
[Federal Register: May 2, 2008 (Volume 73, Number 86)]
[Proposed Rules]               
[Page 24351-24487]
From the Federal Register Online via GPO Access [wais.access.gpo.gov]
[DOCID:fr02my08-28]                         
 

[[Page 24351]]

-----------------------------------------------------------------------

Part II





Department of Transportation





-----------------------------------------------------------------------



 National Highway Traffic Safety Administration



-----------------------------------------------------------------------



49 CFR Parts 523, 531, 533, 534, 536 and 537



Average Fuel Economy Standards, Passenger Cars and Light Trucks; Model 
Years 2011-2015; Proposed Rule


[[Page 24352]]


-----------------------------------------------------------------------

DEPARTMENT OF TRANSPORTATION

National Highway Traffic Safety Administration

49 CFR Parts 523, 531, 533, 534, 536 and 537

[Docket No. NHTSA-2008-0089]
RIN 2127-AK29

 
Average Fuel Economy Standards, Passenger Cars and Light Trucks; 
Model Years 2011-2015

AGENCY: National Highway Traffic Safety Administration (NHTSA), 
Department of Transportation (DOT).

ACTION: Notice of Proposed Rulemaking (NPRM).

-----------------------------------------------------------------------

SUMMARY: This document proposes substantial increases in the Corporate 
Average Fuel Economy (CAFE) standards for passenger cars and light 
trucks that would enhance energy security by improving fuel economy. 
Since the carbon dioxide (CO2) emitted from the tailpipes of 
new motor vehicles is the natural by-product of the combustion of fuel, 
the increased standards would also address climate change by reducing 
tailpipe emissions of CO2. Those emissions represent 97 
percent of the total greenhouse gas emissions from motor vehicles. 
Implementation of the new standards would dramatically add to the 
billions of barrels of fuel already saved since the beginning of the 
CAFE program in 1975.

DATES: Comments must be received on or before July 1, 2008.

ADDRESSES: You may submit comments to the docket number identified in 
the heading of this document by any of the following methods:
     Federal eRulemaking Portal: Go to http://
www.regulations.gov. Follow the online instructions for submitting 
comments.
     Mail: 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.
     Hand Delivery or Courier: West Building Ground Floor, Room 
W12-140, 1200 New Jersey Avenue, SE., between 9 a.m. and 5 p.m. Eastern 
Time, Monday through Friday, except Federal holidays.
     Fax: (202) 493-2251.
    Regardless of how you submit your comments, you should mention the 
docket number of this document.
    You may call the Docket Management Facility at 202-366-9826.
    Instructions: For detailed instructions on submitting comments and 
additional information on the rulemaking process, see the Public 
Participation heading of the Supplementary Information section of this 
document. Note that all comments received will be posted without change 
to http://www.regulations.gov, including any personal information 
provided.
    Privacy Act: Please see the Privacy Act heading under Rulemaking 
Analyses and Notices.

FOR FURTHER INFORMATION CONTACT: For policy and technical issues: Ms. 
Julie Abraham or Mr. Peter Feather, Office of Rulemaking, National 
Highway Traffic Safety Administration, 1200 New Jersey Avenue, SE., 
Washington, DC 20590. Telephone: Ms. Abraham (202) 366-1455; Mr. 
Feather (202) 366-0846.
    For legal issues: Mr. Stephen Wood or Ms. Rebecca Schade, Office of 
the Chief Counsel, National Highway Traffic Safety Administration, 1200 
New Jersey Avenue, SE., Washington, DC 20590. Telephone: (202) 366-
2992.

SUPPLEMENTARY INFORMATION:

Table of Contents

I. Executive overview
    A. Summary
    B. Energy Independence and Security Act of 2007
    C. Proposal
    1. Standards
    a. Stringency
    b. Benefits
    c. Costs
    d. Flexibilities
    2. Credits
II. Background
    A. Contribution of fuel economy improvements to addressing 
energy independence and security and climate change
    1. Relationship between fuel economy and CO2 tailpipe 
emissions
    2. Fuel economy improvements/CO2 tailpipe emission 
reductions since 1975
    B. Chronology of events since the National Academy of Sciences 
called for reforming and increasing CAFE standards
    1. National Academy of Sciences CAFE report (February 2002)
    a. Significantly increasing CAFE standards without reforming 
them would adversely affect safety
    b. Environmental and other externalities justify increasing the 
CAFE standards
    2. Final rule establishing reformed (attribute-based) CAFE 
standards for MY 2008-2011 light trucks (March 2006)
    3. Twenty-in-Ten Initiative (January 2007)
    4. Request for passenger car and light truck product plans 
(February 2007)
    5. Supreme Court decision in Massachusetts v. EPA (April 2007)
    6. Coordination between NHTSA and EPA on development of 
rulemaking proposals (Summer-Fall 2007)
    7. Ninth Circuit decision re final rule for MY 2008-2011 light 
trucks (November 2007)
    8. Enactment of Energy Independence and Security Act of 2007 
(December 2007)
    C. Energy Policy and Conservation Act, as amended
    1. Vehicles subject to standards for automobiles
    2. Mandate to set standards for automobiles
    3. Structure of standards
    4. Factors governing or considered in the setting of standards
    5. Consultation in setting standards
    6. Compliance flexibility and enforcement
III. Fuel economy enhancing technologies
    A. Data sources for technology assumptions
    B. Technologies and estimates of costs and effectiveness
    1. Engine technologies
    2. Transmission technologies
    3. Vehicle technologies
    4. Accessory technologies
    5. Hybrid technologies
    C. Technology synergies
    D. Technology cost learning curve
    E. Ensuring sufficient lead time
    1. Linking to redesign and refresh
    2. Technology phase-in caps
IV. Basis for attribute-based structure for setting fuel economy 
standards
    A. Why attribute-based instead of a single industry-wide 
average?
    B. Which attribute is most appropriate?
    1. Footprint-based function
    2. Functions based on other attributes
    C. The continuous function
V. Volpe model/analysis/generic description of function
    A. The Volpe model
    1. What is the Volpe model?
    2. How does the Volpe model apply technologies to manufacturers' 
future fleets?
    3. What effects does the Volpe model estimate?
    4. How can the Volpe model be used to calibrate and evaluate 
potential CAFE standards?
    5. How has the Volpe model been updated since the April 2006 
light truck CAFE final rule?
    a. Technology synergies
    b. Technology learning curves
    c. Calibration of reformed CAFE standards
    6. What manufacturer information does the Volpe model use?
    7. What economic information does the Volpe model use?
    a. Costs of fuel economy technologies
    b. Potential opportunity costs of improved fuel economy
    c. The on-road fuel economy `gap'
    d. Fuel prices and the value of saving fuel
    e. Consumer valuation of fuel economy and payback period
    f. Vehicle survival and use assumptions
    g. Growth in total vehicle use
    h. Accounting for the rebound effect of higher fuel economy
    i. Benefits from increased vehicle use
    j. Added costs from congestion, crashes and noise
    k. Petroleum consumption and import externalities
    l. Air pollutant emissions
    (i) Impacts on criteria air pollutant emissions
    (ii) Reductions in CO2 emissions
    (iii) Economic value of reductions in CO2 emissions

[[Page 24353]]

    m. The value of increased driving range
    n. Discounting future benefits and costs
    o. Accounting for uncertainty in benefits and costs
    B. How has NHTSA used the Volpe model to select the proposed 
standards?
    1. Establishing a continuous function standard
    2. Calibration of initial continuous function standards
    3. Adjustments to address policy considerations
    a. Curve crossings
    b. Steep curve for passenger cars
    c. Risk of upsizing
VI. Proposed fuel economy standards
    A. Standards for passenger cars and light trucks
    1. Proposed passenger car standards MY 2011-2015
    2. Proposed light truck standards MY 2011-2015
    3. Energy and environmental backstop
    4. Combined fleet performance
    B. Estimated technology utilization under proposed standards
    C. Costs and benefits of proposed standards
    D. Flexibility mechanisms
    E. Consistency of proposed standards with EPCA statutory factors
    1. Technological feasibility
    2. Economic practicability
    3. Effect of other motor vehicle standards of the Government on 
fuel economy
    4. Need of the U.S. to conserve energy
    F. Other considerations in setting standards under EPCA
    1. Safety
    2. Alternative fuel vehicle incentives
    3. Manufacturer credits
    G. Environmental impacts of the proposed standards
    H. Balancing the factors to determine maximum feasible CAFE 
levels
VII. Standards for commercial medium- and heavy-duty on-highway 
vehicles and ``work trucks''
VIII. Vehicle classification
    A. Origins of the regulatory definitions
    B. Rationale for the regulatory definitions in light of the 
current automobile market
    C. NHTSA is not proposing to change regulatory definitions at 
this time
IX. Enforcement
    A. Overview
    B. CAFE credits
    1. Credit trading
    2. Credit transferring
    3. Credit carry-forward/carry-back
    C. Extension and phasing out of flexible-fuel incentive program
X. Regulatory alternatives
XI. Sensitivity and Monte Carlo analysis
XII. Public participation
XIII. Regulatory notices and analyses
    A. Executive Order 12866 and DOT Regulatory Policies and 
Procedures
    B. National Environmental Policy Act
    C. Regulatory Flexibility Act
    D. Executive Order 13132 (Federalism)
    E. Executive Order 12988 (Civil Justice Reform)
    F. Unfunded Mandates Reform Act
    G. Paperwork Reduction Act
    H. Regulation Identifier Number (RIN)
    I. Executive Order 13045
    J. National Technology Transfer and Advancement Act
    K. Executive Order 13211
    L. Department of Energy Review
    M. Plain Language
    N. Privacy Act
XIV. Regulatory Text

I. Executive overview

A. Summary

    This document is being issued pursuant to the Energy Independence 
and Security Act of 2007 (EISA), which Congress passed in December 
2007. EISA mandates the setting of separate maximum feasible standards 
for passenger cars and for light trucks at levels sufficient to ensure 
that the average fuel economy of the combined fleet of all passenger 
cars and light trucks sold by all manufacturers in the U.S. in model 
year (MY) 2020 equals or exceeds 35 miles per gallon. That is a 40 
percent increase above the average of approximately 25 miles per gallon 
for the current combined fleet.
    Congress enabled NHTSA to require these substantial increases in 
fuel economy by requiring that passenger car standards be reformed 
through basing them on one or more vehicle attributes. The attribute-
based approach was originally recommended by the National Academy of 
Sciences in 2002 and adopted by NHTSA for light trucks in 2006. The new 
approach is a substantial improvement over the old approach of 
specifying the same numerical standard for each manufacturer. It avoids 
creating undue risks of adverse safety and employment impacts and 
distributes compliance responsibilities among the vehicle manufacturers 
more equitably.
    This document proposes standards for MYs 2011-2015, the maximum 
number of model years for which NHTSA can establish standards in a 
single rulemaking under EISA. Since lead time is a significant 
consideration in determining the stringency of future standards, the 
agency needs to establish the standards as far in advance as possible 
so as to maximize the amount of lead time for manufacturers to develop 
and implement plans for making the vehicle design changes necessary to 
achieve the requirements of EISA.
    In developing the proposed standards, the agency considered the 
four statutory factors underlying maximum feasibility (technological 
feasibility, economic practicability, the effect of other standards of 
the Government on fuel economy, and the need of the nation to conserve 
energy) as well as other relevant considerations such as safety. After 
assessing what fuel saving technologies would be available, how 
effective they are, and how quickly they could be introduced, and then 
factoring that information into the computer model its uses for 
applying technologies to particular vehicle models, the agency then 
balanced the factors relevant to standard setting. In its decision 
making, the agency used a marginal benefit-cost analysis that placed 
monetary values on relevant externalities (both energy security and 
environmental externalities, including the benefits of reductions in 
CO2 emissions). In the above process, the agency consulted 
with the Department of Energy and particularly the Environmental 
Protection Agency regarding a wide variety of matters, including, for 
example, the cost and effectiveness of available technologies, 
improvements to the computer model, and the selection of appropriate 
analytical assumptions.
    This document also proposes to add a new regulation designed to 
give manufacturers added flexibility in using credits earned by 
exceeding CAFE standards. The regulation would authorize the trading of 
credits between manufacturers. In addition, it would permit a 
manufacturer to transfer its credits from one of its compliance 
categories to another of its categories.
    NHTSA is also publishing two companion documents, one requesting 
vehicle manufacturers to provide up-to-date product plans for the model 
years covered by this document, and the other inviting Federal, State, 
and local agencies, Indian tribes, and the public to participate in 
identifying the environmental issues and reasonable alternatives to be 
examined in an environmental impact statement.

B. Energy Independence and Security Act of 2007

    The Energy Independence and Security Act of 2007 (EISA)\1\ builds 
on the President's ``Twenty in Ten'' initiative, which was announced in 
January 2007. That initiative sought to reduce gasoline usage by 20 
percent in the next 10 years. The enactment of EISA represents a major 
step forward in expanding the production of renewable fuels, reducing 
oil consumption, and confronting global climate change.
---------------------------------------------------------------------------

    \1\ Pub. L. 110-140, 121 Stat. 1492 (Dec. 18, 2007).
---------------------------------------------------------------------------

    EISA will help reduce America's dependence on oil by reducing U.S. 
demand for oil by setting a national fuel economy standard of at least 
35 miles per gallon by 2020--which will increase fuel economy standards 
by 40 percent and save billions of gallons of fuel. In January 2007, 
the President called for the first statutory increase in fuel economy 
standards for passenger

[[Page 24354]]

automobiles (referred to below as ``passenger cars'') since those 
standards were mandated in 1975, and EISA delivers on that request. 
EISA also includes an important reform the President has called for 
that allows the Transportation Department to issue ``attribute-based 
standards,'' which will ensure that increased fuel efficiency does not 
come at the expense of automotive safety. EISA also mandates increases 
in the use of renewable fuels by setting a mandatory Renewable Fuel 
Standard requiring fuel producers to use at least 36 billion gallons of 
renewable fuels in 2022.
    As the President noted in signing EISA, the combined effect of the 
various actions required by the Act will be to produce some of the 
largest CO2 emission reductions in our nation's history.
    EISA made a number of important changes to the Energy Policy and 
Conservation Act (EPCA) (Pub. L. 94-163), the 1975 statute that governs 
the CAFE program. EISA:
     Replaces the old statutory default standard of 27.5 mpg 
for passenger cars with a mandate to establish separate passenger car 
and light truck standards annually, beginning with MY 2011, set at the 
maximum feasible level. The standards for MYs 2011-2020 must, as a 
minimum, be set sufficiently high to ensure that the average fuel 
economy of the combined industrywide fleet of all new passenger cars 
and light trucks sold in the United States during MY 2020 is at least 
35 mpg.\2\
---------------------------------------------------------------------------

    \2\ Although NHTSA established an attribute-based standard for 
MY 2011 light trucks in its 2006 final rule, EISA mandates a new 
rulemaking, reflecting new statutory considerations and a new, up-
to-date administrative record, and consistent with EPCA as amended 
by EISA, to establish the standard for those light trucks.
---------------------------------------------------------------------------

     Limits to five the number of years for which standards can 
be established in a single rulemaking. That requirement, in combination 
with the requirement to start rulemaking with MY 2011, necessitates 
limiting this rulemaking to MYs 2011-2015.
     Mandates the reforming of CAFE standards for passenger 
cars by requiring that all CAFE standards be based on one or more 
vehicle attributes, thus ensuring that the improvements in fuel economy 
do not come at the expense of safety. NHTSA pioneered that approach in 
its last rulemaking on CAFE standards for light trucks.
     Requires that for each model year, beginning with MY 2011, 
the domestic passenger cars of each manufacturer of those cars must 
achieve a measured average fuel economy that is not less than 92 
percent of the average fuel economy of the combined fleet of domestic 
and non-domestic passenger cars sold in the United States in that model 
year.
     Provides greater flexibility for automobile manufacturers 
by (a) increasing from three to five the number of years that a 
manufacturer can carry forward the compliance credits it earns for 
exceeding CAFE standards, (b) allowing a manufacturer to transfer the 
credits it has earned from one of its classes of automobiles to 
another, and (c) authorizing the trading of credits between 
manufacturers.

C. Proposal

1. Standards
a. Stringency
    This document proposes to set attribute-based fuel economy 
standards for passenger cars and light trucks consistent with the 
Reformed CAFE approach that NHTSA used in establishing the light truck 
standards for MY 2008-2011 light trucks. Separate passenger car 
standards would be set for MYs 2011-2015, and light truck standards 
would be set for MYs 2011-2015. As noted above, EISA limits the number 
of model years for which standards may be established in a single 
rulemaking to five. We are proposing to establish standards for five 
years to maximize the amount of lead time that we can provide the 
manufacturers. This is necessary to make it possible to achieve the 
levels of average fuel economy required by MY 2020.
    Each vehicle manufacturer's required level of CAFE would be based 
on target levels of average fuel economy set for vehicles of different 
sizes and on the distribution of that manufacturer's vehicles among 
those sizes. Size would be defined by vehicle footprint. The level of 
the performance target for each footprint would reflect the 
technological and economic capabilities of the industry. The target for 
each footprint would be the same for all manufacturers, regardless of 
differences in their overall fleet mix. Compliance would be determined 
by comparing a manufacturer's harmonically averaged fleet fuel economy 
levels in a model year with a required fuel economy level calculated 
using the manufacturer's actual production levels and the targets for 
each footprint of the vehicles that it produces.
    The proposed standards were developed using a computer model (known 
as the ``Volpe Model'') that, for any given model year, applies 
technologies to a manufacturer's fleet until the manufacturer reaches 
compliance with the standard under consideration. The standards were 
tentatively set at levels such that, considering the seven largest 
manufacturers, the cost of the last technology application equaled the 
benefits of the improvement in fuel economy resulting from that 
application. We reviewed these proposed standards to consider the 
underlying increased use of technologies and the associated impact on 
the industry. This process recognizes that the relevance of costs in 
achieving benefits, and uses benefit figures that include the value of 
reducing the negative externalities (economic and environmental) from 
producing and consuming fuel. These environmental externalities 
include, among other things, reducing tailpipe emissions of CO2.\3\ In 
view of the process used to develop the proposed standards, they are 
also referred to as ``optimized standards.''
---------------------------------------------------------------------------

    \3\ The externalities included in our analysis do not, however, 
include those associated with the reduction of the other GHG emitted 
by automobiles, i.e., methane (CH4), nitrous oxide (N2O), and 
hydroflurocarbons (HFCs). Actual air conditioner operation is not 
included in the test procedures used to obtain both (1) emission 
rates for purposes of determining compliance with EPA criteria 
pollutant emission standards and (2) fuel economy values for 
purposes of determining compliance with NHTSA CAFE standards, 
although air conditioner operation is included in ``supplemental'' 
federal test procedures used to determine compliance with 
corresponding and separate EPA criteria pollutant emission 
standards.
---------------------------------------------------------------------------

    Compared to the 2006 rulemaking that established the MY 2008-11 
CAFE standards for light trucks, this rulemaking much more fully 
captures the value of the costs and benefits of setting CAFE standards. 
This is important because assumptions regarding gasoline price 
projections, along with assumptions for externalities, are based on 
changed economic and environmental and energy security conditions and 
play a big role in the agency's balancing of the statutory 
considerations in arriving at a determination of maximum feasible. In 
light of EISA and the need to balance the statutory considerations in a 
way that reflects the current need of the nation to conserve energy, 
including the current assessment of the climate change problem, the 
agency revisited the various assumptions used in the Volpe Model to 
determine the level of the standards. Specifically, in running the 
Volpe Model and stopping at a point where marginal costs equaled 
marginal benefits or where net benefits to society are maximized, the 
agency used higher gasoline prices and higher estimates for energy 
security values ($0.29 per gallon instead of $0.09 per gallon). The 
agency also monetized carbon dioxide (at

[[Page 24355]]

$7.00/ton), which it did not do in the previous rulemaking, and 
expanded its technology list. In addition, the agency used cost 
estimates that reflect economies of scale and estimated ``learning''-
driven reductions in the cost of technologies as well as quicker 
penetration rates for advanced technologies. These changes to the 
inputs to the model had a major impact on increasing the benefits in 
certain model years by allowing for greater penetration of 
technologies.
    The agency cannot set out the exact level of CAFE that each 
manufacturer will be required to meet for each model year under the 
proposed passenger car or light truck standards since the levels will 
depend on information that will not be available until the end of each 
of the model years, i.e., the final actual production figures for each 
of those years. The agency can, however, project what the industry wide 
level of average fuel economy would be for passenger cars and for light 
trucks if each manufacturer produced its expected mix of automobiles 
and just met its obligations under the proposed ``optimized'' standards 
for each model year. Adjacent to each average fuel economy figure is 
the estimated associated level of tailpipe emissions of CO2 that would 
be achieved.\4\
---------------------------------------------------------------------------

    \4\ Given the contributions made by CAFE standards to addressing 
not only energy independence and security, but also to reducing 
tailpipe emissions of CO2, fleet performance is stated in the above 
discussion both in terms of fuel economy and the associated 
reductions in tailpipe emissions of CO2 since the CAFE standard will 
have the practical effect of limiting those emissions approximately 
to the indicated levels during the official CAFE test procedures 
established by EPA. The relationship between fuel consumption and 
carbon dioxide emissions is discussed ubiquitously, such as at 
www.fueleconomy.gov, a fuel economy-related Web site managed by DOE 
and EPA (see http://www.fueleconomy.gov/feg/contentIncludes/co2_
inc.htm, which provides a rounded value of 20 pounds of CO2 per 
gallon of gasoline). (Last accessed April 20, 2008.) The CO2 
emission rates shown are based on gasoline characteristics. Because 
diesel fuel contains more carbon (per gallon) than gasoline, the 
presence of diesel engines in the fleet--which NHTSA expects to 
increase in response to the proposed CAFE standards--will cause the 
actual CO2 emission rate corresponding to any given CAFE level to be 
slightly higher than shown here. (The agency projects that 4 percent 
of the MY 2015 passenger car fleet and 10 percent of the MY 2015 
light truck fleet will have diesel engines.) Conversely (and 
hypothetically), applying the same CO2 emission standard to both 
gasoline and diesel vehicles would discourage manufacturers from 
improving diesel engines, which show considerable promise as a means 
to improve fuel economy.

    For passenger cars:
MY 2011: 31.2 mpg (285 g/mi of tailpipe emissions of CO2)
MY 2012: 32.8 mpg (271 g/mi of tailpipe emissions of CO2)
MY 2013: 34.0 mpg (261 g/mi of tailpipe emissions of CO2)
MY 2014: 34.8 mpg (255 g/mi of tailpipe emissions of CO2)
MY 2015: 35.7 mpg (249 g/mi of tailpipe emissions of CO2)

    For light trucks:

MY 2011: 25.0 mpg (355 g/mi of tailpipe emissions of CO2)
MY 2012: 26.4 mpg (337 g/mi of tailpipe emissions of CO2)
MY 2013: 27.8 mpg (320 g/mi of tailpipe emissions of CO2)
MY 2014: 28.2 mpg (315 g/mi of tailpipe emissions of CO2)
MY 2015: 28.6 mpg (310 g/mi of tailpipe emissions of CO2)

    The combined industry wide average fuel economy (in miles per 
gallon, or mpg) levels (in grams per mile, or g/mi) for both cars and 
light trucks, if each manufacturer just met its obligations under the 
proposed ``optimized'' standards for each model year, would be as 
follows:

MY 2011: 27.8 mpg (2.5 mpg increase above MY 2010; 320 g/mi CO2)
MY 2012: 29.2 mpg (1.4 mpg increase above MY 2011; 304 g/mi CO2)
MY 2013: 30.5 mpg (1.3 mpg increase above MY 2012; 291 g/mi CO2)
MY 2014: 31.0 mpg (0.5 mpg increase above MY 2013; 287 g/mi CO2)
MY 2015: 31.6 mpg (0.6 mpg increase above MY 2014; 281 g/mi CO2)

    The annual average increase during this five year period is 
approximately 4.5 percent. Due to the uneven distribution of new model 
introductions during this period and to the fact that significant 
technological changes can be most readily made in conjunction with 
those introductions, the annual percentage increases are greater in the 
early years in this period.
    Given a starting point of 31.8 mpg in MY 2015, the average annual 
increase for MYs 2016-2020 would need to be only 2.1 percent in order 
for the projected combined industry wide average to reach at least 35 
mpg by MY 2020, as mandated by EISA.
    In addition, per EISA, each manufacturer's domestic passenger fleet 
is required in each model year to achieve 27.5 mpg or 92 percent of the 
CAFE of the industry wide combined fleet of domestic and non-domestic 
passenger cars \5\ for that model year, whichever is higher. This 
requirement results in the following alternative minimum standard (not 
attribute-based) for domestic passenger cars:
---------------------------------------------------------------------------

    \5\ Those numbers set out several paragraphs above.

MY 2011: 28.7 mpg (310 g/mi of tailpipe emissions of CO2)
MY 2012: 30.2 mpg (294 g/mi of tailpipe emissions of CO2)
MY 2013: 31.3 mpg (284 g/mi of tailpipe emissions of CO2)
MY 2014: 32.0 mpg (278 g/mi of tailpipe emissions of CO2)
MY 2015: 32.9 mpg (270 g/mi of tailpipe emissions of CO2)

    The agency is also issuing, along with this document, a notice 
requesting updated product plan information and other data to assist in 
developing a final rule. We recognize that the manufacturer product 
plans relied upon in developing this proposal--those plans received in 
late spring of 2007 in response to an early 2007 request for 
information--may already be outdated in some respects. We fully expect 
that manufacturers have revised those plans to reflect subsequent 
developments, especially the enactment of EISA.
    We solicit comment on all aspects of this proposal, including the 
methodology, economic assumptions, analysis and tentative conclusions. 
In particular, we solicit comment on whether the proposed levels of 
CAFE satisfy EPCA, e.g., reflect an appropriate balancing of the 
explicit statutory factors and other relevant factors. Other specific 
areas where we request comments are identified elsewhere in this 
preamble and in the Preliminary Regulatory Impact Analysis (PRIA). 
Based on public comments and other information, including new data and 
analysis, and updated product plans,\6\ the standards adopted in the 
final rule could well be different from those proposed in this 
document.
---------------------------------------------------------------------------

    \6\ The proposed standards are, in the first instance, based on 
the confidential product plans submitted by the manufacturers in the 
spring of 2006. The final rule will be based on the confidential 
plans submitted in the next several months. The agency anticipates 
that those new plans, which presumably will reflect in some measure 
the enactment of EISA and the issuance of this proposal, will 
project higher levels of average fuel economy than the 2006 product 
plans.
---------------------------------------------------------------------------

b. Benefits
    We estimate that the proposed standards for passenger cars would 
save approximately 18.7 billion gallons of fuel and avoid tailpipe 
CO2 emissions by 178 billion metric tons over the lifetime 
of the passenger cars sold during those model years, compared to the 
fuel savings and emissions reductions that would occur if the standards 
remained at the adjusted baseline (i.e., the higher of manufacturer's 
plans and the manufacturer's required level of average fuel economy for 
MY 2010).
    We estimate that the value of the total benefits of the proposed 
passenger car standards would be approximately $31 billion \7\ over the 
lifetime of the 5 model

[[Page 24356]]

years combined. This estimate of societal benefits includes direct 
impacts from lower fuel consumption as well as externalities and also 
reflects offsetting societal costs resulting from the rebound effect.
---------------------------------------------------------------------------

    \7\ The $22 billion estimate is based on a 7% discount rate for 
valuing future impacts. NHTSA estimated benefits using both 7% and 
3% discount rates. Under a 3% rate, net consumer benefits for 
passenger car CAFE improvements total $28 million.
---------------------------------------------------------------------------

    We estimate that the proposed standards for light trucks would save 
approximately 36 billion gallons of fuel and prevent the tailpipe 
emission of 343 million metric tons of CO2 over the lifetime 
of the light trucks sold during those model years, compared to the fuel 
savings and emissions reductions that would occur if the standards 
remained at the adjusted baseline. We estimate that the value of the 
total benefits of the proposed light truck standards would be 
approximately $57 billion \8\ over the lifetime of the 5 model years of 
light trucks combined. This estimate of societal benefits includes 
direct impacts from lower fuel consumption as well as externalities and 
also reflects offsetting societal costs resulting from the rebound 
effect.
---------------------------------------------------------------------------

    \8\ The $56 billion estimate is based on a 7% discount rate for 
valuing future impacts. NHTSA estimated benefits using both 7% and 
3% discount rates. Under a 3% rate, net consumer benefits for light 
truck CAFE improvements total $70 million.
---------------------------------------------------------------------------

c. Costs
    The total costs for manufacturers just complying with the standards 
for MY 2011-2015 passenger cars would be approximately $16 billion, 
compared to the costs they would incur if the standards remained at the 
adjusted baseline. The resulting vehicle price increases to buyers of 
MY 2015 passenger cars would be recovered or paid back \9\ in 
additional fuel savings in an average of 56 months, assuming fuel 
prices ranging from $2.26 per gallon in 2016 to $2.51 per gallon in 
2030.\10\
---------------------------------------------------------------------------

    \9\ See Section V.A.7 below for discussion of payback period.
    \10\ The fuel prices (shown here in 2006 dollars) used to 
calculate the length of the payback period are those projected 
(Annual Energy Outlook 2008, revised early release) by the Energy 
Information Administration over the life of the MY 2011-2015 light 
trucks, not current fuel prices.
---------------------------------------------------------------------------

    The total costs for manufacturers just complying with the standards 
for MY 2011-2015 light trucks would be approximately $31 billion, 
compared to the costs they would incur if the standards remained at the 
adjusted baseline. The resulting vehicle price increases to buyers of 
MY 2015 light trucks would be paid back in additional fuel savings in 
an average of 50 months, assuming fuel prices ranging from $2.26 to 
$2.51 per gallon.
d. Flexibilities
    The agency's benefit and cost estimates do not reflect the 
availability and use of flexibility mechanisms, such as compliance 
credits and credit trading because EPCA prohibits NHTSA from 
considering the effects of those mechanisms in setting CAFE standards. 
EPCA has precluded consideration of the FFV adjustments ever since it 
was amended to provide for those adjustments. The prohibition against 
considering compliance credits was added by EISA.
    The benefit and compliance cost estimates used by the agency in 
determining the maximum feasible level of the CAFE standards assume 
that manufacturers will rely solely on the installation of fuel economy 
technology to achieve compliance with the proposed standards. In 
reality, however, manufacturers are likely to rely to some extent on 
flexibility mechanisms provided by EPCA (as described in Section VI) 
and will thereby reduce the cost of complying with the proposed 
standards to a meaningful extent.
2. Credits
    NHTSA is also proposing a new Part 536 on use of ``credits'' earned 
for exceeding applicable CAFE standards. Part 536 will implement the 
provisions in EISA authorizing NHTSA to establish by regulation a 
credit trading program and directing it to establish by regulation a 
credit transfer program.\11\ Since its enactment, EPCA has permitted 
manufacturers to earn credits for exceeding the standards and to apply 
those credits to compliance obligations in years other than the model 
year in which it was earned. EISA extended the ``carry-forward'' period 
to five model years, and left the ``carry-back'' period at three model 
years. Under the proposed Part 536, credit holders (including, but not 
limited to, manufacturers) will have credit accounts with NHTSA, and 
will be able to hold credits, apply them to compliance with CAFE 
standards, transfer them to another ``compliance category'' for 
application to compliance there, or trade them. A credit may also be 
cancelled before its expiry date, if the credit holder so chooses. 
Traded credits will be subject to an ``adjustment factor'' to ensure 
total oil savings are preserved, as required by EISA. EISA also 
prohibits credits earned before MY 2011 from being transferred, so 
NHTSA has developed several regulatory restrictions on trading and 
transferring to facilitate Congress' intent in this regard. Additional 
information on the proposed Part 536 is available in section IX below.
---------------------------------------------------------------------------

    \11\ Congress required that DOT establish a credit 
``transferring'' regulation, to allow individual manufacturers to 
move credits from one of their fleets to another (e.g., using a 
credit earned for exceeding the light truck standard for compliance 
in the domestic passenger car standard). Congress allowed DOT to 
establish a credit ``trading'' regulation, so that credits may be 
bought and sold between manufacturers and other parties.
---------------------------------------------------------------------------

II. Background

A. Contribution of Fuel Economy Improvements to Addressing Energy 
Independence and Security and Climate Change

1. Relationship Between Fuel Economy and CO2 Tailpipe Emissions
    Improving fuel economy reduces the amount of tailpipe emissions of 
CO2. CO2 emissions are directly linked to fuel consumption because CO2 
is the ultimate end product of burning gasoline. The more fuel a 
vehicle burns, the more CO2 it emits. Since the CO2 emissions are 
essentially constant per gallon of fuel combusted, the amount of fuel 
consumption per mile is directly related to the amount of CO2 emissions 
per mile. Thus, requiring improvements in fuel economy indirectly, but 
necessarily requires reductions in tailpipe emissions of CO2 emissions. 
This can be seen in the table below. To take the first value of fuel 
economy from the table below as an example, a standard of 21.0 mpg 
would indirectly place substantially the same limit on tailpipe CO2 
emissions as a tailpipe CO2 emission standard of 423.2 g/mi of CO2, and 
vice versa.\12\
---------------------------------------------------------------------------

    \12\ To the extent that manufacturers comply with a CAFE 
standard with diesel automobiles instead of gasoline ones, the level 
of CO2 tailpipe emissions would be less. As noted above, the agency 
projects that 4 percent of the MY 2015 passenger car fleet and 10 
percent of the MY 2015 light truck fleet will have diesel engines. 
The CO2 tailpipe emissions of a diesel powered passenger car are 15 
percent higher than those of a comparable gasoline power passenger 
car.

