Light-Duty Vehicle Greenhouse Gas Emission Standards and Corporate Average Fuel Economy Standards; Final Rule

AGENCY:

Environmental Protection Agency (EPA) and National Highway Traffic Safety Administration (NHTSA).

ACTION:

Final rule.

SUMMARY:

EPA and NHTSA are issuing this joint Final Rule to establish a National Program consisting of new standards for light-duty vehicles that will reduce greenhouse gas emissions and improve fuel economy. This joint Final Rule is consistent with the National Fuel Efficiency Policy announced by President Obama on May 19, 2009, responding to the country's critical need to address global climate change and to reduce oil consumption. EPA is finalizing greenhouse gas emissions standards under the Clean Air Act, and NHTSA is finalizing Corporate Average Fuel Economy standards under the Energy Policy and Conservation Act, as amended. These standards apply to passenger cars, light-duty trucks, and medium-duty passenger vehicles, covering model years 2012 through 2016, and represent a harmonized and consistent National Program. Under the National Program, automobile manufacturers will be able to build a single light-duty national fleet that satisfies all requirements under both programs while ensuring that consumers still have a full range of vehicle choices. NHTSA's final rule also constitutes the agency's Record of Decision for purposes of its National Environmental Policy Act (NEPA) analysis.

DATES:

This final rule is effective on July 6, 2010, sixty days after date of publication in theFederal Register. The incorporation by reference of certain publications listed in this regulation is approved by the Director of the Federal Register as of July 6, 2010.

ADDRESSES:

EPA and NHTSA have established dockets for this action under Docket ID No. EPA-HQ-OAR-2009-0472 and NHTSA-2009-0059, respectively. All documents in the docket are listed on the http://www.regulations.gov Web site. 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. Publicly available docket materials are available either electronically through http://www.regulations.gov or in hard copy at the following locations: EPA: EPA Docket Center, EPA/DC, EPA West, Room 3334, 1301 Constitution Ave., NW., Washington, DC. The Public Reading Room is open from 8:30 a.m. to 4:30 p.m., Monday through Friday, excluding legal holidays. The telephone number for the Public Reading Room is (202) 566-1744. NHTSA: Docket Management Facility, M-30, U.S. Department of Transportation, West Building, Ground Floor, Rm. W12-140, 1200 New Jersey Avenue, SE., Washington, DC 20590. The Docket Management Facility is open between 9 a.m. and 5 p.m. Eastern Time, Monday through Friday, except Federal holidays.

FOR FURTHER INFORMATION CONTACT:

EPA: Tad Wysor, Office of Transportation and Air Quality, Assessment and Standards Division, Environmental Protection Agency, 2000 Traverwood Drive, Ann Arbor MI 48105; telephone number: 734-214-4332; fax number: 734-214-4816; e-mail address: wysor.tad@epa.gov, or Assessment and Standards Division Hotline; telephone number (734) 214-4636; e-mail address asdinfo@epa.gov. NHTSA: Rebecca Yoon, Office of Chief Counsel, National Highway Traffic Safety Administration, 1200 New Jersey Avenue, SE., Washington, DC 20590. Telephone: (202) 366-2992.

SUPPLEMENTARY INFORMATION:

Does this action apply to me?

This action affects companies that manufacture or sell new light-duty vehicles, light-duty trucks, and medium-duty passenger vehicles, as defined under EPA's CAA regulations, and passenger automobiles (passenger cars) and non-passenger automobiles (light trucks) as defined under NHTSA's CAFE regulations. Regulated categories and entities include:

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

Table of Contents

I. Overview of Joint EPA/NHTSA National Program

A. Introduction

1. Building Blocks of the National Program

2. Public Participation

B. Summary of the Joint Final Rule and Differences From the Proposal

1. Joint Analytical Approach

2. Level of the Standards

3. Form of the Standards

4. Program Flexibilities

5. Coordinated Compliance

C. Summary of Costs and Benefits of the National Program

1. Summary of Costs and Benefits of NHTSA's CAFE Standards

2. Summary of Costs and Benefits of EPA's GHG Standards

D. Background and Comparison of NHTSA and EPA Statutory Authority

II. Joint Technical Work Completed for This Final Rule

A. Introduction

B. Developing the Future Fleet for Assessing Costs, Benefits, and Effects

1. Why did the agencies establish a baseline and reference vehicle fleet?

2. How did the agencies develop the baseline vehicle fleet?

3. How did the agencies develop the projected MY 2011-2016 vehicle fleet?

4. How was the development of the baseline and reference fleets for this Final Rule different from NHTSA's historical approach?

5. How does manufacturer product plan data factor into the baseline used in this Final Rule?

C. Development of Attribute-Based Curve Shapes

D. Relative Car-Truck Stringency

E. Joint Vehicle Technology Assumptions

1. What technologies did the agencies consider?

2. How did the agencies determine the costs and effectiveness of each of these technologies?

F. Joint Economic Assumptions

G. What are the estimated safety effects of the final MYs 2012-2016 CAFE and GHG standards?

1. What did the agencies say in the NPRM with regard to potential safety effects?

2. What public comments did the agencies receive on the safety analysis and discussions in the NPRM?

3. How has NHTSA refined its analysis for purposes of estimating the potential safety effects of this Final Rule?

4. What are the estimated safety effects of this Final Rule?

5. How do the agencies plan to address this issue going forward?

III. EPA Greenhouse Gas Vehicle Standards

A. Executive Overview of EPA Rule

1. Introduction

2. Why is EPA establishing this Rule?

3. What is EPA adopting?

4. Basis for the GHG Standards Under Section 202(a)

B. GHG Standards for Light-Duty Vehicles, Light-Duty Trucks, and Medium-Duty Passenger Vehicles

1. What fleet-wide emissions levels correspond to the CO₂ standards?

2. What are the CO₂ attribute-based standards?

3. Overview of How EPA's CO₂ Standards Will Be Implemented for Individual Manufacturers

4. Averaging, Banking, and Trading Provisions for CO₂ Standards

5. CO₂ Temporary Lead-Time Allowance Alternative Standards

6. Deferment of CO₂ Standards for Small Volume Manufacturers With Annual Sales Less Than 5,000 Vehicles

7. Nitrous Oxide and Methane Standards

8. Small Entity Exemption

C. Additional Credit Opportunities for CO₂ Fleet Average Program

1. Air Conditioning Related Credits

2. Flexible Fuel and Alternative Fuel Vehicle Credits

3. Advanced Technology Vehicle Incentives for Electric Vehicles, Plug-in Hybrids, and Fuel Cell Vehicles

4. Off-Cycle Technology Credits

5. Early Credit Options

D. Feasibility of the Final CO₂ Standards

1. How did EPA develop a reference vehicle fleet for evaluating further CO₂ reductions?

2. What are the effectiveness and costs of CO₂-reducing technologies?

3. How can technologies be combined into “packages” and what is the cost and effectiveness of packages?

4. Manufacturer's Application of Technology

5. How is EPA projecting that a manufacturer decides between options to improve CO₂ performance to meet a fleet average standard?

6. Why are the final CO₂ standards feasible?

7. What other fleet-wide CO₂ levels were considered?

E. Certification, Compliance, and Enforcement

1. Compliance Program Overview

2. Compliance With Fleet-Average CO₂ Standards

3. Vehicle Certification

4. Useful Life Compliance

5. Credit Program Implementation

6. Enforcement

7. Prohibited Acts in the CAA

8. Other Certification Issues

9. Miscellaneous Revisions to Existing Regulations

10. Warranty, Defect Reporting, and Other Emission-Related Components Provisions

11. Light Duty Vehicles and Fuel Economy Labeling

F. How will this Final Rule reduce GHG emissions and their associated effects?

1. Impact on GHG Emissions

2. Overview of Climate Change Impacts From GHG Emissions

3. Changes in Global Climate Indicators Associated With the Rule's GHG Emissions Reductions

G. How will the standards impact non-GHG emissions and their associated effects?

1. Upstream Impacts of Program

2. Downstream Impacts of Program

3. Health Effects of Non-GHG Pollutants

4. Environmental Effects of Non-GHG Pollutants

5. Air Quality Impacts of Non-GHG Pollutants

H. What are the estimated cost, economic, and other impacts of the program?

1. Conceptual Framework for Evaluating Consumer Impacts

2. Costs Associated With the Vehicle Program

3. Cost per Ton of Emissions Reduced

4. Reduction in Fuel Consumption and Its Impacts

5. Impacts on U.S. Vehicle Sales and Payback Period

6. Benefits of Reducing GHG Emissions

7. Non-Greenhouse Gas Health and Environmental Impacts

8. Energy Security Impacts

9. Other Impacts

10. Summary of Costs and Benefits

I. Statutory and Executive Order Reviews

1. Executive Order 12866: Regulatory Planning and Review

2. Paperwork Reduction Act

3. Regulatory Flexibility Act

4. Unfunded Mandates Reform Act

5. Executive Order 13132 (Federalism)

6. Executive Order 13175 (Consultation and Coordination With Indian Tribal

Governments)

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

8. Executive Order 13211 (Energy Effects)

9. National Technology Transfer Advancement Act

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

J. Statutory Provisions and Legal Authority

IV. NHTSA Final Rule and Record of Decision for Passenger Car and Light Truck CAFE Standards for MYs 2012-2016

A. Executive Overview of NHTSA Final Rule

1. Introduction

2. Role of Fuel Economy Improvements in Promoting Energy Independence, Energy Security, and a Low Carbon Economy

3. The National Program

4. Review of CAFE Standard Setting Methodology per the President's January 26, 2009 Memorandum on CAFE Standards for MYs 2011 and Beyond

5. Summary of the Final MY 2012-2016 CAFE Standards

B. Background

1. Chronology of Events Since the National Academy of Sciences Called for Reforming and Increasing CAFE Standards

2. Energy Policy and Conservation Act, as Amended by the Energy Independence and Security Act

C. Development and Feasibility of the Final Standards

1. How was the baseline and reference vehicle fleet developed?

2. How were the technology inputs developed?

3. How did NHTSA develop the economic assumptions?

4. How does NHTSA use the assumptions in its modeling analysis?

5. How did NHTSA develop the shape of the target curves for the final standards?

D. Statutory Requirements

1. EPCA, as Amended by EISA

2. Administrative Procedure Act

3. National Environmental Policy Act

E. What are the final CAFE standards?

1. Form of the Standards

2. Passenger Car Standards for MYs 2012-2016

3. Minimum Domestic Passenger Car Standards

4. Light Truck Standards

F. How do the final standards fulfill NHTSA's statutory obligations?

G. Impacts of the Final CAFE Standards

1. How will these standards improve fuel economy and reduce GHG emissions for MY 2012-2016 vehicles?

2. How will these standards improve fleet-wide fuel economy and reduce GHG emissions beyond MY 2016?

3. How will these final standards impact non-GHG emissions and their associated effects?

4. What are the estimated costs and benefits of these final standards?

5. How would these standards impact vehicle sales?

6. Potential Unquantified Consumer Welfare Impacts of the Final Standards

7. What other impacts (quantitative and unquantifiable) will these final standards have?

H. Vehicle Classification

I. Compliance and Enforcement

1. Overview

2. How does NHTSA determine compliance?

3. What compliance flexibilities are available under the CAFE program and how do manufacturers use them?

4. Other CAFE Enforcement Issues—Variations in Footprint

5. Other CAFE Enforcement Issues—Miscellaneous

J. Other Near-Term Rulemakings Mandated by EISA

1. Commercial Medium- and Heavy-Duty On-Highway Vehicles and Work Trucks

2. Consumer Information on Fuel Efficiency and Emissions

K. NHTSA's Record of Decision

L. Regulatory Notices and Analyses

1. Executive Order 12866 and DOT Regulatory Policies and Procedures

2. National Environmental Policy Act

3. Clean Air Act (CAA)

4. National Historic Preservation Act (NHPA)

5. Executive Order 12898 (Environmental Justice)

6. Fish and Wildlife Conservation Act (FWCA)

7. Coastal Zone Management Act (CZMA)

8. Endangered Species Act (ESA)

9. Floodplain Management (Executive Order 11988 & DOT Order 5650.2)

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

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

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

13. Regulatory Flexibility Act

14. Executive Order 13132 (Federalism)

15. Executive Order 12988 (Civil Justice Reform)

16. Unfunded Mandates Reform Act

17. Regulation Identifier Number

18. Executive Order 13045

19. National Technology Transfer and Advancement Act

20. Executive Order 13211

21. Department of Energy Review

22. Privacy Act

I. Overview of Joint EPA/NHTSA National Program

A. Introduction

The National Highway Traffic Safety Administration (NHTSA) and the Environmental Protection Agency (EPA) are each announcing final rules whose benefits will address the urgent and closely intertwined challenges of energy independence and security and global warming. These rules will implement a strong and coordinated Federal greenhouse gas (GHG) and fuel economy program for passenger cars, light-duty-trucks, and medium-duty passenger vehicles (hereafter light-duty vehicles), referred to as the National Program. The rules will achieve substantial reductions of GHG emissions and improvements in fuel economy from the light-duty vehicle part of the transportation sector, based on technology that is already being commercially applied in most cases and that can be incorporated at a reasonable cost. NHTSA's final rule also constitutes the agency's Record of Decision for purposes of its NEPA analysis.

This joint rulemaking is consistent with the President's announcement on May 19, 2009 of a National Fuel Efficiency Policy of establishing consistent, harmonized, and streamlined requirements that would reduce GHG emissions and improve fuel economy for all new cars and light-duty trucks sold in the United States. The National Program will deliver additional environmental and energy benefits, cost savings, and administrative efficiencies on a nationwide basis that would likely not be available under a less coordinated approach. The National Program also represents regulatory convergence by making it possible for the standards of two different Federal agencies and the standards of California and other states to act in a unified fashion in providing these benefits. The National Program will allow automakers to produce and sell a single fleet nationally, mitigating the additional costs that manufacturers would otherwise face in having to comply with multiple sets of Federal and State standards. This joint notice is also consistent with the Notice of Upcoming Joint Rulemaking issued by DOT and EPA on May 19, 2009 and responds to the President's January 26, 2009 memorandum on CAFE standards for model years 2011 and beyond, the details of which can be found in Section IV of this joint notice.

Climate change is widely viewed as a significant long-term threat to the global environment. As summarized in the Technical Support Document for EPA's Endangerment and Cause or Contribute Findings under Section 202(a) of the Clear Air Act, anthropogenic emissions of GHGs are very likely (90 to 99 percent probability) the cause of most of the observed global warming over the last 50 years. The primary GHGs of concern are carbon dioxide (CO₂), methane, nitrous oxide, hydrofluorocarbons, perfluorocarbons, and sulfur hexafluoride. Mobile sources emitted 31 percent of all U.S. GHGs in 2007 (transportation sources, which do not include certain off-highway sources, account for 28 percent) and have been the fastest-growing source of U.S. GHGs since 1990. Mobile sources addressed in the recent endangerment and contribution findings under CAA section 202(a)—light-duty vehicles, heavy-duty trucks, buses, and motorcycles—accounted for 23 percent of all U.S. GHG in 2007. Light-duty vehicles emit CO₂, methane, nitrous oxide, and hydrofluorocarbons and are responsible for nearly 60 percent of all mobile source GHGs and over 70 percent of Section 202(a) mobile source GHGs. For light-duty vehicles in 2007, CO₂ emissions represent about 94 percent of all greenhouse emissions (including HFCs), and the CO₂ emissions measured over the EPA tests used for fuel economy compliance represent about 90 percent of total light-duty vehicle GHG emissions.

Improving energy security by reducing our dependence on foreign oil has been a national objective since the first oil price shocks in the 1970s. Net petroleum imports now account for approximately 60 percent of U.S. petroleum consumption. World crude oil production is highly concentrated, exacerbating the risks of supply disruptions and price shocks. Tight global oil markets led to prices over $100 per barrel in 2008, with gasoline reaching as high as $4 per gallon in many parts of the U.S., causing financial hardship for many families. The export of U.S. assets for oil imports continues to be an important component of the historically unprecedented U.S. trade deficits. Transportation accounts for about two-thirds of U.S. petroleum consumption. Light-duty vehicles account for about 60 percent of transportation oil use, which means that they alone account for about 40 percent of all U.S. oil consumption.

1. Building Blocks of the National Program

The National Program is both needed and possible because the relationship between improving fuel economy and reducing CO₂ tailpipe emissions is a very direct and close one. The amount of those CO₂ emissions is essentially constant per gallon combusted of a given type of fuel. Thus, the more fuel efficient a vehicle is, the less fuel it burns to travel a given distance. The less fuel it burns, the less CO₂ it emits in traveling that distance. While there are emission control technologies that reduce the pollutants (e.g., carbon monoxide) produced by imperfect combustion of fuel by capturing or converting them to other compounds, there is no such technology for CO₂. Further, while some of those pollutants can also be reduced by achieving a more complete combustion of fuel, doing so only increases the tailpipe emissions of CO₂. Thus, there is a single pool of technologies for addressing these twin problems, i.e., those that reduce fuel consumption and thereby reduce CO₂ emissions as well.

a. DOT's CAFE Program

In 1975, Congress enacted the Energy Policy and Conservation Act (EPCA), mandating that NHTSA establish and implement a regulatory program for motor vehicle fuel economy to meet the various facets of the need to conserve energy, including ones having energy independence and security, environmental and foreign policy implications. Fuel economy gains since 1975, due both to the standards and market factors, have resulted in saving billions of barrels of oil and avoiding billions of metric tons of CO₂ emissions. In December 2007, Congress enacted the Energy Independence and Securities Act (EISA), amending EPCA to require substantial, continuing increases in fuel economy standards.

The CAFE standards address most, but not all, of the real world CO₂ emissions because a provision in EPCA as originally enacted in 1975 requires the use of the 1975 passenger car test procedures under which vehicle air conditioners are not turned on during fuel economy testing. Fuel economy is determined by measuring the amount of CO₂ and other carbon compounds emitted from the tailpipe, not by attempting to measure directly the amount of fuel consumed during a vehicle test, a difficult task to accomplish with precision. The carbon content of the test fuel is then used to calculate the amount of fuel that had to be consumed per mile in order to produce that amount of CO₂. Finally, that fuel consumption figure is converted into a miles-per-gallon figure. CAFE standards also do not address the 5-8 percent of GHG emissions that are not CO₂, i.e., nitrous oxide (N₂ O), and methane (CH₄) as well as emissions of CO₂ and hydrofluorocarbons (HFCs) related to operation of the air conditioning system.

b. EPA's GHG Standards for Light-duty Vehicles

Under the Clean Air Act EPA is responsible for addressing air pollutants from motor vehicles. On April 2, 2007, the U.S. Supreme Court issued its opinion in Massachusetts v. EPA, a case involving EPA's a 2003 denial of a petition for rulemaking to regulate GHG emissions from motor vehicles under section 202(a) of the Clean Air Act (CAA). The Court held that GHGs fit within the definition of air pollutant in the Clean Air Act and further held that the Administrator must determine whether or not emissions from new motor vehicles cause or contribute to air pollution which may reasonably be anticipated to endanger public health or welfare, or whether the science is too uncertain to make a reasoned decision. The Court further ruled that, in making these decisions, the EPA Administrator is required to follow the language of section 202(a) of the CAA. The Court rejected the argument that EPA cannot regulate CO₂ from motor vehicles because to do so would de facto tighten fuel economy standards, authority over which has been assigned by Congress to DOT. The Court stated that “[b]ut that DOT sets mileage standards in no way licenses EPA to shirk its environmental responsibilities. 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.” The Court concluded that “[t]he two obligations may overlap, but there is no reason to think the two agencies cannot both administer their obligations and yet avoid inconsistency.” The case was remanded back to the Agency for reconsideration in light of the Court's decision.

On December 15, 2009, EPA published two findings (74 FR 66496): That emissions of GHGs from new motor vehicles and motor vehicle engines contribute to air pollution, and that the air pollution may reasonably be anticipated to endanger public health and welfare.

c. California Air Resources Board Greenhouse Gas Program

In 2004, the California Air Resources Board approved standards for new light-duty vehicles, which regulate the emission of not only CO₂, but also other GHGs. Since then, thirteen states and the District of Columbia, comprising approximately 40 percent of the light-duty vehicle market, have adopted California's standards. These standards apply to model years 2009 through 2016 and require CO₂ emissions for passenger cars and the smallest light trucks of 323 g/mi in 2009 and 205 g/mi in 2016, and for the remaining light trucks of 439 g/mi in 2009 and 332 g/mi in 2016. On June 30, 2009, EPA granted California's request for a waiver of preemption under the CAA. The granting of the waiver permits California and the other states to proceed with implementing the California emission standards.

In addition, to promote the National Program, in May 2009, California announced its commitment to take several actions in support of the National Program, including revising its program for MYs 2009-2011 to facilitate compliance by the automakers, and revising its program for MYs 2012-2016 such that compliance with the Federal GHG standards will be deemed to be compliance with California's GHG standards. This will allow the single national fleet produced by automakers to meet the two Federal requirements and to meet California requirements as well. California is proceeding with a rulemaking intended to revise its 2004 regulations to meet its commitments. Several automakers and their trade associations also announced their commitment to take several actions in support of the National Program, including not contesting the final GHG and CAFE standards for MYs 2012-2016, not contesting any grant of a waiver of preemption under the CAA for California's GHG standards for certain model years, and to stay and then dismiss all pending litigation challenging California's regulation of GHG emissions, including litigation concerning preemption under EPCA of California's and other states' GHG standards.

2. Public Participation

The agencies proposed their respective rules on September 28, 2009 (74 FR 49454), and received a large number of comments representing many perspectives on the proposed rule. The agencies received oral testimony at three public hearings in different parts of the country, and received written comments from more than 130 organizations, including auto manufacturers and suppliers, States, environmental and other non-governmental organizations (NGOs), and over 129,000 comments from private citizens.

The vast majority of commenters supported the central tenets of the proposed CAFE and GHG programs. That is, there was broad support from most organizations for a National Program that achieves a level of 250 gram/mile fleet average CO₂, which would be 35.5 miles per gallon if the automakers were to meet this CO₂ level solely through fuel economy improvements. The standards will be phased in over model years 2012 through 2016 which will allow manufacturers to build a common fleet of vehicles for the domestic market. In general, commenters from the automobile industry supported the proposed standards as well as the credit opportunities and other compliance provisions providing flexibility, while also making some recommendations for changes. Environmental and public interest non-governmental organizations (NGOs), as well as most States that commented, were also generally supportive of the National Program standards. Many of these organizations also expressed concern about the possible impact on program benefits, depending on how the credit provisions and flexibilities are designed. The agencies also received specific comments on many aspects of the proposal.

Throughout this notice, the agencies discuss many of the key issues arising from the public comments and the agencies' responses. In addition, the agencies have addressed all of the public comments in the Response to Comments document associated with this final rule.

B. Summary of the Joint Final Rule and Differences From the Proposal

In this joint rulemaking, EPA is establishing GHG emissions standards under the Clean Air Act (CAA), and NHTSA is establishing Corporate Average Fuel Economy (CAFE) standards under the Energy Policy and Conservation Action of 1975 (EPCA), as amended by the Energy Independence and Security Act of 2007 (EISA). The intention of this joint rulemaking is to set forth a carefully coordinated and harmonized approach to implementing these two statutes, in accordance with all substantive and procedural requirements imposed by law.

NHTSA and EPA have coordinated closely and worked jointly in developing their respective final rules. This is reflected in many aspects of this joint rule. For example, the agencies have developed a comprehensive Joint Technical Support Document (TSD) that provides a solid technical underpinning for each agency's modeling and analysis used to support their standards. Also, to the extent allowed by law, the agencies have harmonized many elements of program design, such as the form of the standard (the footprint-based attribute curves), and the definitions used for cars and trucks. They have developed the same or similar compliance flexibilities, to the extent allowed and appropriate under their respective statutes, such as averaging, banking, and trading of credits, and have harmonized the compliance testing and test protocols used for purposes of the fleet average standards each agency is finalizing. Finally, under their respective statutes, each agency is called upon to exercise its judgment and determine standards that are an appropriate balance of various relevant statutory factors. Given the common technical issues before each agency, the similarity of the factors each agency is to consider and balance, and the authority of each agency to take into consideration the standards of the other agency, both EPA and NHTSA are establishing standards that result in a harmonized National Program.

This joint final rule covers passenger cars, light-duty trucks, and medium-duty passenger vehicles built in model years 2012 through 2016. These vehicle categories are responsible for almost 60 percent of all U.S. transportation-related GHG emissions. EPA and NHTSA expect that automobile manufacturers will meet these standards by utilizing technologies that will reduce vehicle GHG emissions and improve fuel economy. Although many of these technologies are available today, the emissions reductions and fuel economy improvements finalized in this notice will involve more widespread use of these technologies across the light-duty vehicle fleet. These include improvements to engines, transmissions, and tires, increased use of start-stop technology, improvements in air conditioning systems, increased use of hybrid and other advanced technologies, and the initial commercialization of electric vehicles and plug-in hybrids. NHTSA's and EPA's assessments of likely vehicle technologies that manufacturers will employ to meet the standards are discussed in detail below and in the Joint TSD.

The National Program is estimated to result in approximately 960 million metric tons of total carbon dioxide equivalent emissions reductions and approximately 1.8 billion barrels of oil savings over the lifetime of vehicles sold in model years (MYs) 2012 through 2016. In total, the combined EPA and NHTSA 2012-2016 standards will reduce GHG emissions from the U.S. light-duty fleet by approximately 21 percent by 2030 over the level that would occur in the absence of the National Program. These actions also will provide important energy security benefits, as light-duty vehicles are about 95 percent dependent on oil-based fuels. The agencies project that the total benefits of the National Program will be more than $240 billion at a 3% discount rate, or more than $190 billion at a 7% discount rate. In the discussion that follows in Sections III and IV, each agency explains the related benefits for their individual standards.

Together, EPA and NHTSA estimate that the average cost increase for a model year 2016 vehicle due to the National Program will be less than $1,000. The average U.S. consumer who purchases a vehicle outright is estimated to save enough in lower fuel costs over the first three years to offset these higher vehicle costs. However, most U.S. consumers purchase a new vehicle using credit rather than paying cash and the typical car loan today is a five year, 60 month loan. These consumers will see immediate savings due to their vehicle's lower fuel consumption in the form of a net reduction in annual costs of $130-$180 throughout the duration of the loan (that is, the fuel savings will outweigh the increase in loan payments by $130-$180 per year). Whether a consumer takes out a loan or purchases a new vehicle outright, over the lifetime of a model year 2016 vehicle, the consumer's net savings could be more than $3,000. The average 2016 MY vehicle will emit 16 fewer metric tons of CO₂-equivalent emissions (that is, CO₂ emissions plus HFC air conditioning leakage emissions) during its lifetime. Assumptions that underlie these conclusions are discussed in greater detail in the agencies' respective regulatory impact analyses and in Section III.H.5 and Section IV.

This joint rule also results in important regulatory convergence and certainty to automobile companies. Absent this rule, there would be three separate Federal and State regimes independently regulating light-duty vehicles to reduce fuel consumption and GHG emissions: NHTSA's CAFE standards, EPA's GHG standards, and the GHG standards applicable in California and other States adopting the California standards. This joint rule will allow automakers to meet both the NHTSA and EPA requirements with a single national fleet, greatly simplifying the industry's technology, investment and compliance strategies. In addition, to promote the National Program, California announced its commitment to take several actions, including revising its program for MYs 2012-2016 such that compliance with the Federal GHG standards will be deemed to be compliance with California's GHG standards. This will allow the single national fleet used by automakers to meet the two Federal requirements and to meet California requirements as well. California is proceeding with a rulemaking intended to revise its 2004 regulations to meet its commitments. EPA and NHTSA are confident that these GHG and CAFE standards will successfully harmonize both the Federal and State programs for MYs 2012-2016 and will allow our country to achieve the increased benefits of a single, nationwide program to reduce light-duty vehicle GHG emissions and reduce the country's dependence on fossil fuels by improving these vehicles' fuel economy.

A successful and sustainable automotive industry depends upon, among other things, continuous technology innovation in general, and low GHG emissions and high fuel economy vehicles in particular. In this respect, this action will help spark the investment in technology innovation necessary for automakers to successfully compete in both domestic and export markets, and thereby continue to support a strong economy.

While this action covers MYs 2012-2016, many stakeholders encouraged EPA and NHTSA to also begin working toward standards for MY 2017 and beyond that would maintain a single nationwide program. The agencies recognize the importance of and are committed to a strong, coordinated national program for light-duty vehicles for model years beyond 2016.

Key elements of the National Program finalized today are the level and form of the GHG and CAFE standards, the available compliance mechanisms, and general implementation elements. These elements are summarized in the following section, with more detailed discussions about EPA's GHG program following in Section III, and about NHTSA's CAFE program in Section IV. This joint final rule responds to the wide array of comments that the agencies received on the proposed rule. This section summarizes many of the major comments on the primary elements of the proposal and describes whether and how the final rule has changed, based on the comments and additional analyses. Major comments and the agencies' responses to them are also discussed in more detail in later sections of this preamble. For a full summary of public comments and EPA's and NHTSA's responses to them, please see the Response to Comments document associated with this final rule.

1. Joint Analytical Approach

NHTSA and EPA have worked closely together on nearly every aspect of this joint final rule. The extent and results of this collaboration are reflected in the elements of the respective NHTSA and EPA rules, as well as the analytical work contained in the Joint Technical Support Document (Joint TSD). The Joint TSD, in particular, describes important details of the analytical work that are shared, as well as any differences in approach. These include the build up of the baseline and reference fleets, the derivation of the shape of the curves that define the standards, a detailed description of the costs and effectiveness of the technology choices that are available to vehicle manufacturers, a summary of the computer models used to estimate how technologies might be added to vehicles, and finally the economic inputs used to calculate the impacts and benefits of the rules, where practicable.

EPA and NHTSA have jointly developed attribute curve shapes that each agency is using for its final standards. Further details of these functions can be found in Sections III and IV of this preamble as well as Chapter 2 of the Joint TSD. A critical technical underpinning of each agency's analysis is the cost and effectiveness of the various control technologies. These are used to analyze the feasibility and cost of potential GHG and CAFE standards. A detailed description of all of the technology information considered can be found in Chapter 3 of the Joint TSD (and for A/C, Chapter 2 of the EPA RIA). This detailed technology data forms the inputs to computer models that each agency uses to project how vehicle manufacturers may add those technologies in order to comply with the new standards. These are the OMEGA and Volpe models for EPA and NHTSA, respectively. The models and their inputs can also be found in the docket. Further description of the model and outputs can be found in Sections III and IV of this preamble, and Chapter 3 of the Joint TSD. This comprehensive joint analytical approach has provided a sound and consistent technical basis for each agency in developing its final standards, which are summarized in the sections below.

The vast majority of public comments expressed strong support for the joint analytical work performed for the proposal. Commenters generally agreed with the analytical work and its results, and supported the transparency of the analysis and its underlying data. Where commenters raised specific points, the agencies have considered them and made changes where appropriate. The agencies' further evaluation of various technical issues also led to a limited number of changes. A detailed discussion of these issues can be found in Section II of this preamble, and the Joint TSD.

2. Level of the Standards

In this notice, EPA and NHTSA are establishing two separate sets of standards, each under its respective statutory authorities. EPA is setting national CO₂ emissions standards for light-duty vehicles under section 202(a) of the Clean Air Act. These standards will require these vehicles to meet an estimated combined average emissions level of 250 grams/mile of CO₂ in model year 2016. NHTSA is setting CAFE standards for passenger cars and light trucks under 49 U.S.C. 32902. These standards will require manufacturers of those vehicles to meet an estimated combined average fuel economy level of 34.1 mpg in model year 2016. The standards for both agencies begin with the 2012 model year, with standards increasing in stringency through model year 2016. They represent a harmonized approach that will allow industry to build a single national fleet that will satisfy both the GHG requirements under the CAA and CAFE requirements under EPCA/EISA.

Given differences in their respective statutory authorities, however, the agencies' standards include some important differences. Under the CO₂ fleet average standards adopted under CAA section 202(a), EPA expects manufacturers to take advantage of the option to generate CO₂-equivalent credits by reducing emissions of hydrofluorocarbons (HFCs) and CO₂ through improvements in their air conditioner systems. EPA accounted for these reductions in developing its final CO₂ standards. NHTSA did not do so because EPCA does not allow vehicle manufacturers to use air conditioning credits in complying with CAFE standards for passenger cars. CO₂ emissions due to air conditioning operation are not measured by the test procedure mandated by statute for use in establishing and enforcing CAFE standards for passenger cars. As a result, improvement in the efficiency of passenger car air conditioners is not considered as a possible control technology for purposes of CAFE.

These differences regarding the treatment of air conditioning improvements (related to CO₂ and HFC reductions) affect the relative stringency of the EPA standard and NHTSA standard for MY 2016. The 250 grams per mile of CO₂ equivalent emissions limit is equivalent to 35.5 mpg if the automotive industry were to meet this CO₂ level all through fuel economy improvements. As a consequence of the prohibition against NHTSA's allowing credits for air conditioning improvements for purposes of passenger car CAFE compliance, NHTSA is setting fuel economy standards that are estimated to require a combined (passenger car and light truck) average fuel economy level of 34.1 mpg by MY 2016.

The vast majority of public comments expressed strong support for the National Program standards, including the stringency of the agencies' respective standards and the phase-in from model year 2012 through 2016. There were a number of comments supporting standards more stringent than proposed, and a few others supporting less stringent standards, in particular for the 2012-2015 model years. The agencies' consideration of comments and their updated technical analyses led to only very limited changes in the footprint curves and did not change the agencies' projections that the nationwide fleet will achieve a level of 250 grams/mile by 2016 (equivalent to 35.5 mpg). The responses to these comments are discussed in more detail in Sections III and IV, respectively, and in the Response to Comments document.

As proposed, NHTSA and EPA's final standards, like the standards NHTSA promulgated in March 2009 for MY 2011, are expressed as mathematical functions depending on vehicle footprint. Footprint is one measure of vehicle size, and is determined by multiplying the vehicle's wheelbase by the vehicle's average track width. The standards that must be met by each manufacturer's fleet will be determined by computing the sales-weighted average (harmonic average for CAFE) of the targets applicable to each of the manufacturer's passenger cars and light trucks. Under these footprint-based standards, the levels required of individual manufacturers will depend, as noted above, on the mix of vehicles sold. NHTSA's and EPA's respective standards are shown in the tables below. It is important to note that the standards are the attribute-based curves established by each agency. The values in the tables below reflect the agencies' projection of the corresponding fleet levels that will result from these attribute-based curves.

As a result of public comments and updated economic and future fleet projections, EPA and NHTSA have updated the attribute based curves for this final rule, as discussed in detail in Section II.B of this preamble and Chapter 2 of the Joint TSD. This update in turn affects costs, benefits, and other impacts of the final standards. Thus, the agencies have updated their overall projections of the impacts of the final rule standards, and these results are only slightly different from those presented in the proposed rule.

As shown in Table I.B.2-1, NHTSA's fleet-wide CAFE-required levels for passenger cars under the final standards are projected to increase from 33.3 to 37.8 mpg between MY 2012 and MY 2016. Similarly, fleet-wide CAFE levels for light trucks are projected to increase from 25.4 to 28.8 mpg. NHTSA has also estimated the average fleet-wide required levels for the combined car and truck fleets. As shown, the overall fleet average CAFE level is expected to be 34.1 mpg in MY 2016. These numbers do not include the effects of other flexibilities and credits in the program. These standards represent a 4.3 percent average annual rate of increase relative to the MY 2011 standards.

Accounting for the expectation that some manufacturers could continue to pay civil penalties rather than achieving required CAFE levels, and the ability to use FFV credits, NHTSA estimates that the CAFE standards will lead to the following average achieved fuel economy levels, based on the projections of what each manufacturer's fleet will comprise in each year of the program:

NHTSA is also required by EISA to set a minimum fuel economy standard for domestically manufactured passenger cars in addition to the attribute-based passenger car standard. The minimum standard “shall be the greater of (A) 27.5 miles per gallon; or (B) 92 percent of the average fuel economy projected by the Secretary for the combined domestic and non-domestic passenger automobile fleets manufactured for sale in the United States by all manufacturers in the model year.* * * ”

Based on NHTSA's current market forecast, the agency's estimates of these minimum standards under the MY 2012-2016 CAFE standards (and, for comparison, the final MY 2011 standard) are summarized below in Table I.B.2-3. For eventual compliance calculations, the final calculated minimum standards will be updated to reflect the average fuel economy level required under the final standards.

EPA is establishing GHG emissions standards, and Table I.B.2-4 provides EPA's estimates of their projected overall fleet-wide CO₂ equivalent emission levels. The g/mi values are CO₂ equivalent values because they include the projected use of air conditioning (A/C) credits by manufacturers, which include both HFC and CO₂ reductions.

As shown in Table I.B.2-4, fleet-wide CO₂ emission level requirements for cars are projected to increase in stringency from 263 to 225 g/mi between MY 2012 and MY 2016. Similarly, fleet-wide CO₂ equivalent emission level requirements for trucks are projected to increase in stringency from 346 to 298 g/mi. As shown, the overall fleet average CO₂ level requirements are projected to increase in stringency from 295 g/mi in MY 2012 to 250 g/mi in MY 2016.

EPA anticipates that manufacturers will take advantage of program flexibilities such as flexible fueled vehicle credits and car/truck credit trading. Due to the credit trading between cars and trucks, the estimated improvements in CO₂ emissions are distributed differently than shown in Table I.B.2-4, where full manufacturer compliance without credit trading is assumed. Table I.B.2-5 shows EPA's projection of the achieved emission levels of the fleet for MY 2012 through 2016, which does consider the impact of car/truck credit transfer and the increase in emissions due to certain program flexibilities including flex fueled vehicle credits and the temporary lead time allowance alternative standards. The use of optional air conditioning credits is considered both in this analysis of achieved levels and of the compliance levels described above. As can be seen in Table I.B.2-5, the projected achieved levels are slightly higher for model years 2012-2015 due to EPA's assumptions about manufacturers' use of the regulatory flexibilities, but by model year 2016 the achieved level is projected to be 250 g/mi for the fleet.

Several auto manufacturers stated that the increasingly stringent requirements for fuel economy and GHG emissions in the early years of the program should follow a more linear phase-in. The agencies' consideration of comments and of their updated technical analyses did not lead to changes to the phase-in of the standards discussed above. This issue is discussed in more detail in Sections II.D, and in Sections III and IV.

NHTSA's and EPA's technology assessment indicates there is a wide range of technologies available for manufacturers to consider in upgrading vehicles to reduce GHG emissions and improve fuel economy. Commenters were in general agreement with this assessment. As noted, these include improvements to the engines such as use of gasoline direct injection and downsized engines that use turbochargers to provide performance similar to that of larger engines, the use of advanced transmissions, increased use of start-stop technology, improvements in tire rolling resistance, reductions in vehicle weight, increased use of hybrid and other advanced technologies, and the initial commercialization of electric vehicles and plug-in hybrids. EPA is also projecting improvements in vehicle air conditioners including more efficient as well as low leak systems. All of these technologies are already available today, and EPA's and NHTSA's assessments are that manufacturers will be able to meet the standards through more widespread use of these technologies across the fleet.

With respect to the practicability of the standards in terms of lead time, during MYs 2012-2016 manufacturers are expected to go through the normal automotive business cycle of redesigning and upgrading their light-duty vehicle products, and in some cases introducing entirely new vehicles not on the market today. This rule allows manufacturers the time needed to incorporate technology to achieve GHG reductions and improve fuel economy during the vehicle redesign process. This is an important aspect of the rule, as it avoids the much higher costs that would occur if manufacturers needed to add or change technology at times other than their scheduled redesigns. This time period also provides manufacturers the opportunity to plan for compliance using a multi-year time frame, again consistent with normal business practice. Over these five model years, there will be an opportunity for manufacturers to evaluate almost every one of their vehicle model platforms and add technology in a cost effective way to control GHG emissions and improve fuel economy. This includes redesign of the air conditioner systems in ways that will further reduce GHG emissions. Various commenters stated that the proposed phase-in of the standards should be introduced more aggressively, less aggressively, or in a more linear manner. However, our consideration of these comments about the phase-in, as well as our revised analyses, leads us to conclude that the general rate of introduction of the standards as proposed remains appropriate. This conclusion is also not affected by the slight difference from the proposal in the final footprint-based curves. These issues are addressed further in Sections III and IV.

Both agencies considered other standards as part of the rulemaking analyses, both more and less stringent than those proposed. EPA's and NHTSA's analyses of alternative standards are contained in Sections III and IV of this preamble, respectively, as well as the agencies' respective RIAs.

The CAFE and GHG standards described above are based on determining emissions and fuel economy using the city and highway test procedures that are currently used in the CAFE program. Some environmental and other organizations commented that the test procedures should be improved to reflect more real-world driving conditions; auto manufacturers in general do not support such changes to the test procedures at this time. Both agencies recognize that these test procedures are not fully representative of real-world driving conditions. For example, EPA has adopted more representative test procedures that are used in determining compliance with emissions standards for pollutants other than GHGs. These test procedures are also used in EPA's fuel economy labeling program. However, as discussed in Section III, the current information on effectiveness of the individual emissions control technologies is based on performance over the CAFE test procedures. For that reason, EPA is using the current CAFE test procedures for the CO₂ standards and is not changing those test procedures in this rulemaking. NHTSA, as discussed above, is limited by statute in what test procedures can be used for purposes of passenger car testing, although there is no such statutory limitation with respect to test procedures for trucks. However, the same reasons for not changing the truck test procedures apply for CAFE as well.

Both EPA and NHTSA are interested in developing programs that employ test procedures that are more representative of real-world driving conditions, to the extent authorized under their respective statutes. This is an important issue, and the agencies intend to continue to evaluate it in the context of a future rulemaking to address standards for model year 2017 and thereafter. This could include consideration of a range of test procedure changes to better represent real-world driving conditions in terms of speed, acceleration, deceleration, ambient temperatures, use of air conditioners, and the like. With respect to air conditioner operation, EPA discusses the public comments on these issues and the final procedures for determining emissions credits for controls on air conditioners in Section III.

Finally, based on the information EPA developed in its recent rulemaking that updated its fuel economy labeling program to better reflect average real-world fuel economy, the calculation of fuel savings and CO₂ emissions reductions that will be achieved by the CAFE and GHG standards includes adjustments to account for the difference between the fuel economy level measured in the CAFE test procedure and the fuel economy actually achieved on average under real-world driving conditions. These adjustments are industry averages for the vehicles' performance as a whole, however, and are not a substitute for the information on effectiveness of individual control technologies that will be explored for purposes of a future GHG and CAFE rulemaking.

3. Form of the Standards

NHTSA and EPA proposed attribute-based standards for passenger cars and light trucks. NHTSA adopted an attribute approach based on vehicle footprint in its Reformed CAFE program for light trucks for model years 2008-2011, and recently extended this approach to passenger cars in the CAFE rule for MY 2011 as required by EISA. The agencies also proposed using vehicle footprint as the attribute for the GHG and CAFE standards. Footprint is defined as a vehicle's wheelbase multiplied by its track width—in other words, the area enclosed by the points at which the wheels meet the ground. Most commenters that expressed a view on this topic supported basing the standards on an attribute, and almost all of these supported the proposed choice of vehicle footprint as an appropriate attribute. The agencies continue to believe that the standards are best expressed in terms of an attribute, and that the footprint attribute is the most appropriate attribute on which to base the standards. These issues are further discussed later in this notice and in Chapter 2 of the Joint TSD.

Under the footprint-based standards, each manufacturer will have a GHG and CAFE target unique to its fleet, depending on the footprints of the vehicle models produced by that manufacturer. A manufacturer will have separate footprint-based standards for cars and for trucks. Generally, larger vehicles (i.e., vehicles with larger footprints) will be subject to less stringent standards (i.e., higher CO₂ grams/mile standards and lower CAFE standards) than smaller vehicles. This is because, generally speaking, smaller vehicles are more capable of achieving lower levels of CO₂ and higher levels of fuel economy than larger vehicles. While a manufacturer's fleet average standard could be estimated throughout the model year based on projected production volume of its vehicle fleet, the standard to which the manufacturer must comply will be based on its final model year production figures. A manufacturer's calculation of fleet average emissions at the end of the model year will thus be based on the production-weighted average emissions of each model in its fleet.

The final footprint-based standards are very similar in shape to those proposed. NHTSA and EPA include more discussion of the development of the final curves in Section II below, with a full discussion in the Joint TSD. In addition, a full discussion of the equations and coefficients that define the curves is included in Section III for the CO₂ curves and Section IV for the mpg curves. The following figures illustrate the standards. First, Figure I.B.3-1 shows the fuel economy (mpg) car standard curve.

Under an attribute-based standard, every vehicle model has a performance target (fuel economy for the CAFE standards, and CO₂ g/mile for the GHG emissions standards), the level of which depends on the vehicle's attribute (for this rule, footprint). The manufacturers' fleet average performance is determined by the production-weighted average (for CAFE, harmonic average) of those targets. NHTSA and EPA are setting CAFE and CO₂ emissions standards defined by constrained linear functions and, equivalently, piecewise linear functions. As a possible option for future rulemakings, the constrained linear form was introduced by NHTSA in the 2007 NPRM proposing CAFE standards for MY 2011-2015.

NHTSA is establishing the attribute curves below for assigning a fuel economy level to an individual vehicle's footprint value, for model years 2012 through 2016. These mpg values will be production weighted to determine each manufacturer's fleet average standard for cars and trucks. Although the general model of the equation is the same for each vehicle category and each year, the parameters of the equation differ for cars and trucks. Each parameter also changes on an annual basis, resulting in the yearly increases in stringency. Figure I.B.3-1 below illustrates the passenger car CAFE standard curves for model years 2012 through 2016 while Figure I.B.3-2 below illustrates the light truck standard curves for model years 2012-2016. The MY 2011 final standards for cars and trucks, which are specified by a constrained logistic function rather than a constrained linear function, are shown for comparison.

EPA is establishing the attribute curves below for assigning a CO₂ level to an individual vehicle's footprint value, for model years 2012 through 2016. These CO₂ values will be production weighted to determine each manufacturer's fleet average standard for cars and trucks. As with the CAFE curves above, the general form of the equation is the same for each vehicle category and each year, but the parameters of the equation differ for cars and trucks. Again, each parameter also changes on an annual basis, resulting in the yearly increases in stringency. Figure I.B.3-3 below illustrates the CO₂ car standard curves for model years 2012 through 2016 while Figure I.B.3-4 shows the CO₂ truck standard curves for model years 2012-2016.

NHTSA and EPA received a number of comments about the shape of the car and truck curves. We address these comments further in Section II.C below as well as in Sections III and IV.

As proposed, NHTSA and EPA will use the same vehicle category definitions for determining which vehicles are subject to the car curve standards versus the truck curve standards. In other words, a vehicle classified as a car under the NHTSA CAFE program will also be classified as a car under the EPA GHG program, and likewise for trucks. Auto industry commenters generally agreed with this approach and believe it is an important aspect of harmonization across the two agencies' programs. Some other commenters expressed concern about potential consequences, especially in how cars and trucks are distinguished. However, EPA and NHTSA are employing the same car and truck definitions for the MY 2012-2016 CAFE and GHG standards as those used in the CAFE program for the 2011 model year standards. This issue is further discussed for the EPA standards in Section III, and for the NHTSA standards in Section IV. This approach of using CAFE definitions allows EPA's CO₂ standards and the CAFE standards to be harmonized across all vehicles for this program. However, EPA is not changing the car/truck definition for the purposes of any other previous rules.

Generally speaking, a smaller footprint vehicle will have higher fuel economy and lower CO₂ emissions relative to a larger footprint vehicle when both have the same degree of fuel efficiency improvement technology. In this final rule, the standards apply to a manufacturers overall fleet, not an individual vehicle, thus a manufacturers fleet which is dominated by small footprint vehicles will have a higher fuel economy requirement (lower CO₂ requirement) than a manufacturer whose fleet is dominated by large footprint vehicles. A footprint-based CO₂ or CAFE standard can be relatively neutral with respect to vehicle size and consumer choice. All vehicles, whether smaller or larger, must make improvements to reduce CO₂ emissions or improve fuel economy, and therefore all vehicles will be relatively more expensive. With the footprint-based standard approach, EPA and NHTSA believe there should be no significant effect on the relative distribution of different vehicle sizes in the fleet, which means that consumers will still be able to purchase the size of vehicle that meets their needs. While targets are manufacturer specific, rather than vehicle specific, Table I.B.3-1 illustrates the fact that different vehicle sizes will have varying CO₂ emissions and fuel economy targets under the final standards.

4. Program Flexibilities

EPA's and NHTSA's programs as established in this rule provide compliance flexibility to manufacturers, especially in the early years of the National Program. This flexibility is expected to provide sufficient lead time for manufacturers to make necessary technological improvements and reduce the overall cost of the program, without compromising overall environmental and fuel economy objectives. The broad goal of harmonizing the two agencies' standards includes preserving manufacturers' flexibilities in meeting the standards, to the extent appropriate and required by law. The following section provides an overview of this final rule's flexibility provisions. Many auto manufacturers commented in support of these provisions as critical to meeting the standards in the lead time provided. Environmental groups, some States, and others raised concerns about the possibility for windfall credits and loss of program benefits. The provisions in the final rule are in most cases the same as those proposed. However consideration of the issues raised by commenters has led to modifications in certain provisions. These comments and the agencies' response are discussed in Sections III and IV below and in the Response to Comments document.

a. CO₂/CAFE Credits Generated Based on Fleet Average Performance

Under this NHTSA and EPA final rule, the fleet average standards that apply to a manufacturer's car and truck fleets are based on the applicable footprint-based curves. At the end of each model year, when production of the model year is complete, a production-weighted fleet average will be calculated for each averaging set (cars and trucks). Under this approach, a manufacturer's car and/or truck fleet that achieves a fleet average CO₂/CAFE level better than the standard can generate credits. Conversely, if the fleet average CO₂/CAFE level does not meet the standard, the fleet would incur debits (also referred to as a shortfall).

Under the final program, a manufacturer whose fleet generates credits in a given model year would have several options for using those credits, including credit carry-back, credit carry-forward, credit transfers, and credit trading. These provisions exist in the MY 2011 CAFE program under EPCA and EISA, and similar provisions are part of EPA's Tier 2 program for light-duty vehicle criteria pollutant emissions, as well as many other mobile source standards issued by EPA under the CAA. The manufacturer will be able to carry back credits to offset a deficit that had accrued in a prior model year and was subsequently carried over to the current model year. EPCA also provides for this. EPCA restricts the carry-back of CAFE credits to three years, and as proposed EPA is establishing the same limitation, in keeping with the goal of harmonizing both sets of standards.

After satisfying any need to offset pre-existing deficits, remaining credits can be saved (banked) for use in future years. Under the CAFE program, EISA allows manufacturers to apply credits earned in a model year to compliance in any of the five subsequent model years. As proposed, under the GHG program, EPA is also allowing manufacturers to use these banked credits in the five years after the year in which they were generated (i.e., five years carry-forward).

EISA required NHTSA to establish by regulation a CAFE credits transferring program, which NHTSA established in a March 2009 final rule codified at 49 CFR Part 536, to allow a manufacturer to transfer credits between its vehicle fleets to achieve compliance with the standards. For example, credits earned by over-compliance with a manufacturer's car fleet average standard could be used to offset debits incurred due to that manufacturer's not meeting the truck fleet average standard in a given year. EPA's Tier 2 program also provides for this type of credit transfer. As proposed for purposes of this rule, EPA allows unlimited credit transfers across a manufacturer's car-truck fleet to meet the GHG standard. This is based on the expectation that this flexibility will facilitate manufacturers' ability to comply with the GHG standards in the lead time provided, and will allow the required GHG emissions reductions to be achieved in the most cost effective way. Under the CAA, unlike under EISA, there is no statutory limitation on car-truck credit transfers. Therefore, EPA is not constraining car-truck credit transfers, as doing so would reduce the flexibility for lead time, and would increase costs with no corresponding environmental benefit. For the CAFE program, however, EISA limits the amount of credits that may be transferred, which has the effects of limiting the extent to which a manufacturer can rely upon credits in lieu of making fuel economy improvements to a particular portion of its vehicle fleet, but also of potentially increasing the costs of improving the manufacturer's overall fleet. EISA also prohibits the use of transferred credits to meet the statutory minimum level for the domestic car fleet standard. These and other statutory limits will continue to apply to the determination of compliance with the CAFE standards.

EISA also allowed NHTSA to establish by regulation a CAFE credit trading program, which NHTSA established in the March 2009 final rule at 40 CFR part 536, to allow credits to be traded (sold) to other vehicle manufacturers. As proposed, EPA allows credit trading in the GHG program. These sorts of exchanges are typically allowed under EPA's current mobile source emission credit programs, although manufacturers have seldom made such exchanges. Under the NHTSA CAFE program, EPCA also allows these types of credit trades, although, as with transferred credits, traded credits may not be used to meet the minimum domestic car standards specified by statute. Comments discussing these provisions supported the proposed approach. These final provisions are the same as proposed.

As further discussed in Section IV of this preamble, NHTSA sought to find a way to provide credits for improving the efficiency of light truck air conditioners (A/Cs) and solicited public comments to that end. The agency did so because the power necessary to operate an A/C compressor places a significant additional load on the engine, thus reducing fuel economy and increasing CO₂ tailpipe emissions. See Section III.C.1 below. The agency would have made a similar effort regarding cars, but a 1975 statutory provision made it unfruitful even to explore the possibility of administratively proving such credits for cars. The agency did not identify a workable way of providing such credits for light trucks in the context of this rulemaking.

b. Air Conditioning Credits Under the EPA Final Rule

Air conditioning (A/C) systems contribute to GHG emissions in two ways. Hydrofluorocarbon (HFC) refrigerants, which are powerful GHGs, can leak from the A/C system (direct A/C emissions). As just noted, operation of the A/C system also places an additional load on the engine, which results in additional CO₂ tailpipe emissions (indirect A/C related emissions). EPA is allowing manufacturers to generate credits by reducing either or both types of GHG emissions related to A/C systems. Specifically, EPA is establishing a method to calculate CO₂ equivalent reductions for the vehicle's full useful life on a grams/mile basis that can be used as credits in meeting the fleet average CO₂ standards. EPA's analysis indicates that this approach provides manufacturers with a highly cost-effective way to achieve a portion of GHG emissions reductions under the EPA program. EPA is estimating that manufacturers will on average generate 11 g/mi GHG credit toward meeting the 250 g/mi by 2016 (though some companies may generate more). EPA will also allow manufacturers to earn early A/C credits starting in MY 2009 through 2011, as discussed further in a later section. There were many comments on the proposed A/C provisions. Nearly every one of these was supportive of EPA including A/C control as part of this rule, though there was some disagreement on some of the details of the program. The HFC crediting scheme was widely supported. The comments mainly were concentrated on indirect A/C related credits. The auto manufacturers and suppliers had some technical comments on A/C technologies, and there were many concerns with the proposed idle test. EPA has made some minor adjustments in both of these areas that we believe are responsive to these concerns. EPA addresses A/C issues in greater detail in Section III of this preamble and in Chapter 2 of EPA's RIA.

c. Flexible-Fuel and Alternative Fuel Vehicle Credits

EPCA authorizes a compliance flexibility incentive under the CAFE program for production of dual-fueled or flexible-fuel vehicles (FFV) and dedicated alternative fuel vehicles. FFVs are vehicles that can run both on an alternative fuel and conventional fuel. Most FFVs are E85 capable vehicles, which can run on either gasoline or a mixture of up to 85 percent ethanol and 15 percent gasoline (E85). Dedicated alternative fuel vehicles are vehicles that run exclusively on an alternative fuel. EPCA was amended by EISA to extend the period of availability of the FFV incentive, but to begin phasing it out by annually reducing the amount of FFV incentive that can be used toward compliance with the CAFE standards. Although NHTSA expressed concern about the non-use of alternative fuel by FFVs in a 2002 report to Congress (Effects of the Alternative Motor Fuels Act CAFE Incentives Policy), EISA does not premise the availability of the FFV credits on actual use of alternative fuel by an FFV vehicle. Under NHTSA's CAFE program, pursuant to EISA, no FFV credits will be available for CAFE compliance after MY 2019. For dedicated alternative fuel vehicles, there are no limits or phase-out of the credits. As required by the statute, NHTSA will continue to allow the use of FFV credits for purposes of compliance with the CAFE standards until the end of the EISA phase-out period.

For the GHG program, as proposed, EPA will allow FFV credits in line with EISA limits, but only during the period from MYs 2012 to 2015. After MY 2015, EPA will only allow FFV credits based on a manufacturer's demonstration that the alternative fuel is actually being used in the vehicles and based on the vehicle's actual performance. EPA discusses this in more detail in Section III.C of the preamble, including a summary of key comments. These provisions are being finalized as proposed, with further discussion in Section III.C of how manufacturers can demonstrate that the alternative fuel is being used.

d. Temporary Lead-Time Allowance Alternative Standards Under the EPA Final Rule

Manufacturers with limited product lines may be especially challenged in the early years of the National Program, and need additional lead time. Manufacturers with narrow product offerings may not be able to take full advantage of averaging or other program flexibilities due to the limited scope of the types of vehicles they sell. For example, some smaller volume manufacturer fleets consist entirely of vehicles with very high baseline CO₂ emissions. Their vehicles are above the CO₂ emissions target for that vehicle footprint, but do not have other types of vehicles in their production mix with which to average. Often, these manufacturers pay fines under the CAFE program rather than meet the applicable CAFE standard. EPA believes that these technological circumstances call for more lead time in the form of a more gradual phase-in of standards.

EPA is finalizing a temporary lead-time allowance for manufacturers that sell vehicles in the U.S. in MY 2009 and for which U.S. vehicle sales in that model year are below 400,000 vehicles. This allowance will be available only during the MY 2012-2015 phase-in years of the program. A manufacturer that satisfies the threshold criteria will be able to treat a limited number of vehicles as a separate averaging fleet, which will be subject to a less stringent GHG standard. Specifically, a standard of 25 percent above the vehicle's otherwise applicable foot-print target level will apply to up to 100,000 vehicles total, spread over the four year period of MY 2012 through 2015. Thus, the number of vehicles to which the flexibility could apply is limited. EPA also is setting appropriate restrictions on credit use for these vehicles, as discussed further in Section III. By MY 2016, these allowance vehicles must be averaged into the manufacturer's full fleet (i.e., they will no longer be eligible for a different standard). EPA discusses this in more detail in Section III.B of the preamble.

EPA received comments from several smaller manufacturers that the TLAAS program was insufficient to allow manufacturers with very limited product lines to comply. These manufacturers commented that they need additional lead time to meet the standards, because their CO₂ baselines are significantly higher and their vehicle product lines are even more limited, reducing their ability to average across their fleets compared even to other TLAAS manufacturers. EPA fully summarizes the public comments on the TLAAS program, including comments not supporting the program, in Section III.B. In summary, in response to the lead time issues raised by manufacturers, EPA is modifying the TLAAS program that applies to manufacturers with between 5,000 and 50,000 U.S. vehicle sales in MY 2009. EPA believes these provisions are necessary given that, compared with other TLAAS manufacturers, these manufacturers have even more limited product offerings across which to average and higher baseline CO₂ emissions, and thus need additional lead-time to meet the standards. These manufacturers would have an increased allotment of vehicles, a total of 250,000, compared to 100,000 vehicles (for other TLAAS-eligible manufacturers). In addition, the TLAAS program for these manufacturers would be extended by one year, through MY 2016 for these vehicles, for a total of five years of eligibility. The other provisions of the TLAAS program would continue to apply, such as the restrictions on credit trading and the level of the standard. Additional restrictions would also apply to these vehicles, as discussed in Section III. In addition, for the smallest volume manufacturers, those with below 5,000 U.S. vehicle sales, EPA is not setting standards at this time but is instead deferring standards until a future rulemaking. This is essentially the same approach we are using for small businesses, which are exempted from this rule. The unique issues involved with these manufacturers will be addressed in that future rulemaking. Further discussion of the public comment on these issues and details on these changes from the proposed program are included in Section III.

e. Additional Credit Opportunities Under the Clean Air Act (CAA)

EPA is establishing additional opportunities for early credits in MYs 2009-2011 through over-compliance with a baseline standard. The baseline standard is set to be equivalent, on a national level, to the California standards. Credits can be generated by over-compliance with this baseline in one of two ways—over-compliance by the fleet of vehicles sold in California and the CAA section 177 States (i.e., those States adopting the California program), or over-compliance with the fleet of vehicles sold in the 50 States. EPA is also providing for early credits based on over-compliance with CAFE, but only for vehicles sold in States outside of California and the CAA section 177 states. Under the early credit provisions, no early FFV credits would be allowed, except those achieved by over-compliance with the California program based on California's provisions that manufacturers demonstrate actual use of the alternative fuel. EPA's early credits provisions are designed to ensure that there would be no double counting of early credits. NHTSA notes, however, that credits for overcompliance with CAFE standards during MYs 2009-2011 will still be available for manufacturers to use toward compliance in future model years, just as before.

EPA received comments from some environmental organizations and States expressing concern that these early credits were inappropriate windfall credits because they provided credits for actions that were not surplus, that is above what would otherwise be required for compliance with either State or Federal motor vehicle standards. This focused on the credits for over-compliance with the California standards generated during model years 2009 and perhaps 2010, where according to commenters the CAFE requirements were in effect more stringent than the California standards. EPA believes that early credits provide a valuable incentive for manufacturers that have implemented fuel efficient technologies in excess of their CAFE compliance obligations prior to MY 2012. With appropriate restrictions, these credits, reflecting over-compliance over a three model year time frame (MY 2009-2011) and not just over one or two model years, will be surplus reductions and not otherwise required by law. Therefore, EPA is finalizing these provisions largely as proposed, but in response to comments, with an additional restriction on the trading of MY 2009 credits. The overall structure of this early credit program addresses concerns about the potential for windfall credits in the first one or two model years. This issue is fully discussed in Section III.C.

EPA is providing an additional temporary incentive to encourage the commercialization of advanced GHG/fuel economy control technologies—including electric vehicles (EVs), plug-in hybrid electric vehicles (PHEVs), and fuel cell vehicles (FCVs)—for model years 2012-2016. EPA's proposal included an emissions compliance value of zero grams/mile for EVs and FCVs, and the electric portion of PHEVs, and a multiplier in the range of 1.2 to 2.0, so that each advanced technology vehicle would count as greater than one vehicle in a manufacturer's fleetwide compliance calculation. EPA received many comments on the proposed incentives. Many State and environmental organization commenters believed that the combination of these incentives could undermine the GHG benefits of the rule, and believed the emissions compliance values should take into account the net upstream GHG emissions associated with electrified vehicles compared to vehicles powered by petroleum based fuel. Auto manufacturers generally supported the incentives, some believing the incentives to be a critical part of the National Program. Most auto makers supported both the zero grams/mile emissions compliance value and the higher multipliers.

Upon considering the public comments on this issue, EPA is finalizing an advanced technology vehicle incentive program that includes a zero gram/mile emissions compliance value for EVs and FCVs, and the electric portion of PHEVs, for up to the first 200,000 EV/PHEV/FCV vehicles produced by a given manufacturer during MY 2012-2016 (for a manufacturer that produces less than 25,000 EVs, PHEVs, and FCVs in MY 2012), or for up to the first 300,000 EV/PHEV/FCV vehicles produced during MY 2012-2016 (for a manufacturer that produces 25,000 or more EVs, PHEVs, and FCVs in MY 2012). For any production greater than this amount, the compliance value for the vehicle will be greater than zero gram/mile, set at a level that reflects the vehicle's net increase in upstream GHG emissions in comparison to the gasoline vehicle it replaces. In addition, EPA is not finalizing a multiplier. EPA will also allow this early advanced technology incentive program beginning in MYs 2009-2011. The purpose of these provisions is to provide a temporary incentive to promote technologies which have the potential to produce very large GHG reductions in the future. The tailpipe GHG emissions from EVs, FCVs, and PHEVs operated on grid electricity are zero, and traditionally the emissions of the vehicle itself are all that EPA takes into account for purposes of compliance with standards set under section 202(a). This has not raised any issues for criteria pollutants, as upstream emissions associated with production and distribution of the fuel are addressed by comprehensive regulatory programs focused on the upstream sources of those emissions. At this time, however, there is no such comprehensive program addressing upstream emissions of GHGs, and the upstream GHG emissions associated with production and distribution of electricity are higher than the corresponding upstream GHG emissions of gasoline or other petroleum based fuels. In the future, vehicle fleet electrification combined with advances in low-carbon technology in the electricity sector have the potential to transform the transportation sector's contribution to the country's GHG emissions. EPA will reassess the issue of how to address EVs, PHEVs, and FCVs in rulemakings for model years 2017 and beyond, based on the status of advanced vehicle technology commercialization, the status of upstream GHG control programs, and other relevant factors. Further discussion of the temporary advanced technology vehicle incentives, including more detail on the public comments and EPA's response, is found in Section III.C.

EPA is also providing an option for manufacturers to generate credits for employing new and innovative technologies that achieve GHG reductions that are not reflected on current test procedures, as proposed. Examples of such “off-cycle” technologies might include solar panels on hybrids, adaptive cruise control, and active aerodynamics, among other technologies. These three credit provisions are discussed in more detail in Section III.

5. Coordinated Compliance

Previous NHTSA and EPA regulations and statutory provisions establish ample examples on which to develop an effective compliance program that achieves the energy and environmental benefits from CAFE and motor vehicle GHG standards. NHTSA and EPA have developed a program that recognizes, and replicates as closely as possible, the compliance protocols associated with the existing CAA Tier 2 vehicle emission standards, and with CAFE standards. The certification, testing, reporting, and associated compliance activities closely track current practices and are thus familiar to manufacturers. EPA already oversees testing, collects and processes test data, and performs calculations to determine compliance with both CAFE and CAA standards. Under this coordinated approach, the compliance mechanisms for both programs are consistent and non-duplicative. EPA will also apply the CAA authorities applicable to its separate in-use requirements in this program.

The compliance approach allows manufacturers to satisfy the new program requirements in the same general way they comply with existing applicable CAA and CAFE requirements. Manufacturers would demonstrate compliance on a fleet-average basis at the end of each model year, allowing model-level testing to continue throughout the year as is the current practice for CAFE determinations. The compliance program design establishes a single set of manufacturer reporting requirements and relies on a single set of underlying data. This approach still allows each agency to assess compliance with its respective program under its respective statutory authority.

NHTSA and EPA do not anticipate any significant noncompliance under the National Program. However, failure to meet the fleet average standards (after credit opportunities are exhausted) would ultimately result in the potential for penalties under both EPCA and the CAA. The CAA allows EPA considerable discretion in assessment of penalties. Penalties under the CAA are typically determined on a vehicle-specific basis by determining the number of a manufacturer's highest emitting vehicles that caused the fleet average standard violation. This is the same mechanism used for EPA's National Low Emission Vehicle and Tier 2 corporate average standards, and to date there have been no instances of noncompliance. CAFE penalties are specified by EPCA and would be assessed for the entire noncomplying fleet at a rate of $5.50 times the number of vehicles in the fleet, times the number of tenths of mpg by which the fleet average falls below the standard. In the event of a compliance action arising out of the same facts and circumstances, EPA could consider CAFE penalties when determining appropriate remedies for the EPA case.

Several stakeholders commented on the proposed coordinated compliance approach. The comments indicated broad support for the overall approach EPA proposed. In particular, both regulated industry and the public interest community appreciated the attempt to streamline compliance by adopting current practice where possible and by coordinating EPA and NHTSA compliance requirements. Thus the final compliance program design is largely unchanged from the proposal. Some commenters requested additional detail or clarification in certain areas and others suggested some relatively narrow technical changes, and EPA has responded to these suggestions. EPA and NHTSA summarize these comments and the agencies' responses in Sections III and IV, respectively, below. The Response to Comments document associated with this document includes all of the comments and responses received during the comment period.

C. Summary of Costs and Benefits of the National Program

This section summarizes the projected costs and benefits of the CAFE and GHG emissions standards. These projections helped inform the agencies' choices among the alternatives considered and provide further confirmation that the final standards are an appropriate choice within the spectrum of choices allowable under their respective statutory criteria. The costs and benefits projected by NHTSA to result from these CAFE standards are presented first, followed by those from EPA's analysis of the GHG emissions standards.

For several reasons, the estimates for costs and benefits presented by NHTSA and EPA, while consistent, are not directly comparable, and thus should not be expected to be identical. Most important, NHTSA and EPA's standards would require slightly different fuel efficiency improvements. EPA's GHG standard is more stringent in part due to its assumptions about manufacturers' use of air conditioning credits, which result from reductions in air conditioning-related emissions of HFCs and CO₂. NHTSA was unable to make assumptions about manufacturers' improving the efficiency of air conditioners due to statutory limitations. In addition, the CAFE and GHG standards offer different program flexibilities, and the agencies' analyses differ in their accounting for these flexibilities (for example, FFVs), primarily because NHTSA is statutorily prohibited from considering some flexibilities when establishing CAFE standards, while EPA is not. These differences contribute to differences in the agencies' respective estimates of costs and benefits resulting from the new standards.

