). The revised definitions for hybrid engines and powertrains, which are proposed as subparagraphs under “hybrid”, are more complementary of each other with less redundancy. As noted in Section III.C.2.x, we propose to update the definitions of hybrid engine and hybrid powertrain to exclude systems recovering kinetic energy for electric heaters.

We propose several editorial revisions to definitions as well. We propose to update the definition of mild hybrid such that it is relating to a hybrid engine or hybrid powertrain. We propose to revise the existing definition of small manufacturer to clarify that the employee and revenue limits include the totals from all affiliated companies and added a reference to the definition of affiliated companies in 40 CFR 1068.30.

xii. Miscellaneous Corrections and Clarifications in 40 CFR Part 1036

We are proposing to update 40 CFR 1036.150(j) to clarify that the alternate standards apply for model year 2023 and earlier loose engines, which is consistent with existing 40 CFR 86.1819–14(k)(8).

We propose to update the provision describing how to determine deterioration factors for exhaust emission standards in 40 CFR 1036.245 so it would also apply for hybrid powertrains.

xiii. Off-Cycle Test Procedure for Engines That Use Fuels Other Than Carbon-Containing Fuel

We are proposing a new paragraph 40 CFR 1036.530(j) for engines that use fuels other than carbon-containing fuel. The off-cycle test procedures in 40 CFR 1036.530 use CO₂ as a surrogate for engine power. This approach works for engines that are fueled with carbon-containing fuel, since power correlates to fuel mass rate and for carbon-containing fuels, fuel mass rate is proportional to the CO₂ mass rate of the exhaust. For fuels other than carbon-containing fuels, the fuel mass rate is not proportional to the CO₂ mass rate of the exhaust. To address this issue, we are proposing, for fuels other than carbon-containing fuels, to use engine power directly instead of relying on CO₂ mass rate to determine engine power. For field testing where engine torque and speed is not directly measured, engine broadcasted speed and torque can be used as described in 40 CFR 1065.915(d)(5).

xiv. Onboard Diagnostic and Inducement Amendments

EPA is proposing to make changes to specific aspects of paragraphs within 40 CFR 1036.110 and 1036.111 to add clarifications and correct minor errors in the OBD and inducement provisions adopted in the HD2027 final rule. Specifically, EPA is proposing the following:

• 40 CFR 1036.110(b)(6): Proposing to correct a reference to the CARB regulation to be consistent with our intent as described in the preamble of the final rule (see 88 FR 4372) to not require manufacturer self-testing and reporting requirements in 13 CCR 1971.1(l)(4).

• 40 CFR 1036.110(b)(9): Proposing to clarify that the list of data parameters readable by a generic scan tool is limited to components that are subject to existing OBD monitoring requirements ( e.g., through comprehensive component requirements in 13 CCR 1971.1(g)(3)). For example, if parking brake status was not included in an engine's OBD certificate, it would not be a required data parameter.

• 40 CFR 1036.110(b)(11): Proposing to add a reference to 13 CCR 1971.5. The final rule referenced 13 CCR 1971.1 to point to OBD testing deadlines; however, there are additional OBD testing deadlines specified in 1971.5.

• 40 CFR 1036.110(c)(1) and 40 CFR 1036.125(h)(8)(iii): Proposing to correct terminology within these provisions by referring to inducements related to “DEF level” instead of “DEF quantity,” to make the intent clearer that the system must use the level of DEF in the DEF tank for purposes of evaluating the specified inducement triggering condition. We separately refer to the quantity of DEF injection for managing the functioning of the SCR catalyst, which is unrelated to the level of DEF in the DEF tank.

• 40 CFR 1036.111: Proposing to edit for clarity, to eliminate confusion with onboard diagnostic terminology. More specifically, proposing edits to adjust inducement-related terminology to refer to “inducement triggering conditions” instead of “fault conditions.” Inducement algorithms are executed through OBD algorithms, but the inducement triggers are separate from OBD fault conditions related to the malfunction indicator light.

• 40 CFR 1036.111(a)(2): Proposing to clarify how to determine the speed category when there is less than 30 hours of accumulated data. The regulation as adopted sets the inducement schedule based on average vehicle speed over the preceding 30 hours of non-idle operation. That instruction will cover most circumstances; however, there is no specific instruction for an inducement triggering condition that occurs before the vehicle accumulates 30 hours of non-idle operation. As described in the final rule, we depend on 30 hours of non-idle operation to establish which inducement schedule is appropriate for a vehicle. We are also aware that a newly purchased vehicle would have accumulated several hours of very low-speed operation before being placed into service. We are therefore proposing to specify that engines should not be designed to assess the speed category for inducement triggering conditions until the vehicle has accumulated 30 hours of non-idle operation. We are proposing that manufacturers should program engines with a setting categorizing them as high-speed vehicles until they accumulate 30 hours of data to avoid applying an inappropriate speed schedule.

• 40 CFR 1036.111(d)(1), Table 2: Proposing to correct a typographical error for the middle set of columns that should read “Medium-speed” instead of repeating “Low-speed.” The table was correctly published in the preamble to the final rule (see 88 FR 4378). We are proposing to add an inadvertently omitted notation in the table to identify the placement of a footnote to the table.

xv. Engine Data and Information To Support Vehicle Certification

We are proposing to update 40 CFR 1036.505 to clarify that when certifying vehicles with GEM, for any fuel type not identified in Table 1 of 40 CFR 1036.550, the manufacturer would identify the fuel type as diesel fuel for engines subject to compression-ignition standards, and would identify the fuel type as gasoline for engines subject to spark-ignition standards. This proposed change to 40 CFR 1036.505, is intended to clarify what was originally intended for fuels that are not specified in Table 1 of 40 CFR 1036.550. This proposed clarification would address the potential situation where, if a fuel is input into GEM other than the fuel types identified in Table 1 of 40 CFR 1036.550, GEM will output an error.