[[Page 24357]]



                       Table II-1.--CAFE Standards (mpg) and the Limits They Indirectly Place on Tailpipe Emissions of CO2 (g/mi)*
--------------------------------------------------------------------------------------------------------------------------------------------------------
                  CAFE Std                     CO2     CAFE Std    CO2     CAFE Std    CO2     CAFE Std    CO2     CAFE Std    CO2     CAFE Std    CO2
--------------------------------------------------------------------------------------------------------------------------------------------------------
21.0.......................................    444.4       26.0    341.8       31.0    286.7       36.0    246.9       41.0    216.8       46.0    193.2
22.0.......................................    404.0       27.0    329.1       32.0    277.7       37.0    240.2       42.0    211.6       47.0    188.3
23.0.......................................    386.4       28.0    317.4       33.0    269.3       38.0    233.9       43.0    206.7       48.0    189.1
24.0.......................................    370.3       29.0    306.4       34.0    261.4       39.0    227.9       44.0    202.0       49.0    181.4
25.0.......................................    355.5       30.0    296.2       35.0    253.9       40.0    222.2       45.0    197.5       50.0   177.7
--------------------------------------------------------------------------------------------------------------------------------------------------------
 This table is based on calculations that use the figure of 8,887 grams of CO2 per gallon of gasoline consumed, based on characteristics of gasoline
  vehicle certification fuel. To convert a mpg value into CO2 g/mi, divide 8,887 by the mpg value.

2. Fuel Economy Improvements/CO2 Tailpipe Emission 
Reductions Since 1975
    The need to take action to reduce greenhouse gas emissions, e.g., 
motor vehicle tailpipe emissions of CO2, in order to forestall and even 
mitigate climate change is well recognized.\13\ Less well recognized 
are two related facts. First, improving fuel economy is the only method 
available to motor vehicle manufacturers for making significant 
reductions in the CO2 tailpipe emissions of motor vehicles and thus 
must be the core element of any effort to achieve those reductions. 
Second, the significant improvements in fuel economy since 1975, due to 
the CAFE standards and in some measure to market conditions as well, 
have directly caused reductions in the rate of CO2 tailpipe emissions 
per vehicle.
---------------------------------------------------------------------------

    \13\ IPCC (2007): Climate Change 2007: Mitigation of Climate 
Change. Contribution of Working Group III to the Fourth Assessment 
Report of the Intergovernmental Panel on Climate Change [B. Metz, O. 
Davidson, P. Bosch, R. Dave, and L. Meyer (eds.)]. Cambridge 
University Press, Cambridge, United Kingdom and New York, NY, USA.
---------------------------------------------------------------------------

    In 1975, passenger cars manufactured for sale in the U.S. averaged 
only 15.8 mpg (562.5 grams of CO2 per mile or 562.5 g/mi of CO2). By 
2007, the average fuel economy of passenger cars had increased to 31.3 
mpg, causing g/mi of CO2 to fall to 283.9. Similarly, in 1975, light 
trucks averaged 13.7 mpg (648.7 g/mi of CO2). By 2007, the average fuel 
economy of light trucks had risen to 23.1 mpg, causing g/mi of CO2 to 
fall to 384.7.

  Table II-2.--Improvements in MPG/Reductions in G/MI of CO2 Passenger
                                  Cars
                               [1975-2007]
------------------------------------------------------------------------
                                                    MPG      G/MI of CO2
------------------------------------------------------------------------
1975..........................................         15.8        562.5
2007..........................................         31.3        283.9
------------------------------------------------------------------------


 Table II-3.--Improvements in MPG/Reductions in G/MI of CO2 Light Trucks
                               [1975-2007]
------------------------------------------------------------------------
                                                    MPG      G/MI of CO2
------------------------------------------------------------------------
1975..........................................         13.7        648.7
2007..........................................         23.1        384.7
------------------------------------------------------------------------

    If fuel economy had not increased above the 1975 level, cars and 
light trucks would have emitted an additional 11 billion metric tons of 
CO2 into the atmosphere between 1975 and 2005. That is nearly the 
equivalent of emissions from all U.S. fossil fuel combustion for two 
years (2004 and 2005). The figure below shows the amount of CO2 
emissions avoided due to increases in fuel economy.
BILLING CODE 4910-59-P

[[Page 24358]]

[GRAPHIC] [TIFF OMITTED] TP02MY08.000


[[Page 24359]]



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

1. National Academy of Sciences CAFE Report (February 2002)
a. Significantly Increasing CAFE Standards Without Reforming Them Would 
Adversely Affect Safety
    In the congressionally-mandated report entitled ``Effectiveness and 
Impact of Corporate Average Fuel Economy (CAFE) Standards,'' \14\ a 
committee of the National Academy of Sciences (NAS) (``2002 NAS 
Report'') concluded that the then-existing form of passenger car and 
light truck CAFE standards created an incentive for vehicle 
manufacturers to comply in part by downweighting and even downsizing 
their vehicles and that these actions had led to additional fatalities. 
The committee explained that these problems arose because the CAFE 
standards subjected all passenger cars to the same fuel economy target 
and all light trucks to the same target, regardless of their weight, 
size, or load-carrying capacity. The committee said that this 
experience suggests that consideration should be given to developing a 
new system of fuel economy targets that reflects differences in such 
vehicle attributes.
---------------------------------------------------------------------------

    \14\ National Research Council, ``Effectiveness and Impact of 
Corporate Average Fuel Economy (CAFE) Standards,'' National Academy 
Press, Washington, DC (2002). Available at http://www.nap.edu/
openbook.php?isbn=0309076013 (last accessed April 20, 2008). The 
conference committee report for the Department of Transportation and 
Related Agencies Appropriations Act for FY 2001 (Pub. L. 106-346) 
directed NHTSA to fund a study by NAS to evaluate the effectiveness 
and impacts of CAFE standards (H. Rep. No. 106-940, p. 117-118). In 
response to the direction from Congress, NAS published this lengthy 
report.
---------------------------------------------------------------------------

    Looking to the future, the committee said that while it is 
technically feasible and potentially economically practicable to 
improve fuel economy without reducing vehicle weight or size and, 
therefore, without significantly affecting the safety of motor vehicle 
travel, the actual strategies chosen by manufacturers to improve fuel 
economy will depend on a variety of factors. In the committee's 
judgment, the extensive downweighting and downsizing that occurred 
after fuel economy requirements were established in the 1970s suggested 
that the likelihood of a similar response to further increases in fuel 
economy requirements must be considered seriously. Any reduction in 
vehicle size and weight would have safety implications.
    The committee cautioned that the safety effects of downsizing and 
downweighting are likely to be hidden by the generally increasing 
safety of the light-duty vehicle fleet.\15\ It said that some might 
argue that this improving safety picture means that there is room to 
improve fuel economy without adverse safety consequences; however, such 
an approach would not achieve the goal of avoiding the adverse safety 
consequences of fuel economy increases. Rather, the safety penalty 
imposed by increased fuel economy (if weight reduction is one of the 
measures) will be more difficult to identify in light of the continuing 
improvement in traffic safety. Although it is anticipated that these 
safety innovations will improve the safety of vehicles of all sizes, 
that does not mean that downsizing to achieve fuel economy improvements 
will not have any safety costs. If two vehicles of the same size are 
modified, one both by downsizing it and adding the safety innovations 
and the other just by adding the safety innovations, the latter vehicle 
will in all likelihood be safer.
---------------------------------------------------------------------------

    \15\ Two of the 12 members of the committee dissented from the 
majority's safety analysis and conclusions.
---------------------------------------------------------------------------

    The committee concluded that if an increase in fuel economy were 
implemented pursuant to standards that are structured in a way that 
encourages either downsizing or the increased production of smaller 
vehicles, some additional traffic fatalities would be expected. Without 
a thoughtful restructuring of the program, there would be the trade-
offs that must be made if CAFE standards were increased by any 
significant amount.\16\
---------------------------------------------------------------------------

    \16\ NAS, p. 9.
---------------------------------------------------------------------------

    In response to these conclusions, NHTSA began issuing attribute-
based CAFE standards for light trucks and sought legislative authority 
to issue attribute-based CAFE standards for passenger cars before 
undertaking to raise the car standards. Congress went a step further in 
enacting EISA, not only authorizing the issuance of attribute-based 
standards, but also mandating them.
    Fully realizing all of the safety and other \17\ benefits of these 
reforms will depend in part on whether the unreformed, non-attribute 
based greenhouse standards adopted by California and other states are 
implemented. Apart from issues of relative stringency, the effects on 
vehicle manufacturers of implementing those state emission standards 
should be substantially similar to the effects of implementing non-
attribute-based CAFE standards, given the nearly identical nature of 
most aspects of those emission standards and CAFE standards in terms of 
technological means of compliance and methods of measuring performance.
---------------------------------------------------------------------------

    \17\ Reformed CAFE has several advantages compared to Unreformed 
CAFE:
    First, Reformed CAFE increases energy savings. The energy-saving 
potential of Unreformed CAFE is limited because only a few full-line 
manufacturers are required to make improvements. Under Reformed 
CAFE, which accounts for size differences in product mix, virtually 
all manufacturers will be required to use advanced fuel-saving 
technologies to achieve the requisite fuel economy for their 
automobiles.
    Second, Reformed CAFE reduces the chances of adverse safety 
consequences. Downsizing of vehicles as a CAFE compliance strategy 
is discouraged under Reformed CAFE since as vehicles become smaller, 
the applicable fuel economy target becomes more stringent.
    Third, Reformed CAFE provides a more equitable regulatory 
framework for different vehicle manufacturers. Under Unreformed 
CAFE, the cost burdens and compliance difficulties have been imposed 
nearly exclusively on the full-line manufacturers.
    Fourth, Reformed CAFE is more market-oriented because it more 
fully respects economic conditions and consumer choice. Reformed 
CAFE does not force vehicle manufacturers to adjust fleet mix toward 
smaller vehicles although they can make adjustments if that is what 
consumers are demanding. Instead, it allows the manufacturers to 
adjust the mix of their product offerings in response to the market 
place.
---------------------------------------------------------------------------

b. Environmental and Other Externalities Justify Increasing the CAFE 
Standards
    The 2002 NAS report also concluded that the CAFE standards have 
contributed to increased fuel economy, which in turn has reduced 
dependence on imported oil, improved the nation's terms of trade, and 
reduced emissions of carbon dioxide (a principal greenhouse gas), 
relative to what they otherwise would have been. If fuel economy had 
not improved, gasoline consumption (and crude oil imports) would be 
about 2.8 million barrels per day (mmbd) greater than it is.\18\ 
Reducing fuel consumption in vehicles also reduces carbon dioxide 
emissions. If the nation were using 2.8 mmbd more gasoline, carbon 
emissions would be more than 100 million metric tons of carbon (mmtc) 
higher. Thus, improvements in light-duty vehicle (4 wheeled motor 
vehicles under 10,000 pounds gross vehicle weight rating) fuel economy 
have reduced overall U.S. emissions by about 7 percent.\19\
---------------------------------------------------------------------------

    \18\ NAS, pp. 3 and 20.
    \19\ NAS, p. 20.
---------------------------------------------------------------------------

    The report concluded that technologies exist that could 
significantly further reduce fuel consumption by passenger cars and 
light trucks within 15 years, while maintaining vehicle size, weight, 
utility and performance.\20\ Light duty trucks

[[Page 24360]]

were said to offer the greatest potential for reducing fuel 
consumption.\21\ The report also noted that vehicle development 
cycles--as well as future economic, regulatory, safety and consumer 
preferences--would influence the extent to which these technologies 
could lead to increased fuel economy in the U.S. market. To assess the 
economic trade-offs associated with the introduction of existing and 
emerging technologies to improve fuel economy, the NAS conducted what 
it called a ``cost-efficient analysis'' based on the direct benefits 
(value of saved fuel) to the consumer--``that is, the committee 
identified packages of existing and emerging technologies that could be 
introduced over the next 10 to 15 years that would improve fuel economy 
up to the point where further increases in fuel economy would not be 
reimbursed by fuel savings.'' \22\
---------------------------------------------------------------------------

    \20\ NAS, p. 3 (Finding 5).
    \21\ NAS, p. 4 (Finding 5).
    \22\ NAS, pp. 4 (Finding 6) and 64.
---------------------------------------------------------------------------

    The committee emphasized that it is critically important to be 
clear about the reasons for considering improved fuel economy. While 
the dollar value of the saved fuel would be largest portion of the 
potential benefits, the committee noted that there is theoretically 
insufficient reason for the government to issue higher standards just 
to obtain those direct benefits since consumers have a wide variety of 
opportunities to buy a fuel-efficient vehicle.\23\
---------------------------------------------------------------------------

    \23\ NAS, pp. 8-9.
---------------------------------------------------------------------------

    The committee said that there are two compelling concerns that 
justify a government mandated increase in fuel economy, both relating 
to externalities. The most important concern, it argued, is the one 
about the accumulation in the atmosphere of greenhouse gases, 
principally carbon dioxide.\24\
---------------------------------------------------------------------------

    \24\ NAS, pp. 2, 13, and 83.
---------------------------------------------------------------------------

    A second concern is that petroleum imports have been steadily 
rising because of the nation's increasing demand for gasoline without a 
corresponding increase in domestic supply. The high cost of oil imports 
poses two risks: Downward pressure on the strength of the dollar (which 
drives up the cost of goods that Americans import) and an increase in 
U.S. vulnerability to macroeconomic shocks that cost the economy 
considerable real output.
    To determine how much the fuel economy standards should be 
increased, the committee urged that all social benefits be considered. 
That is, it urged not only that the dollar value of the saved fuel be 
considered, but also that the dollar value to society of the resulting 
reductions in greenhouse gas emissions and in dependence on imported 
oil should be calculated and considered. The committee said that if it 
is possible to assign dollar values to these favorable effects, it 
becomes possible to make at least crude comparisons between the 
socially beneficial effects of measures to improve fuel economy on the 
one hand, and the costs (both out-of-pocket and more subtle) on the 
other. The committee chose a value of about $0.30/gal of gasoline for 
the externalities associated with the combined impacts of fuel 
consumption on greenhouse gas emissions and on world oil market 
conditions.\25\
---------------------------------------------------------------------------

    \25\ NAS, pp. 4 and 85-86.
---------------------------------------------------------------------------

    The report expressed concerns about increasing the standards under 
the CAFE program as currently structured. While raising CAFE standards 
under the existing structure would reduce fuel consumption, doing so 
under alternative structures ``could accomplish the same end at lower 
cost, provide more flexibility to manufacturers, or address inequities 
arising from the present'' structure.\26\ Further, the committee said, 
``to the extent that the size and weight of the fleet have been 
constrained by CAFE requirements * * * those requirements have caused 
more injuries and fatalities on the road than would otherwise have 
occurred.'' \27\ Specifically, it noted: ``The downweighting and 
downsizing that occurred in the late 1970s and early 1980s, some of 
which was due to CAFE standards, probably resulted in an additional 
1300 to 2600 traffic fatalities in 1993.'' \28\
---------------------------------------------------------------------------

    \26\ NAS, pp. 4-5 (Finding 10).
    \27\ NAS, p. 29.
    \28\ NAS, p. 3 (Finding 2).
---------------------------------------------------------------------------

    To address those structural problems, the report suggested various 
possible reforms. The report found that the ``CAFE program might be 
improved significantly by converting it to a system in which fuel 
targets depend on vehicle attributes.'' \29\ The report noted further 
that under an attribute-based approach, the required CAFE levels could 
vary among the manufacturers based on the distribution of their product 
mix. NAS stated that targets could vary among passenger cars and among 
trucks, based on some attribute of these vehicles such as weight, size, 
or load-carrying capacity. The report explained that a particular 
manufacturer's average target for passenger cars or for trucks would 
depend upon the fractions of vehicles it sold with particular levels of 
these attributes.\30\
---------------------------------------------------------------------------

    \29\ NAS, p. 5 (Finding 12).
    \30\ NAS, p. 87.
---------------------------------------------------------------------------

    In February 2002, Secretary Mineta asked Congress ``to provide the 
Department of Transportation with the necessary authority to reform the 
CAFE program, guided by the NAS report's suggestions.''
2. Final Rule Establishing Reformed (Attribute-Based) CAFE Standards 
for MY 2008-2011 Light Trucks (March 2006)
    The 2006 final rule reformed the structure of the CAFE program for 
light trucks and established higher CAFE standards for MY 2008-2011 
light trucks.\31\ Reforming the CAFE program enables it to achieve 
larger fuel savings, while enhancing safety and preventing adverse 
economic consequences.
---------------------------------------------------------------------------

    \31\ 71 FR 17566; April 6, 2006.
---------------------------------------------------------------------------

    During a transition period of MYs 2008-2010, manufacturers may 
comply with CAFE standards established under the reformed structure 
(Reformed CAFE) or with standards established in the traditional way 
(Unreformed CAFE). This permits manufacturers and the agency to gain 
experience with implementing the Reformed CAFE standards. Under the 
2006 rule, all manufacturers were required to comply with a Reformed 
CAFE standard in MY 2011.
    Under Reformed CAFE, fuel economy standards were restructured so 
that they are based on a measure of vehicle size called ``footprint,'' 
which is the product of multiplying a vehicle's wheelbase by average 
its track width. A target level of fuel economy was established for 
each increment in footprint (0.1 ft\2\). Trucks with smaller footprints 
have higher fuel economy targets; conversely, larger ones have lower 
targets. A particular manufacturer's compliance obligation for a model 
year will be calculated as the harmonic average of the fuel economy 
targets for the manufacturer's vehicles, weighted by the distribution 
of manufacturer's production volumes among the footprint increments. 
Thus, each manufacturer will be required to comply with a single 
overall average fuel economy level for each model year of production.
    The approach for determining the fuel economy targets was to set 
them just below the level where the increased cost of technologies that 
could be adopted by manufacturers to improve fuel economy would first 
outweigh the added benefits that would result from such technology. 
These targets translate into required levels of average fuel economy 
that are technologically feasible because manufacturers can achieve 
them using available technologies. Those levels also reflect the need 
of the nation to reduce

[[Page 24361]]

energy consumption because they reflect the economic value of the 
savings in resources, as well as of the reductions in economic and 
environmental externalities that result from producing and using less 
fuel.
    The Unreformed CAFE standards are: 22.5 miles per gallon (mpg) for 
MY 2008, 23.1 mpg for MY 2009, and 23.5 mpg for MY 2010. To aid the 
transition to Reformed CAFE, the Reformed CAFE standards for those 
years were set at levels intended to ensure that the industry-wide 
costs of the Reformed standards are roughly equivalent to the industry-
wide costs of the Unreformed CAFE standards in those model years. For 
MY 2011, the Reformed CAFE standard was set at the level that maximizes 
net benefits. Net benefits include the increase in light truck prices 
due to technology improvements, the decrease in fuel consumption, and a 
number of other factors. All of the standards were set at the maximum 
feasible level, while accounting for technological feasibility, 
economic practicability and other relevant factors.
    We carefully balanced the costs of the rule with the benefits of 
reducing energy consumption. Compared to Unreformed CAFE, Reformed CAFE 
enhances overall fuel savings while providing vehicle manufacturers 
with the flexibility they need to respond to changing market 
conditions. Reformed CAFE will also provide a more equitable regulatory 
framework by creating a level-playing field for manufacturers, 
regardless of whether they are full-line or limited-line manufacturers. 
We were particularly encouraged that Reformed CAFE will eliminate the 
incentive to downsize some of their fleet as a CAFE compliance 
strategy, thereby reducing the adverse safety risks associated with the 
Unreformed CAFE program.
3. Twenty-in-Ten Initiative (January 2007)
    In his January 2007 State of the Union address, the President 
announced his Twenty-in-Ten initiative for increasing the supply of 
renewable and alternative fuels and reforming and increasing the CAFE 
standards. Consistent with the NAS report, he urged the authority be 
provided to reform CAFE for passenger cars by adopting an attribute-
based system (for example, a size-based system) reduces the risk that 
vehicle safety is compromised, helps preserve consumer choice, and 
helps spread the burden of compliance across all product lines and 
manufacturers. He also urged that authority be provided to set the CAFE 
standards, based on cost/benefit analysis, using sound science, and 
without impacting safety.
4. Request for Passenger Car and Light Truck Product Plans (February 
2007)
    In late February 2007, NHTSA published a notice to acquire new and 
updated information regarding vehicle manufacturers' future product 
plans to aid in implementing the President's plan for reforming and 
increasing CAFE standards for passenger cars and further increasing the 
already reformed light truck standards. More specifically, the agency 
said:

    * * * we are seeking information related to fuel economy 
improvements for MY 2007-2017 passenger cars and MY 2010-2017 light 
trucks. The agency is seeking information in anticipation of 
obtaining statutory authority to reform the passenger car CAFE 
program and to set standards under that structure for MY 2010-2017 
passenger cars. The agency is also seeking this information in 
anticipation of setting standards for MY 2012-2017 light trucks.\32\
---------------------------------------------------------------------------

    \32\ 72 FR 8664; February 27, 2007.
---------------------------------------------------------------------------

5. Supreme Court Decision in Massachusetts v. EPA (April 2007)
    On April 2, 2007, the U.S. Supreme Court issued its opinion in 
Massachusetts v. EPA.\33\ The Court ruled that the state of 
Massachusetts had standing because it had already lost a small amount 
of land and stood to lose more due to global warming induced increases 
in sea level; that some portion of this harm was traceable to the 
absence of a regulation issued by EPA requiring reductions in GHG 
emissions (CO2 emissions, most notably) by motor vehicles; 
and that issuance of such an EPA regulation by EPA would reduce the 
risk of further harm to Massachusetts. On the merits, the Court ruled 
that greenhouse gases are ``pollutants'' under the Clean Air Act and 
that the Act therefore authorizes EPA to regulate greenhouse gas 
emissions from motor vehicles if EPA makes the necessary findings and 
determinations under section 202 of the Act.
---------------------------------------------------------------------------

    \33\ 127 S.Ct. 1438 (2007).
---------------------------------------------------------------------------

    The Court considered EPCA briefly, noting that it and the Clean Air 
Act have different overall purposes. It noted further that the two acts 
overlap, but did not define the nature or extent of that overlap. It 
concluded that EPCA did not relieve EPA of its statutory obligations 
and expressed confidence that the two acts could be consistently 
administered. The Court did not address the express preemption 
provision in EPCA.
6. Coordination Between NHTSA and EPA on Development of Rulemaking 
Proposals (Summer-Fall 2007)
    In the wake of the Supreme Court's decision and in the absence of 
the legislation he called for in his 2007 State of the Union message, 
the President called on NHTSA and EPA to take the first steps toward 
regulations that would cut gasoline consumption and greenhouse gas 
emissions from motor vehicles, using his Twenty-in-Ten initiative as a 
starting point. He asked them ``to listen to public input, to carefully 
consider safety, science, and available technologies, and evaluate the 
benefits and costs before they put forth the new regulation.'' He also 
issued an executive order directing all of the departments and agencies 
to work together on the proposal.
    Pursuant to the President's directive, NHTSA and EPA staff jointly 
assessed which technologies would be available and their effectiveness 
and cost. They also jointly assessed the key economic and other 
assumptions affecting the stringency of future standards. Finally, they 
worked together in updating and further improving the Volpe model that 
had been used to help determine the stringency of the MY 2008-2011 
light truck CAFE standards. Much of the work between NHTSA and EPA 
staff was reflected in rulemaking proposals being developed by NHTSA 
prior to the enactment of EISA and was substantially retained when 
NHTSA revised its proposals to be consistent with that legislation. 
Ultimately, the proposals being published today are based on NHTSA's 
assessments of how they meet EPCA, as amended by EISA.
7. Ninth Circuit Decision Re Final Rule for MY 2008-2011 Light Trucks 
(November 2007)
    On November 15, 2007, the United States Court of Appeals for the 
Ninth Circuit issued its decision in Center for Biological Diversity v. 
NHTSA,\34\ the challenge to the MY 2008-11 light truck CAFE rule. The 
Court rejected the petitioners' argument that EPCA precludes the use of 
a marginal cost-benefit analysis that attempted to weigh all of the 
social benefits (i.e., externalities as well as direct benefits to 
consumers) of improved fuel savings in determining the stringency of 
the CAFE standards. It cautioned, however, that it had not reviewed 
whether the agency's balancing of the statutory factors in setting 
those standards was arbitrary and capricious. In that regard, it noted 
that much had changed since a court of appeals had last (i.e., in the 
late 1980's) reviewed the agency's balancing of those factors in a 
rulemaking. Specifically, it noted increases in scientific knowledge of 
climate change

[[Page 24362]]

and in the need to reduce importation of petroleum since that time.
---------------------------------------------------------------------------

    \34\ 508 F.3d 508.
---------------------------------------------------------------------------

    Further, the Court found that NHTSA had been arbitrary and 
capricious in its treatment of the following issues:
     NHTSA's decision not to monetize the benefit of reducing 
CO2 emissions and use that value in conducting its marginal 
benefit-cost analysis based on its view that the value of the benefit 
of CO2 emission reductions resulting from fuel consumption 
reductions was too uncertain to permit the agency to determine a value 
for those emission reductions;\35\
---------------------------------------------------------------------------