NHTSA performed two analyses: a primary analysis that shows the estimates of costs, fuel savings, and related benefits that the agency considered for purposes of establishing new CAFE standards, and a supplemental analysis that reflects the agency's best estimate of the potential real-world effects of the CAFE standards, including manufacturers' potential use of FFV credits in accordance with the provisions of EISA concerning their availability. Because EPCA prohibits NHTSA from considering the ability of manufacturers to use of FFV credits to increase their fleet average fuel economy when establishing CAFE standards, the agency's primary analysis does not include them. However, EPCA does not prohibit NHTSA from considering the fact that manufacturers may pay civil penalties rather than complying with CAFE standards, and NHTSA's primary analysis accounts for some manufacturers' tendency to do so. In addition, NHTSA's supplemental analysis of the effect of FFV credits on benefits and costs from its CAFE standards, demonstrates the real-world impacts of FFVs, and the summary estimates presented in Section IV include these effects. Including the use of FFV credits reduces estimated per-vehicle compliance costs of the program. However, as shown below, including FFV credits does not significantly change the projected fuel savings and CO₂ reductions, because FFV credits reduce the fuel economy levels that manufacturers achieve not only under the standards, but also under the baseline MY 2011 CAFE standards.

Also, EPCA, as amended by EISA, allows manufacturers to transfer credits between their passenger car and light truck fleets. However, EPCA also prohibits NHTSA from considering manufacturers' ability to increase their average fuel economy through the use of CAFE credits when determining the stringency of the CAFE standards. Because of this prohibition, NHTSA's primary analysis does not account for the extent to which credit transfers might actually occur. For purposes of its supplemental analysis, NHTSA considered accounting for the possibility that some manufacturers might utilize the opportunity under EPCA to transfer some CAFE credits between the passenger car and light truck fleets, but determined that in NHTSA's year-by-year analysis, manufacturers' credit transfers cannot be reasonably estimated at this time.

EPA made explicit assumptions about manufacturers' use of FFV credits under both the baseline and control alternatives, and its estimates of costs and benefits from the GHG standards reflect these assumptions. However, under the GHG standards, FFV credits would be available through MY 2015; starting in MY 2016, EPA will only allow FFV credits based on a manufacturer's demonstration that the alternative fuel is actually being used in the vehicles and the actual GHG performance for the vehicle run on that alternative fuel.

EPA's analysis also assumes that manufacturers would transfer credits between their car and truck fleets in the MY 2011 baseline subject to the maximum value allowed by EPCA, and that unlimited car-truck credit transfers would occur under the GHG standards. Including these assumptions in EPA's analysis increases the resulting estimates of fuel savings and reductions in GHG emissions, while reducing EPA's estimates of program compliance costs.

Finally, under the EPA GHG program, there is no ability for a manufacturer to intentionally pay fines in lieu of meeting the standard. Under EPCA, however, vehicle manufacturers are allowed to pay fines as an alternative to compliance with applicable CAFE standards. NHTSA's analysis explicitly estimates the level of voluntary fine payment by individual manufacturers, which reduces NHTSA's estimates of both the costs and benefits of its CAFE standards. In contrast, the CAA does not allow for fine payment (civil penalties) in lieu of compliance with emission standards, and EPA's analysis of benefits from its standard thus assumes full compliance. This assumption results in higher estimates of fuel savings, of reductions in GHG emissions, and of manufacturers' compliance costs to sell fleets that comply with both NHTSA's CAFE program and EPA's GHG program.

In summary, the projected costs and benefits presented by NHTSA and EPA are not directly comparable, because the GHG emission levels established by EPA include air conditioning-related improvements in equivalent fuel efficiency and HFC reductions, because of the assumptions incorporated in EPA's analysis regarding car-truck credit transfers, and because of EPA's projection of complete compliance with the GHG standards. It should also be expected that overall, EPA's estimates of GHG reductions and fuel savings achieved by the GHG standards will be slightly higher than those projected by NHTSA only for the CAFE standards because of the reasons described above. For the same reasons, EPA's estimates of manufacturers' costs for complying with the passenger car and light trucks GHG standards are slightly higher than NHTSA's estimates for complying with the CAFE standards.

A number of stakeholders commented on NHTSA's and EPA's analytical assumptions in estimating costs and benefits of the program. These comments and any changes from the proposed values are summarized in Section II.F, and further in Sections III (for EPA) and IV (for NHTSA); the Response to Comments document presents the detailed responses to each of the comments.

1. Summary of Costs and Benefits of NHTSA's CAFE Standards

NHTSA has analyzed in detail the costs and benefits of the final CAFE standards. Table I.C.1-1 presents the total costs, benefits, and net benefits for NHTSA's final CAFE standards. The values in Table I.C.1-1 display the total costs for all MY 2012-2016 vehicles and the benefits and net benefits represent the impacts of the standards over the full lifetime of the vehicles projected to be sold during model years 2012-2016. It is important to note that there is significant overlap in costs and benefits for NHTSA's CAFE program and EPA's GHG program and therefore combined program costs and benefits, which together comprise the National Program, are not a sum of the two individual programs.

NHTSA estimates that these new CAFE standards will lead to fuel savings totaling 61 billion gallons throughout the useful lives of vehicles sold in MYs 2012-2016. At a 3% discount rate, the present value of the economic benefits resulting from those fuel savings is $143 billion. At a 7% discount rate, the present value of the economic benefits resulting from those fuel savings is $112 billion.

The agency further estimates that these new CAFE standards will lead to corresponding reductions in CO₂ emissions totaling 655 million metric tons (mmt) during the useful lives of vehicles sold in MYs 2012-2016. The present value of the economic benefits from avoiding those emissions is $14.5 billion, based on a global social cost of carbon value of approximately $21 per metric ton (in 2010, and growing thereafter). It is important to note that NHTSA's CAFE standards and EPA's GHG standards will both be in effect, and each will lead to increases in average fuel economy and CO₂ emissions reductions. The two agencies' standards together comprise the National Program, and this discussion of costs and benefits of NHTSA's CAFE standards does not change the fact that both the CAFE and GHG standards, jointly, are the source of the benefits and costs of the National Program.

Considering manufacturers' ability to earn credit toward compliance by selling FFVs, NHTSA estimates very little change in incremental fuel savings and avoided CO₂ emissions, assuming FFV credits would be used toward both the baseline and final standards:

NHTSA estimates that these fuel economy increases would produce other benefits both to drivers (e.g., reduced time spent refueling) and to the U.S. (e.g., reductions in the costs of petroleum imports beyond the direct savings from reduced oil purchases, as well as some disbenefits (e.g., increase traffic congestion) caused by drivers' tendency to travel more when the cost of driving declines (as it does when fuel economy increases). NHTSA has estimated the total monetary value to society of these benefits and disbenefits, and estimates that the standards will produce significant net benefits to society. Using a 3% discount rate, NHTSA estimates that the present value of these benefits would total more than $180 billion over the useful lives of vehicles sold during MYs 2012-2016. More discussion regarding monetized benefits can be found in Section IV of this notice and in NHTSA's Regulatory Impact Analysis. Note that the benefit calculation in Tables I.C.1-4 through 1-7 includes the benefits of reducing CO₂ emissions, but not the benefits of reducing other GHG emissions.

Using a 7% discount rate, NHTSA estimates that the present value of these benefits would total more than $145 billion over the same time period.

NHTSA estimates that FFV credits could reduce achieved benefits by about 3.8%:

NHTSA attributes most of these benefits—about $143 billion (at a 3% discount rate and excluding consideration of FFV credits), as noted above—to reductions in fuel consumption, valuing fuel (for societal purposes) at the future pre-tax prices projected in the Energy Information Administration's (AEO's) reference case forecast from the Annual Energy Outlook (AEO) 2010 Early Release. NHTSA's Final Regulatory Impact Analysis (FRIA) accompanying this rule presents a detailed analysis of specific benefits of the rule.

NHTSA estimates that the increases in technology application necessary to achieve the projected improvements in fuel economy will entail considerable monetary outlays. The agency estimates that incremental costs for achieving its standards—that is, outlays by vehicle manufacturers over and above those required to comply with the MY 2011 CAFE standards—will total about $52 billion (i.e., during MYs 2012-2016).

NHTSA estimates that use of FFV credits could significantly reduce these outlays:

The agency projects that manufacturers will recover most or all of these additional costs through higher selling prices for new cars and light trucks. To allow manufacturers to recover these increased outlays (and, to a much lesser extent, the civil penalties that some companies are expected to pay for noncompliance), the agency estimates that the standards would lead to increases in average new vehicle prices ranging from $457 per vehicle in MY 2012 to $985 per vehicle in MY 2016:

NHTSA estimates that use of FFV credits could significantly reduce these costs, especially in earlier model years:

NHTSA estimates, therefore, that the total benefits of these CAFE standards will be more than three times the magnitude of the corresponding costs. As a consequence, its standards would produce net benefits of $130.7 billion at a 3 percent discount rate (with FFV credits, $138.2 billion) or $94.5 billion at a 7 percent discount rate over the useful lives of vehicles sold during MYs 2012-2016.

2. Summary of Costs and Benefits of EPA's GHG Standards

EPA has analyzed in detail the costs and benefits of the final GHG standards. Table I.C.2-1 shows EPA's estimated lifetime discounted cost, benefits and net benefits for all vehicles projected to be sold in model years 2012-2016. It is important to note that there is significant overlap in costs and benefits for NHTSA's CAFE program and EPA's GHG program and therefore combined program costs and benefits are not a sum of the individual programs.

Table I.C.2-2 shows EPA's estimated lifetime fuel savings and CO₂ equivalent emission reductions for all vehicles sold in the model years 2012-2016. The values in Table I.C.2-2 are projected lifetime totals for each model year and are not discounted. As documented in EPA's Final RIA, the potential credit transfer between cars and trucks may change the distribution of the fuel savings and GHG emission impacts between cars and trucks. As discussed above with respect to NHTSA's CAFE standards, it is important to note that NHTSA's CAFE standards and EPA's GHG standards will both be in effect, and each will lead to increases in average fuel economy and reductions in CO₂ emissions. The two agencies' standards together comprise the National Program, and this discussion of costs and benefits of EPA's GHG standards does not change the fact that both the CAFE and GHG standards, jointly, are the source of the benefits and costs of the National Program.

Table I.C.2-3 shows EPA's estimated lifetime discounted benefits for all vehicles sold in model years 2012-2016. Although EPA estimated the benefits associated with four different values of a one ton GHG reduction ($5, $21, $35, $65), for the purposes of this overview presentation of estimated benefits EPA is showing the benefits associated with one of these marginal values, $21 per ton of CO₂, in 2007 dollars and 2010 emissions. Table I.C.2-3 presents benefits based on the $21 value. Section III.H presents the four marginal values used to estimate monetized benefits of GHG reductions and Section III.H presents the program benefits using each of the four marginal values, which represent only a partial accounting of total benefits due to omitted climate change impacts and other factors that are not readily monetized. The values in the table are discounted values for each model year of vehicles throughout their projected lifetimes. The benefits include all benefits considered by EPA such as fuel savings, GHG reductions, PM benefits, energy security and other externalities such as reduced refueling and accidents, congestion and noise. The lifetime discounted benefits are shown for one of four different social cost of carbon (SCC) values considered by EPA. The values in Table I.C.2-3 do not include costs associated with new technology required to meet the GHG standard.

Table I.C.2-4 shows EPA's estimated lifetime fuel savings, lifetime CO₂ emission reductions, and the monetized net present values of those fuel savings and CO₂ emission reductions. The gallons of fuel and CO₂ emission reductions are projected lifetime values for all vehicles sold in the model years 2012-2016. The estimated fuel savings in billions of barrels and the GHG reductions in million metric tons of CO₂ shown in Table I.C.2-4 are totals for the five model years throughout their projected lifetime and are not discounted. The monetized values shown in Table I.C.2-4 are the summed values of the discounted monetized-fuel savings and monetized-CO₂ reductions for the five model years 2012-2016 throughout their lifetimes. The monetized values in Table I.C.2-4 reflect both a 3 percent and a 7 percent discount rate as noted.

Table I.C.2-5 shows EPA's estimated incremental and total technology outlays for cars and trucks for each of the model years 2012-2016. The technology outlays shown in Table I.C.2-5 are for the industry as a whole and do not account for fuel savings associated with the program.

Table I.C.2-6 shows EPA's estimated incremental cost increase of the average new vehicle for each model year 2012-2016. The values shown are incremental to a baseline vehicle and are not cumulative. In other words, the estimated increase for 2012 model year cars is $342 relative to a 2012 model year car absent the National Program. The estimated increase for a 2013 model year car is $507 relative to a 2013 model year car absent the National Program (not $342 plus $507).

D. Background and Comparison of NHTSA and EPA Statutory Authority

Section I.C of the proposal contained a detailed overview discussion of the NHTSA and EPA statutory authorities. In addition to the discussion in the proposal, each agency discusses comments pertaining to its statutory authority and the agency's responses in Sections III and IV of this notice, respectively.

II. Joint Technical Work Completed for This Final Rule

A. Introduction

In this section NHTSA and EPA discuss several aspects of the joint technical analyses on which the two agencies collaborated. These analyses are common to the development of each agency's final standards. Specifically we discuss: the development of the vehicle market forecast used by each agency for assessing costs, benefits, and effects, the development of the attribute-based standard curve shapes, the determination of the relative stringency between the car and truck fleet standards, the technologies the agencies evaluated and their costs and effectiveness, and the economic assumptions the agencies included in their analyses. The Joint Technical Support Document (TSD) discusses the agencies' joint technical work in more detail.

B. Developing the Future Fleet for Assessing Costs, Benefits, and Effects

1. Why did the agencies establish a baseline and reference vehicle fleet?

In order to calculate the impacts of the EPA and NHTSA regulations, it is necessary to estimate the composition of the future vehicle fleet absent these regulations, to provide a reference point relative to which costs, benefits, and effects of the regulations are assessed. As in the proposal, EPA and NHTSA have developed this comparison fleet in two parts. The first step was to develop a baseline fleet based on model year 2008 data. The second step was to project that fleet into model years 2011-2016. This is called the reference fleet. The third step was to modify that MY 2011-2016 reference fleet such that it had sufficient technology to meet the MY 2011 CAFE standards. This final version of the reference fleet is the light-duty fleet estimated to exist in MY 2012-2016 in the absence of today's standards, based on the assumption that manufacturers would continue to meet the MY 2011 CAFE standards (or pay civil penalties allowed under EPCA ) in the absence of further increases in the stringency of CAFE standards. Each agency used this approach to develop a final reference fleet to use in its modeling. All of the agencies' estimates of emission reductions, fuel economy improvements, costs, and societal impacts are developed in relation to the respective reference fleets.

EPA and NHTSA proposed a transparent approach to developing the baseline and reference fleets, largely working from publicly available data. This proposed approach differed from previous CAFE rules, which relied on confidential manufacturers' product plan information to develop the baseline. Most of the public comments to the NPRM addressing this issue supported this methodology for developing the inputs to the rule's analysis. Because the input sheets can be made public, stakeholders can verify and check EPA's and NHTSA's modeling, and perform their own analyses with these datasets. In this final rulemaking, EPA and NHTSA are using an approach very similar to that proposed, continuing to rely on publicly available data as the basis for the baseline and reference fleets.

2. How did the agencies develop the baseline vehicle fleet?

At proposal, EPA and NHTSA developed a baseline fleet comprised of model year 2008 data gathered from EPA's emission certification and fuel economy database. MY 2008 was used as the basis for the baseline vehicle fleet because it was the most recent model year for which a complete set of data is publicly available. This remains the case. Manufacturers are not required to submit final sales and mpg figures for MY 2009 until April 2010, after the CAFE standard's mandated promulgation date. Consequently, in this final rule, EPA and NHTSA made no changes to the method or the results of the MY 2008 baseline fleet used at proposal, except for some specific corrections to engineering inputs for some vehicle models reflected in the market forecast input to NHTSA's CAFE model. More details about how the agencies constructed this baseline fleet can be found in Chapter 1.2 of the Joint TSD. Corrections to engineering inputs for some vehicle models in the market forecast input to NHTSA's CAFE model are discussed in Chapter 2 of the Joint TSD.

3. How did the agencies develop the projected MY 2011-2016 vehicle fleet?

EPA and NHTSA have based the projection of total car and total light truck sales for MYs 2011-2016 on projections made by the Department of Energy's Energy Information Administration (EIA). EIA publishes a mid-term projection of national energy use called the Annual Energy Outlook (AEO). This projection utilizes a number of technical and econometric models which are designed to reflect both economic and regulatory conditions expected to exist in the future. In support of its projection of fuel use by light-duty vehicles, EIA projects sales of new cars and light trucks. In the proposal, the agencies used the three reports published by EIA as part of the AEO 2009. We also stated that updated versions of these reports could be used in the final rules should AEO timely issue a new version. EIA published an early version of its AEO 2010 in December 2009, and the agencies are making use of it in this final rulemaking. The differences in projected sales in the 2009 report (used in the NPRM) and the early 2010 report are very small, so NHTSA and EPA have decided to simply scale the NPRM volumes for cars and trucks (in the aggregate) to match those in the 2010 report. We thus employ the sales projections from the scaled updated 2009 Annual Energy Outlook, which is equivalent to AEO 2010 Early Release, for the final rule. The scaling factors for each model year are presented in Chapter 1 of the Joint TSD for this final rule.

The agencies recognize that AEO 2010 Early Release does include some impacts of future projected increases in CAFE stringency. We have closely examined the difference between AEO 2009 and AEO 2010 Early Release and we believe the differences in total sales and the car/truck split attributed to considerations of the standard in the final rule are small.

In the AEO 2010 Early Release, EIA projects that total light-duty vehicle sales will gradually recover from their currently depressed levels by around 2013. In 2016, car sales are projected to be 9.4 million (57 percent) and truck sales are projected to be 7.1 million (43 percent). Although the total level of sales of 16.5 million units is similar to pre-2008 levels, the fraction of car sales is projected to be higher than that existing in the 2000-2007 timeframe. This projection reflects the impact of higher fuel prices, as well as EISA's requirement that the new vehicle fleet average at least 35 mpg by MY 2020. The agencies note that AEO does not represent the fleet at a level of detail sufficient to explicitly account for the reclassification—promulgated as part of NHTSA's final rule for MY 2011 CAFE standards—of a number of 2-wheel drive sport utility vehicles from the truck fleet to the car fleet for MYs 2011 and after. Sales projections of cars and trucks for future model years can be found in the Joint TSD for these final rules.

In addition to a shift towards more car sales, sales of segments within both the car and truck markets have been changing and are expected to continue to change. Manufacturers are introducing more crossover models which offer much of the utility of SUVs but use more car-like designs. The AEO 2010 report does not, however, distinguish such changes within the car and truck classes. In order to reflect these changes in fleet makeup, EPA and NHTSA considered several other available forecasts. EPA purchased and shared with NHTSA forecasts from two well-known industry analysts, CSM Worldwide (CSM), and J.D. Powers. NHTSA and EPA decided to use the forecast from CSM, modified as described below, for several reasons presented in the NPRM preamble and draft Joint TSD. The changes between company market share and industry market segments were most significant from 2011-2014, while for 2014-2015 the changes were relatively small. Noting this, and lacking a credible forecast of company and segment shares after 2015, the agencies assumed 2016 market share and market segments to be the same as for 2015.

CSM Worldwide provides quarterly sales forecasts for the automotive industry. In the NPRM, the agencies identified a concern with the 2nd quarter CSM forecast that was used as a basis for the projection. CSM projections at that time were based on an industry that was going through a significant financial transition, and as a result the market share forecasts for some companies were impacted in surprising ways. As the industry's situation has settled somewhat over the past year, the 4th quarter projection appears to address this issue—for example, it shows nearly a two-fold increase in sales for Chrysler compared to significant loss of market share shown for Chrysler in the 2nd quarter projection. Additionally, some commenters, such as GM, recognized that the fleet appeared to include an unusually high number of large pickup trucks. In fact, the agencies discovered (independently of the comments) that CSM's standard forecast included all vehicles below 14,000 GVWR, including class 2b and 3 heavy duty vehicles, which are not regulated by this final rule. The commenters were thus correct that light duty reference fleet projections at proposal had more full size trucks and vans due to the mistaken inclusion of the heavy duty versions of those vehicles. The agencies requested a separate data forecast from CSM that filtered their 4th quarter projection to exclude these heavy duty vehicles. The agencies then used this filtered 4th quarter forecast for the final rule. A detailed comparison of the market by manufacturer can be found in the final TSD. For the public's reference, copies of the 2nd, 3rd, and 4th quarter CSM forecasts have been placed in the docket for this rulemaking.

We then projected the CSM forecasts for relative sales of cars and trucks by manufacturer and by market segment onto the total sales estimates of AEO 2010. Tables II.B.3-1 and II.B.3-2 show the resulting projections for the reference 2016 model year and compare these to actual sales that occurred in baseline 2008 model year. Both tables show sales using the traditional definition of cars and light trucks.

Determining which traditionally-defined trucks will be defined as cars for purposes of this final rule using the revised definition established by NHTSA for MYs 2011 and beyond requires more detailed information about each vehicle model. This is described in greater detail in Chapter 1 of the final TSD.

The forecasts obtained from CSM provided estimates of car and truck sales by segment and by manufacturer, but not by manufacturer for each market segment. Therefore, NHTSA and EPA needed other information on which to base these more detailed projected market splits. For this task, the agencies used as a starting point each manufacturer's sales by market segment from model year 2008, which is the baseline fleet. Because of the larger number of segments in the truck market, the agencies used slightly different methodologies for cars and trucks.

The first step for both cars and trucks was to break down each manufacturer's 2008 sales according to the market segment definitions used by CSM. For example, the agencies found that Ford's cars sales in 2008 were broken down as shown in Table II.B.3-3:

EPA and NHTSA then adjusted each manufacturer's sales of each of its car segments (and truck segments, separately) so that the manufacturer's total sales of cars (and trucks) matched the total estimated for each future model year based on AEO and CSM forecasts. For example, as indicated in Table II.B.3-1, Ford's total car sales in 2008 were 709,583 units, while the agencies project that they will increase to 1,113,333 units by 2016. This represents an increase of 56.9 percent. Thus, the agencies increased the 2008 sales of each Ford car segment by 56.9 percent. This produced estimates of future sales which matched total car and truck sales per AEO and the manufacturer breakdowns per CSM. However, the sales splits by market segment would not necessarily match those of CSM (shown for 2016 in Table II.B.3-2).

In order to adjust the market segment mix for cars, the agencies first adjusted sales of luxury, specialty and other cars. Since the total sales of cars for each manufacturer were already set, any changes in the sales of one car segment had to be compensated by the opposite change in another segment. For the luxury, specialty and other car segments, it is not clear how changes in sales would be compensated. For example, if luxury car sales decreased, would sales of full-size cars increase, mid-size cars, and so on? The agencies have assumed that any changes in the sales of cars within these three segments were compensated for by proportional changes in the sales of the other four car segments. For example, for 2016, the figures in Table II.B.3-2 indicate that luxury car sales in 2016 are 1,548,242 units. Luxury car sales are 1,048,341 units in 2008. However, after adjusting 2008 car sales by the change in total car sales for 2016 projected by EIA and a change in manufacturer market share per CSM, luxury car sales decreased to 1,523,171 units. Thus, overall for 2016, luxury car sales had to increase by 25,071 units or 6 percent. The agencies accordingly increased the luxury car sales by each manufacturer by this percentage. The absolute decrease in luxury car sales was spread across sales of full-size, mid-size, compact and subcompact cars in proportion to each manufacturer's sales in these segments in 2008. The same adjustment process was used for specialty cars and the “other cars” segment defined by CSM.

The agencies used a slightly different approach to adjust for changing sales of the remaining four car segments. Starting with full-size cars, the agencies again determined the overall percentage change that needed to occur in future year full-size car sales after 1) adjusting for total sales per AEO 2010, 2) adjusting for manufacturer sales mix per CSM and 3) adjusting the luxury, specialty and other car segments, in order to meet the segment sales mix per CSM. Sales of each manufacturer's large cars were adjusted by this percentage. However, instead of spreading this change over the remaining three segments, the agencies assigned the entire change to mid-size vehicles. The agencies did so because the CSM data followed the trend of increasing volumes of smaller cars while reducing volumes of larger cars. If a consumer had previously purchased a full-size car, we thought it unlikely that their next purchase would decrease by two size categories, down to a subcompact. It seemed more reasonable to project that they would drop one vehicle size category smaller. Thus, the change in each manufacturer's sales of full-size cars was matched by an opposite change (in absolute units sold) in mid-size cars.

The same process was then applied to mid-size cars, with the change in mid-size car sales being matched by an opposite change in compact car sales. This process was repeated one more time for compact car sales, with changes in sales in this segment being matched by the opposite change in the sales of subcompacts. The overall result was a projection of car sales for model years 2012-2016—the reference fleet—which matched the total sales projections of the AEO forecast and the manufacturer and segment splits of the CSM forecast. These sales splits can be found in Chapter 1 of the Joint TSD for this final rule.

As mentioned above, the agencies applied a slightly different process to truck sales, because the agencies could not confidently project how the change in sales from one segment preferentially went to or came from another particular segment. Some trend from larger vehicles to smaller vehicles would have been possible. However, the CSM forecasts indicated large changes in total sport utility vehicle, multi-activity vehicle and cross-over sales which could not be connected. Thus, the agencies applied an iterative, but straightforward process for adjusting 2008 truck sales to match the AEO and CSM forecasts.

The first three steps were exactly the same as for cars. EPA and NHTSA broke down each manufacturer's truck sales into the truck segments as defined by CSM. The agencies then adjusted all manufacturers' truck segment sales by the same factor so that total truck sales in each model year matched AEO projections for truck sales by model year. The agencies then adjusted each manufacturer's truck sales by segment proportionally so that each manufacturer's percentage of total truck sales matched that forecast by CSM. This again left the need to adjust truck sales by segment to match the CSM forecast for each model year.

In the fourth step, the agencies adjusted the sales of each truck segment by a common factor so that total sales for that segment matched the combination of the AEO and CSM forecasts. For example, projected sales of large pickups across all manufacturers were 1,286,184 units in 2016 after adjusting total sales to match AEO's forecast and adjusting each manufacturer's truck sales to match CSM's forecast for the breakdown of sales by manufacturer. Applying CSM's forecast of the large pickup segment of truck sales to AEO's total sales forecast indicated total large pickup sales of 1,379,036 units. Thus, we increased each manufacturer's sales of large pickups by 7 percent. The agencies applied the same type of adjustment to all the other truck segments at the same time. The result was a set of sales projections which matched AEO's total truck sales projection and CSM's market segment forecast. However, after this step, sales by manufacturer no longer met CSM's forecast. Thus, we repeated step three and adjusted each manufacturer's truck sales so that they met CSM's forecast. The sales of each truck segment (by manufacturer) were adjusted by the same factor. The resulting sales projection matched AEO's total truck sales projection and CSM's manufacturer forecast, but sales by market segment no longer met CSM's forecast. However, the difference between the sales projections after this fifth step was closer to CSM's market segment forecast than it was after step three. In other words, the sales projection was converging to the desired result. The agencies repeated these adjustments, matching manufacturer sales mix in one step and then market segment in the next a total of 19 times. At this point, we were able to match the market segment splits exactly and the manufacturer splits were within 0.1 percent of our goal, which is well within the needs of this analysis.

The next step in developing the reference fleets was to characterize the vehicles within each manufacturer-segment combination. In large part, this was based on the characterization of the specific vehicle models sold in 2008—i.e., the vehicles comprising the baseline fleet. EPA and NHTSA chose to base our estimates of detailed vehicle characteristics on 2008 sales for several reasons. One, these vehicle characteristics are not confidential and can thus be published here for careful review by interested parties. Two, because it is constructed beginning with actual sales data, this vehicle fleet is limited to vehicle models known to satisfy consumer demands in light of price, utility, performance, safety, and other vehicle attributes.

As noted above, the agencies gathered most of the information about the 2008 baseline vehicle fleet from EPA's emission certification and fuel economy database. The data obtained from this source included vehicle production volume, fuel economy, engine size, number of engine cylinders, transmission type, fuel type, etc. EPA's certification database does not include a detailed description of the types of fuel economy-improving/CO₂-reducing technologies considered in this final rule. Thus, the agencies augmented this description with publicly available data which includes more complete technology descriptions from Ward's Automotive Group. In a few instances when required vehicle information (such as vehicle footprint) was not available from these two sources, the agencies obtained this information from publicly accessible Internet sites such as Motortrend.com and Edmunds.com.

The projections of future car and truck sales described above apply to each manufacturer's sales by market segment. The EPA emissions certification sales data are available at a much finer level of detail, essentially vehicle configuration. As mentioned above, the agencies placed each vehicle in the EPA certification database into one of the CSM market segments. The agencies then totaled the sales by each manufacturer for each market segment. If the combination of AEO and CSM forecasts indicated an increase in a given manufacturer's sales of a particular market segment, then the sales of all the individual vehicle configurations were adjusted by the same factor. For example, if the Prius represented 30 percent of Toyota's sales of compact cars in 2008 and Toyota's sales of compact cars in 2016 was projected to double by 2016, then the sales of the Prius were doubled, and the Prius sales in 2016 remained 30 percent of Toyota's compact car sales.

The projection of average footprint for both cars and trucks remained virtually constant over the years covered by the final rulemaking. This occurrence is strictly a result of the CSM projections. There are a number of trends that occur in the CSM projections that caused the average footprint to remain constant. First, as the number of subcompacts increases, so do the number of 2-wheel drive crossover vehicles (that are regulated as cars). Second, truck volumes have many segment changes during the rulemaking time frame. There is no specific footprint related trend in any segment that can be linked to the unchanging footprint, but there is a trend that non-pickups' volumes will move from truck segments that are ladder frame to those that are unibody-type vehicles. A table of the footprint projections is available in the TSD as well as further discussion on this topic.

4. How was the development of the baseline and reference fleets for this Final Rule different from NHTSA's historical approach?

NHTSA has historically based its analysis of potential new CAFE standards on detailed product plans the agency has requested from manufacturers planning to produce light vehicles for sale in the United States. Although the agency has not attempted to compel manufacturers to submit such information, most major manufacturers and some smaller manufacturers have voluntarily provided it when requested.

The proposal discusses many of the advantages and disadvantages of the market forecast approach used by the agencies, including the agencies' interest in examining product plans as a check on the reference fleet developed by the agencies for this rulemaking. One of the primary reasons for the request for data in 2009 was to obtain permission from the manufacturers to make public their product plan information for model years 2010 and 2011. There are a number of reasons that this could be advantageous in the development of a reference fleet. First, some known changes to the fleet may not be captured by the approach of solely using publicly available information. For example, the agencies' current market forecast includes some vehicles for which manufacturers have announced plans for elimination or drastic production cuts such as the Chevrolet Trailblazer, the Chrysler PT Cruiser, the Chrysler Pacifica, the Dodge Magnum, the Ford Crown Victoria, the Mercury Sable, the Pontiac Grand Prix, the Pontiac G5 and the Saturn Vue. These vehicle models appear explicitly in market inputs to NHTSA's analysis, and are among those vehicle models included in the aggregated vehicle types appearing in market inputs to EPA's analysis. However, although the agencies recognize that these specific vehicles will be discontinued, we continue to include them in the market forecast because they are useful as a surrogate for successor vehicles that may appear in the rulemaking time frame to replace the discontinued vehicles in that market segment.

Second, the agencies' market forecast does not include some forthcoming vehicle models, such as the Chevrolet Volt, the Ford Fiesta and several publicly announced electric vehicles, including the announcements from Nissan regarding the Leaf. Nor does it include several MY 2009 or 2010 vehicles, such as the Honda Insight, the Hyundai Genesis and the Toyota Venza, as our starting point for defining specific vehicle models in the reference fleet was Model Year 2008. Additionally, the market forecast does not account for publicly announced technology introductions, such as Ford's EcoBoost system, whose product plans specify which vehicles and how many are planned to have this technology. Chrysler Group LLC has announced plans to offer small- and medium-sized cars using Fiat powertrains. Were the agencies to rely on manufacturers' product plans (that were submitted), the market forecast would account for not only these specific examples, but also for similar examples that have not yet been announced publicly.

Some commenters, such as CBD and NESCAUM, suggested that the agencies' omission of known future vehicles and technologies in the reference fleet causes inaccuracies, which CBD further suggested could lead the agencies to set lower standards. On the other hand, CARB commented that “the likely impact of this omission is minor.” Because the agencies' analysis examines the costs and benefits of progressively adding technology to manufacturers' fleets, the omission of future vehicles and technologies primarily affects how much additional technology (and, therefore, how much incremental cost and benefit) is available relative to the point at which the agencies' examination of potential new standards begins. Thus, in fact, the omission only reflects the reference fleet, rather than the agencies' conclusions regarding how stringent the standards should be. This is discussed further below. The agencies believe the above-mentioned comments by CBD, NESCAUM, and others are based on a misunderstanding of the agencies' approach to analyzing potential increases in regulatory stringency. The agencies also note that manufacturers do not always use technology solely to increase fuel economy, and that use of technology to increase vehicles' acceleration performance or utility would probably make that technology unavailable toward more stringent standards. Considering the incremental nature of the agencies' analysis, and the counterbalancing aspects of potentially omitted technology in the reference fleet, the agencies believe their determination of the stringency of new standards has not been impacted by any such omissions.

Moreover, EPA and NHTSA believe that not including such vehicles after MY 2008 does not significantly impact our estimates of the technology required to comply with the standards. If included, these vehicles could increase the extent to which manufacturers are, in the reference case, expected to over-comply with the MY 2011 CAFE standards, and could thereby make the new standards appear to cost less and yield less benefit relative to the reference case. However, in the agencies' judgment, production of the most advanced technology vehicles, such as the Chevy Volt or the Nissan Leaf (for example), will most likely be too limited during MY 2011 through MY 2016 to significantly impact manufacturers' compliance positions. While we are projecting the characteristics of the future fleet by extrapolating from the MY 2008 fleet, the primary difference between the future fleet and the 2008 fleet in the same vehicle segment is the use of additional CO₂-reducing and fuel-saving technologies. Both the NHTSA and EPA models add such technologies to evaluate means of complying with the standards, and the costs of doing so. Thus, our future projections of the vehicle fleet generally shift vehicle designs towards those more likely to be typical of newer vehicles. Compared to using product plans that show continued fuel economy increases planned based on expectations that CAFE standards will continue to increase, this approach helps to clarify the costs and benefits of the new standards, as the costs and benefits of all fuel economy improvements beyond those required by the MY 2011 CAFE standards are being assigned to the final rules. In some cases, the “actual” (vs. projected or “modeled”) new vehicles being introduced into the market by manufacturers are done so in anticipation of this rulemaking. On the other hand, manufacturers may plan to continue using technologies to improve vehicle performance and/or utility, not just fuel economy. Our approach prevents some of these actual technological improvements and their associated cost and fuel economy improvements from being assumed in the reference fleet. Thus, the added technology will not be considered to be free (or having no benefits) for the purposes of this rule.

In this regard, the agencies further note that manufacturer announcements regarding forward models (or future vehicle models) need not be accepted automatically. Manufacturers tend to limit accurate production intent information in these releases for reasons such as: (a) Competitors will closely examine their information for data in their product planning decisions; (b) the press coverage of forward model announcements is not uniform, meaning highly anticipated models have more coverage and materials than models that may be less exciting to the public and consistency and uniformity cannot be ensured with the usage of press information; and (c) these market projections are subject to change (sometimes significant), and manufacturers may not want to give the appearance of being indecisive, or under/over-confident to their shareholders and the public with premature release of information.

NHTSA has evaluated the use of public manufacturer forward model press information to update the vehicle fleet inputs to the baseline and reference fleet. The challenges in this approach are evidenced by the continuous stream of manufacturer press releases throughout a defined rulemaking period. Manufacturers' press releases suffer from the same types of inaccuracies that many commenters believe can affect product plans. Manufacturers can often be overly optimistic in their press releases, both on projected date of release of new models and on sales volumes.

More generally and more critically, as discussed in the proposal and as endorsed by many of the public comments, there are several advantages to the approach used by the agencies in this final rule. Most importantly, today's market forecast is much more transparent. The information sources used to develop today's market forecast are all either in the public domain or available commercially. Another significant advantage of today's market forecast is the agencies' ability to assess more fully the incremental costs and benefits of the proposed standards. In addition, by developing baseline and reference fleets from common sources, the agencies have been able to avoid some errors—perhaps related to interpretation of requests—that have been observed in past responses to NHTSA's requests. An additional advantage of the approach used for this rule is a consistent projection of the change in fuel economy and CO₂ emissions across the various vehicles from the application of new technology. With the approach used for this final rule, the baseline market data comes from actual vehicles (on the road today) which have actual fuel economy test data (in contrast to manufacturer estimates of future product fuel economy)—so there is no question what is the basis for the fuel economy or CO₂ performance of the baseline market data as it is.

5. How does manufacturer product plan data factor into the baseline used in this Final Rule?

In the spring and fall of 2009, many manufacturers submitted product plans in response to NHTSA's recent requests that they do so. NHTSA and EPA both have access to these plans, and both agencies have reviewed them in detail. A small amount of product plan data was used in the development of the baseline. The specific pieces of data are:

• Wheelbase.
• Track Width Front.
• Track Width Rear.
• EPS (Electric Power Steering).
• ROLL (Reduced Rolling Resistance).
• LUB (Advance Lubrication i.e. low weight oil).
• IACC (Improved Electrical Accessories).
• Curb Weight.
• GVWR (Gross Vehicle Weight Rating).

The track widths, wheelbase, curb weight, and GVWR for vehicles could have been looked up on the Internet (159 were), but were taken from the product plans when available for convenience. To ensure accuracy, a sample from each product plan was used as a check against the numbers available from Motortrend.com. These numbers will be published in the baseline file since they can be easily looked up on the internet. On the other hand, EPS, ROLL, LUB, and IACC are difficult to determine without using manufacturer's product plans. These items will not be published in the baseline file, but the data has been aggregated into the agencies' baseline in the technology effectiveness and cost effectiveness for each vehicle in a way that allows the baseline for the model to be published without revealing the manufacturer's data.

Also, some technical information that manufacturers have provided in product plans regarding specific vehicle models is, at least insofar as NHTSA and EPA have been able to determine, not available from public or commercial sources. While such gaps do not bear significantly on the agencies' analysis, the diversity of pickup configurations necessitated utilizing a sales-weighted average footprint value for many manufacturers' pickups. Since our modeling only utilizes footprint in order to estimate each manufacturer's CO₂ or fuel economy standard and all the other vehicle characteristics are available for each pickup configuration, this approximation has no practical impact on the projected technology or cost associated with compliance with the various standards evaluated. The only impact which could arise would be if the relative sales of the various pickup configurations changed, or if the agencies were to explore standards with a different shape. This would necessitate recalculating the average footprint value in order to maintain accuracy.

Additionally, as discussed in the NPRM, in an effort to update the 2008 baseline to account for the expected changes in the fleet in the near-term model years 2009-2011 described above, NHTSA requested permission from the manufacturers to make this limited product plan information public. Unfortunately, virtually no manufacturers agreed to allow the use of their data after 2009 model year. A few manufacturers, such as GM and Ford, stated we could use their 2009 product plan data after the end of production (December 31), but this would not have afforded us sufficient time to do the analysis for the final rule. Since the agencies were unable to obtain consistent updates, the baseline and reference fleets were not updated beyond 2008 model year for the final rule. The 2008 baseline fleet and projections were instead updated using the latest AEO and CSM data as discussed earlier.

NHTSA and EPA recognize that the approach applied for the current rule gives transparency and openness of the vehicle market forecast high priority, and accommodates minor inaccuracies that may be introduced by not accounting for future product mix changes anticipated in manufacturers' confidential product plans. For any future fleet analysis that the agencies are required to perform, NHTSA and EPA plan to request that manufacturers submit product plans and allow some public release of information. In performing this analysis, the agencies plan to reexamine potential tradeoffs between transparency and technical reasonableness, and to explain resultant choices.

C. Development of Attribute-Based Curve Shapes

In the NPRM, NHTSA and EPA proposed to set attribute-based CAFE and CO₂ standards that are defined by a mathematical function for MYs 2012-2016 passenger cars and light trucks. EPCA, as amended by EISA, expressly requires that CAFE standards for passenger cars and light trucks be based on one or more vehicle attributes related to fuel economy, and be expressed in the form of a mathematical function. The CAA has no such requirement, though in past rules, EPA has relied on both universal and attribute-based standards (e.g., for nonroad engines, EPA uses the attribute of horsepower). However, given the advantages of using attribute-based standards and given the goal of coordinating and harmonizing CO₂ standards promulgated under the CAA and CAFE standards promulgated under EPCA, EPA also proposed to issue standards that are attribute-based and defined by mathematical functions. There was consensus in the public comments that EPA should develop attribute-based CO₂ standards.

Comments received in response to the agencies' decision to base standards on vehicle footprint were largely supportive. Several commenters (BMW, NADA, NESCAUM) expressed support for attribute-based (as opposed to flat or universal) standards generally, and agreed with EPA's decision to harmonize with NHTSA in this respect. Many commenters (Aluminum Association, BMW, ICCT, NESCAUM, NY DEC, Schade, Toyota) also supported the agencies' decision to continue setting CAFE standards, and begin setting GHG standards, on the basis of vehicle footprint, although one commenter (NJ DEP) opposed the use of footprint due to concern that it encourages manufacturers to upsize vehicles and undercut the gains of the standard. Of the commenters supporting the use of footprint, several focused on the benefits of harmonization—both between EPA and NHTSA, and between the U.S. and the rest of the world. BMW commented, for example, that many other countries use weight-based standards rather than footprint-based. While BMW did not object to NHTSA's and EPA's use of footprint-based standards, it emphasized the impact of this non-harmonization on manufacturers who sell vehicles globally, and asked the agencies to consider these effects. NADA supported the use of footprint, but cautioned that the agencies must be careful in setting the footprint curve for light trucks to ensure that manufacturers can continue to provide functionality like 4WD and towing/hauling capacity.

Some commenters requested that the agencies consider other or more attributes in addition to footprint, largely reiterating comments submitted to the MYs 2011-2015 CAFE NPRM. Cummins supported the agencies using a secondary attribute to account for towing and hauling capacity in large trucks, for example, while Ferrari asked the agencies to consider a multi-attribute approach incorporating curb weight, maximum engine power or torque, and/or engine displacement, as it had requested in the previous round of CAFE rulemaking. An individual, Mr. Kenneth Johnson, commented that weight-based standards would be preferable to footprint-based ones, because weight correlates better with fuel economy than footprint, because the use of footprint does not necessarily guarantee safety the way the agencies say it does, and because weight-based standards would be fairer to manufacturers.

In response, EPA and NHTSA continue to believe that the benefits of footprint-attribute-based standards outweigh any potential drawbacks raised by commenters, and that harmonization between the two agencies should be the overriding goal on this issue. As discussed by NHTSA in the MY 2011 CAFE final rule, the agencies believe that the possibility of gaming is lowest with footprint-based standards, as opposed to weight-based or multi-attribute-based standards. Specifically, standards that incorporate weight, torque, power, towing capability, and/or off-road capability in addition to footprint would not only be significantly more complex, but by providing degrees of freedom with respect to more easily-adjusted attributes, they would make it less certain that the future fleet would actually achieve the average fuel economy and CO₂ levels projected by the agencies. The agencies recognize that based on economic and consumer demand factors that are external to this rule, the distribution of footprints in the future may be different (either smaller or larger) than what is projected in this rule. However, the agencies continue to believe that there will not be significant shifts in this distribution as a direct consequence of this rule. The agencies are therefore finalizing MYs 2012-2016 CAFE and GHG standards based on footprint.

The agencies also recognize that there could be benefits for a number of manufacturers if there was greater international harmonization of fuel economy and GHG standards, but this is largely a question of how stringent standards are and how they are enforced. It is entirely possible that footprint-based and weight-based systems can coexist internationally and not present an undue burden for manufacturers if they are carefully crafted. Different countries or regions may find different attributes appropriate for basing standards, depending on the particular challenges they face—from fuel prices, to family size and land use, to safety concerns, to fleet composition and consumer preference, to other environmental challenges besides climate change. The agencies anticipate working more closely with other countries and regions in the future to consider how to mitigate these issues in a way that least burdens manufacturers while respecting each country's need to meet its own particular challenges.

Under an attribute-based standard, every vehicle model has a performance target (fuel economy and CO₂ emissions for CAFE and CO₂ emissions standards, respectively), the level of which depends on the vehicle's attribute (for the proposal, footprint). The manufacturers' fleet average performance is determined by the production-weighted average (for CAFE, harmonic average) of those targets. NHTSA and EPA are promulgating CAFE and CO₂ emissions standards defined by constrained linear functions and, equivalently, piecewise linear functions. As a possible option for future rulemakings, the constrained linear form was introduced by NHTSA in the 2007 NPRM proposing CAFE standards for MY 2011-2015. Described mathematically, the proposed constrained linear function was defined according to the following formula:

Image Not Available

Where

TARGET = the fuel economy target (in mpg) applicable to vehicles of a given footprint (FOOTPRINT, in square feet),

a = the function's upper limit (in mpg),

b = the function's lower limit (in mpg),

c = the slope (in gpm per square foot) of the sloped portion of the function,

d = the intercept (in gpm) of the sloped portion of the function (that is, the value the sloped portion would take if extended to a footprint of 0 square feet, and the MIN and MAX functions take the minimum and maximum, respectively, of the included values; for example, MIN (1,2) = 1, MAX (1,2) = 2, and

MIN[MAX (1,2),3)]=2.

Because the format is linear on a gallons-per-mile basis, not on a miles-per-gallon basis, it is plotted as fuel consumption below. Graphically, the constrained linear form appears as shown in Figure II.C-1.

Image Not Available

The specific form and stringency for each fleet (passenger car and light trucks) and model year are defined through specific values for the four coefficients shown above.

EPA proposed the equivalent equation below for assigning CO₂ targets to an individual vehicle's footprint value. Although the general model of the equation is the same for each vehicle category and each year, the parameters of the equation differ for cars and trucks and for each model year. Described mathematically, EPA's proposed piecewise linear function was as follows:

Target = a, if x ≤ l

Target = cx + d, if l < x ≤ h

Target = b, if x > h

In the constrained linear form similar in form to the fuel economy equation above, this equation takes the simplified form:

Target = MIN [ MAX (c * x + d, a), b]

Where

Target = the CO₂ target value for a given footprint (in g/mi)

a = the minimum target value (in g/mi CO₂)

b = the maximum target value (in g/mi CO₂)

c = the slope of the linear function (in g/mi per sq ft CO₂)

d = is the intercept or zero-offset for the line (in g/mi CO₂)

x = footprint of the vehicle model (in square feet, rounded to the nearest tenth)

l & h are the lower and higher footprint limits or constraints or (“kinks”) or the boundary between the flat regions and the intermediate sloped line (in sq ft)

Graphically, piecewise linear form, like the constrained linear form, appears as shown in Figure II.C-2.

As for the constrained linear form, the specific form and stringency of the piecewise linear function for each fleet (passenger car and light trucks) and model year are defined through specific values for the four coefficients shown above.

For purposes of the proposed rules, NHTSA and EPA developed the basic curve shapes using methods similar to those applied by NHTSA in fitting the curves defining the MY 2011 standards. The first step involved defining the relevant vehicle characteristics in the form used by NHTSA's CAFE model (e.g., fuel economy, footprint, vehicle class, technology) described in Section II.B of this preamble and in Chapter 1 of the Joint TSD. However, because the baseline fleet utilizes a wide range of available fuel saving technologies, NHTSA used the CAFE model to develop a fleet to which all of the technologies discussed in Chapter 3 of the Joint TSD were applied, except dieselization and strong hybridization. This was accomplished by taking the following steps: (1) Treating all manufacturers as unwilling to pay civil penalties rather than applying technology, (2) applying any technology at any time, irrespective of scheduled vehicle redesigns or freshening, and (3) ignoring “phase-in caps” that constrain the overall amount of technology that can be applied by the model to a given manufacturer's fleet. These steps helped to increase technological parity among vehicle models, thereby providing a better basis (than the baseline or reference fleets) for estimating the statistical relationship between vehicle size and fuel economy.

In fitting the curves, NHTSA and EPA also continued to fit the sloped portion of the function to vehicle models between the footprint values at which the agencies continued to apply constraints to limit the function's value for both the smallest and largest vehicles. Without a limit at the smallest footprints, the function—whether logistic or linear—can reach values that would be unfairly burdensome for a manufacturer that elects to focus on the market for small vehicles; depending on the underlying data, an unconstrained form, could result in stringency levels that are technologically infeasible and/or economically impracticable for those manufacturers that may elect to focus on the smallest vehicles. On the other side of the function, without a limit at the largest footprints, the function may provide no floor on required fuel economy. Also, the safety considerations that support the provision of a disincentive for downsizing as a compliance strategy apply weakly, if at all, to the very largest vehicles. Limiting the function's value for the largest vehicles leads to a function with an inherent absolute minimum level of performance, while remaining consistent with safety considerations.

Before fitting the sloped portion of the constrained linear form, NHTSA and EPA selected footprints above and below which to apply constraints (i.e., minimum and maximum values) on the function. The agencies believe that the linear form performs well in describing the observed relationship between footprint and fuel consumption or CO₂ emissions for vehicle models within the footprint ranges covering most vehicle models, but that the single (as opposed to piecewise) linear form does not perform well in describing this relationship for the smallest and largest vehicle models. For passenger cars, the agency noted that several manufacturers offer small, sporty coupes below 41 square feet, such as the BMW Z4 and Mini, Honda S2000, Mazda MX-5 Miata, Porsche Carrera and 911, and Volkswagen New Beetle. Because such vehicles represent a small portion (less than 10 percent) of the passenger car market, yet often have performance, utility, and/or structural characteristics that could make it technologically infeasible and/or economically impracticable for manufacturers focusing on such vehicles to achieve the very challenging average requirements that could apply in the absence of a constraint, EPA and NHTSA proposed to “cut off” the linear portion of the passenger car function at 41 square feet. The agencies recognize that for manufacturers who make small vehicles in this size range, this cut off creates some incentive to downsize (i.e., further reduce the size, and/or increase the production of models currently smaller than 41 square feet) to make it easier to meet the target. The cut off may also create the incentive for manufacturers who do not currently offer such models to do so in the future. However, at the same time, the agencies believe that there is a limit to the market for cars smaller than 41 square feet—most consumers likely have some minimum expectation about interior volume, among other things. The agencies thus believe that the number of consumers who will want vehicles smaller than 41 square feet (regardless of how they are priced) is small, and that the incentive to downsize in response to this final rule, if present, will be minimal. For consistency, the agency proposed to “cut off” the light truck function at the same footprint, although no light trucks are currently offered below 41 square feet. The agencies further noted that above 56 square feet, the only passenger car model present in the MY 2008 fleet were four luxury vehicles with extremely low sales volumes—the Bentley Arnage and three versions of the Rolls Royce Phantom. NHTSA and EPA therefore also proposed to “cut off” the linear portion of the passenger car function at 56 square feet. Finally, the agencies noted that although public information is limited regarding the sales volumes of the many different configurations (cab designs and bed sizes) of pickup trucks, most of the largest pickups (e.g., the Ford F-150, GM Sierra/Silverado, Nissan Titan, and Toyota Tundra) appear to fall just above 66 square feet in footprint. EPA and NHTSA therefore proposed to “cut off” the linear portion of the light truck function at 66 square feet.

Having developed a set of vehicle emissions and footprint data which represent the benefit of all non-diesel, non-hybrid technologies, we determined the initial values for parameters c and d were determined for cars and trucks separately. c and d were initially set at the values for which the average (equivalently, sum) of the absolute values of the differences was minimized between the “maximum technology” fleet fuel consumption (within the footprints between the upper and lower limits) and the straight line of the function defined above at the same corresponding vehicle footprints. That is, c and d were determined by minimizing the average absolute residual, commonly known as the MAD (Mean Absolute Deviation) approach, of the corresponding straight line.

Finally, NHTSA calculated the values of the upper and lower parameters (a and b) based on the corresponding footprints discussed above (41 and 56 square feet for passenger cars, and 41 and 66 square feet for light trucks).

The result of this methodology is shown below in Figures II.C-3 and II.C-4 for passenger cars and light trucks, respectively. The fitted curves are shown with the underlying “maximum technology” passenger car and light truck fleets. For passenger cars, the mean absolute deviation of the sloped portion of the function was 14 percent. For trucks, the corresponding MAD was 10 percent.

The agencies used these functional forms as a starting point to develop mathematical functions defining the actual proposed standards as discussed above. The agencies then transposed these functions vertically (i.e., on a gpm or CO₂ basis, uniformly downward) to produce the same fleetwide fuel economy (and CO₂ emission levels) for cars and light trucks described in the NPRM.

A number of public comments generally supported the agencies' choice of attribute-based mathematical functions, as well as the methods applied to fit the function. Ferrari indicated support for the use of a constrained linear form rather than a constrained logistic form, support for the application of limits on the functions' values, support for a generally less steep passenger car curve compared to MY 2011, and support for the inclusion of all manufacturers in the analysis used to fit the curves. ICCT also supported the use of a constrained linear form. Toyota expressed general support for the methods and outcome, including a less-steep passenger car curve, and the application of limits on fuel economy targets applicable to the smallest vehicles. The UAW commented that the shapes and levels of the curves are reasonable.

Other commenters suggested that changes to the agencies' methods and results would yield better outcomes. GM suggested that steeper curves would provide a greater incentive for limited-line manufacturers to apply technology to smaller vehicles. GM argued that steeper and, in their view, fairer curves could be obtained by using sales-weighted least-squares regression rather than minimization of the unweighted mean absolute deviation. Conversely, students from UC Santa Barbara commented that the passenger car and light truck curves should be flatter and should converge over time in order to encourage the market to turn, as the agencies' analysis assumes it will, away from light trucks and toward passenger cars.

NADA commented that there should be no “cut-off” points (i.e., lower limits or floors), because these de facto “backstops” might limit consumer choice, especially for light trucks—a possibility also suggested by the Alliance. The Alliance and several individual manufacturers also commented that the cut-off point for light trucks should be shifted to 72 square feet (from the proposed 66 square feet), arguing that the preponderance of high-volume light truck models with footprints greater than 66 square feet is such that a 72 square foot cut-off point makes it unduly challenging for manufacturers serving the large pickup market and thereby constitutes a de facto backstop. Also, with respect to the smallest light truck models, Honda commented that the cut-off point should be set at the point defining the smallest 10 percent of the fleet, both for consistency with the passenger car cut-off point, and to provide a greater incentive for manufacturers to downsize the smallest light truck models (which provide greater functionality than passenger cars).

Other commenters focused on whether the agencies should have separate curves for different fleets or whether they should have a single curve that applied to both passenger cars and light trucks. This issue is related, to some extent, to commenters who discussed whether car and truck definitions should change. CARB, Ford, and Toyota supported separate curves for cars and trucks, generally stating that different fleets have different functional characteristics and these characteristics are appropriately addressed by separate curves. Likewise, AIAM, Chrysler, and NADA supported leaving the current definitions of car and truck the same. CBD, ICCT, and NESCAUM supported a single curve, based on concerns about manufacturers gaming the system and reclassifying passenger cars as light trucks in order to obtain the often-less stringent light truck standard, which could lead to lower benefits than anticipated by the agencies.

In addition, the students from UC Santa Barbara reported being unable to reproduce the agencies' analysis to fit curves to the passenger car and light truck fleets, even when using the model, inputs, and external analysis files posted to NHTSA's Web site when the NPRM was issued.

Having considered public comments, NHTSA and EPA have re-examined the development of curves underlying the standards proposed in the NPRM, and are promulgating standards based on the same underlying curves. The agencies have made this decision considering that, while EISA mandates that CAFE standards be defined by a mathematical function in terms of one or more attributes related to fuel economy, neither EISA nor the CAA require that the mathematical function be limited to the observed or theoretical dependence of fuel economy on the selected attribute or attributes. As a means by which CAFE and GHG standards are specified, the mathematical function can and does properly play a normative role. Therefore, NHTSA and EPA have concluded that, as supported by comments, the mathematical function can reasonably be based on a blend of analytical and policy considerations, as discussed below and in the Joint Technical Support Document.

With respect to GM's recommendation that NHTSA and EPA use weighted least-squares analysis, the agencies find that the market forecast used for analysis supporting both the NPRM and the final rule exhibits the two key characteristics that previously led NHTSA to use minimization of the unweighted Mean Absolute Deviation (MAD) rather than weighted least-squares analysis. First, projected model-specific sales volumes in the agencies' market forecast cover an extremely wide range, such that, as discussed in NHTSA's rulemaking for MY 2011, while unweighted regression gives low-selling vehicle models and high-selling vehicle models equal emphasis, sales-weighted regression would give some vehicle models considerably more emphasis than other vehicle models. The agencies' intention is to fit a curve that describes a technical relationship between fuel economy and footprint, given comparable levels of technology, and this supports weighting discrete vehicle models equally. On the other hand, sales weighted regression would allow the difference between other vehicle attributes to be reflected in the analysis, and also would reflect consumer demand.

Second, even after NHTSA's “maximum technology” analysis to increase technological parity of vehicle models before fitting curves, the agencies' market forecast contains many significant outliers. As discussed in NHTSA's rulemaking for MY 2011, MAD is a statistical procedure that has been demonstrated to produce more efficient parameter estimates than least-squares analysis in the presence of significant outliers. In addition, the agencies remain concerned that the steeper curves resulting from weighted least-squares analysis would increase the risk that energy savings and environmental benefits would be lower than projected, because the steeper curves would provide a greater incentive to increase sales of larger vehicles with lower fuel economy levels. Based on these technical considerations and these concerns regarding potential outcomes, the agencies have decided not to re-fit curves using weighted least-squares analysis, but note that they may reconsider using least-squares regression in future analysis.

NHTSA and EPA have considered GM's comment that steeper curves would provide a greater incentive for limited-line manufacturers to apply technology to smaller vehicles. While the agencies agree that a steeper curve would, absent any changes in fleet mix, tend to shift average compliance burdens away from GM and toward companies that make smaller vehicles, the agencies are concerned, as stated above, that steeper curves would increase the risk that induced increases in vehicle size could erode projected energy and environmental benefits.

NHTSA and EPA have also considered the comments by the students from UC Santa Barbara indicating that the passenger car and light truck curves should be flatter and should converge over time. The agencies conclude that flatter curves would reduce the incentives intended in shifting from “flat” CAFE standards to attribute-based CAFE and GHG standards—those being the incentive to respond to attribute-based standards in ways that minimize compromises in vehicle safety, and the incentive for more manufacturers (than primarily those selling a wider range of vehicles) across the range of the attribute to have to increase the application of fuel-saving technologies. With regard to whether the agencies should set separate curves or a single one, NHTSA also notes that EPCA requires NHTSA to establish standards separately for passenger cars and light trucks, and thus concludes that the standards for each fleet should be based on the characteristics of vehicles in each fleet. In other words, the passenger car curve should be based on the characteristics of passenger cars, and the light truck curve should be based on the characteristics of light trucks—thus to the extent that those characteristics are different, an artificially-forced convergence would not accurately reflect those differences. However, such convergence could be appropriate depending on future trends in the light vehicle market, specifically further reduction in the differences between passenger car and light truck characteristics. While that trend was more apparent when car-like 2WD SUVs were classified as light trucks, it seems likely to diminish for the model year vehicles subject to these rules as the truck fleet will be more purely “truck-like” than has been the case in recent years.

NHTSA and EPA have also considered comments on the maxima and minima that the agencies have applied to “cut off” the linear function underlying the proposed curves for passenger cars and light trucks. Contrary to NADA's suggestion that there should be no such cut-off points, the agencies conclude that curves lacking maximum fuel economy targets (i.e., minimum CO₂ targets) would result in average fuel economy and GHG requirements that would not be technologically feasible or economically practicable for manufacturers concentrating on those market segments. In addition, minimum fuel economy targets (i.e., maximum CO₂ targets) are important to mitigate the risk to energy and environmental benefits of potential market shifts toward large vehicles. The agencies also disagree with comments by the Alliance and several individual manufacturers that the cut-off point for light trucks should be shifted to 72 square feet (from the proposed 66 square feet) to ease compliance burdens facing manufacturers serving the large pickup market. Such a shift would increase the risk that energy and environmental benefits of the standards would be compromised by induced increases in the sales of large pickups, in situations where the increased compliance burden is feasible and appropriate. Also, the agencies' market forecast suggests that most of the light trucks models with footprints larger than 66 square feet have curb weights near or above 5,000 pounds. This suggests, in turn, that in terms of highway safety, there is little or no need to discourage downsizing of light trucks with footprints larger than 66 square feet. Based on these energy, environmental, technological feasibility, economic practicability, and safety considerations, the agencies conclude that the light truck curve should be cut off at 66 square feet, as proposed, rather than at 72 square feet. The agencies also disagree with Honda's suggestion that the cut-off point for the smallest trucks be shifted to a larger footprint value, because doing so could potentially increase the incentive to reclassify vehicles in that size range as light trucks, and could thereby increase the possibility that energy and environmental benefits of the rule would be less than projected.

Finally, considering comments by the UC Santa Barbara students regarding difficulties reproducing NHTSA's analysis, NHTSA reexamined its analysis, and discovered some erroneous entries in model inputs underlying the analysis used to develop the curves proposed in the NPRM. These errors are discussed in NHTSA's final Regulatory Impact Analysis (FRIA) and have since been corrected. They include the following: Incorrect valvetrain phasing and lift inputs for many BMW engines, incorrect indexing for some Daimler models, incorrectly enabled valvetrain technologies for rotary engines and Atkinson cycle engines, omitted baseline applications of cylinder deactivation in some Honda and GM engines, incorrect valve phasing codes for some 4-cylinder Chrysler engines, omitted baseline applications of advanced transmissions in some VW models, incorrectly enabled advanced electrification technologies for several hybrid vehicle models, and incorrect DCT effectiveness estimates for subcompact passenger cars. These errors, while not significant enough to impact the overall analysis of stringency, did affect the fitted slope for the passenger car curve and would have prevented precise replication of NHTSA's NPRM analysis by outside parties.

After correcting these errors and repeating the curve development analysis presented in the NPRM, NHTSA obtained the curves shown below in Figures II.C-5 and II.C-6 for passenger cars and light trucks, respectively. The fitted curves are shown with the underlying “maximum technology” passenger car and light truck fleets. For passenger cars, the mean absolute deviation of the sloped portion of the function was 14 percent. For trucks, the corresponding MAD was 10 percent.

This refitted passenger car curve is similar to that presented in the NPRM, and the refitted light truck curve is nearly identical to the corresponding curve in the NPRM. However, the slope of the refitted passenger car curve is about 27 percent steeper (on a gpm per sf basis) than the curve presented in the NPRM. For passenger cars and light trucks, respectively, Figures II.C-7 and II.C-8 show the results of adjustment—discussed in the next section—of the above curves to yield the average required fuel economy levels corresponding to the final standards.

While the resultant light truck curves are visually indistinguishable from one another, the refitted curve for passenger cars would increase stringency for the smallest cars, decrease stringency for the largest cars, and provide a greater incentive to increase vehicle size throughout the range of footprints within which NHTSA and EPA project most passenger car models will be sold through MY 2016. The agencies are concerned that these changes would make it unduly difficult for manufacturers to introduce new small passenger cars in the United States, and unduly risk losses in energy and environmental benefits by increasing incentives for the passenger car market to shift toward larger vehicles.

Also, the agencies note that the refitted passenger car curve produces only a slightly closer fit to the corrected fleet than would the curve estimated in the NPRM; with respect to the corrected fleet (between the “cut off” footprint values, and after the “maximum technology” analysis discussed above), the mean absolute deviation for the refitted curve is 13.887 percent, and that of a refitted curve held to the original slope is 13.933 percent. In other words, the data support the original slope very nearly as well as they support the refitted slope.

Considering NHTSA's and EPA's concerns regarding the change in incentives that would result from a refitted curve for passenger cars, and considering that the data support the original curves about as well as they would support refitted curves, the agencies are finalizing CAFE and GHG standards based on the curves presented in the NPRM.

Finally, regarding some commenters' inability to reproduce the agencies' NPRM analysis, NHTSA believes that its correction of the errors discussed above and its release (on NHTSA's Web site) of the updated Volpe model and all accompanying inputs and external analysis files should enable outside parties to independently reproduce the agencies' analysis. If outside parties continue to experience difficulty in doing so, we encourage them to contact NHTSA, and the agency will do its best to provide assistance.

Thus, in summary, the agencies' approach to developing the attribute-based mathematical functions for MY 2012-2016 CAFE and CO₂ standards represents the agencies' best technical judgment and consideration of potential outcomes at this time, and we are confident that the conclusions have resulted in appropriate and reasonable standards. The agencies recognize, however, that aspects of these decisions may merit updating or revision in future analysis to support CAFE and CO₂ standards or for other purposes. Consistent with best rulemaking practices, the agencies will take a fresh look at all assumptions and approaches to curve fitting, appropriate attributes, and mathematical functions in the context of future rulemakings.

The agencies also recognized in the NPRM the possibility that lower fuel prices could lead to lower fleetwide fuel economy (and higher CO₂ emissions) than projected in this rule. One way of addressing that concern is through the use of a universal standard—that is, an average standard set at a (single) absolute level. This is often described as a “backstop standard.” The agencies explained that under the CAFE program, EISA requires such a minimum average fuel economy standard for domestic passenger cars, but is silent with regard to similar backstops for imported passenger cars and light trucks, while under the CAA, a backstop could be adopted under section 202(a) assuming it could be justified under the relevant statutory criteria. NHTSA and EPA also noted that the flattened portions of the curves at the largest footprints directionally address the issue of a backstop (i.e., the mpg “floor” or gpm “ceiling” applied to the curves provides a universal and absolute value for that range of footprints). The agencies sought comment on whether backstop standards, or any other method within the agencies' statutory authority, should and can be implemented in order to guarantee a level of CO₂ emissions reductions and fuel savings under the attribute-based standards.

The agencies received a number of comments regarding the need for a backstop beyond NHTSA's alternative minimum standard. Comments were divided fairly evenly between support for and opposition to additional backstop standards. The following organizations supported the need for EPA and NHTSA to have explicit backstop standards: American Council for an Energy Efficient Economy (ACEEE), American Lung Association, California Air Resources Board (CARB), Environment America, Environment Defense Fund, Massachusetts Department of Environmental Protection, Natural Resources Defense Council (NRDC), Northeast States for Coordinated Air Use Management (NESCAUM), Public Citizen and Safe Climate Campaign, Sierra Club, State of Washington Department of Ecology, Union of Concerned Scientists, and a number of private citizens. Commenters in favor of additional backstop standards for all fleets for both NHTSA and EPA generally stated that the emissions reductions and fuel savings expected to be achieved by MY 2016 depended on assumptions about fleet mix that might not come to pass, and that various kinds of backstop standards or “ratchet mechanisms” were necessary to ensure that those reductions were achieved in fact. In addition, some commenters stated that manufacturers might build larger vehicles or more trucks during MYs 2012-2016 than the agencies project, for example, because (1) any amount of slope in target curves encourages manufacturers to upsize, and (2) lower targets for light trucks than for passenger cars encourage manufacturers to find ways to reclassify vehicles as light trucks, such as by dropping 2WD versions of SUVs and offering only 4WD versions, perhaps spurred by NHTSA's reclassification of 2WD SUVs as passenger cars. Both of these mechanisms will be addressed further below. Some commenters also discussed EPA authority under the CAA to set backstops, agreeing with EPA's analysis that section 202(a) allows such standards since EPA has wide discretion under that section to craft standards.

The following organizations opposed a backstop: Alliance of Automobile Manufacturers (AAM), Association of International Automobile Manufacturers (AIAM), Ford Motor Company, National Automobile Dealers Association (NADA), Toyota Motor Company, and the United Auto Workers Union. Commenters stating that additional backstops would not be necessary disagreed that upsizing was likely, and emphasized the anti-backsliding characteristics of the target curves. Others argued that universal absolute standards as backstops could restrict consumer choice of vehicles. Commenters making legal arguments under EPCA/EISA stated that Congress' silence regarding backstops for imported passenger cars and light trucks should be construed as a lack of authority for NHTSA to create further backstops. Commenters making legal arguments under the CAA focused on the lack of clear authority under the CAA to create multiple GHG emissions standards for the same fleets of vehicles based on the same statutory criteria, and opposed EPA taking steps that would reduce harmonization with NHTSA in standard setting. Furthermore, AIAM indicated that EISA's requirement that the combined (car and truck) fuel economy level reach at least 35 mpg by 2020 itself constitutes a backstop. One individual commented that while additional backstop standards might be necessary given optimism of fleet mix assumptions, both agencies' authorities would probably need to be revised by Congress to clarify that backstop standards (whether for individual fleets or for the national fleet as a whole) were permissible.

In response, EPA and NHTSA remain confident that their projections of the future fleet mix are reliable, and that future changes in the fleet mix of footprints and sales are not likely to lead to more than modest changes in projected emissions reductions or fuel savings. Both agencies thus remain confident in these fleet projections and the resulting emissions reductions and fuel savings from the standards. As explained in Section II.B above, the agencies' projections of the future fleet are based on the most transparent information currently available to the agencies. In addition, there are only a relatively few model years at issue. Moreover, market trends today are consistent with the agencies' estimates, showing shifts from light trucks to passenger cars and increased emphasis on fuel economy from all vehicles.

Finally, the shapes of the curves, including the “flattening” at the largest footprint values, tend to avoid or minimize regulatory incentives for manufacturers to upsize their fleet to change their compliance burden. Given the way the curves are fit to the data points (which represent vehicle models' fuel economy mapped against their footprint), the agencies believe that there is little real benefit to be gained by a manufacturer upsizing their vehicles. As discussed above, the agencies' analysis indicates that, for passenger car models with footprints falling between the two flattened portions of the corresponding curve, the actual slope of fuel economy with respect to footprint, if fit to that data by itself, is about 27 percent steeper than the curve the agencies are promulgating today. This difference suggests that manufacturers would, if anything, have more to gain by reducing vehicle footprint than by increasing vehicle footprint. For light trucks, the agencies' analysis indicates that, for models with footprints falling between the two flatted portions of the corresponding curve, the slope of fuel economy with respect to footprint is nearly identical to the curve the agencies are promulgating today. This suggests that, within this range, manufacturers would typically have little incentive to either incrementally increase or reduce vehicle footprint. The agencies recognize that based on economic and consumer demand factors that are external to this rule, the distribution of footprints in the future may be different (either smaller or larger) than what is projected in this rule. However, the agencies continue to believe that there will not be significant shifts in this distribution as a direct consequence of this rule.