3. Updates to 40 CFR Part 1037 Heavy-Duty Motor Vehicle Provisions

i. Standards for Qualifying Small Businesses

As noted in Section II.I, we are proposing that qualifying small manufacturers would continue to be subject to the existing MY 2027 and later standards. We are proposing to revise 40 CFR 1037.150(c) to specify the standards that apply for qualifying small business vehicle manufacturers in light of this proposal to adopt new standards for those model years. Specifically, we are renumbering the current paragraphs to apply through MY 2026 and adding new paragraphs that would apply for MY 2027 and later, including three tables that show the small business CO₂ emission standards for vocational vehicles, custom chassis vocational vehicles, and tractors. The proposed updates also include the proposed limitations on generating credits for averaging only (no banking, trading, or use of credit multipliers) unless the small manufacturer certifies to the Phase 3 standards.

ii. Vehicles With Engines Using Fuels Other Than Carbon-Containing Fuels

In the HD2027 final rule, we adopted revisions to 40 CFR 1037.150(f) to include fuel cell electric vehicles, in addition to battery electric vehicles, in the provision that deems tailpipe emissions of regulated GHG pollutants as zero and does not require CO₂ -related emission testing. As discussed in Section II.D.1, hydrogen-fueled internal combustion engines are a newer technology under development and since hydrogen has no carbon, H2 ICEs fueled with neat hydrogen would produce zero HC, CO, and CO₂ engine-out emissions. Therefore, we are proposing to include vehicles using engines fueled with neat hydrogen in 40 CFR 1037.150(f) so that their CO₂ tailpipe emissions are deemed to be zero and manufacturers are not required to perform any engine testing for CO₂ emissions. This proposed revision would not change the requirements for H2 ICE engines, including those fueled with neat hydrogen, to meet the N₂ O GHG standards or the criteria pollutant emission standards in 40 CFR part 1036. We request comment on this proposed revision to include H2 ICE in 40 CFR 1037.150(f).

Additionally, we are proposing to revise 40 CFR 1037.150(f) to replace “electric vehicles” with “battery electric vehicles”, and “hydrogen fuel cell vehicles” with “fuel cell electric vehicles”, consistent with proposed revisions to those definitions (see Section III.C.3.xiii).

iii. ABT Calculations

We are proposing clarifying revisions to the definitions of two variables of the emission credit calculation for ABT in 40 CFR 1037.705. As noted in Section II.C, we propose to update the emission standard variable (variable “Std”) to establish a common reference emission standard when calculating ABT emission credits for vocational vehicles with tailpipe CO₂ emissions deemed to be zero ( i.e., BEVs, FCEVs, and vehicles with engines fueled with pure hydrogen), which would be the CI Multi-Purpose vehicle regulatory subcategory standard for the applicable weight class. We also propose to revise the “Volume” variable to replace the term “U.S.-directed production volume” with a reference to the paragraph (c) where we are also proposing updates consistent with the proposed revision to the definition of U.S.-directed production volume. With the proposed revision to paragraph (c), we intend for 40 CFR 1037.705(c) to replace “U.S.-directed production volume” as the primary reference for the appropriate production volume to apply with respect to the ABT program and propose to generally replace throughout part 1037.

iv. U.S.-Directed Production Volume

The CAA requires that every HD engine and vehicle be covered by a certificate of conformity indicating compliance with the applicable EPA regulations. In the existing 40 CFR 1037.205, which describes requirements for the application for certification, we currently use the term U.S.-directed production volume and are now proposing that manufacturers should, instead, be reporting total nationwide production volumes that include any production volumes certified to different state standards.

In the recent HD2027 rule, we amended the corresponding heavy-duty highway engine provision in 40 CFR 1036.205 to replace “U.S.-directed production volume” with the more general term “nationwide”, noting that manufacturers were already reporting the intended total nationwide production, including production that meets different state standards. In this rule, for the reasons explained in Section III.A.1, we are proposing a broader change to the definition of “U.S.-directed production volume” and the proposed new definition would not require us to change the term used in 1037.205 to ensure manufacturers report nationwide production volumes. We are proposing revisions to the introductory paragraph of 40 CFR 1037.705(c), consistent with the proposed revisions to the corresponding HD engine provisions, to establish this paragraph as the reference for which engines are excluded from the production volume used to calculate emission credits for HD highway (see Section III.C.2.iv). Similarly, the proposed changes include replacing several instances of “U.S.-directed production volume” with a more general “production volume” where the text clearly is connected to ABT or a more specific reference to the production volume specified in 40 CFR 1037.705(c).

v. Revisions to Hybrid Powertrain Testing and Axle Efficiency Testing

We are proposing to add a new figure to 40 CFR 1037.550 to give an overview on how to carry out hybrid powertrain testing in that section. We are proposing in the axle efficiency test in 40 CFR 1037.560(e)(2) to allow the use of an alternate lower gear oil temperature range on a test point by test point basis in addition to the current alternate that requires the use of the same lower temperature range for all test points within the test matrix. This would provide more representative test results as not all test points within a matrix for a given axle test will result in gear oil temperatures within the same range.

vi. Removal of Trailer Provisions

As part of the HD GHG Phase 2 rulemaking, we set standards for certain types of trailers used in combination with tractors (see 81 FR 73639, October 25, 2016). We are proposing to remove the regulatory provisions related to trailers in 40 CFR part 1037 to carry out a decision by the U.S. Court of Appeals for the D.C. Circuit, which vacated the portions of the HD GHG Phase 2 final rule that apply to trailers. The proposed revisions include removal of specific sections and paragraphs describing trailer provisions and related references throughout the part. Additionally, we are proposing new regulatory text for an existing test procedure that currently refers to a trailer test procedure. The existing 40 CFR 1037.527 describes a procedure for manufacturers to measure aerodynamic performance of their vocational vehicles by referring to the A to B testing methodology for trailers in 40 CFR 1037.525. We are proposing to copy the regulatory text describing A to B testing from the trailer procedure into 40 CFR 1037.527 (such that it replaces the cross-referencing regulatory text).

vii. Removal of 40 CFR 1037.205(q)

We are proposing to correct an inadvertent error and remove the existing 40 CFR 1037.205(q). This paragraph contains requirements we proposed in HD2027 but did not finalize and thus did not intend to include in the final rule's amendatory instructions, regarding information for battery electric vehicles and fuel cell electric vehicles to show they meet the standards of 40 CFR part 1037.

viii. Adding Full Cylinder Deactivation to 40 CFR 1037.520(j)(1)

We are proposing to credit vehicles with engines that include full cylinder deactivation during coasting at 1.5 percent. We believe this is appropriate since the same 1.5 percent credit is currently provided for tractors and vocational vehicles with neutral coasting, and both technologies reduce CO₂ emissions by reducing the engine braking during vehicle coasting. Cylinder deactivation can reduce engine braking by closing both the intake and exhaust valves when there is no operator demand to reduce the pumping losses of the engine when motoring. Because of this, only vehicles with engines where both exhaust and intake valves are closed when the vehicle is coasting would qualify for the 1.5 percent credit.

ix. Removal of Chassis Testing Option Under 40 CFR 1037.510 and Reference Update

We are proposing to remove the chassis dynamometer testing option for testing over the duty cycles as described in 40 CFR 1037.510(a). The chassis dynamometer testing was available as an option for Phase 1 testing in 40 CFR 1037.615. We are proposing to remove it to avoid confusion as the chassis dynamometer testing option is only allowed when performing off-cycle testing following 40 CFR 1037.610 and is not allowed for creating the cycle average fuel map for input into GEM. Note that manufacturers may continue to test vehicles on a chassis dynamometer to quantify off-cycle credits under 40 CFR 1037.610.