    \35\ The agency has developed a value for those reductions and 
used it in the analyses underlying the standards proposed in this 
NPRM. For further discussion, see section V of this preamble.
---------------------------------------------------------------------------

     NHTSA's decision not to establish a ``backstop'' (i.e., a 
fixed minimum CAFE standard applicable to manufacturers); \36\
---------------------------------------------------------------------------

    \36\ EISA's requirement that standards be based on one or more 
vehicle attributes and its specification for domestic passenger 
cars, but not for nondomestic passenger cars or light trucks of an 
absolute CAFE level appear to preclude the specification of such a 
backstop standard for the latter two categories of automobiles. For 
further discussion, see Section VI of this preamble.
---------------------------------------------------------------------------

     NHTSA's decision not to proceed to revise the regulatory 
definitions for the passenger car and light truck categories of 
automobiles so that some vehicles currently classified as light trucks 
are instead classified as passenger cars; \37\
---------------------------------------------------------------------------

    \37\ In this NPRM, NHTSA examines the legislative history of the 
statutory definitions of ``automobile'' and ``passenger automobile'' 
and the term ``nonpassenger automobile'' and analyses the impact of 
that moving any vehicles out of the nonpassenger automobile (light 
truck) category into the passenger automobile (passenger car) 
category would have the level of standards for both groups of 
automobiles. For further discussion, see Section VIII of this 
preamble.
---------------------------------------------------------------------------

     NHTSA's decision not to subject most medium- and heavy-
duty pickups and most medium- and heavy-duty cargo vans (i.e., those 
between 8,500 and 10,000 pounds gross vehicle weight rating (GVWR,) to 
the CAFE standards; \38\
---------------------------------------------------------------------------

    \38\ EISA removed these vehicles from the statutory definition 
of ``automobile'' and mandated the establishment of CAFE standards 
for them following the completion of reports by the National Academy 
of Sciences and NHTSA.
---------------------------------------------------------------------------

     NHTSA's limited assessment of cumulative impacts and 
regulatory alternatives in its Environmental Assessment (EA) under the 
National Environmental Policy Act (NEPA), and its decision to prepare 
and publish an EA, coupled with a finding of no significant impact, 
instead of an Environmental Impact Statement (EIS).\39\
---------------------------------------------------------------------------

    \39\ On February 9, NHTSA filed a petition with the Ninth 
Circuit for rehearing en banc on the issue of whether the panel in 
CBD acted within its authority in ordering the agency to prepare an 
EIS instead of remanding the issue to the agency and directing it to 
conduct a new, fuller environmental analysis and decide whether an 
EIS is required. In addition, NHTSA has published a notice of intent 
to prepare an environmental impact statement, thus beginning the EIS 
process for this rulemaking, as discussed in Section XIII.B. of this 
NPRM.
---------------------------------------------------------------------------

    The Court did not vacate the standards, but instead said it would 
remand the rule to NHTSA to promulgate new standards consistent with 
its opinion ``as expeditiously as possible and for the earliest model 
year practicable.\40\ Under the decision, the standards established by 
the April 2006 final rule would remain in effect unless and until 
amended by NHTSA.
---------------------------------------------------------------------------

    \40\ The deadline in EPCA for issuing a final rule establishing, 
for the first time, a CAFE standard for a model year is 18 months 
before the beginning of that model year. 49 U.S.C. 32902(g)(2). The 
same deadline applies to issuing a final rule amending an existing 
CAFE standard so as to increase its stringency. Given that the 
agency has long regarded October 1 as the beginning of a model year, 
the statutory deadline for increasing the MY 2009 standard was March 
30, 2007, and the deadline for increasing the MY 2010 standard is 
March 30, 2008. Thus, the only model year for which there is 
sufficient time to gather all of the necessary information, conduct 
the necessary analyses and complete a rulemaking is MY 2011. As 
noted earlier in this document, however, EISA requires that a new 
standard be established for that model year. This rulemaking is 
being conducted pursuant to that requirement.
---------------------------------------------------------------------------

    On February 6, 2008, the Government petitioned for en banc 
rehearing by the Ninth Circuit on the limited issue of whether it was 
appropriate for the panel, having held that the agency insufficiently 
explored the environmental implications of the MY 2008-11 rulemaking in 
its EA, to order the agency to prepare an EIS rather than simply 
remanding the matter to the agency for further analysis.
    As of the date of the issuance of this proposal, the Court has not 
yet issued its mandate in this case.
8. Enactment of Energy Security and Independence Act of 2007 (December 
2007)
    As noted above in section I.B., EISA significantly changed the 
provisions of EPCA governing the establishment of future CAFE 
standards. These changes made it necessary for NHTSA to pause in its 
efforts so that it could assess the implications of the amendments made 
by EISA and then, as required, revise some aspects of the proposals it 
had been developing (e.g., the model years covered and credit issues).

C. Energy Policy and Conservation Act, as Amended

    EPCA, which was enacted in 1975, mandates a motor vehicle fuel 
economy regulatory program to improve the nation's energy security and 
energy efficiency. It gives the authority under EPCA to regulate fuel 
economy to DOT, which has delegated that authority to NHTSA at 49 CFR 
1.50. EPCA allocates the responsibility for implementing the program as 
follows: NHTSA sets CAFE standards for passenger cars and light trucks; 
EPA calculates the average fuel economy of each manufacturer's 
passenger cars and light trucks; and NHTSA enforces the standards based 
on EPA's calculations.
    We have summarized below EPCA, as amended by EISA. We request 
comment on how EPCA should be implemented to achieve the goals and meet 
the requirements of EISA. For example, what assumptions, methodologies 
and computations should be used in establishing and implementing the 
new standards?
1. Vehicles Subject to Standards for Automobiles
    With two exceptions, all four-wheeled motor vehicles with a gross 
vehicle weight rating of 10,000 pounds or less will be subject to the 
CAFE standards, beginning with MY 2011. The exceptions will be work 
trucks \41\ and multi-stage vehicles. Work trucks are defined as 
vehicles that are:
---------------------------------------------------------------------------

    \41\ While EISA excluded work trucks from ``automobiles,'' it 
did not exclude them from regulation under EPCA. EISA requires that 
work trucks be subjected to CAFE standards, but only first after the 
National Academy of Sciences completes a study and then after NHTSA 
completes a follow-on study. Congress thus recognized and made 
allowances for the practical difficulties that led NHTSA to decline 
to include work trucks in its final rule for MY 2008-11 light 
trucks.

--rated at between 8,500 and 10,000 pounds gross vehicle weight; and
--are not a medium-duty passenger vehicle (as defined in section 
86.1803-01 of title 40, Code of Federal Regulations, as in effect on 
the date of the enactment of the Ten-in-Ten Fuel Economy Act).\42\
---------------------------------------------------------------------------

    \42\ 49 U.S.C. 32902(a)(19).

Medium-duty passenger vehicles (MDPV) include 8,500 to 10,000 lb. GVWR 
sport utility vehicles (SUVs), short bed pick-up trucks, and passenger 
vans, but exclude pickup trucks with longer beds and cargo vans rated 
at between 8,500 and 10,000 lbs GVWR. It is those excluded pickup 
trucks and cargo vans that are work trucks. ``Multi-stage vehicle'' 
includes any vehicle manufactured in different stages by 2 or more 
manufacturers, if no intermediate or final-stage manufacturer of that 
vehicle manufactures more than 10,000 multi-stage vehicles per 
year.\43\
---------------------------------------------------------------------------

    \43\ 49 U.S.C. 32902(a)(3).
---------------------------------------------------------------------------

    Under EPCA, as it existed before EISA, the agency had discretion 
whether to regulate vehicles with a GVWR between 6,000 and 10,000 lbs., 
GVWR. It could regulate the fuel

[[Page 24363]]

economy of vehicles with a GVWR within that range under CAFE if it 
determined that (1) standards were feasible for these vehicles, and (2) 
either (a) that these vehicles were used for the same purpose as 
vehicles rated at not more than 6,000 lbs. GVWR, or (b) that their 
regulation would result in significant energy conservation.
    EISA eliminated the need for administrative determinations in order 
to subject vehicles between 6,000 and 10,000 lbs. GVWR to the CAFE 
standards for automobiles. Congress did so by making the determination 
itself that all vehicles within that GVWR range should be included, 
with the exceptions noted above.
2. Mandate To Set Standards for Automobiles
    As amended by EISA, EPCA requires that the agency establish 
standards for all new automobiles for each model year at the maximum 
feasible levels for that model year. A manufacturer's individual 
passenger cars and light trucks are not required to meet a particular 
fuel economy level. Instead, the harmonically averaged fuel economy of 
a manufacturer's production of passenger cars (or light trucks) in a 
particular model year must meet the standard for those automobiles for 
that model year.
    For model years 2011-2020, several special requirements, in 
addition to the maximum feasible requirement, are specified.\44\ Each 
of the requirements must be interpreted in light of the other 
requirements. For those model years, separate standards for passenger 
cars and for light trucks must be set at high enough levels to ensure 
that the CAFE of the industry wide combined fleet of new passenger cars 
and light trucks for MY 2020 is not less than 35 mpg. The 35 mpg figure 
is not a standard applicable to any individual manufacturer. It is a 
requirement, applicable to the agency, regarding the combined effect of 
the separate standards for passenger cars and light trucks that NHTSA 
is to establish for MY 2020. EISA does not specify precisely how 
compliance with this requirement is to be ensured or how or when the 
CAFE of the industry wide combined fleet for MY 2020 is to be 
calculated for purposes of determining compliance. As a practical 
matter, to ensure that this level is achieved, the standard for MY 2020 
passenger cars would have to be above 35 mpg and the one for MY 2020 
light trucks might or might not be below 35 mpg. Similarly, the CAFE of 
some manufacturers' combined fleet of passenger cars and light trucks 
would be above 35 mpg, while the combined fleet of others might or 
might not be below 35 mpg. The standards for passenger cars and those 
for light trucks must increase ratably each year. The CAFE of each 
manufacturer's fleet of domestic passenger cars must meet a sliding, 
absolute minimum level in each model year: 27.5 mpg or 92 percent of 
the projected CAFE of the industry wide fleet of new domestic passenger 
cars for that model year.
---------------------------------------------------------------------------

    \44\ Under EPCA, prior to its amendment by EISA, the standard 
for passenger cars was 27.5 mpg unless amended to a higher or lower 
level by DOT. Per EISA, the standard will remain at 27.5 mpg through 
MY 2010.
---------------------------------------------------------------------------

    EPCA, as it existed before EISA, EPCA required that light truck 
standards be set at the maximum feasible level for each model year, but 
simply specified a default standard of 27.5 mpg for passenger cars for 
MY 1985 and thereafter. It permitted, but did not require that NHTSA 
establish a higher or lower standard for passenger cars if the agency 
found that the maximum feasible level of fuel economy is higher or 
lower than 27.5 mpg.
3. Structure of Standards
    The standards for passenger cars and light trucks must be based on 
one or more vehicle attributes and expressed in terms of a mathematical 
function. This makes it possible to increase the CAFE standards for 
both passenger cars and light trucks significantly without creating 
incentives to improve fuel economy in ways that reduce safety. 
Formerly, EPCA provided authority for this approach for light trucks, 
but not passenger cars.
4. Factors Governing or Considered in the Setting of Standards
    In determining the maximum feasible level of average fuel economy 
for a model year, EPCA requires that the agency consider four factors: 
technological feasibility, economic practicability, the effect of other 
standards of the Government on fuel economy, and the need of the nation 
to conserve energy. EPCA does not define these terms or specify what 
weight to give each concern in balancing them; thus, NHTSA defines them 
and determines the appropriate weighting based on the circumstances in 
each CAFE standard rulemaking.
    ``Technological feasibility'' means whether a particular method of 
improving fuel economy can be available for commercial application in 
the model year for which a standard is being established.
    ``Economic practicability'' means whether a standard is one 
``within the financial capability of the industry, but not so stringent 
as to'' lead to ``adverse economic consequences, such as a significant 
loss of jobs or the unreasonable elimination of consumer choice.'' \45\ 
In an attempt to ensure the economic practicability of attribute based 
standards, the agency considers a variety of factors, including the 
annual rate at which manufacturers can increase the percentage of its 
fleet that has a particular type of fuel saving technology, and cost to 
consumers. Since consumer acceptability is an element of economic 
practicability, the agency has limited its consideration of fuel saving 
technologies to be added to vehicles to those that provide benefits 
that match their costs. Disproportionately expensive technologies are 
not likely to be accepted by consumers.
---------------------------------------------------------------------------

    \45\ 67 FR 77015, 77021; December 16, 2002.
---------------------------------------------------------------------------

    At the same time, the law does not preclude a CAFE standard that 
poses considerable challenges to any individual manufacturer. The 
Conference Report for EPCA, as enacted in 1975, makes clear, and the 
case law affirms, ``(A) determination of maximum feasible average fuel 
economy should not be keyed to the single manufacturer which might have 
the most difficulty achieving a given level of average fuel 
economy.''\46\ Instead, the agency is compelled ``to weigh the benefits 
to the nation of a higher fuel economy standard against the 
difficulties of individual automobile manufacturers.'' Id. The law 
permits CAFE standards exceeding the projected capability of any 
particular manufacturer as long as the standard is economically 
practicable for the industry as a whole. Thus, while a particular CAFE 
standard may pose difficulties for one manufacturer, it may also 
present opportunities for another. The CAFE program is not necessarily 
intended to maintain the competitive positioning of each particular 
company. Rather, it is intended to enhance fuel economy of the vehicle 
fleet on American roads, while protecting motor vehicle safety and the 
totality of American jobs and the overall United States economy.
---------------------------------------------------------------------------

    \46\ CEI-I, 793 F.2d 1322, 1352 (DC Cir. 1986).
---------------------------------------------------------------------------

    ``The effect of other motor vehicle standards of the Government on 
fuel economy'' means ``the unavoidable adverse effects on fuel economy 
of compliance with emission, safety, noise, or damageability 
standards.'' In the case of emission standards, this includes standards 
adopted by the Federal government and can include standards adopted by 
the States as well, since in certain circumstances the Clean Air Act

[[Page 24364]]

permits States to adopt and enforce State standards in lieu of the 
Federal ones. It does not, however, include State standards expressly 
preempted by EPCA.\47\
---------------------------------------------------------------------------

    \47\ 49 U.S.C. 32919 and 71 FR 17566, 17654-70; April 6, 2006.
---------------------------------------------------------------------------

    ``The need of the United States to conserve energy'' means ``the 
consumer cost, national balance of payments, environmental, and foreign 
policy implications of our need for large quantities of petroleum, 
especially imported petroleum.'' Environmental implications principally 
include reductions in emissions of criteria pollutants and carbon 
dioxide. A prime example of foreign policy implications are energy 
independence and security concerns.
    The agency has considered environmental issues in making decisions 
about the setting of standards from the earliest days of the CAFE 
program. As the three courts of appeal have noted in decisions 
stretching over the last 20 years,\48\ the agency defined the ``need of 
the Nation to conserve energy'' in the late 1970's as including ``the 
consumer cost, national balance of payments, environmental, and foreign 
policy implications of our need for large quantities of petroleum, 
especially imported petroleum.'' \49\ Pursuant to that view, the agency 
declined to include diesel engines in determining the maximum feasible 
level of average fuel economy for passenger cars and for light trucks 
because particulate emissions from diesels were then both a source of 
concern and unregulated.\50\ In the late 1980's, NHTSA cited concerns 
about climate change as one of its reasons for limiting the extent of 
its reduction of the CAFE standard for MY 1989 passenger cars \51\ and 
for declining to reduce the standard for MY 1990 passenger cars.\52\ 
Since then, DOT has considered the indirect benefits of reducing 
tailpipe carbon dioxide emissions in its fuel economy rulemakings 
pursuant to the statutory requirement to consider the nation's need to 
conserve energy by reducing consumption. In this rulemaking, consistent 
with the Ninth Circuit's decision and its observations about the 
potential effect of changing information about climate change on the 
balancing of the EPCA factors and aided by the 2007 reports of the 
United Nations Intergovernmental Panel on Climate Change \53\ and other 
information, NHTSA is monetizing the reductions in tailpipe emissions 
of CO2 that will result from the CAFE standards and is 
proposing to set the MY 2011-15 CAFE standards at levels that reflect 
the value of those reductions in CO2. as well as the value 
of other benefits of those standards. In setting CAFE standards, NHTSA 
also considers environmental impacts under NEPA, 42 U.S.C. 4321-4347.
---------------------------------------------------------------------------

    \48\ Center for Auto Safety v. NHTSA, 793 F.2d 1322, 1325 n. 12 
(DC Cir. 1986); Public Citizen v. NHTSA, 848 F.2d 256, 262-3 n. 27 
(DC Cir. 1988) (noting that ``NHTSA itself has interpreted the 
factors it must consider in setting CAFE standards as including 
environmental effects''); and Center for Biological Diversity v. 
NHTSA, 508 F.3d 508, 529 (9th Cir. 2007).
    \49\ 42 FR 63,184, 63,188 (Dec. 15, 1977) (emphasis added).
    \50\ For example, the final rules establishing CAFE standards 
for MY 1981-84 passenger cars, 42 FR 33,533, 33,540-1 and 33,551; 
June 30, 1977, and for MY 1983-85 light trucks, 45 FR 81,593, 
81,597; December 11, 1980.
    \51\ 53 FR 39,275, 39,302; October 6, 1988.
    \52\ 54 FR 21985,
    \53\ The IPCC 2007 reports can be found at http://www.ipcc.ch/. 
(Last accessed April 20, 2008.)
---------------------------------------------------------------------------

    In addition, the agency is permitted to consider additional 
relevant societal considerations. For example, historically, it has 
considered the potential for adverse safety consequences when deciding 
upon a maximum feasible level. This practice is sanctioned in case 
law.\54\
---------------------------------------------------------------------------

    \54\ See, e.g., Center for Auto Safety v. NHTSA (CAS), 793 F. 2d 
1322 (DC Cir. 1986) (Administrator's consideration of market demand 
as component of economic practicability found to be reasonable); 
Public Citizen 848 F.2d 256 (Congress established broad guidelines 
in the fuel economy statute; agency's decision to set lower standard 
was a reasonable accommodation of conflicting policies). As the 
United States Court of Appeals pointed out in upholding NHTSA's 
exercise of judgment in setting the 1987-1989 passenger car 
standards, ``NHTSA has always examined the safety consequences of 
the CAFE standards in its overall consideration of relevant factors 
since its earliest rulemaking under the CAFE program.'' Competitive 
Enterprise Institute v. NHTSA (CEI I), 901 F.2d 107, 120 at n.11 (DC 
Cir. 1990).
---------------------------------------------------------------------------

    EPCA requires that the MY 2011-2019 CAFE standards for passenger 
cars and for light trucks must both increase ratably to at least the 
levels necessary to meet 35 mpg requirement for MY 2020. NHTSA 
interprets this to mean that the standards must make steady progress 
toward the levels necessary for the average fuel economy of the 
combined industry wide fleet of all new passenger cars and light trucks 
sold in the United States during MY 2020 to reach at least 35 mpg.
    Finally, EPCA provides that in determining the level at which it 
should set CAFE standards for a particular model year, NHTSA may not 
consider the ability of manufacturers to take advantage of several EPCA 
provisions that facilitate compliance with the CAFE standards and 
thereby reduce the costs of compliance. As noted below in Section II, 
manufacturers can earn compliance credits by exceeding the CAFE 
standards and then use those credits to achieve compliance in years in 
which their measured average fuel economy falls below the standards. 
Manufacturers can also increase their CAFE levels through MY 2019 by 
producing alternative fuel vehicles. EPCA provides an incentive for 
producing these vehicles by specifying that their fuel economy is to be 
determined using a special calculation procedure that results in those 
vehicles being assigned a high fuel economy level.
5. Consultation in Setting Standards
    EPCA provides that NHTSA is to consult with the Department of 
Energy (DOE) and Environmental Protection Agency in prescribing CAFE 
standards. It provides further that NHTSA is to provide DOE with an 
opportunity to provide written comments on draft proposed and final 
CAFE standards.\55\
---------------------------------------------------------------------------

    \55\ In addition, Executive Order No. 13432 provides that a 
Federal agency undertaking a regulatory action that can reasonably 
be expected to directly regulate emissions, or to substantially and 
predictably affect emissions, of greenhouse gases from motor 
vehicles, shall act jointly and consistently with other agencies to 
the extent possible and to consider the views of other agencies 
regarding such action.
---------------------------------------------------------------------------

6. Compliance Flexibility and Enforcement
    EPCA specifies a precise formula for determining the amount of 
civil penalties for failure to comply with a standard. The penalty, as 
adjusted for inflation by law, is $5.50 for each tenth of a mpg that a 
manufacturer's average fuel economy falls short of the standard for a 
given model year multiplied by the total volume of those vehicles in 
the affected fleet (i.e., import or domestic passenger car, or light 
truck), manufactured for that model year. The amount of the penalty may 
not be reduced except under the unusual or extreme circumstances 
specified in the statute.
    Likewise, EPCA provides that manufacturers earn credits for 
exceeding a standard. The amount of credit earned is determined by 
multiplying the number of tenths of a mpg by which a manufacturer 
exceeds a standard for a particular category of automobiles by the 
total volume of automobiles of that category manufactured by the 
manufacturer for a given model year.
    EPA is responsible for measuring automobile manufacturers' CAFE so 
that NHTSA can determine compliance with the CAFE standards. In making 
these measurements for passenger cars, EPA is required by EPCA \56\ to 
use the EPA test

[[Page 24365]]

procedures in place as of 1975 (or procedures that give comparable 
results), which are the city and highway tests of today, with 
adjustments for procedural changes that have occurred since 1975.
---------------------------------------------------------------------------

    \56\ 49 U.S.C. 32904(c).
---------------------------------------------------------------------------

    EPA's fuel economy test procedures specify equations for 
calculating fuel economy. These equations are based on the carbon 
balance technique which allows fuel economy to be determined from 
measurement of exhaust emissions. This technique relies upon the 
premise that the quantity of carbon in a vehicle's exhaust gas is equal 
to the quantity of carbon consumed by the engine as fuel.
    When NHTSA finds that a manufacturer is not in compliance, it 
notifies the manufacturer. Surplus credits generated from the five 
previous years can be used to make up the deficit. If there are no (or 
not enough) credits available, then the manufacturer can either pay the 
fine, or submit a carry back plan to the agency. A carry back plan 
describes what the manufacturer plans to do in the following three 
model years to make up for the deficit in credits. NHTSA must examine 
and determine whether to approve the plan.

III. Fuel Economy Enhancing Technologies

    In the Agency's last two rulemakings covering light truck CAFE 
standards for MYs 2005-2007 and MYs 2008-2011, the agency relied on the 
2002 National Academy of Sciences' report, Effectiveness and Impact of 
Corporate Average Fuel Economy Standards (``the 2002 NAS Report'') \57\ 
for estimating potential fuel economy benefits and associated retail 
costs of applying combinations of technologies in 10 classes of 
production vehicles. The NAS cost and effectiveness numbers were the 
best available estimates at this time, determined by a panel of experts 
formed by the National Academy of Sciences, and the report had been 
peer reviewed by individuals chosen for their diverse perspectives and 
technical expertise in accordance with procedures approved by the 
Report Review Committee of the National Research Council. However, 
since the publication of the 2002 NAS Report, there has been 
substantial advancement in fuel-saving technologies, including 
technologies not discussed in the NAS Report that are expected to 
appear on vehicles in the MY 2011-2015 timeframe. There also have been 
reports issued and studies conducted by several other organizations and 
companies that discuss fuel economy technologies and their benefits and 
costs. NHTSA has contracted with the NAS to update the fuel economy 
section, Chapter 3, of the 2002 NAS Report. However, this update will 
not be available in time for this rulemaking. Due to the expedited 
nature of this rulemaking, NHTSA, in consultation with the 
Environmental Protection Agency (EPA), developed an updated technology 
cost and effectiveness list to be used in this document.
---------------------------------------------------------------------------

    \57\ National Research Council, ``Effectiveness and Impact of 
Corporate Average Fuel Economy (CAFE) Standards,'' National Academy 
Press, Washington, DC (2002). Available at http://www.nap.edu/
openbook.php?isbn=0309076013 (last accessed April 20, 2008).
---------------------------------------------------------------------------

    This list presents NHTSA and EPA technical staff's current 
assessment of the costs and effectiveness from a broad range of 
technologies which can be applied to cars and light-duty trucks. EPA 
published the results of this collaboration in a report and submitted 
it to the NAS committee.\58\ A copy of the report and other studies 
used in the technology update will be placed in NHTSA's docket.
---------------------------------------------------------------------------

    \58\ EPA Staff Technical Report: Cost and Effectiveness 
Estimates of Technologies Used to Reduce Light-duty Vehicle Carbon 
Dioxide Emissions. EPA420-R-08-008, March, 2008.
---------------------------------------------------------------------------

    NHTSA believes that the estimates used for this document, which 
rely on the best available public and confidential information, are 
defensible and reasonable predictions for the next five years. 
Nevertheless, NHTSA still believes that the ideal source for this 
information comes from a peer reviewed process such as the NAS. NHTSA 
will continue to work with NAS to update this list on a five year 
interval as required by the Energy Independence and Security Act of 
2007.
    The majority of the technologies discussed in this section are in 
production and available on vehicles today, either in the United 
States, Japan, or Europe. A number of the technologies are commonly 
available, while others have only recently been introduced into the 
market. In a few cases, we provide estimates on technologies which are 
not currently in production, but are expected to be so in the next few 
years. These are technologies which can be applied to cars and trucks 
that are capable of achieving significant improvements in fuel economy 
and reductions in carbon dioxide emissions, and improve vehicle fuel 
economy, at reasonable costs.
    NHTSA and EPA conducted the technology examination using concepts 
from the 2002 NAS report which constituted a starting point for the 
analysis. In the NAS Report, there were three exemplary technology 
paths or scenarios identified for each class of production vehicles, 
which lead to successively greater improvements in fuel consumption and 
greater costs. Path I included production-intent technologies that will 
be available within 10 years and could be implemented under current 
economic and regulatory conditions. Path II included more costly 
production-intent technologies that are technically feasible for 
introduction within 10 years if economic and regulatory conditions 
justify their use. Path III included emerging technologies that will be 
available within 10 to 15 years but that may require further 
development prior to commercial introduction. These three paths 
represented vehicle development steps that would offer increasing 
levels of fuel economy gains (as incremental gains) at incrementally 
increasing cost. As stated earlier, since the publication of the 2002 
NAS Report, automotive technology has continued to advance and many of 
the technologies that were identified in the report as emerging have 
already entered the marketplace.
    In this rulemaking, NHTSA in consultation with EPA have examined a 
variety of technologies, looking beyond path I and path II to path III 
and to emerging technologies beyond path III. These technologies were 
in their infancy when the 2002 NAS Report was being formulated. In 
addition, unlike for past rulemakings where NHTSA projected the use of 
different variants of a technology as a combined technology, in this 
rulemaking, NHTSA working with EPA examined advanced forms and 
subcategories of existing technologies and reflected the effectiveness 
and cost for each of the variants separately for all ten vehicle 
classes. The specific technologies affected are variable valve timing 
(VVT), variable valve lift and timing (VVLT) and cylinder deactivation. 
Manufacturers are currently using many different types of VVTs and 
VVLTs, which have a variety of different names and methods. This 
rulemaking employs specific cost and effectiveness estimates for 
variants of VVT, including Intake Camshaft Phasing (ICP), Coupled 
Camshaft Phasing (CCP), and Dual (Independent) Camshaft Phasing (DCP). 
It also employs specific cost and effectiveness estimates for variants 
of VVLT, including Discrete Variable Valve Lift (DVVL) and Continuous 
Variable Valve Lift (CVVL). We also now include the effectiveness and 
cost estimates for each of the variants of cylinder deactivation. The 
most common type of cylinder deactivation is one in which an eight-
cylinder overhead

[[Page 24366]]

valve engine disables four of its cylinders under light loads. Cylinder 
deactivation could be incorporated on overhead cam engines, and can be 
applied to four and six cylinder engines as well (we have restricted 
application to 6 and 8 cylinder engines). Thus, the variants of 
cylinder deactivation that now have specific cost and effectiveness 
estimates include both overhead valve engine cylinder deactivation and 
overhead cam engine cylinder deactivation.
    The update also revisited technology lead time issues and took a 
fresh look at technology application rates, how to link certain 
technologies to certain redesign and refresh patterns, synergistic 
impacts resulting from adding technology packaging, and learning costs.