At the same time, adding another backstop standard would have virtually no effect if the standard was weak, but a more stringent backstop could compromise the objectives served by attribute-based standards—that they distribute compliance burdens more equally among manufacturers, and at the same time encourage manufacturers to apply fuel-saving technologies rather than simply downsizing their vehicles, as they did in past decades under flat standards. This is why Congress mandated attribute-based CAFE standards in EISA. This compromise in objectives could occur for any manufacturer whose fleet average was above the backstop, irrespective of why they were above the backstop and irrespective of whether the industry as a whole was achieving the emissions and fuel economy benefits projected for the final standards, the problem the backstop is supposed to address. For example, the projected industry wide level of 250 gm/mile for MY 2016 is based on a mix of manufacturer levels, ranging from approximately 205 to 315 gram/mile but resulting in an industry wide basis in a fleet average of 250 gm/mile. Unless the backstop was at a very weak level, above the high end of this range, then some percentage of manufacturers would be above the backstop even if the performance of the entire industry remains fully consistent with the emissions and fuel economy levels projected for the final standards. For these manufacturers and any other manufacturers who were above the backstop, the objectives of an attribute based standard would be compromised and unnecessary costs would be imposed. This could directionally impose increased costs for some manufacturers. It would be difficult if not impossible to establish the level of a backstop standard such that costs are likely to be imposed on manufacturers only when there is a failure to achieve the projected reductions across the industry as a whole. An example of this kind of industry wide situation could be when there is a significant shift to larger vehicles across the industry as a whole, or if there is a general market shift from cars to trucks. The problem the agencies are concerned about in those circumstances is not with respect to any single manufacturer, but rather is based on concerns over shifts across the fleet as a whole, as compared to shifts in one manufacturer's fleet that may be more than offset by shifts the other way in another manufacturer's fleet. However, in this respect, a traditional backstop acts as a manufacturer specific standard.

The concept of a ratchet mechanism recognizes this problem, and would impose the new more stringent standard only when the problem arises across the industry as a whole. While the new more stringent standards would enter into force automatically, any such standards would still need to provide adequate lead time for the manufacturers. Given the limited number of model years covered by this rulemaking and the short lead-time already before the 2012 model year, a ratchet mechanism in this rulemaking that would automatically tighten the standards at some point after model year 2012 is finished and apply the new more stringent standards for model years 2016 or earlier, would fail to provide adequate lead time for any new, more stringent standards

Additionally, we do not believe that the risk of vehicle upsizing or changing vehicle offerings to “game” the passenger car and light truck definitions is as great as commenters imply for the model years in question. The changes that commenters suggest manufacturers might make are neither so simple nor so likely to be accepted by consumers. For example, 4WD versions of vehicles tend to be more expensive and, other things being equal, have inherently lower fuel economy than their 2WD equivalent models. Therefore, although there is a market for 4WD vehicles, and some consumers might shift from 2WD vehicles to 4WD vehicles if 4WD becomes available at little or no extra cost, many consumers still may not desire to purchase 4WD vehicles because of concerns about cost premium and additional maintenance requirements; conversely, many manufacturers often require the 2WD option to satisfy demand for base vehicle models. Additionally, increasing the footprint of vehicles requires platform changes, which usually requires a product redesign phase (the agencies estimate that this occurs on average once every 5 years for most models). Alternatively, turning many 2WD SUVs into 2WD light trucks would require manufacturers to squeeze a third row of seats in or significantly increase their GVWR, which also requires a significant change in the vehicle. The agencies are confident that the anticipated increases in average fuel economy and reductions in average CO₂ emission rates can be achieved without backstops under EISA or the CAA. As noted above, the agencies plan to conduct retrospective analysis to monitor progress. Both agencies have the authority to revise standards if warranted, as long as sufficient lead time is provided.

The agencies acknowledge that the MY 2016 fleet emissions and fuel economy goals of 250 g/mi and 34.1 mpg for EPA and NHTSA respectively are estimates and not standards (the MY 2012-2016 curves are the standards). Changes in fuel prices, consumer preferences, and/or vehicle survival and mileage accumulation rates could result in either smaller or larger oil and GHG savings. As explained above and elsewhere in the rule, the agencies believe that the possibility of not meeting (or, alternatively, exceeding) fuel economy and emissions goals exists, but is not likely. Given this, and given the potential complexities in designing an appropriate backstop, the agencies believe the balance here points to not adopting additional backstops at this time for the MYs 2012-2016 standards other than NHTSA's finalizing of the ones required by EPCA/EISA for domestic passenger cars. Nevertheless, the agencies recognize there are many factors that are inherently uncertain which can affect projections in the future, including fuel price and other factors which are unrelated to the standards contained in this final rule. Such factors can affect consumer preferences and are difficult to predict. At this time and based on the available information, the agencies have not included a backstop for model years 2012-2016. However, if circumstances change in the future in unanticipated ways, the agencies may revisit the issue of a backstop in the context of a future rulemaking either for model years 2012-2016 or as needed for standards for model years beyond 2016. This issue will be discussed further in Sections III and IV.

D. Relative Car-Truck Stringency

The agencies proposed fleetwide standards with the projected levels of stringency of 34.1 mpg or 250 g/mi in MY 2016 (as well as the corresponding intermediate year fleetwide standards) for NHTSA and EPA respectively. To determine the relative stringency of passenger car and light truck standards for those model years, the agencies were concerned that increasing the difference between the car and truck standards (either by raising the car standards or lowering the truck standards) could encourage manufacturers to build fewer cars and more trucks, likely to the detriment of fuel economy and CO₂ reductions. In order to maintain consistent car/truck standards, the agencies applied a constant ratio between the estimated average required performance under the passenger car and light truck standards, in order to maintain a stable set of incentives regarding vehicle classification.

To calculate relative car-truck stringency for the proposal, the agencies explored a number of possible alternatives, and for the reasons described in the proposal used the Volpe model in order to estimate stringencies at which net benefits would be maximized. The agencies have followed the same approach in calculating the relative car-truck stringency for the final standards promulgated today. Further details of the development of this approach can be found in Section IV of this preamble as well as in NHTSA's RIA and EIS. NHTSA examined passenger car and light truck standards that would produce the proposed combined average fuel economy levels from Table I.B.2-2 above. NHTSA did so by shifting downward the curves that maximize net benefits, holding the relative stringency of passenger car and light truck standards constant at the level determined by maximizing net benefits, such that the average fuel economy required of passenger cars remained 31 percent higher than the average fuel economy required of light trucks. This methodology resulted in the average fuel economy levels for passenger cars and light trucks during MYs 2012-2016 as shown in Table I.B.1-1. The following chart illustrates this methodology of shifting the standards from the levels maximizing net benefits to the levels consistent with the combined fuel economy standards in this final rule.

The final car and truck standards for EPA (Table I.B.1-4 above) were subsequently determined by first converting the average required fuel economy levels to average required CO₂ emission rates, and then applying the expected air conditioning credits for 2012-2016. These A/C credits are shown in the following table. Further details of the derivation of these factors can be found in Section III of this preamble or in the EPA RIA.

The agencies sought comment on the use of this methodology for apportioning the fleet stringencies to relative car and truck standards for 2012-2016. General Motors commented that, compared to the passenger car standard, the light truck standard is too stringent because “the most fuel efficient cars and small trucks already meet the 2016 MY requirements” but “the most fuel efficient large trucks must increase fuel economy by 20 percent to meet the 2016 MY requirements.” GM recommended that the agencies relax stringency specifically for large pickups, such as the Silverado.

The agencies disagree with the premise of the comment that the standard is too stringent under the applicable statutory provisions because some existing large trucks are not already meeting a later model year standard. Our analysis shows that the standards are not too stringent for manufacturers selling these vehicles. The agencies' analyses demonstrate a means by which manufacturers could apply cost-effective technologies in order to achieve the standards, and we have provided adequate lead time for the technology to be applied. More important, the agencies' analysis demonstrate that the fleetwide emission standards for MY 2016 are technically feasible, for example by implementing technologies such as engine downsizing, turbocharging, direct injection, improving accessories and tire rolling resistance, etc.

GM did not comment on the use of the methodology applied by the agencies to develop the gap between the passenger car and light truck standards—only on the outcome of the methodology. For the reasons discussed below, the agencies maintain that the methodology applied above provides an appropriate basis to determine the gap between the passenger car and light truck standards, and disagree with GM's arguments that the outcome is unfair.

First, GM's argument incorrectly suggests that every individual vehicle model must achieve its fuel economy and emissions targets. CAFE standards and new GHG emissions standards apply to fleetwide average performance, not model-specific performance, even though average required levels are based on average model-specific targets, and the agencies' analysis demonstrates that GM and other manufacturers of large trucks can cost-effectively comply with the new standards.

Second, GM implies that every manufacturer must be challenged equally with respect to fuel economy and emissions. Although NHTSA and EPA maintain that attribute-based CAFE and GHG emissions standards can more evenly balance compliance challenges, attribute-based standards are not intended to and cannot make these challenges equal, and while the agencies are mindful of the potential impacts of the standards on the relative competitiveness of different vehicle manufacturers, there is nothing in EPCA or the CAA requiring that these challenges be equal.

We have also already addressed and rejected GM's suggestion of shifting the “cut off” point for light trucks from 66 square feet to 72 square feet, thereby “dropping the floor” of the target function for light trucks. As discussed in the preceding section, this is so as not to forego the rules' energy and environmental benefits, and because there is little or no safety basis to discourage downsizing of the largest light trucks.

Finally, NHTSA and EPA disagree with GM's claim that the outcome of the agencies' approach is unfairly burdensome for light trucks as compared to passenger cars. Based on the agencies' market forecast, NHTSA's analysis indicates that incremental technology outlays could, on average, be comparable for passenger cars and light trucks under the final CAFE standards, and further indicates that the ratio of total benefits to total costs could be greater under the final light truck standards than under the final passenger car standards.

E. Joint Vehicle Technology Assumptions

Vehicle technology assumptions, i.e., assumptions about technologies' cost, effectiveness, and the rate at which they can be incorporated into new vehicles, are often controversial as they have a significant impact on the levels of the standards. The agencies must, therefore, take great care in developing and justifying these estimates. In developing technology inputs for the analysis of the MY 2012-2016 standards, the agencies reviewed the technology assumptions that NHTSA used in setting the MY 2011 standards, the comments that NHTSA received in response to its May 2008 Notice of Proposed Rulemaking (NPRM), and the comments received in response to the NPRM for this rule. This review is consistent with the request by President Obama in his January 26 memorandum to DOT. In addition, the agencies reviewed the technology input estimates identified in EPA's July 2008 Advance Notice of Proposed Rulemaking. The review of these documents was supplemented with updated information from more current literature, new product plans from manufacturers, and from EPA certification testing.

As a general matter, EPA and NHTSA believe that the best way to derive technology cost estimates is to conduct real-world tear down studies. Most of the commenters on this issue agreed. The advantages not only lie in the rigor of the approach, but also in its transparency. These studies break down each technology into its respective components, evaluate the costs of each component, and build up the costs of the entire technology based on the contribution of each component and the processes required to integrate them. As such, tear down studies require a significant amount of time and are very costly. EPA has been conducting tear down studies to assess the costs of vehicle technologies under a contract with FEV. Further details for this methodology is described below and in the TSD.

Due to the complexity and time incurred in a tear down study, only a few technologies evaluated in this rulemaking have been costed in this manner thus far. The agencies prioritized the technologies to be costed first based on how prevalent the agencies believed they might be likely to be during the rulemaking time frame, and based on their anticipated cost-effectiveness. The agencies believe that the focus on these important technologies (listed below) is sufficient for the analysis in this rule, but EPA is continuing to analyze more technologies beyond this rule as part of studies both already underway and in the future. For most of the other technologies, because tear down studies were not yet available, the agencies decided to pursue, to the extent possible, the Bill of Materials (BOM) approach as outlined in NHTSA's MY 2011 final rule. A similar approach was used by EPA in the EPA 2008 Staff Technical Report. This approach was recommended to NHTSA by Ricardo, an international engineering consulting firm retained by NHTSA to aid in the analysis of public comments on its proposed standards for MYs 2011-2015 because of its expertise in the area of fuel economy technologies. A BOM approach is one element of the process used in tear down studies. The difference is that under a BOM approach, the build up of cost estimates is conducted based on a review of cost and effectiveness estimates for each component from available literature, while under a tear down study, the cost estimates which go into the BOM come from the tear down study itself. To the extent that the agencies departed from the MY 2011 CAFE final rule estimates, the agencies explained the reasons and provided supporting analyses in the Technical Support Document.

Similarly, the agencies followed a BOM approach for developing the technology effectiveness estimates, insofar as the BOM developed for the cost estimates helped to inform the appropriate effectiveness values derived from the literature review. The agencies supplemented the information with results from available simulation work and real world EPA certification testing.

The agencies would also like to note that per the Energy Independence and Security Act (EISA), the National Academies of Sciences has been conducting a study for NHTSA to update Chapter 3 of their 2002 NAS Report, which presents technology effectiveness estimates for light-duty vehicles. The update takes a fresh look at that list of technologies and their associated cost and effectiveness values. The updated NAS report was expected to be available on September 30, 2009, but has not been completed and released to the public. The results from this study thus are unavailable for this rulemaking. The agencies look forward to considering the results from this study as part of the next round of rulemaking for CAFE/GHG standards.

1. What technologies did the agencies consider?

The agencies considered over 35 vehicle technologies that manufacturers could use to improve the fuel economy and reduce CO₂ emissions of their vehicles during MYs 2012-2016. The majority of the technologies described in this section are readily available, well known, and could be incorporated into vehicles once production decisions are made. Other technologies considered may not currently be in production, but are beyond the research phase and under development, and are expected to be in production in the next few years. These are technologies which can, for the most part, be applied both to cars and trucks, and which are capable of achieving significant improvements in fuel economy and reductions in CO₂ emissions, at reasonable costs. The agencies did not consider technologies in the research stage because the lead time available for this rule is not sufficient to move most of these technologies from research to production.

The technologies considered in the agencies' analysis are briefly described below. They fall into five broad categories: Engine technologies, transmission technologies, vehicle technologies, electrification/accessory technologies, and hybrid technologies. For a more detailed description of each technology and their costs and effectiveness, we refer the reader to Chapter 3 of the Joint TSD, Chapter III of NHTSA's FRIA, and Chapter 1 of EPA's final RIA. Technologies to reduce CO₂ and HFC emissions from air conditioning systems are discussed in Section III of this preamble and in EPA's final RIA.

Types of engine technologies that improve fuel economy and reduce CO₂ emissions include the following:

• Low-friction lubricants—low viscosity and advanced low friction lubricants oils are now available with improved performance and better lubrication. If manufacturers choose to make use of these lubricants, they would need to make engine changes and possibly conduct durability testing to accommodate the low-friction lubricants.
• Reduction of engine friction losses—can be achieved through low-tension piston rings, roller cam followers, improved material coatings, more optimal thermal management, piston surface treatments, and other improvements in the design of engine components and subsystems that improve engine operation.
• Conversion to dual overhead cam with dual cam phasing—as applied to overhead valves designed to increase the air flow with more than two valves per cylinder and reduce pumping losses.
• Cylinder deactivation— deactivates the intake and exhaust valves and prevents fuel injection into some cylinders during light-load operation. The engine runs temporarily as though it were a smaller engine which substantially reduces pumping losses.
• Variable valve timing—alters the timing of the intake valve, exhaust valve, or both, primarily to reduce pumping losses, increase specific power, and control residual gases.
• Discrete variable valve lift—increases efficiency by optimizing air flow over a broader range of engine operation which reduces pumping losses. Accomplished by controlled switching between two or more cam profile lobe heights.
• Continuous variable valve lift—is an electromechanically controlled system in which valve timing is changed as lift height is controlled. This yields a wide range of performance optimization and volumetric efficiency, including enabling the engine to be valve throttled.
• Stoichiometric gasoline direct-injection technology—injects fuel at high pressure directly into the combustion chamber to improve cooling of the air/fuel charge within the cylinder, which allows for higher compression ratios and increased thermodynamic efficiency.
• Combustion restart—can be used in conjunction with gasoline direct-injection systems to enable idle-off or start-stop functionality. Similar to other start-stop technologies, additional enablers, such as electric power steering, accessory drive components, and auxiliary oil pump, might be required.
• Turbocharging and downsizing—increases the available airflow and specific power level, allowing a reduced engine size while maintaining performance. This reduces pumping losses at lighter loads in comparison to a larger engine.
• Exhaust-gas recirculation boost—increases the exhaust-gas recirculation used in the combustion process to increase thermal efficiency and reduce pumping losses.
• Diesel engines—have several characteristics that give superior fuel efficiency, including reduced pumping losses due to lack of (or greatly reduced) throttling, and a combustion cycle that operates at a higher compression ratio, with a very lean air/fuel mixture, relative to an equivalent-performance gasoline engine. This technology requires additional enablers, such as NO_X trap catalyst after-treatment or selective catalytic reduction NO_X after-treatment. The cost and effectiveness estimates for the diesel engine and aftertreatment system utilized in this final rule have been revised from the NHTSA MY 2011 CAFE final rule. Additionally, the diesel technology option has been made available to small cars in the Volpe and OMEGA models. Though this is not expected to make a significant difference in the modeling results, the agencies agreed with the commenters that supported such a revision.

Types of transmission technologies considered include:

• Improved automatic transmission controls— optimizes shift schedule to maximize fuel efficiency under wide ranging conditions, and minimizes losses associated with torque converter slip through lock-up or modulation.
• Six-, seven-, and eight-speed automatic transmissions— the gear ratio spacing and transmission ratio are optimized to enable the engine to operate in a more efficient operating range over a broader range of vehicle operating conditions.
• Dual clutch or automated shift manual transmissions—are similar to manual transmissions, but the vehicle controls shifting and launch functions. A dual-clutch automated shift manual transmission uses separate clutches for even-numbered and odd-numbered gears, so the next expected gear is pre-selected, which allows for faster and smoother shifting.
• Continuously variable transmission— commonly uses V-shaped pulleys connected by a metal belt rather than gears to provide ratios for operation. Unlike manual and automatic transmissions with fixed transmission ratios, continuously variable transmissions can provide fully variable and an infinite number of transmission ratios that enable the engine to operate in a more efficient operating range over a broader range of vehicle operating conditions.
• Manual 6-speed transmission—offers an additional gear ratio, often with a higher overdrive gear ratio, than a 5-speed manual transmission.

Types of vehicle technologies considered include:

• Low-rolling-resistance tires—have characteristics that reduce frictional losses associated with the energy dissipated in the deformation of the tires under load, thereby improving fuel economy and reducing CO₂ emissions.
• Low-drag brakes—reduce the sliding friction of disc brake pads on rotors when the brakes are not engaged because the brake pads are pulled away from the rotors.
• Front or secondary axle disconnect for four-wheel drive systems—provides a torque distribution disconnect between front and rear axles when torque is not required for the non-driving axle. This results in the reduction of associated parasitic energy losses.
• Aerodynamic drag reduction—is achieved by changing vehicle shape or reducing frontal area, including skirts, air dams, underbody covers, and more aerodynamic side view mirrors.
• Mass reduction and material substitution— Mass reduction encompasses a variety of techniques ranging from improved design and better component integration to application of lighter and higher-strength materials. Mass reduction is further compounded by reductions in engine power and ancillary systems (transmission, steering, brakes, suspension, etc.). The agencies recognize there is a range of diversity and complexity for mass reduction and material substitution technologies and there are many techniques that automotive suppliers and manufacturers are using to achieve the levels of this technology that the agencies have modeled in our analysis for the final standards.

Types of electrification/accessory and hybrid technologies considered include:

• Electric power steering (EPS)—is an electrically-assisted steering system that has advantages over traditional hydraulic power steering because it replaces a continuously operated hydraulic pump, thereby reducing parasitic losses from the accessory drive.
• Improved accessories (IACC)—may include high efficiency alternators, electrically driven (i.e., on-demand) water pumps and cooling fans. This excludes other electrical accessories such as electric oil pumps and electrically driven air conditioner compressors. The latter is covered explicitly within the A/C credit program.
• Air Conditioner Systems—These technologies include improved hoses, connectors and seals for leakage control. They also include improved compressors, expansion valves, heat exchangers and the control of these components for the purposes of improving tailpipe CO₂ emissions as a result of A/C use. These technologies are discussed later in this preamble and covered separately in the EPA RIA.
• 12-volt micro-hybrid (MHEV)—also known as idle-stop or start-stop and commonly implemented as a 12-volt belt-driven integrated starter-generator, this is the most basic hybrid system that facilitates idle-stop capability. Along with other enablers, this system replaces a common alternator with a belt-driven enhanced power starter-alternator, and a revised accessory drive system.
• Higher Voltage Stop-Start/Belt Integrated Starter Generator (BISG)—provides idle-stop capability and uses a higher voltage battery with increased energy capacity over typical automotive batteries. The higher system voltage allows the use of a smaller, more powerful electric motor. This system replaces a standard alternator with an enhanced power, higher voltage, higher efficiency starter-alternator, that is belt driven and that can recover braking energy while the vehicle slows down (regenerative braking).
• Integrated Motor Assist (IMA)/Crank integrated starter generator (CISG)—provides idle-stop capability and uses a high voltage battery with increased energy capacity over typical automotive batteries. The higher system voltage allows the use of a smaller, more powerful electric motor and reduces the weight of the wiring harness. This system replaces a standard alternator with an enhanced power, higher voltage, higher efficiency starter-alternator that is crankshaft mounted and can recover braking energy while the vehicle slows down (regenerative braking).
• 2-mode hybrid (2MHEV)—is a hybrid electric drive system that uses an adaptation of a conventional stepped-ratio automatic transmission by replacing some of the transmission clutches with two electric motors that control the ratio of engine speed to vehicle speed, while clutches allow the motors to be bypassed. This improves both the transmission torque capacity for heavy-duty applications and reduces fuel consumption and CO₂ emissions at highway speeds relative to other types of hybrid electric drive systems.
• Power-split hybrid (PSHEV)— a hybrid electric drive system that replaces the traditional transmission with a single planetary gearset and a motor/generator. This motor/generator uses the engine to either charge the battery or supply additional power to the drive motor. A second, more powerful motor/generator is permanently connected to the vehicle's final drive and always turns with the wheels. The planetary gear splits engine power between the first motor/generator and the drive motor to either charge the battery or supply power to the wheels.
• Plug-in hybrid electric vehicles (PHEV)—are hybrid electric vehicles with the means to charge their battery packs from an outside source of electricity (usually the electric grid). These vehicles have larger battery packs with more energy storage and a greater capability to be discharged than other hybrids. They also use a control system that allows the battery pack to be substantially depleted under electric-only or blended mechanical/electric operation.
• Electric vehicles (EV)—are vehicles with all-electric drive and with vehicle systems powered by energy-optimized batteries charged primarily from grid electricity.

The cost estimates for the various hybrid systems have been revised from the estimates used in the MY 2011 CAFE final rule, in particular with respect to estimated battery costs.

2. How did the agencies determine the costs and effectiveness of each of these technologies?

As mentioned above, EPA and NHTSA believe that the best way to derive technology cost estimates is to conduct real-world tear down studies. To date, the costs of the following five technologies have been evaluated with respect to their baseline (or replaced) technologies. For these technologies noted below, the agencies relied on the tear down data available and scaling methodologies used in EPA's ongoing study with FEV. Only the cost estimate for the first technology on the list below was used in the NPRM. The others were completed subsequent to the publication of the NPRM.

1. Stoichiometric gasoline direct injection and turbo charging with engine downsizing (T-DS) for a large DOHC 4 cylinder engine to a small DOHC (dual overhead cam) 4 cylinder engine.

2. Stoichiometric gasoline direct injection and turbo charging with engine downsizing for a SOHC single overhead cam) 3 valve/cylinder V8 engine to a SOHC V6 engine.

3. Stoichiometric gasoline direct injection and turbo charging with engine downsizing for a DOHC V6 engine to a DOHC 4 cylinder engine.

4. 6-speed automatic transmission replacing a 5-speed automatic transmission.

5. 6-speed wet dual clutch transmission (DCT) replacing a 6-speed automatic transmission.

This costing methodology has been published and gone through a peer review. Using this tear down costing methodology, FEV has developed costs for each of the above technologies. In addition, FEV and EPA extrapolated the engine downsizing costs for the following scenarios that were outside of the noted study cases:

1. Downsizing a SOHC 2 valve/cylinder V8 engine to a DOHC V6.

2. Downsizing a DOHC V8 to a DOHC V6.

3. Downsizing a SOHC V6 engine to a DOHC 4 cylinder engine.

4. Downsizing a DOHC 4 cylinder engine to a DOHC 3 cylinder engine.

The agencies relied on the findings of FEV in part for estimating the cost of these technologies in this rulemaking. However, for some of the technologies, NHTSA and EPA modified FEV's estimated costs. FEV made the assumption that these technologies would be mature when produced in large volumes (450,000 units or more). The agencies believe that there is some uncertainty regarding each manufacturer's near-term ability to employ the technology at the volumes assumed in the FEV analysis. There is also the potential for near term (earlier than 2016) supplier-level Engineering, Design and Testing (ED&T) costs to be in excess of those considered in the FEV analysis as existing equipment and facilities are converted to production of new technologies. The agencies have therefore decided to average the FEV results with the NPRM values in an effort to account for these near-term factors. This methodology was done for the following technologies:

1. Converting a port-fuel injected (PFI) DOHC I4 to a turbocharged-downsized-stoichiometric GDI DOHC I3.

2. Converting a PFI DOHC V6 engine to a T-DS-stoichiometric GDI DOHC I4.

3. Converting a PFI SOHC V6 engine to a T-DS-stoichiometric GDI DOHC I4.

4. Converting a PFI DOHC V8 engine to a T-DS-stoichiometric GDI DOHC V6.

5. Converting a PFI SOHC 3V V8 engine to a T-DS-stoichiometric GDI DOHC V6.

6. Converting a PFI SOHC 2V V8 engine to a T-DS-stoichiometric GDI DOHC V6.

7. Replacing a 4-speed automatic transmission with a 6-speed automatic transmission.

8. Replacing a 5-speed automatic transmission with a 6-speed automatic transmission.

9. Replacing a 6-speed automatic transmission with a 6-speed wet dual clutch transmission.

For the I4 to Turbo GDI I4 study applied in the NPRM, the agencies requested from FEV an adjusted cost estimate which accounted for these uncertainties as an adjustment to the base technology burden rate. These new costs are used in the final rules. These details are also further described in the memo to the docket. The confidential information provided by manufacturers as part of their product plan submissions to the agencies or discussed in meetings between the agencies and the manufacturers and suppliers served largely as a check on publicly-available data.

For the other technologies, considering all sources of information (including public comments) and using the BOM approach, the agencies worked together intensively to determine component costs for each of the technologies and build up the costs accordingly. Where estimates differ between sources, we have used our engineering judgment to arrive at what we believe to be the best available cost estimate, and explained the basis for that exercise of judgment in the TSD. Building on NHTSA's estimates developed for the MY 2011 CAFE final rule and EPA's Advance Notice of Proposed Rulemaking, which relied on the EPA 2008 Staff Technical Report, the agencies took a fresh look at technology cost and effectiveness values for purposes of the joint rulemaking under the National Program. For costs, the agencies reconsidered both the direct or “piece” costs and indirect costs of individual components of technologies. For the direct costs, the agencies followed a bill of materials (BOM) approach employed in NHTSA's MY 2011 final rule based on recommendation from Ricardo, Inc., as described above. EPA used a similar approach in the EPA 2008 Staff Technical Report. A bill of materials, in a general sense, is a list of components or sub-systems that make up a system—in this case, an item of fuel economy-improving technology. In order to determine what a system costs, one of the first steps is to determine its components and what they cost.

NHTSA and EPA estimated these components and their costs based on a number of sources for cost-related information. The objective was to use those sources of information considered to be most credible for projecting the costs of individual vehicle technologies. For example, while NHTSA and Ricardo engineers had relied considerably in the MY 2011 final rule on the 2008 Martec Report for costing contents of some technologies, upon further joint review and for purposes of the MY 2012-2016 standards, the agencies decided that some of the costing information in that report was no longer accurate due to downward trends in commodity prices since the publication of that report. The agencies reviewed, then revalidated or updated cost estimates for individual components based on new information. Thus, while NHTSA and EPA found that much of the cost information used in NHTSA's MY 2011 final rule and EPA's staff report was consistent to a great extent, the agencies, in reconsidering information from many sources, revised several component costs of several major technologies: turbocharging with engine downsizing (as described above), mild and strong hybrids, diesels, stoichiometric gasoline direct injection fuel systems, and valve train lift technologies. These are discussed at length in the Joint TSD and in NHTSA's final RIA.

Once costs were determined, they were adjusted to ensure that they were all expressed in 2007 dollars using a ratio of GDP values for the associated calendar years, and indirect costs were accounted for using the ICM (indirect cost multiplier) approach explained in Chapter 3 of the Joint TSD, rather than using the traditional Retail Price Equivalent (RPE) multiplier approach. A report explaining how EPA developed the ICM approach can be found in the docket for this rule. The comments addressing the ICM approach were generally positive and encouraging. However, one commenter suggested that we had mischaracterized the complexity of a few of our technologies, which would result in higher or lower markups than presented in the NPRM. That commenter also suggested that we had used the ICMs as a means of placing a higher level of manufacturer learning on the cost estimates. The latter comment is not true and the methodology behind the ICM approach is explained in detail in the reports that are available in the docket for this rule. The former is open to debate given the subjective nature of the engineering analysis behind it, but upon further thought both agencies believe that the complexities used in the NPRM were appropriate and have, therefore, carried those forward into the final rule. We discuss this in greater detail in the Response to Comments document.

Regarding estimates for technology effectiveness, NHTSA and EPA also reexamined the estimates from NHTSA's MY 2011 final rule and EPA's ANPRM and 2008 Staff Technical Report, which were largely consistent with NHTSA's 2008 NPRM estimates. The agencies also reconsidered other sources such as the 2002 NAS Report, the 2004 NESCCAF report, recent CAFE compliance data (by comparing similar vehicles with different technologies against each other in fuel economy testing, such as a Honda Civic Hybrid versus a directly comparable Honda Civic conventional drive), and confidential manufacturer estimates of technology effectiveness. NHTSA and EPA engineers reviewed effectiveness information from the multiple sources for each technology and ensured that such effectiveness estimates were based on technology hardware consistent with the BOM components used to estimate costs. The agencies also carefully examined the pertinent public comments. Together, they compared the multiple estimates and assessed their validity, taking care to ensure that common BOM definitions and other vehicle attributes such as performance, refinement, and drivability were taken into account. However, because the agencies' respective models employ different numbers of vehicle subclasses and use different modeling techniques to arrive at the standards, direct comparison of BOMs was somewhat more complicated. To address this and to confirm that the outputs from the different modeling techniques produced the same result, NHTSA and EPA developed mapping techniques, devising technology packages and mapping them to corresponding incremental technology estimates. This approach helped compare the outputs from the incremental modeling technique to those produced by the technology packaging approach to ensure results that are consistent and could be translated into the respective models of the agencies.

In general, most effectiveness estimates used in both the MY 2011 final rule and the 2008 EPA staff report were determined to be accurate and were carried forward without significant change first into the NPRM, and now into these final rules. When NHTSA and EPA's estimates for effectiveness diverged slightly due to differences in how the agencies apply technologies to vehicles in their respective models, we report the ranges for the effectiveness values used in each model. There were only a few comments on the technology effectiveness estimates used in the NPRM. Most of the technologies that were mentioned in the comments were the more advanced technologies that are not assumed to have large penetrations in the market within the timeframe of this rule, notably hybrid technologies. Even if the effectiveness figures for hybrid vehicles were adjusted, it would have made little difference in the NHTSA and EPA analysis of the impacts and costs of the rule. The response to comments document has more specific responses to these comments.

The agencies note that the effectiveness values estimated for the technologies considered in the modeling analyses may represent average values, and do not reflect the enormous spectrum of possible values that could result from adding the technology to different vehicles. For example, while the agencies have estimated an effectiveness of 0.5 percent for low friction lubricants, each vehicle could have a unique effectiveness estimate depending on the baseline vehicle's oil viscosity rating. Similarly, the reduction in rolling resistance (and thus the improvement in fuel economy and the reduction in CO₂ emissions) due to the application of low rolling resistance tires depends not only on the unique characteristics of the tires originally on the vehicle, but on the unique characteristics of the tires being applied, characteristics which must be balanced between fuel efficiency, safety, and performance. Aerodynamic drag reduction is much the same—it can improve fuel economy and reduce CO₂ emissions, but it is also highly dependent on vehicle-specific functional objectives. For purposes of the final standards, NHTSA and EPA believe that employing average values for technology effectiveness estimates, as adjusted depending on vehicle subclass, is an appropriate way of recognizing the potential variation in the specific benefits that individual manufacturers (and individual vehicles) might obtain from adding a fuel-saving technology.

Chapter 3 of the Joint Technical Support Document contains a detailed description of our assessment of vehicle technology cost and effectiveness estimates. The agencies note that the technology costs included in this final rule take into account only those associated with the initial build of the vehicle. Although comments were received to the NPRM that suggested there could be additional maintenance required with some new technologies (e.g., turbocharging, hybrids, etc.), and that additional maintenance costs could occur as a result, the agencies do not believe that the amount of additional cost will be significant in the timeframe of this rulemaking, based on the relatively low application rates for these technologies. The agencies will undertake a more detailed review of these potential costs in preparation for the next round of CAFE/GHG standards.

F. Joint Economic Assumptions

The agencies' final analysis of alternative CAFE and GHG standards for the model years covered by this final rulemaking rely on a range of forecast information, economic estimates, and input parameters. This section briefly describes the agencies' choices of specific parameter values. These economic values play a significant role in determining the benefits of both CAFE and GHG standards.

In reviewing these variables and the agency's estimates of their values for purposes of this final rule, NHTSA and EPA reconsidered previous comments that NHTSA had received, reviewed newly available literature, and reviewed comments received in response to the proposed rule. For this final rule, we made three major changes to the economic assumptions. First, we revised the technology costs to reflect more recently available data. Second, we updated fuel price and transportation demand assumptions to reflect the Annual Energy Outlook (AEO) 2010 Early Release. Third, we have updated our estimates of the social cost of carbon (SCC) based on a recent interagency process. The key economic assumptions are summarized below, and are discussed in greater detail in Section III (EPA) and Section IV (NHTSA), as well as in Chapter 4 of the Joint TSD, Chapter VIII of NHTSA's RIA and Chapter 8 of EPA's RIA.

• Costs of fuel economy-improving technologies—These estimates are presented in summary form above and in more detail in the agencies' respective sections of this preamble, in Chapter 3 of the Joint TSD, and in the agencies' respective RIAs. The technology cost estimates used in this analysis are intended to represent manufacturers' direct costs for high-volume production of vehicles with these technologies and sufficient experience with their application so that all cost reductions due to “learning curve” effects have been fully realized. Costs are then modified by applying near-term indirect cost multipliers ranging from 1.11 to 1.64 to the estimates of vehicle manufacturers' direct costs for producing or acquiring each technology to improve fuel economy, depending on the complexity of the technology and the time frame over which costs are estimated. This accounts for both the direct and indirect costs associated with implementing new technologies in response to this final rule. The technology cost estimates for a select group of technologies have changed since the NPRM. These changes, as summarized in Section II.E and in Chapter 3 of the Joint TSD, were made in response to updated cost estimates available to the agencies shortly after publication of the NPRM, not in response to comments. In general, commenters were supportive of the cost estimates used in the NPRM and the transparency of the methodology used to generate them.
• Potential opportunity costs of improved fuel economy—This estimate addresses the possibility that achieving the fuel economy improvements required by alternative CAFE or GHG standards would require manufacturers to compromise the performance, carrying capacity, safety, or comfort of their vehicle models. If it did so, the resulting sacrifice in the value of these attributes to consumers would represent an additional cost of achieving the required improvements, and thus of manufacturers' compliance with stricter standards. Currently the agencies assume that these vehicle attributes do not change, and include the cost of maintaining these attributes as part of the cost estimates for technologies. However, it is possible that the technology cost estimates do not include adequate allowance for the necessary efforts by manufacturers to maintain vehicle performance, carrying capacity, and utility while improving fuel economy and reducing GHG emissions. While, in principle, consumer vehicle demand models can measure these effects, these models do not appear to be robust across specifications, since authors derive a wide range of willingness-to-pay values for fuel economy from these models, and there is not clear guidance from the literature on whether one specification is clearly preferred over another. This issue is discussed in EPA's RIA, Section 8.1.2 and NHTSA's RIA Section VIII.H. The agencies requested comment on how to estimate explicitly the changes in vehicle buyers' welfare from the combination of higher prices for new vehicle models, increases in their fuel economy, and any accompanying changes in vehicle attributes such as performance, passenger- and cargo-carrying capacity, or other dimensions of utility. Commenters did not provide recommendations for how to evaluate the quality of different models or identify a model appropriate for the agencies' purposes. Some commenters expressed various concerns about the use of existing consumer vehicle choice models. While EPA and NHTSA are not using a consumer vehicle choice model to analyze the effects of this rule, we continue to investigate these models.
• The on-road fuel economy “gap”—Actual fuel economy levels achieved by light-duty vehicles in on-road driving fall somewhat short of their levels measured under the laboratory-like test conditions used by NHTSA and EPA to establish compliance with the final CAFE and GHG standards. The agencies use an on-road fuel economy gap for light-duty vehicles of 20 percent lower than published fuel economy levels. For example, if the measured CAFE fuel economy value of a light truck is 20 mpg, the on-road fuel economy actually achieved by a typical driver of that vehicle is expected to be 16 mpg (20*.80). NHTSA previously used this estimate in its MY 2011 final rule, and the agencies confirmed it based on independent analysis for use in this FRM. No substantive comments were received on this input.

• Fuel prices and the value of saving fuel—Projected future fuel prices are a critical input into the preliminary economic analysis of alternative standards, because they determine the value of fuel savings both to new vehicle buyers and to society. For the proposed rule, the agencies had relied on the then most recent fuel price projections from the U.S. Energy Information Administration's (EIA) Annual Energy Outlook (AEO) 2009 (Revised Updated). However, for this final rule, the agencies have updated the analyses based on AEO 2010 (December 2009 Early Release) Reference Case forecasts of inflation-adjusted (constant-dollar) retail gasoline and diesel fuel prices, which represent the EIA's most up-to-date estimate of the most likely course of future prices for petroleum products. AEO 2010 includes slightly lower petroleum prices compared to AEO 2009.

The forecasts of fuel prices reported in EIA's AEO 2010 Early Release Reference Case extends through 2035, compared to the AEO 2009 which only went through 2030. As in the proposal, fuel prices beyond the time frame of AEO's forecast were estimated using an average growth rate.

While EIA revised AEO 2010, the vehicle MPG standards are similar to those that were published in AEO 2009. No substantive comments were received on the use of AEO as a source of fuel prices.

• Consumer valuation of fuel economy and payback period—In estimating the impacts on vehicle sales, the agencies assume that potential buyers value the resulting fuel savings improvements that would result from alternative CAFE and GHG standards over only part of the expected lifetime of the vehicles they purchase. Specifically, we assume that buyers value fuel savings over the first five years of a new vehicle's lifetime, and that buyers discount the value of these future fuel savings using rates of 3% and 7%. The five-year figure represents the current average term of consumer loans to finance the purchase of new vehicles. One commenter argued that higher-fuel-economy vehicles should have higher resale prices than vehicles with lower fuel economy, but did not provide supporting data. This revision, if made, would increase the net benefits of the rule. Another commenter supported the use of a five-year payback period for this analysis. In the absence of data to support changes, EPA and NHTSA have kept the same assumptions. In the analysis of net benefits, EPA and NHTSA assume that vehicle buyers benefit from the full fuel savings over the vehicle's lifetime, discounted for present value calculations at 3 and 7 percent.
• Vehicle sales assumptions—The first step in estimating lifetime fuel consumption by vehicles produced during a model year is to calculate the number of vehicles expected to be produced and sold. The agencies relied on the AEO 2010 Early Release for forecasts of total vehicle sales, while the baseline market forecast developed by the agencies (see Section II.B) divided total projected sales into sales of cars and light trucks.

• Vehicle survival assumptions—We then applied updated values of age-specific survival rates for cars and light trucks to these adjusted forecasts of passenger car and light truck sales to determine the number of these vehicles remaining in use during each year of their expected lifetimes. No substantive comments were received on vehicle survival assumptions.
• Total vehicle use—We then calculated the total number of miles that cars and light trucks produced in each model year will be driven during each year of their lifetimes using estimates of annual vehicle use by age tabulated from the Federal Highway Administration's 2001 National Household Transportation Survey (NHTS), adjusted to account for the effect on vehicle use of subsequent increases in fuel prices. Due to the lower fuel prices projected in AEO 2010, the average vehicle is estimated to be used slightly more (~3 percent) over its lifetime than assumed in the proposal. In order to insure that the resulting mileage schedules imply reasonable estimates of future growth in total car and light truck use, we calculated the rate of growth in annual car and light truck mileage at each age that is necessary for total car and light truck travel to increase at the rates forecast in the AEO 2010 Early Release Reference Case. The growth rate in average annual car and light truck use produced by this calculation is approximately 1.1 percent per year. This rate was applied to the mileage figures derived from the 2001 NHTS to estimate annual mileage during each year of the expected lifetimes of MY 2012-2016 cars and light trucks. While commenters requested further detail on the assumptions regarding total vehicle use, no specific issues were raised.

• Accounting for the rebound effect of higher fuel economy—The rebound effect refers to the fraction of fuel savings expected to result from an increase in vehicle fuel economy—particularly an increase required by the adoption of more stringent CAFE and GHG standards—that is offset by additional vehicle use. The increase in vehicle use occurs because higher fuel economy reduces the fuel cost of driving, typically the largest single component of the monetary cost of operating a vehicle, and vehicle owners respond to this reduction in operating costs by driving slightly more. We received comments supporting our proposed value of 10 percent, although we also received comments recommending higher and lower values. However, we did not receive any new data or comments that justify revising the 10 percent value for the rebound effect at this time.
• Benefits from increased vehicle use—The increase in vehicle use from the rebound effect provides additional benefits to their owners, who may make more frequent trips or travel farther to reach more desirable destinations. This additional travel provides benefits to drivers and their passengers by improving their access to social and economic opportunities away from home. These benefits are measured by the net “consumer surplus” resulting from increased vehicle use, over and above the fuel expenses associated with this additional travel. We estimate the economic value of the consumer surplus provided by added driving using the conventional approximation, which is one half of the product of the decline in vehicle operating costs per vehicle-mile and the resulting increase in the annual number of miles driven. Because it depends on the extent of improvement in fuel economy, the value of benefits from increased vehicle use changes by model year and varies among alternative standards.
• The value of increased driving range—By reducing the frequency with which drivers typically refuel their vehicles, and by extending the upper limit of the range they can travel before requiring refueling, improving fuel economy and reducing GHG emissions thus provides some additional benefits to their owners. No direct estimates of the value of extended vehicle range are readily available, so the agencies' analysis calculates the reduction in the annual number of required refueling cycles that results from improved fuel economy, and applies DOT-recommended values of travel time savings to convert the resulting time savings to their economic value. Please see the Chapter 4 of the Joint TSD for details.

• Added costs from congestion, crashes and noise—Although it provides some benefits to drivers, increased vehicle use associated with the rebound effect also contributes to increased traffic congestion, motor vehicle accidents, and highway noise. Depending on how the additional travel is distributed over the day and on where it takes place, additional vehicle use can contribute to traffic congestion and delays by increasing traffic volumes on facilities that are already heavily traveled during peak periods. These added delays impose higher costs on drivers and other vehicle occupants in the form of increased travel time and operating expenses, increased costs associated with traffic accidents, and increased traffic noise. The agencies rely on estimates of congestion, accident, and noise costs caused by automobiles and light trucks developed by the Federal Highway Administration to estimate the increased external costs caused by added driving due to the rebound effect.

• Petroleum consumption and import externalities—U.S. consumption and imports of petroleum products also impose costs on the domestic economy that are not reflected in the market price for crude petroleum, or in the prices paid by consumers of petroleum products such as gasoline. In economics literature on this subject, these costs include (1) higher prices for petroleum products resulting from the effect of U.S. oil import demand on the world oil price (“monopsony costs”); (2) the expected costs from the risk of disruptions to the U.S. economy caused by sudden reductions in the supply of imported oil to the U.S.; and (3) expenses for maintaining a U.S. military presence to secure imported oil supplies from unstable regions, and for maintaining the strategic petroleum reserve (SPR) to cushion against resulting price increases. Reducing U.S. imports of crude petroleum or refined fuels can reduce the magnitude of these external costs. Any reduction in their total value that results from lower fuel consumption and petroleum imports represents an economic benefit of setting more stringent standards over and above the dollar value of fuel savings itself. Since the agencies are taking a global perspective with respect to the estimate of the social cost of carbon for this rulemaking, the agencies do not include the value of any reduction in monopsony payments as a benefit from lower fuel consumption, because those payments from a global perspective represent a transfer of income from consumers of petroleum products to oil suppliers rather than a savings in real economic resources. Similarly, the agencies do not include any savings in budgetary outlays to support U.S. military activities among the benefits of higher fuel economy and the resulting fuel savings. Based on a recently-updated ORNL study, we estimate that each gallon of fuel saved that results in a reduction in U.S. petroleum imports (either crude petroleum or refined fuel) will reduce the expected costs of oil supply disruptions to the U.S. economy by $0.169 (2007$). Each gallon of fuel saved as a consequence of higher standards is anticipated to reduce total U.S. imports of crude petroleum or refined fuel by 0.95 gallons.

The energy security analysis conducted for this rule estimates that the world price of oil will fall modestly in response to lower U.S. demand for refined fuel. One potential result of this decline in the world price of oil would be an increase in the consumption of petroleum products outside the U.S., which would in turn lead to a modest increase in emissions of greenhouse gases, criteria air pollutants, and airborne toxics from their refining and use. While additional information would be needed to analyze this “leakage effect” in detail, NHTSA provides a sample estimate of its potential magnitude in its Final EIS. This analysis indicates that the leakage effect is likely to offset only a modest fraction of the reductions in emissions projected to result from the rule.

EPA and NHTSA received comments about the treatment of the monopsony effect, macroeconomic disruption effect, and the military costs associated with the energy security benefits of this rule. The agencies did not receive any comments that justify changing the energy security analysis. As a result, the agencies continue to only use the macroeconomic disruption component of the energy security analysis under a global context when estimating the total energy security benefits associated with this rule. Further, the Agencies did not receive any information that they could use to quantity that component of military costs directly related to energy security, and thus did not modify that part of its analysis. A more complete discussion of the energy security analysis can be found in Chapter 4 of the Joint TSD, and Sections III and IV of this preamble.

• Air pollutant emissions

○ Impacts on criteria air pollutant emissions— While reductions in domestic fuel refining and distribution that result from lower fuel consumption will reduce U.S. emissions of criteria pollutants, additional vehicle use associated with the rebound effect will increase emissions of these pollutants. Thus the net effect of stricter standards on emissions of each criteria pollutant depends on the relative magnitudes of reduced emissions from fuel refining and distribution, and increases in emissions resulting from added vehicle use. Criteria air pollutants emitted by vehicles and during fuel production include carbon monoxide (CO), hydrocarbon compounds (usually referred to as “volatile organic compounds,” or VOC), nitrogen oxides (NO_X), fine particulate matter (PM_2.5), and sulfur oxides (SO_X). It is assumed that the emission rates (per mile) stay constant for future year vehicles.

○ Economic value of reductions in criteria air pollutants—For the purpose of the joint technical analysis, EPA and NHTSA estimate the economic value of the human health benefits associated with reducing exposure to PM_2.5 using a “benefit-per-ton” method. These PM_2.5-related benefit-per-ton estimates provide the total monetized benefits to human health (the sum of reductions in premature mortality and premature morbidity) that result from eliminating one ton of directly emitted PM_2.5, or one ton of a pollutant that contributes to secondarily-formed PM_2.5 (such as NO_X, SO_X, and VOCs), from a specified source. Chapter 4.2.9 of the Technical Support Document that accompanies this rule includes a description of these values. Separately, EPA also conducted air quality modeling to estimate the change in ambient concentrations of criteria pollutants and used this as a basis for estimating the human health benefits and their economic value. Section III.H.7 presents these benefits estimates.

○ Reductions in GHG emissions—Emissions of carbon dioxide and other GHGs occur throughout the process of producing and distributing transportation fuels, as well as from fuel combustion itself. By reducing the volume of fuel consumed by passenger cars and light trucks, higher standards will thus reduce GHG emissions generated by fuel use, as well as throughout the fuel supply cycle. The agencies estimated the increases of GHGs other than CO_2, including methane and nitrous oxide, from additional vehicle use by multiplying the increase in total miles driven by cars and light trucks of each model year and age by emission rates per vehicle-mile for these GHGs. These emission rates, which differ between cars and light trucks as well as between gasoline and diesel vehicles, were estimated by EPA using its recently-developed Motor Vehicle Emission Simulator (Draft MOVES 2010). Increases in emissions of non-CO₂ GHGs are converted to equivalent increases in CO₂ emissions using estimates of the Global Warming Potential (GWP) of methane and nitrous oxide.

○ Economic value of reductions in CO₂emissions —EPA and NHTSA assigned a dollar value to reductions in CO₂ emissions using the marginal dollar value (i.e., cost) of climate-related damages resulting from carbon emissions, also referred to as “social cost of carbon” (SCC). The SCC is intended to measure the monetary value society places on impacts resulting from increased GHGs, such as property damage from sea level rise, forced migration due to dry land loss, and mortality changes associated with vector-borne diseases. Published estimates of the SCC vary widely as a result of uncertainties about future economic growth, climate sensitivity to GHG emissions, procedures used to model the economic impacts of climate change, and the choice of discount rates.

EPA and NHTSA received extensive comments about how to improve the characterization of the SCC and have since developed new estimates through an interagency modeling exercise. The comments addressed various issues, such as discount rate selection, treatment of uncertainty, and emissions and socioeconomic trajectories, and justified the revision of SCC for the final rule. The modeling exercise involved running three integrated assessment models using inputs agreed upon by the interagency group for climate sensitivity, socioeconomic and emissions trajectories, and discount rates. A more complete discussion of SCC can be found in the Technical Support Document, Social Cost of Carbon for Regulatory Impact Analysis Under Executive Order 12866 (hereafter, “SCC TSD”); revised SCC estimates corresponding to assumed values of the discount rate are shown in Table II.F-1.

• Discounting future benefits and costs—Discounting future fuel savings and other benefits is intended to account for the reduction in their value to society when they are deferred until some future date, rather than received immediately. The discount rate expresses the percent decline in the value of these benefits—as viewed from today's perspective—for each year they are deferred into the future. In evaluating the non-climate related benefits of the final standards, the agencies have employed discount rates of both 3 percent and 7 percent. We received some comments on the discount rates used in the proposal, most of which were directed at the discount rates used to value future fuel savings and the rates used to value of the social cost of carbon. In general, commenters were supporting one of the discount rates over the other, although some suggested that our rates were too high or too low. We have revised the discounting used when calculating the net present value of social cost of carbon as explained in Sections III.H. and VI but have not revised our discounting procedures for other costs or benefits.

For the reader's reference, Table II.F-2 below summarizes the values used to calculate the impacts of each final standard. The values presented in this table are summaries of the inputs used for the models; specific values used in the agencies' respective analyses may be aggregated, expanded, or have other relevant adjustments. See the respective RIAs for details.

The agencies recognize that each of these values has some degree of uncertainty, which the agencies further discuss in the Joint TSD. The agencies have conducted a range of sensitivities and present them in their respective RIAs. For example, NHTSA has conducted a sensitivity analysis on several assumptions including (1) forecasts of future fuel prices, (2) the discount rate applied to future benefits and costs, (3) the magnitude of the rebound effect, (4) the value to the U.S. economy of reducing carbon dioxide emissions, (5) inclusion of the monopsony effect, and (6) the reduction in external economic costs resulting from lower U.S. oil imports. This information is provided in NHTSA's RIA.

G. What are the estimated safety effects of the final MYs 2012-2016 CAFE and GHG standards?

The primary goals of the final CAFE and GHG standards are to reduce fuel consumption and GHG emissions, but in addition to these intended effects, the agencies must consider the potential of the standards to affect vehicle safety, which the agencies have assessed in evaluating the appropriate levels at which to set the final standards. Safety trade-offs associated with fuel economy increases have occurred in the past, and the agencies must be mindful of the possibility of future ones. These past safety trade-offs occurred because manufacturers chose, at the time, to build smaller and lighter vehicles—partly in response to CAFE standards—rather than adding more expensive fuel-saving technologies (and maintaining vehicle size and safety), and the smaller and lighter vehicles did not fare as well in crashes as larger and heavier vehicles. Historically, as shown in FARS data analyzed by NHTSA, the safest vehicles have been heavy and large, while the vehicles with the highest fatal-crash rates have been light and small, both because the crash rate is higher for small/light vehicles and because the fatality rate per crash is higher for small/light vehicle crashes.

Changes in relative safety are related to shifts in the distribution of vehicles on the road. A policy that induces a widening in the size distribution of vehicles on the road, could result in negative impacts on safety, The primary mechanism in this rulemaking for mitigating the potential negative effects on safety is the application of footprint-based standards, which create a disincentive for manufacturers to produce smaller-footprint vehicles. This is because as footprint decreases, the corresponding fuel economy/GHG emission target becomes more stringent. The shape of the footprint curves themselves have also been designed to be approximately “footprint neutral” within the sloped portion of the functions—that is, to neither encourage manufacturers to increase the footprint of their fleets, nor to decrease it. Upsizing also is discouraged through a “cut-off” at larger footprints. For both cars and light trucks there is a “cut-off” that affects vehicles smaller than 41 square feet. The agencies recognize that for manufacturers who make small vehicles in this size range, this cut off creates some incentive to downsize (i.e. further reduce the size and/or increase the production of models currently smaller than 41 square feet) to make it easier to meet the target. The cut off may also create some incentive for manufacturers who do not currently offer such models to do so in the future. However, at the same time, the agencies believe that there is a limit to the market for cars smaller than 41 square feet—most consumers likely have some minimum expectation about interior volume, among other things. In addition, vehicles in this market segment are the lowest price point for the light-duty automotive market, with a number of models in the $10,000 to $15,000 range. In order to justify selling more vehicles in this market in order to generate fuel economy or CO₂ credits (that is, for this final rule to be the incentive for selling more vehicles in this small car segment), a manufacturer would need to add additional technology to the lowest price segment vehicles, which could be challenging. Therefore, due to these two reasons (a likely limit in the market place for the smallest sized cars and the potential consumer acceptance difficulty in adding the necessary technologies in order to generate fuel economy and CO₂ credits), the agencies believe that the incentive for manufacturers to increase the sale of vehicles smaller than 41 square feet due to this rulemaking, if present, is small. For further discussion on these aspects of the standards, please see Section II.C above and Chapter 2 of the Joint TSD.

Manufacturers have stated, however, that they will reduce vehicle weight as one of the cost-effective means of increasing fuel economy and reducing CO₂ emissions, and the agencies have incorporated this expectation into our modeling analysis supporting today's final standards. NHTSA's previous analyses examining the relationship between vehicle mass and fatalities found fatality increases as vehicle weight and size were reduced, but these previous analyses did not differentiate between weight reductions and size (i.e., weight and footprint) reductions.

The question of the effect of changes in vehicle mass on safety in the context of fuel economy is a complex question that poses serious analytic challenges and has been a contentious issue for many years, as discussed by a number of commenters to the NPRM. This contentiousness arises, at least in part, from the difficulty of isolating vehicle mass from other confounding factors (e.g., driver behavior, or vehicle factors such as engine size and wheelbase). In addition, several vehicle factors have been closely related historically, such as vehicle mass, wheelbase, and track width. The issue has been reviewed and analyzed in the literature for more than two decades. For the reader's reference, much more information about safety in the CAFE context is available in Chapter IX of NHTSA's FRIA. Chapter 7.6 of EPA's final RIA also contained additional discussion on mass and safety.

Over the past several years, as also discussed by a number of commenters to the NPRM, contention has arisen with regard to the applicability of analysis of historical crash data to future safety effects due to mass reduction. The agencies recognize that there are a host of factors that may make future mass reduction different than what is reflected in the historical data. For one, the footprint-based standards have been carefully developed by the agencies so that they do not encourage vehicle footprint reductions as a way of meeting the standards, but so that they do encourage application of fuel-saving technologies, including mass reduction. This in turn encourages manufacturers to find ways to separate mass reduction from footprint reduction, which will very likely result in a future relationship between mass and fatalities that is safer than the historical relationship. However, as manufacturers pursue these methods of mass reduction, the fleet moves further away from the historical trends, which the agencies recognize.

NHTSA's NPRM analysis of the safety effects of the proposed CAFE standards was based on NHTSA's 2003 report concerning mass and size reduction in MYs 1991-1999 vehicles, and evaluated a “worst-case scenario” in which the safety effects of the combined reductions of both mass and size for those vehicles were determined for the future passenger car and light truck fleets. In the NPRM analysis, mass and size could not be separated from one another, resulting in what NHTSA recognized was a larger safety disbenefit than was likely under the MYs 2012-2016 footprint-based CAFE standards. NHTSA emphasized, however, that actual fatalities would likely be less than these “worst-case” estimates, and possibly significantly less, based on the various factors discussed in the NPRM that could reduce the estimates, such as careful mass reduction through material substitution, etc.

For the final rule, as discussed in the NPRM and in recognition of the importance of conducting analysis that better reflects, within the limits of our current knowledge, the potential safety effects of future mass reduction in response to the final CAFE and GHG standards that is highly unlikely to involve concurrent reductions in footprint, NHTSA has revised its analysis in consultation with EPA. Perhaps the most important change has been that NHTSA agreed with commenters that it was both possible and appropriate to separate the effect of mass reductions from the effect of footprint reductions. NHTSA thus performed a new statistical analysis, hereafter referred to as the 2010 Kahane analysis, of the MYs 1991-99 vehicle database from its 2003 report (now including rather than excluding 2-door cars in the passenger car fleet), assessing relationships between fatality risk, mass, and footprint for both passenger cars and LTVs (light trucks and vans). As part of its results, the new report presents an “upper-estimate scenario,” a “lower-estimate scenario,” as well as an “actual regression result scenario” representing potential safety effects of future mass reductions without corresponding vehicle size reductions, that assume, by virtue of being a cross-sectional analysis of historical data, that historical relationships between vehicle mass and fatalities are maintained. The “upper-estimate scenario” and “lower-estimate scenario” are based on NHTSA's judgment as a vehicle safety agency, and are not meant to convey any more or less likelihood in the results, but more to convey a sense of bounding for potential safety effects of reducing mass while holding footprint constant. The upper-estimate scenario reflects potential safety effects given the report's finding that, using the one-step regression method of the 2003 Kahane report, the regression coefficients show that mass and footprint each accounted for about half the fatality increase associated with downsizing in a cross-sectional analysis of MYs 1991-1999 cars. A similar effect was found for lighter LTVs. Applying the same regression method to heavier LTVs, however, the coefficients indicated a significant societal fatality reduction when mass, but not footprint, is reduced in the heavier LTVs. Fatalities are reduced primarily because mass reduction in the heavier LTVs will reduce risk to occupants of the other cars and lighter LTVs involved in collisions with these heavier LTVs. Thus, even in the “upper-estimate scenario,” the potential fatality increases associated with mass reduction in the passenger cars would be to a large extent offset by the benefits of mass reduction in the heavier LTVs.

The lower-estimate scenario, in turn, reflects NHTSA's estimate of potential safety effects if future mass reduction is accomplished entirely by material substitution, smart design, and component integration, among other things, that can reduce mass without perceptibly changing a vehicle's shape, functionality, or safety performance, maintaining structural strength without compromising other aspects of safety. If future mass reduction follows this path, it could limit the added risk close to only the effects of mass per se (the ability to transfer momentum to other vehicles or objects in a collision), resulting in estimated effects in passenger cars that are substantially smaller than in the upper-estimate scenario based directly on the regression results. The lower-estimate scenario also covers both passenger cars and LTVs.

Overall, based on the new analyses, NHTSA estimated that fatality effects could be markedly less than those estimated in the “worst-case scenario” presented in the NPRM. The agencies believe that the overall effect of mass reduction in cars and LTVs may be close to zero, and may possibly be beneficial in terms of the fleet as a whole if mass reduction is carefully applied in the future (as with careful material substitution and other methods of mass reduction that can reduce mass without perceptibly changing a car's shape, functionality, or safety performance, and maintain its structural strength without making it excessively rigid). This is especially important if the mass reduction in the heavier LTVs is greater (in absolute terms) than in passenger cars, as discussed further below and in the 2010 Kahane report.

The following sections will address how the agencies addressed potential safety effects in the NPRM for the proposed standards, how commenters responded, and the work that NHTSA has done since the NPRM to revise its estimates of potential safety effects for the final rule. The final section discusses some of the agencies' plans for the future with respect to potential analysis and studies to further enhance our understanding of this important and complex issue.

1. What did the agencies say in the NPRM with regard to potential safety effects?

In the NPRM preceding these final standards, NHTSA's safety assessment derived from the agency's belief that some of these vehicle factors, namely vehicle mass and footprint, could not be accurately separated. NHTSA relied on the 2003 study by Dr. Charles Kahane, which estimates the effect of 100-pound reductions in MYs 1991-1999 heavy light trucks and vans (LTVs), light LTVs, heavy passenger cars, and light passenger cars. The study compares the fatality rates of LTVs and cars to quantify differences between vehicle types, given drivers of the same age/gender, etc. In that analysis, the effect of “weight reduction” is not limited to the effect of mass per se, but includes all the factors, such as length, width, structural strength, safety features, and size of the occupant compartment, that were naturally or historically confounded with mass in MYs 1991-1999 vehicles. The rationale was that adding length, width, or strength to a vehicle historically also made it heavier.

NHTSA utilized the relationships between mass and safety from Kahane (2003), expressed as percentage increases in fatalities per 100-pound mass reduction, and examined the mass effects assumed in the NPRM modeling analysis. While previous CAFE rulemakings had limited mass reduction as a “technology option” to vehicles over 5,000 pounds GVWR, both NHTSA's and EPA's modeling analyses in the NPRM included mass reduction of up to 5-10 percent of baseline curb weight, depending on vehicle subclass, in response to recently-submitted manufacturer product plans as well as public statements indicating that these levels were possible and likely. 5-10 percent represented a maximum bound; EPA's modeling, for example, included average vehicle weight reductions of 4 percent between MYs 2011 and 2016, although the average per-vehicle mass reduction was greater in absolute terms for light trucks than for passenger cars. NHTSA's assumptions for mass reduction were also limited by lead time such that mass reductions of 1.5 percent were included for redesigns occurring prior to MY 2014, and mass reductions of 5-10 percent were only “achievable” in redesigns occurring in MY 2014 or later. NHTSA further assumed that mass reductions would be limited to 5 percent for small vehicles (e.g., subcompact passenger cars), and that reductions of 10 percent would only be applied to the larger vehicle types (e.g., large light trucks).

Based on these assumptions of how manufacturers might comply with the standards, NHTSA examined the effects of the identifiable safety trends over the lifetime of the vehicles produced in each model year. The effects were estimated on a year-by-year basis, assuming that certain known safety trends would result in a reduction in the target population of fatalities from which the mass effects are derived. Using this method, NHTSA found a 12.6 percent reduction in fatality levels between 2007 and 2020. The estimates derived from applying Kahane's 2003 percentages to a baseline of 2007 fatalities were then multiplied by 0.874 to account for changes that the agency believed would take place in passenger car and light truck safety between the 2007 baseline on-road fleet used for that particular analysis and year 2020.

NHTSA and EPA both emphasized that the safety effect estimates in the NPRM needed to be understood in the context of the 2003 Kahane report, which is based upon a cross-sectional analysis of the actual on-road safety experience of 1991-1999 vehicles. For those vehicles, heavier usually also meant larger-footprint. Hence, the numbers in those analyses were used to predict the safety-related fatalities that could occur in the unlikely event that weight reduction for MYs 2012-2016 is accomplished entirely by reducing mass and reducing footprint. Any estimates derived from those analyses represented a “worst-case” estimate of safety effects, for several reasons.

First, manufacturers are far less likely to reduce mass by “downsizing” (making vehicles smaller overall) under the current attribute-based standards, because the standards are based on vehicle footprint. The selection of footprint as the attribute in setting CAFE and GHG standards helps to reduce the incentive to alter a vehicle's physical dimensions. This is because as footprint decreases, the corresponding fuel economy/GHG emission target becomes more stringent. The shape of the footprint curves themselves have also been designed to be approximately “footprint neutral” within the sloped portion of the functions—that is, to neither encourage manufacturers to increase the footprint of their fleets, nor to decrease it. For further discussion on these aspects of the standards, please see Section II.C above and Chapter 2 of the Joint TSD. However, as discussed in Sections III.H.1 and IV.G.6 below, the agencies acknowledge some uncertainty regarding how consumer purchases will change in response to the vehicles designed to meet the MYs 2012-2016 standards. This could potentially affect the mix of vehicles sold in the future, including the mass and footprint distribution.

As a result, the agencies found it likely that a significant portion of the mass reduction in the MY 2012-2016 vehicles would be accomplished by strategies, such as material substitution, smart design, reduced powertrain requirements, and mass compounding, that have a lesser safety effect than the prevalent 1980s strategy of simply making the vehicles smaller. The agencies noted that to the extent that future mass reductions could be achieved by these methods—without any accompanying reduction in the size or structural strength of the vehicle—then the fatality increases associated with the mass reductions anticipated by the model as a result of the proposed standards could be significantly smaller than those in the worst-case scenario.

However, even though the agencies recognized that these methods of mass reduction could be technologically feasible in the rulemaking time frame, and included them as such in our modeling analyses, the agencies diverged as to how potential safety effects accompanying such methods of mass reduction could be evaluated, particularly in relation to the worst-case scenario presented by NHTSA. NHTSA stated that it could not predict how much smaller those increases would be for any given mixture of mass reduction methods, since the data on the safety effects of mass reduction alone (without size reduction) was not available due to the low numbers of vehicles in the current on-road fleet that have utilized these technologies extensively. Further, to the extent that mass reductions were accomplished through use of light, high-strength materials, NHTSA emphasized that there would be significant additional costs that would need to be determined and accounted for than were reflected in the agency's proposal.

Additionally, NHTSA emphasized that while it thought material substitution and other methods of mass reduction could considerably lessen the potential safety effects compared to the historical trend, NHTSA also stated that it did not believe the effects in passenger cars would be smaller than zero. EPA disagreed with this, and stated in the NPRM that the safety effects could very well be smaller than zero. Even though footprint-based standards discourage downsizing as a way of “balancing out” sales of larger/heavier vehicles, they do not discourage manufacturers from reducing crush space in overhang areas or from reducing structural support as a way of taking out mass. Moreover, NHTSA's analysis had also found that lighter cars have a higher involvement rate in fatal crashes, even after controlling for the driver's age, gender, urbanization, and region of the country. Being unable to explain this clear trend in the crash data, NHTSA stated that it must assume that mass reduction is likely to be associated with higher fatal-crash rates, no matter how the weight reduction is achieved.

NHTSA also noted in the NPRM that several studies by Dynamic Research, Inc. (DRI) had been repeatedly cited to the agency in support of the proposition that reducing vehicle mass while maintaining track width and wheelbase would lead to significant safety benefits. In its 2005 studies, one of which was published and peer-reviewed through the Society of Automotive Engineers as a technical paper, DRI attempted to assess the independent effects of vehicle weight and size (in terms of wheelbase and track width) on safety, and presented results indicating that reducing vehicle weight tends to reduce fatalities, but that reducing vehicle wheelbase and track width tends to increase fatalities. DRI's analysis was based on FARS data for MYs 1985-1998 passenger cars and 1985-1997 light trucks, similar to the MYs 1991-1999 car and truck data used in the 2003 Kahane report. However, DRI included 2-door passenger cars, while the 2003 Kahane report excluded those vehicles out of concern that their inclusion could bias the results of the regression analysis, because a significant proportion of MYs 1991-1999 2-door cars were sports and “muscle” cars, which have particularly high fatal crash rates for their relatively short wheelbases compared to the rest of the fleet. While in the NPRM NHTSA rejected the results of the DRI studies based in part on this concern, the agencies note that upon further consideration, NHTSA has agreed for this final rule that the inclusion of 2-door cars in regression analysis of historical data is appropriate, and indeed has no overly-biasing effects.

The 2005 DRI studies also differed from the 2003 Kahane report in terms of their estimates of the effect of vehicle weight on rollover fatalities. The 2003 Kahane report analyzed a single variable, curb weight, as a surrogate for both vehicle size and weight, and found that curb weight reductions would increase rollover fatalities. The DRI study, in contrast, attempted to analyze curb weight, wheelbase, and track width separately, and found that curb weight reduction would decrease rollover fatalities, while wheelbase reduction and track width reduction would increase them. DRI suggested that heavier vehicles may have higher rollover fatalities for two reasons: first, because taller vehicles tend to be heavier, so the correlation between vehicle height and weight and vehicle center-of-gravity height may make heavier vehicles more rollover-prone; and second, because heavier vehicles may have been less rollover-crashworthy due to FMVSS No. 216's constant (as opposed to proportional) requirements for MYs 1995-1999 vehicles weighing more than 3,333 lbs unloaded.