We are also proposing to correct paragraph reference errors in 40 CFR 1037.510(a)(2)(iii) and (iv). These paragraphs reference the warmup procedure in 40 CFR 1036.520(c)(1). The warmup procedure is actually located in 40 CFR 1036.520(d).

x. Utility Factor Clarification for Testing Engines With a Hybrid Power Takeoff Shaft

We are proposing to clarify the variable description for the utility factor fraction UF_RCD in 40 CFR 1037.540(f)(3)(ii). The current description references the use of an “approved utility factor curve”. The original intent was to use the power take off utility factors that reside in Appendix E to 40 CFR part 1036 to generate a utility factor curve to determine UF_RCD. We are proposing to clarify this by replacing “approved utility factor curve” with a reference to the utility factors in Appendix E.

xi. Heavy-Duty Vehicles at or Below 14,000 Pounds GVWR

The standards proposed in this rule would apply for all heavy-duty vehicles above 14,000 pounds GVWR, except as noted in existing 40 CFR 1037.150(l). We are not proposing changes to the option for manufacturers to voluntarily certify incomplete vehicles at or below 14,000 pounds GVWR to 40 CFR part 1037 instead of certifying under 40 CFR part 86, subpart S; the proposed standards in this rule would also apply for those incomplete heavy-duty vehicles. We propose to remove 40 CFR 1037.104, which currently states that HD vehicles subject to 40 CR part 86, subpart S, are not subject to the 40 CFR 1037 standards; instead, we propose that manufacturers refer to 40 CFR 1037.5 for excluded vehicles.

In a parallel rulemaking to set new emission standards for light-duty and medium-duty vehicles under 40 CFR part 86, subpart S, we intend to propose a requirement for those vehicles at or below 14,000 pounds GVWR with a high tow rating to have installed engines that have been certified to the engine-based criteria emission standards in 40 CFR part 1036. This would apply for both complete vehicles and incomplete vehicles with Gross Combined Weight Rating above 22,000 pounds. Some of those vehicles would continue to meet GHG standards under 40 CFR 86.1819 instead of meeting the engine-based GHG standards in 40 CFR part 1036 and the vehicle-based GHG standards in 40 CFR part 1037. In particular, under the parallel proposed rule, manufacturers of incomplete vehicles at or below 14,000 pounds GVWR with a high tow rating would continue to have the option of either meeting the greenhouse gas standards under 40 CFR parts 1036 and 1037, or instead meeting the greenhouse gas standards with chassis-based measurement procedures under 40 CFR part 86, subpart S.

xii. Updates to Optional Standards for Tractors at or Above 120,000 Pounds

In HD GHG Phase 2 and in a subsequent rulemaking, we adopted optional heavy Class 8 tractor CO₂ emission standards for tractors with a GCWR above 120,000 pounds (see 40 CFR 1037.670). We did this because most manufacturers tend to rely on U.S. certificates as their evidence of conformity for products sold into Canada to reduce compliance burden. Therefore, in Phase 2 we adopted provisions that allow the manufacturers the option to meet standards that reflect the appropriate technology improvements, along with the powertrain requirements that go along with higher GCWR. While these heavy Class 8 tractor standards are optional for tractors sold into the U.S. market, Canada adopted these as mandatory requirements as part of their regulatory development and consultation process. We propose to sunset the optional standards after MY 2026.

xiii. Updates to 40 CFR Part 1037 Definitions

We are proposing several updates to the definitions in 40 CFR 1037.801. As noted in Section III.C.3.vi, we are proposing to remove the trailer provisions, which include removing the following definitions: Box van, container chassis, flatbed trailer, standard tractor, and tank trailer. We also propose to revise several definitions to remove references to trailers or trailer-specific sections, including definitions for: Class, heavy-duty vehicle, low rolling resistance tire, manufacturer, model year, Phase 1, Phase 2, preliminary approval, small manufacturer, standard payload, tire rolling resistance, trailer, and vehicle.

We also propose new and updated definitions in support of several proposed requirements in Section II or this Section III. We propose to replace the existing definition of “electric vehicle” with more specific definitions for the different vehicle technologies and energy sources that could be used to power these vehicles. Specifically, we propose new definitions for battery electric vehicle, fuel cell electric vehicle, and plug-in hybrid electric vehicle. We also propose to replace the existing definition of “hybrid engine or hybrid powertrain” with a definition of “hybrid” that refers to a revised definition in 40 CFR part 1036. We also propose to update U.S.-directed production volume to be equivalent to nationwide production.

We propose several editorial revisions to definitions as well. We propose to revise the definition of vehicle to remove the text of existing paragraph (2)(iii) and move the main phrase of that removed paragraph ( i.e., “when it is first sold as a vehicle”) to the description of “complete vehicle” to further clarify that aspect of the existing definition. We propose to revise the existing definition of small manufacturer, in addition to the proposed revisions removing reference to trailers, to clarify that the employee and revenue limits include the totals from all affiliated companies and added a reference to the definition of affiliated companies in 40 CFR 1068.30.

xiv. Miscellaneous Corrections and Clarifications in 40 CFR Part 1037

We are proposing to revise several references to 40 CFR part 86 revisions. Throughout 40 CFR part 1037, we are proposing to replace references to 40 CFR 86.1816 or 86.1819 with a more general reference to the standards of part 86, subpart S. We propose these revisions to reduce the need to update references to specific part 86 sections if new standards are added to a different section in a future rule. We are not proposing to revise any references to specific part 86 paragraphs ( e.g.,40 CFR 86.1819–14(j)).

We propose to move the duplicative statements in 40 CFR 1036.105(c) and 1037.106(c) regarding CH₄ and N₂ O standards from their current locations to 40 CFR 1037.101(a)(2)(i) where we currently describe the standards that apply in part 1037. We also propose to update 40 CFR 1037.101(a)(2)(i) to more accurately state that only CO₂ standards are described in 40 CFR 1037.105 and 1037.106, by removing reference to CH₄ and N₂ O in that sentence. We propose to update the section title for 40 CFR 1037.102 to include the term “Criteria” and the list of components ( i.e., NO_X, HC, PM, and CO) covered by the section to be consistent with the naming convention used in 40 CFR part 1036.