A. Data Sources for Technology Assumptions

    A large number of technical reports and papers are available which 
contain data and estimates of the fuel economy improvements of various 
vehicle technologies. In addition to specific peer-reviewed papers 
respecting individual technologies, we also utilized a number of recent 
reports which had been utilized by various State and Federal Agencies 
and which were specifically undertaken for the purpose of estimating 
future vehicle fuel economy reduction effectiveness or improvements in 
fuel economy. The reports we utilized most frequently were:
     2002 National Academy of Science (NAS) report titled 
``Effectiveness and Impact of Corporate Average Fuel Economy 
Standards''. At the time it was published, the NAS report was 
considered by many to be the most comprehensive summary of current and 
future fuel efficiencies improvements which could be obtained by the 
application of individual technologies. The focus of this report was 
fuel economy, which can be directly correlated with CO2 
emissions. The 2002 NAS report contains effectiveness estimates for ten 
different vehicle classifications (small car, mid-SUV, large truck, 
etc), but did not differentiate these effectiveness values across the 
classes. Where other sources or engineering principles indicated that a 
differentiation was warranted, we utilized the 2002 NAS effectiveness 
estimates as a starting point and further refined the estimate to one 
of the vehicle classes using engineering judgment or by consulting 
additional reliable sources.
     2004 Northeast States Center for a Clean Air Future 
(NESCCAF) report ``Reducing Greenhouse Gas Emissions from Light-Duty 
Motor Vehicles''. This report, which was utilized by the California Air 
Resources Board for their 2004 regulatory action on vehicle 
CO2 emissions, includes a comprehensive vehicle simulation 
study undertaken by AVL, a world-recognized leader in automotive 
technology and engineering. In addition, the report included cost 
estimates developed by the Martec Group, a market-based research and 
consulting firm which provides services to the automotive industry. The 
NESCCAF report considered a number of technologies not examined in the 
2002 NAS report. In addition, through the use of vehicle simulation 
modeling, the 2004 NESCCAF report provides a scientifically rigorous 
estimation of the synergistic impacts of applying multiple fuel economy 
technologies to a given vehicle.
     2006 Energy and Environmental Analysis Inc (EEA) report 
``Technology to Improve the Fuel Economy of Light Duty Trucks to 2015'' 
Prepared for The U.S. Department of Energy and The U.S. Department of 
Transportation. This update of technology characteristics is based on 
new data obtained by EEA from technology suppliers and auto-
manufacturers, and these data are compared to data from studies 
conducted earlier by EEA, the National Academy of Sciences (NAS), the 
Northeast States Center for a Clean Air future (NESCCAF) and California 
Air Resources Board (CARB).
     Data from Vehicle Manufacturers, Component Suppliers, and 
other reports. We also evaluated confidential data from a number of 
vehicle manufacturers as well as a number of technology component 
suppliers. In February of 2007, the NHTSA published a detailed Request 
for Comment (RFC) in the Federal Register. This RFC included, among 
other items, a request for information from automotive manufacturers 
and the public on the fuel economy improvement potential of a large 
number of vehicle technologies. The manufacturer's submissions to this 
RFC were supplemented by confidential briefing and data provided by 
vehicle component suppliers, who for many of the technologies 
considered are the actual manufacturers of the specific technology and 
often undertake their own development and testing efforts to 
investigate the fuel economy improvement potential of their products. 
Manufacturers that provided NHTSA and EPA with fuel economy cost and 
effectiveness estimates include BMW, Chrysler, Ford, General Motors, 
Honda, Nissan, Toyota and Volkswagen. The major suppliers that provided 
NHTSA with fuel economy cost and effectiveness estimates include Borg-
Warner, Bosch, Corning, Delphi, and Siemens.
     Finally, to verify that the fuel economy cost and 
effectiveness estimates for each of the technologies was reasonable and 
within currently available estimates for these technologies, NHTSA 
examined those estimates provided by other reports or sources, such as 
the Martec (contained in the 2004 NESCAFF report) and Sierra Research 
reports.\59\

B. Technologies and Estimates of Costs and Effectiveness

    This section describes each technology and associated cost and 
effectiveness numbers. The technologies can be classified into five 
main groups similar to how they were classified in the NAS Report: 
engine technologies; transmission technologies; accessory technologies; 
vehicle technologies; and hybrid technologies.
    While NHTSA and EPA followed the general approach taken by the NAS 
in estimating the cost and effectiveness numbers, we decided to update 
some of these estimates to reflect better the changed marketplace and 
regulatory environment, as well as the advancement in and greater 
penetration of some production-intent and emerging technologies, which 
have led to lower costs. The values contained in the 2002 NAS report 
were used to establish a baseline for the fuel economy cost and 
effectiveness estimates for each of the technologies. We then examined 
all other estimates provided by manufacturers and major suppliers or 
other sources. In examining these values, we gave more weight to values 
or estimates provided by manufacturers that have already implemented 
these technologies in their fleet, especially those that have 
introduced them in the largest quantities. Likewise, for technologies 
that have not penetrated the fleet to date, but will by early in the 
next decade (according to confidential manufacturer plans), we gave 
more weight to values or estimates provided by manufacturers that have 
stated that they will be introducing these technologies in their fleet, 
especially those that plan to introduce them in the largest quantities. 
In addition, for the technologies that will appear on vehicles by early 
in the next decade, we carefully examined the values provided

[[Page 24367]]

by those suppliers who have developed these technologies and may have 
contracts in place to provide them to manufacturers.
    Because not all technologies can be applied on all types of 
vehicles, engines or transmissions, we separately evaluated 10 classes 
of vehicles to estimate fuel economy cost and effectiveness for each of 
the technologies. As discussed above, these ten classes, also used in 
NHTSA's 2006 light truck CAFE rule, were derived from the 2002 NAS 
Report, which estimated the feasibility, potential incremental fuel 
consumption benefit and the incremental cost of three product 
development paths for the following ten vehicle classes: Subcompact 
passenger cars, compact passenger cars, midsize passenger cars, large 
passenger cars, small sport utility vehicles, midsize sport utility 
vehicles, large sport utility vehicles, small pickups, large pickups, 
and minivans.
---------------------------------------------------------------------------

    \59\ ``Alternative and Future Technologies for Reducing 
Greenhouse Gas Emission from Road Vehicles'' Sierra Research Report 
for Environment Canada, 1999 (SR99-07-01). http://
www.sierraresearch.com/ReportListing.htm (Last accessed April 20, 
2008.)
---------------------------------------------------------------------------

    The application of technologies to a vehicle class is limited not 
only by whether the manufacturer is capable of applying it within a 
particular development cycle, but also by whether the technology may 
physically be applied to the vehicle. For example, continuously 
variable transmissions (CVTs) were only allowed to be projected on 
vehicles with unibody construction, which includes all passenger cars 
and minivans and some small and midsize SUVs. CVTs could not be 
projected for use on vehicles with ladder-frame construction, which 
includes all pickups and large SUVs and some small and midsize SUVs. 
Another example is cylinder deactivation being limited to vehicles with 
6- or 8-cylinder engines. To simplify the analysis, NHTSA assumed that 
each class of vehicles would typically have vehicle construction and 
engines with a specific number of cylinders that is most representative 
of that vehicle class.
    Although we looked at ten vehicle classes separately, for some 
technologies the estimated incremental fuel consumption benefit and 
incremental cost were the same across all vehicle classes (as for 
engine accessory improvement), while for other technologies the 
estimated incremental fuel consumption benefit and incremental cost 
differed across classes (as for hybrid drivetrains). The main 
difference was with which path(s) each technology was expected to be 
associated.
    The exact cost and benefit of a given technology depends on 
specific vehicle characteristics (size, weight, base engine, etc.) and 
the existence of additional technologies that were already applied to 
the vehicle. In the section below, ranges of incremental cost and fuel 
consumption reduction values are listed where the values depend on 
vehicle characteristics and are independent of the order in which they 
are applied to a vehicle. All costs, which are reflective of estimated 
retail price equivalents (RPEs) were inflated by the producer price 
index (if needed) and are presented in year 2006 dollars, because this 
is the last year for which final economic indexing is available. Some 
cost estimates are based on supplier costs. In those instances, 
multipliers were included in those costs so that they would be treated 
in the same manner as cost estimates that are based on manufacturer 
costs. These incremental values were calculated by subtracting out all 
same-path synergies associated with a given technology and any 
preceding items on the same path. Essentially, the incremental percent 
reduction in fuel consumption and cost impacts represent improvements 
beyond the ones realized due to technologies already applied to the 
vehicle. As an example, a 5-speed automatic transmission could 
incrementally reduce fuel consumption by 2 to 3 percent at an 
incremental cost of $75 to $165 per vehicle, relative to a 4-speed 
automatic transmission. In turn, a 6-speed automatic transmission could 
incrementally reduce fuel consumption by 4.5 to 6.5 percent at an 
incremental cost of $10 to $20 per vehicle, relative to a 5-speed 
transmission.
    NHTSA acknowledges that this approach is different from the one it 
followed in establishing the reformed light truck standards for MYs 
2008-2011, where we relied nearly exclusively on the 2002 NAS report's 
estimates. Our preference remains to rely upon peer-review and credible 
studies, such as the 2002 NAS report; however we believe that the 
estimates made by the joint EPA/NHTSA team are accurate and defensible. 
The agency seeks comments on our assumptions and the cost, 
effectiveness and availability estimates provided. NHTSA also seeks 
comments on whether the order in which these technologies was applied 
by the Volpe model is proper and whether we have accurately accounted 
for technologies already included on vehicles and whether we have 
accurately accounted for technologies that are projected to be applied 
to vehicles. The agency also seeks comments on the ``synergy'' factors 
(discussed below) it has applied in order to adjust the estimated 
incremental effectiveness of some pairs of technology and on whether 
similar adjustments to the estimated incremental cost of some 
technologies should be made. In preparation for a final rule, NHTSA 
intends to update its technology-related methodologies and estimates, 
and expects that these anticipated updates will affect the form and 
stringency of the final standards.
a. Engine Technologies

Low-Friction Lubricants

    The use of lower viscosity engine and transmission lubricants can 
reduce fuel consumption. More advanced multi-viscosity engine and 
transmission oils are now available with improved performance in a 
wider temperature band, with better lubricating properties. However, 
even without any changes to fuel economy standards, most MY 2011-2015 
vehicles are likely to use 5W-30 motor oil, and some will use even less 
viscous oils, such as 5W-20 or possibly even 0W-20 to reduce cold start 
friction. This may directionally benefit the fuel economy improvements 
of valvetrain technologies such as cylinder deactivation, which rely on 
a minimum oil temperature (viscosity) for operation. Most manufacturers 
therefore attributed smaller potential fuel economy reductions and cost 
increases to lubricant improvements.
    The NAS Report estimated that low-friction lubricants could 
incrementally reduce fuel consumption by 1 percent at an incremental 
cost of $8 to $11.\60\ The NESCCAF study projected that low-friction 
lubricants could incrementally reduce fuel consumption by 1 percent at 
an incremental cost of $5 to $15; while the EEA report projected that 
low-friction lubricants could incrementally reduce fuel consumption by 
1 percent at an incremental cost of $10 to $20. In contrast, 
manufacturer data projected an estimated fuel consumption potential of 
0 percent to 1 percent at an incremental cost that ranged from $1 to 
$11, with many of them stating the costs as ranging from $1 to $5. 
NHTSA believes that these manufacturer estimates are more accurate and 
estimates that low-friction lubricants could reduce fuel consumption by 
0.5 percent for all vehicle types at an incremental cost of $3, which 
represents the mid-point of $2.50, rounded up to the next dollar.
---------------------------------------------------------------------------

    \60\ The price increases noted in this chapter are slightly 
higher than shown in the NAS study, since they have been converted 
into calendar year 2006 prices.
---------------------------------------------------------------------------

Reduction of Engine Friction Losses

    All reciprocating and rotating components in the engine are 
candidates for friction reduction, and minute improvements in several

[[Page 24368]]

components can add to a measurable fuel economy improvement. The amount 
of energy an engine loses to friction can be reduced in a variety of 
ways. Improvements in the design of engine components and subsystems 
will result in friction reduction, improved engine operation, greater 
fuel economy and reduced emissions. Examples include low-tension piston 
rings, roller cam followers, crankshaft design, improved material 
coatings, material substitution, more optimal thermal management, 
piston surface treatments, and as lubricant friction reduction. 
Additionally, as computer-aided modeling software continues to improve, 
more opportunities for incremental friction reduction might become 
apparent. Even without any changes to fuel economy standards, most MY 
2010-2015 vehicles are likely to employ one or more such techniques to 
reduce engine friction and other mechanical and hydrodynamic losses.
    The NAS Report estimated that such technologies could incrementally 
reduce fuel consumption by 1 to 5 percent at an incremental cost of $36 
to $146. NESCCAF predicted that such technologies could incrementally 
reduce fuel consumption by 0.5 percent at an incremental cost of $5 to 
$15; while the EEA report predicted that such technologies could reduce 
fuel consumption at an incremental cost of $10 to $55. Confidential 
manufacturer data indicates that engine friction reduction could 
incrementally reduce fuel consumption by 1 to 3 percent at an 
incremental cost of $0 to $168. Based on available information from 
these reports and confidential manufacturer data, NHTSA estimates that 
friction reduction could reduce fuel consumption for all vehicles by 1 
to 3 percent at a cost of $21 per cylinder. Thus, the incremental cost 
of engine friction reduction for a 4-cylinder engine is $0 to $84 
(applicable to subcompact and compact cars); for a 6-cylinder engine is 
$0 to $126 (applicable to midsize cars, large cars, small pickups, 
small SUVs, minivans and midsize SUVs); and for an 8-cylinder engine is 
$0 to $168 (applicable to large pickups and SUVs).

Multi-Valve Overhead Camshaft Engine

    It appears likely that many vehicles would still use overhead valve 
(OHV) engines with pushrods and one intake and one exhaust valve per 
cylinder during the early part of the next decade. Engines with 
overhead cams (OHC) and more than two valves per cylinder achieve 
increased airflow at high engine speeds and reductions of the valve 
train's moving mass and enable central positioning of spark plugs. Such 
engines, which are already used in some light trucks, typically develop 
higher power at high engine speeds. The NAS Report projected that 
multi-valve OHC engines could incrementally reduce fuel consumption by 
2 percent to 5 percent at an incremental cost of $109 to $146, and 
NHTSA found no sources to update these projections.
    For purposes of this rule, OHV engines and OHC engines were 
considered separately, and the model was generally not allowed to apply 
multivalve OHC technology to OHV engines, except where continuous 
variable valve timing and lift (CVVL) is applied to OHV engines. In 
that case, the model assumes conversion to DOHC valvetrain, because 
DOHC valvetrains are prerequisites for the application of any advanced 
engine technology over and above CVVL. Since applying CVVL to an OHV is 
the last improvement that could be made to such an engine, it's logical 
to assume that manufacturers would redesign that engine as a DOHC and 
include CVVL as part of that redesign.
    For 4-cylinder engines we estimated that the cost to redesign an 
OHV engine as a DOHC that includes CVVL would be $599 ($169 for 
conversion to DVVL, $254 for conversion to CVVL, and $176 for 
conversion to DOHC, which comprises an additional camshaft and valves), 
with estimated fuel consumption reduction of 2 to 3 percent. For 6-
cylinder engines we estimated that the cost to redesign an OHV engine 
as a DOHC that includes CVVL would be $1262 ($246 for conversion to 
DVVL, $488 for conversion to CVVL, and $550 for conversion to DOHC, 
which comprises an additional camshaft and valves), with estimated fuel 
consumption reduction of 1 to 4 percent. For 8-cylinder engines we 
estimated that the cost to redesign an OHV engine as a DOHC that 
includes CVVL would be $1380 ($322 for conversion to DVVL, $508 for 
conversion to CVVL, and $550 for conversion to DOHC, which comprises an 
additional camshaft and valves), with estimated fuel consumption 
reduction of 2 to 3 percent. Incremental cost estimates for DVVL and 
CVVL are discussed below.
    NHTSA believes that the NESCCAF report and confidential 
manufacturer data are more accurate, and thereby estimates that a 
conversion of an OHV engine to a DOHC engine with CVVL could 
incrementally reduce fuel consumption by 1 to 4 percent at an 
incremental cost of $599 to $1,380 compared to an OHV with VVT.

Cylinder Deactivation

    For the vast majority of vehicles, each cylinder is always active 
while the engine is running. Under partial load conditions, the 
engine's specific fuel consumption could be reduced if some cylinders 
could be disabled, such that the active cylinders operate at higher 
load. In cylinder deactivation, some (usually half) of the cylinders 
are ``shut down'' during light load operation--the valves are kept 
closed, and no fuel is injected--as a result, the trapped air within 
the deactivated cylinders is simply compressed and expanded as an air 
spring, with minimal friction and heat losses. The active cylinders 
combust at almost double the load required if all of the cylinders were 
operating. Pumping losses are significantly reduced as long as the 
engine is operated in this ``part-cylinder'' mode.
    The theoretical engine operating region for cylinder deactivation 
is limited to no more than roughly 50 percent of peak power at any 
given engine speed. In practice, however, cylinder deactivation is 
employed primarily at lower engine cruising loads and speeds, where the 
transitions in and out of deactivation mode are less apparent to the 
operator and where the noise and vibration (NVH) associated with fewer 
firing cylinders may be less of an issue. Manufacturers are exploring 
the possibilities of increasing the amount of time that part-cylinder 
mode might be suitable to a vehicle with more refined powertrain and 
NVH treatment strategies.
    General Motors and Chrysler Group have incorporated cylinder 
deactivation across a substantial portion of their V8-powered lineups. 
Honda (Odyssey, Pilot) and General Motors (Impala, Monte Carlo) offer 
V6 models with cylinder deactivation.
    There are two variants of cylinder deactivation. The most common 
type of cylinder deactivation is one in which an eight-cylinder 
overhead valve engine disables four cylinders under light loads. Thus 
an eight-cylinder engine could disable four cylinders under light 
loads, such as when the vehicle is cruising at highway speed. This 
technology could be applied to four and six cylinder engines as well. 
General Motors and Chrysler Group have incorporated cylinder 
deactivation across a substantial portion of their V8-powered overhead 
valve lineups.
    Cylinder deactivation could be incorporated on overhead cam engines 
and can be applied to four and six cylinder engines as well. Honda has 
already begun offering three V6 models

[[Page 24369]]

with cylinder deactivation (Accord, Odyssey, and Pilot) and GM will 
soon release cylinder deactivation on its 3.9L 6-cylinder engine. Fuel 
economy improvement potential scales roughly with engine displacement-
to-vehicle weight ratio: the higher displacement-to-weight vehicles, 
operating at lower relative loads for normal driving, have the 
potential to operate in part-cylinder mode more frequently.
    Honda's technology includes the use of active engine mounts and 
noise damping amongst other items added to its V6 engines with cylinder 
deactivation. This, of course, increases the cost relative to a four or 
eight cylinder OHC engine.
    Some manufacturers are getting results in excess of 6 percent and 
most are at the high end of the range. This higher number is supported 
by official fuel economy test data on a V6 Honda Odyssey with cylinder 
deactivation compared to the same vehicle (and engine displacement) 
without cylinder deactivation and by confidential manufacturer 
information.
    The NAS Report projected that cylinder deactivation could 
incrementally reduce fuel consumption by 3 percent to 6 percent at an 
incremental cost of $112 to $252. The NESCCAF study projected that 
cylinder deactivation could incrementally reduce fuel consumption by 
1.7 percent to 4.2 percent at an incremental cost of $161 to $210; 
while the EEA report projected that cylinder deactivation could 
incrementally reduce fuel consumption by 5.2 percent to 7.2 percent at 
an incremental cost of $105 to $135. Confidential manufacturer data and 
official fuel economy test data indicates that cylinder deactivation 
could incrementally reduce fuel consumption by at least 6 percent at an 
incremental cost of $203 to $229. NHTSA believes that these 
manufacturer estimates are more accurate and thus estimates that 
cylinder deactivation could reduce fuel consumption by 4.5 percent to 6 
percent at an incremental cost of $203 to $229.

Variable Valve Timing

    Variable valve timing is a classification of valvetrain designs 
that alter the timing of the intake valve, exhaust valve, or both, 
primarily to reduce pumping losses, increase specific power, and 
control residual gases. VVT reduces pumping losses when the engine is 
lightly loaded by positioning the valve at the optimum position needed 
to sustain horsepower and torque. VVT can also improve thermal 
efficiency at higher engine speeds and loads. Additionally, VVT can be 
used to alter (and optimize) the effective compression ratio where it 
is advantageous for certain engine operating modes.
    Variable valve timing has been available in the market for quite a 
while. By the early 1990s, VVT had made a significant market 
penetration with the arrival of Honda's ``VTEC'' line of engines. VVT 
has now become a widely adopted technology: for the 2007 model year, 
over half of all new cars and light trucks have engines with some 
method of variable valve timing. Therefore, the degree of further 
improvement across the fleet is limited to vehicles that have not 
already implemented this technology.
    Manufacturers are currently using many different types of variable 
valve timing, which have a variety of different names and methods. The 
major types of VVT are listed below:

Intake Camshaft Phasing (ICP)

    Valvetrains with ICP--the simplest type of cam phasing--can modify 
the timing of the intake valve while the exhaust valve timing remains 
fixed. This requires the addition of a cam phaser for each bank of 
intake valves on the engine. An in-line 4-cylinder engine has one bank 
of intake valves, while V-configured engines would have two banks of 
intake valves. The NAS Report projected that ICP could incrementally 
reduce fuel consumption by 3 percent to 6 percent at an incremental 
cost of $35; while the EEA report projected that ICP could reduce fuel 
consumption at an incremental cost of $35. The NESCCAF study projected 
that ICP could incrementally reduce fuel consumption by 1 percent to 2 
percent at an incremental cost of $49. Consistent with the EEA report 
and NESCCAF study, we have used this $35 manufacturer cost to arrive at 
incremental cost of $59 per cam phaser or $59 for an in-line 4 cylinder 
and $119 for a V-type, thus NHTSA estimates that ICP could 
incrementally reduce fuel consumption by 1 to 2 percent at an 
incremental cost of $59 to $119.

Coupled Camshaft Phasing (CCP)

    Coupled (or coordinated) cam phasing is a design in which both the 
intake and exhaust valve timing are varied with the same cam phaser. 
For an overhead cam engine, the same phaser added for ICP would be used 
for CCP control. As a result, its costs should be identical to those 
for ICP. For an overhead valve engine, only one phaser would be 
required for both inline and V-configured engines since only one 
camshaft exists. Therefore, for overhead valve engines, the cost is 
estimated at $59 regardless of engine configuration, using the logic 
provided for ICP.
    The NESCCAF study projected that CCP could incrementally reduce 
fuel consumption by 1 percent to 3 percent above that obtained by ICP. 
Confidential manufacturer data also projects that that CCP could 
incrementally reduce fuel consumption by 1 percent to 3 percent above 
that obtained by ICP. According to the NESCCAF report and confidential 
manufacturer data, NHTSA estimates that CCP could incrementally reduce 
fuel consumption by 1 to 3 percent at an incremental cost of $59 to 
$119 above ICP valvetrains.

Dual (Independent) Camshaft Phasing (DCP)

    The most flexible VVT design is dual cam phasing, where the intake 
and exhaust valve opening and closing events are controlled 
independently. This design allows the option of controlling valve 
overlap, which can be used as an internal EGR strategy. Our estimated 
incremental compliance cost for this technology is built upon that for 
VVT-ICP where an additional cam phaser is added to control each bank of 
exhaust valves less the cost to the manufacturer of the removed EGR 
valve. The incremental compliance cost for a 4-cylinder engine is 
estimated to be $59 for each bank of valves, plus an estimated piece 
cost of $30 for the valves, for a total incremental compliance cost of 
$89. The incremental compliance cost for a V6 or a V8 engine is 
estimated to be $59 for each bank of intake valves (i.e., two banks 
times $59/bank = $119), $59 for each bank of exhaust valves (i.e., 
another $119) minus an estimated $29 incremental compliance cost for 
the removed EGR valve; the total incremental compliance cost being 
$209.
    According to the NESCCAF report and confidential manufacturer data, 
it is estimated that DCP could incrementally reduce fuel consumption by 
1 to 3 percent at an incremental cost of $89 to $209 compared to 
engines with ICP or CCP.
    Because ICP and CCP have the same cost and similar effectiveness, 
it is assumed that manufacturers will choose the technology that best 
fits the specific engine architecture and application.

Variable Valve Lift and Timing

    Some vehicles have engines for which both valve timing and lift can 
be at least partially optimized based on engine operating conditions. 
Engines with variable valve timing and lift (VVLT) can achieve further 
reductions in pumping losses and further increases in thermal 
efficiency. Controlling the lift

[[Page 24370]]

height of the valves provides additional flexibility and potential for 
further fuel consumption reduction. By reducing the valve lift, engines 
can decrease the volumetric flow at lower operating loads, improving 
fuel-air mixing and in-cylinder mixture motion which results in 
improved thermodynamic efficiency and also potentially reduced overall 
valvetrain friction. Also, by moving the throttling losses further 
downstream of the throttle valve, the heat transfer losses that occur 
from the throttling process are directed into the fresh charge-air 
mixture just prior to compression, delaying the onset of knock-limited 
combustion processes. At the same time, such systems may also incur 
increased parasitic losses associated with their actuation mechanisms.
    The NAS report projected that VVLT could incrementally reduce fuel 
consumption by 1 to 2 percent over VVT alone at an incremental cost of 
$73 to 218.
    Manufacturers are currently using many different types of variable 
valve lift and timing, which have a variety of different names and 
methods. The major types of VVLT are listed below:

Discrete Variable Valve Lift

    Discrete variable valve lift (DVVL) is a method in which the 
valvetrain switches between multiple cam profiles, usually 2 or 3, for 
each valve. These cam profiles consist of a low and a high-lift lobe, 
and may include an inert or blank lobe to incorporate cylinder 
deactivation (in the case of a 3-step DVVL system). According to the 
NESCCAF report and confidential manufacturer data, it is estimated that 
DVVL could incrementally reduce fuel consumption by 0.5 to 3 percent at 
an incremental cost of $169 to $322 compared to VVT depending on engine 
size and overhead cam versus overhead valve engines. Included in this 
cost estimate is $25 for controls and associated oil supply needs 
(these costs not reflected in the NESCCAF study). We also project that 
a single valve lifter could control valve pairs, thus engines with dual 
intake and/or dual exhaust valves would require only one lifter per 
pair of valves. Due to this, the estimated costs for applying DVVL to 
overhead cam and overhead valve engines are the same.

Continuous Variable Valve Lift

    Continuous variable valve lift (CVVL) employs a mechanism that 
varies the pivot point in the rocker arm. This design is realistically 
limited to overhead cam engines. Currently, BMW has implemented this 
type of system in its Valvetronic engines, which employs fully flexible 
valve timing to allow an extra set of rocker arms to vary the valve 
lift height. CVVL enables intake valve throttling in engines, which 
allows for the use of more complex systems of sensors and electronic 
controls to enable further optimization of valve lift.
    The NESCCAF study projected incremental costs from $210 to $420, 
depending on vehicle class, while the EEA report projected incremental 
costs of $180 to $350, depending on vehicle class. Confidential 
manufacturer data projects that CVVL could incrementally reduce fuel 
consumption by 1.5 by 4 percent at an incremental cost of $200 to $515. 
NHTSA believes that these manufacturer estimates are more accurate than 
NESCCAF estimates, thus it gives more weight to them. According to the 
NESCCAF report and confidential manufacturer data, NHTSA estimates that 
CVVL could incrementally reduce fuel consumption by 1.5 by 4 percent at 
an incremental cost of $254 to $508 compared to VVT with cost estimates 
varying from $254, $466, and $508 for a 4-, 6-, and 8-cylinder engine, 
respectively.

Camless Valve Actuation

    Camless valve actuation relies on electromechanical actuators 
instead of camshafts to open and close the cylinder valves. When 
electromechanical actuators are used to replace cams and coupled with 
sensors and microprocessor controls, valve timing and lift can be 
optimized over all conditions. An engine valvetrain that operates 
independently of any mechanical means provides the ultimate in 
flexibility for intake and exhaust timing and lift optimization. With 
it comes infinite valve overlap variability, the rapid response 
required to change between operating modes (such as HCCI and GDI), 
intake valve throttling, cylinder deactivation, and elimination of the 
camshafts (reduced friction). This level of control can enable even 
further incremental reductions in fuel consumption.
    Camless valvetrains have been under research for many decades due 
to the design flexibility and the attractive fuel economy improvement 
potential they might provide. Despite the promising features of camless 
valvetrains, significant challenges remain. High costs and design 
complexity have reduced manufacturers' enthusiasm for camless engines 
in light of other competing valvetrain technologies. The advances in 
VVT, VVLT, and cylinder deactivation systems demonstrated in recent 
years have reduced the potential efficiency advantage of camless 
valvetrains.
    The NAS Report projected that camless valve actuation could 
incrementally reduce fuel consumption by 5 to 10 percent over VVLT at 
an incremental cost of $336 to $673. Confidential manufacturer 
information provides incremental fuel consumption losses that range 
from 2 to 10 percent at costs that range from $300 to $1,100. The 
NESCCAF study projected that camless valve actuation could 
incrementally reduce fuel consumption by 11 to 13 percent at an 
incremental cost of $805 to $1,820; while the EEA report projected that 
camless valve actuation could incrementally reduce fuel consumption by 
10 to 14 percent at an incremental cost of $210 to $600. These benefits 
and costs are believed to be incremental to engines with VVT.
    In reviewing our sources for costs, we have determined that the 
adjusted costs presented in the 2002 NAS study, which ranged from $336 
to $673--depending on vehicle class--represent the best available 
estimates. Subtracting out the improvements associated with the 
application of VVLT provides an estimated fuel consumption reduction of 
2.5 percent.