Overall, DRI's 2005 studies found a reduction in fatalities for cars (580 in the first study, and 836 in the second study) and for trucks (219 in the first study, 682 in the second study) for a 100 pound reduction in curb weight without accompanying wheelbase or track width reductions. In the NPRM, NHTSA disagreed with the results of the DRI studies, out of concern that DRI's inclusion of 2-door cars in its analysis biased the results, and because NHTSA was unable to reproduce DRI's results despite repeated attempts. NHTSA stated that it agreed intuitively with DRI's conclusion that vehicle mass reductions without accompanying size reductions (as through substitution of a heavier material for a lighter one) would be less harmful than downsizing, but without supporting real-world data and unable to verify DRI's results, NHTSA stated that it could not conclude that mass reductions would result in safety benefits. EPA, in contrast, believed that DRI's results contained some merit, in particular because the study separated the effects of mass and size and EPA stated that applying them using the curb weight reductions in EPA's modeling analysis would show an overall reduction of fatalities for the proposed standards.

On balance, both agencies recognized that mass reduction could be an important tool for achieving higher levels of fuel economy and reducing CO₂ emissions, and emphasized that NHTSA's fatality estimates represented a worst-case scenario for the potential effects of the proposed standards, and that actual fatalities will be less than these estimates, possibly significantly less, based on the various factors discussed in the NPRM that could reduce the estimates. The agencies sought comment on the safety analysis and discussions presented in the NPRM.

2. What public comments did the agencies receive on the safety analysis and discussions in the NPRM?

Several dozen commenters addressed the safety issue. Claims and arguments made by commenters in response to the safety effects analysis and discussion in the NPRM tended to follow several general themes, as follows:

• NHTSA's safety effects estimates are inaccurate because they do not account for:

○ While NHTSA's study only considers vehicles from MYs 1991-1999, more recently-built vehicles are safer than those, and future vehicles will be safer still;

○ Lighter vehicles are safer than heavier cars in terms of crash-avoidance, because they handle and brake better;

○ Fatalities are linked more to other factors than mass;

○ The structure of the standards reduces/contributes to potential safety effects from mass reduction;

○ NHTSA could mitigate additional safety effects from mass reduction, if there are any, by simply regulating safety more;

○ Casualty risks range widely for vehicles of the same weight or footprint, which skews regression analysis and makes computer simulation a better predictor of the safety effects of mass reduction;

• DRI's analysis shows that lighter vehicles will save lives, and NHTSA reaches the opposite conclusion without disproving DRI's analysis;

○ Possible reasons that NHTSA and DRI have reached different conclusions:

■ NHTSA's study should distinguish between reductions in size and reductions in weight like DRI's;

■ NHTSA's study should include two-door cars;

■ NHTSA's study should have used different assumptions;

■ NHTSA's study should include confidence intervals;

• NHTSA should include a “best-case” estimate in its study;
• NHTSA should not include a “worst-case” estimate in its study;

The agencies recognize that the issue of the potential safety effects of mass reduction, which was one of the many factors considered in the balancing that led to the agencies' conclusion as to appropriate stringency levels for the MYs 2012-2016 standards, is of great interest to the public and could possibly be a more significant factor in regulators' and manufacturers' decisions with regard to future standards beyond MY 2016. The agencies are committed to analyzing this issue thoroughly and holistically going forward, based on the best available science, in order to further their closely related missions of safety, energy conservation, and environmental protection. We respond to the issues and claims raised by commenters in turn below.

NHTSA's estimates are inaccurate because NHTSA's study only considers vehicles from MYs 1991-1999, but more recently-built vehicles are safer than those, and future vehicles will be safer still

A number of commenters (CAS, Adcock, NACAA, NJ DEP, NY DEC, UCS, and Wenzel) argued that the 2003 Kahane report, on which the “worst-case scenario” in the NPRM was based, is outdated because it considers the relationship between vehicle weight and safety in MYs 1991-1999 passenger cars. These commenters generally stated that data from MYs 1991-1999 vehicles provide an inaccurate basis for assessing the relationship between vehicle weight and safety in current or future vehicles, because the fleets of vehicles now and in the future are increasingly different from that 1990s fleet (more crossovers, fewer trucks, lighter trucks, etc.), with different vehicle shapes and characteristics, different materials, and more safety features. Several of these commenters argued that NHTSA should conduct an updated analysis for the final rule using more recent data—Wenzel, for example, stated that an updated regression analysis that accounted for the recent introduction of crossover SUVs would likely find reduced casualty risk, similar to DRI's previous finding using fatality data. CEI, in contrast, argued that the “safety trade-off” would not be eliminated by new technologies and attribute-based standards, because additional weight inherently makes a vehicle safer to its own occupants, citing the 2003 Kahane report, while AISI argued that Desapriya had found that passenger car drivers and occupants are two times more likely to be injured than drivers and occupants in larger pickup trucks and SUVs.

Several commenters (Adcock, CARB, Daimler, NESCAUM, NRDC, Public Citizen, UCS, Wenzel) suggested that NHTSA's analysis was based on overly pessimistic assumptions about how manufacturers would choose to reduce mass in their vehicles, because manufacturers have a strong incentive in the market to build vehicles safely. Many of these commenters stated that several manufacturers have already committed publicly to fairly ambitious mass reduction goals in the mid-term, but several stated further that NHTSA should not assume that manufacturers will reduce the same amount of mass in all vehicles, because it is likely that they will concentrate mass reduction in the heaviest vehicles, which will improve compatibility and decrease aggressivity in the heaviest vehicles. Daimler emphasized that all vehicles will have to comply with the Federal Motor Vehicle Safety Standards, and will likely be designed to test well in NHTSA's NCAP tests.

Other commenters (Aluminum Association, CARB, CAS, ICCT, MEMA, NRDC, U.S. Steel) also emphasized the need for NHTSA to account for the safety benefits to be expected in the future from use of advanced materials for lightweighting purposes and other engineering advances. The Aluminum Association stated that advanced vehicle design and construction techniques using aluminum can improve energy management and minimize adverse safety effects of their use, but that NHTSA's safety analysis could not account for those benefits if it were based on MYs 1991-1999 vehicles. CAS, ICCT, and U.S. Steel discussed similar benefits for more recent and future vehicles built with high strength steel (HSS), although U.S. Steel cautioned that given the stringency of the proposed standards, manufacturers would likely be encouraged to build smaller and lighter vehicles in order to achieve compliance, which fare worse in head-on collisions than larger, heavier vehicles. AISI, in contrast to U.S. Steel, stated that in its research with the Auto/Steel Partnership and in programs supported by DOE, it had found that the use of new Advanced HSS steel grades could enable mass of critical crash structures, such as front rails and bumper systems, to be reduced by 25 percent without degrading performance in standard NHTSA frontal or IIHS offset instrumented crash tests compared to their “heavier counterparts.”

Agencies' response: NHTSA, in consultation with EPA and DOE, plans to begin updating the MYs 1991-1999 database on which NHTSA's safety analyses in the NPRM and final rule are based in the next several months in order to analyze the differences in safety effects against vehicles built in more recent model years. As this task will take at least a year to complete, beginning it immediately after the NPRM would not have enabled the agency to complete it and then conduct a new analysis during the period between the NPRM and the final rule.

For purposes of this final rule, however, we believe that using the same MYs 1991-1999 database as that used in the 2003 Kahane study provides a reasonable basis for attempting to estimate safety effects due to reductions in mass. While commenters often stated that updating the database would help to reveal the effect of recently-introduced lightweight vehicles with extensive material substitution, there have in fact not yet been a significant number of vehicles with substantial mass reduction/material substitution to analyze, and they must also show up in the crash databases for NHTSA to be able to add them to its analysis. Based on NHTSA's research, specifically, on three statistical analyses over a 12-year period (1991-2003) covering a range of 22 model years (1978-1999), NHTSA believes that the relationships between mass, size, and safety has only changed slowly over time, although we recognize that they may change somewhat more rapidly in the future. As the on-road fleet gains increasing numbers of vehicles with increasing amounts of different methods of mass reduction applied to them, we may begin to discern changes in the crash databases due to the presence of these vehicles, but any such changes are likely to be slow and evolutionary, particularly in the context of MYs 2000-2009 vehicles. The agencies do expect that further analysis of historical data files will continue to provide a robust and practicable basis for estimating the potential safety effects that might occur with future reductions in vehicle mass. However, we recognize that estimates derived from analysis of historical data, like estimates from any other type of analysis (including simulation-based analysis, which cannot feasibly cover all relevant scenarios), will be uncertain in terms of predicting actual future outcomes with respect to a vehicle fleet, driving population, and operating environment that does not yet exist.

The agencies also recognize that more recent vehicles have more safety features than 1990s vehicles, which are likely to make them safer overall. To account for this, NHTSA did adjust the results of both its NPRM and final rule analysis to include known safety improvements, like ESC and increases in seat belt use, that have occurred since MYs 1991-1999. However, simply because newer vehicles have more safety countermeasures, does not mean that the weight/safety relationship necessarily changes. More likely, it would change the target population (the number of fatalities) to which one would apply the weight/safety relationship. Thus, we still believe that some mass reduction techniques for both passenger cars and light trucks can make them less safe, in certain crashes as discussed in NHTSA's FRIA, than if mass had not been reduced.

As for NHTSA's assumptions about mass reduction, in its analysis, NHTSA generally assumed that lighter vehicles could be reduced in weight by 5 percent while heavier light trucks could be reduced in weight by 10 percent. NHTSA recognizes that manufacturers might choose a different mass reduction scheme than this, and that its quantification of the estimated effect on safety would be different if they did. We emphasize that our estimates are based on the assumptions we have employed and are intended to help the agency consider the potential effect of the final standards on vehicle safety. Thus, based on the 2010 Kahane analysis, reductions in weight for the heavier light trucks would have positive overall safety effects, while mass reductions for passenger cars and smaller light trucks would have negative overall safety effects.

NHTSA's estimates are inaccurate because they do not account for the fact that lighter vehicles are safer than heavier cars in terms of crash-avoidance, because they handle and brake better

ICCT stated that lighter vehicles are better able to avoid crashes because they “handle and brake slightly better,” arguing that size-based standards encourage lighter-weight car-based SUVs with “significantly better handling and crash protection” than 1996-1999 mid-size SUVs, which will reduce both fatalities and fuel consumption. ICCT stated that NHTSA did not include these safety benefits in its analysis. DRI also stated that its 2005 report found that crash avoidance improves with reduction in curb weight and/or with increases in wheelbase and track, because “Crash avoidance can depend, amongst other factors, on the vehicle directional control and rollover characteristics.” DRI argued that, therefore, “These results indicate that vehicle weight reduction tends to decrease fatalities, but vehicle wheelbase and track reduction tends to increase fatalities.”

Agencies' response: In fact, NHTSA's regression analysis of crash fatalities per million registration years measures the effects of crash avoidance, if there are any, as well as crashworthiness. Given that the historical empirical data for passenger cars show a trend of higher crash rates for lighter cars, it is unclear whether lighter cars have, in the net, superior crash avoidance, although the agencies recognize that they may have advantages in certain individual situations. EPA presents a discussion of improved accident avoidance as vehicle mass is reduced in Chapter 7.6 of its final RIA. The important point to emphasize is that it depends on the situation—it would oversimplify drastically to point to one situation in which extra mass helps or hurts and then extrapolate effects for crash avoidance across the board based on only that.

For example, the relationship of vehicle mass to rollover and directional stability is more complex than commenters imply. For rollover, it is true that if heavy pickups were always more top-heavy than lighter pickups of the same footprint, their higher center of gravity could make them more rollover-prone, yet some mass can be placed so as to lower a vehicle's center of gravity and make it less rollover-prone. For mass reduction to be beneficial in rollover crashes, then, it must take center of gravity height into account along with other factors such as passenger compartment design and structure, suspension, the presence of various safety equipment, and so forth.

Similarly, for directional stability, it is true that having more mass increases the “understeer gradient” of cars—i.e., it reinforces their tendency to proceed in a straight line and slows their response to steering input, which would be harmful where prompt steering response is essential, such as in a double-lane-change maneuver to avoid an obstacle. Yet more mass and a higher understeer gradient could help when it is better to remain on a straight path, such as on a straight road with icy patches where wheel slip might impair directional stability. Thus, while less vehicle mass can sometimes improve crash avoidance capability, there can also be situations when more vehicle mass can help in other kinds of crash avoidance.

Further, NHTSA's research suggests that additional vehicle mass may be even more helpful, as discussed in Chapter IX of NHTSA's FRIA, when the average driver's response to a vehicle's maneuverability is taken into account. Lighter cars have historically (1976-2009) had higher collision-involvement rates than heavier cars—even in multi-vehicle crashes where directional and rollover stability is not particularly an issue. Based on our analyses using nationally-collected FARS and GES data, drivers of lighter cars are more likely to be the culpable party in a 2-vehicle collision, even after controlling for footprint, the driver's age, gender, urbanization, and region of the country.

Thus, based on this data, it appears that lighter cars may not be driven as well as heavier cars, although it is unknown why this is so. If poor drivers intrinsically chose light cars (self-selection), it might be evidenced by an increase in antisocial driving behavior (such as DWI, drug involvement, speeding, or driving without a license) as car weight decreases, after controlling for driver age and gender—in addition to the increases in merely culpable driver behavior (such as failure to yield the right of way). But analyses in NHTSA's 2003 report did not show an increase in antisocial driver behavior in the lighter cars paralleling their increase in culpable involvements.

NHTSA also hypothesizes that certain aspects of lightness and/or smallness in a car may give a driver a perception of greater maneuverability that ultimately results in driving with less of a “safety margin,” e.g., encouraging them to weave in traffic. That may appear paradoxical at first glance, as maneuverability is, in the abstract, a safety plus. Yet the situation is not unlike powerful engines that could theoretically enable a driver to escape some hazards, but in reality have long been associated with high crash and fatality rates.

NHTSA's estimates are inaccurate because fatalities are linked more to other factors than mass

Tom Wenzel stated that the safety record of recent model year crossover SUVs indicates that weight reduction in this class of vehicles (small to mid-size SUVs) resulted in a reduction in fatality risk. Wenzel argued that NHTSA should acknowledge that other vehicle attributes may be as important, if not more important, than vehicle weight or footprint in terms of occupant safety, such as unibody construction as compared to ladder-frame, lower bumpers, and less rigid frontal structures, all of which make crossover SUVs more compatible with cars than truck-based SUVs.

Marc Ross commented that fatalities are linked more strongly to intrusion than to mass, and stated that research by safety experts in Japan and Europe suggests the main cause of serious injuries and deaths is intrusion due to the failure of load-bearing elements to properly protect occupants in a severe crash. Ross argued that the results from this project have “overturned the original views about compatibility,” which thought that mass and the mass ratio were the dominant factors. Since footprint-based standards will encourage the reduction of vehicle weight through materials substitution while maintaining size, Ross stated, they will help to reduce intrusion and consequently fatalities, as the lower weight reduces crash forces while maintaining size preserves crush space. Ross argued that this factor was not considered by NHTSA in its discussion of safety. ICCT agreed with Ross' comments on this issue.

In previous comments on NHTSA rulemakings and in several studies, Wenzel and Ross have argued generally that vehicle design and “quality” is a much more important determinant of vehicle safety than mass. In comments on the NPRM, CARB, NRDC, Sierra Club, and UCS echoed this theme.

ICCT commented as well that fatality rates in the EU are much lower than rates in the U.S., even though the vehicles in the EU fleet tend to be smaller and lighter than those in the U.S. fleet. Thus, ICCT argued, “This strongly supports the idea that vehicle and highway design are far more important factors than size or weight in vehicle safety.” ICCT added that “It also suggests that the rise in SUVs in the U.S. has not helped reduce fatalities.” CAS also commented that Germany's vehicle fleet is both smaller and lighter than the American fleet, and has lower fatality rates.

Agencies' response: NHTSA and EPA agree that there are many features that affect safety. While crossover SUVs have lower fatality rates than truck-based SUVs, there are no analyses that attribute the improved safety to mass alone, and not to other factors such as the lower center of gravity or the unibody construction of these vehicles. While a number of improvements in safety can be made, they do not negate the potential that another 100 lbs. could make a passenger car or crossover vehicle safer for its occupants, because of the effects of mass per se as discussed in NHTSA's FRIA, albeit similar mass reductions could make heavier LTVs safer to other vehicles without necessarily harming their own drivers and occupants. Moreover, in the 2004 response to docket comments, NHTSA explained that the significant relationship between mass and fatality risk persisted even after controlling for vehicle price or nameplate, suggesting that vehicle “quality” as cited by Wenzel and Ross is not necessarily more important than vehicle mass.

As for reductions in intrusions due to material substitution, the agencies agree generally that the use of new and innovative materials may have the potential to reduce crash fatalities, but such vehicles have not been introduced in large numbers into the vehicle fleet. The agencies will continue to monitor the situation, but ultimately the effects of different methods of mass reduction on overall safety in the real world (not just in simulations) will need to be analyzed when vehicles with these types of mass reduction are on the road in sufficient quantities to provide statistically significant results. For example, a vehicle that is designed to be much stiffer to reduce intrusion is likely to have a more severe crash pulse and thus impose greater forces on the occupants during a crash, and might not necessarily be good for elderly and child occupant safety in certain types of crashes. Such trade-offs make it difficult to estimate overall results accurately without real world data. The agencies will continue to evaluate and analyze such real world data as it becomes available, and will keep the public informed as to our progress.

ICCT's comment illustrates the fact that different vehicle fleets in different countries can face different challenges. NHTSA does not believe that the fact that the EU vehicle fleet is generally lighter than the U.S. fleet is the exclusive reason, or even the primary factor, for the EU's lower fatality rates. The data ICCT cites do not account for significant differences between the U.S. and EU such as in belt usage, drunk driving, rural/urban roads, driving culture, etc.

The structure of the standards reduces/contributes to potential safety risks from mass reduction

Since switching in 2006 to setting attribute-based light truck CAFE standards, NHTSA has emphasized that one of the benefits of a footprint-based standard is that it discourages manufacturers from building smaller, less safe vehicles to achieve CAFE compliance by “balancing out” their larger vehicles, and thus avoids a negative safety consequence of increasing CAFE stringency. Some commenters on the NPRM (Daimler, IIHS, NADA, NRDC, Sierra Club et al.) agreed that footprint-based standards would protect against downsizing and help to mitigate safety risks, while others stated that there would still be safety risks even with footprint-based standards—CEI, for example, argued that mass reduction inherently creates safety risks, while IIHS and Porsche expressed concern about footprint-based standards encouraging manufacturers to manipulate wheelbase, which could reduce crush space and worsen vehicle handling. U.S. Steel and AISI both commented that the “aggressive schedule” for the proposed increases in stringency could encourage manufacturers to build smaller, lighter vehicles in order to comply.

Some commenters also focused on the shape and stringency of the target curves and their potential effect on vehicle safety. IIHS agreed with the agencies' tentative decision to cut off the target curves at the small-footprint end. Regarding the safety effect of the curves requiring less stringent targets for larger vehicles, while IIHS stated that increasing footprint is good for safety, CAS, Wenzel, and the UCSB students stated that decreasing footprint may be better for safety in terms of risk to occupants of other vehicles. Daimler, Wenzel, and the University of PA Environmental Law Project commented generally that more similar passenger car and light truck targets at identical footprints (as Wenzel put it, a single target curve) would improve fleet compatibility and thus, safety, by encouraging manufacturers to build more passenger cars instead of light trucks.

Agencies' response: The agencies continue to believe that footprint-based standards help to mitigate potential safety risks from downsizing if the target curves maintain sufficient slope, because, based on NHTSA's analysis, larger-footprint vehicles are safer than smaller-footprint vehicles. The structure of the footprint-based curves will also discourage the upsizing of vehicles. Nevertheless, we recognize that footprint-based standards are not a panacea—NHTSA's analysis continues to show that there was a historical relationship between lower vehicle mass and increased safety risk in passenger cars even if footprint is maintained, and there are ways that manufacturers may increase footprint that either improve or reduce vehicle safety, as indicated by IIHS and Porsche.

With regard to whether the agencies should set separate curves or a single one, NHTSA also notes in Section II.C that EPCA requires NHTSA to establish standards separately for passenger cars and light trucks, and thus concludes that the standards for each fleet should be based on the characteristics of vehicles in each fleet. In other words, the passenger car curve should be based on the characteristics of passenger cars, and the light truck curve should be based on the characteristics of light trucks—thus to the extent that those characteristics are different, an artificially-forced convergence would not accurately reflect those differences. However, such convergence could be appropriate depending on future trends in the light vehicle market, specifically further reduction in the differences between passenger car and light truck characteristics. While that trend was more apparent when car-like 2WD SUVs were classified as light trucks, it seems likely to diminish for the model year vehicles subject to these rules as the truck fleet will be more purely “truck-like” than has been the case in recent years.

NHTSA's estimates are inaccurate because NHTSA could mitigate additional safety risks from mass reduction, if there are any, by simply regulating safety more

Since NHTSA began considering the potential safety risks from mass reduction in response to increased CAFE standards, some commenters have suggested that NHTSA could mitigate those safety risks, if any, by simply regulating more. In response to the safety analysis presented in the NPRM, several commenters stated that NHTSA should develop additional safety regulations to require vehicles to be designed more safely, whether to improve compatibility (Adcock, NY DEC, Public Citizen, UCS), to require seat belt use (CAS, UCS), to improve rollover and roof crush resistance (UCS), or to improve crashworthiness generally by strengthening NCAP and the star rating system (Adcock). Wenzel commented further that “Improvements in safety regulations will have a greater effect on occupant safety than FE standards that are structured to maintain, but may actually increase, vehicle size.”

Agencies' response: NHTSA appreciates the commenters' suggestions and notes that the agency is continually striving to improve motor vehicle safety consistent with its mission. As noted above, improving safety in other areas affects the target population that the mass/footprint relationship could affect, but it does not necessarily change the relationship.

The 2010 Kahane analysis discussed in this final rule evaluates the relative safety risk when vehicles are made lighter than they might otherwise be absent the final MYs 2012-2016 standards. It does consider the effect of known safety regulations as they are projected to affect the target population.

Casualty risks range widely for vehicles of the same weight or footprint, which skews regression analysis and makes computer simulation a better predictor of the safety effects of mass reduction

Wenzel commented that he had found, in his most recent work, after accounting for drivers and crash location, that there is a wide range in casualty risk for vehicles with the same weight or footprint. Wenzel stated that for drivers, casualty risk does generally decrease as weight or footprint increases, especially for passenger cars, but the degree of variation in the data for vehicles (particularly light trucks) at a given weight or footprint makes it difficult to say that a decrease in weight or footprint will necessarily result in increased casualty risk. In terms of risk imposed on the drivers of other vehicles, Wenzel stated that risk increases as light truck weight or footprint increases.

Wenzel further stated that because a regression analysis can only consider the average trend in the relationship between vehicle weight/size and risk, it must “ignore” vehicles that do not follow that trend. Wenzel therefore recommended that the agency employ computer crash simulations for analyzing the effect of vehicle weight reduction on safety, because they can “pinpoint the effect of specific vehicle designs on safety,” and can model future vehicles which do not yet exist and are not bound to analyzing historical data. Wenzel cited, as an example, a DRI simulation study commissioned by the Aluminum Association (Kebschull 2004), which used a computer model to simulate the effect of changing SUV mass or footprint (without changing other attributes of the vehicle) on crash outcomes, and showed a 15 percent net decrease in injuries, while increasing wheelbase by 4.5 inches while maintaining weight showed a 26 percent net decrease in serious injuries.

Agencies' response: The agencies have reviewed Mr. Wenzel's draft report for DOE to which he referred in his comments, but based on NHTSA's work do not find such a wide range of safety risk for vehicles with the same weight, although we agree there is a range of risk for a given footprint. Wenzel found that for drivers, casualty risk does generally decrease as weight or footprint increases, especially for passenger cars, and that in terms of risk imposed on the drivers of other vehicles, risk increases as light truck weight or footprint increases, but concluded that the variation in the data precluded the possibility of drawing any conclusions. In the 2010 Kahane study presented in the FRIA, NHTSA undertook a similar analysis in which it correlated weight to fatality risk for vehicles of essentially the same footprint. The “decile analysis,” provided as a check on the trend/direction of NHTSA's regression analysis, shows that societal fatality risk generally increases and rarely decreases for lighter relative to heavier cars of the same footprint. Thus, while Mr. Wenzel was reluctant to draw a conclusion, NHTSA believes that both our research and Mr. Wenzel's appear to point to the same conclusion. We agree that there is a wide range in casualty risk among cars of the same footprint, but we find that that casualty risk is correlated with weight. The correlation shows that heavier cars have lower overall societal fatality rates than lighter cars of very similar footprint.

The agencies agree that simulation can be beneficial in certain circumstances. NHTSA cautions, however, that it is difficult for a simulation analysis to capture the full range of variations in crash situations in the way that a statistical regression analysis does. Vehicle crash dynamics are complex, and small changes in initial crash conditions (such as impact angle or closing speed) can have large effects on injury outcome. This condition is a consequence of variations in the deformation mode of individual components (e.g., buckling, bending, crushing, material failure, etc.) and how those variations affect the creation and destruction of load paths between the impacting object and the occupant compartment during the crash event. It is therefore difficult to predict and assess structural interactions using computational methods when one does not have a detailed, as-built geometric and material model. Even when a complete model is available, prudent engineering assessments require extensive physical testing to verify crash behavior and safety. Despite all this, the agencies recognize that detailed crash simulations can be useful in estimating the relative structural effects of design changes over a limited range of crash conditions, and will continue to evaluate the appropriate use of this tool in the future.

Simplified crash simulations can also be valuable tools, but only when employed as part of a comprehensive analytical program. They are especially valuable in evaluating the relative effect and associated confidence intervals of feasible design alternatives. For example, the method employed by Nusholtz et al. could be used by a vehicle designer to estimate the benefit of incremental changes in mass or wheelbase as well as the tradeoffs that might be made between them once that designer has settled on a preliminary design. A key difference between the research by Nusholtz and the research by Kebschull that Mr. Wenzel cited is in their suggested applications. The former is useful in evaluating proposed alternatives early in the design process—Nusholtz specifically warns that the model provides only “general insights into the overall risk * * * and cannot be used to obtain specific response characteristics.” Mr. Wenzel implies the latter can “isolate the effect of specific design changes, such as weight reduction” and thus quantify the fleet-wide effect of substantial vehicle redesigns. Yet while Kebschull reports injury reductions to three significant digits, there is no validation that vehicle structures of the proposed weight and stiffness are even feasible with current technology. Thus, while the agencies agree that computer simulations can be useful tools, we also recognize the value of statistical regression analysis for determining fleet-wide effects, because it inherently incorporates real-world factors in historical safety assessments.

DRI's analysis shows that lighter vehicles will save lives, and NHTSA reaches the opposite conclusion without disproving DRI's analysis

The difference between NHTSA's results and DRI's results for the relationship between vehicle mass and vehicle safety has been at the crux of this issue for several years. While NHTSA offered some theories in the NPRM as to why DRI might have found a safety benefit for mass reduction, NHTSA's work since then has enabled it to identify what we believe is the most likely reason for DRI's findings. The potential near multicollinearity of the variables of curb weight, track width, and wheelbase creates some degree of concern that any regression models with those variables could inaccurately calibrate their effects. However, based on its own experience with statistical analysis, NHTSA believes that the specific two-step regression model used by DRI increases this concern, because it weakens relationships between curb weight and dependent variables by splitting the effect of curb weight across the two regression steps.

The comments below are in response to NHTSA's theories in the NPRM about the source of the differences between NHTSA's and DRI's results. The majority of them are answered more fully in the 2010 Kahane report included in NHTSA's FRIA, but we respond to them in this document as well for purposes of completeness.

NHTSA and DRI may have reached different conclusions because NHTSA's study does not distinguish between reductions in size and reductions in weight like DRI's

Several commenters (CARB, CBD, EDF, ICCT, NRDC, and UCS) stated that DRI had been able to separate the effect of size and weight in its analysis, and in so doing proved that there was a safety benefit to reducing weight without reducing size. The commenters suggested that if NHTSA properly distinguished between reductions in size and reductions in weight, it would find the same result as DRI.

Agencies' response: In the 2010 Kahane analysis presented in the FRIA, NHTSA did attempt to separate the effects of vehicle size and weight by performing regression analyses with footprint (or alternatively track width and wheelbase) and curb weight as separate independent variables. For passenger cars, NHTSA found that the regressions attribute the fatality increase due to downsizing about equally to mass and footprint—that is, the effect of reducing mass alone is about half the effect of reducing mass and reducing footprint. Unlike DRI's results, NHTSA's regressions for passenger cars and for lighter LTVs did not find a safety benefit to reducing weight without reducing size; while NHTSA did find a safety benefit for reducing weight in the heaviest LTVs, the magnitude of the benefit as compared to DRI's was significantly smaller. NHTSA believes that these differences in results may be an artifact of DRI's two-step regression model, as explained above.

NHTSA and DRI may have reached different conclusions because NHTSA's study does not include two-door cars like DRI's

One of NHTSA's primary theories in the NPRM as to why NHTSA and DRI's results differed related to DRI's inclusion in its analysis of 2-door cars. NHTSA had excluded those vehicles from its analysis on the grounds that 2-door cars had a disproportionate crash rate (perhaps due to their inclusion of muscle and sports cars) which appeared likely to skew the regression. Several commenters argued that NHTSA should have included 2-door cars in its analysis. DRI and James Adcock stated that 2-door cars should not be excluded because they represent a significant portion of the light-duty fleet, while CARB and ICCT stated that because DRI found safety benefits whether 2-door cars were included or not, NHTSA should include 2-door cars in its analysis. Wenzel also commented that NHTSA should include 2-door cars in subsequent analyses, stating that while his analysis of MY 2000-2004 crash data from 5 states indicates that, in general, 4-door cars tend to have lower fatality risk than 2-door cars, the risk is even lower when he accounts for driver age/gender and crash location. Wenzel suggested that the increased fatality risk in the 2-door car population seemed primarily attributable to the sports cars, and that that was not sufficient grounds to exclude all 2-door cars from NHTSA's analysis.

Agencies' response: The agencies agree that 2-door cars can be included in the analysis, and NHTSA retracts previous statements that DRI's inclusion of them was incorrect. In its 2010 analysis, NHTSA finds that it makes little difference to the results whether 2-door cars are included, partially included, or excluded from the analysis. Thus, analyses of 2-door and 4-door cars combined, as well as other combinations, have been included in the analysis. That said, no combination of 2-door and 4-door cars resulted in NHTSA's finding a safety benefit for passenger cars due to mass reduction.

NHTSA and DRI may have reached different conclusions due to different assumptions

DRI commented that the differences found between its study and NHTSA's may be due to the different assumptions about the linearity of the curb weight effect and control variable for driver age, vehicle age, road conditions, and other factors. NHTSA's analysis was based on a two-piece linear model for curb weight with two different weight groups (less than 2,950 lbs., and greater than or equal to 2.950 lbs). The DRI analysis assumed a linear model for curb weight with a single weight group. Additionally, DRI stated that NHTSA's use of eight control variables (rather than three control variables like DRI used) for driver age introduces additional degrees of freedom into the regressions, which it suggested may be correlated with the curb weight, wheelbase, and track width, and/or other control variables. DRI suggested that this may also affect the results and cause or contribute to the differences in outcomes between NHTSA and DRI.

Agencies' response: NHTSA's FRIA documents that NHTSA analyzed its database using both a single parameter for weight (a linear model) and two parameters for weight (a two-piece linear model). In both cases, the logistic regression responded identically, allocating the same way between weight, wheelbase, track width, or footprint. Thus, NHTSA does not believe that the differences between its results and DRI's results are due to whether the studies used a single weight group or two weight groups.

The FRIA also documents that NHTSA examined NHTSA's use of eight control variables for driver age (ages 14-30, 30-50, 50-70, 70+ for males and females separately, versus DRI's use of three control variables for age (FEMALE = 1 for females, 0 for males, YOUNGDRV = 35-AGE for drivers under 35, 0 for all others, OLDMAN = AGE-50 for males over 50, 0 for all others; OLDWOMAN = AGE-45 for females over 45, 0 for all others) to see if that affected the results. NHTSA ran its analysis using the eight control variables and again using three control variables for age, and obtained similar results each time. Thus, NHTSA does not believe that the differences between its results and DRI's results are due to the number of control variables used for driver age.

NHTSA's and DRI's conclusions may be similar if confidence intervals are taken into account

DRI commented that NHTSA has not reported confidence intervals, while DRI has reported them in its studies. Thus, DRI argued, it is not possible to determine whether the confidence intervals overlap and whether the differences between NHTSA's and DRI's analyses are statistically significant.

Agencies' response: NHTSA has included confidence intervals for the main results of the 2010 Kahane analysis, as shown in Chapter IX of NHTSA's FRIA. For passenger cars, the NHTSA results are a statistically significant increase in fatalities with a 100 pound reduction while maintaining track width and wheelbase (or footprint); the DRI results are a statistically significant decrease in fatalities with a 100 pound reduction while maintaining track width and wheelbase. The DRI results are thus outside the confidence bounds of the NHTSA results and do not overlap.

NHTSA should include a “best-case” estimate in its study

Several commenters (Center for Auto Safety, NRDC, Public Citizen, Sierra Club et al., and Wenzel) urged NHTSA to include a “best-case” estimate in the final rule, showing scenarios in which lives were saved rather than lost. Public Citizen stated that there would be safety benefits to reducing the weight of the heaviest vehicles while leaving the weight of the lighter vehicles unchanged, and that increasing the number of smaller vehicles would provide safety benefits to pedestrians, bicyclists, and motorcyclists. Sierra Club et al. stated that new materials, smart design, and lighter, more advanced engines can all improve fuel economy while maintaining or increasing vehicle safety. Both Center for Auto Safety and Sierra Club argued that the agency should have presented a “best-case” scenario to balance out the “worst-case” scenario presented in the NPRM, especially if NHTSA itself believed that the worst-case scenario was not inevitable. NRDC requested that NHTSA present both a “best-case” and a “most likely” scenario. Wenzel simply stated that NHTSA did not present a “best-case” scenario, despite DRI's finding in 2005 that fatalities would be reduced if track width was held constant.

Agencies' response: NHTSA has included an “upper estimate” and a “lower estimate” in the new 2010 Kahane analysis. The lower estimate assumes that mass reduction will be accomplished entirely by material substitution or other techniques that do not perceptibly change a vehicle's shape, structural strength, or ride quality. The lower estimate examines specific crash modes and is meant to reflect the increase in fatalities for the specific crash modes in which a reduction in mass per se in the case vehicle would result in a reduction in safety: namely, collisions with larger vehicles not covered by the regulations (e.g., trucks with a GVWR over 10,000 lbs), collisions with partially-movable objects (e.g., some trees, poles, parked cars, etc.), and collisions of cars or light LTVs with heavier LTVs—as well as the specific crash modes where a reduction in mass per se in the case vehicle would benefit safety: namely, collisions of heavy LTVs with cars or lighter LTVs. NHTSA believes that this is the effect of mass per se, i.e., the effects of reduced mass will generally persist in these crashes regardless of how the mass is reduced. The lower estimate attempts to quantify that scenario, although any such estimate is hypothetical and subject to considerable uncertainty. NHTSA believes that a “most likely” scenario cannot be determined with any certainty, and would depend entirely upon agency assumptions about how manufacturers intend to reduce mass in their vehicles. While we can speculate upon the potential effects of different methods of mass reduction, we cannot predict with certainty what manufacturers will ultimately do.

NHTSA should not include a “worst-case” estimate in its study

NRDC, Public Citizen and Sierra Club et al. commented that NHTSA should remove the “worst-case scenario” estimate from the rulemaking, generally because it was based on an analysis that evaluated historical vehicles, and future vehicles would be sufficiently different to render the “worst-case scenario” inapplicable.

Agencies' response: NHTSA stated in the NPRM that the “worst-case scenario” addressed the effect of a kind of downsizing (i.e., mass reduction accompanied by footprint reduction) that was not likely to be a consequence of attribute-based CAFE standards, and that the agency would refine its analysis of such a scenario for the final rule. NHTSA has not used the “worst-case scenario” in the final rule. Instead, we present three scenarios: the first is an estimate based directly on the regression coefficients of weight reduction while maintaining footprint in the statistical analyses of historical data. As discussed above, presenting this scenario is possible because NHTSA attempted to separate the effects of weight and footprint reduction in the new analysis. However, even the new analysis of LTVs produced some coefficients that NHTSA did not consider entirely plausible. NHTSA also presents an “upper estimate” in which those coefficients for the LTVs were adjusted based on additional analyses and expert opinion as a safety agency and a “lower estimate,” which estimates the effect if mass reduction is accomplished entirely by safety-conscious technologies such as material substitution.

3. How has NHTSA refined its analysis for purposes of estimating the potential safety effects of this Final Rule?

During the past months, NHTSA has extensively reviewed the literature on vehicle mass, size, and fatality risk. NHTSA now agrees with DRI and other commenters that it is essential to analyze the effect of mass independently from the effects of size parameters such as wheelbase, track width, or footprint—and that the NPRM's “worst-case” scenario based on downsizing (in which weight, wheelbase, and track width could all be changed) is not useful for that purpose. The agency should instead provide estimates that better reflect the more likely effect of the regulation—estimating the effect of mass reduction that maintains footprint.

Yet it is more difficult to analyze multiple, independent parameters than a single parameter (e.g., curb weight), because there is a potential concern that the near multicollinearity of the parameters—the strong, natural and historical correlation of mass and size—can lead to inaccurate statistical estimates of their effects. NHTSA has performed new statistical analyses of its historical database of passenger cars, light trucks, and vans (LTVs) from its 2003 report (now including also 2-door cars), assessing relationships between fatality risk, mass, and footprint. They are described in Subsections 2.2 (cars) and 3.2 (LTVs) of the 2010 Kahane report presented in Chapter IX of the FRIA. While the potential concerns associated with near multicollinearity are inherent in regression analyses with multiple size/mass parameters, NHTSA believes that the analysis approach in the 2010 Kahane report, namely a single-step regression analysis, generally reduces those concerns and models the trends in the historical data. The results differ substantially from DRI's, based on a two-step regression analysis. Subsections 2.3 and 2.4 of the 2010 Kahane report attempt to account for the differences primarily by applying selected techniques from DRI's analyses to NHTSA's database.

The statistical analyses—logistic regressions—of trends in MYs 1991-1999 vehicles generate one set of estimates of the possible effects of reducing mass by 100 pounds while maintaining footprint. While these effects might conceivably carry over to future mass reductions, there are two reasons that future safety effects of mass reduction could differ from projections from historical data:

• The statistical analyses are “cross-sectional” analyses that estimate the increase in fatality rates for vehicles weighing n-100 pounds relative to vehicles weighing n pounds, across the spectrum of vehicles on the road, from the lightest to the heaviest. They do not directly compare the fatality rates for a specific make and model before and after a 100-pound reduction from that model. Instead, they use the differences across makes and models as a surrogate for the effects of actual reductions within a specific model; those cross-sectional differences could include trends that are statistically, but not causally related to mass.
• The manner in which mass changed across MY 1991-1999 vehicles might not be consistent with future mass reductions, due to the availability of newer materials and design methods.

Therefore, Subsections 2.5 and 3.4 of the 2010 Kahane report supplement those estimates with one or more scenarios in which some of the logistic regression coefficients are replaced by numbers based on additional analyses and NHTSA's judgment of the likely effect of mass per se (the ability to transfer momentum to other vehicles or objects in a collision) and of what trends in the historical data could be avoided by current mass-reduction technologies such as materials substitution. The various scenarios may be viewed as a plausible range of point estimates for the effects of mass reduction while maintaining footprint, but they should not be construed as upper and lower bounds. Furthermore, being point estimates, they are themselves subject to uncertainties, such as, for example, the sampling errors associated with statistical analyses.

The principal findings and conclusions of the 2010 Kahane report are as follows:

Passenger cars: This database with the one-step regression method of the 2003 Kahane report estimates an increase of 700-800 fatalities when curb weight is reduced by 100 pounds and footprint is reduced by 0.65 square feet (the historic average footprint reduction per 100-pound mass reduction in cars). The regression attributes the fatality increase about equally to curb weight and to footprint. The results are approximately the same whether 2-door cars are fully included or partially included in the analysis or whether only 4-door cars are included (as in the 2003 report). Regressions by curb weight, track width and wheelbase produce findings quite similar to the regressions by curb weight and footprint, but the results with the single “size” variable, footprint, rather than the two variables, track width and wheelbase vary even less with the inclusion or exclusion of 2-door cars.

In Subsection 2.3 of the new report, a two-step regression method that resembles (without exactly replicating) the approach by DRI, when applied to the same (NHTSA's) crash and registration data, estimates a large benefit when mass is reduced, offset by even larger fatality increases when track width and wheelbase (or footprint) are reduced. NHTSA believes that the benefit estimated by this method is inaccurate, due to the potential concerns with the near multicollinearity of the parameters (curb weight, track width, and wheelbase) even though the analysis is theoretically unbiased. Almost any analysis incorporating those parameters has a possibility of inaccurate coefficients due to near multicollinearity; however, based on our own experience with other regression analyses of crash data, NHTSA believes a DRI-type two-step method augments the possibility of estimating inaccurate coefficients for curb weight, because it weakens relationships between curb weight and dependent variables by splitting the effect of curb weight across the two regression steps.

In Subsection 2.4 of the new report, as a check on the results from the regression methods, NHTSA also performed what we refer to as “decile” analyses: Simpler, tabular data analysis that compares fatality rates of cars of different mass but similar footprint. Decile analysis is not a precise tool because it does not control for confounding factors such as driver age/gender or the specific type of car, but it may be helpful in identifying the general directional trend in the data when footprint is held constant and curb weight varies. The decile analyses show that fatality risk in MY 1991-1999 cars generally increased and rarely decreased for lighter relative to heavier cars of the same footprint. They suggest that the historical, cross-sectional trend was generally in the lighter ↔ more fatalities direction and not in the opposite direction, as might be suggested by the regression coefficients from the method that resembles DRI's approach.

The regression coefficients from NHTSA's one-step method suggest that mass and footprint each accounted for about half the fatality increase associated with downsizing in a cross-sectional analysis of 1991-1999 cars. They estimate the historical difference in societal fatality rates (i.e., including fatalities to occupants of all the vehicles involved in the collisions, plus any pedestrians) of cars of different curb weights but the same footprint. They may be considered an “upper-estimate scenario” of the effect of future mass reduction—if it were accomplished in a manner that resembled the historical cross-sectional trend—i.e., without any particular regard for safety (other than not to reduce footprint).

However, NHTSA believes that future vehicle design is likely to take advantage of safety-conscious technologies such as materials substitution that can reduce mass without perceptibly changing a car's shape or ride and maintain its structural strength. This could avoid much of the risk associated with lighter and smaller vehicles in the historical analyses, especially the historical trend toward higher crash-involvement rates for lighter and smaller vehicles. It could thereby shrink the added risk close to just the effects of mass per se (the ability to transfer momentum to other vehicles or objects in a collision). Subsection 2.5 of the 2010 Kahane report attempts to quantify a “lower-estimate scenario” for the potential effect of mass reduction achieved by safety-conscious technologies; the estimated effects are substantially smaller than in the upper-estimate scenario based directly on the regression results.

We note, again, that the preceding paragraph is conditional. Nothing in the CAFE standard requires manufacturers to use material substitution or, more generally, take a safety-conscious approach to mass reduction. Federal Motor Vehicle Safety Standards include performance tests that verify historical improvements in structural strength and crashworthiness, but few FMVSS provide test information that sheds light about how a vehicle rides or otherwise helps explain the trend toward higher crash-involvement rates for lighter and smaller vehicles. It is possible that using material substitution and other current mass reduction methods could avoid the historical trend in this area, but that remains to be studied as manufacturers introduce more of these vehicles into the on-road fleet in coming years. A detailed discussion of methods currently used for reducing the mass of passenger cars and light trucks is included in Chapter 3 of the Technical Support Document.

LTVs: The principal difference between LTVs and passenger cars is that mass reduction in the heavier LTVs is estimated to have significant societal benefits, in that it reduces the fatality risk for the occupants of cars and light LTVs that collide with the heavier LTVs. By contrast, footprint (size) reduction in LTVs has a harmful effect (for the LTVs' own occupants), as in cars. The regression method of the 2003 Kahane report applied to the database of that report estimates a societal increase of 231 fatalities when curb weight is reduced by 100 pounds and footprint is reduced by 0.975 square feet (the historic average footprint reduction per 100-pound mass reduction in LTVs). But the regressions attribute an overall reduction of 266 fatalities to the 100-pound mass reduction and an increase of 497 fatalities to the .975-square-foot footprint reduction. The regression results constitute one of the scenarios for the possible societal effects of future mass reduction in LTVs.

However, NHTSA cautions that some of the regression coefficients, even by NHTSA's preferred method, might not accurately model the historical trend in the data, possibly due to near multicollinearity of curb weight and footprint or because of the interaction of both of these variables with LTV type. Based on supplementary analyses and discussion in Subsections 3.3 and 3.4, the new report defines an additional upper-estimate scenario that NHTSA believes may more accurately reflect the historical trend in the data and a lower-estimate scenario that may come closer to the effects of mass per se. All three scenarios, however, attribute a societal fatality reduction to mass reduction in the heavier LTVs.

Overall effects of mass reduction while maintaining footprint in cars and LTVs: The immediate purpose of the new report's analyses of relationships between fatality risk, mass, and footprint is to develop the four parameters that the Volpe model needs in order to predict the safety effects, if any, of the modeled mass reductions in MYs 2012-2016 cars and LTVs over the lifetime of those vehicles. The four numbers are the overall percentage increases or decreases, per 100-pound mass reduction while holding footprint constant, in crash fatalities involving: (1) Cars < 2,950 pounds (which was the median curb weight of cars in MY 1991-1999), (2) cars ≥ 2,950 pounds, (3) LTVs < 3,870 pounds (which was the median curb weight of LTVs in those model years), and (4) LTVs ≥ 3,870 pounds. Here are the percentage effects for each of the three alternative scenarios, again, the “upper-estimate scenario” and the “lower-estimate scenario” have been developed based on NHTSA's expert opinion as a vehicle safety agency:

In all three scenarios, the estimated effects of a 100-pound mass reduction while maintaining footprint are an increase in fatalities in cars < 2,950 pounds, substantially smaller increases in cars ≥ 2,950 pounds and LTVs < 3,870 pounds, and a societal benefit for LTVs ≥ 3,870 pounds (because it reduces fatality risk to occupants of cars and lighter LTVs they collide with). These are the estimated effects of reducing each vehicle by exactly 100 pounds. However, the actual mass reduction will vary by make, model, and year. The aggregate effect on fatalities can only be estimated by attempting to forecast, as NHTSA has using inputs to the Volpe model, the mass reductions by make and model. It should be noted, however, that a 100-pound reduction would be 5 percent of the mass of a 2000-pound car but only 2 percent of a 5000-pound LTV. Thus, a forecast that mass will decrease by an equal or greater percentage in the heavier vehicles than in the lightest cars would be proportionately more influenced by the benefit for mass reduction in the heavy LTVs than by the fatality increases in the other groups; it is likely to result in an estimated net benefit under one or more of the scenarios. It should also be noted, again, that the three scenarios are point estimates and are subject to uncertainties, such as the sampling errors associated with the regression results. In the scenario based on actual regression results, the 1.96-sigma sampling errors in the above estimates are ± 0.91 percentage points for cars < 2,950 pounds and also for cars ≥ 2,950 pounds, ± 0.82 percentage points for LTVs < 3,870 pounds, and ± 1.18 percentage points for LTVs ≥ 3,870 pounds. In other words, the fatality increase in the cars < 2,950 pounds and the societal fatality reduction attributed to mass reduction in the LTVs ≥ 3,870 pounds are statistically significant. The sampling errors associated with the scenario based on actual regression results perhaps also indicate the general level of statistical noise in the other two scenarios.

4. What are the estimated safety effects of this Final Rule?

The table below shows the estimated safety effects of the modeled reduction in vehicle mass provided in the NPRM and in this final rule in order to meet the MYs 2012-2016 standards, based on the analysis described briefly above and in much more detail in Chapter IX of the FRIA. These are combined results for passenger cars and light trucks. A positive number is an estimated increase in fatalities and a negative number (shown in parentheses) is an estimated reduction in fatalities over the lifetime of the model year vehicles compared to the MY 2011 baseline fleet.

NHTSA emphasizes that the table above is based on the NHTSA's assumptions about how manufacturers might choose to reduce the mass of their vehicles in response to the final rule, which are very similar to EPA's assumptions. In general, as discussed above, the agencies assume that mass will be reduced by as much as 10 percent in the heaviest LTVs but only by as much as 5 percent in other vehicles and that substantial mass reductions will take place only in the year that models are redesigned. The actual mass reduction that is likely to occur in response to the standards will of course vary by make and model, depending on each manufacturer's particular approach, with likely more opportunity for the largest LTVs that still use separate frame construction.

The “upper estimate” presented above, as discussed in the FRIA, assumes only that manufacturers will reduce vehicle mass without reducing footprint. Thus, under such a scenario, safety effects could be somewhat adverse if, for example, manufacturers chose to reduce crush space associated with vehicle overhang as a way of reducing mass without changing footprint. The “lower estimate,” in turn, is based on the assumption that manufacturers will reduce vehicle mass solely through methods like material substitution, which (under these assumptions) fully maintain not only footprint but also all structural integrity, and other aspects of vehicle safety. Under these scenarios, safety effects could be worse if mass reduction was not undertaken thoughtfully to maintain existing safety levels, but could also be better if it was undertaken with a thorough and extensive vehicle redesign to maximize both mass reduction and safety.

And finally, while NHTSA does not believe that the “worst-case” scenario presented in the NPRM is likely to occur during the MYs 2012-2016 timeframe, we cannot guarantee that manufacturers will never choose to reduce vehicle footprint, particularly if market forces lead to increased sales of small vehicles in response to sharp increases in the price of petroleum, though this situation would not be in direct response to the CAFE/GHG standards. Thus, we cannot completely reject the worst-case scenario for all vehicles, although we can and do recognize that the footprint-based standards will significantly limit the likelihood of its occurrence within the context of this rulemaking.

In summary, the agencies recognize the balancing inherent in achieving higher levels of fuel economy and lower levels of CO₂ emissions through reduction of vehicle mass. Based on the 2010 Kahane analysis that attempts to separate the effects of mass reductions and footprint reductions, and to account better for the possibility that mass reduction will be accomplished entirely through methods that preserves structural strength and vehicle safety, the agencies now believe that the likely deleterious safety effects of the MYs 2012-2016 standards may be much lower than originally estimated. They may be close to zero, or possibly beneficial if mass reduction is carefully undertaken in the future and if the mass reduction in the heavier LTVs is greater (in absolute terms) than in passenger cars. In light of these findings, we believe that the balancing is reasonable.

5. How do the agencies plan to address this issue going forward?

NHTSA and EPA believe that it is important for the agencies to conduct further study and research into the interaction of mass, size and safety to assist future rulemakings. The agencies intend to begin working collaboratively and to explore with DOE, CARB, and perhaps other stakeholders an interagency/intergovernmental working group to evaluate all aspects of mass, size and safety. It would also be the goal of this team to coordinate government supported studies and independent research, to the extent possible, to help ensure the work is complementary to previous and ongoing research and to guide further research in this area. DOE's EERE office has long funded extensive research into component advanced vehicle materials and vehicle mass reduction. Other agencies may have additional expertise that will be helpful in establishing a coordinated work plan. The agencies are interested in looking at the weight-safety relationship in a more holistic (complete vehicle) way, and thanks to this CAFE rulemaking NHTSA has begun to bring together parts of the agency—crashworthiness, and crash avoidance rulemaking offices and the agency's Research & Development office—in an interdisciplinary way to better leverage the expertise of the agency. Extending this effort to other agencies will help to ensure that all aspects of the weight-safety relationship are considered completely and carefully with our future research. The agencies also intend to carefully consider comments received in response to the NPRM in developing plans for future studies and research and to solicit input from stakeholders.

The agencies also plan to watch for safety effects as the U.S. light-duty vehicle fleet evolves in response both to the CAFE/GHG standards and to consumer preferences over the next several years. Additionally, as new and advanced materials and component smart designs are developed and commercialized, and as manufacturers implement them in more vehicles, it will be useful for the agencies to learn more about them and to try to track these vehicles in the fleet to understand the relationship between vehicle design and injury/fatality data. Specifically, the agencies intend to follow up with study and research of the following:

First, NHTSA is in the process of contracting with an independent institution to review the statistical methods that NHTSA and DRI have used to analyze historical data related to mass, size and safety, and to provide recommendation on whether the existing methods or other methods should be used for future statistical analysis of historical data. This study will include a consideration of potential near multicollinearity in the historical data and how best to address it in a regression analysis. This study is being initiated because, in response to the NPRM, NHTSA received a number of comments related to the methodology NHTSA used for the NPRM to determine the relationship between mass and safety, as discussed in detail above.

Second, NHTSA and EPA, in consultation with DOE, intend to begin updating the MYs 1991-1999 database on which the safety analyses in the NPRM and final rule are based with newer vehicle data in the next several months. This task will take at least a year to complete. This study is being initiated in response to the NPRM comments related to the use of data from MYs 1991-1999 in the NHTSA analysis, as discussed in detail above.

Third, in order to assess if the design of recent model year vehicles that incorporate various mass reduction methods affect the relationships among vehicle mass, size and safety, NHTSA and EPA intend to conduct collaborative statistical analysis, beginning in the next several months. The agencies intend to work with DOE to identify vehicles that are using material substitution and smart design. After these vehicles are identified, the agencies intend to assess if there are sufficient data for statistical analysis. If there are sufficient data, statistical analysis would be conducted to compare the relationship among mass, size and safety of these smart design vehicles to vehicles of similar size and mass with more traditional designs. This study is being initiated because, in response to the NPRM, NHTSA received comments related to the use of data from MYs 1991-1999 in the NHTSA analysis that did not include new designs that might change the relationship among mass, size and safety, as discussed in detail above.

NHTSA may initiate a two-year study of the safety of the fleet through an analysis of the trends in structural stiffness and whether any trends identified impact occupant injury response in crashes. Vehicle manufacturers may employ stiffer light weight materials to limit occupant compartment intrusion while controlling for mass that may expose the occupants to higher accelerations resulting in a greater chance of injury in real-world crashes. This study would provide information that would increase the understanding of the effects on safety of newer vehicle designs.

In addition, NHTSA and EPA, possibly in collaboration with DOE, may conduct a longer-term computer modeling-based design and analysis study to help determine the maximum potential for mass reduction in the MYs 2017-2021 timeframe, through direct material substitution and smart design while meeting safety regulations and guidelines, and maintaining vehicle size and functionality. This study may build upon prior research completed on vehicle mass reduction. This study would further explore the comprehensive vehicle effects, including dissimilar material joining technologies, manufacturer feasibility of both supplier and OEM, tooling costs, and crash simulation and perhaps eventual crash testing.

III. EPA Greenhouse Gas Vehicle Standards

A. Executive Overview of EPA Rule

1. Introduction

The Environmental Protection Agency (EPA) is establishing GHG emissions standards for the largest sources of transportation GHGs—light-duty vehicles, light-duty trucks, and medium-duty passenger vehicles (hereafter light vehicles). These vehicle categories, which include cars, sport utility vehicles, minivans, and pickup trucks used for personal transportation, are responsible for almost 60% of all U.S. transportation related emissions of the six gases discussed above (Section I.A). This action represents the first-ever EPA rule to regulate vehicle GHG emissions under the Clean Air Act (CAA) and will establish standards for model years 2012-2016 and later light vehicles sold in the United States.

EPA is adopting three separate standards. The first and most important is a set of fleet-wide average carbon dioxide (CO₂) emission standards for cars and trucks. These standards are CO₂ emissions-footprint curves, where each vehicle has a different CO₂ emissions compliance target depending on its footprint value. Vehicle CO₂ emissions will be measured over the EPA city and highway tests. The rule allows for credits based on demonstrated improvements in vehicle air conditioner systems, including both efficiency and refrigerant leakage improvement, which are not captured by the EPA tests. The EPA projects that the average light vehicle tailpipe CO₂ level in model year 2011 will be 325 grams per mile while the average vehicle fleetwide average CO₂ emissions compliance level for the model year 2016 standard will be 250 grams per mile, an average reduction of 23 percent from today's CO₂ levels.

EPA is also finalizing standards that will cap tailpipe nitrous oxide (N₂ O) and methane (CH₄) emissions at 0.010 and 0.030 grams per mile, respectively. Even after adjusting for the higher relative global warming potencies of these two compounds, nitrous oxide and methane emissions represent less than one percent of overall vehicle greenhouse gas emissions from new vehicles. Accordingly, the goal of these two standards is to limit any potential increases of tailpipe emissions of these compounds in the future but not to force reductions relative to today's low levels.

This final rule responds to the Supreme Court's 2007 decision in Massachusetts v. EPA which found that greenhouse gases fit within the definition of air pollutant in the Clean Air Act. The Court held that the Administrator must determine whether or not emissions from new motor vehicles cause or contribute to air pollution which may reasonably be anticipated to endanger public health or welfare, or whether the science is too uncertain to make a reasoned decision. The Court further ruled that, in making these decisions, the EPA Administrator is required to follow the language of section 202(a) of the CAA. The case was remanded back to the Agency for reconsideration in light of the court's decision.

The Administrator has responded to the remand by issuing two findings under section 202(a) of the Clean Air Act. First, the Administrator found that the science supports a positive endangerment finding that the mix of six greenhouse gases (carbon dioxide (CO₂), methane (CH₄), nitrous oxide (N₂ O), hydrofluorocarbons (HFCs), perfluorocarbons (PFCs), and sulfur hexafluoride (SF₆)) in the atmosphere endangers the public health and welfare of current and future generations. This is referred to as the endangerment finding. Second, the Administrator found that the combined emissions of the same six gases from new motor vehicles and new motor vehicle engines contribute to the atmospheric concentrations of these key greenhouse gases and hence to the threat of climate change. This is referred to as the cause and contribute finding. Motor vehicles and new motor vehicle engines emit carbon dioxide, methane, nitrous oxide, and hydrofluorocarbons. EPA provides more details below on the legal and scientific bases for this final rule.

As discussed in Section I, this GHG rule is part of a joint National Program such that a large majority of the projected benefits are achieved jointly with NHTSA's CAFE rule which is described in detail in Section IV of this preamble. EPA projects total CO₂ equivalent emissions savings of approximately 960 million metric tons as a result of the rule, and oil savings of 1.8 billion barrels over the lifetimes of the MY 2012-2016 vehicles subject to the rule. EPA projects that over the lifetimes of the MY 2012-2016 vehicles, the rule will cost $52 billion but will result in benefits of $240 billion at a 3 percent discount rate, or $192 billion at a 7 percent discount rate (both values assume the average SCC value at 3%, i.e., the $21/ton SCC value in 2010). Accordingly, these light vehicle greenhouse gas emissions standards represent an important contribution under the Clean Air Act toward meeting long-term greenhouse gas emissions and import oil reduction goals, while providing important economic benefits as well. The results of our analysis of 2012-2016 MY vehicles, which we refer to as our “model year analysis,” are summarized in Tables III.H.10-4 to III.H.10-7.

We have also looked beyond the lifetimes of 2012-2016 MY vehicles at annual costs and benefits of the program for the 2012 through 2050 timeframe. We refer to this as our “calendar year” analysis (as opposed to the costs and benefits mentioned above which we refer to as our “model year analysis”). In our calendar year analysis, the new 2016 MY standards are assumed to apply to all vehicles sold in model years 2017 and later. The net present values of annual costs for the 2012 through 2050 timeframe are $346 billion for new vehicle technology which will provide $1.5 billion in fuel savings, both values at a 3 percent discount rate. At a 7 percent discount rate over the same period, the technology costs are estimated at $192 billion which will provide $673 billion in fuel savings. The social benefits during the 2012 through 2050 timeframe are estimated at $454 billion and $305 billion at a 3 and 7 percent discount rate, respectively. Both of these benefit estimates assume the average SCC value at 3% (i.e., the $21/ton SCC value in 2010). The net benefits during this time period are then $1.7 billion and $785 million at a 3 and 7 percent discount rate, respectively. The results of our “calendar year” analysis are summarized in Tables III.H 10-1 to III.H.10-3.

2. Why is EPA establishing this Rule?

This rule addresses only light vehicles. EPA is addressing light vehicles as a first step in control of greenhouse gas emissions under the Clean Air Act for four reasons. First, light vehicles are responsible for almost 60% of all mobile source GHG emissions, a share three times larger than any other mobile source subsector, and represent about one-sixth of all U.S. greenhouse gas emissions. Second, technology exists that can be readily and cost-effectively applied to these vehicles to reduce their greenhouse gas emissions in the near term. Third, EPA already has an existing testing and compliance program for these vehicles, refined since the mid-1970s for emissions compliance and fuel economy determinations, which would require only minor modifications to accommodate greenhouse gas emissions regulations. Finally, this rule is an important step in responding to the Supreme Court's ruling in Massachusetts v. EPA, which applies to other emissions sources in addition to light-duty vehicles. In fact, EPA is currently evaluating controls for motor vehicles other than those covered by this rule, and is also reviewing seven motor vehicle related petitions submitted by various states and organizations requesting that EPA use its Clean Air Act authorities to take action to reduce greenhouse gas emissions from aircraft (under § 231(a)(2)), ocean-going vessels (under § 213(a)(4)), and other nonroad engines and vehicle sources (also under § 213(a)(4)).

a. Light Vehicle Emissions Contribute to Greenhouse Gases and the Threat of Climate Change

Greenhouse gases are gases in the atmosphere that effectively trap some of the Earth's heat that would otherwise escape to space. Greenhouse gases are both naturally occurring and anthropogenic. The primary greenhouse gases of concern that are directly emitted by human activities include carbon dioxide, methane, nitrous oxide, hydrofluorocarbons, perfluorocarbons, and sulfur hexafluoride.

These gases, once emitted, remain in the atmosphere for decades to centuries. Thus, they become well mixed globally in the atmosphere and their concentrations accumulate when emissions exceed the rate at which natural processes remove greenhouse gases from the atmosphere. The heating effect caused by the human-induced buildup of greenhouse gases in the atmosphere is very likely the cause of most of the observed global warming over the last 50 years. The key effects of climate change observed to date and projected to occur in the future include, but are not limited to, more frequent and intense heat waves, more severe wildfires, degraded air quality, heavier and more frequent downpours and flooding, increased drought, greater sea level rise, more intense storms, harm to water resources, continued ocean acidification, harm to agriculture, and harm to wildlife and ecosystems. A detailed explanation of observed and projected changes in greenhouse gases and climate change and its impact on health, society, and the environment is included in EPA's technical support document for the recently promulgated Endangerment and Cause or Contribute Findings for Greenhouse Gases Under Section 202(a) of the Clean Air Act.

Mobile sources represent a large and growing share of United States greenhouse gases and include light-duty vehicles, light-duty trucks, medium-duty passenger vehicles, heavy duty trucks, airplanes, railroads, marine vessels and a variety of other sources. In 2007, all mobile sources emitted 31% of all U.S. GHGs, and were the fastest-growing source of U.S. GHGs in the U.S. since 1990. Transportation sources, which do not include certain off-highway sources such as farm and construction equipment, account for 28% of U.S. GHG emissions, and Section 202(a) sources, which include light-duty vehicles, light-duty trucks, medium-duty passenger vehicles, heavy-duty trucks, buses, and motorcycles account for 23% of total U.S. GHGs.

Light vehicles emit carbon dioxide, methane, nitrous oxide and hydrofluorocarbons. Carbon dioxide (CO₂) is the end product of fossil fuel combustion. During combustion, the carbon stored in the fuels is oxidized and emitted as CO₂ and smaller amounts of other carbon compounds. Methane (CH₄) emissions are a function of the methane content of the motor fuel, the amount of hydrocarbons passing uncombusted through the engine, and any post-combustion control of hydrocarbon emissions (such as catalytic converters). Nitrous oxide (N₂ O) (and nitrogen oxide (NO_X)) emissions from vehicles and their engines are closely related to air-fuel ratios, combustion temperatures, and the use of pollution control equipment. For example, some types of catalytic converters installed to reduce motor vehicle NO_X, carbon monoxide (CO) and hydrocarbon emissions can promote the formation of N₂ O. Hydrofluorocarbons (HFC) emissions are progressively replacing chlorofluorocarbons (CFC) and hydrochlorofluorocarbons (HCFC) in these vehicles' cooling and refrigeration systems as CFCs and HCFCs are being phased out under the Montreal Protocol and Title VI of the CAA. There are multiple emissions pathways for HFCs with emissions occurring during charging of cooling and refrigeration systems, during operations, and during decommissioning and disposal.

b. Basis for Action Under the Clean Air Act

Section 202(a)(1) of the Clean Air Act (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.” As noted above, the Administrator has found that the elevated concentrations of greenhouse gases in the atmosphere may reasonably be anticipated to endanger public health and welfare. The Administrator defined the “air pollution” referred to in CAA section 202(a) to be the combined mix of six long-lived and directly emitted GHGs: Carbon dioxide (CO₂), methane (CH₄), nitrous oxide (N₂ O), hydrofluorocarbons (HFCs), perfluorocarbons (PFCs), and sulfur hexafluoride (SF₆). The Administrator has further found under CAA section 202(a) that emissions of the single air pollutant defined as the aggregate group of these same six greenhouse gases from new motor vehicles and new motor vehicle engines contribute to air pollution. As a result of these findings, section 202(a) requires EPA to issue standards applicable to emissions of that air pollutant. New motor vehicles and engines emit CO_2, methane, N₂ O and HFC. This preamble describes the provisions that control emissions of CO₂, HFCs, nitrous oxide, and methane. For further discussion of EPA's authority under section 202(a), see Section I.C.2 of the preamble to the proposed rule (74 FR at 49464-66).

There are a variety of other CAA Title II provisions that are relevant to standards established under section 202(a). The standards are applicable to motor vehicles for their useful life. EPA has the discretion in determining what standard applies over the vehicles' useful life and has exercised that discretion in this rule. See Section II I.E. 4 below.

The standards established under CAA section 202(a) are implemented and enforced through various mechanisms. Manufacturers are required to obtain an EPA certificate of conformity before they may sell or introduce their new motor vehicle into commerce, according to CAA section 206(a). The introduction into commerce of vehicles without a certificate of conformity is a prohibited act under CAA section 203 that may subject a manufacturer to civil penalties and injunctive actions (see CAA sections 204 and 205). Under CAA section 206(b), EPA may conduct testing of new production vehicles to determine compliance with the standards. For in-use vehicles, if EPA determines that a substantial number of vehicles do not conform to the applicable regulations then the manufacturer must submit and implement a remedial plan to address the problem (see CAA section 207(c)). There are also emissions-based warranties that the manufacturer must implement under CAA section 207(a). Section III.E describes the rule's certification, compliance, and enforcement mechanisms.

c. EPA's Endangerment and Cause or Contribute Findings for Greenhouse Gases Under Section 202(a) of the Clean Air Act

On December 7, 2009 EPA's Administrator signed an action with two distinct findings regarding greenhouse gases under section 202(a) of the Clean Air Act. On December 15, 2009, the final findings were published in the Federal Register. This action is called the Endangerment and Cause or Contribute Findings for Greenhouse Gases under Section 202(a) of the Clean Air Act (Endangerment Finding). Below are the two distinct findings:

• Endangerment Finding: The Administrator finds that the current and projected concentrations of the six key well-mixed greenhouse gases—carbon dioxide (CO₂), methane (CH₄), nitrous oxide (N₂ O), hydrofluorocarbons (HFCs), perfluorocarbons (PFCs), and sulfur hexafluoride (SF₆)—in the atmosphere threaten the public health and welfare of current and future generations.
• Cause or Contribute Finding: The Administrator finds that the combined emissions of these well-mixed greenhouse gases from new motor vehicles and new motor vehicle engines contribute to the greenhouse gas pollution which threatens public health and welfare.

Specifically, the Administrator found, after a thorough examination of the scientific evidence on the causes and impact of current and future climate change, and careful review of public comments, that the science compellingly supports a positive finding that atmospheric concentrations of these greenhouse gases result in air pollution which may reasonably be anticipated to endanger both public health and welfare. In her finding, the Administrator relied heavily upon the major findings and conclusions from the recent assessments of the U.S. Climate Change Science Program and the U.N. Intergovernmental Panel on Climate Change. The Administrator made a positive endangerment finding after considering both observed and projected future effects of climate change, key uncertainties, and the full range of risks and impacts to public health and welfare occurring within the United States. In addition, the finding focused on impacts within the U.S. but noted that the evidence concerning risks and impacts occurring outside the U.S. provided further support for the finding.

The key scientific findings supporting the endangerment finding are that:

— Concentrations of greenhouse gases are at unprecedented levels compared to recent and distant past. These high concentrations are the unambiguous result of anthropogenic emissions and are very likely the cause of the observed increase in average temperatures and other climatic changes.

— The effects of climate change observed to date and projected to occur in the future include more frequent and intense heat waves, more severe wildfires, degraded air quality, heavier downpours and flooding, increasing drought, greater sea level rise, more intense storms, harm to water resources, harm to agriculture, and harm to wildlife and ecosystems. These impacts are effects on public health and welfare within the meaning of the Clean Air Act.