4. Updates to 40 CFR Part 1065 Engine Testing Procedures

i. Engine Testing and Certification With Fuels Other Than Carbon-Containing Fuels

Alternative fuels and fuels other than carbon-containing fuels are part of the fuel pathway for sustainable biofuel, e-fuel, and clean hydrogen development under the U.S. National Blueprint for Transportation Decarbonization. This blueprint anticipates a mix of battery electric, sustainable fuel, and hydrogen use to achieve a net zero carbon emissions level by 2050 for the heavy-duty sector. EPA is proposing updates to 40 CFR part 1065 to facilitate certification of engines using fuels other than carbon-containing fuels for all sectors that use engine testing to show compliance with the standards. This includes a new definition of “carbon-containing fuel” in 40 CFR 1065.1001, and the proposed addition of a new chemical balance procedure in section 40 CFR 1065.656 that would be used in place of the carbon-based chemical balance procedure in 40 CFR 1065.655 when an engine is certified for operation using fuels other than carbon-containing fuels ( e.g., hydrogen or ammonia). Since these fuels do not contain carbon, the current carbon-based chemical balance cannot be used as it is designed based on comparisons of the amount of carbon in the fuel to the amount measured post combustion in the exhaust. The chemical balance for fuels other than carbon-containing fuels looks at the amount of hydrogen in the fuel versus what is measured in the exhaust. The proposed amendments also facilitate certification of an engine on a mix of carbon-containing fuels and fuels other than carbon-containing fuels.

The proposed addition of the certification option for fuels other than carbon-containing fuels relies on inputs requiring hydrogen, ammonia, and water concentration measurement from the exhaust. Therefore, we are proposing the addition of new sections in 40 CFR part 1065 and proposing revisions to some existing sections to support the procedure in 40 CFR 1065.656. We are proposing a new 40 CFR 1065.255 to provide specifications for hydrogen measurement devices, a new 40 CFR 1065.257 to provide specifications for water measurement using a Fourier Transform Infrared (FTIR) analyzer, and a new 40 CFR 1065.277 to provide specifications for ammonia measurement devices. These additions also require a proposed new 40 CFR 1065.357 to address CO₂ interference when measuring water using an FTIR analyzer, a proposed new 40 CFR 1065.377 to address H₂ O interference and any other interference species as deemed by the instrument manufacturer or using good engineering judgment when measuring NH₃ using an FTIR or laser infrared analyzers, and the proposed addition of calibration gases for these new analyzer types to 40 CFR 1065.750. We are also proposing to add drift check requirements to 40 CFR 1065.550(b) to address drift correction of the H₂, O₂, H₂ O, and NH₃ measurements needed in the 40 CFR 1065.656 procedure. This also includes the proposed addition of drift check requirements in 40 CFR 1065.935(g)(5)(ii) for testing with PEMS. We are also proposing to add a new 40 CFR 1065.750(a)(6) to address the uncertainty of the water concentrations generated to perform the linearity verification of the water FTIR analyzer in 40 CFR 1065.257. We are proposing two options to generate a humid gas stream. The first is via a heated bubbler where dry gas is passed through the bubbler at a controlled water temperature to generate a gas with the desired water content. The second is a device that injects heated liquid water into a gas stream. We are proposing linearity verification of the humidity generator once a year to an uncertainty of ± 3 percent; however, we are not proposing to require that the calibration of the humidity generator should be NIST traceable and request comment on whether that calibration should be NIST traceable. We are proposing a requirement for a leak check after the humidity generator is assembled, as these devices are typically disassembled and stored when not in use and subsequent assembly prior to use could lead to leaks in the system. We are proposing to include calculations to determine the uncertainty of the humidity generator from measurements of dewpoint and absolute pressure. We are proposing a new definition for “carbon-containing fuel” and “lean-burn” in 40 CFR 1065.1001 to further support the addition of the certification option for engines using fuels other than carbon-containing fuels. We request comment on these proposed changes and their ability to allow certification of engines using fuels other than carbon-containing fuels.

We also request comment on whether we should add specifications for alternative test fuels, like methanol, and fuels other than carbon-containing fuels like hydrogen and ammonia, to 40 CFR part 1065, subpart H. Currently, 40 CFR 1065.701(c) allows the use of test fuels that we do not specify in 40 CFR part 1065, subpart H, with our approval. If a comment is submitted that fuel specifications should be included for these alternate test fuels, we request that the comment include specifications for the fuels the comment specifies should be included.

ii. Engine Speed Derate for Exhaust Flow Limitation

We are proposing a change to 40 CFR 1065.512(b)(1) to address the appearance of three options for generating new reference duty-cycle points for the engine to follow. The option in the existing 40 CFR 1065.512(b)(1)(i) isn't actually an option and instead gives direction on how to operate the dynamometer (torque control mode). Under our proposed revision, this sentence would be retained and moved into a new 40 CFR 1065.512(b)(1)(i) that contains some existing text split off from the current 40 CFR 1065.512(b)(1). The two remaining options in the current 40 CFR 1065.512(b)(1)(ii) and (iii) would be redesignated as 40 CFR 1065.512(b)(1)(i)(A) and (B). The proposed restructuring of 40 CFR 1065.512(b)(1) and its subparagraphs address the proposed edits described in the following paragraph.

We are proposing a change to 40 CFR 1065.512(b)(1) to address cycle validation issues where an engine with power derate intended to limit exhaust mass flowrate might include controls that reduce engine speed under cold-start conditions, resulting in reduced exhaust flow that assists other aftertreatment thermal management technologies ( e.g. electric heater). In this case, normalized speeds would generate reference speeds above this engine speed derate, which would adversely affect cycle validation. To address this, the proposed changes would provide two options. The first option is if the engine control module (ECM) broadcasts the engine derate speed that is below the denormalized speed, the broadcast speed would then be used as the reference speed for duty-cycle validation. The second option is if an ECM broadcast signal is not available, the engine would be operated over one or more practice cycles to determine the engine derate speed as a function of cycle time. Under this option, any cycle reference speed that is greater than the engine derate speed would be replaced with the engine derate speed.

iii. Accelerated Aftertreatment Aging

We recently finalized a new accelerated aftertreatment aging procedure for use in deterioration factor determination in 40 CFR 1065.1131 through 1065.1145. We request comment on the need for potential changes to the procedure based on experience that manufacturers and test labs have gained since the procedure was finalized.