Stoichiometric Gasoline Direct Injection Technology

    Gasoline direct injection (GDI, or SIDI) engines inject fuel at 
high pressure directly into the combustion chamber (rather than the 
intake port in port fuel injection). Direct injection improves cooling 
of the air/fuel charge within the cylinder, which allows for higher 
compression ratios and increased thermodynamic efficiency. Injector 
design advances and increases in fuel pressure have promoted better 
mixing of the air and fuel, enhancing combustion rates, increasing 
exhaust gas tolerance and improving cold start emissions. GDI engines 
achieve higher power density and match well with other technologies, 
such as boosting and variable valvetrain designs.
    Several manufacturers (Audi, BMW, and Volkswagen) have recently 
released GDI engines while General Motors and Toyota will be 
introducing GDI engines. In addition, BMW and GM have announced their 
plans to dramatically increase the number of GDI engines in their 
portfolios.
    The NESCCAF report projected that the incremental cost for GDI of 
$189 to $294; while the EEA report projected an incremental cost of $77 
to $135. Confidential manufacturer data provides data with higher upper 
end costs than these estimates, with incremental fuel consumption 
estimates ranging from 1

[[Page 24371]]

to 2 percent. For our analysis, we have estimated the costs of 
individual components of a GDI system and used a ``bottom up'' approach 
looking at incremental costs for injectors, fuel pumps, etc., to arrive 
at system incremental compliance costs ranging from $122 to $420 for 
small cars and up to $228 to $525 for large trucks. The lower end of 
the ranges represents our best estimate using a bottom up approach 
while the upper end of the ranges represent levels more consistent with 
the manufacturer CBI submittals. As a result, we estimate that 
stoichiometric GDI could incrementally reduce fuel consumption by 1 to 
2 percent at an incremental cost of $122 to $525 compared to engines of 
similar power output.

Gasoline Engine Turbocharging and Engine Downsizing

    The specific power of a naturally aspirated engine is limited, in 
part, by the rate at which the engine is able to draw air into the 
combustion chambers. Turbocharging and supercharging are two methods to 
increase the intake manifold pressure and cylinder charge-air mass 
above naturally aspirated levels. By increasing the pressure 
differential between the atmosphere and the charging cylinders, 
superchargers and turbochargers increase this available airflow, and 
thus increase the specific power level, and with it the ability to 
reduce engine size while maintaining performance. This effectively 
reduces the pumping losses at lighter loads in comparison to a larger, 
naturally aspirated engine, while at the same time reducing net 
friction losses
    Almost every major manufacturer currently markets a vehicle with 
some form of boosting. While boosting has been a common practice for 
increasing performance for several decades, it has considerable fuel 
economy potential when the engine displacement is reduced. Specific 
power levels for a boosted engine often exceed 100 hp/L--compared to 
average naturally aspirated engine power density of roughly 70 hp/L. As 
a result, engines can conservatively be downsized roughly 30 percent to 
achieve similar peak output levels.
    In the last decade, improvements to turbine design have improved 
their reliability and performance across the entire engine operating 
range. New variable geometry turbines spool up to speed faster 
(eliminating the once-common ``turbo lag'') while maintaining high flow 
rates for increased boost at high speeds.
    Turbocharging and downsizing involve the addition of a boost 
system, removal of two cylinders in most cases (from an 8-cylinder to a 
6, or a 6 to a 4) and associated valves, and the addition of some form 
of cold start control system (e.g., air injection) to address possible 
cold start emission control. The NAS Report projected that 
turbocharging and downsizing could incrementally reduce fuel 
consumption by 5 to 7 percent at an incremental cost of $364 to $582. 
The EEA report projected turbocharging and downsizing could 
incrementally reduce fuel consumption by 5.2 to 7.8 percent.
    In developing estimated costs for turbocharging and downsizing an 
engine, NHTSA, in conjunction with EPA, relied upon piece cost 
estimates contained in the NESCCAF report. The cost estimates provided 
by the NESCCAF report are as follows: $600 for the turbocharger and 
associated parts; $90 for an air injection pump and associated parts 
(each turbocharger requires an air injection pump); $75 per cylinder 
and associated components; $15 per each valve and associated 
components; and $150 per camshaft.
    In developing the cost estimates for each of the 10 classes of 
vehicles, we determined the most logical type of downsizing that would 
occur for each class and starting with the turbocharger and air 
injector cost, either added or deleted cost, depending on the 
situation. For subcompact and compact cars, we determined that the 
downsizing wouldn't involve the removal of any cylinders, valves and 
camshafts, but instead would result in a manufacturer using a smaller 
displacement 4-cylinder engine and adding the turbocharger and the air 
injector to the smaller engine. Thus, for subcompact and compact cars, 
we estimated the cost of turbocharging and downsizing to be $690 ($600 
for the turbocharger plus $90 for the air injector).
    For large trucks and large SUVs we determined that the most logical 
engine downsizing would involve replacing an 8-cylinder overhead valve 
engine with a turbocharged 6-cylinder dual overhead cam engine. This 
change would result in the removal of 2 cylinders, and the addition of 
a turbocharger, an air injector, 8 valves and 2 camshafts. Thus, we 
have estimated the cost of turbocharging and downsizing to be $810 
($600 for the turbocharger plus $90 for the air injector, plus $120 for 
eight valves plus $150 for a camshaft and minus $150 for the removal of 
two cylinders).
    For midsize cars, large cars, small trucks, small SUVs, midsize 
SUVs and minivans, we determined that the most logical engine 
downsizing would involve replacing a 6-cylinder dual overhead cam 
engine with a turbocharged 4-cylinder dual overhead cam engine. This 
change would result in the removal of 2 cylinders, 8 valves and 2 
camshafts and the addition of a turbocharger and air injector. Thus, we 
have estimated the cost of turbocharging and downsizing to be $120 
($600 for the turbocharger plus $90 for the air injector, minus $150 
for the removal of two cylinders, minus $120 for the removal of eight 
valves and minus $300 for the removal of two camshafts).
    Thus, we have estimated the cost for a boosted/downsized engine 
system at $690 for small cars, $810 for large trucks, and $120 for 
other vehicle classes. Projections of the fuel consumption reduction 
potential of a turbocharged and downsized engine from the NAS Report 
are backed by EEA estimates and confidential manufacturer data. 
According to the NAS Report, the EEA report, cost estimates developed 
in conjunction with EPA and confidential manufacturer data, NHTSA 
estimates that downsized turbocharged engines could incrementally 
reduce fuel consumption from 5 to 7.5 percent at an incremental cost of 
$120 to $810.

Diesel Engine

    Diesel engines have several characteristics that give them superior 
fuel efficiency to conventional gasoline, spark-ignited engines. 
Pumping losses are greatly reduced due to lack of (or greatly reduced) 
throttling. The diesel combustion cycle operates at a higher 
compression ratio, with a very lean air/fuel mixture, and typically at 
much higher torque levels than an equivalent-displacement gasoline 
engine. Turbocharged light-duty diesels typically achieve much higher 
torque levels at lower engine speeds than equivalent-displacement 
naturally-aspirated gasoline engines. Additionally, diesel fuel has 
higher energy content per gallon. However, diesel engines have 
emissions characteristics that present challenges to meeting Tier 2 
emissions standards.
    Compliance strategies are expected to include a combination of 
combustion improvements and after-treatment. Several key advances in 
diesel technology have made it possible to reduce emissions coming from 
the engine (prior to after-treatment). These technologies include 
improved fuel systems (higher pressures and more responsive injectors), 
advanced controls and sensors to optimize combustion and emissions 
performance, higher EGR levels to reduce NOX, lower

[[Page 24372]]

compression ratios and advanced turbocharging systems.
    For after-treatment, the traditional 3-way catalyst found on 
gasoline-powered vehicles is ineffective due to the lean-burn 
combustion of a diesel. All diesels will require a particulate filter, 
an oxidation catalyst, and a NOX reduction strategy to 
comply with Tier 2 emissions standards.
    The NOX reduction strategies most common are outlined 
below:

Lean NOX Trap Catalyst After-Treatment

    A lean NOX trap (LNT) operates, in principle, by storing 
NOX (NO and NO2) when the engine is running in 
its normal (lean) state. When the control system determines (via 
mathematical model or a NOX sensor) that the trap is 
saturated with NOX, it switches to a rich operating mode. 
This rich mode produces excess hydrocarbons that act as a reducing 
agent to convert the stored NOX to N2 and water, 
thereby ``regenerating'' the LNT and opening up more locations for 
NOX to be stored. LNTs are sensitive to sulfur deposits 
which can reduce catalytic performance, but periodically undergo a 
desulfation engine operating mode to clean it of sulfur buildup.
    According to confidential manufacturer data, NHTSA estimates that 
LNT-based diesels can incrementally reduce fuel consumption by 8 to 15 
percent at an incremental cost of $1,500 to $1,600 compared to a direct 
injected turbocharged and downsized internal combustion engine. These 
costs are based on a ``bottom up'' cost analysis that was performed 
with EPA which then subtracted the costs of all previous steps on the 
decision tree prior to diesel engines.

Selective Catalytic Reduction NOX After-Treatment

    SCR uses a reductant (typically, ammonia derived from urea) 
continuously injected into the exhaust stream ahead of the SCR 
catalyst. Ammonia combines with NOX in the SCR catalyst to 
form N2 and water. The hardware configuration for an SCR 
system is more complicated than that of an LNT, due to the onboard urea 
storage and delivery system (which requires a urea pump and injector 
into the exhaust stream). While there is no required rich engine 
operating mode prescribed for NOX reduction, the urea is 
typically injected at a rate of 3 to 4 percent of that of fuel 
consumed. Manufacturers designing SCR systems are intending to align 
urea tank refills with standard maintenance practices such as oil 
changes. Incremental fuel consumption reduction estimates for diesel 
engines with an SCR system range from 11 to 20 percent at an 
incremental cost of $2,051 to $2,411 compared to a direct injected 
turbocharged and downsized internal combustion engine. These costs are 
based on a ``bottom up'' cost analysis that was performed with EPA, 
which then subtracted the costs of all previous steps on the decision 
tree prior to diesel engines.
    Based on public information and on recent discussions that NHTSA 
and EPA have had with auto manufacturers and aftertreatment device 
manufacturers, NHTSA has received strong indications that LNT systems 
would probably be used on smaller vehicles while the SCR systems would 
be used on larger vehicles and trucks. The primary reason given for 
this choice is the trade off between the rhodium needed for the LNT and 
the urea injection system needed for SCR. The breakeven point between 
these two cost factors appears to occur around 3.0 liters. Thus, it is 
believed that it is cheaper to manufacture diesel engines smaller than 
3.0 liters with an LNT system, and that conversely, it is cheaper to 
manufacture diesel engines larger than 3.0 liters with a SCR system. Of 
course, there are other factors that influence a manufacturer's 
decision on which system to use, but we have used this rule-of-thumb 
for our analysis.
b. Transmission Technologies

Five-, Six-, Seven-, and Eight-Speed Automatic Transmissions

    The number of available transmission speeds influences the width of 
gear ratio spacing and overall coverage and, therefore, the degree of 
transmission ratio optimization available under different operating 
conditions. In general, transmissions can offer a greater available 
degree of engine optimization and can therefore achieve higher fuel 
economy when the number of gears is increased. However, potential gains 
may be reduced by increases in transmission weight and rotating mass. 
Regardless of possible changes to fuel economy standards, manufacturers 
are increasingly introducing 5- and 6-speed automatic transmissions on 
their vehicles. Additionally, some manufacturers are introducing 7-, 
and 8-speed automatic transmissions, with 7-speed automatic 
transmissions appearing with increasing frequency.

Automatic 5-Speed Transmissions

    As automatic transmissions have been developed over the years, more 
forward speeds have been added to improve fuel efficiency and 
performance. Increasing the number of available ratios provides the 
opportunity to optimize engine operation under a wider variety of 
vehicle speeds and load conditions. Also, additional gears allow for 
overdrive ratios (where the output shaft of the transmission is turning 
at a higher speed than the input shaft) which can lower the engine 
speed at a given road speed (provided the engine has sufficient power 
at the lower rpm point) to reduce pumping losses. However, additional 
gears can add weight, rotating mass, and friction. Nevertheless, 
manufacturers are increasingly adding 5-speed automatic transmissions 
to replace 3- and 4-speed automatic transmissions.
    The 2002 NAS study projected that 5-speed automatic transmissions 
could incrementally reduce fuel consumption by 2 to 3 percent at an 
incremental cost of $76 to $167. The NESCCAF study projected that 5-
speed automatic transmissions could incrementally reduce fuel 
consumption by 1 percent at an incremental cost of $140; while the EEA 
report projected that 5-speed automatic transmissions could 
incrementally reduce fuel consumption by 2 to 3 percent at an 
incremental cost of $130. Confidential manufacturer data projected that 
5-speed automatic transmissions could incrementally reduce fuel 
consumption by 1 to 6 percent at an incremental cost of from $60 to 
$281. NHTSA believes that the NAS study's estimates are still valid and 
estimates that 5-speed automatic transmissions could incrementally 
reduce fuel consumption by 2.5 percent at an incremental cost of $76 to 
$167 (relative to a 4-speed automatic transmission).

Automatic 6-, 7-, and 8-Speed Transmissions

    In addition to 5-speed automatic transmissions, manufacturers can 
also choose to utilize 6-, 7-, or 8-speed automatic transmissions. 
Additional ratios allow for further optimization of engine operation 
over a wider range of conditions, but this is subject to diminishing 
returns as the number of speeds increases. As additional planetary gear 
sets are added (which may be necessary in some cases to achieve the 
higher number of ratios), additional weight and friction are 
introduced. Also, the additional shifting of such a transmission can be 
perceived as bothersome to some consumers, so manufacturers need to 
develop strategies for smooth shifts. Some manufacturers are replacing 
4-speed automatics with 6-speed automatics (there are also increasing 
numbers of 5-speed automatic transmissions that are

[[Page 24373]]

being replaced by 6-speed automatic transmissions), and 7-, and 8-speed 
automatics have entered production, albeit in lower-volume 
applications.
    The NAS study projected that 6-, 7- or 8-speed transmissions could 
incrementally reduce fuel consumption by 1 to 2 percent at an 
incremental cost of $70 to $126. Confidential manufacturer data 
projected that 6-, 7-or 8-speed transmissions could incrementally 
reduce fuel consumption by 1 to 3 percent at an incremental cost of $20 
to $120. However, according to the EEA report, a Lepelletier gear set 
design provides for 6-speeds at the same cost as a 5-speed automatic. 
Based on that analysis, we have estimated the cost of a 6-speed 
automatic to be equivalent to that for a 5-speed automatic. We have not 
developed any estimate costs for 7-or 8-speed transmissions because of 
the diminishing returns in efficiency versus the costs for 
transmissions beyond 6-speeds. NHTSA estimates that 6-, 7-, or 8-speed 
automatic transmissions could incrementally reduce fuel consumption by 
0.5 to 2.5 percent at an incremental cost of $0 to $20 (relative to a 
5-speed automatic transmission). We are estimating up to an additional 
$20 in costs because we have tried to account for the engineering 
effort in addition to the hardware which we believe the EEA did not and 
we wanted to capture some of the higher costs reported by 
manufacturers.

Aggressive Shift Logic

    In operation, an automatic transmission's controller decides when 
to upshift or downshift based on a variety of inputs such as vehicle 
speed and throttle position according to programmed logic. Aggressive 
shift logic (ASL) can be employed so that a transmission is engineered 
in such a way as to maximize fuel efficiency by upshifting earlier and 
inhibiting downshifts under some conditions. Through partial lock-up 
under some operating conditions and early lock-up under others, 
automatic transmissions can achieve some reduction in overall fuel 
consumption. Aggressive shift logic is applicable to all vehicle types 
with automatic transmissions, and since in most cases it would require 
no significant hardware modifications, it can be adopted during vehicle 
redesign or refresh or even in the middle of a vehicle's product cycle. 
The application of this technology does, however, require a 
manufacturer to confirm that driveability, durability, and noise, 
vibration, and harshness (NVH) are not significantly degraded.
    The NAS study projected that aggressive shift logic could 
incrementally reduce fuel consumption by 1 to 2 percent at an 
incremental cost of $0 to $70. Confidential manufacturer data projected 
that aggressive shift logic could incrementally reduce fuel consumption 
by 0.5 to 3 percent at an incremental cost of $18 to $70. The NAS study 
estimates and confidential manufacturer data are within the same 
ranges, thus NHTSA believes that the NAS estimates are still accurate. 
Thus, NHTSA estimates aggressive shift logic could incrementally reduce 
fuel consumption by 1 to 2 percent at an incremental cost of $38, which 
is approximately the average of the midpoint of the NAS cost range and 
the manufacturer cost range.

Early Torque Converter Lockup

    A torque converter is a fluid coupling located between the engine 
and transmission in vehicles with automatic transmissions and 
continuously-variable transmissions (CVTs). This fluid coupling allows 
for slip so the engine can run while the vehicle is idling in gear, 
provides for smoothness of the powertrain, and also provides for torque 
multiplication during acceleration. During light acceleration and 
cruising, this slip causes increased fuel consumption, so modern 
automatic transmissions utilize a clutch in the torque converter to 
lock it and prevent this slippage. Fuel consumption can be further 
reduced by locking up the torque converter early, and/or by using 
partial-lockup strategies to reduce slippage.
    Some torque converters will require upgraded clutch materials to 
withstand additional loading and the slipping conditions during partial 
lock-up. As with aggressive shift logic, confirmation of acceptable 
driveability, performance, durability and NVH characteristics is 
required to successfully implement this technology.
    The 2002 NAS study did not include any estimates for this 
technology. The NESCCAF study projected that early torque converter 
lockup could incrementally reduce fuel consumption by 0.5 percent at an 
incremental cost of $0 to $10; while the EEA report projected that low-
friction lubricants could incrementally reduce fuel consumption by 0.5 
percent at an incremental cost of $5. NHTSA estimates the cost of this 
technology (i.e., the calibration effort) at $30 based in part on 
NESCCAF and the CBI submissions which provided costs with a midpoint of 
$30. We have used a higher value here than NESCCAF and EEA because we 
have tried to account for the engineering effort in addition to the 
hardware which we believe NESCCAF and EEA did not do and which were 
captured in the manufacturers' higher costs.
    NHTSA estimates that early torque converter lockup could 
incrementally reduce fuel consumption by approximately 0.5 percent at 
an incremental cost of approximately $30.

Automated Shift Manual Transmissions

    An automated manual transmission (AMT) is mechanically similar to a 
conventional transmission, but shifting and launch functions are 
controlled by the vehicle. There are two basic types of AMTs, single-
clutch and dual-clutch. A single-clutch AMT is essentially a manual 
transmission with automated clutch and shifting. Because there are some 
shift quality issues with single-clutch designs, dual-clutch AMTs are 
more common. A dual-clutch AMT uses separate clutches for the even-
numbered gears and odd-numbered gears. In this way, the next expected 
gear is pre-selected, which allows for faster and smoother shifting.
    Overall, AMTs likely offer the greatest potential for fuel 
consumption reduction among the various transmission options presented 
in this report because they offer the inherently lower losses of a 
manual transmission with the efficiency and shift quality advantages of 
computer control. AMTs offer the lower losses of a manual transmission 
with the efficiency advantages of computer control. The lower losses 
stem from the elimination of the conventional lock-up torque converter 
and a greatly reduced need for high pressure hydraulic circuits to hold 
clutches to maintain gear ratios (in automatic transmissions) or hold 
pulleys in position to maintain gear ratio (in continuously variable 
transmissions, discussed below). However, the lack of a torque 
converter will affect how the vehicle launches from rest, so an AMT 
will most likely be paired with an engine that offers enough torque in 
the low-RPM range to allow for adequate launch performance.
    An AMT is mechanically similar to a conventional manual 
transmission, but shifting and launch functions are controlled by the 
vehicle rather than the driver. A switch from a conventional automatic 
transmission with torque converter to an AMT incurs some costs but also 
allows for some cost savings. Savings can be realized through 
elimination of the torque converter which is a very costly part of a 
traditional automatic transmission, and through reduced need for high 
pressure hydraulic circuits to hold clutches (to maintain gear ratios 
in automatic transmissions) or hold pulleys (to maintain gear ratios in 
Continuously

[[Page 24374]]

Variable Transmissions). Cost increases would be incurred in the form 
of calibration efforts since transmission calibrations would have to be 
redone, and the addition of a clutch assembly for launce and gear 
changes.
    The NESCCAF study projected that AMTs could incrementally reduce 
fuel consumption by 5 to 8 percent at an incremental cost of $0 to 
$280; while the EEA report projected that low-friction lubricants could 
incrementally reduce fuel consumption by 6 to 7 percent at an 
incremental cost of $195 to $225. Confidential manufacturer data 
projected that AMTs could incrementally reduce fuel consumption by 2 to 
5 percent at an incremental cost of $70 to $400.
    Taking all these estimates into consideration, NHTSA estimates that 
AMTs could incrementally reduce fuel consumption by 4.5 to 7.5 percent 
at an incremental cost of approximately $141. We believe that, overall, 
the hardware associated with an AMT, whether single clutch or dual 
clutch, is no more costly than that for a traditional automatic 
transmission given the savings associated with removal of the torque 
converter and high pressure hydraulic circuits, which is estimated to 
amount to at least $30. Nonetheless, given the need for engineering 
effort (e.g., calibration and vehicle integration work) when 
transitioning from a traditional automatic to an AMT, we have estimated 
the incremental compliance cost at $141, independent of vehicle class, 
which is the midpoint of the NESCCAF estimates and within the range 
provided confidential manufacturer data.

Continuously Variable Transmission

    A Continuously Variable Transmission (CVT) is unique in that it 
does not use gears to provide ratios for operation. Unlike manual and 
automatic transmissions with fixed transmission ratios, CVTs provide, 
within their operating ranges, fully variable transmission ratios with 
an infinite number of gears. This enables even finer optimization of 
the transmission ratio under different operating conditions and, 
therefore, some reduction of pumping and engine friction losses. CVTs 
use either a belt or chain on a system of two pulleys.
    The main advantage of a CVT is that the engine can operate at its 
most efficient point more often, since there are no fixed ratios. Also, 
CVTs often have a wider range of ratios than conventional automatic 
transmissions.
    The most common CVT design uses two V-shaped pulleys connected by a 
metal belt. Each pulley is split in half and a hydraulic actuator moves 
the pulley halves together or apart. This causes the belt to ride on 
either a larger or smaller diameter section of the pulley which changes 
the effective ratio of the input to the output shafts.
    It is assumed that CVTs will only be used on cars, small SUVs, 
midsize crossover vehicles and minivans because they are currently used 
mainly in lower-torque applications. While a high-torque CVT could be 
developed for small pickup trucks and large pickup trucks and large 
SUVs, it would likely have to be treated separately in terms of 
effectiveness. We do not see development in the area of high-torque 
CVTs and therefore did not include this type in our analysis.
    The 2002 NAS study projected that CVTs could incrementally reduce 
fuel consumption by 4 to 8 percent at an incremental cost of $140 to 
$350. The NESCCAF study projected that CVTs could incrementally reduce 
fuel consumption by 4 percent at an incremental cost of $210 to $245. 
Confidential manufacturer data projected that CVTs could incrementally 
reduce fuel consumption by 3 to 9 percent at an incremental cost of 
$140 to $800. These values are incremental to a 4-speed transmission.
    Based on an aggregation of manufacturers' information, we estimate 
a CVT benefit of about 6 percent over a 4-speed automatic. This is 
above the NESCCAF value, but in the range of NAS. In reviewing our 
sources for costs, we have determined that the adjusted costs presented 
in the 2002 NESCCAF study represent the best available estimates. 
Subtracting the estimated fuel consumption reduction and costs of 
replacing a 4-speed automatic transmission with a 5-speed automatic 
transmission results in NHTSA's projecting that CVTs could 
incrementally reduce fuel consumption by 3.5 percent when compared to a 
conventional 5-speed automatic transmission at an incremental cost of 
$100 to $139.

Manual 6-, 7-, and 8-Speed Transmissions

    As with automatic transmissions, increasing the number of available 
ratios in a manual transmission can improve fuel economy by allowing 
the driver to select a ratio that optimizes engine operation at a given 
speed. Typically, this is achieved through adding additional overdrive 
ratios to reduce engine speed (which saves fuel through reduced pumping 
losses). Six-speed manual transmissions have already achieved 
significant market penetration, so manufacturers have considerable 
experience with them and the associated costs. For those vehicles with 
five-speed manual transmissions, an upgrade to a six-speed could 
incrementally reduce fuel consumption by 0.5 percent. Based on CBI 
submissions, which provided costs with a midpoint of $107, NHTSA 
estimates that 6-speed manual transmissions could incrementally reduce 
fuel consumption by 0.5 percent when compared to 5-speed automatic 
transmission at an incremental cost of $107.
c. Vehicle Technologies

Rolling Resistance Reduction

    Tire characteristics (e.g., materials, construction, and tread 
design) influence durability, traction control, vehicle handling, and 
comfort. They also influence rolling resistance--the 30 frictional 
losses associated mainly with the energy dissipated in the deformation 
of the tires under load--and therefore, CO2 emissions. This 
technology is applicable to all vehicles, except for body-on-frame 
light trucks and performance vehicles (described in the next section). 
Based on a 2006 NAS/NRC report, a 10 percent rolling resistance 
reduction would provide an increase in fuel economy of 1 to 2 percent. 
The same report estimates a $1 per tire cost for low rolling resistance 
tires. For four tires, our incremental compliance cost estimate is $6 
per vehicle, independent of vehicle class, although not applicable to 
large trucks.

Low Drag Brakes

    Low drag brakes reduce the sliding friction of disc brake pads on 
rotors when the brakes are not engaged because the brake shoes are 
pulled away from the rotating drum. While most passenger cars have 
already adopted this technology, there are indications that this 
technology is still available for body-on-frame trucks. According to 
confidential manufacturer data, low drag brakes could incrementally 
reduce fuel consumption by 1 to 2 percent at an incremental cost of $85 
to $90. NHTSA has adopted these values for its analysis.

Front or Secondary Axle Disconnect for Four-Wheel Drive Systems

    To provide shift-on-the-fly capabilities, many part-time four-wheel 
drive systems use some type of axle disconnect: Front axle disconnect 
in ladder-frame vehicles, and secondary (i.e., either front or rear) 
axle disconnect in unibody vehicles. Front and secondary axle 
disconnects serve two basic purposes. Using front axle

[[Page 24375]]

disconnect as an example, in two-wheel drive mode, the technology 
disengages the front axle from the front driveline so the front wheels 
do not turn the front driveline at road speed, saving wear and tear. 
Then, when shifting from two- to four-wheel drive ``on the fly'' (while 
moving), the front axle disconnect couples the front axle to the front 
differential side gear only when the transfer case's synchronizing 
mechanism has spun the front driveshaft up to the same speed as the 
rear driveshaft.
    Four-wheel drive systems that have axle disconnect typically do not 
have either manual- or automatic-locking hubs. To isolate (for example) 
the front wheels from the rest of the front driveline, front axle 
disconnects use a sliding sleeve to connect or disconnect an axle shaft 
from the front differential side gear.
    This technology has been used by ladder-frame vehicles for some 
time, but has only started to appear on unibody vehicles recently. The 
incremental costs and benefits of applying front axle disconnect 
differ, depending on the vehicle's type of construction. According to 
confidential manufacturer data, front axle disconnects for ladder frame 
vehicles could achieve incremental fuel consumption reductions of 1.5 
percent at an incremental cost of $114, while secondary axle 
disconnects for unibody vehicles could achieve incremental fuel 
consumption reductions of 1 percent at an incremental cost of $676. 
NHTSA has adopted these estimates for its analysis.

Aerodynamic Drag Reduction

    A vehicle's size and shape determine the amount of power needed to 
push the vehicle through the air at different speeds. Changes in 
vehicle shape or frontal area can therefore reduce CO2 
emissions. Areas for potential aerodynamic drag improvements include 
skirts, air dams, underbody covers, and more aerodynamic side view 
mirrors. NHTSA and EPA estimate a fleet average of 20 percent total 
aerodynamic drag reduction is attainable for passenger cars, whereas a 
fleet average of 10 percent reduction is more realistic for trucks 
(with a caveat for ``high-performance'' vehicles, described below). 
These drag reductions equate to increases in fuel economy of 2 percent 
and 3 percent for trucks and cars, respectively. These numbers are in 
agreement with the technical literature and supported by confidential 
manufacturer information. The CBI submittals generally showed the RPE 
associated with these changes at less than $100. NHTSA and EPA estimate 
that the incremental compliance cost to range from $0 to $75, 
independent of vehicle class.
    Aerodynamic drag reduction technologies are readily available 
today, although the phase-in time required to distribute over a 
manufacturer's fleet is relatively long (6 years or so).