The Administrator found that emissions of the single air pollutant defined as the aggregate group of these same six greenhouse gases from new motor vehicles and new motor vehicle engines contribute to the air pollution and hence to the threat of climate change. Key facts supporting this cause and contribute finding for on-highway vehicles regulated under section 202(a) of the Clean Air Act are that these sources are responsible for 24% of total U.S. greenhouse gas emissions, and more than 4% of total global greenhouse gas emissions. As noted above, these findings require EPA to issue standards under section 202(a) “applicable to emission” of the air pollutant that EPA found causes or contributes to the air pollution that endangers public health and welfare. The final emissions standards satisfy this requirement for greenhouse gases from light-duty vehicles. Under section 202(a) the Administrator has significant discretion in how to structure the standards that apply to the emission of the air pollutant at issue here, the aggregate group of six greenhouse gases. EPA has the discretion under section 202(a) to adopt separate standards for each gas, a single composite standard covering various gases, or any combination of these. In this rulemaking EPA is finalizing separate standards for nitrous oxide and methane, and a CO₂ standard that provides for credits based on reductions of HFCs, as the appropriate way to issue standards applicable to emission of the single air pollutant, the aggregate group of six greenhouse gases. EPA is not setting any standards for perfluorocarbons (PFCs) or sulfur hexafluoride (SF₆) as they are not emitted by motor vehicles.

3. What is EPA adopting?

a. Light-Duty Vehicle, Light-Duty Truck, and Medium-Duty Passenger Vehicle Greenhouse Gas Emission Standards and Projected Compliance Levels

The following section provides an overview of EPA's final rule. The key public comments are not discussed here, but are discussed in the sections that follow which provide the details of the program. Comments are also discussed in the Response to Comments document.

The CO₂ emissions standards are by far the most important of the three standards and are the primary focus of this summary. As proposed, EPA is adopting an attribute-based approach for the CO₂ fleet-wide standard (one for cars and one for trucks), using vehicle footprint as the attribute. These curves establish different CO₂ emissions targets for each unique car and truck footprint. Generally, the larger the vehicle footprint, the higher the corresponding vehicle CO₂ emissions target. Table III.A.3-1 shows the greenhouse gas standards for light vehicles that EPA is finalizing for model years (MY) 2012 and later:

One important flexibility associated with the CO₂ standard is the option for manufacturers to obtain credits associated with improvements in their air conditioning systems. EPA is adopting the air conditioning provisions with minor modifications. As will be discussed in greater detail in later sections, EPA is establishing test procedures and design criteria by which manufacturers can demonstrate improvements in both air conditioner efficiency (which reduces vehicle tailpipe CO₂ by reducing the load on the engine) and air conditioner refrigerants (using lower global warming potency refrigerants and/or improving system design to reduce GHG emissions associated with leaks). Neither of these strategies to reduce GHG emissions from air conditioners will be reflected in the EPA FTP or HFET tests. These improvements will be translated to a g/mi CO₂-equivalent credit that can be subtracted from the manufacturer's tailpipe CO₂ compliance value. EPA expects a high percentage of manufacturers to use this flexibility to earn air conditioning-related credits for MY 2012-2016 vehicles such that the average credit earned is about 11 grams per mile CO₂-equivalent in 2016.

A second flexibility, being finalized essentially as proposed, is CO₂ credits for flexible and dual fuel vehicles, similar to the CAFE credits for such vehicles which allow manufacturers to gain up to 1.2 mpg in their overall CAFE ratings. The Energy Independence and Security Act of 2007 (EISA) mandated a phase-out of these flexible fuel vehicle CAFE credits beginning in 2015, and ending after 2019. EPA is allowing comparable CO₂ credits for flexible fuel vehicles through MY 2015, but for MY 2016 and beyond, the GHG rule treats flexible and dual fuel vehicles on a CO₂-performance basis, calculating the overall CO₂ emissions for flexible and dual fuel vehicles based on a fuel use-weighted average of the CO₂ levels on gasoline and on the alternative fuel, and on a manufacturer's demonstration of actual usage of the alternative fuel in its vehicle fleet.

Table III.A.3-2 summarizes EPA projections of industry-wide 2-cycle CO₂ emissions and fuel economy levels that will be achieved by manufacturer compliance with the GHG standards for MY 2012-2016.

For MY 2011, Table III.A.3-2 uses the NHTSA projections of the average fuel economy level that will be achieved by the MY 2011 fleet of 30.8 mpg for cars and 23.3 mpg for trucks, converted to an equivalent combined car and truck CO₂ level of 326 grams per mile. EPA believes this is a reasonable estimate with which to compare the MY 2012-2016 CO₂ emission standards. Identifying the proper MY 2011 estimate is complicated for many reasons, among them being the turmoil in the current automotive market for consumers and manufacturers, uncertain and volatile oil and gasoline prices, the ability of manufacturers to use flexible fuel vehicle credits to meet MY 2011 CAFE standards, and the fact that most manufacturers have been surpassing CAFE standards (particularly the car standard) in recent years. Taking all of these considerations into account, EPA believes that the MY 2011 projected CAFE achieved values, converted to CO₂ emissions levels, represent a reasonable estimate.

Table III.A.3-2 shows projected industry-wide average CO₂ emissions values. The Projected CO₂ Emissions for the Footprint-Based Standard column shows the CO₂ g/mi level corresponding with the footprint standard that must be met. It is based on the promulgated CO₂-footprint curves and projected footprint values, and will decrease each year to 250 grams per mile (g/mi) in MY 2016. For MY 2012-2016, the emissions impact of the projected utilization of flexible fuel vehicle (FFV) credits and the temporary lead-time allowance alternative standard (TLAAS, discussed below) are shown in the next two columns. The Projected CO₂ Emissions column gives the CO₂ emissions levels projected to be achieved given use of the flexible fuel credits and temporary lead-time allowance program. This column shows that, relative to the MY 2011 estimate, EPA projects that MY 2016 CO₂ emissions will be reduced by 23 percent over five years. The Projected A/C Credit column represents the industry wide average air conditioner credit manufacturers are expected to earn on an equivalent CO₂ gram per mile basis in a given model year. In MY 2016, the projected A/C credit of 10.6 g/mi represents 14 percent of the 76 g/mi CO₂ emissions reductions associated with the final standards. The Projected 2-cycle CO₂ Emissions column shows the projected CO₂ emissions as measured over the EPA 2-cycle tests, which will allow compliance with the standard assuming projected utilization of the FFV, TLAAS, and A/C credits.

EPA is also finalizing a series of flexibilities for compliance with the CO₂ standard which are not expected to significantly affect the projected compliance and achieved values shown above, but which should reduce the costs of achieving those reductions. These flexibilities include the ability to earn: Annual credits for a manufacturer's over-compliance with its unique fleet-wide average standard, early credits from MY 2009-2011, credit for “off-cycle” CO₂ reductions from new and innovative technologies that are not reflected in CO₂/fuel economy tests, as well as the carry-forward and carry-backward of credits, and the ability to transfer credits between a manufacturer's car and truck fleets. These flexibilities are being adopted with only very minor changes from the proposal, as discussed in Section III.C.

EPA is finalizing an incentive to encourage the commercialization of advanced GHG/fuel economy control technologies, including electric vehicles (EVs), plug-in hybrid electric vehicles (PHEVs), and fuel cell vehicles (FCVs), for model years 2012-2016. EPA's proposal included an emissions compliance value of zero grams/mile for EVs and FCVs, and the electric portion of PHEVs, and a multiplier in the range of 1.2 to 2.0, so that each advanced technology vehicle would count as greater than one vehicle in a manufacturer's fleet-wide compliance calculation. Several commenters were very concerned about these credits and upon considering the public comments on this issue, EPA is finalizing an advanced technology vehicle incentive program to assign a zero gram/mile emissions compliance value for EVs and FCVs, and the electric portion of PHEVs, for up to the first 200,000 EV/PHEV/FCV vehicles produced by a given manufacturer during MY 2012-2016. For any production greater than this amount, the compliance value for the vehicle will be greater than zero gram/mile, set at a level that reflects the vehicle's average net increase in upstream greenhouse gas emissions in comparison to the gasoline or diesel vehicle it replaces. EPA is not finalizing a multiplier based on the concerns potentially excessive credits using that incentive. EPA agrees that the multiplier, in combination with the zero grams/mile compliance value, would be excessive. EPA will also allow this early advanced technology incentive program beginning in MYs 2009 through 2011. Further discussion on the advanced technology vehicle incentives, including more detail on the public comments and EPA's response, is found in Section III.C.

EPA is also finalizing a temporary lead-time allowance (TLAAS) for manufacturers that sell vehicles in the U.S. in MY 2009 and for which U.S. vehicle sales in that model year are below 400,000 vehicles. This allowance will be available only during the MY 2012-2015 phase-in years of the program. A manufacturer that satisfies the threshold criteria will be able to treat a limited number of vehicles as a separate averaging fleet, which will be subject to a less stringent GHG standard. Specifically, a standard of 125 percent of the vehicle's otherwise applicable foot-print target level will apply to up to 100,000 vehicles total, spread over the four-year period of MY 2012 through 2015. Thus, the number of vehicles to which the flexibility could apply is limited. EPA also is setting appropriate restrictions on credit use for these vehicles, as discussed further in Section III. By MY 2016, these allowance vehicles must be averaged into the manufacturer's full fleet (i.e., they will no longer be eligible for a different standard). EPA discusses this in more detail in Section III.B of the preamble.

EPA received comments from several smaller manufacturers that the TLAAS program was insufficient to allow manufacturers with very limited product lines to comply. These manufacturers commented that they need additional lead-time to meet the standards, because their CO₂ baselines are significantly higher and their vehicle product lines are even more limited, reducing their ability to average across their fleets compared even to other TLAAS manufacturers. EPA fully summarizes the public comments on the TLAAS program, including comments not supporting the program, in Section III.B. In summary, in response to the lead time issues raised by manufacturers, EPA is modifying the TLAAS program that applies to manufacturers with between 5,000 and 50,000 U.S. vehicle sales in MY 2009. These manufactures would have an increased allotment of vehicles, a total of 250,000, compared to 100,000 vehicles for other TLAAS-eligible manufacturers. In addition, the TLAAS program for these manufacturers would be extended by one year, through MY 2016 for these vehicles, for a total of five years of eligibility. The other provisions of the TLAAS program would continue to apply, such as the restrictions on credit trading and the level of the standard. Additional restrictions would also apply to these vehicles, as discussed in Section III.B.5. In addition, for the smallest volume manufacturers, those with U.S. sales of below 5,000 vehicles, EPA is not setting standards at this time but is instead deferring standards until a future rulemaking. This is the same approach we are using for small businesses. The unique issues involved with these manufacturers will be addressed in that future rulemaking. Further discussion of the public comment on these issues and details on these changes from the proposed program are included in Section III.B.6. The agency received comments on its compliance with the Regulatory Flexibility Act. As stated in Section III.I.3, small entities are not significantly impacted by this rulemaking.

EPA is also adopting caps on the tailpipe emissions of nitrous oxide (N₂ O) and methane (CH₄)—0.010 g/mi for N₂ O and 0.030 g/mi for CH₄—over the EPA FTP test. While N₂ O and CH₄ can be potent greenhouse gases on a relative mass basis, their emission levels from modern vehicle designs are extremely low and represent only about 1% of total late model light vehicle GHG emissions. These cap standards are designed to ensure that N₂ O and CH₄ emissions levels do not rise in the future, rather than to force reductions in the already low emissions levels. Accordingly, these standards are not designed to require automakers to make any changes in current vehicle designs, and thus EPA is not projecting any environmental or economic costs or benefits associated with these standards.

EPA has attempted to build on existing practice wherever possible in designing a compliance program for the GHG standards. In particular, the program structure will streamline the compliance process for both manufacturers and EPA by enabling manufacturers to use a single data set to satisfy both the new GHG and CAFE testing and reporting requirements. Timing of certification, model-level testing, and other compliance activities also follow current practices established under the Tier 2 emissions and CAFE programs.

EPA received numerous comments on issues related to the impacts on stationary sources, due to the Clean Air Act's provisions for permitting requirements related to the issuance of the proposed GHG standards for new motor vehicles. Some comments suggested that EPA had underestimated the number of stationary sources that may be subject to GHG permitting requirements; other comments suggested that EPA did not adequately consider the permitting impact on small business sources. Other comments related to EPA's interpretation of the CAA's provisions for subjecting stationary sources to permit regulation after GHG standards are set. EPA's response to these comments is contained in the Response to Comments document; however, many of these comments pertain to issues that EPA is addressing in its consideration of the final Greenhouse Gas Permit Tailoring Rule, Prevention of Significant Deterioration and Title V Greenhouse Gas Tailoring Rule; Proposed Rule, 74 FR 55292 (October 27, 2009) and will thus be fully addressed in that rulemaking.

Some of the comments relating to the stationary source permitting issues suggested that EPA should defer setting GHG standards for new motor vehicles to avoid such stationary source permitting impacts. EPA is issuing these final GHG standards for light-duty vehicles as part of its efforts to expeditiously respond to the Supreme Court's nearly three year old ruling in Massachusetts v. EPA, 549 U.S. 497 (2007). In that case, the Court held that greenhouse gases fit within the definition of air pollutant in the Clean Air Act, and that EPA is therefore compelled to respond to the rulemaking petition under section 202(a) by determining whether or not emissions from new motor vehicles cause or contribute to air pollution which may reasonably be anticipated to endanger public health or welfare, or whether the science is too uncertain to make a reasoned decision. The Court further ruled that, in making these decisions, the EPA Administrator is required to follow the language of section 202(a) of the CAA. The Court stated that under section 202(a), “[i]f EPA makes [the endangerment and cause or contribute findings], the Clean Air Act requires the agency to regulate emissions of the deleterious pollutant.” 549 U.S. at 534. As discussed above, EPA has made the two findings on contribution and endangerment. 74 FR 66496 (December 15, 2009). Thus, EPA is required to issue standards applicable to emissions of this air pollutant from new motor vehicles.

The Court properly noted that EPA retained “significant latitude” as to the “timing * * * and coordination of its regulations with those of other agencies” (id.). However it has now been nearly three years since the Court issued its opinion, and the time for delay has passed. In the absence of these final standards, there would be three separate Federal and State regimes independently regulating light-duty vehicles to increase fuel economy and reduce GHG emissions: NHTSA's CAFE standards, EPA's GHG standards, and the GHG standards applicable in California and other states adopting the California standards. This joint EPA-NHTSA program will allow automakers to meet all of these requirements with a single national fleet because California has indicated that it will accept compliance with EPA's GHG standards as compliance with California's GHG standards. 74 FR at 49460. California has not indicated that it would accept NHTSA's CAFE standards by themselves. Without EPA's vehicle GHG standards, the states will not offer the Federal program as an alternative compliance option to automakers and the benefits of a harmonized national program will be lost. California and several other states have expressed strong concern that, without comparable Federal vehicle GHG standards, the states will not offer the Federal program as an alternative compliance option to automakers. Letter dated February 23, 2010 from Commissioners of California, Maine, New Mexico, Oregon and Washington to Senators Harry Reid and Mitch McConnell (Docket EPA-HQ-OAR-2009-0472-11400). The automobile industry also strongly supports issuance of these rules to allow implementation of the national program and avoid “a myriad of problems for the auto industry in terms of product planning, vehicle distribution, adverse economic impacts and, most importantly, adverse consequences for their dealers and customers.” Letter dated March 17, 2010 from Alliance of Automobile Manufacturers to Senators Harry Reid and Mitch McConnell, and Representatives Nancy Pelosi and John Boehner (Docket EPA-HQ-OAR-2009-0472-11368). Thus, without EPA's GHG standards as part of a Federal harmonized program, important GHG reductions as well as benefits to the automakers and to consumers would be lost. In addition, delaying the rule would impose significant burdens and uncertainty on automakers, who are already well into planning for production of MY 2012 vehicles, relying on the ability to produce a single national fleet. Delaying the issuance of this final rule would very seriously disrupt the industry's plans.

Instead of delaying the LDV rule and losing the benefits of this rule and the harmonized national program, EPA is directly addressing concerns about stationary source permitting in other actions that EPA is taking with regard to such permitting. That is the proper approach to address the issue of stationary source permitting, as compared to delaying the issuance of this rule for some undefined, indefinite time period.

Some parties have argued that EPA's issuance of this light-duty vehicle rule amounts to a denial of various administrative requests pending before EPA, in which parties have requested that EPA reconsider and stay the GHG endangerment finding published on December 15, 2009. That is not an accurate characterization of the impact of this final rule. EPA has not taken final action on these administrative requests, and issuance of this vehicle rule is not final agency action, explicitly or implicitly, on those requests. Currently, while we carefully consider the pending requests for reconsideration on endangerment, these final findings on endangerment and contribution remain in place. Thus under section 202(a) EPA is obligated to promulgate GHG motor vehicle standards, although there is no statutory deadline for issuance of the light-duty vehicle rule or other motor vehicle rules. In that context, issuance of this final light-duty vehicle rule does no more than recognize the current status of the findings—they are final and impose a rulemaking obligation on EPA, unless and until we change them. In issuing the vehicle rule we are not making a decision on requests to reconsider or stay the endangerment finding, and are not in any way prejudicing or limiting EPA's discretion in making a final decision on these administrative requests.

For discussion of comments on impacts on small entities and EPA's compliance with the Regulatory Flexibility Act, see the discussion in Section III.I.3.

b. Environmental and Economic Benefits and Costs of EPA's Standards

In Table III.A.3-3 EPA presents estimated annual net benefits for the indicated calendar years. The table also shows the net present values of those benefits for the calendar years 2012-2050 using both a 3 percent and a 7 percent discount rate. As discussed previously, EPA recognizes that much of these same costs and benefits are also attributable to the CAFE standard contained in this joint final rule.

4. Basis for the GHG Standards Under Section 202(a)

EPA statutory authority under section 202(a)(1) of the Clean Air Act (CAA) is discussed in more detail in Section I.C.2 of the proposed rule (74 FR at 49464-65). The following is a summary of the basis for the final GHG standards under section 202(a), which is discussed in more detail in the following portions of Section III.

With respect to CO₂ and HFCs, EPA is adopting attribute-based light-duty car and truck standards that achieve large and important emissions reductions of GHGs. EPA has evaluated the technological feasibility of the standards, and the information and analysis performed by EPA indicates that these standards are feasible in the lead time provided. EPA and NHTSA have carefully evaluated the effectiveness of individual technologies as well as the interactions when technologies are combined. EPA's projection of the technology that would be used to comply with the standards indicates that manufacturers will be able to meet the standards by employing a wide variety of technologies that are already commercially available and can be incorporated into their vehicles at the time of redesign. In addition to the consideration of the manufacturers' redesign cycle, EPA's analysis also takes into account certain flexibilities that will facilitate compliance especially in the early years of the program when potential lead time constraints are most challenging. These flexibilities include averaging, banking, and trading of various types of credits. For the industry as a whole, EPA's projections indicate that the standards can be met using technology that will be available in the lead-time provided. At the same time, it must be noted that because technology is commercially available today does not mean it can automatically be incorporated fleet-wide during the model years in question. As discussed below, and in detail in Section III.D.7, EPA and NHTSA carefully analyzed issues of adequacy of lead time in determining the level of the standards, and the agencies are convinced both that lead time is sufficient to meet the standards but that major further additions of technology across the fleet is not possible during these model years.

To account for additional lead-time concerns for various manufacturers of typically higher performance vehicles, EPA is adopting a Temporary Lead-time Allowance similar to that proposed that will further facilitate compliance for limited volumes of such vehicles in the program's initial years. For a few very small volume manufacturers, EPA is deferring standards pending later rulemaking.

EPA has also carefully considered the cost to manufacturers of meeting the standards, estimating piece costs for all candidate technologies, direct manufacturing costs, cost markups to account for manufacturers' indirect costs, and manufacturer cost reductions attributable to learning. In estimating manufacturer costs, EPA took into account manufacturers' own practices such as making major changes to model technology packages during a planned redesign cycle. EPA then projected the average cost across the industry to employ this technology, as well as manufacturer-by-manufacturer costs. EPA considers the per vehicle costs estimated from this analysis to be within a reasonable range in light of the emissions reductions and benefits received. EPA projects, for example, that the fuel savings over the life of the vehicles will more than offset the increase in cost associated with the technology used to meet the standards.

EPA has also evaluated the impacts of these standards with respect to reductions in GHGs and reductions in oil usage. For the lifetime of the model year 2012-2016 vehicles we estimate GHG reductions of approximately 960 million metric tons CO₂ eq. and fuel reductions of 1.8 billion barrels of oil. These are important and significant reductions. EPA has also analyzed a variety of other impacts of the standards, ranging from the standards' effects on emissions of non-GHG pollutants, impacts on noise, energy, safety and congestion. EPA has also quantified the cost and benefits of the standards, to the extent practicable. Our analysis to date indicates that the overall quantified benefits of the standards far outweigh the projected costs. Utilizing a 3% discount rate, we estimate the total net social benefits over the life of the model year 2012-2016 vehicles is $192 billion, and the net present value of the net social benefits of the standards through the year 2050 is $1.9 trillion dollars. These values are estimated at $136 billion and $787 billion, respectively, using a 7% discount rate and the SCC discounted at 3 percent.

Under section 202(a) EPA is called upon to set standards that provide adequate lead-time for the development and application of technology to meet the standards. EPA's standards satisfy this requirement, as discussed above. In setting the standards, EPA is called upon to weigh and balance various factors, and to exercise judgment in setting standards that are a reasonable balance of the relevant factors. In this case, EPA has considered many factors, such as cost, impacts on emissions (both GHG and non-GHG), impacts on oil conservation, impacts on noise, energy, safety, and other factors, and has, where practicable, quantified the costs and benefits of the rule. In summary, given the technical feasibility of the standard, the moderate cost per vehicle in light of the savings in fuel costs over the life time of the vehicle, the very significant reductions in emissions and in oil usage, and the significantly greater quantified benefits compared to quantified costs, EPA is confident that the standards are an appropriate and reasonable balance of the factors to consider under section 202(a). See Husqvarna AB v. EPA, 254 F. 3d 195, 200 (DC 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); see also Hercules Inc. v. EPA, 598 F. 2d 91, 106 (DC 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”); Permian Basin Area Rate Cases, 390 U.S. 747, 797 (1968) (same); Federal Power Commission v. Conway Corp., 426 U.S. 271, 278 (1976) (same); Exxon Mobil Gas Marketing Co. v. FERC, 297 F. 3d 1071, 1084 (DC Cir. 2002) (same).

EPA recognizes that the vast majority of technologies which we are considering for purposes of setting standards under section 202(a) are commercially available and already being utilized to a limited extent across the fleet. The vast majority of the emission reductions, which would result from this rule, would result from the increased use of these technologies. EPA also recognizes that this rule would enhance the development and limited use of more advanced technologies, such as PHEVs and EVs. In this technological context, there is no clear cut line that indicates that only one projection of technology penetration could potentially be considered feasible for purposes of section 202(a), or only one standard that could potentially be considered a reasonable balancing of the factors relevant under section 202(a). EPA therefore evaluated two sets of alternative standards, one more stringent than the promulgated standards and one less stringent.

The alternatives are 4% per year increase in standards which would be less stringent and a 6% per year increase in the standards which would be more stringent. EPA is not adopting either of these. As discussed in Section III.D.7, the 4% per year forgoes CO₂ reductions which can be achieved at reasonable cost and are achievable by the industry within the rule's timeframe. The 6% per year alternative requires a significant increase in the projected required technology penetration which appears inappropriate in this timeframe due to the limited available lead time and the current difficult financial condition of the automotive industry. (See Section III.D.7 for a detailed discussion of why EPA is not adopting either of the alternatives.) EPA also believes that the no backsliding standards it is adopting for N₂ O and CH₄ are appropriate under section 202(a).

B. GHG Standards for Light-Duty Vehicles, Light-Duty Trucks, and Medium-Duty Passenger Vehicles

EPA is finalizing new emission standards to control greenhouse gases (GHGs) from light-duty vehicles. First, EPA is finalizing an emission standard for carbon dioxide (CO₂) on a gram per mile (g/mile) basis that will apply to a manufacturer's fleet of cars, and a separate standard that will apply to a manufacturer's fleet of trucks. CO₂ is the primary greenhouse gas resulting from the combustion of vehicular fuels, and the amount of CO₂ emitted is directly correlated to the amount of fuel consumed. Second, EPA is providing auto manufacturers with the opportunity to earn credits toward the fleet-wide average CO₂ standards for improvements to air conditioning systems, including both hydrofluorocarbon (HFC) refrigerant losses (i.e., system leakage) and indirect CO₂ emissions related to the increased load on the engine. Third, EPA is finalizing separate emissions standards for two other GHGs: Methane (CH₄) and nitrous oxide (N₂ 0). CH₄ and N₂ O emissions relate closely to the design and efficient use of emission control hardware (i.e., catalytic converters). The standards for CH₄ and N₂ O will be set as a cap that will limit emissions increases and prevent backsliding from current emission levels. The final standards described below will apply to passenger cars, light-duty trucks, and medium-duty passenger vehicles (MDPVs). As an overall group, they are referred to in this preamble as light vehicles or simply as vehicles. In this preamble section passenger cars may be referred to simply as “cars”, and light-duty trucks and MDPVs as “light trucks” or “trucks.”

EPA's program includes a number of credit opportunities and other flexibilities to help manufacturers comply, especially in the early years of the program. EPA is establishing a system of averaging, banking, and trading of credits integral to the fleet averaging approach, based on manufacturer fleet average CO₂ performance, as discussed in Section III.B.4. This approach is similar to averaging, banking, and trading (ABT) programs EPA has established in other programs and is also similar to provisions in the CAFE program. In addition to traditional ABT credits based on the fleet emissions average, EPA is also including A/C credits as an aspect of the standards, as mentioned above. EPA is also including several additional credit provisions that apply only in the initial model years of the program. These include flex fuel vehicle credits, incentives for the early commercialization of certain advanced technology vehicles, credits for new and innovative “off-cycle” technologies that are not captured by the current test procedures, and generation of credits prior to model year 2012. The A/C credits and additional credit opportunities are described in Section III.C. These credit programs will provide flexibility to manufacturers, which may be especially important during the early transition years of the program. EPA will also allow a manufacturer to carry a credit deficit into the future for a limited number of model years. A parallel provision, referred to as credit carry-back, will be part of the CAFE program. Finally, EPA is finalizing an optional compliance flexibility, the Temporary Leadtime Allowance Alternative Standard program, for intermediate volume manufacturers, and is deferring standards for the smallest manufacturers, as discussed in Sections III.B.5 and 6 below.

1. What fleet-wide emissions levels correspond to the CO₂ standards?

The attribute-based CO₂ standards are projected to achieve a national fleet-wide average, covering both light cars and trucks, of 250 grams/mile of CO₂ in model year (MY) 2016. This includes CO₂-equivalent emission reductions from A/C improvements, reflected as credits in the standard. The standards will begin with MY 2012, with a generally linear increase in stringency from MY 2012 through MY 2016. EPA will have separate standards for cars and light trucks. The tables in this section below provide overall fleet average levels that are projected for both cars and light trucks over the phase-in period which is estimated to correspond with the standards. The actual fleet-wide average g/mi level that will be achieved in any year for cars and trucks will depend on the actual production for that year, as well as the use of the various credit and averaging, banking, and trading provisions. For example, in any year, manufacturers may generate credits from cars and use them for compliance with the truck standard. Such transfer of credits between cars and trucks is not reflected in the table below. In Section III.F, EPA discusses the year-by-year estimate of emissions reductions that are projected to be achieved by the standards.

In general, the schedule of standards acts as a phase-in to the MY 2016 standards, and reflects consideration of the appropriate lead-time for each manufacturer to implement the requisite emission reductions technology across its product line. Note that 2016 is the final model year in which standards become more stringent. The 2016 CO₂ standards will remain in place for 2017 and later model years, until revised by EPA in a future rulemaking.

EPA estimates that, on a combined fleet-wide national basis, the 2016 MY standards will achieve a level of 250 g/mile CO₂, including CO₂-equivalent credits from A/C related reductions. The derivation of the 250 g/mile estimate is described in Section III.B.2.

EPA has estimated the overall fleet-wide CO₂-equivalent emission levels that correspond with the attribute-based standards, based on the projections of the composition of each manufacturer's fleet in each year of the program. Tables III.B.1-1 and III.B.1-2 provides these estimates for each manufacturer.

As a result of public comments and updated economic and future fleet projections, the attribute based curves have been updated for this final rule, as discussed in detail in Section II.B of this preamble and Chapter 2 of the Joint TSD. This update in turn affects costs, benefits, and other impacts of the final standards—thus EPA's overall projection of the impacts of the final rule standards have been updated and the results are different than for the NPRM, though in general not by a large degree.

These estimates were aggregated based on projected production volumes into the fleet-wide averages for cars and trucks (Table III.B.1-3).

As shown in Table III.B.1-3, fleet-wide CO₂-equivalent emission levels for cars under the approach are projected to decrease from 263 to 225 grams per mile between MY 2012 and MY 2016. Similarly, fleet-wide CO₂-equivalent emission levels for trucks are projected to decrease from 346 to 398 grams per mile. These numbers do not include the effects of other flexibilities and credits in the program. The estimated achieved values can be found in Chapter 5 of the Regulatory Impact Analysis (RIA).

EPA has also estimated the average fleet-wide levels for the combined car and truck fleets. These levels are provided in Table III.B.1-4. As shown, the overall fleet average CO₂ level is expected to be 250 g/mile in 2016.

As noted above, EPA is finalizing standards that will result in increasingly stringent levels of CO₂ control from MY 2012 though MY 2016—applying the CO₂ footprint curves applicable in each model year to the vehicles expected to be sold in each model year produces fleet-wide annual reductions in CO₂ emissions. Comments from the Center for Biological Diversity (CBD) challenged EPA to increase the stringency of the standards for all of the years of the program, and even argued that 2016 standards should be feasible in 2012. Other commenters noted the non-linear increase in the standards from 2011 (CAFE) to the 2012 GHG standards. As explained in greater detail in Section III.D below and the relevant support documents, EPA believes that the level of improvement achieves important CO₂ emissions reductions through the application of feasible control technology at reasonable cost, considering the needed lead time for this program. EPA further believes that the averaging, banking and trading provisions, as well as other credit-generating mechanisms, allow manufacturers further flexibilities which reduce the cost of the CO₂ standards and help to provide adequate lead time. EPA believes this approach is justified under section 202(a) of the Clean Air Act.

EPA has analyzed the feasibility under the CAA of achieving the CO₂ standards, based on projections of what actions manufacturers are expected to take to reduce emissions. The results of the analysis are discussed in detail in Section III.D below and in the RIA. EPA also presents the estimated costs and benefits of the car and truck CO₂ standards in Section III.H. In developing the final rule, EPA has evaluated the kinds of technologies that could be utilized by the automobile industry, as well as the associated costs for the industry and fuel savings for the consumer, the magnitude of the GHG reductions that may be achieved, and other factors relevant under the CAA.

With respect to the lead time and cost of incorporating technology improvements that reduce GHG emissions, EPA and NHTSA place important weight on the fact that during MYs 2012-2016 manufacturers are expected to redesign and upgrade their light-duty vehicle products (and in some cases introduce entirely new vehicles not on the market today). Over these five model years there will be an opportunity for manufacturers to evaluate almost every one of their vehicle model platforms and add technology in a cost-effective way to control GHG emissions and improve fuel economy. This includes redesign of the air conditioner systems in ways that will further reduce GHG emissions. The time-frame and levels for the standards, as well as the ability to average, bank and trade credits and carry a deficit forward for a limited time, are expected to provide manufacturers the time needed to incorporate technology that will achieve GHG reductions, and to do this as part of the normal vehicle redesign process. This is an important aspect of the final rule, as it will avoid the much higher costs that will occur if manufacturers needed to add or change technology at times other than these scheduled redesigns. This time period will also provide manufacturers the opportunity to plan for compliance using a multi-year time frame, again in accord with their normal business practice. Further details on lead time, redesigns and feasibility can be found in Section III-D.

Consistent with the requirement of CAA section 202(a)(1) that standards be applicable to vehicles “for their useful life,” EPA is finalizing CO₂ vehicle standards that will apply for the useful life of the vehicle. Under section 202(i) of the Act, which authorized the Tier 2 standards, EPA established a useful life period of 10 years or 120,000 miles, whichever first occurs, for all Tier 2 light-duty vehicles and light-duty trucks. Tier 2 refers to EPA's standards for criteria pollutants such as NO_X, HC, and CO. EPA is finalizing new CO₂ standards for the same group of vehicles, and therefore the Tier 2 useful life will apply for CO₂ standards as well. The in-use emission standard will be 10% higher than the model-level certification emission test results, to address issues of production variability and test-to-test variability. The in-use standard is discussed in Section II I.E.

EPA is requiring manufacturers to measure CO₂ for certification and compliance purposes using the same test procedures currently used by EPA for measuring fuel economy. These procedures are the Federal Test Procedure (FTP or “city” test) and the Highway Fuel Economy Test (HFET or “highway” test). This corresponds with the data used to develop the footprint-based CO₂ standards, since the data on control technology efficiency was also developed in reference to these test procedures. Although EPA recently updated the test procedures used for fuel economy labeling, to better reflect the actual in-use fuel economy achieved by vehicles, EPA is not using these test procedures for the CO₂ standards in this final rule, given the lack of data on control technology effectiveness under these procedures. There were a number of commenters that advocated for a change in either the test procedures or the fuel economy calculation weighting factors. The U.S. Coalition for Advanced Diesel Cars urged a changing of the city/highway weighting factors from their current values of 45/55 to 43/57 to be more consistent with the EPA (5-cycle) fuel economy labeling rule. EPA has decided that such a change would not be appropriate, nor consistent with the technical analyses supporting the 5-cycle fuel economy label rulemaking. The city/highway weighting of 43/57 was found to be appropriate when the city fuel economy is based on a combination of Bags 2 and 3 of the FTP and the city portion of the US06 test cycle, and when the highway fuel economy is based on a combination of the HFET and the highway portion of the US06 cycle. When city and highway fuel economy are based on the FTP and HFET cycles, respectively, the appropriate city/highway weighting is not 43/57, but very close to 55/45. Therefore, the weighting of the city and highway fuel economy values contained in this rule is appropriate for and consistent with the use of the FTP and HFET cycles to measure city and highway fuel economy.

The American Council for an Energy-Efficient Economy (ACEEE), Cummins, and Sierra Club all suggested using more real-world test procedures. It is not feasible at this time to base the final CO₂ standards on EPA's five-cycle fuel economy formulae. Consistent with its name, these formulae require vehicle testing over five test cycles, the two cycles associated with the proposed CO₂ standards, plus the cold temperature FTP, the US06 high speed, high acceleration cycle and the SC03 air conditioning test. EPA considered employing the five-cycle calculation of fuel economy and GHG emissions for this rule, but there were a number of reasons why this was not practical. As discussed extensively in the Joint TSD, setting the appropriate levels of CO₂ standards requires extensive knowledge of the CO₂ emission control effectiveness over the certification test cycles. Such knowledge has been gathered over the FTP and HFET cycles for decades, but is severely lacking for the other three test cycles. EPA simply lacks the technical basis to project the effectiveness of the available technologies over these three test cycles and therefore, could not adequately support a rule which set CO₂ standards based on the five-cycle formulae. The benefits of today's rule do presume a strong connection between CO₂ emissions measured over the FTP and HFET cycles and onroad operation. Since CO₂ emissions determined by the five-cycle formulae are believed to correlate reasonably with onroad emissions, this implies a strong connection between emissions over the FTP and HFET cycles and the five cycle formulae. However, while we believe that this correlation is reasonable on average for the vehicle fleet, it may not be reasonable on a per vehicle basis, nor for any single manufacturer's vehicles. Thus, we believe that it is reasonable to project a direct relationship between the percentage change in CO₂ emissions over the two certification cycles and onroad emissions (a surrogate of which is the five-cycle formulae), but not reasonable to base the certification of specific vehicles on that untested relationship. Furthermore, EPA is allowing for off-cycle credits to encourage technologies that may not be not properly captured on the 2-cycle city/highway test procedure (although these credits could apply toward compliance with EPA's standards, not toward compliance with the CAFE standards). For future analysis, EPA will consider examining new drive cycles and test procedures for fuel economy.

EPA is finalizing standards that include hydrocarbons (HC) and carbon monoxide (CO) in its CO₂ emissions calculations on a CO₂-equivalent basis. It is well accepted that HC and CO are typically oxidized to CO₂ in the atmosphere in a relatively short period of time and so are effectively part of the CO₂ emitted by a vehicle. In terms of standard stringency, accounting for the carbon content of tailpipe HC and CO emissions and expressing it as CO₂-equivalent emissions will add less than one percent to the overall CO₂-equivalent emissions level. This will also ensure consistency with CAFE calculations since HC and CO are included in the “carbon balance” methodology that EPA uses to determine fuel usage as part of calculating vehicle fuel economy levels.

2. What are the CO₂ attribute-based standards?

EPA is finalizing the same vehicle category definitions that are used in the CAFE program for the 2011 model year standards. This approach allows EPA's CO₂ standards and the CAFE standards to be harmonized across all vehicles. In other words, vehicles will be subject to either car standards or truck standards under both programs, and not car standards under one program and trucks standards under the other. The CAFE vehicle category definitions differ slightly from the EPA definitions for cars and light trucks used for the Tier 2 program and other EPA vehicle programs. However, EPA is not changing the vehicle category definitions for any other light-duty mobile source programs, except the GHG standards.

EPA is finalizing separate car and truck standards, that is, vehicles defined as cars have one set of footprint-based curves for MY 2012-2016 and vehicles defined as trucks have a different set for MY 2012-2016. In general, for a given footprint the CO₂ g/mi target for trucks is less stringent then for a car with the same footprint.

Some commenters requested a single or converging curve for both cars and trucks. EPA is not finalizing a single fleet standard where all cars and trucks are measured against the same footprint curve for several reasons. First, some vehicles classified as trucks (such as pick-up trucks) have certain attributes not common on cars which attributes contribute to higher CO₂ emissions—notably high load carrying capability and/or high towing capability. Due to these differences, it is reasonable to separate the light-duty vehicle fleet into two groups. Second, EPA wishes to harmonize key program design elements of the GHG standards with NHTSA's CAFE program where it is reasonable to do so. NHTSA is required by statute to set separate standards for passenger cars and for non-passenger cars. As discussed in Section IV, EPCA does not preclude NHTSA from issuing converging standards if its analysis indicates that these are the appropriate standards under the statute applicable separately to each fleet.

Finally, most of the advantages of a single standard for all light duty vehicles are also present in the two-fleet standards finalized here. Because EPA is allowing unlimited credit transfer between a manufacturer's car and truck fleets, the two fleets can essentially be viewed as a single fleet when manufacturers consider compliance strategies. Manufacturers can thus choose on which vehicles within their fleet to focus GHG reducing technology and then use credit transfers as needed to demonstrate compliance, just as they will if there was a single fleet standard. The one benefit of a single light-duty fleet not captured by a two-fleet approach is that a single fleet prevents potential “gaming” of the car and truck definitions to try and design vehicles which are more similar to passenger cars but which may meet the regulatory definition of trucks. Although this is of concern to EPA, we do not believe at this time that concern is sufficient to outweigh the other reasons for finalizing separate car and truck fleet standards. However, it is possible that in the future, recent trends may continue such that cars may become more truck-like and trucks may become more car-like. Therefore, EPA will reconsider whether it is appropriate to use converging curves if justified by future analysis.

For model years 2012 and later, EPA is finalizing a series of CO₂ standards that are described mathematically by a family of piecewise linear functions (with respect to vehicle footprint). The form of the function is as follows:

CO₂ = a, if x ≤ l

CO₂ = cx + d, if l < x ≤ h

CO₂ = b, if x > h

Where:

CO₂ = the CO₂ target value for a given footprint (in g/mi)

a = the minimum CO₂ target value (in g/mi)

b = the maximum CO₂ target value (in g/mi)

c = the slope of the linear function (in g/mi per sq ft)

d = is the zero-offset for the line (in g/mi CO₂)

x = footprint of the vehicle model (in square feet, rounded to the nearest tenth)

l & h are the lower and higher footprint limits, constraints, or the boundary (“kinks”) between the flat regions and the intermediate sloped line

EPA's parameter values that define the family of functions for the CO₂ fleetwide average car and truck standards are as follows:

The equations can be shown graphically for each vehicle category, as shown in Figures III.B.2-1 and III.B.2-2. These standards (or functions) decrease from 2012-2016 with a vertical shift.

The EPA received a number of comments on both the attribute and the shape of the curve. For reasons described in Section IIC and Chapter 2 of the TSD, the EPA feels that footprint is the most appropriate choice of attribute for this rule. More background discussion on other alternative attributes and curves EPA explored can be found in the EPA RIA. EPA recognizes that the CAA does not mandate that EPA use an attribute based standard, as compared to NHTSA's obligations under EPCA. The EPA believes that a footprint-based program will harmonize EPA's program and the CAFE program as a single national program, resulting in reduced compliance complexity for manufacturers. EPA's reasons for using an attribute based standard are discussed in more detail in the Joint TSD. Also described in these other sections are the reasons why EPA is finalizing the slopes and the constraints as shown above. For future analysis, EPA will consider other options and suggestions made by commenters.

EPA also received public comments from three manufacturers, General Motors, Ford Motor Company, and Chrysler, suggesting that the GHG program should harmonize with an EPCA provision that allows a manufacturer to exclude emergency vehicles from its CAFE fleet by providing written notice to NHTSA. These manufacturers believe this provision is necessary because law enforcement vehicles (e.g., police cars) must be designed with special performance and features necessary for police work—but which tend to raise GHG emissions and reduce fuel economy relative to the base vehicle. These commenters provided several examples of features unique to these special purpose vehicles that negatively impact GHG emissions, such as heavy-duty suspensions, unique engine and transmission calibrations, and heavy-duty components (e.g., batteries, stabilizer bars, engine cooling). These manufacturers believe consistency in addressing these vehicles between the EPA and NHTSA programs is critical, as a manufacturer may be challenged to continue providing the performance needs of the Federal, State, and local government purchasers of emergency vehicles.

EPA is not finalizing such an emergency vehicle provision in this rule, since we believe that it is feasible for manufacturers to apply the same types of technologies to the base emergency vehicle as they would to other vehicles in their fleet. However, EPA also recognizes that, because of the unique “performance upgrading” needed to convert a base vehicle into one that meets the performance demands of the law enforcement community—which tend to reduce GHGs relative to the base vehicles—there could be situations where a manufacturer is more challenged in meeting the GHG standards than the CAFE standards, simply due to inclusion of these higher-emitting vehicles in the GHG program fleet. While EPA is not finalizing such an exclusion for emergency vehicles today, we do believe it is important to assess this issue in the future. EPA plans to assess the unique characteristics of these emergency vehicles and whether special provisions for addressing them are warranted. EPA plans to undertake this evaluation as part of a follow-up rulemaking in the next 18 months (this rulemaking is discussed in the context of small volume manufacturers in Section III.B.6. below).

3. Overview of How EPA's CO₂ Standards Will Be Implemented for Individual Manufacturers

This section provides a brief overview of how EPA will implement the CO₂ standards. Section III.E explains EPA's approach to certification and compliance in detail. As proposed, EPA is finalizing two kinds of standards—fleet average standards determined by a manufacturer's fleet makeup, and in-use standards that will apply to the individual vehicles that make up the manufacturer's fleet. Although this is similar in concept to the current light-duty vehicle Tier 2 program, there are important differences. In explaining EPA's CO₂ standards, it is useful to summarize how the Tier 2 program works.

Under Tier 2, manufacturers select a test vehicle prior to certification and test the vehicle and/or its emissions hardware to determine both its emissions performance when new and the emissions performance expected at the end of its useful life. Based on this testing, the vehicle is assigned to one of several specified bins of emissions levels, identified in the Tier 2 rule, and this bin level becomes the emissions standard for the test group the test vehicle represents. All of the vehicles in the group must meet the emissions level for that bin throughout their useful life. The emissions level assigned to the bin is also used in calculating the manufacturer's fleet average emissions performance.

Since compliance with the Tier 2 fleet average depends on actual test group sales volumes and bin levels, it is not possible to determine compliance at the time the manufacturer applies for and receives a certificate of conformity for a test group. Instead, at certification, the manufacturer demonstrates that the vehicles in the test group are expected to comply throughout their useful life with the emissions bin assigned to that test group, and makes a good faith demonstration that its fleet is expected to comply with the Tier 2 average when the model year is over. EPA issues a certificate for the vehicles covered by the test group based on this demonstration, and includes a condition in the certificate that if the manufacturer does not comply with the fleet average then production vehicles from that test group will be treated as not covered by the certificate to the extent needed to bring the manufacturer's fleet average into compliance with Tier 2.

EPA is retaining the Tier 2 approach of requiring manufacturers to demonstrate in good faith at the time of certification that vehicles in a test group will meet applicable standards throughout useful life. EPA is also retaining the practice of conditioning certificates upon attainment of the fleet average standard. However, there are several important differences between a Tier 2 type of program and the CO₂ standards program. These differences and resulting modifications to EPA's certification protocols are summarized below and are described in detail in Section II I.E.

EPA will continue to certify test groups as it does for Tier 2, and the CO₂ emission results for the test vehicle will serve as the initial or default standard for all of the vehicles in the test group. However, manufacturers will later collect and submit data for individual vehicle model types within each test group, based on the extensive fuel economy testing that occurs through the course of the model year. This model type data will be used to assign a distinct certification level for each model type, thus replacing the initial test group data as the compliance value for each model. It is these model type values that will be used to calculate the fleet average after the end of the model year. The option to substitute model type data for the test group data is at the manufacturer's discretion, except they are required, as they are under the CAFE test protocols, to submit sufficient vehicle test data to represent no less than 90 percent of their actual model year production. The test group emissions data will continue to apply for any model type that is not covered by vehicle test data specific to that model type.

EPA's CO₂ standards also differ from Tier 2 in that the fleet average calculation for Tier 2 is based on test group bin levels and test group sales whereas under the CO₂ program the CO₂ fleet average could be based on a combination of test group and model type emissions and model type production. For the new CO₂ standards, the final regulations use production rather than sales in calculating the fleet average in order to closely conform with the CAFE program, which is a production-based program. Production as defined in the regulations is relatively easy for manufacturers to track, but once the vehicle is delivered to dealerships the manufacturer becomes once step removed from the sale to the ultimate customer, and it becomes more difficult to track that final transaction. There is no environmental impact of using production instead of actual sales, and many commenters supported maintaining alignment between EPA's program and the CAFE program where possible.

4. Averaging, Banking, and Trading Provisions for CO₂ Standards

As explained above, EPA is finalizing a fleet average CO₂ program for passenger cars and light trucks. EPA has previously implemented similar averaging programs for a range of motor vehicle types and pollutants, from the Tier 2 fleet average for NO_X to motorcycle hydrocarbon (HC) plus oxides of nitrogen (NO_X) emissions to NO_X and particulate matter (PM) emissions from heavy-duty engines. The program will operate much like EPA's existing averaging programs in that manufacturers will calculate production-weighted fleet average emissions at the end of the model year and compare their fleet average with a fleet average emission standard to determine compliance. As in other EPA averaging programs, the Agency is also finalizing a comprehensive program for averaging, banking, and trading of credits which together will help manufacturers in planning and implementing the orderly phase-in of emissions control technology in their production, consistent with their typical redesign schedules.

Averaging, Banking, and Trading (ABT) of emissions credits has been an important part of many mobile source programs under CAA Title II, both for fuels programs as well as for engine and vehicle programs. ABT is important because it can help to address many issues of technological feasibility and lead-time, as well as considerations of cost. ABT is an integral part of the standard setting itself, and is not just an add-on to help reduce costs. In many cases, ABT resolves issues of lead-time or technical feasibility, allowing EPA to set a standard that is either numerically more stringent or goes into effect earlier than could have been justified otherwise. This provides important environmental benefits and at the same time it increases flexibility and reduces costs for the regulated industry. A wide range of commenters expressed general support for the ABT provisions. Some commenters noted issues regarding specific provisions of the ABT program, which will be discussed in the appropriate context below. Several commenters requested that EPA publicly release manufacturer-specific ABT data to improve the transparency of credit transactions. These comments are addressed in Section II I.E.

This section discusses generation of credits by achieving a fleet average CO₂ level that is lower than the manufacturer's CO₂ fleet average standard. The final rule includes a variety of additional ways credits may be generated by manufacturers. Section III.C describes these additional opportunities to generate credits in detail. Manufacturers may earn credits through A/C system improvements beyond a specified baseline. Credits can also be generated by producing alternative fuel vehicles, by producing advanced technology vehicles including electric vehicles, plug-in hybrids, and fuel cell vehicles, and by using technologies that improve off-cycle emissions. In addition, early credits can be generated prior to the program's MY 2012 start date. The credits will be used to determine a manufacturer's compliance at the end of the model year. These credit generating opportunities are described below in Section III.C.

As explained earlier, manufacturers will determine the fleet average standard that applies to their car fleet and the standard for their truck fleet from the applicable attribute-based curve. A manufacturer's credit or debit balance will be determined by comparing their fleet average with the manufacturer's CO₂ standard for that model year. The standard will be calculated from footprint values on the attribute curve and actual production levels of vehicles at each footprint. A manufacturer will generate credits if its car or truck fleet achieves a fleet average CO₂ level lower than its standard and will generate debits if its fleet average CO₂ level is above that standard. At the end of the model year, each manufacturer will calculate a production-weighted fleet average for each averaging set (cars and trucks). A manufacturer's car or truck fleet that achieves a fleet average CO₂ level lower than its standard will generate credits, and if its fleet average CO₂ level is above that standard its fleet will generate debits.

The regulations will account for the difference in expected lifetime vehicle miles traveled (VMT) between cars and trucks in order to preserve CO₂ reductions when credits are transferred between cars and trucks. As directed by EISA, NHTSA accomplishes this in the CAFE program by using an adjustment factor that is applied to credits when they are transferred between car and truck compliance categories. The CAFE adjustment factor accounts for two different influences that can cause the transfer of car and truck credits (expressed in tenths of a mpg), if left unadjusted, to potentially negate fuel reductions. First, mpg is not linear with fuel consumption, i.e., a 1 mpg improvement above a standard will imply a different amount of actual fuel consumed depending on the level of the standard. Second, NHTSA's conversion corrects for the fact that the typical lifetime miles for cars is less than that for trucks, meaning that credits earned for cars and trucks are not necessarily equal. NHTSA's adjustment factor essentially converts credits into vehicle lifetime gallons to ensure preservation of fuel savings and the transfer credits on an equal basis, and then converts back to the statutorily-required credit units of tenths of a mile per gallon. To convert to gallons NHTSA's conversion must take into account the expected lifetime mileage for cars and trucks. Because EPA's standards are expressed on a CO₂ gram per mile basis, which is linear with fuel consumption, EPA's credit calculations do not need to account for the first issue noted above. However, EPA is accounting for the second issue by expressing credits when they are generated in total lifetime Megagrams (metric tons), rather than through the use of conversion factors that would apply at certain times. In this way credits may be freely exchanged between car and truck compliance categories without the need for adjustment. Additional detail regarding this approach, including a discussion of the vehicle lifetime mileage estimates for cars and trucks can be found in Section II I.E. 5. A discussion of the derivation of the estimated vehicle lifetime miles traveled can be found in Chapter 4 of the Joint Technical Support Document.

A manufacturer that generates credits in a given year and vehicle category may use those credits in essentially four ways, although with some limitations. These provisions are very similar to those of other EPA averaging, banking, and trading programs. These provisions have the potential to reduce costs and compliance burden, and support the feasibility of the standards in terms of lead time and orderly redesign by a manufacturer, thus promoting and not reducing the environmental benefits of the program.

First, EPA proposed that the manufacturer must use any credits earned to offset any deficit that had accrued in the current year or in a prior model year that had been carried over to the current model year. NRDC commented that such a provision is necessary to prevent credit “shell games” from delaying the adoption of new technologies. EPA's Tier 2 program includes such a restriction, and EPA is applying an identical restriction to the GHG program. Simply stated, a manufacturer may not bank (or carry forward) credits if that manufacturer is also carrying a deficit. In such a case, the manufacturer is obligated to use any current model year credits to offset that deficit. Using current model year credits to offset a prior model year deficit is referred to in the CAFE program as credit carry-back. EPA's deficit carry-forward, or credit carry-back provisions are described further, below.

Second, after satisfying any needs to offset pre-existing deficits, remaining credits may be banked, or saved for use in future years. Credits generated in this program will be available to the manufacturer for use in any of the five model years after the model year in which they were generated, consistent with the CAFE program under EISA. This is also referred to as a credit carry-forward provision.

EPA received a number of comments regarding the credit carry-back and carry-forward provisions. Many supported the proposed consistency of these provisions with EISA and the flexibility provided by these provisions, and several offered qualified or tentative support. For example, NRDC encouraged EPA to consider further restrictions in the 2017 and later model years. Public Citizen expressed concern regarding the complexity of the program and how these provisions might obscure a straightforward determination of compliance in any given model year. At least two automobile manufacturers suggested modeling the program after California, which allows credits to be carried forward for three additional years following a discounting schedule.

For other new emission control programs, EPA has sometimes initially restricted credit life to allow time for the Agency to assess whether the credit program is functioning as intended. When EPA first offered averaging and banking provisions in its light-duty emissions control program (the National Low Emission Vehicle Program), credit life was restricted to three years. The same is true of EPA's early averaging and banking program for heavy-duty engines. As these programs matured and were subsequently revised, EPA became confident that the programs were functioning as intended and that the standards were sufficiently stringent to remove the restrictions on credit life. EPA is therefore acting consistently with our past practice in finalizing reasonable restrictions on credit life in this new program. The Agency believes that a credit life of five years represents an appropriate balance between promoting orderly redesign and upgrade of the emissions control technology in the manufacturer's fleet and the policy goal of preventing large numbers of credits accumulated early in the program from interfering with the incentive to develop and transition to other more advanced emissions control technologies. As discussed below in Section III.C, early credits generated by a manufacturer are also be subject to the five year credit carry-forward restriction based on the year in which they are generated. This limits the effect of the early credits on the long-term emissions reductions anticipated to result from the new standards.

Third, the new program enables manufacturers to transfer credits between the two averaging sets, passenger cars and trucks, within a manufacturer. For example, credits accrued by over-compliance with a manufacturer's car fleet average standard may be used to offset debits accrued due to that manufacturer's not meeting the truck fleet average standard in a given year. EPA believes that such cross-category use of credits by a manufacturer provides important additional flexibility in the transition to emissions control technology without affecting overall emission reductions. Comments regarding the credit transfer provisions expressed general support, noting that it does not matter to the environment whether a gram of greenhouse gas is generated from a car or a truck. Additional comments regarding EPA's streamlined megagram approach and method of accounting for expected vehicle lifetime miles traveled are summarized in Section II I.E.

Finally, accumulated credits may be traded to another vehicle manufacturer. As with intra-company credit use, such inter-company credit trading provides flexibility in the transition to emissions control technology without affecting overall emission reductions. Trading credits to another vehicle manufacturer could be a straightforward process between the two manufacturers, but could also involve third parties that could serve as credit brokers. Brokers may not own the credits at any time. These sorts of exchanges are typically allowed under EPA's current emission credit programs, e.g., the Tier 2 light-duty vehicle NO_X fleet average standard and the heavy-duty engine NO_X fleet average standards, although manufacturers have seldom made such exchanges. Comments generally reflected support for the credit trading flexibility, although some questioned the extent to which trading might actually occur. As noted above, comments regarding program transparency are addressed in Section II I.E.

If a manufacturer has accrued a deficit at the end of a model year—that is, its fleet average level failed to meet the required fleet average standard—the manufacturer may carry that deficit forward (also referred to credit carry-back) for a total of three model years after the model year in which that deficit was generated. EPA continues to believe that three years is an appropriate amount of time that gives the manufacturers adequate time to respond to a deficit situation but does not create a lengthy period of prolonged non-compliance with the fleet average standards. As noted above, such a deficit carry-forward may only occur after the manufacturer has applied any banked credits or credits from another averaging set. If a deficit still remains after the manufacturer has applied all available credits, and the manufacturer did not obtain credits elsewhere, the deficit may be carried forward for up to three model years. No deficit may be carried into the fourth model year after the model year in which the deficit occurred. Any deficit from the first model year that remains after the third model year will constitute a violation of the condition on the certificate, which will constitute a violation of the Clean Air Act and will be subject to enforcement action.

The averaging, banking, and trading provisions are generally consistent with those included in the CAFE program, with a few notable exceptions. As with EPA's approach, CAFE allows five year carry-forward of credits and three year carry-back. Under CAFE, transfers of credits across a manufacturer's car and truck averaging sets are also allowed, but with limits established by EISA on the use of transferred credits. The amount of transferred credits that can be used in a year is limited, and transferred credits may not be used to meet the CAFE minimum domestic passenger car standard. CAFE allows credit trading, but again, traded credits cannot be used to meet the minimum domestic passenger car standard. EPA did not propose, and is not finalizing, these constraints on the use of transferred credits.

Additional details regarding the averaging, banking, and trading provisions and how EPA will implement these provisions can be found in Section II I.E.

5. CO₂ Temporary Lead-Time Allowance Alternative Standards

EPA proposed adopting a limited and narrowly prescribed option, called the Temporary Lead-time Allowance Alternative Standards (TLAAS), to provide additional lead time for a certain subset of manufacturers. As noted in the proposal, this option was designed to address two different situations where we project that more lead time is needed, based on the level of emissions control technology and emissions control performance currently exhibited by certain vehicles. One situation involves manufacturers who have traditionally paid CAFE fines instead of complying with the CAFE fleet average, and as a result at least part of their vehicle production currently has significantly higher CO₂ and lower fuel economy levels than the industry average. More lead time is needed in the program's initial years to upgrade these vehicles to meet the aggressive CO₂ emissions performance levels required by the final rule. The other situation involves manufacturers who have a limited line of vehicles and are therefore unable to average emissions performance across a full line of production. For example, some smaller volume manufacturers produce only vehicles with emissions above the corresponding CO₂ footprint target, and do not have other types of vehicles (that exceed their compliance targets) in their production mix with which to average. Often, these manufacturers also pay fines under the CAFE program rather than meeting the applicable CAFE standard. Because voluntary non-compliance through payment of civil penalties is impermissible for the GHG standards under the CAA, both of these types of manufacturers need additional lead time to upgrade vehicles and meet the standards. EPA proposed that this subset of manufacturers be allowed to produce up to 100,000 vehicles over model years 2012-2015 that would be subject to a somewhat less stringent CO₂ standard of 1.25 times the standard that would otherwise apply to those vehicles. Only manufacturers with total U.S. sales of less than 400,000 vehicles per year in MY 2009 would be eligible for this allowance. Those manufacturers would have to exhaust designated program flexibilities in order to be eligible, and credit generating and trading opportunities for the eligible vehicles would be restricted. See 74 FR 49522-224.

EPA is finalizing the optional TLAAS provisions, with certain limited modifications, so that these manufacturers can have sufficient lead time to meet the tougher MY 2016 GHG standards, while preserving consumer choice of vehicles during this time. EPA is finalizing modified provisions to address the unique lead-time issues of smaller volume manufacturers. One provision involves additional flexibility under the TLAAS program for manufacturers below 50,000 U.S. vehicle sales, as discussed further in Section III.B.5.b below. Another provision defers the CO₂ standards for the smallest volume manufacturers, those below 5,000 U.S. vehicle sales, as discussed in Section III.B.6.

Comments from several manufacturers strongly supported the TLAAS program as critical to provide the lead time needed for manufacturers to meet the standards. Volkswagen commented that TLAAS is an important aspect of EPA's proposal and that it responds to the needs of some smaller manufacturers for additional lead time and flexibility under the CAA. Daimler Automotive Group commented that TLAAS is a critical element of the program and falls squarely within EPA's discretion to provide appropriate lead time to limited-line low-volume manufacturers. BMW also commented that TLAAS is needed because most of the companies with limited lines will have to meet a more stringent fleet standard by 2016 than full-line manufacturers because they sell “feature-dense” vehicles (as opposed to light-weight large wheel-base vehicles) and no pick-up trucks. BMW commented that their MY 2016 footprint-based standard is projected to be 4 percent more stringent than the fleet average standard of 250 g/mile. The Alliance of Automobile Manufacturers supported the flexibilities proposed by EPA, including TLAAS. As discussed in detail below, EPA received extensive comments from many smaller volume manufacturers that the proposed TLAAS program was insufficient to address lead time and feasibility issues they will face under the program.

In contrast, EPA also received comments from the Center for Biological Diversity opposing the TLAAS program, commenting that an exception for high performance vehicles is not allowed under EPCA or the CAA and that it rewards manufacturers that pay penalties under CAFE and penalizes those that have complied with CAFE. This commenter suggests that manufacturers could decrease vehicle mass or power output of engines, purchase credits from another manufacturer, or earn off-cycle credits. EPA responds to these comments below.

After carefully considering the public comments, EPA continues to believe that the TLAAS program is essential in providing necessary lead time and flexibility to eligible manufacturers in the early years of the standards. First, EPA believes that it is acting well within its legal authority in adopting the various TLAAS provisions. EPA is required to provide sufficient lead time for industry as a whole for standards under section 202(a)(1), which mandates that standards are to take effect only “after providing 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.” Thus, although section 202(a)(1) does not explicitly authorize this or any other specific lead time provision, it affords ample leeway for EPA to craft provisions designed to provide adequate lead time, and to tailor those provisions as appropriate. We show below that the types of technology penetrations required for TLAAS-eligible vehicles in the program's earlier years raise critical issues as to adequacy of lead time. As discussed in the EPA feasibility analysis provided in Section III.D.6 and III.D.7 several manufacturers eligible for TLAAS are projected to face a compliance shortfall in MY 2016 without the TLAAS program, even with the full application of technologies assumed by the OMEGA Model, including hybrid use of up to 15 percent. These include BMW, Jaguar Land Rover, Daimler, Porsche, and Volkswagen In addition, the smaller volume manufacturers of this group (i.e., Jaguar Land Rover and Porsche) face the greatest shortfall (see Table III.D.6-4). Even with TLAAS, these manufacturers will need to take technology steps to comply with standards above and beyond those of other manufacturers. These manufacturers have relatively few models with high baseline emissions and this flexibility allows them additional lead time to adapt to a longer term strategy of meeting the final standards within their vehicle redesign cycles.

Second, EPA has carefully evaluated other means of eligible manufacturers to meet the standards, such as utilizing available credit opportunities. Indeed, eligibility for the TLAAS, and for temporary deferral of regulation for very small volume manufacturers, is conditioned on first exhausting the various programmatic flexibilities including credit utilization. At the same time, a basic reason certain manufacturers are faced with special lead time difficulties is their inability to generate credits which can be then be averaged across their fleet because of limited product lines. And although purchasing credits is an option under the program, there are no guarantees that credits will be available. Historic practice in fact suggests that manufacturers do not sell credits to competitors. While some of the smaller manufacturers covered by the TLAAS program may be in a position to obtain credits, they are not likely to be available for the TLAAS manufacturers across the board in the volume needed to comply without the TLAAS provisions. At the same time the TLAAS provisions have been structured such that any credits that do become available would likely be used before a manufacturer would turn to the more restricted and limiting TLAAS provisions.

As discussed in Section III.C., off-cycle credits are available if manufacturers are able to employ new and innovative technologies not already in widespread use, which provide real-world emissions reductions not captured on the current test cycles. Further, these credits are eligible only for technologies that are newly introduced on just a few vehicle models, and are not yet in widespread use across the fleet. The magnitude of these credits are highly uncertain because they are based on new technologies, and EPA is not aware of any such technologies that would provide enough credits to bring these manufacturers into compliance without TLAAS lead time flexibility. Manufacturers first must develop these technologies and then demonstrate their emissions reductions capabilities, which will require lead time. Moreover, the technologies mentioned in the proposal which are the most likely to be eligible based on present knowledge, including solar panels and active aerodynamics, are likely to provide only small incremental emissions reductions.

We agree with the comment that reducing vehicle mass or power are potential methods for reducing emissions that should be employed by TLAAS-eligible manufacturers to help them meet standards. However, based on our assessment of the lead time needed for these manufacturers to comply with the standards, especially given their more limited product offerings and higher baseline emissions, we believe that additional time is needed for them to come into compliance. EPA can permissibly consider the TLAAS and other manufacturers' lead time, cost, and feasibility issues in developing the primary standards and has discretion in setting the overall stringency of the standards to account for these factors. Natural Resources Defense Council v. Thomas, 805 F. 2d 410, 421 (DC Cir. 1986) (even when implementing technology-forcing provisions of Title II, EPA may base standards on an industry-wide capability “taking into account the broad spectrum of technological capabilities as well as cost and other factors” across the industry). EPA is not legally required to set standards that drive these manufacturers or their products out of the market, nor is EPA legally required to preserve a certain product line or vehicle characteristic. Instead EPA has broad discretion under section 202(a)(1) to set standards that reasonably balance lead time needs across the industry as a whole and vehicle availability. In this rulemaking, EPA has consistently emphasized the importance of obtaining very significant reductions in emissions of GHGs from the industry as a whole, and obtaining those reductions through regulatory approaches that avoid limiting the ability of manufacturers to provide model availability and choice for consumers. The primary mechanism to achieve this is the use of a footprint attribute curve in setting the increasingly stringent model year standards. The TLAAS provisions are a temporary and strictly limited modification to these attribute standards allowing the TLAAS manufacturers lead time to upgrade their product lines to meet the 2016 GHG standards. EPA has made a reasonable choice here to preserve the overall stringency of the program, and to afford increased flexibility in the program's early years to a limited class of vehicles to assure adequate lead time for all manufacturers to meet the strictest of the standards by MY 2016.

As described below, EPA also carefully considered the comments of smaller volume manufacturers and believes additional lead time is needed. Therefore, EPA is finalizing the TLAAS program, similar to that proposed, and is also finalizing an additional TLAAS option for manufacturers with annual U.S. sales under 50,000 vehicles. EPA is also deferring standards for manufacturers with annual sales of less than 5,000 vehicles. These new TLAAS provisions and the small volume manufacturer deferment are discussed in detail below and in Section III.B.6.

a. Base TLAAS Program

As proposed, EPA is establishing the TLAAS program for a specified subset of manufacturers. This alternative standard is an option only for manufacturers with total U.S. sales of less than 400,000 vehicles per year, using 2009 model year final sales numbers to determine eligibility for these alternative standards. For manufacturers with annual U.S. sales of 50,000 or more but less than 400,000 vehicles, EPA is finalizing the TLAAS program largely as proposed. EPA proposed that under the TLAAS, qualifying manufacturers would be allowed to produce up to 100,000 vehicles that would be subject to a somewhat less stringent CO₂ standard of 1.25 times the standard that would otherwise apply to those vehicles. This 100,000 volume is not an annual limit, but is an absolute limit for the total number of vehicles which can use the TLAAS program over the model years 2012-2015. Any additional production would be subject to the same standards as any other manufacturer. EPA is retaining this limit for manufacturers with baseline MY 2009 sales of 50,000 but less than 400,000. In addition, as discussed further below, EPA is finalizing a variety of restrictions on the use of the TLAAS program, to ensure that only manufacturers who need more lead time for the kinds of reasons noted above are likely to use the program.

Volvo and Saab commented that basing eligibility strictly on MY 2009 sales would be problematic for these companies, which are being spun-off from larger manufacturer in the MY 2009 time frame due to the upheaval in the auto industry over the past few years. These commenters offered a variety of suggestions including using MY 2010 as the eligibility cut-off instead of MY 2009, reassessing eligibility on a year-by-year basis as corporate relationships change, or allowing companies separated from a larger parent company by the end of 2010 to use their MY 2009 branded U.S. sales to qualify for TLAAS. In response to these concerns, EPA recognizes that these companies currently being sold by larger manufacturers will share the same characteristics of the manufacturers for which the TLAAS program was designed. As newly independent companies, these firms will face the challenges of a narrower fleet of vehicles across which to average, and may potentially be in a situation, at least in the first few years, of paying fines under CAFE. Lead time concerns in the program's initial years are in fact particularly acute for these manufacturers since they will be newly independent, and thus would have even less of an opportunity to modify their vehicles to meet the standards. Therefore, EPA is finalizing an approach that allows manufacturers with U.S. “branded sales” in MY 2009 under the umbrella of a larger manufacturer that become independent by the end of calendar year 2010 to use their MY 2009 branded sales to qualify for TLAAS eligibility. In other words, a manufacturer will be eligible for TLAAS if it produced vehicles for the U.S. market in MY 2009, its branded sales of U.S. vehicles were less than 400,000 in MY 2009 but whose vehicles were sold as part of a larger manufacturer, and it becomes independent by the end of calendar year 2010, if the new entity has sales below 400,000 vehicles.

Manufacturers with no U.S. sales in MY 2009 are not eligible to utilize the TLAAS program. EPA does not support the commenter's suggestion of a year-by-year eligibility determination because it opens up the TLAAS program to an unknown universe of potential eligible manufacturers, with the potential for gaming. EPA does not believe the TLAAS program should be available to new entrants to the U.S. market since these manufacturers are not transitioning from the CAFE regime which allows fine paying as a means of compliance to a CAA regime which does not, and hence do not present the same types of lead time issues. Manufacturers entering the U.S. market for the first time thus will be fully subject to the GHG fleet-average standards.