iv. Nonmethane Cutter Water Interference Correction

We recently finalized options and requirements for gaseous fueled engines to allow a correction for the effect of water on the nonmethane cutter (NMC) performance, as gaseous fueled engines produce much higher water content in the exhaust than gasoline or diesel fuels, impacting the final measured emission result. The correction is done by adjusting the methane and ethane response factors used for the Total Hydrocarbon (THC) Flame Ionization Detector (FID) and the combine methane response factor and penetration fraction and combined ethane response factor and penetration fraction of the NMC FID. These response factors and penetration fractions are then used to determine NMHC and methane concentrations based on the molar water concentration in the raw or diluted exhaust. EPA is aware that test labs that have attempted to implement this correction have reported that this new option is lacking clarity with respect to the implementation of these corrections from both a procedural and emission calculation perspective. Test labs and manufacturers have also requested the option to use the water correction for all fuels, not just gaseous fuels. Test labs and manufacturers have also stated that in their view, as written, 40 CFR 1065.360(d)(12) indicates that the water correction for the methane response factor on the THC FID is required; we note that was not our intent and are thus proposing to clarify that provision.

In addition to general edits that improve the consistency of terminology and the rearrangement of some paragraphs to improve the flow of the procedure, we are proposing the following changes to 40 CFR 1065.360, 1065.365, and 1065.660 to address the concerns raised regarding implementation and use of the NMC performance corrections. In 40 CFR 1065.360 and 1065.365, we are proposing to allow the optional use of the water correction for the applicable response factors and penetration fractions for engines operated on any fuel, as the use of the correction improves the quality of the emission measurement even though the effect is less pronounced for liquid fuels. In 40 CFR 1065.360, we are proposing revisions to clarify that determination of the FID methane response factor as a function of molar water concentration is optional for all fuels. In 40 CFR 1065.365, we are proposing to remove the recommendation of a methane penetration fraction of greater than 0.85 for the NMC FID because the procedure will account for the effect of the penetration fraction regardless of the level of NMC methane penetration. We are also proposing a corresponding change in relation to another change proposed in this rule, such that the requirements for linearity performance of the humidity generator would meet the proposed uncertainty requirements in 40 CFR 1065.750(a)(6) that we are proposing to address the accuracy of humidity generators used in the calibration of the FTIRs used for water measurement. In 40 CFR 1065.660, we are proposing to modify equations 1065.660–2 and 1065.660–9 by adding the variable for the methane response factor and penetration fraction for the NMC FID back into the equations, which we previously removed for simplification because the value was set to a constant of one. This modification would have no effect on the outcome of the calculations in the event that the effect of water on the NMC performance is not being accounted for because the procedure directs that the methane response factor and penetration fraction for the NMC FID are set to one. In the event that the effect of water is being accounted for, these modified equations would make it easier to understand the requirements of the procedure.

v. ISO 8178 Exceptions in 40 CFR 1065.601

40 CFR 1065.601(c)(1) allows the use of ISO 8178 mass-based emission calculations instead of the calculations specified in 40 CFR part 1065 subpart G with two exceptions. We are proposing to update the section reference to the exception in 40 CFR 1065.601(c)(1)(i) for NO_X humidity and temperature correction from ISO 8178–1 Section 14.4 to ISO 8178–4 Section 9.1.6 to address updates made to ISO 8178 over the last 20 years that changed the location of this correction. We are also proposing to remove the exception for the use of the particulate correction factor for humidity in ISO 8178–1 Section 15.1 because this correction factor no longer exists in ISO 8178.

vi. Work System Boundary in 40 CFR 1065.210

Figure 1 in 40 CFR 1065.210 provides diagrams for the work inputs, outputs, and system boundaries for engines. We are proposing to update the diagram for liquid cooled engines in Figure 1 to paragraph (a) of 40 CFR 1065.210 to include electric heaters that use work from an external power source. We are also proposing to update 40 CFR 1065.210(a) to include an example of an engine exhaust electrical heater and direction on how to simulate the efficiency of the electrical generator, to account for the work of the electrical heater. We are proposing an efficiency of 67 percent, as this is the value used in 40 CFR 86.1869–12(b)(4)(xiii) as the baseline alternator efficiency when determining off-cycle improvements of high efficiency alternators. We request comment on the proposed value of 67 percent and request that commenters provide data if you comment that a value different than 67 percent should be used.

IV. Proposed Program Costs

In this section, we present the costs we estimate would be incurred by manufacturers and purchasers of HD vehicles impacted by the proposed standards. We also present the social costs of the proposed standards. Our analyses characterize the costs of the technology package described in section II.E of the preamble; however, as we note there, manufacturers may elect to comply using a different combination of HD vehicle and engine technologies than what we have identified. We break the costs into the following categories and subcategories:

(1) Technology Package Costs, which are the sum of direct manufacturing costs (DMC) and indirect costs. This may also be called the “package RPE.” This includes:

a. DMC, which include the costs of materials and labor to produce a product or piece of technology.

b. Indirect costs, which include research and development (R&D), warranty, corporate operations (such as salaries, pensions, health care costs, dealer support, and marketing), and profits. We estimate indirect costs using retail price equivalent (RPE) markups.

(2) Manufacturer Costs, or “manufacturer RPE,” which is the package RPE less any applicable battery tax credits. This includes:

a. Package RPE. Traditionally, the package RPE is the manufacturer RPE in EPA cost analyses.

b. Battery tax credit from IRA section 13502, “Advanced Manufacturing Production Credit,” which serve to reduce manufacturer costs. The battery tax credit is described further in Sections I and II of this preamble and Chapters 1 and 2 of the DRIA.

(3) Purchaser Costs, which are the sum of purchaser upfront vehicle costs and operating costs. This includes:

a. Manufacturer RPE. In other words, the purchaser incurs the manufacturer's package costs less any applicable battery tax credits. We refer to this as the “manufacturer RPE” in relation to the manufacturer and, at times, the “purchaser RPE” in relation to the purchaser. These two terms are equivalent in this analysis.

b. Vehicle tax credit from IRA section 13403, “Qualified Commercial Clean Vehicles,” which serve to reduce purchaser costs. The vehicle tax credit is described further in Sections I and II of this preamble and Chapters 1 and 2 of the DRIA.

c. Electric Vehicle Supply Equipment (EVSE) costs, which are the costs associated with charging equipment. Our EVSE cost estimates include indirect costs so are sometimes referred to as “EVSE RPE.”

d. Purchaser upfront vehicle costs, which include the manufacturer (also referred to as purchaser) RPE plus EVSE costs less any applicable vehicle tax credits.

e. Operating costs, which include fuel costs, electricity costs, costs for diesel exhaust fluid (DEF), and maintenance and repair costs.