Weight Reduction

    The term weight reduction encompasses a variety of techniques with 
a variety of costs and lead times. These include lighter-weight 
materials, higher strength materials, component redesign, and size 
matching of components. Lighter-weight materials involve using lower 
density materials in vehicle components, such as replacing steel parts 
with aluminum or plastic. The use of higher strength materials involves 
the substitution of one material for another that possesses higher 
strength and less weight. An example would be using high strength alloy 
steel versus cold rolled steel. Component redesign is an on-going 
process to reduce costs and/or weight of components, while improving 
performance and reliability. An example would be a subsystem replacing 
multiple components and mounting hardware.
    The cost of reducing weight is difficult to determine and is 
dependent upon the methods used. For example, a change in design that 
reduces weight on a new model may or may not save money. On the other 
hand, material substitution can result in an increase in price per 
application of the technology if more expensive materials are used.
    For purposes of this proposed rule, NHTSA has considered only 
vehicles weighing greater than 5,000 pounds for weight reduction 
through materials substitution. Provided that those vehicles remain 
above 5,000 pounds weight, vehicles may realize up to roughly 2 percent 
incremental fuel consumption through materials substitution 
(corresponding to a 3 percent reduction in vehicle weight) at 
incremental costs of $0.75 to $1.25 per pound reduced.

d. Accessory Technologies

Electric Power Steering

    Electric power steering (EPS) is advantageous over hydraulic 
steering in that it only draws power when the wheels are being turned, 
which is only a small percentage of a vehicle's operating time. EPS may 
be implemented on many vehicles with a standard 12V system; however, 
for heavier vehicles, a 42V system may be required, which adds cost and 
complexity.
    The NAS study projected that a 12V EPS system could incrementally 
reduce fuel consumption by 1.5 to 2.5 percent at an incremental cost of 
$105 to $150. The NESCCAF study projected that a 12V EPS could 
incrementally reduce fuel consumption by 1 percent at an incremental 
cost of $28 to $56; while the EEA report projected that a 12V EPS could 
incrementally reduce fuel consumption by 1.5 to 1.9 percent at an 
incremental cost of $70 to $90. According to confidential manufacturer 
data, electric power steering could achieve incremental fuel 
consumption reductions of 1.5 to 2.0 percent at an incremental cost of 
$118 to $197.
    NHTSA believes that these manufacturer estimates are more accurate 
and thus estimates that a 12V EPS system could incrementally reduce 
fuel consumption by 1.5 to 2 percent at an incremental cost of $118 to 
$197, independent of vehicle class.

Engine Accessory Improvement

    The accessories on an engine, like the alternator, coolant, and oil 
pumps, are traditionally driven by the accessory belt. Improving the 
efficiency or outright electrification (12V) of these accessories (in 
the case of the mechanically driven pumps) would provide an opportunity 
to reduce the accessory loads on the engine. However, the potential for 
such replacement will be greater for vehicles with 42V electrical 
systems. Some large trucks also employ mechanical fans, some of which 
could also be improved or electrified. Additionally, there are now 
higher efficiency alternators which require less of an accessory load 
to achieve the same power flow to the battery.
    According to the NAS Report engine accessory improvement could 
achieve incremental fuel consumption reductions of 1 to 2 percent at an 
incremental cost of $124 to $166. Confidential manufacturer information 
is also within these ranges. The NESCCAF study estimated a cost of $56, 
but that estimate included only a high efficiency generator and did not 
include electrification of other accessories. In reviewing our sources 
for costs, we have determined that the adjusted costs presented in the 
2002 NAS study, which ranged from $124 to $166--depending on vehicle 
class--represent the best available estimates. Based on the NAS study 
and confidential manufacturer information, NHTSA estimates that 
accessory improvement could incrementally reduce fuel consumption by 1 
to 2 percent at an incremental cost of $124 to $166.

[[Page 24376]]

Forty-Two Volt (42V) Electrical System

    Most vehicles today (aside from hybrids) operate on 12V electrical 
systems. At higher voltages, which appear to be under consideration to 
meet expected increases in on-board electrical demands, the power 
density of motors, solenoids, and other electrical components may 
increase to the point that new and more efficient systems, such as 
electric power steering, may be feasible. A 42V system can also 
accommodate an integrated starter generator. According to the NAS 
Report, 42V engine accessory improvement could achieve incremental fuel 
consumption reductions of 1 to 2 percent at an incremental cost of $194 
to $259. According to confidential manufacturer data, a 42V system 
could achieve incremental fuel consumption reductions of 0 to 4 percent 
at an incremental cost of $62 to $280.
    We believe that the state of 42V technology has evolved to where it 
is on par with the incremental costs and benefits of 12V engine 
accessory improvement. In reviewing our sources, we have determined 
that the numbers provided in the 2002 NAS study, which estimated that 
engine accessory improvement could achieve incremental fuel consumption 
reductions of 1 to 2 percent at an incremental cost of $124 to $166--
depending on vehicle class--represent the best available estimates for 
both 12V and 42V systems. Thus, we are estimating that a 42V electrical 
system could achieve incremental fuel consumption reductions of 1 to 2 
percent at an incremental cost of $124 to $166. These estimates are 
independent of vehicle class and exclusive of improvements to the 
efficiencies or electrification of 12V accessories. These estimates are 
incremental to a 12V system, regardless of whether the 12V system has 
improved efficiency or not.
e. Hybrid Technologies
    A hybrid describes a vehicle that combines two or more sources of 
propulsion energy, where one uses a consumable fuel (like gasoline) and 
one is rechargeable (during operation, or by another energy source). 
Hybrids reduce fuel consumption through three major mechanisms: by 
optimizing the operation of the internal combustion engine (through 
downsizing, or other control techniques) to operate at or near its most 
efficient point more of the time; by recapturing lost braking energy 
and storing it for later use; and by turning off the engine when it is 
not needed, such as when the vehicle is coasting or when stopped.
    Hybrid vehicles utilize some combination of the above three 
mechanisms to reduce fuel consumption. The effectiveness of a hybrid 
depends on the utilization of the above mechanisms and how aggressively 
they are pursued. Different hybrid concepts utilize these mechanisms 
differently, so they are treated separately in this analysis. Below is 
a discussion of the major hybrid concepts judged to be available for 
use within the timeframe of this rulemaking.

Integrated Starter-Generator With Idle-Off

    Integrated Starter-Generator (ISG) systems are the most basic of 
hybrid systems and offer mainly idle-stop capability. They offer the 
least power assist and regeneration capability of the hybrid 
approaches, but their low cost and easy adaptability to existing 
powertrains and platforms can make them attractive for some 
applications. ISG systems operate at around 42V and so have smaller 
electric motors and less battery capacity than other HEV designs 
because of their lower power demand.
    ISG systems replace the conventional belt-driven alternator with a 
belt-driven, higher power starter-alternator. The starter-alternator 
starts the engine during idle-stop operation, but often a conventional 
12V gear-reduction starter is retained to ensure cold-weather 
startability. Also, during idle-stop, some functions such as power 
steering and automatic transmission hydraulic pressure are lost with 
conventional arrangements, so electric power steering and an auxiliary 
transmission pump are added. These components are similar to those that 
would be used in other hybrid designs. An ISG system could be capable 
of providing some launch assist, but it would be limited in comparison 
to other hybrid concepts. According to the NAS Report, an EEA report 
and confidential manufacturer data, ISG systems could achieve 
incremental fuel consumption reductions that range from 5 to 10 
percent.
    In addition, when idle-off is used (i.e., the petroleum fuelled 
engine is shut off during idle operation), an electric power steering 
and auxiliary transmission pump are added to provide for functioning of 
these systems which, in a traditional vehicle, were powered by the 
petroleum engine. The 2002 NAS study estimated the cost of these 
systems at $210 to $350 with a 12V electrical system and independent of 
vehicle class, while the NESCCAF study estimated the cost for these 
systems at $280 with a 12 Volt electrical system for a small car. The 
2002 NAS study estimated the cost of these systems to be $210 to $350 
with a 12 volt electrical system and independent of vehicle class, 
while the NESCCAF study estimated the cost for these systems of $280 
with a 12 volt electrical system for a small car. Confidential 
manufacturer information provides cost estimates for ISGs that range 
from $418 to $800. We believe that the NAS and the NESCCAF estimates 
are still accurate for ISGs with a 12V system. Thus, if you add these 
cost estimates to those we estimated for 42V systems plus associated 
equipment, which results an estimated incremental compliance cost of 
these systems, including the costs associated with upgrading to a 42 
volt electrical system of $563 to $600, depending on vehicle class.
    Therefore, NHTSA estimates that ISG systems could achieve 
incremental fuel consumption reductions of 5 to 10 percent at 
incremental costs of $563 to $600, depending on vehicle class (this 
includes the costs associated with upgrading to a 42 volt electrical 
system).

Integrated Motor Assist (IMA)/Integrated Starter-Alternator-Dampener 
(ISAD) Hybrid

    Honda is the only manufacturer that uses Integrated Motor Assist 
(IMA), which utilizes a thin axial electric motor bolted to the 
engine's crankshaft and connected to the transmission through a torque 
converter or clutch. This electric motor acts as both a motor for 
helping to launch the vehicle and a generator for recovering energy 
while slowing down. It also acts as the starter for the engine and the 
electrical system's main generator. Since it is rigidly fixed to the 
engine, if the motor turns, the engine must turn also, but combustion 
does not necessarily need to occur. The Civic Hybrid uses cylinder 
deactivation on all four cylinders for decelerations and some cruise 
conditions.
    The main advantage of the IMA system is that it is relatively low 
cost and adapts readily to conventional vehicles and powertrains, while 
providing excellent efficiency gains. Packaging space is a concern for 
the physically longer engine-motor-transmission assembly as well as the 
necessary battery pack, cabling and power electronics. According to EPA 
test data and confidential manufacturer data, the IMA system could 
achieve incremental fuel consumption reductions of 3.5 to 8.5 
percent.\61\ NHTSA has adopted these estimates for its analysis.
---------------------------------------------------------------------------

    \61\ The cost estimates are protected as confidential business 
information.

---------------------------------------------------------------------------

[[Page 24377]]

    The 2002 NAS study did not consider this technology while the 
NESCCAF study estimated the cost for these systems at $2,310 to $2,940 
for a small car and large car, respectively. We have used these 
estimates combined with confidential manufacturer data as the basis for 
our incremental compliance costs of $1,636 for the small car and $2,274 
for the large car, expressed in 2006 dollars. We have not estimated 
incremental compliance costs for the other vehicle classes because we 
do not believe those classes would use this technology and would, 
instead, use the hybrid technologies discussed below.

2-Mode Hybrids

    GM, DaimlerChrysler, and BMW have formed a joint venture to develop 
a new HEV system based on HEV transmission technology originally 
developed by GM's Allison Transmission Division for heavy-duty vehicles 
like city buses. This technology uses an adaptation of a conventional 
stepped-ratio automatic transmission by replacing some of the 
transmission clutches with two electric motors, which makes the 
transmission act like a CVT. Like Toyota's Power Split design, these 
motors control the ratio of engine speed to vehicle speed. But unlike 
the Power Split system, clutches allow the motors to be bypassed, which 
improves both the transmission's torque capacity for heavy-duty 
applications and fuel economy at highway speeds. According to 
confidential manufacturer data, 2-mode hybrids could achieve 
incremental fuel consumption reductions of 25 to 40 percent. NHTSA 
estimates that 2-mode hybrids could achieve fuel reductions of 3.5 
percent to 7 percent incremental to an Integrated Motor Assist (IMA)/
Integrated Starter-Alternator-Dampener (ISAD) Hybrid.
    The 2002 NAS study did not consider this technology, while the 
NESCCAF study estimated the costs to range from $4,340 to $5,600, 
depending on vehicle class. These estimates are not incremental to an 
Integrated Motor Assist (IMA)/Integrated Starter-Alternator-Dampener 
(ISAD) Hybrid. To accurately project the cost of 2-mode hybrids when 
they were applied to midsize and large cars, we subtracted the 
estimated costs of an Integrated Motor Assist (IMA)/Integrated Starter-
Alternator-Dampener (ISAD) Hybrid. We have used the NESCCAF estimates 
as the basis for our incremental compliance costs of $1,501 to $5,127 
in 2006 dollars, incremental to an Integrated Motor Assist (IMA)/
Integrated Starter-Alternator-Dampener (ISAD) Hybrid or an ISG system 
depending on vehicle class.\62\ We have not estimated incremental 
compliance costs for small cars because we believe that this ISG or 
IMA/ISAD technology is a better fit for small cars.
---------------------------------------------------------------------------

    \62\ GM's cost estimates are protected as confidential business 
information.
---------------------------------------------------------------------------

Power Split Hybrid

    Toyota's Hybrid Synergy Drive system as used in the Prius is a 
completely different approach than Honda's IMA system and uses a 
``Power Split'' device in place of a conventional transmission. The 
Power Split system replaces the vehicle's transmission with a single 
planetary gear and a motor/generator. 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 the engine's 
torque between the first motor/generator and the drive motor. The first 
motor/generator uses its engine torque to either charge the battery or 
supply additional power to the drive motor. The speed of the first 
motor/generator determines the relative speed of the engine to the 
wheels. In this way, the planetary gear allows the engine to operate 
completely independently of vehicle speed, much like a CVT.
    The Power Split system allows for outstanding fuel economy in city 
driving. The vehicle also avoids the cost of a conventional 
transmission, replacing it with a much simpler single planetary and 
motor/generator. However, it is less efficient at highway speeds due to 
the requirement that the first motor/generator must be constantly 
spinning at a relatively high speed to maintain the correct ratio. 
Also, load capacity is limited to the first motor/generator's capacity 
to resist the reaction torque of the drive train.
    A version of Toyota's Power Split system is also used in the Lexus 
RX400h and Toyota Highlander sport utility vehicles. This version has 
more powerful motor/generators to handle higher loads and also adds a 
third motor/generator on the rear axle of four-wheel-drive models. This 
provides the vehicle with four wheel drive capability and four wheel 
regenerative braking capability. Ford's eCVT system used in the hybrid 
Escape is another version of the Power Split system, but four-wheel-
drive models use a conventional transfer case and drive shaft to power 
the rear wheels.
    Other versions of this system are used in the Lexus GS450h and 
Lexus LS600h luxury sedans. These systems have modifications and 
additional hardware for sustained high-speed operation and/or all-
wheel-drive capability. However, the Power Split system isn't planned 
for usage on full-size trucks and SUVs due to its limited ability to 
provide the torque needed by these vehicles. It's anticipated that 
full-size trucks and SUVs would use the 2-mode hybrid system. The 2002 
NAS study didn't consider this technology, while the NESCCAF study 
estimated the incremental costs at to be $3,500 prior to any cost 
adjustment. Based on the NESCCAF study and fuel economy test data from 
EPA's certification database which shows these systems being capable of 
reducing fuel consumption by 25 to 35 percent, NHTSA estimates that 
Power Split hybrids can achieve incremental fuel consumption reductions 
of 25 to 35 percent over conventionally powered vehicles at an 
incremental cost of $3,700 to $3,850. Because NHTSA applies 
technologies incrementally to the technologies preceding them on our 
decision trees, the incremental fuel consumption reductions for Power 
Split hybrids are estimated to be 5 to 6.5 percent incremental to 2-
Mode Hybrids (the technology that precedes Power Split hybrids on the 
decision tree), because the technologies applied prior to and including 
2-Mode hybrids are estimated to have incremental fuel consumption 
reductions of 20 to 28.5 percent over conventionally powered vehicles. 
The technologies discussed below were not projected for use during the 
MY 2011 to 2015 timeframes because NHTSA isn't aware that any 
manufacturer is including these technologies in any vehicle for which 
we have production plans for nor has any manufacturer publicly stated 
that any of these technologies will definitively be included on future 
products. If NHTSA receives such information regarding one or more 
technologies, it will revisit this decision for the final rule. NHTSA 
is including its discussion of these technologies and their estimated 
costs and fuel consumption reductions as a reference for commenters and 
in anticipation of their possible inclusion in the final rule.

Variable Compression Ratio

    A spark-ignited engine's specific power is limited by the engine's 
compression ratio, which is, in turn, currently limited by the engine's 
susceptibility to knock, particularly under high load conditions. 
Engines with variable compression ratio (VCR) improve fuel economy by 
the use of higher compression ratios at lower loads and lower 
compression ratios under higher loads. The NAS Report projected that 
VCR could incrementally reduce

[[Page 24378]]

fuel consumption by 2 to 6 percent over 4-valve VVT at an incremental 
cost of $218 to $510. NHTSA has no information which suggests that VCR 
will be included on any vehicles during the MY 2011-2015 timeframe, 
thus NHTSA does not use this technology in its analysis. Additionally, 
no updates to these estimates were sought.

Lean-Burn Gasoline Direct Injection Technology

    One way to improve dramatically an engine's thermodynamic 
efficiency is by operating at a lean air-fuel mixture (excess air). 
Fuel system improvements, changes in combustion chamber design and 
repositioning of the injectors have allowed for better air/fuel mixing 
and combustion efficiency. There is currently a shift from wall-guided 
injection to spray guided injection, which improves injection precision 
and targeting towards the spark plug, increasing lean combustion 
stability. Combined with advances in NOX after-treatment, 
lean-burn GDI engines may be a possibility in North America. However, a 
key technical requirement for lean-burn GDI engines to meet EPA's Tier 
2 NOX emissions levels is the availability of low-sulfur 
gasoline, which is projected to be unavailable during MY 2011-2015.
    According to the NESCCAF report and confidential manufacturer data 
NHTSA estimates that lean-burn GDI engines could incrementally reduce 
fuel consumption from 9 to 16 percent at an incremental cost of $500 to 
$750 compared to a port-fueled (stoichiometric) engine. NHTSA did not 
project the use of this technology during the time frame covered by 
this proposal, due to large uncertainties surrounding the availability 
of low-sulfur gasoline. Nonetheless, we have estimated the incremental 
compliance cost for these systems at $750, independent of vehicle 
class, and incremental to a stoichiometric GDI engine.

Homogeneous Charge Compression Ignition

    Homogeneous charge compression ignition (HCCI), also referred to as 
controlled auto ignition (CAI), is an alternate engine operating mode 
that does not rely on a spark event to initiate combustion. The 
principles are more closely aligned with a diesel combustion cycle, in 
which the compressed charge exceeds a temperature and pressure 
necessary for spontaneous ignition. The resulting burn is much shorter 
in duration with higher thermal efficiency.
    An HCCI engine has inherent advantages in its overall efficiency 
for several reasons. An extremely lean fuel/air charge increases 
thermodynamic efficiency. Shorter combustion times and higher EGR 
tolerance permit very high compression ratios (which also increase 
thermodynamic efficiency). Additionally, pumping losses are reduced 
because the engine can run unthrottled.
    However, due to the nature of its combustion process, HCCI is 
difficult to control, requiring in-cylinder pressure sensors and very 
fast engine control logic to optimize combustion timing, especially 
considering the variable nature of operating conditions seen in a 
vehicle. To be used in a commercially acceptable vehicle application, 
an HCCI-equipped engine would most likely be ``dual-mode,'' in which 
HCCI operation is complemented with a traditional SI combustion process 
at idle and at higher loads and speeds.
    Until recently, HCCI technology was considered to still be in the 
research phase. However, several manufacturers have made public 
statements about the viability of incorporating HCCI into production 
vehicles over the next 10 years. The NESCCAF study estimated the cost 
to range from $560 to $840, depending on vehicle class, including the 
costs for a stoichiometric GDI system with DVVL. We have based our 
estimated incremental compliance cost on the NESCCAF estimates and, 
after subtracting out the estimated incremental cost for a 
stoichiometric GDI system with DVVL, we estimate the incremental cost 
for HCCI to be from $263 to $685, depending on vehicle class. This 
estimated incremental compliance cost is incremental to a 
stoichiometric GDI engine.
    According to the NESCCAF report and confidential manufacturer data, 
NHTSA estimates that gasoline HCCI/GDI dual-mode engines could 
incrementally reduce fuel consumption from 10 to 12 percent at an 
incremental cost of $233 to $606, compared to a comparable GDI engine.

Advanced CVT

    Advanced CVTs have the ability to deliver higher torques than 
existing CVTs and have the potential for broader market penetration. 
These new designs incorporate toroidal friction elements or cone-and-
ring assemblies with varying diameters. According to the NAS Report, 
advanced CVT could incrementally reduce fuel consumption by up to 2 
percent at an incremental cost of $364 to $874. NHTSA has no 
information which suggests that VCR will be included on any vehicles 
during the MY 2011-2015 timeframe, thus NHTSA does not use this 
technology in its analysis. Additionally, no updates to these estimates 
were sought.

Plug-in Hybrids

    Plug-In Hybrid Electric Vehicles (PHEVs) are very similar to hybrid 
electric vehicles, but with three significant functional differences. 
The first is the addition of a means to charge the battery pack from an 
outside source of electricity (usually the electric grid). Second, a 
PHEV would have a larger battery pack with more energy storage, and a 
greater capability to be discharged. Finally, a PHEV would have a 
control system that allows the battery pack to be significantly 
depleted during normal operation.
    Deriving some of their propulsion energy from the electric grid 
provides several advantages for PHEVs. PHEVs offer a significant 
opportunity to replace petroleum used for transportation energy with 
domestically-produced electricity. The reduction in petroleum usage 
does, of course, depend on the amount of electric drive the vehicle is 
capable of under its duty cycle.
    The fuel consumption reduction potential of PHEVs depends on many 
factors, the most important being the electrical capacity designed into 
the battery pack. To estimate the fuel consumption reduction potential 
of PHEVs, EPA has developed an in-house vehicle energy model (PEREGRIN) 
which is based on the PERE (Physical Emission Rate Estimator) physics-
based model used as a fuel consumption input for EPA's MOVES mobile 
source emissions modelB.
    EPA modeled the PHEV small car, large car, minivan and small trucks 
using parameters from a midsize car similar to today's hybrids and 
scaled to each vehicle's weight. The large truck PHEV was modeled 
separately assuming very little engine downsizing. Each PHEV was 
assumed to have enough battery capacity for a 20-mile-equivalent all-
electric range and a power requirement to provide similar performance 
to a hybrid vehicle. A twenty mile range was selected because it offers 
a good compromise for vehicle performance, weight, battery packaging 
and cost.
    To calculate the total energy use of a PHEV, a vehicle can be 
thought of as operating in two distinct modes, electric (EV) mode, and 
hybrid (HEV) mode. The energy consumed during EV operation can be 
accounted for and calculated in terms of gasoline-equivalent MPG by 
using 10CFR474, Electric and Hybrid Vehicle Research, Development, and 
Demonstration Program; Petroleum-Equivalent Fuel Economy Calculation. 
The EV mode fuel economy can then be

[[Page 24379]]

combined with the HEV mode fuel economy using the Utility Factor 
calculation in SAE J1711 to determine a total MPG value for the 
vehicle. Calculating a total fuel consumption reduction based on model 
outputs, gasoline-equivalent calculations, and the Utility Factor 
calculations, results in a 28 percent fuel consumption reduction for 
small cars, large cars, minivans, and small trucks and a 31 percent 
fuel consumption reduction for large trucks.
    The fuel consumption reduction potential of PHEVs will vary based 
on the electrical capacity designed into the battery pack. Assuming a 
20-mile ``all-electric range'' design, a PHEV might incrementally 
reduce fuel consumption by 28 to 31 percent.\63\ Based on discussions 
with EPA, we have estimated the incremental cost of PHEVs to be from 
$4,500 to $10,200, depending on vehicle class.
---------------------------------------------------------------------------

    \63\ This estimate is based on the EPA test cycle. We are unable 
to provide cost estimates for PHEV technology due to the great 
amount of uncertainty in deciding the appropriate battery chemistry 
to be used.
---------------------------------------------------------------------------

    However, all indications suggest that any PHEVs that may be 
available within the time frame of this rulemaking will be concept 
vehicles and not production vehicles. Additionally, NHTSA is unaware of 
the existence of any batteries that are deemed acceptable for the 
performance characteristics necessary for a plug-in hybrid. Therefore, 
although we discuss them here, the model does not apply them.
    NHTSA would like to note that if it receives new and/or updated 
information from manufacturers regarding the likelihood of PHEV 
production during the MY 2011 to 2015 timeframe, it will make every 
effort to include PHEVs as a technology in its final rule. To enable 
the possible inclusion of PHEVs as a technology, NHTSA would also have 
to configure the Volpe model to account for the estimated source(s) 
that would supply the electricity for electrical grid charging of the 
battery. Work has started on this effort, but has not yet been 
completed.
    Tables III-1 through III-3 below summarize for each of the 10 
classes of vehicles the cost and effectiveness assumptions used in this 
rulemaking as well as the year of availability of each technology. The 
agency seeks comments on our assumptions and the cost and effectiveness 
estimates provided.

                                                         Table III-1.--Technology Cost Estimates
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                           Vehicle technology incremental retail price equivalent per vehicle ($) by vehicle class
                                                   -----------------------------------------------------------------------------------------------------
                   Technologies                     Subcompact   Compact   Midsize    Large     Small     Small              Midsize    Large     Large
                                                        car        car       car       car     pickup      SUV     Minivan     SUV     pickup      SUV
--------------------------------------------------------------------------------------------------------------------------------------------------------
Low friction lubricants--incremental to base                3          3         3         3         3         3         3         3         3         3
 engine...........................................
Engine friction reduction--incremental to base           0-84       0-84     0-126     0-126     0-126     0-126     0-126     0-126     0-168     0-168
 engine...........................................
Overhead Cam Branch...............................  ..........  ........  ........  ........  ........  ........  ........  ........  ........  ........
VVT--intake cam phasing...........................         59         59       119       119       119       119       119       119       119       119
VVT--coupled cam phasing..........................         59         59       119       119       119       119       119       119       119       119
VVT--dual cam phasing.............................         89         89       209       209       209       209       209       209       209       209
Cylinder deactivation.............................       n.a.       n.a.       203       203       203       203       203       203       229       229
Discrete VVLT.....................................        169        169       246       246       246       246       246       246       322       322
Continuous VVLT...................................        254        254       466       466       466       466       466       466       508       508
Overhead Valve Branch.............................  ..........  ........  ........  ........  ........  ........  ........  ........  ........  ........
Cylinder deactivation.............................       n.a.       n.a.       203       203       203       203       203       203       229       229
 VVT--coupled cam phasing.........................         59         59        59        59        59        59        59        59        59        59
 Discrete VVLT....................................        169        169       246       246       246       246       246       246       322       322
Continuous VVLT (includes conversion to Overhead          599        599      1262      1262      1262      1262      1262      1262      1380      1380
 Cam).............................................
Camless valvetrain (electromagnetic)..............    336-673    336-673   336-673   336-673   336-673   336-673   336-673   336-673   336-673   336-673
 GDI--stoichiometric..............................    122-420    122-420   204-525   204-525   204-525   204-525   204-525   204-525   228-525   228-525
GDI--lean burn....................................        750        750       750       750       750       750       750       750       750       750
Gasoline HCCI dual-mode...........................        263        263       390       390       390       390       390       390       685       685
Turbocharge & downsize............................        690        690       120       120       120       120       120       120       810       810
Diesel--Lean NOX trap.............................       1586       1586  ........  ........  ........  ........  ........  ........  ........  ........
Diesel--urea SCR..................................  ..........  ........      2051      2051      2411      2411      2126      2411      2261      2261
Aggressive shift logic............................         38         38        38        38        38        38        38        38        38        38
Early torque converter lockup.....................         30         30        30        30        30        30        30        30        30        30
5-speed automatic.................................     76-167     76-167    76-167    76-167    76-167    76-167    76-167    76-167    76-167    76-167
6-speed automatic.................................     76-187     76-187    76-187    76-187    76-187    76-187    76-187    76-187    76-187    76-187
6-speed AMT.......................................        141        141       141       141       141       141       141       141       141       141
6-speed manual....................................        107        107       107       107       107       107       107       107       107       107
CVT...............................................        100        100       139       139      n.a.       139       139       139      n.a.      n.a.
Stop-Start with 42 volt system....................        563        563       600       600       600       600       600       600       600       600
IMA/ISA/BSG (includes engine downsize)............       1636       1636      2274      2274       n.a       n.a       n.a       n.a       n.a       n.a
2-Mode hybrid electric vehicle....................       n.a.       n.a.      4655      4655      4655      4655      4655      4655      6006      6006
Power-split hybrid electric vehicle (P-S HEV).....  3700-3850   3700-385  3700-385  3700-385  3700-385  3700-385  3700-385  3700-385  ........  ........
                                                                       0         0         0         0         0         0         0
 Plug-in hybrid electric vehicle (PHEV)...........       4500       4500      6750      6750      6750      6750      6750      6750     10200     10200
Improved high efficiency alternator &                 124-166    124-166   124-166   124-166   124-166   124-166   124-166   124-166   124-166   124-166
 electrification of accessories (12 volt).........
Electric power steering (12 or 42 volt)...........    118-197    118-197   118-197   118-197   118-197   118-197   118-197   118-197   118-197   118-197
Improved high efficiency alternator &                 124-166    124-166   124-166   124-166   124-166   124-166   124-166   124-166   124-166   124-166
 electrification of accessories (42 volt).........
Aero drag reduction (20% on cars, 10% on trucks)..       0-75       0-75      0-75      0-75      0-75      0-75      0-75      0-75      0-75      0-75
Low rolling resistance tires (10%)................          6          6         6         6         6         6         6         6  ........  ........
Low drag brakes (ladder frame only)...............  ..........  ........  ........  ........        87        87  ........        87        87        87
Secondary axle disconnect (unibody only)..........        676        676       676       676       676       676       676       676  ........  ........
Front axle disconnect (ladder frame only).........  ..........  ........  ........  ........       114       114  ........       114       114       114
Weight reduction (1%)--above 5,000 lbs only.......  ..........  ........  ........  ........  ........  ........  ........  ........       \1\       \1\
Weight reduction (2%)--incremental to 1%..........  ..........  ........  ........  ........  ........  ........  ........  ........       \1\       \1\

[[Page 24380]]


Weight reduction (3%)--incremental to 2%..........  ..........  ........  ........  ........  ........  ........  ........  ........       \2\       \2\
--------------------------------------------------------------------------------------------------------------------------------------------------------
\1\ 2/pound.
\2\ 3/pound.