As proposed, manufacturers qualifying for TLAAS will be allowed to meet slightly less stringent standards for a limited number of vehicles. An eligible manufacturer could have a total of up to 100,000 units of cars or trucks combined over model years 2012-2015 which would be subject to a standard 1.25 times the standard that would otherwise apply to those vehicles under the primary program. In other words, the footprint curves upon which the individual manufacturer standards for the TLAAS fleets are based would be less stringent by a factor of 1.25 for up to 100,000 of an eligible manufacturer's vehicles for model years 2012-2015. EPA believes that 100,000 units over four model years achieves an appropriate balance, as the emissions impact is quite small, but does provide companies with necessary lead time during MY 2012-2015. For example, for a manufacturer producing 400,000 vehicles per year, this would be a total of up to 100,000 vehicles out of a total production of up to 1.6 million vehicles over the four year period, or about 6 percent of total production.

Finally, for manufacturers of 50,000 but less than 400,000 U.S. vehicles sales during 2009, the program expires at the end of MY 2015 as proposed. EPA continues to believe the program reasonably addresses a real world lead time constraint for these manufacturers, and does so in a way that balances the need for more lead time with the need to minimize any resulting loss in potential emissions reductions. In MY 2016, the TLAAS option thus ends for all but the smallest manufacturers opting for TLAAS, and manufacturers must comply with the same CO₂ standards as non-TLAAS manufacturers; under the CAFE program companies would continue to be allowed to pay civil penalties in lieu of complying with the CAFE standards. However, because companies must meet both the CAFE standards and the EPA CO₂ standards, the National Program will have the practical impact of providing a level playing field for almost all except the smallest companies beginning in MY 2016. This option, even with the modifications being adopted, thereby results in more fuel savings and CO₂ reductions than would be the case under the CAFE program by itself.

EPA proposed that manufacturers meeting the cut-point of below 400,000 sales for MY 2009 but whose U.S. sales grew above 400,000 in any subsequent model years would remain eligible for the TLAAS program. The total sales number applies at the corporate level, so if a corporation owns several vehicle brands the aggregate sales for the corporation must be used. These provisions would help prevent gaming of the provisions through corporate restructuring. Corporate ownership or control relationships would be based on determinations made under CAFE for model year 2009 (except in the case of a manufacturer being sold by a larger manufacturer by the end of calendar year 2010, as discussed above). In other words, corporations grouped together for purposes of meeting CAFE standards in MY 2009, must be grouped together for determining whether or not they are eligible under the 400,000 vehicle cut point. EPA is finalizing these provisions with the following modifications. EPA recognizes the dynamic corporate restructuring occurring in the auto industry and believes it is important to structure additional provisions to ensure there is no ability to game the TLAAS provisions and to ensure no unintended loss of feasible environmental benefits. Therefore, EPA is finalizing a provision that if two or more TLAAS eligible companies are later merged, with one company having at least 50% or more ownership of the other, or if the companies are combined for the purposes of EPA certification and compliance, the TLAAS allotment is not additive. The merged company will only be allowed the allotment for what is considered the parent company under the new corporate structure. Further, if the newly formed company would have exceeded the 400,000 vehicle cut point based on combined MY 2009 sales, the new entity is not eligible for TLAAS in the model year following the merger. EPA believes that such mergers and acquisitions would give the parent company additional opportunities to average across its fleet, eliminating one of the primary needs for the TLAAS program. This provision will not be retroactive and will not affect the TLAAS program in the year of the merger or for previous model years. EPA believes these additional provisions are essential to ensure the integrity of the TLAAS program by ensuring that it does not become available to large manufacturers through mergers and acquisitions.

As proposed, the TLAAS vehicles will be separate car and truck fleets for that model year and subject to the less stringent footprint-based standards of 1.25 times the primary fleet average that would otherwise apply. The manufacturer will determine what vehicles are assigned to these separate averaging sets for each model year. As proposed, credits from the primary fleet average program can be transferred and used in the TLAAS program. Credits generated within the TLAAS program may also be transferred between the TLAAS car and truck averaging sets (but not to the primary fleet as explained below) for use through MY 2015 when the TLAAS ends.

EPA is finalizing a number of restrictions on credit trading within the TLAAS program, as proposed. EPA is concerned that if credit use in the TLAAS program were unrestricted, some manufacturers would be able to place relatively clean vehicles in the TLAAS fleet, and generate credits for the primary program fleet. First, credits generated under TLAAS may not be transferred or traded to the primary program. Therefore, any unused credits under TLAAS expire after model year 2015 (or 2016 for manufacturers with annual sales less than 50,000 vehicles). EPA believes that this is necessary to limit the program to situations where it is needed and to prevent the allowance from being inappropriately transferred to the long-term primary program where it is not needed. EPA continues to believe this provision is necessary to prevent credits from being earned simply by removing some high-emitting vehicles from the primary fleet. Absent this restriction, manufacturers would be able to choose to use the TLAAS for these vehicles and also be able to earn credits under the primary program that could be banked or traded under the primary program without restriction. Second, EPA is finalizing two additional restrictions on the use of TLAAS by requiring that for any of the 2012-2015 model years for which an eligible manufacturer would like to use the TLAAS, the manufacturer must use two of the available flexibilities in the GHG program first in order to try and comply with the primary standard before accessing the TLAAS—i.e., TLAAS eligibility is not available to those manufacturers with other readily-available means of compliance. Specifically, before using the TLAAS a manufacturer must: (1) Use any banked emission credits from previous model years; and, (2) use any available credits from the companies' car or truck fleet for the specific model year (i.e., use credit transfer from cars to trucks or from trucks to cars). That is, before using the TLAAS for either the car fleet or the truck fleet, the company must make use of any available intra-manufacturer credit transfers first. Finally, EPA is restricting the use of banking and trading between companies of credits in the primary program in years in which the TLAAS is being used. No such restriction is in place for years when the TLAAS is not being used.

EPA received several comments in support of these credit restrictions for the TLAAS program. On the negative side, one manufacturer commented that the restrictions were not necessary, saying that the restrictions are counter to providing manufacturers with flexibility and that the emissions impacts estimated by EPA due to the full use of the program are small. However, EPA continues to believe that the restrictions are appropriate to prevent the potential gaming described above, and to ensure that the TLAAS program is used only by those manufacturers that have exhausted all other readily available compliance mechanisms and consequently have legitimate lead time issues.

One manufacturer commented that the program is restrictive due to the requirement that manufacturers must decide prior to the start of the model year whether or not and how to use the TLAAS program. EPA did not intend for manufactures to have to make this determination prior to the start of the model year. EPA expects that manufacturers will provide a best estimate of their plans to use the TLAAS program during certification based on projected model year sales, as part of their pre model year report projecting their overall plan for compliance (as required by § 600.514-12 of the regulations). Manufacturers must determine the program's actual use at the end of the model year during the process of demonstrating year-end compliance. EPA recognizes that depending on actual sales for a given model year, a manufacturer's use of TLAAS may change from the projections used in the pre-model year report.

b. Additional TLAAS Flexibility for Manufacturers With MY 2009 Sales of Less Than 50,000 Vehicles

EPA received extensive comments that the TLAAS program would not provide sufficient lead time and flexibility for companies with sales of significantly less than 400,000 vehicles. Jaguar Land Rover, which separated from Ford in 2008, commented that it sells products only in the middle and large vehicle segments and that its total product range remains significantly more limited in terms of segments in comparison with its main competitors which typically have approximately 75% of their passenger car fleet in the small and middle segments. Jaguar Land Rover also commented that it has already committed $1.3 billion of investment to reducing CO₂ from its vehicle fleet and that this investment is already delivering a range of technologies to improve the fuel economy and CO₂ performance of its existing vehicles. Jaguar Land Rover submitted confidential business information regarding their future product plans and emissions performance capabilities of their vehicles which documents their assertions.

Porsche commented that their passenger car footprint-based standard is the most stringent of any manufacturer and this, combined with their high baseline emissions level, means that it would need to reduce emissions by about 10 percent per year over the 2012-2016 time-frame. Porsche commented that such reductions were not feasible. They commented that their competitors will be able to continue to offer their full line of products because the competitors have a wider range of products with which to average. Porsche further commented that their product development cycles are longer than larger competitors. Porsche recommended for small limited line niche manufacturers that EPA require an annual 5 percent reduction in emissions from baseline up to a total reduction of 25 percent, or to modify the TLAAS program to require such reductions. Porsche noted that this percent reduction would be in line with the average emissions reductions required for larger manufacturers.

EPA also received comments from several very small volume manufacturers that, even with the TLAAS program, the proposed standards are not feasible for them, certainly not in the MY 2012-2016 MY time frame. These manufacturers included Aston Martin, McLaren, Lotus, and Ferrari. Their comments consistently focused on the need for separate, less stringent standards for small volume manufacturers. The manufacturers commented that they are willing to make progress in reducing emissions, but that separate, less-stringent small volume manufacturer standards are needed for them to remain in the U.S. market. The commenters note that their product line consists entirely of high end sports cars. Most of these manufacturers have only a few vehicle models, have annual sales on the order of a few hundred to a few thousand vehicles, and several have average baseline CO₂ emissions in excess of 500 g/mile—nearly twice the industry average. McLaren commented that its vehicle model to be introduced in MY 2011 will have class leading CO₂ performance but that it would not be able to offer the vehicle in the U.S. market because it does not have other vehicle models with which to average. Similarly, Aston Martin commented that it is of utmost importance that it is not required to reduce emissions significantly more than equivalent vehicles from larger manufacturers, which would render them uncompetitive due purely to the size of its business. Manufacturers also noted that they launch new products less frequently than larger manufacturers (e.g., Ferrari noted that their production period for models is 7-8 years), and that suppliers serve large manufacturers first because they can buy in larger volumes. Some manufacturers also noted that they would be willing to purchase credits at a reasonable price, but they believed that credit availability from other manufacturers was highly unlikely due to the competitive nature of the auto industry. Several of these manufacturers provided confidential business information indicating their preliminary plans for reducing GHG emissions across their product lines through MY 2016 and beyond.

The Association of International Automobile Manufacturers (AIAM) also commented that, because of their essential features, vehicles produced by small volume manufacturers would not be able to meet the proposed greenhouse gas standards. AIAM commented that “while it is possible that these small volume manufacturers (SVMs) might be able to comply with greenhouse gas standards by purchasing credits from other manufacturers, this is far too speculative a solution. The market for credits is unpredictable at this point. Other than exiting the U.S. market, therefore, the only other possible solution for an independent SVM would be to sell an equity interest in the company to a larger, full-line manufacturer, so that the emissions of the luxury vehicles could be averaged in with the much larger volume of other vehicles produced by the major manufacturer. This cannot possibly be the outcome EPA intends, especially when measured against the minimal, if any, environmental benefit that would result.” AIAM commented further that “there is ample legal authority for EPA to provide SVMs a more generous lead-time allowance or an alternative standard. Indeed, EPA recognizes such authority in the proposal for a small entity exemption (for those companies defined under the Small Business Administration's regulations), see 74 FR at 49574, and in the TLAAS. These provisions are consistent with previous EPA rulemaking under the Clean Air Act which offer relief to SVMs.” AIAM recommended deferring standards for SVMs to a future rulemaking, providing EPA with adequate time to assess relevant product plans and technology feasibility information from SVMs, conduct the necessary reviews and modeling that may be needed, and consult with the stakeholders.

These commenters noted that standards for the smallest manufacturers were deferred in the California program until MY 2016 and that California's program would have established standards for small volume manufacturers in MY 2016 at a level that would be technologically feasible. The commenters also suggested that California's approach is similar to the approach being taken by EPA for small business entities. Further, these commenters noted that in Tier 2 and other light-duty vehicle programs, EPA has allowed small volume manufacturers (SVMs) until the end of the phase-in period to comply with standards. The commenters recommended that EPA should defer standards for SVMs, and conduct a future rulemaking to establish appropriate standards for SVMs starting in model year 2016. Alternatively, some manufacturers recommended establishing much less stringent standards for SVMs as part of the current rulemaking.

In summary, the manufacturers commented that their range of products was insufficient to allow them to meet the standards in the time provided, even with the proposed TLAAS program. Many of these manufacturers have baseline emissions significantly higher than their larger-volume competitors, and thus the CO₂ reductions required from baseline under the program are larger for many of these companies than for other companies. Although they are investing substantial resources to reduce CO₂ emissions, they believe that they will not be able to achieve the standards under the proposed approach.

EPA also received comments urging us not to expand the TLAAS program. The commenters are concerned about the loss of benefits that would occur with any expansion.

EPA has considered the comments carefully and concludes that additional flexibility is needed for these companies. After assessing the issues raised by commenters, EPA believes there are two groups of manufacturers that need additional lead time. The first group includes manufacturers with annual U.S. sales of less than 5,000 vehicles per year. Standards for these small volume manufacturers are being deferred until a future rulemaking in the 2012 timeframe, as discussed in Section III.B.6, below. This will allow EPA to determine the appropriate level of standards for these manufacturers, as well as the small business entities, at a later time. The second group includes manufacturers with MY 2009 U.S. sales of less than 50,000 vehicles but above the 5,000 vehicle threshold being established for small volume manufacturers. EPA has selected a cut point of 50,000 vehicles in order to limit the additional flexibility to only the smaller manufacturers with much more limited product lines over which to average. EPA has tailored these provisions as narrowly as possible to provide additional lead time only as needed by these smaller manufacturers. We estimate that the TLAAS program, including the changes below will result in a total decrease in overall emissions reductions of about one percent of the total projected GHG program emission benefits. These estimates are provided in RIA Chapter 5 Appendix A.

For some of the companies, the reduction from baseline CO₂ emissions required to meet the standards is clearly greater than for other TLAAS-eligible manufacturers. Compared with other TLAAS-eligible manufacturers, these companies also have more limited fleets across which to average the standards. Some companies have only a few vehicle models all of a similar utility, and thus their averaging abilities are extremely limited posing lead time issues of greater severity than other TLAAS-eligible manufacturers. EPA's feasibility analysis provided in Section III.D., shows that these companies face a compliance shortfall significantly greater than other TLAAS companies (see Table III.D.6-4). This shortfall is primarily due to their narrow product lines and more limited ability to average across their vehicle fleets. In addition, with fewer models with which to average, there is a higher likelihood that phase-in requirements may conflict with normal product redesign cycles.

Therefore, for manufacturers with MY 2009 U.S. sales of less than 50,000 vehicles, EPA is finalizing additional TLAAS compliance flexibility through model year 2016. These manufacturers will be allowed to place up to 200,000 vehicles in the TLAAS program in MY 2012-2015 and an additional 50,000 vehicles in MY 2016. To be eligible for the additional allotment above the base TLAAS level of 100,000 vehicles, manufacturers must annually demonstrate that they have diligently made a good faith effort to purchase credits from other manufacturers in order to comply with the base TLAAS program, but that sufficient credits were not available. Manufacturers must secure credits to the extent they are reasonably available from other manufacturers to offset the difference between their emissions reductions obligations under the base TLAAS program and the expanded TLAAS program. Manufacturers must document their efforts to purchase credits as part of their end of year compliance report. All other aspects of the TLAAS program including the 1.25x adjustment to the standards and the credits provision restrictions remain the same as described above for the same reasons. This will still require the manufacturers to reduce emissions significantly in the 2012-2016 time-frame and to meet the final emissions standards in MY 2017. The standards remain very challenging for these manufacturers but these additional provisions will allow them the necessary lead time for implementing their strategy for compliance with the final, most stringent standards.

The eligibility limit of 50,000 vehicles will be treated in a similar way as the 400,000 vehicle eligibility limit is treated, as described above. Manufacturers with model year 2009 U.S. sales of less than 50,000 vehicles are eligible for the expanded TLAAS flexibility. Manufacturers whose sales grow in later years above 50,000 vehicles without merger or acquisition will continue to be eligible for the expanded TLAAS program. However, manufacturers that exceed the 50,000 vehicle limit through mergers or acquisitions will not be eligible for the expanded TLAAS program in the model year following the merger or acquisition, but may continue to be eligible for the base TLAAS program if the MY 2009 sales of the new company would have been below the 400,000 vehicle eligibility cut point. The use of TLAAS by all the entities within the company in years prior to the merger must be counted against the 100,000 vehicle limit of the base program. If the 100,000 vehicle limit has been exceeded, the company is no longer eligible for TLAAS.

6. Deferment of CO₂ Standards for Small Volume Manufacturers With Annual Sales Less Than 5,000 Vehicles

In the proposal, in the context of the TLAAS program, EPA recognized that there would be a wide range of companies within the eligible manufacturers with sales less than 400,000 vehicles in model year 2009. As noted in the proposal, some of these companies, while having relatively small U.S. sales volumes, are large global automotive firms, including companies such as Mercedes and Volkswagen. Other companies are significantly smaller niche firms, with sales volumes closer to 10,000 vehicles per year worldwide, such as Aston Martin. EPA anticipated that there is a small number of such smaller volume manufacturers, which may face greater challenges in meeting the standards due to their limited product lines across which to average. EPA requested comment on whether the proposed TLAAS program would provide sufficient lead-time for these smaller firms to incorporate the technology needed to comply with the proposed GHG standards. See 74 FR at 49524.

EPA received comments from several very small volume manufacturers that the TLAAS program would not provide sufficient lead time, as described above. EPA agrees with comments that the standards would be extremely challenging and potentially infeasible for these small volume manufacturers, absent credits from other manufacturers, and that credit availability at this point is highly uncertain—although these companies are planning to introduce significant GHG-reducing technologies to their product lines, they are still highly unlikely to meet the standards by MY 2016. Because the products produced by these manufacturers are so unique, these manufacturers were not included in EPA's OMEGA modeling assessment of the technology feasibility and costs to meet the proposed standards. As noted above, these manufacturers have only a few models and have very high baseline emissions. TLAAS manufacturers are projected to be required to reduce emissions by up to 39%, whereas SVMs in many cases would need to cut their emissions by more than half to comply with MY 2016 standards.

Given the unique feasibility issues raised for these manufacturers, EPA is deferring establishing CO₂ standards for manufacturers with U.S. sales of less than 5,000 vehicles. This will provide EPA more time to consider the unique challenges faced by these manufacturers. EPA expects to conduct this rulemaking in the 2012 timeframe. The deferment only applies to CO₂ standards and SVMs must meet N₂ O and CH₄ standards. EPA plans to set standards for these manufacturers as part of a future rulemaking in the next 18 months. This future rulemaking will allow EPA to fully examine the technologies and emissions levels of vehicles offered by small manufacturers and to determine the potential emissions control capabilities, costs, and necessary lead time. This timing may also allow a credits market to develop, so that EPA may consider the availability of credits during the rulemaking process. See State of Mass. v. EPA, 549 U.S. at 533 (EPA retains discretion as to timing of any regulations addressing vehicular GHG emissions under section 202(a)(1)). We expect that standards would begin to be implemented in the MY 2016 timeframe. This approach is consistent with that envisioned by California for these manufacturers. EPA estimates that eligible small volume manufacturers currently comprise less than 0.1 percent of the total light-duty vehicle sales in the U.S., and therefore the deferment will have a very small impact on the GHG emissions reductions from the standards.

In addition to the 5,000 vehicle per year cut point, to be eligible for deferment each year, manufacturers must also demonstrate due diligence in attempting to secure credits from other manufacturers. Manufacturers must make a good faith effort to secure credits to the extent they are reasonably available from other manufacturers to offset the difference between their baseline emissions and what their obligations would be under the TLAAS program starting in MY 2012.

Eligibility will be determined somewhat differently compared to the TLAAS program. Manufacturers with either MY 2008 or MY 2009 U.S. sales of less than 5,000 vehicles will be initially eligible. This includes “branded sales” for companies that sold vehicles under a larger manufacturer but has become independent by the end of calendar year 2010. EPA is including MY 2008 as well as MY 2009 because some manufacturers in this market segment have such limited sales that they often drop in and out of the market from year to year.

In determining eligibility, manufacturers must be aggregated according to the provisions of 40 CFR 86.1838-01(b)(3), which requires the sales of different firms to be aggregated in various situations, including where one firm has a 10% or more equity ownership of another firm, or where a third party has a 10% or more equity ownership of two or more firms. EPA received public comment from a manufacturer requesting that EPA should allow a manufacturer to apply to EPA to establish small volume manufacturer status based on the independence of its research, development, testing, design, and manufacturing from another firm that may have an ownership interest in that manufacturer. EPA has reviewed this comment, but is not finalizing such a provision at this time. EPA believes that this issue likely presents some competitive issues, which we would like to be fully considered through the public comment process. Therefore, EPA plans to consider this issue and seek public comments in our proposal for small volume manufacturer CO₂ standards, which we expect to complete within 18 months.

To remain eligible for the deferral from standards, the rolling average of three consecutive model years of sales must remain below 5,000 vehicles. EPA is establishing the 5,000 vehicle threshold to allow for some sales growth by SVMs, as SVMs typically have annual sales of below 2,000 vehicles. However, EPA wants to ensure that standards for as few vehicles as possible are deferred and therefore believes it is appropriate that manufacturers with U.S. sales growing to above 5,000 vehicles per year be required to comply with standards (including TLAAS, as applicable). Manufacturers with unusually strong sales in a given year would still likely remain eligible, based on the three year rolling average. However, if a manufacturer takes steps to expand in the U.S. market on a permanent basis such that they consistently sell more than 5,000 vehicles per year, they must meet the TLAAS standards. EPA believes a manufacturer will be able to consider these provisions, along with other factors, in its planning to significantly expand in the U.S. market.

For manufacturers exceeding the 5,000 vehicle rolling average through mergers or acquisitions of other manufacturers, those manufacturers will lose eligibility in the MY immediately following the last year of the rolling average. For manufacturers exceeding this level through sales growth, but remaining below a 50,000 vehicle threshold, the manufacturer will lose eligibility for the deferred standards in the second model year following the last year of the rolling average. For example, if the rolling average of MYs 2009-2011 exceeded 5,000 vehicles but was below 50,000 vehicles, the manufacturer would not be eligible for the deferred standards in MY 2013. For manufacturers with a 3-year rolling average exceeding 50,000 vehicles, the manufacturer would lose eligibility in the MY immediately following the last model year in the rolling average. For example, if the rolling average of MYs 2009-2011 exceeded 50,000 vehicles, the manufacturer would not be eligible for the deferred standards in MY 2012. Such manufacturers may continue to be eligible for TLAAS, or the expanded TLAAS program, per the provisions described above. EPA believes these provisions are needed to ensure that the SVM deferment remains targeted to true small volume manufacturers and does not become available to larger manufacturers through mergers or acquisitions. EPA is including the 50,000 vehicle criteria to differentiate between manufacturers that may slowly gain more sales and manufacturers that have taken major steps to significantly increase their presence in the U.S. market, such as by introducing new vehicle models. EPA believes manufacturers selling more than 50,000 vehicles should not be able to take advantage of the deferment, as they should be able to meet the applicable TLAAS standards through averaging across their larger product line.

EPA is requiring that potential SVMs submit a declaration to EPA containing a detailed written description of how the manufacturer qualifies as a small volume manufacturer. The declaration must contain eligibility information including MY 2008 and 2009 U.S. sales, the last three completed MYs sales information, detailed information regarding ownership relationships with other manufacturers, and documentation of efforts to purchase credits from other manufacturers. Because such manufacturers are not automatically exempted from other EPA regulations for light-duty vehicles and light-duty trucks, entities are subject to the greenhouse gas control requirements in this program until such a declaration has been submitted and approved by EPA. The declaration must be submitted annually at the time of vehicle emissions certification under the EPA Tier 2 program, beginning in MY 2012.

7. Nitrous Oxide and Methane Standards

In addition to fleet-average CO₂ standards, as proposed, EPA is establishing separate per-vehicle standards for nitrous oxide (N₂ O) and methane (CH₄) emissions. The agency's intention is to set emissions standards that act to cap emissions to ensure that future vehicles do not increase their N₂ O and CH₄ emissions above levels typical of today's vehicles. EPA proposed to cap N₂ O at a level of 0.010 g/mi and to cap CH₄ at a level of 0.03 g/mi. Both of these compounds are more potent contributors to global warming than CO₂; N₂ O has a global warming potential, or GWP, of 298 and CH₄ has a GWP of 25.

EPA received many comments on the proposed N₂ O and CH₄ standards. A range of stakeholders supported the proposed approach of “cap” standards and the proposed emission levels, including most states and environmental organizations that addressed this topic, and the Manufacturers of Emissions Control Association. These commenters stated that EPA needs to address all mobile GHGs under the Clean Air Act, and N₂ O and CH₄ are both more potent contributors to global warming than CO₂. The Center for Biological Diversity commented that in light of the potency of these GHGs, EPA should develop standards which reduce emissions over current levels and that EPA had not analyzed either the technologies or the costs of doing so. EPA discusses these comments and our responses below and in the Response to Comments Document.

Auto manufacturers generally did not support standards for these GHGs, stating that the levels of these GHGs from current vehicles are too small to warrant standards at this time. These commenters also stated that if EPA were to proceed with “cap” standards, the stringency of the proposed levels could restrict the introduction of some new technologies. Commenters specifically raised this concern with the examples of diesel and lean-burn gasoline for N₂ O, or natural gas and ethanol fueled vehicles for CH₄. Only one manufacturer, Volkswagen, submitted actual test data to support these claims; very limited emission data on two concept vehicles—a CNG vehicle and a flexible-fuel vehicle—indicated measured emission levels near or above the proposed standards, but included no indication of whether any technological steps had been taken to reduce emissions below the cap levels. Many commenters support an approach of establishing a CO₂-equivalent standard, where N₂ O and CH₄ could be averaged with CO₂ emissions to result in an overall CO₂-equivalent compliance value, similar to the approach California has used for its GHG standards Under such an approach, the auto industry commenters supported using a default value for N₂ O emissions in lieu of a measured test value. Several auto manufacturers also had concerns that a new requirement to measure N₂ O would require significant equipment and facility upgrades and would create testing challenges with new measurement equipment with which they have little experience.

EPA has considered these comments and is finalizing the cap standards for N₂ O and CH₄ as proposed. EPA agrees with the NGO, State, and other commenters that light-duty vehicle emissions are small but important contributors to the U.S. N₂ O and CH₄ inventories, and that in the absence of a limitation, the potential for significant emission increases exists with the evolution of new vehicle and engine technologies. (Indeed, the industry commenters concede as much in stating that they are contemplating introducing vehicle technologies that could result in emissions exceeding the cap standard levels). EPA also believes that in most cases N₂ O and CH₄ emissions from light-duty vehicles will remain well below the cap standards. Therefore, we are setting cap standards for these GHGs at the proposed levels. However, as described below, the agency is incorporating several provisions intended to address industry concerns about technological feasibility and leadtime, including an optional CO₂-equivalent approach and, for N₂ O, more leadtime before testing will be required to demonstrate compliance with the emissions standard (in interim, manufacturers may certify based on a compliance statement based on good engineering judgment).

a. Nitrous Oxide (N₂ O) Exhaust Emission Standard

As stated above, N₂ O is a global warming gas with a high global warming potential. It accounts for about 2.3% of the current greenhouse gas emissions from cars and light trucks. EPA is setting a per-vehicle N₂ O emission standard of 0.010 g/mi, measured over the traditional FTP vehicle laboratory test cycles. The standard will become effective in model year 2012 for all light-duty cars and trucks. The standard is designed to prevent increases in N₂ O emissions from current levels; i.e., it is a no-backsliding standard.

N₂ O is emitted from gasoline and diesel vehicles mainly during specific catalyst temperature conditions conducive to N₂ O formation. Specifically, N₂ O can be generated during periods of emission hardware warm-up when rising catalyst temperatures pass through the temperature window when N₂ O formation potential is possible. For current Tier 2 compatible gasoline engines with conventional three-way catalyst technology, N₂ O is not generally produced in significant amounts because the time the catalyst spends at the critical temperatures during warm-up is short. This is largely due to the need to quickly reach the higher temperatures necessary for high catalyst efficiency to achieve emission compliance for criteria pollutants. As several auto manufacturer comments noted, N₂ O is a more significant concern with diesel vehicles, and potentially future gasoline lean-burn engines, equipped with advanced catalytic NO_Xemissions control systems. In the absence of N₂ O emission standards, these systems could be designed in a way that emphasizes efficient NO_X control while at the same time allowing the formation of significant quantities of N₂ O. Excess oxygen present in the exhaust during lean-burn conditions in diesel or lean-burn gasoline engines equipped with these advanced systems can favor N₂ O formation if catalyst temperatures are not carefully controlled. Without specific attention to controlling N₂ O emissions in the development of such new NO_X control systems, vehicles could have N₂ O emissions many times greater than are emitted by current gasoline vehicles.

EPA is setting an N₂ O emission standard that the agency believes will be met by current-technology gasoline vehicles at essentially no cost. As just noted, N₂ O formation in current catalyst systems occurs, but the emission levels are relatively low, because the time the catalyst spends at the critical temperatures during warm-up when N₂ O can form is short. At the same time, EPA believes that the standard will ensure that the design of advanced NO_X control systems, especially for future diesel and lean-burn gasoline vehicles, will control N₂ O emission levels. While current NO_X control approaches used on current Tier 2 diesel vehicles do not tend to favor the formation of N₂ O emissions, EPA believes that this N₂ O standard will discourage new emission control designs that achieve criteria emissions compliance at the cost of increased N₂ O emissions. Thus, the standard will cap N₂ O emission levels, with the expectation that current gasoline and diesel vehicle control approaches that comply with the Tier 2 vehicle emission standards for NO_X will not increase their emission levels, and that the cap will ensure that future vehicle designs will be appropriately controlled for N₂ O emissions.

The level of the N₂ O standard is approximately two times the average N₂ O level of current gasoline passenger cars and light-duty trucks that meet the Tier 2 NO_X standards. EPA has not previously regulated N₂ O emissions, and available data on current vehicles is limited. However, EPA derived the standard from a combination of emission factor values used in modeling light duty vehicle emissions and limited recent EPA test data. Because the standard represents a level 100 percent higher than the average current N₂ O level, we continue to believe that most if not all Tier 2 compliant gasoline and diesel vehicles will easily be able to meet the standards. Manufacturers typically use design targets for NO_X emission levels of about 50% of the standard, to account for in-use emissions deterioration and normal testing and production variability, and EPA expects that manufacturers will use a similar approach for N₂ O emission compliance. EPA did not propose and is not finalizing a more stringent standard for current vehicles because we believe that the stringent Tier 2 program and the associated NO_X fleet average requirement already result in significant N₂ O control, and the agency does not expect current N₂ O levels to rise for these vehicles. Moreover, EPA believes that the CO₂ standards will be challenging for the industry and that these standards should be the industry's chief focus in this first phase of vehicular GHG emission controls. See Massachusetts v. EPA, 549 U.S. at 533 (EPA has significant discretion as to timing of GHG regulations); see also Sierra Club v. EPA, 325 F. 3d 374, 379 (DC Cir. 2003) (upholding anti-backsliding standards for air toxics under technology-forcing section 202 (l) because it is reasonable for EPA to assess the effects of its other regulations on the motor vehicle sector before aggressively regulating emissions of toxic vehicular air pollutants.

Diesel cars and light trucks with advanced emission control technology are in the early stages of development and commercialization. As this segment of the vehicle market develops, the N₂ O standard will likely require these manufacturers to incorporate control strategies that minimize N₂ O formation. Available approaches include using electronic controls to limit catalyst conditions that might favor N₂ O formation and consider different catalyst formulations. While some of these approaches may have modest associated costs, EPA believes that they will be small compared to the overall costs of the advanced NO_X control technologies already required to meet Tier 2 standards.

In the proposal, EPA sought comment on an approach of expressing N₂ O and CH₄ in common terms of CO₂-equivalent emissions and combining them into a single standard along with CO₂ emissions. 74 FR at 49524. California's “Pavley” program adopted such a CO₂-equivalent emissions standards approach to GHG emissions. EPA was primarily concerned that such an approach could undermine the stringency of the CO₂ standards, as the proposed standards were designed to “cap” N₂ O and CH₄ emissions, rather than reflecting a level either that is the industry fleet-wide average or that would effect reductions in these GHGs.

As noted above, several auto manufacturers expressed interest in such a CO₂-equivalent approach, due to concerns that the caps could be limiting for some advanced technology vehicles. While we continue to believe that the vast majority of light-duty vehicles will be able to easily meet the standards, we acknowledge that advanced diesel or lean-burn gasoline vehicles of the future may face slightly greater challenges. Therefore, after considering these comments, EPA is finalizing an optional compliance approach to provide flexibility for any advanced technologies that may have challenges in meeting the N₂ O or CH₄ cap standards.

In lieu of complying with the separate N₂ O and CH₄ cap standards, a manufacturer may choose to comply with a CO₂-equivalent standard. A manufacturer choosing this option will convert its N₂ O and CH₄ test results (or, as described below, a default N₂ O value for MY 2012-2014) into CO₂-equivalent values and add this sum to their CO₂ emissions. This CO₂-equivalent value will still need to comply with the manufacturer's footprint-based CO₂ target level. In other words, a manufacturer could offset any N₂ O emissions (or any CH₄ emissions) by taking steps to further reduce CO₂. A manufacturer choosing this option will need to apply this approach to all of the test groups in its fleet. This approach is more environmentally protective overall than the cap standard approach, since the manufacturer will need to reduce its CO₂ emissions to offset the higher N₂ O (or CH₄) levels, but will not be allowed to increase CO₂ above its footprint target level by reducing N₂ O (or CH₄).

The compliance level in g/mi for the optional CO₂-equivalent approach for gasoline vehicles is calculated as CO₂ + (CWF/0.273 × NMHC) + (1.571 × CO) + (298 × N₂ O) + (25 × CH₄). The N₂ O and CH₄ values are the measured emission values for these GHGs, except N₂ O in model years 2012 through 2014. For these model years, manufacturers may use a default N₂ O value of 0.010 g/mi, the same value as the N₂ O cap standard. For MY 2015 and later, the manufacturer would need to provide actual test data on the emission data vehicle for each test group. (That is, N₂ O data would not be required for each model type, since EPA believes that there will likely be little N₂ O variability among model types within a test group.) EPA believes that its selection of 0.010 g/mi as the N₂ O default value is an appropriately protective level, on the high end of current technologies, as further discussed below. Consistent with the other elements of the equation, N₂ O and CH₄ must be included at full useful life deteriorated values. This requires testing using the highway test cycle in addition to the FTP during the manufacturer's deterioration factor (DF) development program. However, EPA recognizes that manufacturers may not be able to develop DFs for N₂ O and CH₄ for all their vehicles in the 2012 model year, and thus EPA is allowing the use of alternative values through the 2014 model year. For N₂ O the alternative value is the DF developed for NO_X emissions, and for CH₄ the alternative value is the DF developed for NMOG emissions. Finally, for manufacturers using this option, the CO₂-equivalent emission level would also be the basis for any credits that the manufacturer might generate.

Manufacturers expressed concerns about their ability to acquire and install N₂ O analytical equipment. However, the agency continues to believe that such burdens, while not trivial, will also not be excessive. While many manufacturers do not appear to have invested yet in adding N₂ O measurement equipment to their test facilities, EPA is not aware of any information to indicate that that suppliers will have difficulty providing sufficient hardware, or that such equipment is unusually expensive or complex compared to existing measurement hardware. EPA allows N₂ O measurement using any of four methods, all of which are commercially available today. The costs of certification and other indirect costs of this rule are accounted for in the Indirect Cost Multipliers, discussed in Section III.H below.

Still, given the short lead-time for this rule and the newness of N₂ O testing to this industry, EPA proposed that manufacturers be able to apply for a certificate of conformity with the N₂ O standard for model year 2012 provided that they supply a compliance statement based on good engineering judgment. Under the proposal, beginning in MY 2013, manufacturers would have needed to base certification on actual N₂ O testing data. This approach was intended to reasonably ensure that the emission standards are being met, while allowing manufacturers lead-time to purchase new N₂ O emissions measurement equipment, modify certification test facilities, and begin N₂ O testing. After consideration of the comments, EPA agrees with manufacturers that one year of additional lead-time to begin actual N₂ O measurement across their vehicle fleets may still be insufficient for manufacturers to efficiently make the necessary facility changes and equipment purchases. Therefore, EPA is extending the ability to certify based on a compliance statement for two additional years, through model year 2014. For 2015 and later model years, manufacturers will need to submit measurements of N₂ O for compliance purposes.

b. Methane (CH₄) Exhaust Emission Standard

Methane (CH₄) is a greenhouse gas with a high global warming potential. It accounts for about 0.2% of the greenhouse gases from cars and light trucks.

EPA is setting a CH₄ emission standard of 0.030 g/mi as measured on the FTP, to apply beginning with model year 2012 for both cars and trucks. EPA believes that this level for the standard will be met by current gasoline and diesel vehicles, and will prevent large increases in future CH₄ emissions. This is particularly a concern in the event that alternative fueled vehicles with high methane emissions, like some past dedicated compressed natural gas (CNG) vehicles and some flexible-fueled vehicles when operated on E85 fuel, become a significant part of the vehicle fleet. Currently EPA does not have separate CH₄ standards because unlike other hydrocarbons it does not contribute significantly to ozone formation. However, CH₄ emissions levels in the gasoline and diesel car and light truck fleet have nevertheless generally been controlled by the Tier 2 standards for non-methane organic gases (NMOG). However, without an emission standard for CH₄, there is no guarantee that future emission levels of CH₄ will remain at current levels as vehicle technologies and fuels evolve.

The standard will cap CH₄ emission levels, with the expectation that emissions levels of current gasoline and diesel vehicles meeting the Tier 2 emission standards will not increase. The level of the standard will generally be achievable for typical vehicles through normal emission control methods already required to meet the Tier 2 emission standards for NMOG. Also, since CH₄ is already measured under the current Tier 2 regulations (so that it may be subtracted to calculate non-methane hydrocarbons), we believe that the standard will not result in any additional testing costs. Therefore, EPA is not attributing any costs to this part of this program. Since CH₄ is produced during fuel combustion in gasoline and diesel engines similarly to other hydrocarbon components, controls targeted at reducing overall NMOG levels are generally also effective in reducing CH₄ emissions. Therefore, for typical gasoline and diesel vehicles, manufacturer strategies to comply with the Tier 2 NMOG standards have to date tended to prevent increases in CH₄ emissions levels. The CH₄ standard will ensure that emissions will be addressed if in the future there are increases in the use of natural gas or other alternative fuels or technologies that may result in higher CH₄ emissions.

As with the N₂ O standard, EPA is setting the level of the CH₄ standard to be approximately two times the level of average CH₄ emissions from Tier 2 gasoline passenger cars and light-duty trucks. EPA believes the standard will easily be met by current gasoline vehicles, and that flexible fuel vehicles operating on ethanol can be designed to resolve any potential CH₄ emissions concerns. Similarly, since current diesel vehicles generally have even lower CH₄ emissions than gasoline vehicles, EPA believes that diesels will also meet the CH₄ standard. However, EPA also believes that to set a CH₄ emission standard more stringent than the proposed standard could effectively make the Tier 2 NMOG standard more stringent and is inappropriate for that reason (and untimely as well, given the challenge of meeting the CO₂ standards, as noted above).

Some CNG-fueled vehicles have historically produced significantly higher CH₄ emissions than gasoline or diesel vehicles. This is because CNG fuel is essentially methane and any unburned fuel that escapes combustion and is not oxidized by the catalyst is emitted as methane. However, in recent model years, the few dedicated CNG vehicles sold in the U.S. meeting the Tier 2 standards have had CH₄ control as effective as that of gasoline or diesel vehicles. Still, even if these vehicles meet the Tier 2 NMOG standard and appear to have effective CH₄ control by nature of the NMOG controls, Tier 2 standards do not require CH₄ control. Although EPA believes that in most cases that the CH₄ cap standard should not require any different emission control designs beyond what is already required to meet Tier 2 NMOG standards on a dedicated CNG vehicle, the cap will ensure that systems maintain the current level of CH₄ control.

Some manufacturers have also expressed some concerns about CH₄ emissions from flexible-fueled vehicles operating on E85 (85% ethanol, 15% gasoline). However, we are not aware of any information that would indicate that if engine-out CH₄ proves to be higher than for a typical gasoline vehicle, that such emissions could not be managed by reasonably available control strategies (perhaps similar to those used in dedicated CNG vehicles).

As described above, in response to the comments, EPA will also allow manufacturers to choose to comply with a CO₂-equivalent standard in lieu of complying with a separate CH₄ cap standard. A manufacturer choosing this option would convert its N₂ O and CH₄ test results into CO₂-equivalent values (using the respective GWP values), and would then compare this value to the manufacturer's footprint-based CO₂ target level to determine compliance. However, as with N₂ O, this approach will not permit a manufacturer to increase its CO₂ by reducing CH₄; the company's footprint-based CO₂ target level would remain the same.

8. Small Entity Exemption

As proposed, EPA is exempting from GHG emissions standards small entities meeting the Small Business Administration (SBA) size criteria of a small business as described in 13 CFR 121.201. EPA will instead consider appropriate GHG standards for these entities as part of a future regulatory action. This includes both U.S.-based and foreign small entities in three distinct categories of businesses for light-duty vehicles: small volume manufacturers, independent commercial importers (ICIs), and alternative fuel vehicle converters.

EPA has identified about 13 entities that fit the Small Business Administration (SBA) size criterion of a small business. EPA estimates there currently are approximately two small volume manufacturers, eight ICIs, and three alternative fuel vehicle converters in the light-duty vehicle market. Further detail is provided in Section III.I.3, below. EPA estimates that these small entities comprise less than 0.1 percent of the total light-duty vehicle sales in the U.S., and therefore the exemption will have a negligible impact on the GHG emissions reductions from the standards.

To ensure that EPA is aware of which companies would be exempt, EPA proposed to require that such entities submit a declaration to EPA containing a detailed written description of how that manufacturer qualifies as a small entity under the provisions of 13 CFR 121.201. EPA has reconsidered the need for this additional submission under the regulations and is deleting it as not necessary. We already have information on the limited number of small entities that we expect would receive the benefits of the exemption, and do not need the proposed regulatory requirement to be able to effectively implement this exemption for those parties who in fact meet its terms. Small entities are currently covered by a number of EPA motor vehicle emission regulations, and they routinely submit information and data on an annual basis as part of their compliance responsibilities.

EPA did not receive adverse comments regarding the proposed small entity exemption. EPA received comments concerning whether or not the small entity exemption applies to foreign manufacturers. EPA clarifies that foreign manufacturers meeting the SBA size criteria are eligible for the exemption, as was EPA's intent during the proposal.

C. Additional Credit Opportunities for CO ₂ Fleet Average Program

The final standards represent a significant multi-year challenge for manufacturers, especially in the early years of the program. Section III.B.4 above describes EPA's provisions for manufacturers to be able to generate credits by achieving fleet average CO₂ emissions below their fleet average standard, and also how manufacturers can use credits to comply with the standards. As described in Section III.B.4, credits can be carried forward five years, carried back three years, transferred between vehicle categories, and traded between manufacturers. The credits provisions described below provide manufacturers with additional ways to earn credits starting in MY 2012. EPA is also including early credits provisions for the 2009-2011 model years, as described below in Section III.C.5.

The provisions described below provide additional flexibility, especially in the early years of the program. This helps to address issues of lead-time or technical feasibility for various manufacturers and in several cases provides an incentive for promotion of technology pathways that warrant further development. EPA is finalizing a variety of credit opportunities because manufacturers are not likely to be in a position to use every credit provision. EPA expects that manufacturers are likely to select the credit opportunities that best fit their future plans.

EPA believes it is critical that manufacturers have options to ease the transition to the final MY 2016 standards. At the same time, EPA believes these credit programs must be and are designed in a way to ensure that they achieve emission reductions that achieve real-world reductions over the full useful life of the vehicle (or, in the case of FFV credits and Advanced Technology incentives, to incentivize the introduction of those vehicle technologies) and are verifiable. In addition, EPA believes that these credit programs do not provide an opportunity for manufacturers to earn “windfall” credits. Comments on the proposed EPA credit programs are summarized below along with EPA's response, and are detailed in the Response to Comments document.

1. Air Conditioning Related Credits

Manufacturers will be able to generate and use credits for improved air conditioner (A/C) systems in complying with the CO₂ fleetwide average standards described above (or otherwise to be able to bank or trade the credits). EPA expects that most manufacturers will choose to utilize the A/C provisions as part of its compliance demonstration (and for this reason cost of compliance with A/C related emission reductions are assumed in the cost analysis). The A/C provisions are structured as credits, unlike the CO₂ standards for which manufacturers will demonstrate compliance using 2-cycle (city/highway) tests (see Sections III.B and III.E.). Those tests do not measure either A/C leakage or tailpipe CO₂ emissions attributable to A/C load. Thus, it is a manufacturer's option to include A/C GHG emission reductions as an aspect of its compliance demonstration. Since this is an elective alternative, EPA is referring to the A/C part of the rule as a credit.

EPA estimates that direct A/C GHG emissions—emissions due to the leakage of the hydrofluorocarbon refrigerant in common use today—account for 5.1% of CO₂-equivalent GHGs from light-duty cars and trucks. This includes the direct leakage of refrigerant as well as the subsequent leakage associated with maintenance and servicing, and with disposal at the end of the vehicle's life. The emissions that are associated with leakage reductions are the direct leakage and the leakage associated with maintenance and servicing. Together these are equivalent to CO₂ emissions of approximately 13.6 g/mi per car and light-truck. EPA also estimates that indirect GHG emissions (additional CO₂ emitted due to the load of the A/C system on the engine) account for another 3.9% of light-duty GHG emissions. This is equivalent to CO₂ emissions of approximately 14.2 g/mi per vehicle. The derivation of these figures can be found in Chapter 2.2 of the EPA RIA.

EPA believes that it is important to address A/C direct and indirect emissions because the technologies that manufacturers will employ to reduce vehicle exhaust CO₂ will have little or no impact on A/C related emissions. Without addressing A/C related emissions, as vehicles become more efficient, the A/C related contribution will become a much larger portion of the overall vehicle GHG emissions.

Over 95% of the new cars and light trucks in the United States are equipped with A/C systems and, as noted, there are two mechanisms by which A/C systems contribute to the emissions of greenhouse gases: Through leakage of refrigerant into the atmosphere and through the consumption of fuel to provide mechanical power to the A/C system. With leakage, it is the high global warming potential (GWP) of the current automotive refrigerant (HFC-134a, with a GWP of 1430) that results in the CO₂-equivalent impact of 13.6 g/mi. Due to the high GWP of this HFC, a small leakage of the refrigerant has a much greater global warming impact than a similar amount of emissions of CO₂ or other mobile source GHGs. Manufacturers can reduce A/C leakage emissions by using leak-tight components. Also, manufacturers can largely eliminate the global warming impact of leakage emissions by adopting systems that use an alternative, low-GWP refrigerant, as discussed below. The A/C system also contributes to increased CO₂ emissions through the additional work required to operate the compressor, fans, and blowers. This additional work typically is provided through the engine's crankshaft, and delivered via belt drive to the alternator (which provides electric energy for powering the fans and blowers) and the A/C compressor (which pressurizes the refrigerant during A/C operation). The additional fuel used to supply the power through the crankshaft necessary to operate the A/C system is converted into CO₂ by the engine during combustion. This incremental CO₂ produced from A/C operation can thus be reduced by increasing the overall efficiency of the vehicle's A/C system, which in turn will reduce the additional load on the engine from A/C operation.

Manufacturers can make very feasible improvements to their A/C systems to address A/C system leakage and efficiency. EPA is finalizing two separate credit approaches to address leakage reductions and efficiency improvements independently. A leakage reduction credit will take into account the various technologies that could be used to reduce the GHG impact of refrigerant leakage, including the use of an alternative refrigerant with a lower GWP. An efficiency improvement credit will account for the various types of hardware and control of that hardware available to increase the A/C system efficiency. For purposes of use of A/C credits at certification, manufacturers will be required to attest to the durability of the leakage reduction and the efficiency improvement technologies over the full useful life of the vehicle.

EPA believes that both reducing A/C system leakage and increasing efficiency are highly cost-effective and technologically feasible. EPA expects most manufacturers will choose to use these A/C credit provisions, although some may not find it necessary to do so.

a. A/C Leakage Credits

The refrigerant used in vehicle A/C systems can get into the atmosphere by many different means. These refrigerant emissions occur from the slow leakage over time that all closed high pressure systems will experience. Refrigerant loss occurs from permeation through hoses and leakage at connectors and other parts where the containment of the system is compromised. The rate of leakage can increase due to deterioration of parts and connections as well. In addition, there are emissions that occur during accidents and maintenance and servicing events. Finally, there are end-of-life emissions if, at the time of vehicle scrappage, refrigerant is not fully recovered.

Because the process of refrigerant leakage has similar root causes as those that cause fuel evaporative emissions from the fuel system, some of the emission control technologies are similar (including hose materials and connections). There are, however, some fundamental differences between the systems that require a different approach, both to controlling and to documenting that control. The most notable difference is that A/C systems are completely closed systems and always under significant pressure, whereas the fuel system is not. Fuel systems are meant to be refilled as liquid fuel is consumed by the engine, while the A/C system ideally should never require “recharging” of the contained refrigerant. Thus it is critical that the A/C system leakages be kept to an absolute minimum. As a result, these emissions are typically too low to accurately measure in most current SHED chambers designed for fuel evaporative emissions measurement, especially for A/C systems that are new or early in life.

A few commenters suggested that we allow manufacturers, as an option, to use an industry-developed “mini-shed” test procedure (SAE J2763—Test Procedure for Determining Refrigerant Emissions from Mobile Air Conditioning Systems) to measure and report annual refrigerant leakage. However, while EPA generally prefers performance testing, for an individual vehicle A/C system or component, there is not a strong inherent correlation between a performance test using SAE J2763 and the design-based approach we are adopting (based on SAE J2727, as discussed below). Establishing such a correlation would require testing of a fairly broad range of current-technology systems in order to establish the effects of such factors as production variability and assembly practices (which are included in J2727 scores, but not in J2763 measurements). To EPA's knowledge, such a correlation study has not been done. At the same time, as discussed below, there are indications that much of the industry will eventually be moving toward alternative refrigerants with very low GWPs. EPA believes such a transition would diminish the value of any correlation studies that might be done to confirm the appropriateness of the SAE J2763 procedure as an option in this rule. For these reasons, EPA is therefore not adopting such an optional direct measurement approach to addressing refrigerant leakage at this time.

Instead, as proposed, EPA is adopting a design-based method for manufacturers to demonstrate improvements in their A/C systems and components. Manufacturers implementing system designs expected to result in reduced refrigerant leakage will be eligible for credits that could then be used to meet their CO₂ emission compliance requirements (or otherwise banked or traded). The A/C Leakage Credit provisions will generally assign larger credits to system designs that would result in greater leakage reductions. In addition, proportionately larger A/C Leakage Credits will be available to manufacturers that substitute a refrigerant with lower GWP than the current HFC-134a refrigerant.

Our method for calculating A/C Leakage Credits is based closely on an industry-consensus leakage scoring method, described below. This leakage scoring method is correlated to experimentally-measured leakage rates from a number of vehicles using the different available A/C components. Under the approach, manufacturers will choose from a menu of A/C equipment and components used in their vehicles in order to establish leakage scores which will characterize their A/C system leakage performance. Credits will be generated from leakage reduction improvements that exceed average fleetwide leakage rates.

EPA believes that the design-based approach will result in estimates of leakage emissions reductions that will be comparable to those that will eventually result from performance-based testing. We believe that this method appropriately approximates the real-world leakage rates for the expected MY 2012-2016 A/C systems.

The cooperative industry and government Improved Mobile Air Conditioning (IMAC) program has demonstrated that new-vehicle leakage emissions can be reduced by 50%. This program has shown that this level of improvement can be accomplished by reducing the number and improving the quality of the components, fittings, seals, and hoses of the A/C system. All of these technologies are already in commercial use and exist on some of today's systems.

As proposed, a manufacturer wishing to generate A/C Leakage Credits will compare the components of its A/C system with a set of leakage-reduction technologies and actions based closely on that developed through IMAC and the Society of Automotive Engineers (as SAE Surface Vehicle Standard J2727, August 2008 version). The J2727 approach was developed from laboratory testing of a variety of A/C related components, and EPA believes that the J2727 leakage scoring system generally represents a reasonable correlation with average real-world leakage in new vehicles. The EPA credit approach addresses the same A/C components as does SAE J2727 and associates each component with the same gram-per-year leakage rate as the SAE method, although, as described below, EPA limits the credits allowed and also modifies it for other factors such as alternative refrigerants.

A manufacturer choosing to generate A/C Leakage Credits will sum the leakage values for an A/C system for a total A/C leakage score according to the following formula. Because the primary GHG program standards are expressed in terms of vehicle exhaust CO₂ emissions as measured in grams per mile, the credits programs adopted in this rule, including A/C related credits, must ultimately be converted to a common metric for proper calculation of credits toward compliance with the primary vehicle standards. This formula describes the conversion of the grams-per-year leakage score to a grams-per-mile CO₂ eq value, taking vehicle miles traveled (VMT) and the GWP of the refrigerant into account:

A/C Leakage Credit = (MaxCredit) * [1−(LeakScore/AvgImpact) * (GWPRefrigerant/1430)]

Where:

MaxCredit is 12.6 and 15.6 g/mi CO₂ eq for cars and trucks, respectively. These values become 13.8 and 17.2 for cars and trucks, respectively, if low-GWP refrigerants are used, since this would generate additional credits from reducing emissions during maintenance events, accidents, and at end-of-life.

LeakScore is the leakage score of the A/C system as measured according to the EPA leakage method (based on the J2727 procedure, as discussed above) in units of g/yr. The minimum score that EPA considers feasible is fixed at 8.3 and 10.4 g/yr for cars and trucks respectively (4.1 and 5.2 g/yr for systems using electric A/C compressors) as discussed below.

Avg Impact is the average current A/C leakage emission rate, which is 16.6 and 20.7 g/yr for cars and trucks, respectively.

GWPRefrigerant is the global warming potential (GWP) for direct radiative forcing of the refrigerant. For purposes of this rule, the GWP of HFC-134a is 1430, the GWP of HFC-152a is 124, the GWP of HFO-1234yf is 4, and the GWP of CO₂ as a refrigerant is 1.

The EPA Final RIA elaborates further on the development of each of the values incorporated in the A/C Leakage Credit formula above, as summarized here. First, as proposed, EPA estimates that leakage emission rates for systems using the current refrigerant (HFC-134a) could be feasibly reduced to rates no less than 50% of current rates—or 8.3 and 10.4 g/yr for cars and trucks, respectively—based on the conclusions of the IMAC study as well as consideration of refrigerant emissions over the full life of the vehicle.

Also, some commenters noted that A/C compressors powered by electric motors (e.g. as used today in several hybrid vehicle models) were not included in the IMAC study and yet allow for leakage emission rate reductions beyond EPA's estimates for systems with conventional belt-driven compressors. EPA agrees with these comments, and we have incorporated lower minimum emission rates into the formula above—4.1 and 5.2 g/yr for cars and trucks, respectively—in order to allow additional leakage reduction credits for vehicles that use sealed electric A/C compressors. The maximum available credits for these two approaches are summarized in Table III.C.1-1 below.

AIAM commented that EPA should not set a lower limit on the leakage score, even for non-electric compressors. EPA has determined not to do so. First, although there do exist vehicles in the Minnesota data with lower scores than our proposed (and now final) minimum scores, there are very few car models that have scores less than 8.3, and these range from 7.0 to about 8.0 and the difference are small compared to our minimum score. More important, lowering the leakage limit would necessarily increase credit opportunities for equipment design changes, and EPA believes that these changes could discourage the environmentally optimal result of using low GWP refrigerants. Introduction of low GWP refrigerants could be discouraged because it may be less costly to reduce leakage than to replace many of the A/C system components. Moreover, due to the likelihood of in-use factors, even a leakless (according to J2727) R134a system will have some emissions due to manufacturing variability, accidents, deterioration, maintenance, and end of life emissions, a further reason to cap the amount of credits available through equipment design. The only way to guarantee a near zero emission system in-use is to use a low GWP refrigerant. The EPA has therefore decided for the purposes of this final rule to not change the minimum score for belt driven compressors due to the reason cited above and to the otherwise overwhelming support for the program as proposed from commenters.

In addition, as discussed above, EPA recognizes that substituting a refrigerant with a significantly lower GWP will be a very effective way to reduce the impact of all forms of refrigerant emissions, including maintenance, accidents, and vehicle scrappage. To address future GHG regulations in Europe and California, systems using alternative refrigerants—including HFO1234yf, with a GWP of 4 and CO₂ with a GWP of 1—are under serious development and have been demonstrated in prototypes by A/C component suppliers. The European Union has enacted regulations phasing in alternative refrigerants with GWP less than 150 starting this year, and the State of California proposed providing credits for alternative refrigerant use in its GHG rule. Within the timeframe of MYs 2012-2016, EPA is not expecting widespread use of low-GWP refrigerants. However, EPA believes that these developments are promising, and, as proposed, has included in the A/C Leakage Credit formula above a factor to account for the effective GHG reductions that could be expected from refrigerant substitution. The A/C Leakage Credits that will be available will be a function of the GWP of the alternative refrigerant, with the largest credits being available for refrigerants with GWPs at or approaching a value of 1. For a hypothetical alternative refrigerant with a GWP of 1 (e.g., CO₂ as a refrigerant), effectively eliminating leakage as a GHG concern, our credit calculation method could result in maximum credits equal to total average emissions, or credits of 13.8 and 17.2 g/mi CO₂ eq for cars and trucks, respectively, as incorporated into the A/C Leakage Credit formula above as the “MaxCredit” term.

Table III.C.1-1 summarizes the maximum A/C leakage credits available to a manufacturer, according to the formula above.

It is possible that alternative refrigerants could, without compensating action by the manufacturer, reduce the efficiency of the A/C system (see related discussion of the A/C Efficiency Credit below.) However, as noted at proposal and discussed further in the following section, EPA believes that manufacturers will have substantial incentives to design their systems to maintain the efficiency of the A/C system. Therefore EPA is not accounting for any potential efficiency degradation due to the use of alternative refrigerants.

Beyond the comments mentioned above, commenters generally supported or were silent about EPA's refrigerant leakage methodology (as based on SAE J2727), including the maximum leakage credits available, the technologies eligible for credit and their associated leakage reduction values, and the potential for alternative refrigerants. All comments related to A/C credits are addressed in the Response to Comments Document.

b. A/C Efficiency Credits

Manufacturers that make improvements in their A/C systems to increase efficiency and thus reduce CO₂ emissions due to A/C system operation may be eligible for A/C Efficiency Credits. As with A/C Leakage Credits, manufacturers could apply A/C Efficiency Credits toward compliance with their overall CO₂ standards (or otherwise bank and trade the credits).

As mentioned above, EPA estimates that the CO₂ emissions due to A/C related loads on the engine account for approximately 3.9% of total greenhouse gas emissions from passenger vehicles in the United States. Usage of A/C systems is inherently higher in hotter and more humid months and climates; however, vehicle owners may use their A/C systems all year round in all parts of the nation. For example, people commonly use A/C systems to cool and dehumidify the cabin air for passenger comfort on hot humid days, but they also use the systems to de-humidify cabin air to assist in defogging/de-icing the front windshield and side glass in cooler weather conditions for improved visibility. A more detailed discussion of seasonal and geographical A/C usage rates can be found in the RIA.

Most of the additional load on the engine from A/C system operation comes from the compressor, which pumps the refrigerant around the system loop. Significant additional load on the engine may also come from electric or hydraulic fans, which are used to move air across the condenser, and from the electric blower, which is used to move air across the evaporator and into the cabin. Manufacturers have several currently-existing technology options for improving efficiency, including more efficient compressors, fans, and motors, and system controls that avoid over-chilling the air (and subsequently re-heating it to provide the desired air temperature with an associated loss of efficiency). For vehicles equipped with automatic climate-control systems, real-time adjustment of several aspects of the overall system (such as engaging the full capacity of the cooling system only when it is needed, and maximizing the use of recirculated air) can result in improved efficiency. Table III.C.1-2 below lists some of these technologies and their respective efficiency improvements.

As discussed in the proposal, EPA is adopting a design-based “menu” approach for estimating efficiency improvements and, thus, quantifying A/C Efficiency Credits. However, EPA's ultimate preference is performance-based standards and credit mechanisms (i.e., using actual measurements) as typically providing a more accurate measure of performance. However, EPA has concluded that a practical, performance-based procedure for the purpose of accurately quantifying A/C-related CO₂ emission reductions, and thus efficiency improvements for assigning credits, is not yet available. Still, EPA is introducing a new specialized performance-based test for the more limited purpose of demonstrating that actual efficiency improvements are being achieved by the design improvements for which a manufacturer is seeking A/C credits. As discussed below, beginning in MY 2014, manufacturers wishing to generate A/C Efficiency Credits will need to show improvement on the new A/C Idle Test in order to then use the “menu” approach to quantify the number of credits attributable to those improvements.

In response to comments concerning the applicability and effectiveness of technologies that were or were not included in our analysis, we have made several changes to the design-based menu. First, we have separated the credit available for `recirculated air' technologies into those with closed-loop control of the air supply and those with open-loop control. By “closed-loop” control, we mean a system that uses feedback from a sensor, or sensors, (e.g., humidity, glass fogging, CO₂, etc.) to actively control the interior air quality. For those systems that use “open-loop” control of the air supply, we project that since this approach cannot precisely adjust to varying ambient humidity or passenger respiration levels, the relative effectiveness will be less than that for systems using closed-loop control.

Second, many commenters indicated that the electronic expansion valve, or EXV, should not be included in the menu of technologies, as its effectiveness may not be as high as we projected. Commenters noted that the SAE IMAC report stated efficiency improvements for an EXV used in conjunction with a more efficient compressor, and not as a stand alone technology and that no manufacturers are considering this technology for their products within the timeframe of this rulemaking. We believe other technologies (improved compressor controls for example) can achieve the same benefit as an EXV, without the need for this unique component, and therefore are not adopting it as an option in the design menu of efficiency-improving A/C technologies.

Third, many commenters requested that an internal heat exchanger, or IHX, be added to the design menu. EPA initially considered adding this technology, but in our initial review of studies on this component, we had understood that the value of the technology is limited to systems using the alternative refrigerant HFO-1234yf. Some manufacturers, however, commented that an IHX can also be used with systems using the current refrigerant HFC-134a to improve efficiency, and that they plan on implementing this technology as part their strategy to improve A/C efficiency. Based on these comments, and projections in a more recent SAE Technical Paper, we project that an IHX in a conventional HFC-134a system can improve system efficiency by 20%, resulting in a credit of 1.1 g/mi. Further discussion of IHX technology can be found in the RIA.

Fourth, we have modified the definition of `improved evaporators and condensers' to recognize that improved versions of these heat exchangers may be used separately or in conjunction with one another, and that an engineering analysis must indicate a COP improvement of 10% or better when using either or both components (and not a 10% COP improvement for each component). Furthermore, we have modified the regulation text to clarify what is considered to be the `baseline' components for this analysis. We consider the baseline component to be the version which a manufacturer most recently had in production on the same vehicle or a vehicle in a similar EPA vehicle classification. The dimensional characteristics (e.g. tube configuration/thickness/spacing, and fin density) of the baseline components are then compared to the new components, and an engineering analysis is required to demonstrate the COP improvement.

For model years 2012 and 2013, a manufacturer wishing to generate A/C Efficiency Credits for a group of its vehicles with similar A/C systems will compare several of its vehicle A/C-related components and systems with a list of efficiency-related technology improvements (see Table III.C.1-2 below). Based on the technologies the manufacturer chooses, an A/C Efficiency Credit value will be established. This design-based approach will recognize the relationships and synergies among efficiency-related technologies. Manufacturers could receive credits based on the technologies they chose to incorporate in their A/C systems and the associated credit value for each technology. The total A/C Efficiency Credit will be the total of these values, up to a maximum allowable credit of 5.7 g/mi CO₂ eq. This will be the maximum improvement from current average efficiencies for A/C systems (see the RIA for a full discussion of our derivation of the reductions and credit values for individual technologies and for the maximum total credit available). Although the total of the individual technology credit values may exceed 5.7 g/mi CO₂ eq, synergies among the technologies mean that the values are not additive. A/C Efficiency Credits as adopted may not exceed 5.7 g/mi CO₂ eq.

The proposal requested comment on adjusting the efficiency credit for alternative refrigerants. Although a few commenters noted that the efficiency of an HFO1234yf system may differ from a current HFC-134a system, we believe that this difference does not take into account any efficiency improvements that may be recovered or gained when the overall system is specifically designed with consideration of the new refrigerant properties (as compared to only substituting the new refrigerant). EPA is therefore not adjusting the credits based on efficiency differences for this rule.

As noted above, for model years 2014 and later, manufacturers seeking to generate design-based A/C Efficiency Credits will also need to use a specific new EPA performance test to confirm that the design changes are resulting in improvements in A/C system efficiency as integrated into the vehicle. As proposed, beginning in MY 2014 manufacturers will need to perform an A/C CO₂ Idle Test for each A/C system (family) for which it desires to generate Efficiency Credits. Manufacturers will need to demonstrate an improvement over current average A/C CO₂ levels (21.3 g/minute on the Idle Test) to qualify for the menu approach credits. Upon qualifying on the Idle Test, the manufacturer will be eligible to use the menu approach above to quantify the potential credits it could generate. To earn the full amount of credits available in the menu approach (limited to the maximum), the test must demonstrate a 30% or greater improvement in CO₂ levels over the current average.

For A/C systems that achieve an improvement between 0-and-30% (or a result between 21.3 and 14.9 g/minute result on the A/C CO₂ Idle Test), a credit can still be earned, but a multiplicative credit adjustment factor will be applied to the eligible credits. As shown in Figure III.C.1-1 this factor will be scaled from 1.0 to 0, with vehicles demonstrating a 30% or better improvement (14.9 g/min or lower) receiving 100% of the eligible credit (adj. factor = 1.0), and vehicles demonstrating a 0% improvement—21.3 g/min or higher result—receiving no credit (adj. factor = 0). We adopted this adjustment factor in response to commenters who were concerned that a vehicle which incorporated many efficiency-improving technologies may not achieve the full 30% improvement, and as a result would receive no credit (thus discouraging them from using any of the technologies). Because there is environmental benefit (reduced CO₂) from the use of even some of these efficiency-improving technologies, EPA believes it is appropriate to scale the A/C efficiency credits to account for these partial improvements.

EPA is adopting the A/C CO₂ Idle Test procedure as proposed in most respects. This laboratory idle test is performed while the vehicle is at idle, similar to the idle carbon monoxide (CO) test that was once a part of EPA vehicle certification. The test determines the additional CO₂ generated at idle when the A/C system is operated. The A/C CO₂ Idle Test will be run with and without the A/C system cooling the interior cabin while the vehicle's engine is operating at idle and with the system under complete control of the engine and climate control system. The test includes tighter restrictions on test cell temperatures and humidity levels than apply for the basic FTP test procedure in order to more closely control the loads from operation of the A/C system. EPA is also adopting additional refinements to the required in-vehicle blower fan settings for manually controlled systems to more closely represent “real world” usage patterns.

Many commenters questioned the ability of this test to measure the improved efficiency of certain A/C technologies, and stated that the test was not representative of real-world driving conditions. However, although EPA acknowledges that this test directly simulates a relatively limited range of technologies and conditions, we determined that it is sufficiently robust for the purpose of demonstrating that the system design changes are indeed implemented properly and are resulting in improved efficiency of a vehicle's A/C system, at idle as well as under a range of operating conditions. Further details of the A/C Idle Test can be found in the RIA and the regulations, as well as in the Response to Comments Document.

The design of the A/C CO₂ Idle Test represents a balancing of the need for performance tests whenever possible to ensure the most accurate quantification of efficiency improvements, with practical concerns for testing burden and facility requirements. EPA believes that the Idle Test adds to the robust quantification of A/C credits that will result in real-world efficiency improvements and reductions in A/C-related CO₂ emissions. The Idle Test will not be required in order to generate A/C Efficiency Credits until MY 2014 to allow sufficient time for manufacturers to make the necessary facilities improvements and to gain experience with the test.

EPA also considered and invited comment on a more comprehensive testing approach to quantifying A/C CO₂ emissions that could be somewhat more technically robust, but would require more test time and test facility improvements for many manufacturers. EPA invited comment on using an adapted version of the SCO3, an existing test procedure that is part of the Supplemental Federal Test Procedure. EPA discussed and invited comment on the various benefits and concerns associated with using an adapted SCO3 test. There were many comments opposed to this proposal, and very few supporters. Most of the comments opposing this approach echoed the concerns made by in the NPRM. These included excessive testing burden, limited test facilities and the cost of adding new ones, and the concern that the SC03 test may not be sufficiently representative of in use A/C usage. Some commenters supported a derivative of the SCO3 test or multiple runs of other urban cycles (such as the LA-4) for quantifying A/C system efficiency. While EPA considers a test cycle that covers a broader range of vehicle speed and climatic conditions to be ideal, developing such a representative A/C test would involve the work of many stakeholders, and would require a significant amount of time, exceeding the scope of this rule. EPA expects to continue working with industry, the California Air Resources Board, and other stakeholders to move toward increasingly robust performance tests and methods for determining the efficiency of mobile A/C systems and the related impact on vehicle CO₂ emissions, including a potential adapted SC03 test.

c. Interaction With Title VI Refrigerant Regulations

Title VI of the Clean Air Act deals with the protection of stratospheric ozone. Section 608 establishes a comprehensive program to limit emissions of certain ozone-depleting substances (ODS). The rules promulgated under section 608 regulate the use and disposal of such substances during the service, repair or disposal of appliances and industrial process refrigeration. In addition, section 608 and the regulations promulgated under it, prohibit knowingly venting or releasing ODS during the course of maintaining, servicing, repairing or disposing of an appliance or industrial process refrigeration equipment. Section 609 governs the servicing of motor vehicle A/C systems. The regulations promulgated under section 609 (40 CFR part 82, subpart B) establish standards and requirements regarding the servicing of A/C systems. These regulations include establishing standards for equipment that recovers and recycles (or, for refrigerant blends, only recovers) refrigerant from A/C systems; requiring technician training and certification by an EPA-approved organization; establishing recordkeeping requirements; imposing sales restrictions; and prohibiting the venting of refrigerants. Section 612 requires EPA to review substitutes for class I and class II ozone depleting substances and to consider whether such substitutes will cause an adverse effect to human health or the environment as compared with other substitutes that are currently or potentially available. EPA promulgated regulations for this program in 1992 and those regulations are located at 40 CFR part 82, subpart G. When reviewing substitutes, in addition to finding them acceptable or unacceptable, EPA may also find them acceptable so long as the user meets certain use conditions. For example, all motor vehicle air conditioning systems must have unique fittings and a uniquely colored label for the refrigerant being used in the system.