(4) Social Costs, which are the sum of package RPE, EVSE RPE, and operating costs and computed on at a fleet level on an annual basis. This includes:

a. Package RPE which excludes applicable tax credits.

b. EVSE RPE.

c. Operating costs which include pre-tax fuel costs, DEF costs and maintenance and repair costs.

d. Note that fuel taxes and battery and vehicle tax credits are not included in the social costs. Taxes and tax credits are transfers as opposed to social costs.

We describe these costs and present our cost estimates in the text that follows. All costs are presented in 2021 dollars, unless noted otherwise. We used the MOVES scenarios discussed in DRIA Chapter 4, the reference and proposed cases, to compute technology costs and operating costs as well as social costs on an annual basis. Our costs and tax credits estimated on a per vehicle basis do not change between the reference and proposal cases, but the estimated vehicle populations that would be ICE vehicles, BEVs or FCEVs do change between the reference and proposal cases. We expect an increase in BEV and FCEV sales and a decrease in ICE vehicle sales in the proposal compared to the reference case and these changes in vehicle populations are the determining factor for total cost differences between the reference and proposal cases.

But first we discuss the relevant IRA tax credits and how we have considered them in our estimates. Note that the analysis that follows sometimes presents undiscounted costs and sometimes presents discounted costs. We discount future costs and benefits to properly characterize their value in the present or, as directed by the Office of Management and Budget in Advisory Circular A–4, in the year costs and benefits begin. Also in Circular A–4, OMB directs use of both 3 and 7 percent discount rates as we have done with some exceptions. We request comment, including data, on all aspects of the cost analysis. In particular, we request comment on our assessment of the IRA tax credits (see Sections IV.C.2 and IV.D.2) and operating costs (see Section IV.D.5). We also request comment, including data, on alternative approaches to estimating cost that may help inform our cost estimates for the final rulemaking.

A. IRA Tax Credits

Our cost analysis quantitatively includes consideration of two IRA tax credits, specifically the battery tax credit and the vehicle tax credit discussed in Sections I.C.2 and II.E.4 of the preamble and Chapters 1.3.2, 2.4.3, and 3.1 of the DRIA. We note that a detailed discussion of how these tax credits were considered in our consideration of costs in our technology packages may be found in Section II.E of the preamble and Chapter 2.4.3 of the DRIA. The battery tax credits are expected to reduce manufacturer costs, and in turn purchaser costs, as discussed in Section IV.C The vehicle tax credits are expected to reduce purchaser costs, as discussed in Section IV.D.2. For the cost analysis discussed in this Section IV, both the battery tax credit and vehicle tax credit were estimated for MYs 2027 through 2032 and then aggregated for each MOVES source type and regulatory class.

We request comment on our assessment of the impact of the IRA tax credits.

B. Technology Package Costs

Technology package costs include estimated technology costs associated with compliance with the proposed MY 2027 and later CO₂ emission standards (see Chapter 3 of the DRIA). Individual technology piece costs are presented in Chapter 2 and 3 of the DRIA. In general, for the first MY of each proposed emission standard, the per vehicle individual technology piece costs consist of the DMC estimated for each vehicle in the model year of the proposed standards and are used as a starting point in estimating both the technology package costs and total incremental costs. Following each year of when costs are first incurred, we have applied a learning effect to represent the cost reductions expected to occur via the “learning by doing” phenomenon. The “learning by doing” phenomenon is the process by which doing something over and over results in learning how to do that thing more efficiently which, in turn, leads to reduced resource usage, i.e., cost savings. This provides a year-over-year cost for each technology as applied to new vehicle production, which is then used to calculate total technology package costs of the proposed standards.

This technology package cost calculation approach presumes that the expected technologies would be purchased by the vehicle original equipment manufacturers (OEMs) from their suppliers. So, while the DMC estimates for the OEM in Section IV.B.1 include the indirect costs and profits incurred by the supplier, the indirect cost markups we apply in Section IV.B.2 cover the indirect costs incurred by OEMs to incorporate the new technologies into their vehicles and profit margins for the OEM typical of the heavy-duty vehicle industry. To address these OEM indirect costs, we then applied industry standard “retail price equivalent” (RPE) markup factors to the DMC to estimate indirect costs associated with the new technology. These factors represent an average price, or retail price equivalent (RPE), for products assuming all products recapture costs in the same way. We recognize that this is rarely the case since manufacturers typically price certain products higher than average and others lower than average ( i.e., they cross-subsidize). For that reason, the RPE should not be considered a price but instead should be considered more like the average cross-subsidy needed to recapture both costs and profits to support ongoing business operations. Both the learning effects applied to direct costs and the application of markup factors to estimate indirect costs are consistent with the cost estimation approaches used in EPA's past HD GHG regulatory programs. The sum of the DMC and indirect costs represents our estimate of technology “package costs” or “package RPE” per vehicle year-over-year. These per vehicle technology package costs are multiplied by estimated sales for the proposed and reference scenarios. Then the total technology package-related costs for manufacturers (total package costs or total package RPE) associated with the proposed HD vehicle CO₂ standards is the difference between the proposed and reference scenarios.

1. Direct Manufacturing Costs

To produce a unit of output, manufacturers incur direct and indirect manufacturing costs. DMC include cost of materials and labor costs. Indirect manufacturing costs are discussed in the following section, IV.A.2. The DMCs presented here include the incremental technology piece costs associated with compliance with the proposed standards as compared to the technology piece costs associated with the comparable baseline vehicle. We based the proposed standards on technology packages that include both ICE vehicle and ZEV technologies. In our analysis, the ICE vehicles include a suite of technologies that represent a vehicle that meets the existing MY 2027 Phase 2 CO₂ emission standards. Therefore, our direct manufacturing costs for the ICE vehicles are considered to be $0 because we did not add additional CO₂ -reducing technologies to the ICE vehicles beyond those in the baseline vehicle. The DMC of the BEVs or FCEVs are the technology piece costs of replacing an ICE powertrain with a BEV or FCEV powertrain for a comparable vehicle.

Throughout this discussion, when we refer to reference case costs we are referring to our cost estimate of the no-action case (impacts absent this proposed rule) which include costs associated with replacing a comparable ICE powertrain with a BEV or FCEV powertrain for ZEV adoption rates in the reference case.