                                                Table III-2.--Technology Percent Effectiveness Estimates
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                               Vehicle technology incremental fuel consumption reduction (%) by vehicle class
                                                   -----------------------------------------------------------------------------------------------------
                   Technologies                     Subcompact   Compact   Midsize    Large     Small     Small              Midsize    Large     Large
                                                        car        car       car       car     pickup      SUV     Minivan     SUV     pickup      SUV
--------------------------------------------------------------------------------------------------------------------------------------------------------
Low friction lubricants--incremental to base              0.5        0.5       0.5       0.5       0.5       0.5       0.5       0.5       0.5       0.5
 engine...........................................
Engine friction reduction--incremental to base            1-3        1-3       1-3       1-3       1-3       1-3       1-3       1-3       1-3       1-3
 engine...........................................
    Overhead Cam Branch
VVT--intake cam phasing...........................          2          2         1         1         1         1         1         1         2         2
VVT--coupled cam phasing..........................          1          1         3         3         2         2         1         1         2         2
VVT--dual cam phasing.............................          1          1         3         3         1         1         1         1         2         2
Cylinder deactivation.............................        n/a        n/a       4.5       4.5       4.5       4.5       4.5       4.5       4.5       4.5
Discrete VVLT.....................................          3          3       1.5       1.5       1.5       1.5       0.5       0.5       1.5       1.5
Continuous VVLT...................................          4          4       3.5       3.5       2.5       2.5       1.5       1.5       2.5       2.5
    Overhead Valve Branch
Cylinder deactivation.............................        n/a        n/a         6         6         6         6         6         6         6         6
VVT--coupled cam phasing..........................          3          3       2.5       2.5       1.5       1.5       0.5       0.5       2.5       2.5
Discrete VVLT.....................................        1.5        1.5       1.5       1.5       1.5       1.5       0.5       0.5       1.5       1.5
Continuous VVLT (includes conversion to Overhead          2.5        2.5       3.5       3.5       2.5       2.5       1.5       1.5       2.5       2.5
 Cam).............................................
Camless valvetrain (electromagnetic)..............        2.5        2.5       2.5       2.5       2.5       2.5       2.5       2.5       2.5       2.5
GDI--stoichiometric...............................        1-2        1-2       1-2       1-2       1-2       1-2       1-2       1-2       1-2       1-2
GDI--lean burn....................................         --         --        --        --        --        --        --        --        --        --
Gasoline HCCI dual-mode...........................      10-12      10-12     10-12     10-12     10-12     10-12     10-12     10-12     10-12     10-12
Turbocharge & Downsize............................    5.0-7.5    5.0-7.5   5.0-7.5   5.0-7.5   5.0-7.5   5.0-7.5   5.0-7.5   5.0-7.5   5.0-7.5   5.0-7.5
Diesel--Lean NOx trap.............................       11.5       11.5       n/a       n/a       n/a       n/a       n/a       n/a       n/a       n/a
Diesel--urea SCR..................................        n/a        n/a      15.5      15.5      15.5      15.5      15.5      15.5      15.5      15.5
Aggressive shift logic............................        1-2        1-2       1-2       1-2       1-2       1-2       1-2       1-2       1-2       1-2
Early torque converter lockup.....................        0.5        0.5       0.5       0.5       0.5       0.5       0.5       0.5       0.5       0.5
5-speed automatic.................................        2.5        2.5       2.5       2.5       2.5       2.5       2.5       2.5       2.5       2.5
6-speed automatic.................................    0.5-2.5    0.5-2.5   0.5-2.5   0.5-2.5   0.5-2.5   0.5-2.5   0.5-2.5   0.5-2.5   0.5-2.5   0.5-2.5
6-speed AMT.......................................    4.5-7.5    4.5-7.5   4.5-7.5   4.5-7.5   4.5-7.5   4.5-7.5   4.5-7.5   4.5-7.5   4.5-7.5   4.5-7.5
6-speed manual....................................        0.5        0.5       0.5       0.5       0.5       0.5       0.5       0.5       0.5       0.5
CVT...............................................        3.5        3.5       3.5       3.5       n/a       3.5       3.5       3.5       n/a       n/a
Stop-Start with 42 volt system....................        7.5        7.5       7.5       7.5       7.5       7.5       7.5       7.5       7.5       7.5
IMA/ISA/BSG (includes engine downsize)............        8.5        8.5       3.5       3.5       n/a       n/a       n/a       n/a       n/a       n/a
2-Mode hybrid electric vehicle....................        n/a        n/a       3.5       3.5         7         7         7         7       3.5       3.5
Power-split hybrid electric vehicle (P-S HEV).....          5          5       6.5       6.5       6.5       6.5       6.5       6.5       n/a       n/a
Plug-in hybrid electric vehicle (PHEV)............         28         28        28        28        28        28        28        28        31        31
Improved high efficiency alternator &                     1-2        1-2       1-2       1-2       1-2       1-2       1-2       1-2       1-2       1-2
 electrification of accessories (12 volt).........
Electric power steering (12 or 42 volt)...........        1.5        1.5     1.5-2     1.5-2         2         2         2         2         2         2
Improved high efficiency alternator &                     1-2        1-2       1-2       1-2       1-2       1-2       1-2       1-2       1-2       1-2
 electrification of accessories (42 volt).........
Aero drag reduction (20% on cars, 10% on trucks)..          3          3         3         3         2         2         3         3         2         2
Low rolling resistance tires (10%)................        1-2        1-2       1-2       1-2       1-2       1-2       1-2       1-2       n/a       n/a
Low drag brakes (ladder frame only)...............        n/a        n/a       n/a       n/a         1         1       n/a       n/a         1         1
Secondary axle disconnect (unibody only)..........          1          1         1         1         1         1         1         1       n/a       n/a
Front axle disconnect (ladder frame only).........        n/a        n/a       n/a       n/a       1.5       1.5       n/a       n/a       1.5       1.5
Weight reduction (1%)--above 5,000 lbs only.......        n/a        n/a       n/a       n/a       n/a       n/a       n/a       n/a       0.7       0.7
Weight reduction (2%)--incremental to 1%..........        n/a        n/a       n/a       n/a       n/a       n/a       n/a       n/a       0.7       0.7
Weight reduction (3%)--incremental to 2%..........        n/a        n/a       n/a       n/a       n/a       n/a       n/a       n/a       0.7       0.7
--------------------------------------------------------------------------------------------------------------------------------------------------------


[[Page 24381]]


                   Table III-3.--Year of Availability
------------------------------------------------------------------------
               Technologies                    Year of  availability
------------------------------------------------------------------------
Low friction lubricants--incremental to    Present.
 base engine.
Engine friction reduction--incremental to  Present.
 base engine.
Overhead Cam Branch
    VVT--intake cam phasing..............  Present.
    VVT--coupled cam phasing.............  Present.
    VVT--dual cam phasing................  Present.
    Cylinder deactivation................  Present.
    Discrete VVLT........................  Present.
    Continuous VVLT......................  Present.
Overhead Valve Branch
    Cylinder deactivation................  Present.
    VVT--coupled cam phasing.............  Present.
    Discrete VVLT........................  Present.
    Continuous VVLT (includes conversion   Present.
     to Overhead Cam).
Camless valvetrain (electromagnetic).....  2020.
GDI--stoichiometric......................  Present.
GDI--lean burn...........................  2020.
Gasoline HCCI dual-mode..................  2016.
Turbocharging & Downsizing...............  2010.
Diesel--Lean NOX trap....................  2010.
Diesel--urea SCR.........................  2010.
Aggressive shift logic...................  Present.
Early torque converter lockup............  Present.
5-speed automatic........................  Present.
6-speed automatic........................  Present.
6-speed AMT..............................  2010.
6-speed manual...........................  Present.
CVT......................................  Present.
Stop-Start with 42 volt system...........  2014.
IMA/ISA/BSG (includes engine downsize)...  2014.
2-Mode hybrid electric vehicle...........  2014.
Power-split hybrid electric vehicle (P-S   2014.
 HEV).
Full-Series hydraulic hybrid.............  NA.
Plug-in hybrid electric vehicle (PHEV)...  NA.
Full electric vehicle (EV)...............  NA.
Improved high efficiency alternator &      Present.
 electrification of accessories (12 volt).
Electric power steering (12 or 42 volt)..  Present.
Improved high efficiency alternator &      Present.
 electrification of accessories (42 volt).
Aero drag reduction (20% on cars, 10% on   Present.
 trucks).
Low rolling resistance tires (10%).......  Present.
Low drag brakes (ladder frame only)......  Present.
Secondary axle disconnect (unibody only).  2012.
Front axle disconnect (ladder frame only)  Present.
Weight reduction (1%)--above 6,000 lbs     Present.
 only.
Weight reduction (2%)--incremental to 1%.  Present.
Weight reduction (3%)--incremental to 2%.  Present.
------------------------------------------------------------------------

C. Technology Synergies

    When two or more technologies are added to a particular vehicle 
model to improve its fuel efficiency, the resultant fuel consumption 
reduction may sometimes be higher or lower than the product of the 
individual effectiveness values for those items. This may occur because 
one or more technologies applied to the same vehicle partially address 
the same source or sources of engine or vehicle losses. Alternately, 
this effect may be seen when one technology shifts the engine operating 
points, and therefore increases or reduces the fuel consumption 
reduction achieved by another technology or set of technologies. The 
difference between the observed fuel consumption reduction associated 
with a set of technologies and the product of the individual 
effectiveness values in that set is sometimes referred to as a 
``synergy.'' Synergies may be positive (increased fuel consumption 
reduction compared to the product of the individual effects) or 
negative (decreased fuel consumption reduction).
    The NAS committee which authored the 2002 Report was aware of 
technology synergies and considered criticisms as part of the peer-
review process that its analysis was ``judgment-simplified,'' but 
concluded overall that its approach was ``sufficiently rigorous'' for 
purposes of the report.\64\ After examining its analysis again, the 
committee stated that ``* * * the path 1 and path 2 estimate average 
fuel consumption improvements * * * appear quite reasonable, although 
the uncertainty in the analysis grows as more technology features are 
considered.''\65\ In essence, as more technology features are 
considered, the features are more likely to overlap and result in 
synergies. Because NAS did not expect vehicle manufacturers to reach 
``path 3'' in the timeframe considered, it did not concern itself 
deeply with the effect of technology synergies in its analysis.
---------------------------------------------------------------------------

    \64\ NAS Report, p. 151.
    \65\ Id.
---------------------------------------------------------------------------

    NHTSA's rulemaking regarding CAFE standards for MY 2008-MY 2011 
light trucks made significant use of NAS' ``path 2'' estimates of the 
effectiveness and cost of available technologies. In part because its 
analysis did not extend to the more aggressive ``path 3,'' the agency 
concluded that the NAS-based multiplicative approach it followed when 
aggregating these technologies was reasonable. In contrast, the 
agency's current proposal is based on an analysis that includes a 
broader range of technologies than was considered by NAS in 2001 and 
2002. Also, the extent to which technologies are included in the 
current analysis is more consistent with NAS' prior ``path 3'' 
approach. Therefore, the agency's current analysis uses estimated 
``synergies'' to address the uncertainties mentioned in the 2002 NAS 
report.
    The Volpe model has been modified to estimate the interactions of 
technologies using estimates of incremental synergies associated with a 
number of technology pairs identified by NHTSA, Volpe Center, and EPA 
staff. The use of discrete technology pair incremental synergies is 
similar to that in DOE's National Energy Modeling System (NEMS).\66\ 
Inputs to the Volpe model incorporate NEMS-identified pairs, as well as 
additional pairs from the set of technologies considered in the Volpe 
model. However, to maintain an approach that was consistent with the 
technology sequencing developed by NHTSA, Volpe Center, and EPA staff, 
new incremental synergy estimates for all pairs were obtained from a 
first-order ``lumped parameter'' analysis tool created by EPA.\67\ 
Results of this analysis were generally consistent with those of full-
scale vehicle simulation modeling performed by Ricardo, Inc.\68\ 
NHTSA's analysis applies these incremental synergy values, obtained 
from the tool using baseline passenger car engine and vehicle inputs, 
to all vehicle classes.
---------------------------------------------------------------------------

    \66\ U.S. Department of Energy, Energy Information 
Administration, Transportation Sector Module of the National Energy 
Modeling System: Model Documentation 2007, May 2007, Washington, DC, 
DOE/EIA-M070(2007), pp. 29-30.
    \67\ This tool is a simple spreadsheet model that represents 
energy consumption in terns of average performance over the fuel 
economy test procedure, rather than explicitly analyzing specific 
drive cycles. The tool begins with an apportionment of fuel 
consumption across several loss mechanisms, and accounts for the 
average extent to which different technologies affect these loss 
mechanisms, using estimates of engine and motor characteristics and 
other variables that are averaged over a driving cycle.
    \68\ EPA contracted with Ricardo, Inc. (an independent 
consulting firm) to study the potential effectiveness of carbon 
dioxide-reducing (and thus, fuel economy-improving) vehicle 
technologies. The Ricardo study is available in the docket for this 
NPRM.
---------------------------------------------------------------------------

    Incremental synergy values are specified in Volpe model input files 
in two ways: as part of the incremental effectiveness values table 
(same path technologies) and in a separate incremental synergies table 
(separate path technologies). In the case of same path technologies, 
each technology's incremental effectiveness value was obtained from the 
technical literature and manufacturers' submitted information, and then 
the sum of all

[[Page 24382]]

incremental synergies associated with that technology and each 
technology located higher on the same path was subtracted to determine 
the incremental effectiveness. For example, all engine technologies 
take into account incremental synergy factors of preceding engine 
technologies; all transmission technologies take into account 
incremental synergy factors of preceding transmission technologies. 
These factors are expressed in the fuel consumption improvement factors 
in the input files used by the Volpe model.
    For applying incremental synergy factors in separate path 
technologies, the Volpe model uses an input table which lists 
technology pairings and incremental synergy factors associated with 
those pairings, most of which are between engine technologies and 
transmission technologies. When a technology is applied to a vehicle by 
the Volpe model, all instances of that technology in the incremental 
synergy table which match technologies already applied to the vehicle 
(either pre-existing or previously applied by the Volpe model) are 
summed and applied to the fuel consumption improvement factor of the 
technology being applied. When the Volpe model applies incremental 
synergies, the fuel consumption improvement factors cannot be reduced 
below zero.
    Incremental synergy values were calculated assuming the prior 
application (implying succession in some cases) of all technologies 
located higher along both paths than the pair considered. This is 
usually a true reflection of a given vehicle's equipment at any point 
in the model run and thus the method is expected to produce reasonable 
results in most cases.
    NHTSA considered other methods for estimating interactions between 
technologies. For example, the agency has considered integrating 
detailed simulation of individual vehicles' performance into the Volpe 
model.\69\ However, while application of such simulation techniques 
could provide a useful source of information when developing inputs to 
the Volpe model, the agency believes that applying detailed simulation 
when analyzing the entire fleet of future vehicles is neither necessary 
nor feasible. NHTSA is charged with setting standards at the maximum 
feasible level. To understand the potential impacts of its standards, 
the agency analyzes entire fleets of vehicles expected to be produced 
in the future. Although some expected engineering characteristics of 
these vehicles are available, the level of detail needed for full 
vehicle simulation--a level of detail that would be important if NHTSA 
were actually designing vehicles--is not available.
---------------------------------------------------------------------------

    \69\ In other words, this would mean having the Volpe model run 
a full vehicle simulation every time the Volpe model is evaluating 
the potential effect of applying a specific technology to a specific 
vehicle model.
---------------------------------------------------------------------------

    As another possible alternative to using ``synergy'' factors, NHTSA 
has also considered modifying the Volpe model to accept as inputs 
different measures of efficiency for each engine, transmission, and 
vehicle model in the product plans. For instance, manufacturers could 
provide estimates of mechanical and drivetrain efficiencies. Mechanical 
efficiency (usually between 70 and 90 percent) gives an estimate of the 
amount of fuel consumed by engine friction and pumping losses. 
Drivetrain efficiency (usually between 80 and 90 percent) gives an 
estimate of the amount of fuel consumed by parasitic loads, gearbox 
friction, and torque converter losses. From these efficiencies along 
with other inputs such as compression ratio, aerodynamic drag, rolling 
resistance, and vehicle mass, the model could estimate the fuel 
consumption associated with each loss mechanism and enforce a maximum 
fuel consumption reduction for each vehicle model based on those 
estimates and the technologies applied. Like the use of incremental 
synergies, this method could help the model avoid double counting fuel 
consumption benefits when applying multiple technologies to the same 
vehicle model.\70\ The agency believes that this approach, like the use 
of ``synergy'' factors currently used by the Volpe model, could 
conceivably provide a means of addressing uncertainty in fuel 
consumption estimation within the context of CAFE analysis. However, 
the agency is not confident that model-by-model estimates of baseline 
fuel consumption partitioning would be available. Also, partitioned 
estimates of the effects of all the technologies considered in the 
analysis of this proposal were not available. If both of these concerns 
could be addressed, NHTSA believes it would be possible to implement 
partitioned accounting of fuel consumption. However, the agency is 
unsure whether and, if so, to what extent doing so would represent an 
improvement over our current approach of using incremental synergy 
factors.
---------------------------------------------------------------------------

    \70\ This approach was proposed in a paper criticizing NAS' 
approach to synergies in the 2001-02 peer-review process for the NAS 
Report. See Patton, et al., ``Aggregating Technologies for Reduced 
Fuel Consumption: A Review of the Technical Content in the 2002 
National Research Council Report on CAFE'', SAE 2002-01-0628, March 
2002.
---------------------------------------------------------------------------

    The agency solicits comments on its use of incremental synergy 
factors to address uncertainty in the estimation of the extent to which 
fuel consumption is reduced by applying technologies. For additional 
detail on the synergies used, please see Section V of this document. In 
particular, the agency solicits comment on (a) the values of the 
factors the agency has applied, (b) possible variations across the ten 
categories of vehicles the agency has considered, and (c) additional 
technology pairs that may involve such interactions. The proposal of 
any additional methodologies, such as prototyping and testing, full 
vehicle simulation, or partitioned accounting, should address 
information and resource requirements, particularly as related to the 
analysis of entire fleets of future vehicles expected to be produced 
through MY 2015. Synergies used for this analysis can be found in 
Section V of this document.

D. Technology Cost Learning Curve

    In past rulemaking analyses, NHTSA did not explicitly account for 
the cost reductions a manufacturer may realize through learning 
achieved from experience in actually applying a given technology. NHTSA 
understood technology cost-estimates to reflect already the full 
learning costs of technology. EPA felt that for some of the newer, 
emerging technologies, cost estimates did not reflect the full impact 
of learning. NHTSA tentatively agreed, but is seeking comment on the 
impact of learning on cost and the production volumes where it occurs. 
NHTSA has modified its previous approach in this rulemaking for that 
reason. In this rulemaking we have included a learning factor for some 
of the technologies. The ``learning curve'' describes the reduction in 
unit incremental production costs as a function of accumulated 
production volume and small redesigns that reduce costs.
    NHTSA implemented technology learning curves by using three 
parameters: (1) The initial production volume that must be reached 
before cost reductions begin to be realized (referred to as ``threshold 
volume''); (2) the percent reduction in average unit cost that results 
from each successive doubling of cumulative production volume (usually 
referred to as the ``learning rate''); and (3) the initial cost of the 
technology. Section V below describing the Volpe model contains 
additional information on learning curve functions.
    Figure III-1 illustrates a learning curve for a vehicle technology 
with an

[[Page 24383]]

initial average unit cost of $100 and a learning rate of approximately 
20 percent. In this hypothetical example, the initial production volume 
before cost reductions begin to be realized is set at 12,000 units and 
the production volume at the cost floor is set at roughly 50,000 units 
with a cost of $64.
[GRAPHIC] [TIFF OMITTED] TP02MY08.001

    Most studies of the effect of the learning curve on production 
costs appear to assume that cost reductions begin only after some 
initial volume threshold has been reached, but not all of these studies 
specify what this threshold volume is. The rate at which costs decline 
beyond the initial threshold is usually expressed as the percent 
reduction in average unit cost that results from each successive 
doubling of cumulative production volume, sometimes referred to as the 
learning rate. Many estimates of learning experience curves do not 
specify a cumulative production volume beyond which cost reductions no 
longer occur, instead depending on the asymptotic behavior of the above 
expression of (CQ) for learning rates below 100 percent to establish a 
floor on costs.
    For this analysis, NHTSA has applied learning curve cost reductions 
on a manufacturer-specific basis, and has assumed that learning-based 
reductions in technology costs occur at the point that a manufacturer 
applies the given technology to the first 25,000 cars or trucks, and 
are repeated a second time as it produces another 25,000 cars or trucks 
for the second learning step (car and truck volumes are treated 
separately for determining these sales volumes). The volumes chosen 
represent our best estimate for where learning would occur. As such, we 
believe that these estimates are better suited to this analysis than a 
more general approach of a single number for the learning curve factor, 
because each manufacturer would be implementing technologies at its own 
pace in this rule, rather than assuming that all manufacturers 
implement identical technology at the same time. NHTSA is aware that 
some of the cost estimates that it has relied upon were derived from 
suppliers and has added multipliers so that these costs are reflective 
of what manufacturers would pay for this technology. NHTSA seeks 
comments on the estimated level of price markups that manufacturers pay 
for technologies purchased from suppliers and whether different 
learning curves should be applied to those types of technologies. In 
addition, NHTSA seeks comments on how learning curves should be 
adjusted if a supplier supplies more than one manufacturer.
    Ideally, we would know the development production cycle and 
maturity level for each technology so that we could calculate learning 
curves precisely. Without that knowledge, we have to use engineering 
judgment. After having produced 25,000 cars or trucks with a specific 
part or system, we believe that sufficient learning will have taken 
place such that costs will be lower by 20 percent for some technologies 
and 10 percent for others. After another 25,000 units, it is expected 
that, for some technologies, such as 6-speed AMTs, another cost 
reduction will have been realized.
    For each of the technologies, we have considered whether we could 
project future cost reductions due to manufacturer learning. In making 
this determination, we considered whether or not the technology was in 
wide-spread use today or expected to be by the model year 2011-2012 
time frame, in which case no future learning curve would apply because 
the technology would already be in wide-spread production by the 
automotive industry by that timeframe, e.g., on the order of multi-
millions of units per year. (Examples of these include 5-speed 
automatic transmissions and intake-cam phasing variable valve timing. 
These technologies have been in production for light-duty vehicles for 
more than 10 years.) In addition, we carefully considered the 
underlying source data for our cost estimates. If the source data 
specifically stated that manufacturer cost reduction from future 
learning would occur, we took that information into account in 
determining whether we would apply manufacturer learning in our cost 
projections. Thus, for many of the technologies, we have not applied 
any future cost reduction learning curve.
    However, there are a number of technologies which are not yet in 
mass production for which we have applied a learning curve. As 
indicated in Table III-4 below, we have applied the learning curve 
beginning in MY 2011 to one set of technologies, and for a number of 
additional technologies we did not apply manufacturer learning until MY 
2014. The distinction between MYs 2011 and 2014 is due to our source 
data for our cost estimates. For those technologies where we have 
applied manufacturer learning in MY 2011, the source of our cost 
estimate did not rely on manufacturer learning to develop the initial 
cost estimate we have used--therefore we apply the manufacturer

[[Page 24384]]

learning methodology beginning in MY 2011.

        Table III.-4.--Learning Curve Application to Technologies
------------------------------------------------------------------------
                                                 First year    Learning
                  Technology                         of         factor
                                                application   (percent)
------------------------------------------------------------------------
Overhead Cam Branch Cylinder deactivation.....         2014           20
Continuous VVLT...............................         2014           20
Camless valvetrain (electromagnetic)..........         2011           20
GDI--lean burn................................         2011           20
Gasoline HCCI dual-mode.......................         2011           20
Turbocharging & downsizing....................         2014           20
Diesel--Lean NOX trap*........................         2011           10
Diesel--urea SCR*.............................         2011           10
6-speed AMT...................................         2011           20
Stop-Start with 42 volt system................         2014           20
IMA/ISA/BSG (includes engine downsize)........         2014           20
2-Mode hybrid electric vehicle................         2014           20
Power-split hybrid electric vehicle (P-S HEV).         2014           20
Plug-in hybrid electric vehicle (PHEV)........         2011           20
Improved high efficiency alternator &                  2011           20
 electrification of accessories (42 volt).....
Secondary axle disconnect (unibody only)......         2011           20
Weight reduction (1%)--above 6,000 lbs only...         2011           20
Weight reduction (2%)--incremental to 1%......         2011           20
Weight reduction (3%)--incremental to 2%......         2011          20
------------------------------------------------------------------------
* For diesel technologies, learning is only applied to the cost of the
  emission control equipment, not the cost for the entire diesel system.

    The technologies for which we do not begin applying learning until 
2014 all have the same reference source, the 2004 NESCCAF study, for 
which the sub-contractor was The Martec Group. In the work done for the 
2004 NESCCAF report, Martec relied upon actual price quotes from Tier 1 
automotive suppliers to develop automotive manufacturer cost estimates. 
Based on information presented by Martec to the National Academy of 
Sciences (NAS) Committee during their January 24, 2008, public meeting 
in Dearborn, Michigan,\71\ we understand that the Martec cost estimates 
incorporated some element of manufacturer learning. Martec stated that 
the Tier 1 suppliers were specifically requested to provide price 
quotes which would be valid for three years (2009-2011), and that for 
some components the Tier 1 supplier included cost reductions in years 
two and three which the supplier anticipated could occur, and which 
they anticipated would be necessary in order for their quote to be 
competitive with other suppliers. Therefore, for this analysis, we did 
not apply any learning curve to any of the Martec-sourced costs for the 
first three years of this proposal (2011-2013). However, the theory of 
manufacturer learning is that it is a continuous process, though the 
rate of improvement decreases as the number of units produced 
increases. While we were not able to gain access to the detailed 
submissions from Tier 1 suppliers which Martec relied upon for their 
estimates, we do believe that additional cost reductions will occur in 
the future for a number of the technologies for which we relied upon 
the Martec cost estimates for the reasons stated above in reference to 
the general learning curve effect. For those technologies we applied a 
learning curve beginning in 2014. Martec has recently submitted a study 
to the NAS Committee comparing the 2004 NESCCAF study with new updated 
cost information. Given that this study had just been completed, the 
agency could not take it into consideration for the NPRM. However, the 
agency will review the new study and consider its findings in time for 
the final rule.
---------------------------------------------------------------------------

    \71\ ``Variable Costs of Fuel Economy Technologies'' Martec 
Group, Inc Report Presented to: Committee to Assess Technologies for 
Improving Light-Duty Vehicle Fuel Economy. Division on Engineering 
and Physical Systems, Board on Energy and Environmental Systems, the 
National Academy of Sciences, January 24, 2008.
---------------------------------------------------------------------------

    Manufacturers' actual costs for applying these technologies to 
specific vehicle models are likely to include significant additional 
outlays for accompanying design or engineering changes to each model, 
development and testing of prototype versions, recalibrating engine 
operating parameters, and integrating the technology with other 
attributes of the vehicle. Manufacturers may also incur additional 
corporate overhead, marketing, or distribution and selling expenses as 
a consequence of their efforts to improve the fuel economy of 
individual vehicle models and their overall product lines.
    In order to account for these additional costs, NHTSA has applied 
an indirect cost multiplier of 1.5 to its estimate of the vehicle 
manufacturers' direct costs for producing or acquiring each fuel 
economy-improving technology to arrive at a consumer cost. This 
estimate was developed by Argonne National Laboratory in a recent 
review of vehicle manufacturers' indirect costs. The Argonne study was 
specifically intended to improve the accuracy of future cost estimates 
for production of vehicles that achieve high fuel economy by employing 
many of the same advanced technologies considered in the agency's 
analysis.\72\ Thus, its recommendation that a multiplier of 1.5 be 
applied to direct manufacturing costs to reflect manufacturers' 
increased indirect costs for deploying advanced fuel economy 
technologies appears to be appropriate for use in the current analysis. 
Historically, NHTSA has used almost the exact same multiplier, a 
multiplier of 1.51, as the markup from variable costs or direct 
manufacturing costs to consumer costs. This markup takes into account 
fixed costs, burden, manufacturer's profit, and dealer's profit. Table 
VII-2 of the PRIA shows the estimated incremental consumer costs for 
each vehicle type.\73\
---------------------------------------------------------------------------

    \72\ Vyas, Anant, Dan Santini, and Roy Cuenca, Comparison of 
Indirect Cost Multipliers for Vehicle Manufacturing, Center for 
Transportation Research, Argonne National Laboratory, April 2000.
    \73\ PRIA, VII-9.

---------------------------------------------------------------------------

[[Page 24385]]

E. Ensuring Sufficient Lead Time

    In analyzing potential technological improvements to the product 
offerings for each manufacturer with a substantial share of the market, 
NHTSA added technologies based on our engineering judgment and 
expertise about possible adjustments to the detailed product plans 
submitted to NHTSA. Our decision whether and when to add a technology 
reflected our consideration of the practicability of applying a 
specific technology and the necessity for lead time in its application. 
NHTSA recognizes that vehicle manufacturers must have sufficient lead 
time to incorporate changes and new features into their vehicles and 
hence added technologies in a cost-minimizing fashion. That is, we 
generally added technologies that were most cost-effective and took 
into account the year of availability of the technologies.
    NHTSA realizes that not all technologies will be available 
immediately or could be applied immediately and that there are 
different phase-in rates (how rapidly a technology is able to be 
applied across a manufacturer's fleet of vehicles) applicable to each 
technology as well as windows of opportunities when certain 
technologies could be applied (i.e., when a product is redesigned or 
refreshed).
a. Linking To Redesign and Refresh
    In the automobile industry there are two terms that describe when 
changes to vehicles occur: redesign and refresh. In projecting the 
technologies that could be applied to specific vehicle models, NHTSA 
tied the application of the majority of the technologies to a vehicle's 
refresh/redesign cycle. Vehicle redesign usually encompasses changes to 
a vehicle's appearance, shape, dimensions, and powertrain and is 
traditionally associated with the introduction of ``new'' vehicles into 
the market, and often is characterized as the next generation of a 
vehicle. In contrast vehicle refresh usually only encompasses changes 
to a vehicle's appearance, and may include an upgraded powertrain and 
is traditionally associated with mid-cycle cosmetic changes to a 
vehicle within its current generation to make it appear ``fresh.'' 
Vehicle refresh traditionally occurs no earlier than two years after a 
vehicle redesign or at least two years before a scheduled redesign. 
Table III-5 below contains a complete list of the technologies that 
were applied and whether NHTSA allowed them to be applied during a 
redesign year, a refresh year or during any model year is shown in the 
table below.