On September 14, 2006, EPA proposed to approve R-744 (CO₂) for use in motor vehicle A/C systems (71 FR 55140) and on October 19, 2009, EPA proposed to approve the low-GWP refrigerant HFO-1234yf for these systems (74 FR 53445), both subject to certain requirements. Final action on both of these proposals is expected later this year. EPA previously issued a final rule allowing the use of HFC-152a as a refrigerant in motor vehicle A/C systems subject to certain requirements (June 12, 2008; 73 FR 33304). As discussed above, manufacturers transitioning to any of the approved refrigerants would be eligible for A/C Leakage Credits, the value of which would depend on the GWP of their refrigerant and the degree of leakage reduction of their systems.

EPA views this rule as complementing these Title VI programs, and not conflicting with them. To the extent that manufacturers choose to reduce refrigerant leakage in order to earn A/C Leakage Credits, this will dovetail with the Title VI section 609 standards which apply to maintenance events, and to end-of-vehicle life disposal. In fact, as noted, a benefit of the A/C credit provisions is that there should be fewer and less impactive maintenance events for MVACs, since there will be less leakage. In addition, the credit provisions will not conflict (or overlap) with the Title VI section 609 standards. EPA also believes the menu of leak control technologies described in this rule will complement the section 612 requirements, because these control technologies will help ensure that HFC-134a (or other refrigerants) will be used in a manner that further minimizes potential adverse effects on human health and the environment.

2. Flexible Fuel and Alternative Fuel Vehicle Credits

EPA is finalizing its proposal to allow flexible-fuel vehicles (FFVs) and alternative fuel vehicles to generate credits for purposes of the GHG rule starting in the 2012 model year. FFVs are vehicles that can run on both an alternative fuel and a conventional fuel. Most FFVs are E85 vehicles, which can run on a mixture of up to 85 percent ethanol and gasoline. Dedicated alternative fuel vehicles are vehicles that run exclusively on an alternative fuel (e.g., compressed natural gas). These credits are designed to complement the treatment of FFVs under CAFE, consistent with the emission reduction objectives of the CAA. As explained at proposal, EPCA includes an incentive under the CAFE program for production of dual-fueled vehicles or FFVs, and dedicated alternative fuel vehicles. For FFVs and dual-fueled vehicles, the EPCA/EISA credits have three elements: (1) The assumption that the vehicle is operated 50% of the time on the conventional fuel and 50% of the time on the alternative fuel, (2) that 1 gallon of alternative fuel is treated as 0.15 gallon of fuel, essentially increasing the fuel economy of a vehicle on alternative fuel by a factor of 6.67, and (3) a “cap” provision that limits the maximum fuel economy increase that can be applied to a manufacturer's overall CAFE compliance value for all CAFE compliance categories (i.e., domestic passenger cars, import passenger cars, and light trucks) to 1.2 mpg through 2014 and 1.0 mpg in 2015. EPCA's provisions were amended by the EISA to extend the period of availability of the FFV credits, but to begin phasing them out by annually reducing the amount of FFV credits that can be used in demonstrating compliance with the CAFE standards. EPCA does not premise the availability of the FFV credits on actual use of alternative fuel. Under EPCA, after MY 2019 no FFV credits will be available for CAFE compliance. Under EPCA, for dedicated alternative fuel vehicles, there are no limits or phase-out. As proposed, FFV and Alternative Fuel Vehicle Credits will be calculated as a part of the calculation of a manufacturer's overall fleet average fuel economy and fleet average carbon-related exhaust emissions (§ 600.510-12).

Manufacturers supported the inclusion of FFV credits in the program. Chrysler noted that the credits encourage manufacturers to continue production of vehicles capable of running on alternative fuels as the production and distribution systems of such fuels are developed. Chrysler believes the lower carbon intensity of such fuels is an opportunity for further greenhouse gas reductions and increased energy independence, and the continuance of such incentives recognizes the important potential of this technology to reduce GHGs. Toyota noted that because actions taken by manufacturers to comply with EPA's regulation will, to a large extent, be the same as those taken to comply with NHTSA's CAFE regulation, it is appropriate for EPA to consider flexibilities contained in the CAFE program that clearly impact product plans and technology deployment plans already in place or nearly in place. Toyota believes that adopting the FFV credit for a transitional period of time appears to recognize this reality, while providing a pathway to eventually phase-out the flexibility.

As proposed, electric vehicles (EVs) or plug-in hybrid electric vehicles (PHEVs) are not eligible to generate this type of credit. These vehicles are covered by the advanced technology vehicle incentives provisions described in Section III.C.3, so including them here would lead to a double counting of credits.

a. Model Year 2012-2015 Credits

i. FFVs

For the GHG program, EPA is allowing FFV credits corresponding to the amounts allowed by the amended EPCA but only during the period from MYs 2012 to 2015. (As discussed below in Section III.E., EPA is not allowing CAFE-based FFV credits to be generated as part of the early credits program.) As noted at proposal, several manufacturers have already taken the availability of FFV credits into account in their near-term future planning for CAFE and this reliance indicates that these credits need to be considered in assessing necessary lead time for the CO₂ standards. Manufacturers commented that the credits are necessary in allowing them to transition to the new standards. EPA thus believes that allowing these credits, in the near term, would help provide adequate lead time for manufacturers to implement the new multi-year standards, but that for the longer term there is adequate lead time without the use of such credits. This will also tend to harmonize the GHG and the CAFE program during these interim years. As discussed below, EPA is requiring for MY 2016 and later that manufacturers will need to reliably estimate the extent to which the alternative fuel is actually being used by vehicles in order to count the alternative fuel use in the vehicle's CO₂ emissions level determination. Beginning in MY 2016, the FFV credits as described above for MY 2012-2015 will no longer be available for EPA's GHG program. Rather, GHG compliance values will be based on actual emissions performance of the FFV on conventional and alternative fuels, weighted by the actual use of these fuels in the FFVs.

As with the CAFE program, EPA will base MY 2012-2015 credits on the assumption that the vehicles would operate 50% of the time on the alternative fuel and 50% of the time on conventional fuel, resulting in CO₂ emissions that are based on an arithmetic average of alternative fuel and conventional fuel CO₂ emissions. In addition, the measured CO₂ emissions on the alternative fuel will be multiplied by a 0.15 volumetric conversion factor which is included in the CAFE calculation as provided by EPCA. Through this mechanism a gallon of alternative fuel is deemed to contain 0.15 gallons of fuel. For example, for a flexible-fuel vehicle that emitted 330 g/mi CO₂ operating on E85 and 350 g/mi CO₂ operating on gasoline, the resulting CO₂ level to be used in the manufacturer's fleet average calculation would be:

Image Not Available

EPA understands that by using the CAFE approach—including the 0.15 factor—the CO₂ emissions value for the vehicle is calculated to be significantly lower than it actually would be otherwise, even if the vehicle were assumed to operate on the alternative fuel at all times. This represents a “credit” being provided to FFVs.

EPA notes also that the above equation and example are based on an FFV that is an E85 vehicle. EPCA, as amended by EISA, also establishes the use of this approach, including the 0.15 factor, for all alternative fuels, not just E85. The 0.15 factor is used for B-20 (20 percent biofuel and 80 percent diesel) FFVs. EPCA also establishes this approach, including the 0.15 factor, for gaseous-fueled dual-fueled vehicles, such as a vehicle able to operate on gasoline and CNG. (For natural gas dual-fueled vehicles, EPCA establishes a factor of 0.823 gallons of fuel for every 100 cubic feet a natural gas used to calculate a gallons equivalent. ) The EISA's use of the 0.15 factor in this way provides a similar regulatory treatment across the various types of alternative fuel vehicles. EPA also will use the 0.15 factor for all FFVs in order not to disrupt manufacturers' near-term compliance planning and assure sufficient lead time. EPA, in any case, expects the vast majority of FFVs to be E85 vehicles, as is the case today.

The FFV credit limits for CAFE are 1.2 mpg for model years 2012-2014 and 1.0 mpg for model year 2015. In CO₂ terms, these CAFE limits translate to declining CO₂ credit limits over the four model years, as the CAFE standards increase in stringency. As the CAFE standard increases numerically, the limit becomes a smaller fraction of the standard. EPA proposed, but is not adopting, credit limits based on the overall industry average CO₂ standards for cars and trucks. EPA also requested comments on basing the calculated CO₂ credit limits on the individual manufacturer fleet-average standards calculated from the footprint curves. EPA received comment from one manufacturer supporting this approach. EPA also received comments from another manufacturer recommending that the credit limits for an individual manufacturer be based instead on that manufacturer's fleet average performance. The commenter noted that this approach is in line with how CAFE FFV credit limits are applied. This is due to the fact that the GHG-equivalent of the CAFE 1.2 mpg cap will vary due to the non-linear relationship between fuel economy and GHGs/fuel consumption. EPA agrees with this approach since it best harmonizes how credit limits are determined in CAFE. EPA intended and continues to believe it is appropriate to provide essentially the same FFV credits under both programs for MYs 2012-2015. Therefore, EPA is finalizing FFV credits limits for MY 2012-2015 based on a manufacturer's fleet-average performance. For example, if a manufacturer's 2012 car fleet average emissions performance was 260 g/mile (34.2 mpg), the credit limit in CO₂ terms would be 9.5 g/mile (34.2 mpg − 1.2 mpg = 33.0 mpg = 269.5 g/mile) and if it were 270 g/mile the limit would be 10.2 g/mile.

ii. Dedicated Alternative Fuel Vehicles

As proposed, EPA will calculate CO₂ emissions from dedicated alternative fuel vehicles for MY 2012-2015 by measuring the CO₂ emissions over the test procedure and multiplying the results by the 0.15 conversion factor described above. For example, for a dedicated alternative fuel vehicle that would achieve 330 g/mi CO₂ while operating on alcohol (ethanol or methanol), the effective CO₂ emissions of the vehicle for use in determining the vehicle's CO₂ emissions would be calculated as follows:

CO₂ = 330 × 0.15 = 49.5 g/mi

b. Model Years 2016 and Later

i. FFVs

EPA is treating FFV credits the same as under EPCA for model years 2012-2015, but is applying a different approach starting with model year 2016. EPA recognizes that under EPCA automatic FFV credits are entirely phased out of the CAFE program by MY 2020, and apply in the prior model years with certain limitations, but without a requirement that the manufacturers demonstrate actual use of the alternative fuel. Unlike EPCA, CAA section 202(a) does not mandate that EPA treat FFVs in a specific way. Instead EPA is required to exercise its own judgment and determine an appropriate approach that best promotes the goals of this CAA section. Under these circumstances, EPA will treat FFVs for model years 2012-2015 the same as under EPCA, as part of providing sufficient lead time given manufacturers' compliance strategies which rely on the existence of these EPCA statutory credits, as explained above.

Starting with model year 2016, as proposed, EPA will no longer allow manufacturers to base FFV emissions on the use of the 0.15 factor credit described above, and on the use of an assumed 50% usage of alternative fuel. Instead, EPA believes the appropriate approach is to ensure that FFV emissions are based on demonstrated emissions performance. This will promote the environmental goals of the final program. EPA received several comments in support of EPA's proposal to use this approach instead of the EPCA approach for MY 2016 and later. Under the EPA program in MY 2016 and later, manufacturers will be allowed to base an FFV's emissions compliance value in part on the vehicle test values run on the alternative fuel, for that portion of its fleet for which the manufacturer demonstrates utilized the alternative fuel in the field. In other words, the default is to assume FFVs operate on 100% gasoline, and the emissions value for the FFV vehicle will be based on the vehicle's tested value on gasoline. However, if a manufacturer can demonstrate that a portion of its FFVs are using an alternative fuel in use, then the FFV emissions compliance value can be calculated based on the vehicle's tested value using the alternative fuel, prorated based on the percentage of the fleet using the alternative fuel in the field. An example calculation is described below. EPA believes this approach will provide an actual incentive to ensure that such fuels are used. The incentive arises since actual use of the flexible fuel typically results in lower tailpipe GHG emissions than use of gasoline and hence improves the vehicles' performance, making it more likely that its performance will improve a manufacturers' average fleetwide performance. Based on existing certification data, E85 FFV CO₂ emissions are typically about 5 percent lower on E85 than CO₂ emissions on 100 percent gasoline. Moreover, currently there is little incentive to optimize CO₂ performance for vehicles when running on E85. EPA believes the above approach would provide such an incentive to manufacturers and that E85 vehicles could be optimized through engine redesign and calibration to provide additional CO₂ reductions.

Under the EPCA credit provisions, there is an incentive to produce FFVs but no actual incentive to ensure that the alternative fuels are used, or that actual vehicle fuel economy improves. GHG and energy security benefits are only achieved if the alternative fuel is actually used and (for GHGs) that performance improves, and EPA's approach for MY 2016 and beyond will now provide such an incentive. This approach will promote greater use of alternative fuels, as compared to a situation where there is a credit but no usage requirement. This is also consistent with the agency's overall commitment to the expanded use of renewable fuels. Therefore, EPA is basing the FFV program for MYs 2016 and thereafter on real-world reductions: i.e., actual vehicle CO₂ emissions levels based on actual use of the two fuels, without the 0.15 conversion factor specified under EISA.

For 2016 and later model years, EPA will therefore treat FFVs similarly to conventional fueled vehicles in that FFV emissions would be based on actual CO₂ results from emission testing on the fuels on which it operates. In calculating the emissions performance of an FFV, manufacturers may base FFV emissions on vehicle testing based on the alternative fuel emissions, if they can demonstrate that the alternative fuel is actually being used in the vehicles. Performance will otherwise be calculated assuming use only of conventional fuel. The manufacturer must establish the ratio of operation that is on the alternative fuel compared to the conventional fuel. The ratio will be used to weight the CO₂ emissions performance over the 2-cycle test on the two fuels. The 0.15 conversion factor will no longer be included in the CO₂ emissions calculation. For example, for a flexible-fuel vehicle that emitted 300 g/mi CO₂ operating on E85 ten percent of the time and 350 g/mi CO₂ operating on gasoline ninety percent of the time, the CO₂ emissions for the vehicles to be used in the manufacturer's fleet average would be calculated as follows:

CO₂ = (300 × 0.10) + (350 × 0.90) = 345 g/mi

The most complex part of this approach is to establish what data are needed for a manufacturer to accurately demonstrate use of the alternative fuel, where the manufacturer intends for its performance to be calculated based on some use of alternative fuels. One option EPA is finalizing is establishing a rebuttable presumption using a national average approach based on national E85 fuel use. Manufacturers could use this value along with their vehicle emissions results demonstrating lower emissions on E85 to determine the emissions compliance values for FFVs sold by manufacturers under this program. For example, national E85 volumes and national FFV sales may be used to prorate E85 use by manufacturer sales volumes and FFVs already in-use. Upon a manufacturer's written request, EPA will conduct an analysis of vehicle miles travelled (VMT) by year for all FFVs using its emissions inventory MOVES model. Using the VMT ratios and the overall E85 sales, E85 usage will be assigned to each vehicle. This method accounts for the VMT of new FFVs and FFVs already in the existing fleet using VMT data in the model. The model will then be used to determine the ratio of E85 and gasoline for new vehicles being sold. Fluctuations in E85 sales and FFV sales will be taken into account to adjust the manufacturers' E85 actual use estimates annually. EPA plans to make this assigned fuel usage factor available through guidance prior to the start of MY 2016 and adjust it annually as necessary. EPA believes this is a reasonable way to apportion E85 use across the fleet.

If manufacturers decide not to use EPA's assigned fuel usage based on the national average analysis, they have a second option of presenting their own data for consideration as the basis for evaluating fuel usage. Manufacturers have suggested demonstrations using vehicle on-board data gathering through the use of on-board sensors and computers. California's program allows FFV credits based on FFV use and envisioned manufacturers collecting fuel use data from vehicles in fleets with on-site refueling. Manufacturers must present a statistical analysis of alternative fuel usage data collected on actual vehicle operation. EPA is not attempting to specify how the data is collected or the amount of data needed. However, the analysis must be based on sound statistical methodology. Uncertainty in the analysis must be accounted for in a way that provides reasonable certainty that the program does not result in loss of emissions reductions.

EPA received comments that the 2016 and later FFV emissions performance methodology should be based on the life cycle emissions (i.e., including the upstream GHG emissions associated with fuel feedstocks, production, and transportation) associated with the use of the alternative fuel. Commenters are concerned that the use of ethanol will not result in lower GHGs on a lifecycle basis. After considering these comments, EPA is not including lifecycle emissions in the calculation of vehicle credits. EPA continues to believe that it is appropriate to base credits for MY 2012-2015 on the EPCA/CAFE credits and to base compliance values for MY 2016 on the demonstrated tailpipe emissions performance on gasoline and E85, and is finalizing this approach as proposed. EPA recently finalized its RFS2 rulemaking which addresses lifecycle emissions from ethanol and the upstream GHG benefits of E85 use are already captured by this program.

ii. Dedicated Alternative Fuel Vehicles

As proposed, for model years 2016 and later dedicated alternative fuel vehicles, CO₂ will be measured over the 2-cycle test in order to be included in a manufacturer's fleet average CO₂ calculations. As noted above, this is different than CAFE methodology which provides a methodology for calculating a petroleum-based mpg equivalent for alternative fuel vehicles so they can be included in CAFE. However, because CO₂ can be measured directly from alternative fuel vehicles over the test procedure, EPA believes this is the simplest and best approach since it is consistent with all other vehicle testing under the CO₂ program. EPA did not receive comments on this approach.

3. Advanced Technology Vehicle Incentives for Electric Vehicles, Plug-in Hybrids, and Fuel Cell Vehicles

EPA is finalizing provisions that provide a temporary regulatory incentive for the commercialization of certain advanced vehicle power trains—electric vehicles (EVs), plug-in hybrid electric vehicles (PHEVs), and fuel cell vehicles (FCVs)—for model year 2012-2016 light-duty and medium-duty passenger vehicles. The purpose of these provisions is to provide a temporary incentive to promote technologies which have the potential to produce very large GHG reductions in the future, but which face major challenges such as vehicle cost, consumer acceptance, and the development of low-GHG fuel production infrastructure. The tailpipe GHG emissions from EVs, PHEVs operated on grid electricity, and hydrogen-fueled FCVs are zero, and traditionally the emissions of the vehicle itself are all that EPA takes into account for purposes of compliance with standards set under section 202(a). Focusing on vehicle tailpipe emissions has not raised any issues for criteria pollutants, as upstream emissions associated with production and distribution of the fuel are addressed by comprehensive regulatory programs focused on the upstream sources of those emissions. At this time, however, there is no such comprehensive program addressing upstream emissions of GHGs, and the upstream GHG emissions associated with production and distribution of electricity are higher than the corresponding upstream GHG emissions of gasoline or other petroleum based fuels. In the future, if there were a program to comprehensively control upstream GHG emissions, then the zero tailpipe levels from these vehicles have the potential to produce very large GHG reductions, and to transform the transportation sector's contribution to nationwide GHG emissions.

This temporary incentive program applies only for the model years 2012-2016 covered by this final rule. EPA will reassess the issue of how to address EVs, PHEVs, and FCVs in rulemakings for model years 2017 and beyond, based on the status of advanced technology vehicle commercialization, the status of upstream GHG emissions control programs, and other relevant factors.

In the Joint Notice of Intent, EPA stated that “EPA is currently considering proposing additional credit opportunities to encourage the commercialization of advanced GHG/fuel economy control technology such as electric vehicles and plug-in hybrid electric vehicles. These `super credits' could take the form of a multiplier that would be applied to the number of vehicles sold such that they would count as more than one vehicle in the manufacturer's fleet average.” Following through, EPA proposed two mechanisms by which these vehicles would earn credits: (1) A zero grams/mile compliance value for EVs, FCVs, and for PHEVs when operated on grid electricity, and (2) a vehicle multiplier in the range of 1.2 to 2.0.

The zero grams/mile compliance value for EVs (and for PHEVs when operated on grid electricity, as well as for FCVs which involve similar upstream GHG issues with respect to hydrogen production) is an incentive that operates like a credit because, while it accurately accounts for tailpipe GHG emissions, it does not reflect the increase in upstream GHG emissions associated with the electricity used by EVs compared to the upstream GHG emissions associated with the gasoline or diesel fuel used by conventional vehicles. For example, based on GHG emissions from today's national average electricity generation (including GHG emissions associated with feedstock extraction, processing, and transportation) and other key assumptions related to vehicle electricity consumption, vehicle charging losses, and grid transmission losses, a midsize EV might have an upstream GHG emissions of about 180 grams/mile, compared to the upstream GHG emissions of a typical midsize gasoline car of about 60 grams/mile. Thus, the EV would cause a net upstream GHG emissions increase of about 120 grams/mile (in general, the net upstream GHG increase would be less for a smaller EV and more for a larger EV). The zero grams/mile compliance value provides an incentive because it is less than the 120 grams/mile value that would fully account for the net increase in GHG emissions, counting upstream emissions. The net upstream GHG impact could change over time, of course, based on changes in electricity generation or gasoline production.

The proposed vehicle multiplier incentive would also have operated like a credit as it would have allowed an EV, PHEV, or FCV to count as more than one vehicle in the manufacturer's fleet average. For example, combining a multiplier of 2.0 with a zero grams/mile compliance value for an EV would allow that EV to be counted as two vehicles, each with a zero grams/mile compliance value, in the manufacturer's fleet average calculations. In effect, a multiplier of 2.0 would double the overall credit associated with an EV, PHEV, or FCV.

EPA explained in the proposal that the potential for large future emissions benefits from these technologies provides a strong reason for providing incentives at this time to promote their commercialization in the 2012-2016 model years. At the same time, EPA acknowledged that the zero grams/mile compliance value did not account for increased upstream GHG emissions. EPA requested comment on providing some type of incentive, the appropriateness of both the zero grams/mile and vehicle multiplier incentive mechanisms, and on any alternative approaches for addressing advanced technology vehicle incentives. EPA received many comments on these issues, which will be briefly summarized below.

Although some environmental organizations and State agencies supported the principle of including some type of regulatory incentive mechanism, almost all of their comments were opposed to the combination of both the zero grams/mile compliance value and multipliers in the higher end of the proposed range of 1.2 to 2.0. The California Air Resources Board stated that the proposed credits “are excessive” and the Union of Concerned Scientists stated that it “strongly objects” to the approach that lacks “technical justification” by not “accounting for upstream emissions.” The Natural Resources Defense Council (NRDC) stated that the credits could “undermine the emissions benefits of the program and will have the unintended consequence of slowing the development of conventional cleaner vehicle emission reduction technologies into the fleet.” NRDC, along with several other commenters who made the same point, cited an example based on Nissan's public statements that it plans on producing up to 150,000 Nissan Leaf EVs in the near future at its plant in Smyrna, Tennessee. NRDC's analysis showed that if EVs were to account for 10% of Nissan's car fleet in 2016, the combination of the zero grams/mile and 2.0 multiplier would allow Nissan to make only relatively small improvements to its gasoline car fleet and still be in compliance. NRDC described a detailed methodology for calculating “true full fuel cycle emissions impacts” for EVs. The Sierra Club suggested that the zero grams/mile credit would “taint” EVs as the public comes to understand that these vehicles are not zero-GHG vehicles, and that the zero grams/mile incentive would allow higher gasoline vehicle GHG emissions.

Most vehicle manufacturers were supportive of both the zero grams/mile compliance value and a higher vehicle multiplier. The Alliance of Automobile Manufacturers supported zero grams/mile “since customers need to receive a clear signal that they have made the right choice by preferring an EV, PHEV, or EREV. * * * However, the Alliance recognizes the need for a comprehensive approach with shared responsibility in order to achieve an overall carbon reduction.” Nissan claimed that zero grams/mile is “legally required,” stating that EPA's 2-cycle test procedures do not account for upstream GHG emissions, that accounting for upstream emissions from electric vehicles but not from other vehicles would be arbitrary, and that including upstream GHG would “disrupt the careful balancing embedded into the National Program.” Several other manufacturers, including Ford, Chrysler, Toyota, and Mitsubishi, also supported the proposed zero grams/mile compliance value. BMW suggested a compliance value approach similar to that used for CAFE compliance (described below), which would yield a very low, non-zero grams/mile compliance value. Honda opposed the zero grams/mile incentive. Honda suggested that EPA should fully account for upstream GHG and “should separate incentives and credits from the measurement of emissions.” Automakers universally supported higher multipliers, many higher than the maximum 2.0 level proposed by EPA. Honda suggested a multiplier of 16.0 for FCVs. Mitsubishi supported the concept of larger, temporary incentives until advanced technology vehicle sales achieved a 10% market share. Finally, some commenters suggested that other technologies should also receive incentives, such as diesel vehicles, hydrogen-fueled internal combustion engines, and natural gas vehicles.

Based on a careful consideration of these comments, EPA is modifying its proposed advanced technology vehicle incentive program for EVs, PHEVs, and FCVs produced in 2012-2016. EPA is not extending the program to include additional technologies at this time. The final incentive program, and our rationale for it, are described below.

One, the incentive program retains the zero grams/mile value for EVs and FCVs, and for PHEVs when operated on grid electricity, subject to vehicle production caps discussed below. EPA acknowledges that, based on current electricity and hydrogen production processes, that EVs, PHEVs, and FCVs yield higher upstream GHG emissions than comparable gasoline vehicles. But EPA reiterates its support for temporarily rewarding advanced emissions control technologies by foregoing modest emissions reductions in the short term in order to lay the foundation for the potential for much larger emission reductions in the longer term. EPA notes that EVs, PHEVs, and FCVs are potential GHG “game changers” if major cost and consumer barriers can be overcome and if there is a nationwide transformation to low-GHG electricity (or hydrogen, in the case of FCVs).

Although EVs and FCVs will have compliance values of zero grams/mile, PHEV compliance values will be determined by combining zero grams/mile for grid electricity operation with the GHG emissions from the 2-cycle test results during operation on liquid fuel, and weighting these values by the percentage of miles traveled that EPA believes will be performed on grid electricity and on liquid fuel, which will vary for different PHEVs. EPA is currently considering different approaches for determining the weighting factor to be used in calculating PHEV GHG emissions compliance values. EPA will consider the work of the Society of Automotive Engineers Hybrid Technical Standards Committee, as well as other relevant factors. EPA will issue a final rule on this methodology by the fall of 2010, when EPA expects some PHEVs to initially enter the market.

EPA agrees with the comments by the environmental organizations, States, and Honda that the zero grams/mile compliance value will reduce the overall GHG benefits of the program. However, EPA believes these reductions in GHG benefits will be relatively small based on the projected production of EVs, PHEVs, and FCVs during the 2012-2016 timeframe, along with the other changes that we are making in the incentive program. EPA believes this modest potential for reduction in near-term emissions control is more than offset by the potential for very large future emissions reductions that commercialization of these technologies could promote.

Two, the incentive program will not include any vehicle multipliers, i.e., an EV's zero grams/mile compliance value will count as one vehicle in a manufacturer's fleet average, not as more than one vehicle as proposed. EPA has concluded that the combination of the zero grams/mile and multiplier credits would be excessive. Compared to the maximum multiplier of 2.0 that EPA had proposed, dropping this multiplier reduces the aggregate impact of the overall credit program by a factor of two (less so for lower multipliers, of course).

Three, EPA is placing a cumulative cap on the total production of EVs, PHEVs, and FCVs for which an individual manufacturer can claim the zero grams/mile compliance value during model years 2012-2016. The cumulative production cap will be 200,000 vehicles, except those manufacturers that sell at least 25,000 EVs, PHEVs, and FCVs in MY 2012 will have a cap of 300,000 vehicles for MY 2012-2016. This higher cap option is an additional incentive for those manufacturers that take an early leadership role in aggressively and successfully marketing advanced technology vehicles. These caps are a second way to limit the potential GHG benefit losses associated with the incentive program and therefore are another response to the concerns that the proposed incentives were excessive and could significantly undermine the program's GHG benefits. If, for example, 500,000 EVs were produced in 2012-2016 that qualified for the zero grams/mile compliance value, the loss in GHG benefits due to this program would be about 25 million metric tons, or less than 3 percent of the total projected GHG benefits of this program. The rationale for these caps is that the incentive for EVs, PHEVs, and FCVs is most critical when individual automakers are beginning to introduce advanced technologies in the market, and less critical once individual automakers have successfully achieved a reasonable market share and technology costs decline due to higher production volumes and experience. EPA believes that cap levels of 200,000-300,000 vehicles over a five model year period are reasonable, as production greater than this would indicate that the manufacturer has overcome at least some of the initial market barriers to these advanced technologies. Further, EPA believes that it is unlikely that many manufacturers will approach these cap levels in the 2012-2016 timeframe.

Production beyond the cumulative vehicle production cap for a given manufacturer in MY 2012-2016 would have its compliance values calculated according to a methodology that accounts in full for the net increase in upstream GHG emissions. For an EV, for example, this would involve: (1) Measuring the vehicle electricity consumption in watt-hours/mile over the 2-cycle test (in the example introduced earlier, a midsize EV might have a 2-cycle test electricity consumption of 230 watt-hours/mile), (2) adjusting this watt-hours/mile value upward to account for electricity losses during transmission and vehicle charging (dividing 230 watt-hours/mile by 0.93 to account for grid/transmission losses and by 0.90 to reflect losses during vehicle charging yields a value of 275 watt-hours/mile), (3) multiplying the adjusted watt-hours/mile value by a nationwide average electricity upstream GHG emissions rate of 0.642 grams/watt-hour at the powerplant (275 watt-hours/mile multiplied by 0.642 grams GHG/watt-hour yields 177 grams/mile), and 4) subtracting the upstream GHG emissions of a comparable midsize gasoline vehicle of 56 grams/mile to reflect a true net increase in upstream GHG emissions (177 grams/mile for the EV minus 56 grams/mile for the gasoline vehicle yields a net increase and EV compliance value of 121 grams/mile). The full accounting methodology for the portion of PHEV operation on grid electricity would use this same approach.

EPA projects that the aggregate impact of the incentive program on advanced technology vehicle GHG compliance values will be similar to the way advanced technologies are treated under DOT's CAFE program. In the CAFE program, the mpg value for an EV is determined using a “petroleum equivalency factor” that has a 1/0.15 factor built into it similar to the flexible fuel vehicle credit. For example, under current regulations, an EV with a 2-cycle electricity consumption of 230 watt-hours/mile would have a CAFE rating of about 360 miles per gallon, which would be equivalent to a gasoline vehicle GHG emissions value of 25 grams/mile, which is close to EPA's zero grams/mile for EV production that is below an individual automaker's cumulative vehicle production cap. The exception would be if a manufacturer exceeded its cumulative vehicle production cap during MY 2012-2016. Then, the same EV would have a GHG compliance value of about 120 grams/mile, which would be significantly higher than the 25 gram/mile implied by the 360 mile/gallon CAFE value.

EPA disagrees with Nissan that excluding upstream GHGs is legally required under section 202(a)(1). In this rulemaking, EPA is adopting standards under section 202(a)(1), which provides EPA with broad discretion in setting emissions standards. This includes authority to structure the emissions standards in a way that provides an incentive to promote advances in emissions control technology. This discretion includes the adjustments to compliance values adopted in the final rule, the multipliers we proposed, and other kinds of incentives. EPA recognizes that we have not previously made adjustments to a compliance value to account for upstream emissions in a section 202(a) vehicle emissions standard, but that does not mean we do not have authority to do so in this case. In addition, EPA is not directly regulating upstream GHG emissions from stationary sources, but instead is deciding how much value to assign to a motor vehicle for purposes of compliance calculations with the motor vehicle standard. While the logical place to start is the emissions level measured under the test procedure, section 202(a)(1) does not require that EPA limit itself to only that level. For vehicles above the production volume cap described above, EPA will adjust the measured value to a level that reflects the net difference in upstream GHG emissions compared to a comparable conventional vehicle. This will account for the actual GHG emissions increase associated with the use of the EV. As shown above, upstream GHG emissions attributable to increased electricity production to operate EVs or PHEVs currently exceed the upstream GHG emissions attributable to gasoline vehicles. There is a rational basis for EPA to account for this net difference, as that best reflects the real world effect on the air pollution problem we are addressing. For vehicles above the cap, EPA is reasonably and fairly accounting for the incremental increase in upstream GHG emissions from both the electric vehicles and the conventional vehicles. EPA is not, as Nissan suggested, arbitrarily counting upstream emissions for electric vehicles but not for conventional fuel vehicles.

EPA recognizes that every motor vehicle fuel and fuel production process has unique upstream GHG emissions impacts. EPA has discretion in this rulemaking under section 202(a) on whether to account for differences in net upstream GHG emissions relative to gasoline produced from oil, and intends to only consider upstream GHG emissions for those fuels that have significantly higher or lower GHG emissions impacts. At this time, EPA is only making such a determination for electricity, given that, as shown above in the example for a midsize car, electricity upstream GHG emissions are about three times higher than gasoline upstream GHG emissions. For example, the difference in upstream GHG emissions for both diesel fuel from oil and CNG from natural gas are relatively small compared to differences associated with electricity. Nor is EPA arbitrarily ignoring upstream GHG emissions of flexible fuel vehicles (FFVs) that can operate on E85. Data show that, on average, FFVs operate on gasoline over 99 percent of the time, and on E85 fuel less than 1 percent of the time. EPA's recently promulgated Renewable Fuel Standard Program shows that, with respect to aggregate lifecycle emissions including non-tailpipe GHG emissions (such as feedstock growth, transportation, fuel production, and land use), lifecycle emissions for ethanol from corn using advanced production technologies are about 20 percent less GHG than gasoline from oil. Given this difference, and that E85 is used in FFVs less than 1 percent of the time, EPA has concluded that it is not necessary to adopt a more complicated upstream accounting for FFVs. Accordingly, EPA's incentive approach here is both reasonable and authorized under section 202(a)(1).

In summary, EPA believes that this program for MY 2012-2016 strikes a reasoned balance by providing a temporary regulatory incentive to help promote commercialization of advanced vehicle technologies which are potential game-changers, but which also face major barriers, while effectively minimizing potential GHG losses by dropping the proposed multiplier and adding individual automaker production volume caps. In the future, if there were a program to control utility GHG emissions, then these advanced technology vehicles have the potential to produce very large reductions in GHG emissions, and to transform the transportation sector's contribution to nationwide GHG emissions. EPA will reassess the issue of how to address EVs, PHEVs, and FCVs in rulemakings for model years 2017 and beyond based on the status of advanced vehicle technology commercialization, the status of upstream GHG control programs, and other relevant factors.

Finally, the criteria and definitions for what vehicles qualify for the advanced technology vehicle incentives are provided in Section III.E. These definitions for EVs, PHEVs, and FCVs ensure that only credible advanced technology vehicles are provided the incentives.

4. Off-Cycle Technology Credits

As proposed, EPA is adopting an optional credit opportunity intended to apply to new and innovative technologies that reduce vehicle CO₂ emissions, but for which the CO₂ reduction benefits are not significantly captured over the 2-cycle test procedure used to determine compliance with the fleet average standards (i.e., “off-cycle”). Eligible innovative technologies are those that are relatively newly introduced in one or more vehicle models, but that are not yet implemented in widespread use in the light-duty fleet. EPA will not approve credits for technologies that are not innovative or do not provide novel approaches to reducing greenhouse gas emissions. Manufacturers must obtain EPA approval for new and innovative technologies at the time of vehicle certification in order to earn credits for these technologies at the end of the model year. This approval must include the testing methodology to be used for quantifying credits. Further, any credits for these off-cycle technologies must be based on real-world GHG reductions not significantly captured on the current 2-cycle tests and verifiable test methods, and represent average U.S. driving conditions.

Similar to the technologies used to reduce A/C system indirect CO₂ emissions by increasing A/C efficiency, eligible technologies would not be primarily active during the 2-cycle test and therefore the associated improvements in CO₂ emissions would not be significantly captured. Because these technologies are not nearly so well developed and understood, EPA is not prepared to consider them in assessing the stringency of the CO₂ standards. However, EPA is aware of some emerging and innovative technologies and concepts in various stages of development with CO₂ reduction potential that might not be adequately captured on the FTP or HFET. EPA believes that manufacturers should be able to generate credit for the emission reductions these technologies actually achieve, assuming these reductions can be adequately demonstrated and verified. Examples include solar panels on hybrids or electric vehicles, adaptive cruise control, and active aerodynamics. EPA believes it would be appropriate to provide an incentive to encourage the introduction of these types of technologies, that bona fide reductions from these technologies should be considered in determining a manufacturer's fleet average, and that a credit mechanism is an effective way to do this. This optional credit opportunity would be available through the 2016 model year.

EPA received comments from a few manufacturers that the “new and innovative” criteria should be broadened. The commenters pointed out that there are technologies already in the marketplace that would provide emissions reductions off-cycle and that their use should be incentivized. One manufacturer suggested that off-cycle credits should be given for start-stop technologies. EPA does not agree that this technology, which EPA's modeling projects will be widely used by manufacturers in meeting the CO₂ standards, should qualify for off-cycle credits. Start-stop technology already achieves a significant CO₂ benefit on the current 2-cycle tests, which is why many manufacturers have announced plans to adopt it across large segments of the fleet. EPA recognizes there may be additional benefits to start-stop technology beyond the 2-cycle tests (e.g., heavy idle use), and that this is likely the case for other technologies that manufacturers will rely on to meet the MY 2012-2016 standards. EPA plans to continue to assess the off-cycle potential for these technologies in the future. However, EPA does not believe that off-cycle credits should be granted for technologies which we expect manufacturers to rely on in widespread use throughout the fleet in meeting the CO₂ standards. Such credits could lead to double counting, as there is already significant CO₂ benefit over the 2-cycle tests. EPA expects that most if not all technologies that reduce CO₂ emission on the 2-cycle test will also reduce CO₂ emissions during the wide variety of in-use operation that is not directly captured in the 2-cycle test. This is no different than what occurs from the control technology on vehicles for criteria pollutants. We expect that the catalytic converter and other emission control technology will operate to reduce emissions throughout in-use driving, and not just when the vehicle is tested on the specified test procedure. The aim for this off-cycle credit provisions is to provide an incentive for technologies that normally would not be chosen as a GHG control strategy, as their GHG benefits are not measured on the specified 2-cycle test. It is not designed to provide credits for technology that does provide significant GHG benefits on the 2-cycle test and as expected will also typically provide GHG benefits in other kinds of operation. Thus, EPA is finalizing the “new and innovative” criteria as proposed. That is, the potential to earn off-cycle credits will be limited to those technologies that are new and innovative, are introduced in only a limited number of vehicle models (i.e., not in widespread use), and are not captured on the current 2-cycle tests. This approach will encourage future innovation, which may lead to the opportunity for future emissions reductions.

As proposed, manufacturers would quantify CO₂ reductions associated with the use of the innovative off-cycle technologies such that the credits could be applied on a g/mile equivalent basis, as is the case with A/C system improvements. Credits must be based on real additional reductions of CO₂ emissions and must be quantifiable and verifiable with a repeatable methodology. As proposed, the technologies upon which the credits are based would be subject to full useful life compliance provisions, as with other emissions controls. Unless the manufacturer can demonstrate that the technology would not be subject to in-use deterioration over the useful life of the vehicle, the manufacturer must account for deterioration in the estimation of the credits in order to ensure that the credits are based on real in-use emissions reductions over the life of the vehicle.

As discussed below, EPA is finalizing a two-tiered process for demonstrating the CO₂ reductions of an innovative and novel technology with benefits not captured by the FTP and HFET test procedures. First, a manufacturer must determine whether the benefit of the technology could be captured using the 5-cycle methodology currently used to determine fuel economy label values. EPA established the 5-cycle test methods to better represent real-world factors impacting fuel economy, including higher speeds and more aggressive driving, colder temperature operation, and the use of air conditioning. If this determination is affirmative, the manufacturer must follow the procedures described below (as codified in today's rules). If the manufacturer finds that the technology is such that the benefit is not adequately captured using the 5-cycle approach, then the manufacturer would have to develop a robust methodology, subject to EPA approval, to demonstrate the benefit and determine the appropriate CO₂ gram per mile credit. As discussed below, EPA is also providing opportunity for public comment as part of the approval process for such non-5-cycle credits.

a. Technology Demonstration Using EPA 5-Cycle Methodology

As noted above, the CO₂ reduction benefit of some innovative technologies could be demonstrated using the 5-cycle approach currently used for EPA's fuel economy labeling program. The 5-cycle methodology was finalized in EPA's 2006 fuel economy labeling rule, which provides a more accurate fuel economy label estimate to consumers starting with 2008 model year vehicles. In addition to the FTP and HFET test procedures, the 5-cycle approach folds in the test results from three additional test procedures to determine fuel economy. The additional test cycles include cold temperature operation, high temperature, high humidity and solar loading, and aggressive and high-speed driving; thus these tests could be used to demonstrate the benefit of a technology that reduces CO₂ over these types of driving and environmental conditions. Using the test results from these additional test cycles collectively with the 2-cycle data provides a more precise estimate of the average fuel economy and CO₂ emissions of a vehicle for both the city and highway independently. A significant benefit of using the 5-cycle methodology to measure and quantify the CO₂ reductions is that the test cycles are properly weighted for the expected average U.S. operation, meaning that the test results could be used without further adjustments.

EPA continues to believe that the use of these supplemental cycles may provide a method by which technologies not demonstrated on the baseline 2-cycles can be quantified and is finalizing this approach as proposed. The cold temperature FTP can capture new technologies that improve the CO₂ performance of vehicles during colder weather operation. These improvements may be related to warm-up of the engine or other operation during the colder temperature. An example of such a new, innovative technology is a waste heat capture device that provides heat to the cabin interior, enabling additional engine-off operation during colder weather not previously enabled due to heating and defrosting requirements. The additional engine-off time would result in additional CO₂ reductions that otherwise would not have been realized without the heat capture technology.

Although A/C credits for efficiency improvements will largely be captured in the A/C credits provisions through the credit menu of known efficiency improving components and controls, certain new technologies may be able to use the high temperatures, humidity, and solar load of the SC03 test cycle to accurately measure their impact. An example of a new technology may be a refrigerant storage device that accumulates pressurized refrigerant during driving operation or uses recovered vehicle kinetic energy during deceleration to pressurize the refrigerant. Much like the waste heat capture device used in cold weather, this device would also allow additional engine-off operation while maintaining appropriate vehicle interior occupant comfort levels. SC03 test data measuring the relative impact of innovative A/C-related technologies could be applied to the 5-cycle equation to quantify the CO₂ reductions of the technology.

The US06 cycle may be used to capture innovative technologies designed to reduce CO₂ emissions during higher speed and more aggressive acceleration conditions, but not reflected on the 2-cycle tests. An example of this is an active aerodynamic technology. This technology recognizes the benefits of reduced aerodynamic drag at higher speeds and makes changes to the vehicle at those speeds. The changes may include active front or grill air deflection devices designed to redirect frontal airflow. Certain active suspension devices designed primarily to reduce aerodynamic drag by lowering the vehicle at higher speeds may also be measured on the US06 cycle. To properly measure these technologies on the US06, the vehicle would require unique load coefficients with and without the technologies. The different load coefficient (properly weighted for the US06 cycle) could effectively result in reduced vehicle loads at the higher speeds when the technologies are active. Similar to the previously discussed cycles, the results from the US06 test with and without the technology could then use the 5-cycle methodology to quantify CO₂ reductions.

If the 5-cycle procedures can be used to demonstrate the innovative technology, then the regulatory evaluation/approval process will be relatively simple. The manufacturer will simply test vehicles with and without the technology installed or operating and compare results. All 5-cycles must be tested with the technology enabled and disabled, and the test results will be used to calculate a combined city/highway CO₂ value with the technology and without the technology. These values will then be compared to determine the amount of the credit; the combined city/highway CO₂ value with the technology operating will be subtracted from the combined city/highway CO₂ value without the technology operating to determine the gram per mile CO₂ credit. It is likely that multiple tests of each of the five test procedures will need to be performed in order to achieve the necessary strong degree of statistical significance of the credit determination results. This will have to be done for each model type for which a credit is sought, unless the manufacturer could demonstrate that the impact of the technology was independent of the vehicle configuration on which it was installed. In this case, EPA may consider allowing the test to be performed on an engine family basis or other grouping. At the end of the model year, the manufacturer will determine the number of vehicles produced subject to each credit amount and report that to EPA in the final model year report. The gram per mile credit value determined with the 5-cycle comparison testing will be multiplied by the total production of vehicles subject to that value to determine the total number of credits.

EPA received a few comments regarding the 5-cycle approach. While not commenting directly on the 5-cycle testing methodology, the Alliance raised general concerns that the proposed approach did not offer manufacturers enough certainty with regard to credit applications and testing in order to take advantage of the credits. The Alliance further commented that the proposal did not provide a level playing field to all manufacturers in terms of possible credit availability. The Alliance recommended that rather than attempting to quantify CO₂ reductions with a prescribed test procedure on unknown technologies, EPA should handle credit applications and testing guidelines via future guidance letters, as technologies emerge and are developed.

EPA believes that 5-cycle testing methodology is one clear and objective way to demonstrate certain off-cycle emissions control technologies, as discussed above. It provides certainty with regard to testing, and is available for all manufacturers. As discussed below, there are also other options for manufactures where the 5-cycle test is not appropriate. EPA is retaining this as a primary methodology for determining off-cycle credits. For technologies not able to be demonstrated on the 5-cycle test, EPA is finalizing an approach that will include a public comment opportunity, as discussed below, which we believe addresses commenter concerns regarding maintaining a level playing field.

b. Alternative Off-Cycle Credit Methodologies

As proposed, in cases where the benefit of a technological approach to reducing CO₂ emissions can not be adequately represented using existing test cycles, manufacturers will need to develop test procedures and analytical approaches to estimate the effectiveness of the technology for the purpose of generating credits. As discussed above, the first step must be a thorough assessment of whether the 5-cycle approach can be used to demonstrate a reduction in emissions. If EPA determines that the 5-cycle process is inadequate for the specific technology being considered by the manufacturer (i.e., the 5-cycle test does not demonstrate any emissions reductions), then an alternative approach may be developed and submitted to EPA for approval. The demonstration program must be robust, verifiable, and capable of demonstrating the real-world emissions benefit of the technology with strong statistical significance.

The CO₂ benefit of some technologies may be able to be demonstrated with a modeling approach, using engineering principles. An example would be where a roof solar panel is used to charge the on-board vehicle battery. The amount of potential electrical power that the panel could supply could be modeled for average U.S. conditions and the units of electrical power could be translated to equivalent fuel energy or annualized CO₂ emission rate reduction from the captured solar energy. The CO₂ reductions from other technologies may be more challenging to quantify, especially if they are interactive with the driver, geographic location, environmental condition, or other aspect related to operation on actual roads. In these cases, manufacturers might have to design extensive on-road test programs. Any such on-road testing programs would need to be statistically robust and based on average U.S. driving conditions, factoring in differences in geography, climate, and driving behavior across the U.S.

Whether the approach involves on-road testing, modeling, or some other analytical approach, the manufacturer will be required to present a proposed methodology to EPA. EPA will approve the methodology and credits only if certain criteria are met. Baseline emissions and control emissions must be clearly demonstrated over a wide range of real world driving conditions and over a sufficient number of vehicles to address issues of uncertainty with the data. The analytical approach must be robust, verifiable, and capable of demonstrating the real-world emissions benefit with strong statistical significance. Data must be on a vehicle model-specific basis unless a manufacturer demonstrated model specific data was not necessary. Approval of the approach to determining a CO₂ benefit will not imply approval of the results of the program or methodology; when the testing, modeling, or analyses are complete the results will likewise be subject to EPA review and approval. EPA believes that manufacturers could work together to develop testing, modeling, or analytical methods for certain technologies, similar to the SAE approach used for A/C refrigerant leakage credits.

In addition, EPA received several comments recommending that the approval process include an opportunity for public comment. As noted above, some manufacturers are concerned that there be a level playing field in terms of all manufacturers having a reasonable opportunity to earn credits under an approved approach. Commenters also want an opportunity for input in the methodology to ensure the accuracy of credit determinations for these technologies. Commenters point out that there are a broad number of stakeholders with experience in the issues pertaining to the technologies that could add value in determining the most appropriate method to assess these technologies' performance. EPA agrees with these comments and is including an opportunity for public comment as part of the approval process. If and when EPA receives an application for off-cycle credits using an alternative non 5-cycle methodology, EPA will publish a notice of availability in the Federal Register with instructions on how to comment on draft off-cycle credit methodology. The public information available for review will focus on the methodology for determining credits but the public review obviously is limited to non-confidential business information. The timing for final approval will depend on the comments received. EPA also believes that a public review will encourage manufacturers to be thorough in their preparation prior to submitting their application for credits to EPA for approval. EPA will take comments into consideration, and where appropriate, work with the manufacturer to modify their approach prior to approving any off-cycle credits methodology. EPA will give final notice of its determination to the general public as well as the applicant. Off-cycle credits would be available in the model year following the final approval. Thus, it will be imperative for a manufacturer pursuing this option to begin the process as early as possible.

EPA also received comments that the off-cycle credits highlights the inadequacy of current test procedures, and that there is a clear need for updated certification test procedures. As discussed in Section III. B., EPA believes the current test procedures are adequate for implementing the standards finalized today. However, EPA is interested in improving test procedures in the future and believes that the off-cycle credits program has the potential to provide useful data and insights both for the 5-cycle test procedures and also other test procedures that capture off-cycle emissions.

5. Early Credit Options

EPA is finalizing a program to allow manufacturers to generate early credits in model years 2009-2011. As described below, credits may be generated through early additional fleet average CO₂ reductions, early A/C system improvements, early advanced technology vehicle credits, and early off-cycle credits. As with other credits, early credits are subject to a five year carry-forward limit based on the model year in which they are generated. Manufacturers may transfer early credits between vehicle categories (e.g., between the car and truck fleet). With the exception of MY 2009 early program credits, as discussed below, a manufacturer may trade other early credits to other manufacturers without limits. The agencies note that CAFE credits earned in MYs prior to MY 2011 will still be available to manufacturers for use in the CAFE program in accordance with applicable regulations.

EPA is not adopting certification, compliance, or in-use requirements for vehicles generating early credits. Since manufacturers are already certifying MY 2010 and in some cases even MY 2011 vehicles, doing so would make certification, compliance, and in-use requirements unworkable. As discussed below, manufacturers are required to submit an early credits report to EPA for approval no later than 90 days after the end of MY 2011. This report must include details on all early credits the manufacturer generates, why the credits are bona fide, how they are quantified, and how they can be verified.

a. Credits Based on Early Fleet Average CO₂ Reductions

As proposed, EPA is finalizing opportunities for early credit generation in MYs 2009-2011 through over-compliance with a fleet average CO₂ baseline established by EPA. EPA is finalizing four pathways for doing so. In order to generate early CO₂ credits, manufacturers must select one of the four paths for credit generation for the entire three year period and may not switch between pathways for different model years. For two pathways, EPA is establishing the baseline equivalent to the California standards for the relevant model year. Generally, manufacturers that over-comply with those CARB standards would earn credits. Two additional pathways, described below, include credits based on over-compliance with CAFE standards in states that have not adopted the California standards.

EPA received comments from manufacturers in support of the early credits program as a necessary compliance flexibility. The Alliance commented that the early credits reward manufacturers for providing fleet performance that exceeds California and Federal standards and do not result in a windfall. AIAM commented that early credits are essential to assure the feasibility of the proposed standards and the need for such credits must be evaluated in the context of the dramatic changes the standards will necessitate in vehicle design and the current economic environment in which manufacturers are called upon to make the changes. Manufacturers also supported retaining all four pathways, commenting that eliminating pathways would diminish the flexibility of the program. EPA also received comments from many environmental organizations and states that the program would provide manufacturers with windfall credits because manufacturers will not have to take any steps to earn credits beyond those that are already planned and in some cases implemented. These commenters were particularly concerned that the California truck standards in MY 2009 are not as stringent as CAFE, so overcompliance with the California standards could be a windfall in MY 2009, and possibly even MY 2010. These commenters supported an early credits program based on overcompliance with the more stringent of either the CAFE or California standards in any given year. EPA is retaining the early credits program because EPA judges that they are not windfall credits, and manufacturers in some cases have reasonably relied on the availability of these credits, and have based early model year compliance strategies on their availability so that the credits are needed to provide adequate lead for the initial years of the program. However, as discussed below, EPA is restricting credit trading for MY 2009 credits earned under the California-based pathways.

Manufacturers selecting Pathway 1 will generate credits by over-complying with the California equivalent baseline established by EPA over the manufacturer's fleet of vehicles sold nationwide. Manufacturers selecting Pathway 2 will generate credits against the California equivalent baseline only for the fleet of vehicles sold in California and the CAA section 177 states. This approach includes all CAA 177 states as of the date of promulgation of the Final Rule in this proceeding. Manufacturers are required to include both cars and trucks in the program. Under Pathways 1 and 2, EPA is requiring manufacturers to cover any deficits incurred against the baseline levels established by EPA during the three year period 2009-2011 before credits can be carried forward into the 2012 model year. For example, a deficit in 2011 would have to be subtracted from the sum of credits earned in 2009 and 2010 before any credits could be applied to 2012 (or later) model year fleets. EPA is including this provision to help ensure the early credits generated under this program are consistent with the credits available under the California program during these model years. In its comments, California supported such an approach.

Table III.C.5-1 provides the California equivalent baselines EPA is finalizing to be used as the basis for CO₂ credit generation under the California-based pathways. These are the California GHG standards for the model years shown. EPA proposed to adjust the California standards by 2.0 g/mile to account for the exclusion of N₂ O and CH₄, which are included in the California GHG standards, but not included in the credits program. EPA received comments from one manufacturer that this adjustment is in error and should not be made. The commenter noted that EPA already includes total hydrocarbons in the carbon balance determination of carbon related exhaust emissions and therefore already accounts for CH₄. EPA also includes CO in the carbon related exhaust emissions determination which acts to offset the need for an N₂ 0 adjustment. The commenter noted that THC and CO add about 0.8 to 3.0 g/mile to the determination of carbon related emissions and therefore EPA should not make the 2.0g/mile adjustment. The commenter is correct, and therefore the final levels shown in the table below are 2.0 g/mile higher than proposed. These comments are further discussed in the Response to Comments document. Manufacturers will generate CO₂ credits by achieving fleet average CO₂ levels below these baselines. As shown in the table, the California-based early credit pathways are based on the California vehicle categories. Also, the California-based baseline levels are not footprint-based, but universal levels that all manufacturers would use. Manufacturers will need to achieve fleet levels below those shown in the table in order to earn credits, using the California vehicle category definitions.

Manufacturers using Pathways 1 or 2 above will use year-end car and truck sales in each category. Although production data is used for the program starting in 2012, EPA is using sales data for the early credits program in order to apportion vehicles by State. This is described further below. Manufacturers must calculate actual fleet average emissions over the appropriate vehicle fleet, either for vehicles sold nationwide for Pathway 1, or California plus 177 states sales for Pathway 2. Early CO₂ credits are based on the difference between the baseline shown in the table above and the actual fleet average emissions level achieved. Any early A/C credits generated by the manufacturer, described below in Section III.C.5.b, will be included in the fleet average level determination. In model year 2009, the California CO₂ standard for cars (323 g/mi CO₂) is equivalent to 323 g/mi CO₂, and the California light-truck standard (437 g/mi CO₂) is less stringent than the equivalent CAFE standard, recognizing that there are some differences between the way the California program and the CAFE program categorize vehicles. Manufacturers are required to show that they over comply over the entire three model year time period, not just the 2009 model year, to generate early credits under either Pathways 1, 2 or 3. A manufacturer cannot use credits generated in model year 2009 unless they offset any debits from model years 2010 and 2011.

EPA received comments that this approach will provide windfall credits to manufacturers because the MY 2009 California light truck standards are less stringent than the corresponding CAFE standards. While this could be accurate if credits were based on performance in just MY 2009, that is not how credits are determined. Credits are based on the performance over a three model year period, MY 2009-2011. As noted in the proposal, EPA expects that the requirement to over comply over the entire time period covering these three model years should mean that the credits that are generated are real and are in excess of what would have otherwise occurred. However, because of the circumstances involving the 2009 model year, in particular for companies with significant truck sales, there is some concern that under Pathways 1, 2, and 3, there is a potential for a large number of credits generated in 2009 against the California standard, in particular for a number of companies who have significantly over-achieved on CAFE in recent model years. Some commenters were very concerned about this issue and commented in support of restricting credit trading between firms of MY 2009 credits based on the California program. EPA requested comments on this approach and is finalizing this credit trading restriction based on continued concerns regarding the issue of windfall credits. EPA wants to avoid a situation where, contrary to expectation, some part of the early credits generated by a manufacturer are in fact not excess, where companies could trade such credits to other manufacturers, risking a delay in the addition of new technology across the industry from the 2012 and later EPA CO₂ standards. Therefore, manufacturers selecting Pathways 1, 2, or 3 will not be allowed to trade any MY 2009 credits that they may generate.

Commenters also recommended basing credits on the more stringent of the standards between CAFE and CARB, which for MY 2009, would be the CAFE standards. However, EPA believes that this would not be necessary in light of the credit provisions requiring manufacturers choosing the California based pathways to use the California pathway for all three MYs 2009-2011, and the credit trading restrictions for MY 2009 discussed above.

In addition, for Pathways 1 and 2, EPA is allowing manufacturers to include alternative compliance credits earned per the California alternative compliance program. These alternative compliance credits are based on the demonstrated use of alternative fuels in flex fuel vehicles. As with the California program, the credits are available beginning in MY 2010. Therefore, these early alternative compliance credits are available under EPA's program for the 2010 and 2011 model years. FFVs are otherwise included in the early credit fleet average based on their emissions on the conventional fuel. This does not apply to EVs and PHEVs. The emissions of EVs and PHEVs are to be determined as described in Section III.C.3. Manufacturers may choose to either include their EVs and PHEVs in one of the four pathways described in this section or under the early advanced technology emissions credits described below, but not both due to issues of credit double counting.

EPA is also finalizing two additional early credit pathways manufacturers could select. Pathways 3 and 4 incorporate credits based on over-compliance with CAFE standards for vehicles sold outside of California and CAA 177 states in MY 2009-2011. Pathway 3 allows manufacturers to earn credits as under Pathway 2, plus earn CAFE-based credits in other states. Credits may not be generated for cars sold in California and CAA 177 states unless vehicle fleets in those states are performing better than the standards which otherwise would apply in those states, i.e., the baselines shown in Table III.C.5-1 above.

Pathway 4 is for manufacturers choosing to forego California-based early credits entirely and earn only CAFE-based credits outside of California and CAA 177 states. Manufacturers may not include FFV credits under the CAFE-based early credit pathways since those credits do not automatically reflect actual reductions in CO₂ emissions.

The baselines for CAFE-based early pathways are provided in Table III.C.5-2 below. They are based on the CAFE standards for the 2009-2011 model years. For CAFE standards in 2009-2011 model years that are footprint-based, the baseline would vary by manufacturer. Footprint-based standards are in effect for the 2011 model year CAFE standards. Additionally, for Reform CAFE truck standards, footprint standards are optional for the 2009-2010 model years. Where CAFE footprint-based standards are in effect, manufacturers will calculate a baseline using the footprints and sales of vehicles outside of California and CAA 177 states. The actual fleet CO₂ performance calculation will also only include the vehicles sold outside of California and CAA 177 states, and as mentioned above, may not include FFV credits.

For the CAFE-based pathways, EPA is using the NHTSA car and truck definitions that are in place for the model year in which credits are being generated. EPA understands that the NHTSA definitions change starting in the 2011 model year, and therefore changes part way through the early credits program. EPA further recognizes that medium-duty passenger vehicles (MDPVs) are not part of the CAFE program until the 2011 model year, and therefore are not part of the early credits calculations for 2009-2010 under the CAFE-based pathways.

Pathways 2 through 4 involve splitting the vehicle fleet into two groups, vehicles sold in California and CAA 177 states and vehicles sold outside of these states. This approach requires a clear accounting of location of vehicle sales by the manufacturer. EPA believes it will be reasonable for manufacturers to accurately track sales by State, based on its experience with the National Low Emissions Vehicle (NLEV) Program. NLEV required manufacturers to meet separate fleet average standards for vehicles sold in two different regions of the country. As with NLEV, the determination is to be based on where the completed vehicles are delivered as a point of first sale, which in most cases would be the dealer.

As noted above, manufacturers choosing to generate early CO₂ credits must select one of the four pathways for the entire early credits program and would not be able to switch among them. Manufacturers must submit their early credits report to EPA when they submit their final CAFE report for MY 2011 (which is required to be submitted no later than 90 days after the end of the model year). Manufacturers will have until then to decide which pathway to select. This gives manufacturers enough time to determine which pathway works best for them. This timing may be necessary in cases where manufacturers earn credits in MY 2011 and need time to assess data and prepare an early credits submittal for final EPA approval.

The table below provides a summary of the four fleet average-based CO₂ early credit pathways EPA is finalizing:

b. Early A/C Credits

As proposed, EPA is finalizing provisions allowing manufacturers to earn early A/C credits in MYs 2009-2011 using the same A/C system design-based EPA provisions being finalized for MYs commencing in 2012, as described in Section III.C.1, above. Manufacturers will be able to earn early A/C CO₂-equivalent credits by demonstrating improved A/C system performance, for both direct and indirect emissions. To earn credits for vehicles sold in California and CAA 177 states, the vehicles must be included in one of the California-based early credit pathways described above in III.C.5.a. EPA is finalizing this constraint in order to avoid credit double counting with the California program in place in those states, which also allows A/C system credits in this time frame. Manufacturers must fold the A/C credits into the fleet average CO₂ calculations under the California-based pathway. For example, the MY 2009 California-based program car baseline would be 323 g/mile (see Table III.C.5-1). If a manufacturer under Pathway 1 had a MY 2009 car fleet average CO₂ level of 320 g/mile and then earned an additional 12 g/mile CO₂-equivalent A/C credit, the manufacturers would earn a total of 10 g/mile of credit. Vehicles sold outside of California and 177 states would be eligible for the early A/C credits whether or not the manufacturers participate in other aspects of the early credits program. The early A/C credits for vehicles sold outside of California and 177 states are based on the NHTSA vehicle categories established for the model year in which early A/C credits are being earned.

c. Early Advanced Technology Vehicle Incentive

As discussed in Section III.C.3, above, EPA is finalizing an incentive for sales of advanced technology vehicles including EVs, PHEVs, and fuel cell vehicles. EPA is not including a multiplier for these vehicles. However, EPA is allowing the use of the 0 g/mile value for electricity operation for up to 200,000 vehicles per manufacturer (or 300,000 vehicles for any manufacturer that sells 25,000 or more advanced technology vehicles in MY 2012). EPA believes that providing an incentive for the sales of such vehicles prior to MY 2012 is consistent with the goal encouraging the introduction of such vehicles as early as possible. Therefore, manufacturers may use the 0 g/mile value for vehicles sold in MY 2009-2011 consistent with the approach being finalized for MY 2012-2016. Any vehicles sold prior to MY 2012 under these provisions must be counted against the cumulative sales cap of 200,000 (or 300,000, if applicable) vehicles. Manufacturers selling such vehicles in MY 2009-2011 have the option of either folding them into the early credits calculation under Pathways 1 through 4 described in III.C.5.a above, or tracking the sales of these vehicles separately for use in their fleetwide average compliance calculation in MY 2012 or later years, but may not do both as this would lead to double counting. Manufacturers tracking the sales of vehicles not folded into Pathways 1-4, may choose to use the vehicle counts along with the 0 g/mi emissions value (up to the applicable vehicle sales cap) to comply with 2012 or later standards. For example, if a manufacturer sells 1,000 EVs in MY 2011, the manufacturer would then be able to include 1,000 vehicles at 0 g/mile in their MY 2012 fleet to decrease the fleet average for that model year. Again, these 1,000 vehicles would be counted against the cumulative cap of 200,000 or 300,000, as applicable, vehicles. Also, these 1,000 EVs would not be included in the early credit pathways discussed above in Section III.C.5.a, otherwise the vehicles would be double counted. As with early credits, these early advanced technology vehicles will be tracked by model year (2009, 2010, or 2011) and subject to the 5-year carry-forward restrictions.

d. Early Off-Cycle Credits

EPA's is finalizing off-cycle innovative technology credit provisions, as described in Section III.C.4. EPA requested comment on beginning these credits in the 2009-2011 time frame, provided manufacturers are able to make the necessary demonstrations outlined in Section III.C.4, above. EPA is finalizing this approach for early off-cycle credits as a way to encourage innovation to lower emissions as early as possible, including the requirements for public review described in Section III.C.4. Upon EPA approval of a manufacturer's application for credits, the credits may be earned retroactively. EPA did not receive comments specifically on early off-cycle credits.

D. Feasibility of the Final CO ₂ Standards

This final rule is based on the need to obtain significant GHG emissions reductions from the transportation sector, and the recognition that there are cost-effective technologies to achieve such reductions for MY 2012-2016 vehicles. As in many prior mobile source rulemakings, the decision on what standard to set is largely based on the effectiveness of the emissions control technology, the cost and other impacts of implementing the technology, and the lead time needed for manufacturers to employ the control technology. The standards derived from assessing these factors are also evaluated in terms of the need for reductions of greenhouse gases, the degree of reductions achieved by the standards, and the impacts of the standards in terms of costs, quantified benefits, and other impacts of the standards. The availability of technology to achieve reductions and the cost and other aspects of this technology are therefore a central focus of this rulemaking.

EPA is taking the same basic approach in this rulemaking, although the technological problems and solutions involved in this rulemaking differ in some ways from prior mobile source rulemakings. Here, the focus of the emissions control technology is on reducing CO₂ and other greenhouse gases. Vehicles combust fuel to perform two basic functions: (1) To transport the vehicle, its passengers and its contents (and any towed loads), and (2) to operate various accessories during the operation of the vehicle such as the air conditioner. Technology can reduce CO₂ emissions by either making more efficient use of the energy that is produced through combustion of the fuel or reducing the energy needed to perform either of these functions.

This focus on efficiency calls for looking at the vehicle as an entire system, and the proposed and now final standards reflect this basic paradigm. In addition to fuel delivery, combustion, and aftertreatment technology, any aspect of the vehicle that affects the need to produce energy must also be considered. For example, the efficiency of the transmission system, which takes the energy produced by the engine and transmits it to the wheels, and the resistance of the tires to rolling both have major impacts on the amount of fuel that is combusted while operating the vehicle. The braking system, the aerodynamics of the vehicle, and the efficiency of accessories, such as the air conditioner, all affect how much fuel is combusted as well.

In evaluating vehicle efficiency, we have excluded fundamental changes in vehicles' size and utility. For example, we did not evaluate converting minivans and SUVs to station wagons, converting vehicles with four wheel drive to two wheel drive, or reducing headroom in order to lower the roofline and reduce aerodynamic drag. We have limited our assessment of technical feasibility and resultant vehicle cost to technologies which maintain vehicle utility as much as possible. Manufacturers may decide to alter the utility of the vehicles which they sell in response to this rule, but this is not a necessary consequence of the rule but rather a matter of automaker choice.

This need to focus on the efficient use of energy by the vehicle as a system leads to a broad focus on a wide variety of technologies that affect almost all the systems in the design of a vehicle. As discussed below, there are many technologies that are currently available which can reduce vehicle energy consumption. These technologies are already being commercially utilized to a limited degree in the current light-duty fleet. These technologies include hybrid technologies that use higher efficiency electric motors as the power source in combination with or instead of internal combustion engines. While already commercialized, hybrid technology continues to be developed and offers the potential for even greater efficiency improvements. Finally, there are other advanced technologies under development, such as lean burn gasoline engines, which offer the potential of improved energy generation through improvements in the basic combustion process. In addition, the available technologies are not limited to powertrain improvements but also include mass reduction, electrical system efficiencies, and aerodynamic improvements.

The large number of possible technologies to consider and the breadth of vehicle systems that are affected mean that consideration of the manufacturer's design and production process plays a major role in developing the final standards. Vehicle manufacturers typically develop many different models by basing them on a limited number of vehicle platforms. The platform typically consists of a common set of vehicle architecture and structural components. This allows for efficient use of design and manufacturing resources. Given the very large investment put into designing and producing each vehicle model, manufacturers typically plan on a major redesign for the models approximately every 5 years. At the redesign stage, the manufacturer will upgrade or add all of the technology and make most other changes supporting the manufacturer's plans for the next several years, including plans related to emissions, fuel economy, and safety regulations.