We have estimated the DMC by starting with the cost of the baseline vehicle, removing the cost of the ICE powertrain, and adding the cost of a BEV or FCEV powertrain, as presented in Chapter 2 and 3 of the DRIA. In other words, net incremental costs reflect adding the total costs of components added to the powertrain to make it a BEV or FCEV, as well as removing the total costs of components removed from a comparable ICE vehicle to make it a BEV or FCEV.

Chapter 4 of the DRIA contains a description of the MOVES vehicle source types and regulatory classes. In short, we estimate costs in MOVES for vehicle source types that have both regulatory class populations and associated emission inventories. Also, throughout this section, LHD refers to light heavy-duty vehicles, MHD refers to medium heavy-duty vehicles, and HHD refers to heavy heavy-duty vehicles.

The direct costs are then adjusted to account for learning effects on BEV, FCEV and ICE vehicle powertrains on an annual basis going forward beginning with the first year of the analysis, e.g. MY 2027, for the proposed and reference scenarios. Overall, we anticipate the number of ICE powertrains (including engines and transmissions) manufactured each year will decrease as more ZEVs enter the market. This scenario may lead to an increase in component costs for ICE powertrains. On the other hand, with the inclusion of new hardware costs projected to meet the HD2027 emission standards, we would expect learning effects would reduce the incremental cost of these technologies. Chapter 3 of the DRIA includes a detailed description of the approach used to apply learning effects in this analysis and we request data and information to refine our learning effects. The resultant DMC per vehicle and how those costs decrease over time on a fleet level are presented in Section IV.E.1 of this preamble. We request comment on this approach, including methods for accounting for the projected future ICE costs.

2. Indirect Manufacturing Costs

Indirect manufacturing costs are all the costs associated with producing the unit of output that are not direct manufacturing costs—for example, they may be related to research and development (R&D), warranty, corporate operations (such as salaries, pensions, health care costs, dealer support, and marketing) and profits. An example of a R&D cost for this proposal includes the engineering resources required to develop a battery state of health monitor as described in Section III.B.1. An example of a warranty cost is the future cost covered by the manufacturer to repair defective BEV or FCEV components and meet the warranty requirements proposed in Section III.B.2. Indirect costs are generally recovered by allocating a share of the indirect costs to each unit of goods sold. Although direct costs can be allocated to each unit of goods sold, it is more challenging to account for indirect costs allocated to a unit of goods sold. To ensure that regulatory analyses capture the changes in indirect costs, markup factors (which relate total indirect costs to total direct costs) have been developed and used by EPA and other stakeholders. These factors are often referred to as retail price equivalent (RPE) multipliers and are typically applied to direct costs to estimate indirect costs. RPE multipliers provide, at an aggregate level, the proportionate share of revenues relative shares of revenue where:

Revenue = Direct Costs + Indirect Costs

Revenue/Direct Costs = 1 + Indirect Costs/Direct Costs = RPE multiplier

Resulting in:

Indirect Costs = Direct Costs × (RPE−1)

If the relationship between revenues and direct costs ( i.e., RPE multiplier) can be shown to equal an average value over time, then an estimate of direct costs can be multiplied by that average value to estimate revenues, or total costs. Further, that difference between estimated revenues, or total costs, and estimated direct costs can be taken as the indirect costs. Cost analysts and regulatory agencies have frequently used these multipliers to predict the resultant impact on costs associated with manufacturers' responses to regulatory requirements and we are using that approach in this analysis.

The proposed cost analysis estimates indirect costs by applying the RPE markup factor used in past EPA rulemakings (such as those setting GHG standards for heavy-duty vehicles and engines). The markup factors are based on company filings with the Securities and Exchange Commission for several engine and engine/vehicle manufacturers in the heavy-duty industry. The RPE factors for the HD vehicle industry as a whole are shown in Table IV–1. Also shown in Table IV–1 are the RPE factors for light-duty vehicle manufacturers.

For this analysis, EPA based indirect cost estimates for diesel and compressed natural gas (CNG) regulatory classes on the HD Truck Industry RPE value shown in Table IV–1. We are using an RPE of 1.42 to compute the indirect costs associated with the replacement of a diesel-fueled or CNG-fueled powertrain with a BEV or FCEV powertrain. For this analysis, EPA based indirect cost estimates for gasoline regulatory classes on the LD Vehicle RPE value shown in Table IV–1. We are using an RPE of 1.5 to compute the indirect costs associated with the replacement of a gasoline-fueled powertrain with a BEV or FCEV powertrain. The heavy-duty vehicle industry is becoming more vertically integrated and the direct and indirect manufacturing costs we are analyzing are those that reflect the technology packages costs OEMs would try to recover at the end purchaser, or retail, level. For that reason, we believe the two respective vehicle industry RPE values represent the most appropriate factors for this analysis. We request data to inform RPE factors for the heavy-duty industry.

3. Vehicle Technology Package RPE

Table IV–2 presents the total fleet-wide incremental technology costs estimated for the proposal relative to the reference case for the projected adoption of ZEVs in our technology package relative to the reference case on an annual basis. As previously explained in this section, the costs shown in Table IV–2 reflect marginal direct and indirect manufacturing costs of the technology package for the proposed CO₂ standards as compared to the baseline vehicle.

It is important to note that these are costs and not prices. We do not attempt to estimate how manufacturers would price their products in the technology package costs. Manufacturers may pass costs along to purchasers via price increases that reflect actual incremental costs to manufacture a ZEV when compared to a comparable ICE vehicle. However, manufacturers may also price products higher or lower than what would be necessary to account for the incremental cost difference. For instance, a manufacturer may price certain products higher than necessary and price others lower with the higher-priced products effectively subsidizing the lower-priced products. This pricing strategy may be true in any market and is not limited to the heavy-duty vehicle industry. It may be used for a variety of reasons, not solely as a response to regulatory programs.

C. Manufacturer Costs

1. Relationship to Technology Package RPE

The manufacturer costs in EPA's past HD GHG rulemaking cost analyses on an average-per-vehicle basis was only the average-per-vehicle technology package RPE described in Section II.F.5.i. However, in the cost analysis for this proposal, we are also taking into account the IRA battery tax credit in our estimates of manufacturer costs (also referred to in this section as manufacturer's RPE), as we expect the battery tax credit to reduce manufacturer costs, and in turn purchaser costs.

2. Battery Tax Credit

Table IV–3 shows the annual estimated fleet-wide battery tax credits from IRA section 13502, “Advanced Manufacturing Production Credit,” for the proposal relative to the reference case in 2021 dollars. These estimates were based on the detailed discussion in DRIA Chapter 2 of how we considered battery tax credits. Both BEVs and FCEVs include a battery in the powertrain system that may meet the IRA battery tax credit requirements if the applicable criteria are met. The battery tax credits begin to phase down starting in CY 2030 and expire after CY 2032.

3. Manufacturer RPE

The manufacturer RPE for BEVs is calculated by subtracting the battery tax credit in Table IV–3 from the corresponding technology package RPE from Table IV–2 and the resultant manufacturer RPE is shown in Table IV–4. Table IV–4 reflects learning effects on vehicle package RPE and battery tax credits from CY 2027 through 2055. The sum of the vehicle package RPE and battery tax credits for each year is shown in the manufacturer RPE column. The difference in manufacturer RPE between the proposal and reference case is presented in Table IV–4.

D. Purchaser Costs

1. Purchaser RPE

The purchaser RPE is the estimated upfront vehicle cost paid by the purchaser prior to considering the IRA vehicle tax credits. Note, as explained in Section IV.C, we do consider the IRA battery tax credit in estimating the manufacturer RPE, which in this analysis we then consider to be equivalent to the purchaser RPE because we assume full pass-through of the IRA battery tax credit from the manufacturer to the purchaser. In other words, in this analysis, the manufacturer RPE and purchaser RPE are equivalent terms. The purchaser RPEs reflect the same values as the corresponding manufacturer RPEs presented in Section IV.C.3.

2. Vehicle Purchase Tax Credit

Table IV–5 shows the annual estimated vehicle tax credit for BEV and FCEV vehicles from IRA section 13403, “Qualified Commercial Clean Vehicles,” for the proposal relative to the reference case, in 2021 dollars. These estimates were based on the detailed discussion in DRIA Chapter 2 of how we considered vehicle tax credits. The vehicle tax credits carry through to MY 2032 with the value diminishing over time as vehicle costs decrease due to the learning effect as shown in DRIA Chapter 2. Beginning in CY 2033, the tax credit program expires.

3. Electric Vehicle Supply Equipment Costs

EVSE and associated costs are described in Chapter 2.6 of the DRIA. EVSE is needed for charging of BEVs and is not needed for FCEVs. The EVSE cost estimates are assumed to include both direct and indirect costs and are sometimes referred to in this proposal as EVSE RPE costs. For these EVSE cost estimates, we assume that up to two vehicles can share one DCFC port if there is sufficient dwell time for both vehicles to meet their daily charging needs. While fleet owners may also choose to share Level 2 chargers across vehicles, we are conservatively assigning one Level 2 charger per vehicle. As discussed in the DRIA, we assume that EVSE costs are incurred by purchasers, i.e. heavy-duty vehicle purchasers/owners. Some purchasers may be eligible for a Federal tax credit for charging equipment. See DRIA Chapter 1.3.2 for a discussion of this tax credit and DRIA Chapter 2.6.5.2 for a description of how we considered it in our cost analysis. We analyzed EVSE costs in 2021 dollars on a fleet-wide basis for this analysis. The annual costs associated with EVSE in the proposal relative to the reference case are shown in Table IV–6.

We request comment on our estimated EVSE costs as well as our proposal to add EVSE costs to each vehicle's purchaser RPE costs in estimating purchaser costs.

4. Purchaser Upfront Vehicle Costs

The expected upfront incremental costs to the purchaser include the purchaser RPE discussed in Section IV.D.1 less the vehicle tax credit discussed in Section IV.D.2 plus the EVSE RPE in IV.D.3. Table IV–7 shows the estimated incremental upfront purchaser costs for BEVs and FCEVs by calendar year for the proposed option relative to the reference case. Note that EVSE costs are associated with BEVs only; FCEVs do not have any associated EVSE costs.

5. Operating Costs

We have estimated three types of operating costs associated with the proposed HD Phase 3 CO₂ emission standards and our potential projected technology pathway to comply with those proposed standards that includes BEV or FCEV powertrains. These three types of operating costs include decreased fuel costs of BEVs compared to comparable ICE vehicles, avoided diesel exhaust fluid (DEF) consumption by BEVs and FCEVs compared to comparable diesel-fueled ICE vehicles, and reduced maintenance and repair costs of BEVs and FCEVs as compared to comparable ICE vehicles. To estimate each of these costs, the results of MOVES runs, as discussed in DRIA Chapter 4, were used to estimate costs associated with fuel consumption, DEF consumption, and VMT. We have estimated the net effect on fuel costs, DEF costs, and maintenance and repair costs. We describe our approach in this Section IV.D.5.

Additional details on our methodology and estimates of operating costs per mile impacts are included in DRIA Chapter 3.4. Chapter 4 of the DRIA contains a description of the MOVES vehicle source types and regulatory classes. In short, we estimate costs in MOVES for vehicle source types that have both regulatory class populations and associated emission inventories. Also, throughout this section, LHD refers to light heavy-duty vehicles, MHD refers to medium heavy-duty vehicles, and HHD refers to heavy heavy-duty vehicles.

i. Costs Associated With Fuel Usage

To determine the total costs associated with fuel usage for MY 2027 vehicles, the fuel usage for each MOVES source type and regulatory class was multiplied by the fuel price from the AEO 2022 reference case for diesel, gasoline, and CNG prices over the first 28 years of the lifetime of the vehicle. Fuel costs per gallon and kWh are discussed in DRIA Chapter 2. We used retail fuel prices since we expect that retail fuel prices are the prices paid by owners of these ICE vehicles. For electric vehicle costs, the electricity price from the AEO 2022 reference case for commercial electricity end-use prices in cents per kWh was multiplied by the fuel usage in kWh. For hydrogen vehicle fuel costs, a value of $6.10/kg starting in 2027 and linearly decreasing to $4/kg in 2030 and held constant until 2055, as discussed in DRIA Chapter 2.5.3.1, was multiplied by fuel usage in kg. To calculate the average cost per mile of fuel usage for each scenario, MOVES source type and regulatory class, the fuel cost was divided by the VMT for each of the MY 2027 vehicles over the 28-year period. The estimates of fuel cost per mile for MY 2027 vehicles under the proposal are shown in Table IV–8 with 3 percent discounting and Table IV–9 with 7 percent discounting. Values shown as a dash (“-”), in Table IV–8 and Table IV–9 represent cases where a given MOVES source type and regulatory class did not use a specific fuel type for MY 2027 vehicles.