This redesign often involves a package of changes designed to work together to meet the various requirements and plans for the model for several model years after the redesign. This often involves significant engineering, development, manufacturing, and marketing resources to create a new product with multiple new features. In order to leverage this significant upfront investment, manufacturers plan vehicle redesigns with several model years' of production in mind. Vehicle models are not completely static between redesigns as limited changes are often incorporated for each model year. This interim process is called a refresh of the vehicle and generally does not allow for major technology changes although more minor ones can be done (e.g., small aerodynamic improvements, valve timing improvements, etc.). More major technology upgrades that affect multiple systems of the vehicle thus occur at the vehicle redesign stage and not in the time period between redesigns. The Center for Biological Diversity commented on EPA's assumptions on redesign cycles, and these comments are addressed in Section III.D.7 below.

As discussed below, there are a wide variety of CO₂ reducing technologies involving several different systems in the vehicle that are available for consideration. Many can involve major changes to the vehicle, such as changes to the engine block and cylinder heads, redesign of the transmission and its packaging in the vehicle, changes in vehicle shape to improve aerodynamic efficiency and the application of aluminum (and other lightweight materials) in body panels to reduce mass. Logically, the incorporation of emissions control technologies would be during the periodic redesign process. This approach would allow manufacturers to develop appropriate packages of technology upgrades that combine technologies in ways that work together and fit with the overall goals of the redesign. It also allows the manufacturer to fit the process of upgrading emissions control technology into its multi-year planning process, and it avoids the large increase in resources and costs that would occur if technology had to be added outside of the redesign process.

This final rule affects five years of vehicle production, model years 2012-2016. Given the now-typical five year redesign cycle, nearly all of a manufacturer's vehicles will be redesigned over this period. However, this assumes that a manufacturer has sufficient lead time to redesign the first model year affected by this final rule with the requirements of this final rule in mind. In fact, the lead time available for the start of model year 2012 (January 2011) is relatively short, less than a year. The time between this final rule and the start of 2013 model year (January 2012) production is under two years. At the same time, manufacturer product plans indicate that they are planning on introducing many of the technologies EPA projects could be used to show compliance with the final CO₂ standards in both 2012 and 2013. In order to account for the relatively short lead time available prior to the 2012 and 2013 model years, albeit mitigated by their existing plans, EPA has factored this reality into how the availability is modeled for much of the technology being considered for model years 2012-2016 as a whole. If the technology to control greenhouse gas emissions is efficiently folded into this redesign process, then EPA projects that 85 percent of each manufacturer's sales will be able to be redesigned with many of the CO₂ emission reducing technologies by the 2016 model year, and as discussed below, to reduce emissions of HFCs from the air conditioner.

In determining the level of this first ever GHG emissions standard under the CAA for light-duty vehicles, EPA uses an approach that accounts for and builds on this redesign process. This provides the opportunity for several control technologies to be incorporated into the vehicle during redesign, achieving significant emissions reductions from the model at one time. This is in contrast to what would be a much more costly approach of trying to achieve small increments of reductions over multiple years by adding technology to the vehicle piece by piece outside of the redesign process.

As described below, the vast majority of technology required by this final rule is commercially available and already being employed to a limited extent across the fleet (although the final rule will necessitate far wider penetration of these technologies throughout the fleet). The vast majority of the emission reductions which will result from this final rule will be produced from the increased use of these technologies. EPA also believes that this final rule will encourage the development and limited use of more advanced technologies, such as PHEVs and EVs, and the final rule is structured to facilitate this result.

In developing the final standard, EPA built on the technical work performed by the State of California during its development of its statewide GHG program. EPA began by evaluating a nationwide CAA standard for MY 2016 that would require the levels of technology upgrade, across the country, which California standards would require for the subset of vehicles sold in California under Pavley 1. In essence, EPA developed an assessment of an equivalent national new vehicle fleet-wide CO₂ performance standards for model year 2016 which would result in the new vehicle fleet in the State of California having CO₂ performance equal to the performance from the California Pavley 1 standards. This assessment is documented in Chapter 3.1 of the RIA. The results of this assessment predicts that a national light-duty vehicle fleet which adopts technology that achieves performance of 250 g/mile CO₂ for model year 2016 will result in vehicles sold in California that would achieve the CO₂ performance equivalent to the Pavley 1 standards.

EPA then analyzed a level of 250 g/mi CO₂ in 2016 using the OMEGA model (described in more detail below), and the car and truck footprint curves' relative stringency discussed in Section II to determine what technology will be needed to achieve a fleet wide average of 250 g/mi CO₂. As discussed later in this section we believe this level of technology application to the light-duty vehicle fleet can be achieved in this time frame, that such standards will produce significant reductions in GHG emissions, and that the costs for both the industry and the costs to the consumer are reasonable. EPA also developed standards for the model years 2012 through 2015 that lead up to the 2016 level.

EPA's independent technical assessment of the technical feasibility of the final MY 2012-2016 standards is described below. EPA has also evaluated a set of alternative standards for these model years, one that is more stringent than the final standards and one that is less stringent. The technical feasibility of these alternative standards is discussed at the end of this section.

Evaluating the feasibility of these standards primarily includes identifying available technologies and assessing their effectiveness, cost, and impact on relevant aspects of vehicle performance and utility. The wide number of technologies which are available and likely to be used in combination requires a more sophisticated assessment of their combined cost and effectiveness. An important factor is also the degree that these technologies are already being used in the current vehicle fleet and thus, unavailable for use to improve energy efficiency beyond current levels. Finally, the challenge for manufacturers to design the technology into their products, and the appropriate lead time needed to employ the technology over the product line of the industry must be considered.

Applying these technologies efficiently to the wide range of vehicles produced by various manufacturers is a challenging task. In order to assist in this task, EPA has developed a computerized model called the Optimization Model for reducing Emissions of Greenhouse gases from Automobiles (OMEGA) model. Broadly, the model starts with a description of the future vehicle fleet, including manufacturer, sales, base CO₂ emissions, footprint and the extent to which emission control technologies are already employed. For the purpose of this analysis, over 200 vehicle platforms were used to capture the important differences in vehicle and engine design and utility of future vehicle sales of roughly 16 million units in the 2016 timeframe. The model is then provided with a list of technologies which are applicable to various types of vehicles, along with their cost and effectiveness and the percentage of vehicle sales which can receive each technology during the redesign cycle of interest. The model combines this information with economic parameters, such as fuel prices and a discount rate, to project how various manufacturers would apply the available technology in order to meet various levels of emission control. The result is a description of which technologies are added to each vehicle platform, along with the resulting cost. While OMEGA can apply technologies which reduce CO₂ emissions and HFC refrigerant emissions associated with air conditioner use, this task is currently handled outside of the OMEGA model. The model can be set to account for various types of compliance flexibilities, such as FFV credits.

The remainder of this section describes the technical feasibility analysis in greater detail. Section III.D.1 describes the development of our projection of the MY 2012-2016 fleet in the absence of this final rule. Section III.D.2 describes our estimates of the effectiveness and cost of the control technologies available for application in the 2012-2016 timeframe. Section III.D.3 combines these technologies into packages likely to be applied at the same time by a manufacturer. In this section, the overall effectiveness of the technology packages vis-à-vis their effectiveness when combined individually is described. Section III.D.4 describes the process which manufacturers typically use to apply new technology to their vehicles. Section III.D.5 describes EPA's OMEGA model and its approach to estimating how manufacturers will add technology to their vehicles in order to comply with CO₂ emission standards. Section III.D.6 presents the results of the OMEGA modeling, namely the level of technology added to manufacturers' vehicles and its cost. Section III.D.7 discusses the feasibility of the alternative 4-percent-per-year and 6-percent-per-year standards. Further detail on all of these issues can be found in EPA and NHTSA's Joint Technical Support Document as well as EPA's Regulatory Impact Analysis.

1. How did EPA develop a reference vehicle fleet for evaluating further CO₂ reductions?

In order to calculate the impacts of this final rule, it is necessary to project the GHG emissions characteristics of the future vehicle fleet absent this regulation. This is called the “reference” fleet. EPA and NHTSA develop this reference fleet using a three step process. Step one develops a set of detailed vehicle characteristics and sales for a specific model year (in this case, 2008). This is called the baseline fleet. Step two adjusts the sales of these vehicles using projections made by AEO and CSM to account for expected changes in market conditions. Step three applies fuel saving and emission control technology to these vehicles to the extent necessary for manufacturers to comply with the MY 2011 CAFE standards. Thus, the reference fleet differs from the MY 2008 baseline fleet in both the level of technology utilized and in terms of the sales of any particular vehicle.

EPA and NHTSA perform steps one and two in an identical manner. The development of the characteristics of the baseline 2008 fleet and the adjustment of sales to match AEO and CSM forecasts is described in detail in Section II.B above. The two agencies perform step three in a conceptually identical manner, but each agency utilizes its own vehicle technology and emission model to project the technology needed to comply with the 2011 CAFE standards. The agencies use the same two models to project the technology and cost of the 2012-2016 standards. Use of the same model for both pre-control and post-control costs ensures consistency.

The agencies received one comment from the Center for Biological Diversity that the use of 2008 vehicles in our baseline and reference fleets inherently includes vehicle models which already have or will be discontinued by the time this rule takes effect and will be replaced by more advanced vehicle models. This is true. However, we believe that the use of 2008 vehicle designs is still the most appropriate approach available. First, as discussed in Section II.B above, the designs of these new vehicles at the level of detail required for emission and cost modeling are not publically available. Even the confidential descriptions of these vehicle designs are usually not of sufficient detail to facilitate the level of technology and emission modeling performed by both agencies. Second, steps two and three of the process used to create the reference fleet adjust both the sales and technology of the 2008 vehicles. Thus, our reference fleet reflects the extent that completely new vehicles are expected to shift the light vehicle market in terms of both segment and manufacturer. Also, by adding technology to facilitate compliance with the 2011 CAFE standards, we account for the vast majority of ways in which these new vehicles will differ from their older counterparts.

The agencies also received a comment that some manufacturers have already announced plans to introduce technology well beyond that required by the 2011 MY CAFE standards. This commenter indicated that the agencies' approach over-estimated the technology and cost required by the proposed standards and resulted in less stringent standards being proposed than a more realistic reference fleet would have supported. First, the agencies agree that limiting the application of additional technology beyond that already on 2008 vehicles to only that required by the 2011 CAFE standards could under-estimate the use of such technology absent this rule. However, it is difficult, if not impossible, to separate future fuel economy improvements made for marketing purposes from those designed to facilitate compliance with anticipated CAFE or CO₂ emission standards. For example, EISA was signed over two years ago, which contained specific minimum limits on light vehicle fuel economy in 2020, while also requiring ratable improvements in the interim. NHTSA proposed fuel economy standards for the 2012-2015 model years under the EISA provisions in April of 2008, although NHTSA finalized only 2011 standards for passenger vehicles. It is also true that manufacturers can change their plans based on market conditions and other factors. Thus, announcements of future plans are not certain. As mentioned above, these plans do not include specific vehicle characteristics. Thus, in order to avoid under-estimating the cost associated with this rule, the agencies have limited the fuel economy improvements in the reference fleet to those projected to result from the existing CAFE standards. We disagree with the commenter that this has caused the standards being promulgated today to be less stringent than would have been the case had we been able to confidently predict additional fuel economy and CO₂ emission improvements which will occur absent this rule. The inclusion of such technology in the reference fleet would certainly have reduced the cost of this final rule, as well as the benefits, but would not have changed the final level of technology required to meet the final standards. Also, we believe that the same impacts would apply to our evaluations of the two alternative sets of standards, the 4% per year and 6% per year standards. We are confident that the vast majority of manufacturers would not comply with the least stringent of these standards (the 4% per year standards) in the absence of this rule. Thus, changes to the reference fleet would not have affected the differences in technology, cost or benefits between the final standards and the two alternatives. As described below, our rejection of the two alternatives in favor of the final standards is based primarily on the relative technology, cost and benefits associated with the three sets of standards than the absolute cost or benefit relative to the reference fleet. Thus, we do not agree with the commenter that our choice of reference fleet adversely impacted the development of the final standards being promulgated today.

The addition of technology to the baseline fleet so that it complies with the MY 2011 CAFE standards is described later in Section III.D.4, as this uses the same methodology used to project compliance with the final CO₂ emission standards. In summary, the reference fleet represents vehicle characteristics and sales in the 2012 and later model years absent this final rule. Technology is then added to these vehicles in order to reduce CO₂ emissions to achieve compliance with the final CO₂ standards. As noted above, EPA did not factor in any changes to vehicle utility or characteristics, or sales in projecting manufacturers' compliance with this final rule.

After the reference fleet is created, the next step aggregates vehicle sales by a combination of manufacturer, vehicle platform, and engine design. As discussed in Section III.D.4 below, manufacturers implement major design changes at vehicle redesign and tend to implement these changes across a vehicle platform. Because the cost of modifying the engine depends on the valve train design (such as SOHC, DOHC, etc.), the number of cylinders and in some cases head design, the vehicle sales are broken down beyond the platform level to reflect relevant engine differences. The vehicle groupings are shown in Table III.D.1-1. These groupings are the same as those used in the NPRM.

As mentioned above, the second factor which needs to be considered in developing a reference fleet against which to evaluate the impacts of this final rule is the impact of the 2011 MY CAFE standards. Since the vehicles which comprise the above reference fleet are those sold in the 2008 MY, when coupled with our sales projections, they do not necessarily meet the 2011 MY CAFE standards.

The levels of the 2011 MY CAFE standards are straightforward to apply to future sales fleets, as is the potential fine-paying flexibility afforded by the CAFE program (i.e., $55 per mpg of shortfall). However, projecting some of the compliance flexibilities afforded by EISA and the CAFE program are less clear. Two of these compliance flexibilities are relevant to EPA's analysis: (1) The credit for FFVs, and (2) the limit on the transferring of credits between car and truck fleets. The FFV credit is limited to 1.2 mpg in 2011 and EISA gradually reduces this credit, to 1.0 mpg in 2015 and eventually to zero in 2020. In contrast, the limit on car-truck transfer is limited to 1.0 mpg in 2011, and EISA increases this to 1.5 mpg beginning in 2015 and then to 2.0 mpg beginning in 2020. The question here is whether to hold the 2011 MY CAFE provisions constant in the future or incorporate the changes in the FFV credit and car-truck credit trading limits contained in EISA.

As was done for the NPRM, EPA has decided to hold the 2011 MY limits on FFV credit and car-truck credit trading constant in projecting the fuel economy and CO₂ emission levels of vehicles in our reference case. This approach treats the changes in the FFV credit and car-truck credit trading provisions consistently with the other EISA-mandated changes in the CAFE standards themselves. All EISA provisions relevant to 2011 MY vehicles are reflected in our reference case fleet, while all post-2011 MY provisions are not. Practically, relative to the alternative, this increases both the cost and benefit of the final standards. In our analysis of this final rule, any quantified benefits from the presence of FFVs in the fleet are not considered. Thus, the only impact of the FFV credit is to reduce onroad fuel economy. By assuming that the FFV credit stays at 1.2 mpg in the future absent this rule, the assumed level of onroad fuel economy that would occur absent this final rule is reduced. As this final rule eliminates the FFV credit (for purposes of CO₂ emission compliance) starting in 2016, the net result is to increase the projected level of fuel savings from our final standards. Similarly, the higher level of FFV credit reduces projected compliance cost for manufacturers to meet the 2011 MY standards in our reference case. This increases the projected cost of meeting the final 2012 and later standards.

As just implied, EPA needs to project the technology (and resultant costs) required for the 2008 MY vehicles to comply with the 2011 MY CAFE standards in those cases where they do not automatically do so. The technology and costs are projected using the same methodology that projects compliance with the final 2012 and later CO₂ standards. The description of this process is described in the following four sections and is essentially the same process used for the NPRM.

A more detailed description of the methodology used to develop these sales projections can be found in the Joint TSD. Detailed sales projections by model year and manufacturer can also be found in the TSD.

2. What are the effectiveness and costs of CO₂-reducing technologies?

EPA and NHTSA worked together to jointly develop information on the effectiveness and cost of the CO₂-reducing technologies, and fuel economy-improving technologies, other than A/C related control technologies. This joint work is reflected in Chapter 3 of the Joint TSD and in Section II of this preamble. A summary of the effectiveness and cost of A/C related technology is contained here. For more detailed information on the effectiveness and cost of A/C related technology, please refer to Section III.C of this preamble and Chapter 2 of EPA's RIA.

A/C improvements are an integral part of EPA's technology analysis and have been included in this section along with the other technology options. While discussed in Section III.C as a credit opportunity, air conditioning-related improvements are included in Table III.D.2-1. because A/C improvements are a very cost-effective technology at reducing CO₂ (or CO₂-equivalent) emissions. EPA expects most manufacturers will choose to use AC improvement credit opportunities as a strategy for meeting compliance with the CO₂ standards. Note that the costs shown in Table III.D.2-1 do not include maintenance savings that would be expected from the new AC systems. Further, EPA does not include AC-related maintenance savings in our cost and benefit analysis presented in Section III.H. EPA discusses the likely maintenance savings in Chapter 2 of the RIA, though these savings are not included in our final cost estimates for the final rule. The EPA approximates that the level of the credits earned will increase from 2012 to 2016 as more vehicles in the fleet are redesigned. The penetrations and average levels of credit are summarized in Table III.D.2-2, though the derivation of these numbers (and the breakdown of car vs. truck credits) is described in the RIA. As demonstrated in the IMAC study (and described in Section III.C as well as the RIA), these levels are feasible and achievable with technologies that are available and cost-effective today.

These improvements are categorized as either leakage reduction, including use of alternative refrigerants, or system efficiency improvements. Unlike the majority of the technologies described in this section, A/C improvements will not be demonstrated in the test cycles used to quantify CO₂ reductions in this final rule. As described earlier, for this analysis A/C-related CO₂ reductions are handled outside of OMEGA model and therefore their CO₂ reduction potential is expressed in grams per mile rather than a percentage used by the OMEGA model. See Section III.C.1 for the method by which potential reductions are calculated or measured. Further discussion of the technological basis for these improvements is included in Chapter 2 of the RIA.

3. How can technologies be combined into “packages” and what is the cost and effectiveness of packages?

Individual technologies can be used by manufacturers to achieve incremental CO₂ reductions. However, as mentioned in Section III.D.1, EPA believes that manufacturers are more likely to bundle technologies into “packages” to capture synergistic aspects and reflect progressively larger CO₂ reductions with additions or changes to any given package. In addition, manufacturers typically apply new technologies in packages during model redesigns that occur approximately once every five years, rather than adding new technologies one at a time on an annual or biennial basis. This way, manufacturers can more efficiently make use of their redesign resources and more effectively plan for changes necessary to meet future standards.

Therefore, as explained at proposal, the approach taken here is to group technologies into packages of increasing cost and effectiveness. EPA determined that 19 different vehicle types provided adequate representation to accurately model the entire fleet. This was the result of analyzing the existing light duty fleet with respect to vehicle size and powertrain configurations. All vehicles, including cars and trucks, were first distributed based on their relative size, starting from compact cars and working upward to large trucks. Next, each vehicle was evaluated for powertrain, specifically the engine size, I4, V6, and V8, and finally by the number of valves per cylinder. Note that each of these 19 vehicle types was mapped into one of the five classes of vehicles mentioned in Section III.D.2. While the five classes provide adequate representation for the cost basis associated with most technology application, they do not adequately account for all existing vehicle attributes such as base vehicle powertrain configuration and mass reduction. As an example, costs and effectiveness estimates for engine friction reduction for the small car class were used to represent cost and effectiveness for three vehicle types: Subcompact cars, compact cars, and small multi-purpose vehicles (MPV) equipped with a 4-cylinder engine, however the mass reduction associated for each of these vehicle types was based on the vehicle type sales-weighted average. In another example, a vehicle type for V8 single overhead cam 3-valve engines was created to properly account for the incremental cost in moving to a dual overhead cam 4-valve configuration. Note also that these 19 vehicle types span the range of vehicle footprint (smaller footprints for smaller vehicles and larger footprints for larger vehicles) which serve as the basis for the standards being promulgated today. A complete list of vehicles and their associated vehicle types is shown above in Table III.D.1-1.

Within each of the 19 vehicle types, multiple technology packages were created in increasing technology content resulting in increasing effectiveness. Important to note that the effort in creating the packages attempted to maintain a constant utility for each package as compared to the baseline package. As such, each package is meant to provide equivalent driver-perceived performance to the baseline package. The initial packages represent what a manufacturer will most likely implement on all vehicles, including low rolling resistance tires, low friction lubricants, engine friction reduction, aggressive shift logic, early torque converter lock-up, improved electrical accessories, and low drag brakes. Subsequent packages include advanced gasoline engine and transmission technologies such as turbo/downsizing, GDI, and dual-clutch transmission. The most technologically advanced packages within a segment included HEV, PHEV and EV designs. The end result is a list of several packages for each of 19 different vehicle types from which a manufacturer could choose in order to modify its fleet such that compliance could be achieved.

Before using these technology packages as inputs to the OMEGA model, EPA calculated the cost and effectiveness for the package. The first step was to apply the scaling class for each technology package and vehicle type combination. The scaling class establishes the cost and effectiveness for each technology with respect to the vehicle size or type. The Large Car class was provided as an example in Section III.D.2. Additional classes include Small Car, Minivan, Small Truck, and Large Truck and each of the 19 vehicle types was mapped into one of those five classes. In the next step, the cost for a particular technology package was determined as the sum of the costs of the applied technologies. The final step, determination of effectiveness, requires greater care due to the synergistic effects mentioned in Section III.D.2. This step is described immediately below.

Usually, the benefits of the engine and transmission technologies can be combined multiplicatively. For example, if an engine technology reduces CO₂ emissions by five percent and a transmission technology reduces CO₂ emissions by four percent, the benefit of applying both technologies is 8.8 percent (100%−(100%−4%) * (100%−5%)). In some cases, however, the benefit of the transmission-related technologies overlaps with many of the engine technologies. This occurs because the primary goal of most of the transmission technologies is to shift operation of the engine to more efficient locations on the engine map. This is accomplished by incorporating more ratio selections and a wider ratio span into the transmissions. Some of the engine technologies have the same goal, such as cylinder deactivation, advanced valvetrains, and turbocharging. In order to account for this overlap and avoid over-estimating emissions reduction effectiveness, EPA has developed a set of adjustment factors associated with specific pairs of engine and transmission technologies.

The various transmission technologies are generally mutually exclusive. As such, the effectiveness of each transmission technology generally supersedes each other. For example, the 9.5-14.5 percent reduction in CO₂ emissions associated with the automated manual transmission includes the 4.5-6.5 percent benefit of a 6-speed automatic transmission. Exceptions are aggressive shift logic and early torque converter lock-up that can be applied to vehicles with several types of automatic transmissions.

EPA has chosen to use an engineering approach known as the lumped-parameter technique to determine these adjustment factors. The results from this approach were then applied directly to the vehicle packages. The lumped-parameter technique is well documented in the literature, and the specific approach developed by EPA is detailed in Chapter 1 of the RIA.

Table III.D.3-1 presents several examples of the reduction in the effectiveness of technology pairs. A complete list and detailed discussion of these synergies is presented in Chapter 3 of the Joint TSD.

Table III.D.3-2 presents several examples of the CO₂-reducing technology vehicle packages used in the OMEGA model for the large car class. Similar packages were generated for each of the 19 vehicle types and the costs and effectiveness estimates for each of those packages are discussed in detail in Chapter 3 of the Joint TSD.

4. Manufacturer's Application of Technology

Vehicle manufacturers often introduce major product changes together, as a package. In this manner the manufacturers can optimize their available resources, including engineering, development, manufacturing and marketing activities to create a product with multiple new features. In addition, manufacturers recognize that a vehicle will need to remain competitive over its intended life, meet future regulatory requirements, and contribute to a manufacturer's CAFE requirements. Furthermore, automotive manufacturers are largely focused on creating vehicle platforms to limit the development of entirely new vehicles and to realize economies of scale with regard to variable cost. In very limited cases, manufacturers may implement an individual technology outside of a vehicle's redesign cycle. In following with these industry practices, EPA has created set of vehicle technology packages that represent the entire light duty fleet.

In evaluating needed lead time, EPA has historically authorized manufacturers of new vehicles or nonroad equipment to phase in available emission control technology over a number of years. Examples of this are EPA's Tier 2 program for cars and light trucks and its 2007 and later PM and NO_X emission standards for heavy-duty vehicles. In both of these rules, the major modifications expected from the rules were the addition of exhaust aftertreatment control technologies. Some changes to the engine were expected as well, but these were not expected to affect engine size, packaging or performance. The CO₂ reduction technologies described above potentially involve much more significant changes to car and light truck designs. Many of the engine technologies involve changes to the engine block and heads. The transmission technologies could change the size and shape of the transmission and thus, packaging. Improvements to aerodynamic drag could involve body design and therefore, the dies used to produce body panels. Changes of this sort potentially involve new capital investment and the obsolescence of existing investment.

At the same time, vehicle designs are not static, but change in major ways periodically. The manufacturers' product plans indicate that vehicles are usually redesigned every 5 years on average. Vehicles also tend to receive a more modest “refresh” between major redesigns, as discussed above. Because manufacturers are already changing their tooling, equipment and designs at these times, further changes to vehicle design at these times involve a minimum of stranded capital equipment. Thus, the timing of any major technological changes is projected to coincide with changes that manufacturers are already making to their vehicles. This approach effectively avoids the need to quantify any costs associated with discarding equipment, tooling, emission and safety certification, etc. when CO₂-reducing equipment is incorporated into a vehicle.

This final rule affects five years of vehicle production, model years 2012-2016. Given the now-typical five year redesign cycle, nearly all of a manufacturer's vehicles will be redesigned over this period. However, this assumes that a manufacturer has sufficient lead time to redesign the first model year affected by this final rule with the requirements of this final rule in mind. In fact, the lead time available for model year 2012 is relatively short. The time between a likely final rule and the start of 2013 model year production is likely to be just over two years. At the same time, the manufacturer product plans indicate that they are planning on introducing many of the technologies projected to be required by this final rule in both 2012 and 2013. In order to account for the relatively short lead time available prior to the 2012 and 2013 model years, albeit mitigated by their existing plans, EPA projects that only 85 percent of each manufacturer's sales will be able to be redesigned with major CO₂ emission-reducing technologies by the 2016 model year. Less intrusive technologies can be introduced into essentially all of a manufacturer's sales. This resulted in three levels of technology penetration caps, by manufacturer. Common technologies (e.g., low friction lubes, aerodynamic improvements) had a penetration cap of 100%. More advanced powertrain technologies (e.g., stoichiometric GDI, turbocharging) had a penetration cap of 85%. The most advanced technologies considered in this analysis (e.g., diesel engines, as well as IMA, powersplit and 2-mode hybrids) had a 15% penetration cap.

This is the same approach as was taken in the NPRM. EPA received several comments commending it on its approach to establishing technical feasibility via its use of the OMEGA model. The only adverse comment received regarding the application of technology was from the Center for Biological Diversity (CBD), which criticized EPA's use of the 5-year redesign cycle. CBD argued that manufacturers occasionally redesign vehicles sooner than 5 years and that EPA did not quantify the cost of shortening the redesign cycle to less than 5 years and compare this cost to the increased benefit of reduced CO₂ emissions. CBD also noted that manufacturers have been recently dropping vehicle lines and entire divisions with very little leadtime, indicating their ability to change product plans much quicker than projected above.

EPA did not explicitly evaluate the cost of reducing the average redesign cycle to less than 5 years for two reasons. One, in the past, manufacturers have usually shortened the redesign cycle to address serious problems with the current design, usually lower than anticipated sales. However, the amortized cost of the capital necessary to produce a new vehicle design will increase by 23%, from one-fifth of the capital cost to one-fourth (and assuming a 3% discount rate). This would be on top of the cost of the emission control equipment itself. The only benefit of this increase in societal cost will be earlier CO₂ emission reductions (and the other benefits associated with CO₂ emission control). The capital costs associated with vehicle redesign go beyond CO₂ emission control and potentially involve every aspect of the vehicle and can represent thousands of dollars. We believe that it would be an inefficient use of societal resources to incur such costs when they can be obtained much more cost effectively just one year later.

Two, the examples of manufacturers dropping vehicle lines and divisions with very short lead time is not relevant to the redesign of vehicles. There is no relationship between a manufacturer's ability to stop selling a vehicle model or to close a vehicle division and a manufacturer's ability to redesign a vehicle. A company could decide to stop selling all of its products within a few weeks—but it would still take a firm approximately 5 years to introduce a major new vehicle line. It is relatively easy to stop the manufacture of a particular product (though this too can incur some cost—such as plant wind-down costs, employee layoff or relocation costs, and dealership related costs). It is much more difficult to perform the required engineering design and development, design, purchase, and install the necessary capital equipment and tooling for components and vehicle manufacturing and develop all the processes associated with the application of a new technology. Further discussion of the CBD comments can be found in III.D.7 below.

5. How is EPA projecting that a manufacturer decides between options to improve CO₂ performance to meet a fleet average standard?

EPA is generally taking the same approach to projecting the application of technology to vehicles as it did for the NPRM. With the exception of two comments, all commenters agreed with the modeling approach taken in the NPRM. One of these two comments is addressed is Section III.D.1 above, while the other is addressed in Section III.D.3. above.

There are many ways for a manufacturer to reduce CO₂-emissions from its vehicles. A manufacturer can choose from a myriad of CO₂ reducing technologies and can apply one or more of these technologies to some or all of its vehicles. Thus, for a variety of levels of CO₂ emission control, there are an almost infinite number of technology combinations which produce the desired CO₂ reduction. As noted earlier, EPA developed a new vehicle model, the OMEGA model in order to make a reasonable estimate of how manufacturers will add technologies to vehicles in order to meet a fleet-wide CO₂ emissions level. EPA has described OMEGA's specific methodologies and algorithms in a memo to the docket for this rulemaking (Docket EPA-HQ-OAR-2009-0472).

The OMEGA model utilizes four basic sets of input data. The first is a description of the vehicle fleet. The key pieces of data required for each vehicle are its manufacturer, CO₂ emission level, fuel type, projected sales and footprint. The model also requires that each vehicle be assigned to one of the 19 vehicle types, which tells the model which set of technologies can be applied to that vehicle. (For a description of how the 19 vehicle types were created, reference Section III.D.3.) In addition, the degree to which each vehicle already reflects the effectiveness and cost of each available technology must also be input. This avoids the situation, for example, where the model might try to add a basic engine improvement to a current hybrid vehicle. Except for this type of information, the development of the required data regarding the reference fleet was described in Section III.D.1 above and in Chapter 1 of the Joint TSD.

The second type of input data used by the model is a description of the technologies available to manufacturers, primarily their cost and effectiveness. Note that the five vehicle classes are not explicitly used by the model, rather the costs and effectiveness associated with each vehicle package is based on the associated class. This information was described in Sections III.D.2 and III.D.3 above as well as Chapter 3 of the Joint TSD. In all cases, the order of the technologies or technology packages for a particular vehicle type is determined by the model user prior to running the model. Several criteria can be used to develop a reasonable ordering of technologies or packages. These are described in the Joint TSD.

The third type of input data describes vehicle operational data, such as annual scrap rates and mileage accumulation rates, and economic data, such as fuel prices and discount rates. These estimates are described in Section II.F above, Section III.H below and Chapter 4 of the Joint TSD.

The fourth type of data describes the CO₂ emission standards being modeled. These include the CO₂ emission equivalents of the 2011 MY CAFE standards and the final CO₂ standards for 2016. As described in more detail below, the application of A/C technology is evaluated in a separate analysis from those technologies which impact CO₂ emissions over the 2-cycle test procedure. Thus, for the percent of vehicles that are projected to achieve A/C related reductions, the CO₂ credit associated with the projected use of improved A/C systems is used to adjust the final CO₂ standard which will be applicable to each manufacturer to develop a target for CO₂ emissions over the 2-cycle test which is assessed in our OMEGA modeling.

As mentioned above for the market data input file utilized by OMEGA, which characterizes the vehicle fleet, our modeling must and does account for the fact that many 2008 MY vehicles are already equipped with one or more of the technologies discussed in Section III.D.2 above. Because of the choice to apply technologies in packages, and 2008 vehicles are equipped with individual technologies in a wide variety of combinations, accounting for the presence of specific technologies in terms of their proportion of package cost and CO₂ effectiveness requires careful, detailed analysis. The first step in this analysis is to develop a list of individual technologies which are either contained in each technology package, or would supplant the addition of the relevant portion of each technology package. An example would be a 2008 MY vehicle equipped with variable valve timing and a 6-speed automatic transmission. The cost and effectiveness of variable valve timing would be considered to be already present for any technology packages which included the addition of variable valve timing or technologies which went beyond this technology in terms of engine related CO₂ control efficiency. An example of a technology which supplants several technologies would be a 2008 MY vehicle which was equipped with a diesel engine. The effectiveness of this technology would be considered to be present for technology packages which included improvements to a gasoline engine, since the resultant gasoline engines have a lower CO₂ control efficiency than the diesel engine. However, if these packages which included improvements also included improvements unrelated to the engine, like transmission improvements, only the engine related portion of the package already present on the vehicle would be considered. The transmission related portion of the package's cost and effectiveness would be allowed to be applied in order to comply with future CO₂ emission standards.

The second step in this process is to determine the total cost and CO₂ effectiveness of the technologies already present and relevant to each available package. Determining the total cost usually simply involves adding up the costs of the individual technologies present. In order to determine the total effectiveness of the technologies already present on each vehicle, the lumped parameter model described above is used. Because the specific technologies present on each 2008 vehicle are known, the applicable synergies and dis-synergies can be fully accounted for.

The third step in this process is to divide the total cost and CO₂ effectiveness values determined in step 2 by the total cost and CO₂ effectiveness of the relevant technology packages. These fractions are capped at a value of 1.0 or less, since a value of 1.0 causes the OMEGA model to not change either the cost or CO₂ emissions of a vehicle when that technology package is added.

As described in Section III.D.3 above, technology packages are applied to groups of vehicles which generally represent a single vehicle platform and which are equipped with a single engine size (e.g., compact cars with four cylinder engine produced by Ford). These grouping are described in Table III.D.1-1. Thus, the fourth step is to combine the fractions of the cost and effectiveness of each technology package already present on the individual 2008 vehicles models for each vehicle grouping. For cost, percentages of each package already present are combined using a simple sales-weighting procedure, since the cost of each package is the same for each vehicle in a grouping. For effectiveness, the individual percentages are combined by weighting them by both sales and base CO₂ emission level. This appropriately weights vehicle models with either higher sales or CO₂ emissions within a grouping. Once again, this process prevents the model from adding technology which is already present on vehicles, and thus ensures that the model does not double count technology effectiveness and cost associated with complying with the 2011 MY CAFE standards and the final CO₂ standards.

Conceptually, the OMEGA model begins by determining the specific CO₂ emission standard applicable for each manufacturer and its vehicle class (i.e., car or truck). Since the final rule allows for averaging across a manufacturer's cars and trucks, the model determines the CO₂ emission standard applicable to each manufacturer's car and truck sales from the two sets of coefficients describing the piecewise linear standard functions for cars and trucks in the inputs, and creates a combined car-truck standard. This combined standard considers the difference in lifetime VMT of cars and trucks, as indicated in the final regulations which govern credit trading between these two vehicle classes. For both the 2011 CAFE and 2016 CO₂ standards, these standards are a function of each manufacturer's sales of cars and trucks and their footprint values. When evaluating the 2011 MY CAFE standards, the car-truck trading was limited to 1.2 mpg. When evaluating the final CO₂ standards, the OMEGA model was run only for MY 2016. OMEGA is designed to evaluate technology addition over a complete redesign cycle and 2016 represents the final year of a redesign cycle starting with the first year of the final CO₂ standards, 2012. Estimates of the technology and cost for the interim model years are developed from the model projections made for 2016. This process is discussed in Chapter 6 of EPA's RIA to this final rule. When evaluating the 2016 standards using the OMEGA model, the final CO₂ standard which manufacturers will otherwise have to meet to account for the anticipated level of A/C credits generated was adjusted. On an industry wide basis, the projection shows that manufacturers will generate 11 g/mi of A/C credit in 2016. Thus, the 2016 CO₂ target for the fleet evaluated using OMEGA was 261 g/mi instead of 250 g/mi.

As noted above, EPA estimated separately the cost of the improved A/C systems required to generate the 11 g/mi credit. This is consistent with our final A/C credit procedures, which will grant manufacturers A/C credits based on their total use of improved A/C systems, and not on the increased use of such systems relative to some base model year fleet. Some manufacturers may already be using improved A/C technology. However, this represents a small fraction of current vehicle sales. To the degree that such systems are already being used, EPA is over-estimating both the cost and benefit of the addition of improved A/C technology relative to the true reference fleet to a small degree.

The model then works with one manufacturer at a time to add technologies until that manufacturer meets its applicable standard. The OMEGA model can utilize several approaches to determining the order in which vehicles receive technologies. For this analysis, EPA used a “manufacturer-based net cost-effectiveness factor” to rank the technology packages in the order in which a manufacturer is likely to apply them. Conceptually, this approach estimates the cost of adding the technology from the manufacturer's perspective and divides it by the mass of CO₂ the technology will reduce. One component of the cost of adding a technology is its production cost, as discussed above. However, it is expected that new vehicle purchasers value improved fuel economy since it reduces the cost of operating the vehicle. Typical vehicle purchasers are assumed to value the fuel savings accrued over the period of time which they will own the vehicle, which is estimated to be roughly five years. It is also assumed that consumers discount these savings at the same rate as that used in the rest of the analysis (3 or 7 percent). Any residual value of the additional technology which might remain when the vehicle is sold is not considered. The CO₂ emission reduction is the change in CO₂ emissions multiplied by the percentage of vehicles surviving after each year of use multiplied by the annual miles travelled by age, again discounted to the year of vehicle purchase.

Given this definition, the higher priority technologies are those with the lowest manufacturer-based net cost-effectiveness value (relatively low technology cost or high fuel savings leads to lower values). Because the order of technology application is set for each vehicle, the model uses the manufacturer-based net cost-effectiveness primarily to decide which vehicle receives the next technology addition. Initially, technology package #1 is the only one available to any particular vehicle. However, as soon as a vehicle receives technology package #1, the model considers the manufacturer-based net cost-effectiveness of technology package #2 for that vehicle and so on. In general terms, the equation describing the calculation of manufacturer-based cost effectiveness is as follows:

Image Not Available

Where

ManufCostEff = Manufacturer-Based Cost Effectiveness (in dollars per kilogram CO₂),

TechCost = Marked up cost of the technology (dollars),

PP = Payback period, or the number of years of vehicle use over which consumers value fuel savings when evaluating the value of a new vehicle at time of purchase,

dFS_i = Difference in fuel consumption due to the addition of technology times fuel price in year i,

dCO₂ = Difference in CO₂ emissions due to the addition of technology,

VMTi = product of annual VMT for a vehicle of age i and the percentage of vehicles of age i still on the road, and

1- Gap = Ratio of onroad fuel economy to two-cycle (FTP/HFET) fuel economy.

The OMEGA model does not currently allow for the VMT used in determining the various technology ranking factors to be a function of the rebound factor. If the user believed that the consideration of rebound VMT was important, they could increase their estimate of the payback period to simulate the impact of the rebound VMT.

EPA describes the technology ranking methodology and manufacturer-based cost effectiveness metric in greater detail in a technical memo to the Docket for this final rule (Docket EPA-HQ-OAR-2009-0472).

When calculating the fuel savings, the full retail price of fuel, including taxes is used. While taxes are not generally included when calculating the cost or benefits of a regulation, the net cost component of the manufacturer-based net cost-effectiveness equation is not a measure of the social cost of this final rule, but a measure of the private cost, (i.e., a measure of the vehicle purchaser's willingness to pay more for a vehicle with higher fuel efficiency). Since vehicle operators pay the full price of fuel, including taxes, they value fuel costs or savings at this level, and the manufacturers will consider this when choosing among the technology options.

This definition of manufacturer-based net cost-effectiveness ignores any change in the residual value of the vehicle due to the additional technology when the vehicle is five years old. As discussed in Chapter 1 of the RIA, based on historic used car pricing, applicable sales taxes, and insurance, vehicles are worth roughly 23% of their original cost after five years, discounted to year of vehicle purchase at 7% per annum. It is reasonable to estimate that the added technology to improve CO₂ level and fuel economy will retain this same percentage of value when the vehicle is five years old. However, it is less clear whether first purchasers, and thus, manufacturers consider this residual value when ranking technologies and making vehicle purchases, respectively. For this final rule, this factor was not included in our determination of manufacturer-based net cost-effectiveness in the analyses performed in support of this final rule.

The values of manufacturer-based net cost-effectiveness for specific technologies will vary from vehicle to vehicle, often substantially. This occurs for three reasons. First, both the cost and fuel-saving component cost, ownership fuel-savings, and lifetime CO₂ effectiveness of a specific technology all vary by the type of vehicle or engine to which it is being applied (e.g., small car versus large truck, or 4-cylinder versus 8-cylinder engine). Second, the effectiveness of a specific technology often depends on the presence of other technologies already being used on the vehicle (i.e., the dis-synergies). Third, the absolute fuel savings and CO₂ reduction of a percentage on incremental reduction in fuel consumption depends on the CO₂ level of the vehicle prior to adding the technology. Chapter 1 of the RIA of this final rule contains further detail on the values of manufacturer-based net cost-effectiveness for the various technology packages.

6. Why are the final CO₂ standards feasible?

The finding that the final standards are technically feasible is based primarily on two factors. One is the level of technology needed to meet the final standards. The other is the cost of this technology. The focus is on the final standards for 2016, as this is the most stringent standard and requires the most extensive use of technology.

With respect to the level of technology required to meet the standards, EPA established technology penetration caps. As described in Section III.D.4, EPA used two constraints to limit the model's application of technology by manufacturer. The first was the application of common fuel economy enablers such as low rolling resistance tires and transmission logic changes. These were allowed to be used on all vehicles and hence had no penetration cap. The second constraint was applied to most other technologies and limited their application to 85% with the exception of the most advanced technologies (e.g., power-split hybrid and 2-mode hybrid) and diesel, whose application was limited to 15%.

EPA used the OMEGA model to project the technology (and resultant cost) required for manufacturers to meet the current 2011 MY CAFE standards and the final 2016 MY CO₂ emission standards. Both sets of standards were evaluated using the OMEGA model. The 2011 MY CAFE standards were applied to cars and trucks separately with the transfer of credits from one category to the other allowed up to an increase in fuel economy of 1.0 mpg as allowed under the applicable MY 2011 CAFE regulations. Chrysler, Ford and General Motors are assumed to utilize FFV credits up to the maximum of 1.2 mpg for both their car and truck sales. Nissan is assumed to utilize FFV credits up to the maximum of 1.2 mpg for only their truck sales. The use of any banked credits from previous model years was not considered. The modification of the reference fleet to comply with the 2011 CAFE standards through the application of technology by the OMEGA model is the final step in creating the final reference fleet. This final reference fleet forms the basis for comparison for the model year 2016 standards.

Table III.D.6-1 shows the usage level of selected technologies in the 2008 vehicles coupled with 2016 sales prior to projecting their compliance with the 2011 MY CAFE standards. These technologies include converting port fuel-injected gasoline engines to direct injection (GDI), adding the ability to deactivate certain engine cylinders during low load operation to overhead cam engines (OHC-DEAC), adding a turbocharger and downsizing the engine (Turbo), diesel engine technology, increasing the number of transmission speeds to 6, or converting automatic transmissions to dual-clutch automated manual transmissions (Dual-Clutch Trans), adding 42 volt start-stop capability (Start-Stop), and converting a vehicle to an intermediate or strong hybrid design. This last category includes three current hybrid designs: Integrated motor assist (IMA), power-split (PS), 2-mode hybrids and electric vehicles.

As can be seen, all of these technologies were already being used on some 2008 MY vehicles, with the exception of direct injection gasoline engines with either cylinder deactivation or turbocharging and downsizing. Transmissions with more gearsets were the most prevalent, with some manufacturers (e.g., BMW, Suzuki) using them on essentially all of their vehicles. Both Daimler and VW equip many of their vehicles with automated manual transmissions, while VW makes extensive use of direct injection gasoline engine technology. Toyota has converted a significant percentage of its 2008 vehicles to strong hybrid design.

Table III.D.6-2 shows the usage level of the same technologies in the reference case fleet after projecting their compliance with the 2011 MY CAFE standards. Except for mass reduction, the figures shown represent the percentages of each manufacturer's sales which are projected to be equipped with the indicated technology. For mass reduction, the overall mass reduction projected for that manufacturer's sales is also shown. The last row in Table III.D.6-2 shows the increase in projected technology penetration due to compliance with the 2011 MY CAFE standards. The results of DOT's Volpe modeling were used to project that all manufacturers would comply with the 2011 MY standards in 2016 without the need to pay fines, with one exception. This exception was Porsche in the case of their car fleet. When projecting Porsche's compliance with the 2011 MY CAFE standard for cars, NHTSA projected that Porsche would achieve a CO₂ emission level of 304.3 g/mi instead of the required 284.8 g/mi level (29.2 mpg instead of 31.2 mpg), and pay fines in lieu of further control.

As can be seen, the 2011 MY CAFE standards, when evaluated on an industry wide basis, require only a modest increase in the use of these technologies. The projected MY 2016 fraction of automatic transmission with more gearsets actually decreases slightly due to conversion of these units to more efficient designs such as automated manual transmissions and hybrids. However, the impact of the 2011 MY CAFE standards is much greater on selected manufacturers, particularly BMW, Daimler, Porsche, Tata (Jaguar/Land Rover) and VW. All of these manufacturers are projected to increase their use of direct injection gasoline engine technology, advanced transmission technology, and start-stop technology. It should be noted that these manufacturers have traditionally paid fines under the CAFE program. However, with higher fuel prices and the lower cost mature technology projected to be available by 2016, these manufacturers would likely find it in their best interest to improve their fuel economy levels instead of continuing to pay fines (again with the exception of Porsche cars). While not shown, no gasoline engines were projected to be converted to diesel technology and no hybrid vehicles were projected. Most manufacturers do not require the level of CO₂ emission control associated with either of these technologies. The few manufacturers that would were projected to choose to pay CAFE fines in 2011 in lieu of adding diesel or hybrid technologies.

This 2008 baseline fleet, modified to meet 2011 standards, becomes our “reference” case. See Section II.B above. This is the fleet against which the final 2016 standards are compared. Thus, it is also the fleet that is assumed to exist in the absence of this rule. No air conditioning improvements are assumed for model year 2011 vehicles. The average CO₂ emission levels of this reference fleet vary slightly from 2012-2016 due to small changes in the vehicle sales by market segments and manufacturer. CO₂ emissions from cars range from 282-284 g/mi, while those from trucks range from 382-384 g/mi. CO₂ emissions from the combined fleet range from 316-320. These estimates are described in greater detail in Section 5.3.2.2 of the EPA RIA.

Conceptually, both EPA and NHTSA perform the same projection in order to develop their respective reference fleets. However, because the two agencies use two different models to modify the baseline fleet to meet the 2011 CAFE standards, the projected technology that could be added will be slightly different. The differences, however, are relatively small since most manufacturers only require modest addition of technology to meet the 2011 CAFE standards.

EPA then used the OMEGA model once again to project the level of technology needed to meet the final 2016 CO₂ emission standards. Using the results of the OMEGA model, every manufacturer was projected to be able to meet the final 2016 standards with the technology described above except for four: BMW, VW, Porsche and Tata (which is comprised of Jaguar and Land Rover vehicles in the U.S. fleet). For these manufacturers, the results presented below are those with the fully allowable application of technology available in EPA's OMEGA modeling analysis and not for the technology projected to enable compliance with the final standards. Described below are a number of potential feasible solutions for how these companies can achieve compliance. The overall level of technology needed to meet the final 2016 standards is shown in Table III.D.6-3. As discussed above, all manufacturers are projected to improve the air conditioning systems on 85% of their 2016 sales.

Table III.D.6-4 shows the 2016 standards, as well as the achieved CO₂ emission levels for the five manufacturers which are not able to meet these standards under the premises of our modeling. It should be noted that the two sets of combined emission levels shown in Table III.D.6-4 are based on sales weighting car and truck emission levels.

As can be seen, BMW and Daimler have the smallest shortfalls, 5-6 g/mi, while Porsche has the largest, 40 g/mi.

On an industry average basis, the technology penetrations are very similar to those projected in the proposal. There is a slight shift from the use of cylinder deactivation to the two advanced transmission technologies. This is due to the fact that the estimated costs for these three technologies have been updated, and thus, their relative cost effectiveness when applied to specific vehicles have also shifted. The reader is referred to Section II.E of this preamble as well as Chapter 3 of the Joint TSD for a detailed description of the cost estimates supporting this final rule and to the RIA for a description of the selection of technology packages for specific vehicle types. The other technologies shown in Table III.D.6-4 changed by 2 percent or less between the proposal and this final rule.

As can be seen, the overall average reduction in vehicle weight is projected to be 4 percent. This reduction varies across the two vehicle classes and vehicle base weight. For cars below 2,950 pounds curb weight, the average reduction is 2.8 percent (75 pounds), while the average was 4.3 percent (153 pounds) for cars above 2,950 curb weight. For trucks below 3,850 pounds curb weight, the average reduction is 4.7 percent (163 pounds), while it was 5.1 percent (240 pounds) for trucks above 3,850 curb weight. Splitting trucks at a higher weight, for trucks below 5,000 pounds curb weight, the average reduction is 4.4 percent (186 pounds), while it was 7.0 percent (376 pounds) for trucks above 5,000 curb weight.

The levels of requisite technologies differ significantly across the various manufacturers. Therefore, several analyses were performed to ascertain the cause. Because the baseline case fleet consists of 2008 MY vehicle designs, these analyses were focused on these vehicles, their technology and their CO₂ emission levels.

Comparing CO₂ emissions across manufacturers is not a simple task. In addition to widely varying vehicle styles, designs, and sizes, manufacturers have implemented fuel efficient technologies to varying degrees, as indicated in Table III.D.6-1. The projected levels of requisite technology to enable compliance with the final 2016 standards shown in Table III.D.6-3 account for two of the major factors which can affect CO₂ emissions (1) Level of technology already being utilized and (2) vehicle size, as represented by footprint.

For example, the fuel economy of a manufacturer's 2008 vehicles may be relatively high because of the use of advanced technologies. This is the case with Toyota's high sales of their Prius hybrid. However, the presence of this technology in a 2008 vehicle eliminates the ability to significantly reduce CO₂ further through the use of this technology. In the extreme, if a manufacturer were to hybridize a high level of its sales in 2016, it does not matter whether this technology was present in 2008 or whether it would be added in order to comply with the standards. The final level of hybrid technology would be the same. Thus, the level at which technology is present in 2008 vehicles does not explain the difference in requisite technology levels shown in Table III.D.6-3.

Similarly, the final CO₂ emission standards adjust the required CO₂ level according to a vehicle's footprint, requiring lower absolute emission levels from smaller vehicles. Thus, just because a manufacturer produces larger vehicles than another manufacturer does not explain the differences seen in Table III.D.6-3.

In order to remove these two factors from our comparison, the EPA lumped parameter model described above was used to estimate the degree to which technology present on each 2008 MY vehicle in our reference fleet was improving fuel efficiency. The effect of this technology was removed and each vehicle's CO₂ emissions were estimated as if it utilized no additional fuel efficiency technology beyond the baseline. The differences in vehicle size were accounted for by determining the difference between the sales-weighted average of each manufacturer's “no technology” CO₂ levels to their required CO₂ emission level under the final 2016 standards. The industry-wide difference was subtracted from each manufacturer's value to highlight which manufacturers had lower and higher than average “no technology” emissions. The results are shown in Figure III.D.6-1.

As can be seen in Table III.D.6-3 the manufacturers projected to require the greatest levels of technology also show the highest offsets relative to the industry. The greatest offset shown in Figure III.D.6-1 is for Tata's trucks (Land Rover). These vehicles are estimated to have 100 g/mi greater CO₂ emissions than the average 2008 MY truck after accounting for differences in the use of fuel saving technology and footprint. The lowest adjustment is for Subaru's trucks, which have 50 g/mi CO₂ lower emissions than the average truck.

While this comparison confirms the differences in the technology penetrations shown in Table III.D.6-3, it does not yet explain why these differences exist. Two well-known factors affecting vehicle fuel efficiency are vehicle weight and acceleration performance (henceforth referred to as “performance”). The footprint-based form of the final CO₂ standard accounts for most of the difference in vehicle weight seen in the 2008 MY fleet. However, even at the same footprint, vehicles can have varying weights. Higher performing vehicles also tend to have higher CO₂ emissions over the two-cycle fuel economy test procedure. So manufacturers with higher average performance levels will tend to have higher average CO₂ emissions for any given footprint. This variability at any given footprint contributes to much of the scatter in the data (shown for example on plots like Figures II.C.1-3 through II.C.1-6).

We developed a methodology to assess the impact of these two factors on each manufacturer's projected compliance with the 2016 standards. First, we had to remove (or isolate) the effect of CO₂ control technology already being employed on 2008 vehicles. As described above, 2008 vehicles exhibit a wide range of control technology and leaving these impacts in place would confound the assessment of performance and weight on CO₂ emissions. Thus, the first step was to estimate each vehicle's “no technology” CO₂ emissions. To do this, we used the EPA lumped parameter model (described in the TSD) to estimate the overall percentage reduction in CO₂ emissions associated with technology already on the vehicle and then backed out this effect mathematically. Second, we performed a least-square linear regression of these no technology CO₂ levels against curb weight and the ratio of rated engine horsepower to curb weight simultaneously. The ratio of rated engine horsepower to curb weight is a good surrogate for acceleration performance and the data is available for all vehicles, whereas the zero to sixty time is not. Both factors were found to be statistically significant at the 95% confidence level. Together, they explained over 80% of the variability in vehicles' CO₂ emissions for cars and over 70% for trucks. Third, we determined the sales-weighted average curb weight per footprint for cars and trucks, respectively, for the fleet as a whole. We also determined the sales-weighted average of the ratio of rated engine horsepower to curb weight for cars and trucks, respectively, for the fleet as a whole. Fourth, we adjusted each vehicle's “no technology” CO₂ emissions to eliminate the degree to which the vehicle had higher or lower acceleration performance or curb weight per footprint relative to the car or truck fleet as a whole. For example, if a car's ratio of horsepower to weight was 0.007 and the average ratio for all cars was 0.006, then the vehicle's “no technology” CO₂ emission level was reduced by the difference between these two values (0.001) times the impact of the ratio of horsepower to weight on car CO₂ emissions from the above linear regression. Finally, we substituted these performance and weight adjusted CO₂ emission levels for the original, “no technology” CO₂ emission levels shown in Figure III.D.6-1. The results are shown in Figure III.D.6-2.

First, note that the scale in Figure III.D.6-2 is much smaller by a factor of 3 than that in Figure III.D.6-1. In other words, accounting for differences in vehicle weight (at constant footprint) and performance dramatically reduces the variability among the manufacturers' CO₂ emissions. Most of the manufacturers with high positive offsets in Figure III.D.6-1 now show low or negative offsets. For example, BMW's and VW's trucks show very low CO₂ emissions. Tata's emissions are very close to the industry average. Daimler's vehicles are no more than 10 g/mi above the average for the industry. This analysis indicates that the primary reasons for the differences in technology penetrations shown for the various manufacturers in Table III.D.6-3 are weight and acceleration performance. EPA has not determined why some manufacturers' vehicle weight is relatively high for its footprint value, or whether this weight provides additional utility for the consumer. Performance is more straightforward. Some consumers desire high-acceleration performance and some manufacturers orient their sales towards these consumers. However, the cost in terms of CO₂ emissions is clear. Manufacturers producing relatively heavy or high performance vehicles presently (with concomitant increased CO₂ emissions) will require greater levels of technology in order to meet the final CO₂ standards in 2016.

As can be seen from Table III.D.6-3 above, widespread use of several technologies is projected due to the final standards. The vast majority of engines are projected to be converted to direct injection, with some of these engines including cylinder deactivation or turbocharging and downsizing. More than 60 percent of all transmissions are projected to be either 6+ speed automatic transmissions or dual-clutch automated manual transmissions. More than one-third of the fleet is projected to be equipped with 42 volt start-stop capability. This technology was not utilized in 2008 vehicles, but as discussed above, promises significant fuel efficiency improvement at a moderate cost.

In their comments, Porsche stated that their vehicles have twice the power-to-weight ratio as the fleet average and that their vehicles presently have a high degree of technology penetration, which allows them to meet the 2009 CAFE standards. Porsche also commented that the 2016 standards are not feasible for their firm, in part due to the high level of technologies already present in their vehicles and due to their “very long production life cycles”. BMW in their comments stated that their vehicles are “feature-dense” thus “requiring additional efforts to comply” with future standards. Ferrari, in their comments, states that the standards are not feasible for high-performance sports cars without compromising on their “distinctiveness”. They also state that because they already have many technologies on the vehicles, “there are limited possibilities for further improvements.” Finally Ferrari states that smaller volume manufacturers have higher costs “because they can be distributed over very limited production volumes”, and they have longer product lifecycles. The latter view was also shared by Lotus. These comments will be addressed below, but are cited here as supporting the conclusions from the above analysis that high-performance and feature-dense vehicles have a greater challenge meeting the 2016 standards. In general, other manufacturers covering the rest of the fleet and other commenters agreed with EPA's analysis in the proposal of projected technology usage, and supported the view that the 2016 model year standards were feasible in the lead-time provided.

In response to the comments above, EPA foresees no significant technical or engineering issues with the projected deployment of these technologies across the fleet by MY 2016, with their incorporation being folded into the vehicle redesign process (with the exception of some of the small volume manufacturers). All of these technologies are commercially available now. The automotive industry has already begun to convert its port fuel-injected gasoline engines to direct injection. Cylinder deactivation and turbocharging technologies are already commercially available. As indicated in Table III.D.6-1, high-speed transmissions are already widely used. However, while more common in Europe, automated manual transmissions are not currently used extensively in the U.S. Widespread use of this technology would require significant capital investment but does not present any significant technical or engineering issues. Start-stop systems based on a 42-volt architecture also represent a challenge because of the complications involved in a changeover to a higher voltage electrical architecture. However, with appropriate capital investments (which are captured in the EPA estimated costs), these technology penetration rates are achievable within the timeframe of this rule. While most manufacturers have some plans for these systems, our projections indicate that their use may exceed 35% of sales, with some manufacturers projected to use higher levels.

Most manufacturers are not projected to hybridize any vehicles to comply with the final standards. The hybrids shown for Toyota are projected to be sold even in the absence of the final standards. However the relatively high hybrid penetrations (14-15%) projected for BMW, Daimler, Porsche, Tata and Volkswagen deserve further discussion. These manufacturers are all projected by the OMEGA model to utilize the maximum application of full hybrids allowed by our model in this timeframe, which is 15 percent.

As discussed in the EPA RIA, a maximum 2016 technology penetration rate of 85% is projected for the vast majority of available technologies, however, for full hybrid systems the projection shows that given the available lead-time full hybrids can only be applied to approximately 15% of a manufacturer's fleet. This number of course can vary by manufacturer. Hybrids are a relatively costly technology option which requires significant changes to a vehicle's powertrain design, and EPA estimates that manufacturers will require a significant amount of lead time and capital investment to introduce this technology into the market in very large numbers. Thus the EPA captures this significant change in production facilities with a lower penetration cap. A more thorough discussion of lead time limitations can be found below and in Section III.B.5.

While the hybridization levels of BMW, Daimler, Porsche, Tata and Volkswagen are relatively high, the sales levels of these five manufacturers are relatively low. Thus, industry-wide, hybridization reaches only 4 percent, compared with 3 percent in the reference case. This 4 percent level is believed to be well within the capability of the hybrid component industry by 2016. Thus, the primary challenge for these five companies would be at the manufacturer level, redesigning a relatively large percentage of sales to include hybrid technology. The final TLAAS provisions will provide significant needed lead time to these manufacturers for pre-2016 compliance, since all qualified companies are able to take advantage of these provisions.

By 2016, it is likely that these manufacturers would also be able to change vehicle characteristics which currently cause their vehicles to emit much more CO₂ than similar sized vehicles produced by other manufacturers. These factors may include changes in model mix, further mass reduction, electric and/or plug-in hybrid vehicles as well as technologies that may not be included in our packages. Also, companies may have technology penetration rates of less costly technologies (listed in the above tables) greater than 85%, and they may also be able to apply hybrid technology to more than 15 percent of their fleet (while the 15% cap on the application of hybrid technology is reasonable for the industry as a whole, higher percentages are certainly possible for individual manufacturers, particularly those with small volumes). For example, a switch to a low GWP alternative refrigerant in a large fraction of a fleet can replace many other much more costly technologies, but this option is not captured in the modeling. In addition, these manufacturers can also take advantage of flexibilities, such as early credits for air conditioning and trading with other manufacturers.

EPA believes it is likely that there will be certain high volume manufacturers that will earn a significant amount of early GHG credits starting in 2010 that would expire 5 years later, by 2015, unused. It is possible that these manufacturers may be willing to sell these credits to manufacturers with whom there is little or no direct competition. Furthermore, a large number of manufacturers have also stated publicly that they support the 2016 standards. The following companies have all submitted letters in support of the national program, including the 2016 MY levels discussed above: BMW, Chrysler, Daimler, Ford, GM, Nissan, Honda, Mazda, Toyota, and Volkswagen. This supports the view that the emissions reductions needed to achieve the standards are technically and economically feasible for all these companies, and that EPA's projection of model year 2016 non-compliance for BMW, Daimler, and Volkswagen is based on an inability of our model at this time to fully account for the full flexibilities of the EPA program as well as the potentially unique technology approaches or new product offerings which these manufacturers are likely to employ.

In addition, manufacturers do not need to apply technology exactly according to our projections. Our projections simply indicate one path which would achieve compliance. Those manufacturers whose vehicles are heavier (feature dense) and higher performing than average in particular have additional options to facilitate compliance and reduce their technological burden closer to the industry average. These options include decreasing the mass of the vehicles and/or decreasing the power output of the engines. Finally, EPA allows compliance to be shown through the use of emission credits obtained from other manufacturers. Especially for the lower volume sales of some manufacturers that could be one component of an effective compliance strategy, reducing the technology that needs to be employed on their vehicles.

For light-duty cars and trucks, manufacturers have available to them a range of technologies that are currently commercially available and can feasibly be employed in their vehicles by MY 2016. Our modeling projects widespread use of these technologies as a technologically feasible approach to complying with the final standards. Comments from the manufacturers provided broad support for this conclusion. A limited number of commenters presented specific concerns about their technology opportunities, and EPA has described above (and elsewhere in the rule) the paths available for them to comply.

In sum, EPA believes that the emissions reductions called for by the final standards are technologically feasible, based on projections of widespread use of commercially available technology, as well as use by some manufacturers of other technology approaches and compliance flexibilities not fully reflected in our modeling.

EPA also projected the cost associated with these projections of technology penetration. Table III.D.6-4 shows the cost of technology in order for manufacturers to comply with the 2011 MY CAFE standards, as well as those associated with the final 2016 CO₂ emission standards. The latter costs are incremental to those associated with the 2011 MY standards and also include $60 per vehicle, on average, for the cost of projected use of improved air-conditioning systems.

As can be seen, the industry average cost of complying with the 2011 MY CAFE standards is quite low, $89 per vehicle. This cost is $11 per vehicle higher than that projected in the NPRM. This change is very small and is due to several factors, mainly changes in the projected sales of each manufacturer's specific vehicles, and changes in estimated technology costs. Similar to the costs projected in the NPRM, the range of costs across manufacturers is quite large. Honda, Mazda and Toyota are projected to face no cost. In contrast, Mitsubishi, Porsche, Tata and Volkswagen face costs of at least $272 per vehicle. As described above, three of these last four manufacturers (all but Mitsubishi) face high costs to meet even the 2011 MY CAFE standards due to either their vehicles' weight per unit footprint or performance. Porsche would have been projected to face lower costs in 2016 if they were not expected to pay CAFE fines in 2011.

As shown in the last row of Table III.D.6-4, the average cost of technology to meet the final 2016 standards for cars and trucks combined relative to the 2011 MY CAFE standards is $948 per vehicle. This is $103 lower than that projected in the NPRM, due primarily to lower technology cost projections for the final rule compared to the NPRM for certain technologies. (See Chapter 1 of the Joint TSD for a detailed description of how our technology costs for the final rule differ from those used in the NPRM). As was the case in the NPRM, Table III.D.6-4 shows that the average cost for cars would be slightly lower than that for trucks. Toyota and Honda show projected costs significantly below the average, while BMW, Porsche, Tata and Volkswagen show significantly higher costs. On average, the $948 per vehicle cost is significant, representing 3.4 percent of the total cost of a new vehicle. However, as discussed below, the fuel savings associated with the final standards exceed this cost significantly. In general, commenters supported EPA's cost projections, as discussed in Section II.

While the CO₂ emission compliance modeling using the OMEGA model focused on the final 2016 MY standards, the final standards for 2012-2015 are also feasible. As discussed above, manufacturers develop their future vehicle designs with several model years in view. Generally, the technology estimated above for 2016 MY vehicles represents the technology which would be added to those vehicles which are being redesigned in 2012-2015. The final CO₂ standards for 2012-2016 reduce CO₂ emissions at a fairly steady rate. Thus, manufacturers which redesign their vehicles at a fairly steady rate will automatically comply with the interim standard as they plan for compliance in 2016.

Manufacturers which redesign much fewer than 20% of their sales in the early years of the final program would face a more difficult challenge, as simply implementing the “2016 MY” technology as vehicles are redesigned may not enable compliance in the early years. However, even in this case, manufacturers would have several options to enable compliance. One, they could utilize the debit carry-forward provisions described above. This may be sufficient alone to enable compliance through the 2012-2016 MY time period, if their redesign schedule exceeds 20% per year prior to 2016. If not, at some point, the manufacturer might need to increase their use of technology beyond that projected above in order to generate the credits necessary to balance the accrued debits. For most manufacturers representing the vast majority of U.S. sales, this would simply mean extending the same technology to a greater percentage of sales. The added cost of this in the later years of the program would be balanced by lower costs in the earlier years. Two, the manufacture could take advantage of the many optional credit generation provisions contained in this final rule, including early-credit generation for model years 2009-2011, credits for advanced technology vehicles, and credits for the application of technology which result in off-cycle GHG reductions. Finally, the manufacturer could buy credits from another manufacturer. As indicated above, several manufacturers are projected to require less stringent technology than the average. These manufacturers would be in a position to provide credits at a reasonable technology cost. Thus, EPA believes the final standards for 2012-2016 would be feasible. Further discussion of the technical feasibility of the interim year standards, including for smaller volume manufacturers can be found in Section III.B, in the discussion on the Temporary Leadtime Allowance Alternative Standards.

7. What other fleet-wide CO₂ levels were considered?

Two alternative sets of CO₂ standards were considered. One set would reduce CO₂ emissions at a rate of 4 percent per year. The second set would reduce CO₂ emissions at a rate of 6 percent per year. The analysis of these standards followed the exact same process as described above for the final standards. The only difference was the level of CO₂ emission standards. The footprint-based standard coefficients of the car and truck curves for these two alternative control scenarios were discussed above. The resultant projected CO₂ standards in 2016 for each manufacturer under these two alternative scenarios and under the final rule are shown in Table III.D.7-1.

Tables III.D.7-2 and III.D.7-3 show the technology penetration levels for the 4 percent per year and 6 percent per year standards in 2016.

With respect to the 4 percent per year standards, the levels of requisite control technology are lower than those under the final standards, as would be expected. Industry-wide, the largest decreases were a 7 percent decrease in use of gasoline direct injection engines, a 4 percent decrease in the use of dual clutch transmissions, and a 2 percent decrease in the application of start-stop technology. On a manufacturer specific basis, the most significant decreases were a 10 percent or larger decrease in the use of stop-start technology for Honda, Kia, Mitsubishi and Suzuki and a 12 percent drop in turbocharger use for Mitsubishi. These are relatively small changes and are due to the fact that the 4 percent per year standards only require 4 g/mi CO₂ less control than the final standards in 2016. Porsche, Tata and Volkswagen continue to be unable to comply with the CO₂ standards in 2016, even under the 4 percent per year standard scenario. BMW just complied under this scenario, so its costs and technology penetrations are the same as under the final standards.

With respect to the 6 percent per year standards, the levels of requisite control technology increased substantially relative to those under the final standards, as again would be expected. Industry-wide, the largest increase was a 25 percent increase in the application of start-stop technology and 13-17 percent increases in the use of gasoline direct injection engines, turbocharging and dual clutch transmissions. On a manufacturer specific basis, the most significant increases were a 10 percent increase in hybrid penetration for Ford and Mitsubishi. These are more significant changes and are due to the fact that the 6 percent per year standards require 20 g/mi CO₂ more control than the final standards in 2016. Our projections for BMW, Porsche, Tata and Volkswagen continue to show they are unable to comply with the CO₂ standards in 2016, so our projections for these manufacturers do not differ relative to the final standards, though the amount of short-fall for each firm increases significantly, by an additional 20 g/mi CO₂ per firm. However, Ford and Mitsubishi join this list as can be seen from Figure III.D.6-2. The CO₂ emissions from Ford's cars are very similar to those of the industry when adjusted for technology, weight and performance. However, their trucks emit more than 25% more CO₂ per mile than the industry average. It is possible that addressing this issue would resolve their difficulty in complying with the 6 percent per year scenario. Both Mitsubishi's cars and truck emit roughly 10% more than the industry average vehicles after adjusting for technology, weight and performance. Again, addressing this issue could resolve their difficulty in complying with the 6 percent per year scenario. Five manufacturers are projected to need to increase their use of start-stop technology by at least 30 percent.

Table III.D.7-4 shows the projected cost of the two alternative sets of standards.