                            Table III-5.--Technology Refresh and Redesign Application
----------------------------------------------------------------------------------------------------------------
                                                                           Can be        Can be
                                                                           applied       applied       Can be
                                                                           during       during a       applied
                Technology                             Abbr.              redesign     redesign or   during any
                                                                         model year      refresh     model year
                                                                            only       model year
----------------------------------------------------------------------------------------------------------------
Low Friction Lubricants...................  LUB.......................  ............            X             X
Engine Friction Reduction.................  EFR.......................  ............            X   ............
Variable Valve Timing (ICP)...............  VVTI......................  ............            X   ............
Variable Valve Timing (CCP)...............  VVTC......................  ............            X   ............
Variable Valve Timing (DCP)...............  VVTD......................  ............            X   ............
Cylinder Deactivation.....................  DISP......................  ............            X   ............
Variable Valve Lift & Timing (CVVL).......  VVLTC.....................            X   ............  ............
Variable Valve Lift & Timing (DVVL).......  VVLTD.....................            X   ............  ............
Cylinder Deactivation on OHV..............  DISPO.....................  ............            X   ............
Variable Valve Timing (CCP) on OHV........  VVTO......................  ............            X   ............
Multivalve Overhead Cam with CVVL.........  DOHC......................            X   ............  ............
Variable Valve Lift & Timing (DVVL) on OHV  VVLTO.....................            X   ............  ............
Camless Valve Actuation...................  CVA.......................            X   ............  ............
Stoichiometric GDI........................  SIDI......................            X   ............  ............
Lean Burn GDI.............................  LBDI......................            X   ............  ............
Turbocharging and Downsizing..............  TURB......................            X   ............  ............
HCCI......................................  HCCI......................            X   ............  ............
Diesel with LNT...........................  DSLL......................            X   ............  ............
Diesel with SCR...........................  DSLS......................            X   ............  ............
5 Speed Automatic Transmission............  5SP.......................  ............            X   ............
Aggressive Shift Logic....................  ASL.......................  ............            X             X
Early Torque Converter Lockup.............  TORQ......................  ............            X   ............
6 Speed Automatic Transmission............  6SP.......................  ............            X   ............
Automatic Manual Transmission.............  AMT.......................            X   ............  ............
Continuously Variable Transmission........  CVT.......................            X   ............  ............
6 Speed Manual............................  6MAN......................            X   ............  ............
Improved Accessories......................  IACC......................  ............  ............            X
Electronic Power Steering.................  EPS.......................  ............            X   ............
42-Volt Electrical System.................  42V.......................            X   ............  ............
Low Rolling Resistance Tires..............  ROLL......................  ............  ............            X
Low Drag Brakes...........................  LDB.......................  ............  ............            X
Secondary Axle Disconnect--Unibody........  SAXU......................  ............            X   ............
Secondary Axle Disconnect--Ladder Frame...  SAXL......................  ............            X   ............
Aero Drag Reduction.......................  AERO......................  ............            X   ............
Material Substitution (1%)................  MS1.......................            X   ............  ............
Material Substitution (2%)................  MS2.......................            X   ............  ............
Material Substitution (5%)................  MS5.......................            X   ............  ............
ISG with Idle-Off.........................  ISGO......................            X   ............  ............
IMA/ISAD/BSG Hybrid (includes engine        IHYB......................            X   ............  ............
 downsizing).
2-Mode Hybrid.............................  2HYB......................            X   ............  ............

[[Page 24386]]


Power Split Hybrid........................  PHYB......................            X   ............  ............
----------------------------------------------------------------------------------------------------------------

    As can be seen in the above table, most technologies would only be 
applied by the Volpe model when a specific vehicle was due for a 
redesign or refresh. However, for a limited set of technologies, the 
model was not restricted to applying them during a refresh/redesign 
year and thus they were made available for application at any time.
    These specific technologies are:
     Low Friction Lubricants
     Improved Accessories
     Low Rolling Resistance Tires
     Low Drag Brakes
    All of these technologies are very cost-effective, can apply to 
multiple vehicle models/platforms and can be applied across multiple 
vehicle models/platforms in one year. Although they can also be applied 
during a refresh/redesign year, they are not restricted to that 
timeframe because their application is not viewed as necessitating a 
major engineering redesign and testing/calibration.
    There is an additional technology whose application is not tied to 
refresh/redesign, which is Aggressive Shift Logic (ASL). ASL is 
accomplished through reprogramming the shift points for a transmission 
to be more like a manual transmission. Upgrading a transmission to 
utilize ASL can happen at refresh/redesign, but because it is not a 
hardware change, it can also occur at other points in a vehicle's 
design cycle. If a model that is scheduled for refresh/redesign has a 
transmission that is being upgraded to ASL, it is possible that all 
other vehicles that utilize the same transmission (which is usually 
produced at the same manufacturing plant) could be upgraded at the same 
time to incorporate ASL and that ASL could permeate other vehicle 
models in years other than a refresh/redesign year.
    NHTSA based the redesign rates used in the Volpe Model on a 
combination of the manufacturers' confidential product plans and 
NHTSA's engineering judgment. In most instances, NHTSA has accepted the 
projected redesign periods from the companies who provided them through 
MY 2013. If companies did not provide product plan date, NHTSA used 
publicly available data about vehicle redesigns to establish the 
redesign rates for the vehicles produced by these companies.
    NHTSA assumes that passenger cars will be redesigned every 5 years, 
based on the trend over the last 10-15 years for passenger cars to be 
redesigned every 5 years. These trends are reflected in the 
manufacturer production plans that NHTSA received in response to its 
request for product plan information and was confirmed by many 
automakers in meetings held with NHTSA to discuss various issues with 
manufacturers.
    NHTSA believes that the vehicle design process has progressed and 
improved rapidly over the last decade and these improvements have 
resulted in the ability of manufacturers to shorten the design process 
and to introduce vehicles more frequently to respond to competitive 
market forces. Almost all passenger cars will be on a 5-year redesign 
cycle by the end of the decade, with the exception being some high 
performance vehicles and vehicles' with specific market niches.
    Currently, light trucks are redesigned every 5 to 7 years, with 
some vehicles having longer redesign periods (e.g., full-size vans). In 
the most competitive SUV and crossover vehicle segments, the redesign 
cycle currently averages slightly above 5 years. It is expected that 
the light truck redesign schedule will be shortened in the future due 
to competitive market forces and in response to fuel economy and other 
regulatory requirements. It is expected that by MY 2014, almost all 
light trucks will be redesigned on a 5-year cycle. Thus, for almost all 
vehicles scheduled for a redesign in model year 2014 and later, NHTSA 
estimated that all vehicles would be redesigned on a 5-year cycle. 
Exceptions were made for high performance vehicles and other vehicles 
that traditionally had longer than average design cycles (e.g., 2-
seater sports cars). For those vehicles, NHTSA attempted to preserve 
the historic redesign cycle rates.
b. Technology Phase-in Caps
    In analyzing potential technological improvements to the product 
offerings for each manufacturer with a substantial share of the market, 
NHTSA added technologies based on our engineering judgment and 
expertise about possible adjustments to the detailed product plans 
submitted to NHTSA. Our decision whether and when to add a technology 
reflected our consideration of the practicability of applying a 
specific technology and the necessity for lead-time in its application.
    NHTSA recognizes that vehicle manufacturers must have sufficient 
lead time to incorporate changes and new features into their vehicles 
and that these changes cannot occur all at once, but must be phased in 
over time. As discussed above, our analysis addresses these realities 
in part by timing the estimated application of most technologies to 
coincide with anticipated vehicle redesigns and/or freshenings. We have 
estimated that future vehicle redesigns can be implemented on a 5-year 
cycle with mid-cycle freshening, except where manufacturers have 
indicated plans for shorter redesign cycles.
    However, the agency further recognizes that engineering, planning 
and financial constraints prohibit most technologies from being applied 
across an entire fleet of vehicles within a year. Thus, as for the 
analysis supporting its 2006 rulemaking regarding light truck CAFE, the 
agency is employing overall constraints on the rates at which each 
technology can penetrate a manufacturer's fleet. The Volpe model 
applies these ``phase-in caps'' by ceasing to add a given technology to 
a manufacturer's fleet in a specific model year once it has increased 
the corresponding penetration rate by at least amount of the cap. 
Having done so, the model proceeds to apply other technologies in lieu 
of the ``capped'' technology.
    For its regulatory analysis in 2006, NHTSA applied phase-in caps 
expected to be consistent with NAS' indication in its 2002 report that 
even existing technologies would require 4 to 8 years to achieve 
widespread penetration of the fleet. The NAS report, which is believed 
to be the only peer-reviewed source which provides phase-in rates, was 
relied upon for establishing the phase-in caps that we used for all

[[Page 24387]]

technologies, except diesels and hybrids, for which the report didn't 
include that information. Most of the phase-in caps applied by the 
agency in 2006 ranged from 25 percent (4 year introduction) to 17 
percent (approximately 6 years, the midpoint of the NAS estimate). The 
agency assumed shorter implementation rates for technologies that did 
not require changes to the manufacturing line. For other technologies 
(e.g., hybrid and diesel powertrains), the agency employed phase-in 
caps as low as 3 percent, to reflect the major redesign efforts and 
capital investments required to implement these technologies.
    Considerable changes have occurred since NHTSA's 2006 analysis, and 
even more since the 2002 NAS report. Not only have fuel prices 
increased, but official forecasts of future fuel prices have increased, 
as well. This suggests a market environment in which consumers are more 
likely to demand fuel-saving technologies than previously anticipated, 
and it suggests a financial environment in which investors are more 
likely to invest in companies developing and producing such 
technologies. Indeed, some technologies have penetrated the marketplace 
more quickly than projected in 2006. Confidential product plan 
information submitted to NHTSA in 2007 and information from suppliers 
confirm that the rate of technology penetration has increased as 
compared to 2006.
    Also, the statutory environment has changed since 2006. With the 
enactment of EISA, Congress has adopted the specific objectives of 
increasing new vehicle fuel economy to at least 35 mpg by 2020 and 
making ratable progress toward that objective in earlier model years. 
This reduces manufacturers' uncertainty about the general direction of 
future fuel economy standards in the United States. Moreover, 
developments in other regions (e.g., Europe) and countries (e.g., 
Canada and China) suggest that the generalized expectation that future 
vehicles will perform well with respect to energy efficiency is not 
unique to the United States. Discussions with manufacturers in late 
2007 and early 2008 indicate that the industry is highly sensitive to 
all of these developments and has been anticipating the need to 
accelerate the rate of technology deployment in response to the passage 
of major energy legislation in the U.S.
    Considering these developments, the agency revisited the phase-in 
caps it had applied in 2006 and determined that it would be appropriate 
to relax many of them. In our judgment, most of the engine technologies 
could penetrate the fleet in as quickly as five years--rather than in 
the six we previously estimated--as long as they are applied during 
redesign. Low friction lubricants are already widely used, and our 
expectation is that they can quickly penetrate the remainder of the 
fleet. Therefore, we relaxed the 25 percent (4-year) phase-in cap to 50 
percent (2 years). Similarly, product plans indicate that transmissions 
with 5 or more forward gears will widely penetrate the fleet even 
without the current proposal. Also, given the technology cost and 
effectiveness estimates discussed above, the Volpe model frequently 
estimates that manufacturers will ``leapfrog'' past 5-speed 
transmissions to apply more advanced transmissions (e.g., 6-speed or 
AMT). We have therefore increased the phase-in cap for 5-speed 
transmissions from 25 percent (4 years) to 100 percent (1 year). 
However, in our judgment, phase-in caps of 17 percent (6 years) are 
currently still appropriate for most other transmission technologies.
    Although NHTSA has applied phase-in caps of 25 percent (4 years) 
for most remaining technologies, we continue to anticipate that phase-
in caps of 3 percent are appropriate for some advanced technologies, 
such as hybrids and diesels. Although engine, vehicle, and exhaust 
aftertreatment manufacturers have, more recently, expressed greater 
optimism than before regarding the outlook for light vehicle diesel 
engines, our expectation is that the phase-in cap that we have chosen 
is appropriate at this time. We also estimate that a 3 percent rate is 
appropriate for hybrid technologies, which are very complex, require 
significant engineering resources to implement, but are just now 
starting to penetrate the market.
    Table III-6 below presents the phase-in caps applied in the current 
analysis, with rates from the analysis of the 2006 final rule provided 
for comparison. NHTSA requests comments on the phase-in caps shown 
here, and on whether slower or faster rates would be more appropriate 
and, if so, why.

                 Table III.--6. Phase-In Cap Application
------------------------------------------------------------------------
                                                 2006 final    Current
                  Technology                        rule         NPRM
------------------------------------------------------------------------
Low Friction Lubricants.......................           25           50
Engine Friction Reduction.....................           17           20
Variable Valve Timing (ICP)...................           17           20
Variable Valve Timing (CCP)...................           17           20
Variable Valve Timing (DCP)...................           17           20
Cylinder Deactivation.........................           17           20
Variable Valve Lift & Timing (CVVL)...........           17           20
Variable Valve Lift & Timing (DVVL)...........           17           20
Cylinder Deactivation on OHV..................           17           20
Variable Valve Timing (CCP) on OHV............           17           20
Multivalve Overhead Cam with CVVL.............           17           20
Variable Valve Lift & Timing (DVVL) on OHV....           17           20
Camless Valve Actuation.......................           10           20
Stoichiometric GDI............................            3           20
Diesel following GDI-S (SIDI).................            3            3
Lean Burn GDI.................................  ...........           20
Turbocharging and Downsizing..................           17           20
Diesel following Turbo D/S....................            3            3
HCCI..........................................  ...........           13
Diesel following HCCI.........................            3            3
5 Speed Automatic Transmission................           17          100
Aggressive Shift Logic........................           17           25
Early Torque Converter Lockup.................  ...........           25
6 Speed Automatic Transmission................           17           17

[[Page 24388]]


Automated Manual Transmission.................           17           17
Continuously Variable Transmission............           17           17
6 Speed Manual................................  ...........           17
Improved Accessories..........................           25           25
Electric Power Steering.......................           17           25
42-Volt Electrical System.....................           17           25
Low Rolling Resistance Tires..................           25           25
Low Drag Brakes...............................           17           25
Secondary Axle Disconnect--Unibody............           17           17
Secondary Axle Disconnect--Ladder Frame.......           17           17
Aero Drag Reduction...........................           17           17
Material Substitution (1%)....................           17           17
Material Substitution (2%)....................           17           17
Material Substitution (5%)....................           17           17
ISG with Idle-Off.............................            5            3
IMA/ISAD/BSG Hybrid (includes engine                      5            3
 downsizing)..................................
2-Mode Hybrid.................................            5            3
Power Split Hybrid............................            5            3
Plug-in Hybrid................................  ...........            3
------------------------------------------------------------------------

IV. Basis for Attribute-Based Structure for Setting Fuel Economy 
Standards

A. Why attribute-based instead of a single industry-wide average?

    NHTSA is obligated under 49 U.S.C. 32902(a)(3)(A), recently added 
by Congress, to set attribute-based fuel economy standards for 
passenger cars and light trucks. NHTSA welcomes Congress' affirmation 
through EISA of the value of setting attribute-based fuel economy 
standards, because we believe that an attribute-based structure is 
preferable to a single industry-wide average standard for the following 
reasons. First, attribute-based standards increase fuel savings and 
reduce emissions when compared to an equivalent industry-wide standard 
under which each manufacturer is subject to the same numerical 
requirement. Under such a single industry-wide average standard, there 
are always some manufacturers that are not required to make any 
improvements for any given year because they already exceed the 
standard. Under an attribute-based system, in contrast, every 
manufacturer can potentially be required to continue improving each 
year. Because each manufacturer produces a different mix of vehicles, 
attribute-based standards are individualized for each manufacturer's 
different product mix. All manufacturers must ensure they have used 
available technologies to enhance fuel economy levels of the vehicles 
they sell. Therefore, fuel savings and emissions reductions will always 
be higher under an attribute-based system than under a comparable 
industry-wide standard.
    Second, attribute-based standards eliminate the incentive for 
manufacturers to respond to CAFE standards in ways harmful to 
safety.\74\ Because each vehicle model has its own target (based on the 
attribute chosen), attribute-based standards provide no incentive to 
build smaller vehicles simply to meet a fleet-wide average, because the 
smaller vehicles will be subject to more stringent fuel economy and 
emissions targets.
---------------------------------------------------------------------------

    \74\ The 2002 NAS Report, on which NHTSA relied in reforming the 
CAFE program for light trucks, described at length and quantified 
the potential safety problem with average fuel economy standards 
that specify a single numerical requirement for the entire industry. 
See National Academy of Sciences, ``Effectiveness and Impact of 
Corporate Average Fuel Economy (CAFE) Standards,'' (``NAS Report'') 
National Academy Press, Washington, DC (2002), 5, finding 12. 
Available at http://www.nap.edu/openbook.php?record-- 
id=10172page=R1 (last accessed April 20, 2008).
---------------------------------------------------------------------------

    Third, attribute-based standards provide a more equitable 
regulatory framework for different vehicle manufacturers.\75\ A single 
industry-wide average standard imposes disproportionate cost burdens 
and compliance difficulties on the manufacturers that need to change 
their product plans and no obligation on those manufacturers that have 
no need to change their plans. Attribute-based standards spread the 
regulatory cost burden for fuel economy more broadly across all of the 
vehicle manufacturers within the industry.
---------------------------------------------------------------------------

    \75\Id. at 4-5, finding 10.
---------------------------------------------------------------------------

    And fourth, attribute-based standards respect economic conditions 
and consumer choice, instead of having the government mandate a certain 
fleet mix. Manufacturers are required to invest in technologies that 
improve the fuel economy achieved by the vehicles they sell, regardless 
of their size.

B. Which attribute is most effective?

    Although NHTSA previously set the MY 2008-2011 light truck fuel 
economy standards based on vehicle footprint as the relevant attribute, 
the agency took a fresh look for purposes of this rulemaking. Although 
several attributes offer benefits, NHTSA has preliminarily concluded 
that a footprint-based function will again be the most effective and 
efficient for both passenger car and light truck standards. The 
discussion below explains our conclusion in favor of footprint, and 
also examines the relative benefits and drawbacks of the other 
attributes considered.
1. Footprint-Based Function
    NHTSA is proposing to set fuel economy standards for manufacturers 
according to vehicle footprint, as light truck CAFE standards are 
currently set by NHTSA. A vehicle's ``footprint'' is the product of the 
average track width (the distance between the centerline of the tires 
\76\ ) and wheelbase (basically, the distance between the centers of 
the axles \77\ ). Each vehicle footprint value is assigned a mile per 
gallon target specific to that footprint value. Footprint-based

[[Page 24389]]

standards have a number of benefits, as described below.
---------------------------------------------------------------------------

    \76\ The proposed definition for track width is the same as that 
used in NHTSA's April 2006 light truck CAFE rule, which is ``the 
lateral distance between the centerlines of the base tires at 
ground, including camber angle.'' 49 CFR 523.2, 71 FR 19450 (Apr. 
14, 2006).
    \77\ The proposed definition for wheelbase is also the same as 
that used in NHTSA's April 2006 light truck CAFE rule. Wheelbase is 
``the longitudinal distance between front and rear wheel 
centerlines.'' Id.
---------------------------------------------------------------------------

    First, NHTSA tentatively concludes that use of the footprint-
attribute helps us achieve greater fuel economy/emissions reductions 
without having a potentially negative impact on safety. While past 
analytic work \78\ focused on the relationship between vehicle weight 
and safety, weight was understood to encompass a constellation of size-
related factors, not just weight. More recent studies \79\ have begun 
to consider whether the relationship between vehicle size and safety 
differs. To the extent that reduction of mass has historically been 
associated with reductions in many other size attributes, and given the 
construct of the current fleet, we believe that the relationship 
between size or weight (on the one hand) and safety (on the other) has 
been similar.
---------------------------------------------------------------------------

    \78\ See Kahane, Charles J., PhD, DOT HS 809 662, ``Vehicle 
Weight, Fatality Risk and Crash Compatibility of Model Year 1991-99 
Passenger Cars and Light Trucks,'' October 2003. Available at http:/
/www.nhtsa.dot.gov/cars/rules/regrev/Evaluate/809662.html (last 
accessed April 20, 2008). See also Van Auken, R.M. and J.W. Zellner, 
``An Assessment of the Effects of Vehicle Weight on Fatality Risk in 
Model Year 1985-98 Passenger Cars and 1985-97 Light Trucks,'' 
Dynamic Research, Inc., February 2002. Available at Docket No. 
NHTSA-2003-16318-2.
    \79\ See Van Auken, R.M. and J.W. Zellner, Supplemental Results 
on the Independent Effects of Curb Weight, Wheelbase, and Track on 
Fatality Risk in 1985-1997 Model Year LTVs, Dynamic Research, Inc., 
May 2005. Available at Docket No. NHTSA-2003-16318-17.
---------------------------------------------------------------------------

    Overall, use of vehicle footprint is ``weight-neutral'' and thus 
does not exacerbate the vehicle compatibility safety problem.\80\ A 
footprint-based system does not encourage manufacturers to add weight 
to move vehicles to a higher footprint category, because additional 
weight makes no difference to the required target. Nor would the system 
penalize manufacturers for making limited weight reductions. By using 
vehicle footprint in lieu of a weight-based metric, the standards would 
also facilitate the use of promising lightweight materials that, 
although perhaps not cost-effective in mass production today, may 
ultimately achieve wider use in the fleet, become less expensive, and 
enhance emissions reductions, vehicle safety, and fuel economy.\81\
---------------------------------------------------------------------------

    \80\ The vehicle compatibility safety problem refers to the 
disparity in effects experienced by smaller lighter vehicles in 
crashes with larger heavier vehicles.
    \81\ For example, the Aluminum Association indicated in the 
April 2006 light truck CAFE rulemaking that using aluminum to 
decrease a vehicle's weight by 10 percent could improve its fuel 
economy (and thus, reduce its CO2 emissions) by 5-8 
percent, without reducing performance in frontal barrier crash 
tests. See comments provided by the Aluminum Association, Inc., at 
Docket No. NHTSA-2003-16128-1120, pp. 5 and 12.
---------------------------------------------------------------------------

    Finally, vehicle footprint is more difficult to modify than other 
attributes. It is more integral to a vehicle's design than either 
vehicle weight or shadow, and cannot easily be altered between model 
years in order to move a vehicle into a different category with a lower 
fuel economy target. Footprint is dictated by the vehicle platform, 
which is typically used for a multi-year model lifecycle. Short-term 
changes to a vehicle's platform would be expensive and difficult to 
accomplish without disrupting multi-year product planning. In some 
cases, several models share a common platform, thus adding to the cost, 
difficulty, and therefore unlikelihood of short-term changes.
    Concurrent with the NPRM, NHTSA will develop a test procedure for 
measuring wheelbase and track width and for calculating footprint. This 
test procedure will be available on NHTSA's Web site. We note that the 
test procedure will be used to validate the corresponding wheelbase, 
track width, and footprint data provided to us by the manufacturers in 
their pre-model year reports but could include other CAFE-related 
enforcement activities in the future. We seek comment on the test 
procedure.
2. Functions Based on Other Attributes
    Although NHTSA has concluded that footprint is the best attribute 
for CAFE standards, we considered a number of other attributes on which 
to base the standards, including, but not limited to, curb weight, 
engine displacement, interior volume, passenger capacity, towing 
capability, and cargo hauling capability. Below we have described the 
relative merits and drawbacks of the other attributes considered.
    Curb weight: One of the benefits of choosing curb weight as the 
relevant attribute for the standards is that it correlates with fuel 
economy and emissions controls better than vehicle footprint. 
Additionally, because reductions in weight would lead to higher 
targets, weight-based standards prevent the systemic downweighting of 
vehicles and the associated detriment to safety. However, weight-based 
standards also discourage the down-weighting of vehicles through the 
use of lightweight materials that could improve fuel economy and safety 
and reduce emissions. Weight-based standards are also more susceptible 
to gaming and creep, because weight can be altered very easily compared 
to other attributes. Weight is also only rarely considered by 
consumers, in contrast to size (which is reflected in footprint and 
shadow), and can be raised considerably (thus decreasing fuel economy/
increasing CO2 emissions) without consumers being aware of 
the change.
    Engine displacement: The primary benefit of choosing engine 
displacement as the relevant attribute for the standards is that it 
correlates well with fuel economy, since a larger engine consumes fuel 
at a faster rate. However, engine-displacement-based standards would be 
highly susceptible to gaming and creep, given that many vehicle 
manufacturers already offer identical models with different size 
engines. Additionally, engine-displacement-based standards would 
discourage the use of small turbo-charged engines, which have the 
potential to improve fuel economy without sacrificing the engine power 
that American consumers generally seek.
    Interior volume: Standards based on interior volume would have 
virtually no correlation with fuel economy, so they were not 
extensively considered. Such standards would have the advantage of not 
encouraging downsizing, so they could have a positive impact on safety 
in that respect, but few other benefits were discerned.
    Passenger capacity: Besides having virtually no correlation with 
fuel economy, passenger capacity has the disadvantage of being 
identical for a substantial portion of the light-duty vehicle 
population (i.e., many vehicles have five seats). Thus, using passenger 
capacity as the attribute on which to base fuel economy standards would 
essentially result in a single industry-wide average standard, which is 
precisely what Congress sought to avoid in requiring attribute-based 
standards.
    Towing or cargo-hauling capability: In its light truck rulemaking 
for MYs 2008-2011, NHTSA sought comment on whether towing or cargo-
hauling capability should be used as an attribute in addition to 
footprint--in other words, whether the footprint attribute should be 
modified in any way due to towing or cargo-hauling capability. The 
reason that NHTSA sought comment was that two vehicles with equal 
footprint would nevertheless achieve different fuel economies if one's 
towing or cargo-hauling capability was greater, because engineering a 
vehicle to provide that kind of power occurs at the expense of 
engineering for fuel economy. NHTSA posited that perhaps for vehicle 
manufacturers that have a product mix weighted toward vehicles with 
superior towing and/or cargo-hauling capabilities, a footprint-based 
Reformed CAFE standard might not provide a

[[Page 24390]]

fully equitable competitive environment. Based on comments to the final 
rule for the MY 2008-2011 light truck rulemaking, however, NHTSA 
concluded that the lack of an objective measure for tow rating and the 
potential for gaming of a system based on this attribute made towing or 
cargo-hauling capacity an inappropriate attribute at that time. NHTSA 
tentatively concludes that such is still the case.
    In summary, then, NHTSA has tentatively decided that a footprint-
based system will be optimal for this rulemaking. However, we seek 
comment on whether the proposed standards should be based on vehicle 
footprint alone, or whether other attributes such as the ones described 
above should be considered. If any commenters advocate one or more 
additional attributes, the agency requests those commenters to supply a 
specific, objective measure for each attribute that is accepted within 
the industry and that can be applied to the full range of light-duty 
vehicles covered by this rulemaking.

C. The Continuous Function

    NHTSA considered this issue of how to set attribute-based functions 
in its 2006 light truck CAFE rulemaking, and examined the relative 
merits of both step functions and continuous functions. In the CAFE 
context, a step function would separate the vehicle models along the 
spectrum of attribute magnitudes into discrete groups, and each group 
would be assigned a fuel economy target (that end up looking like 
steps), so that the average of the groups would be the average fleet 
fuel economy. A continuous function, in contrast, would not separate 
the vehicles into a set of discrete categories. Each vehicle model 
produced by a manufacturer would have its own fuel economy target, 
based on its particular footprint. In other words, a continuous 
function is a mathematical function that defines attribute-based 
targets across the entire range of possible footprint values, and 
applies them through a harmonically weighted formula to derive 
regulatory obligations for fleet averages.
    In proposing the current standards in this rulemaking, NHTSA relied 
on its experience in the last light truck rulemaking. In that 
rulemaking, NHTSA decided in favor of the continuous function for three 
main reasons.
     First, under a step function, manufacturers who build 
vehicle models whose footprints fall near the upper boundary of a step 
have a considerable incentive to upsize the vehicle in order to receive 
the lower target of the next step. A continuous function reduces the 
incentive created by a step function to upsize a vehicle whose 
footprint is near a category boundary, because on an uninterrupted 
spectrum, upsizing slightly can never cause a drastic decrease in the 
stringency of the applicable target.
     Second, the continuous function minimizes the incentive to 
downsize a vehicle as a way to meet the standards, because any 
downsizing results in higher targets being applicable.
     And finally, the continuous function provides 
manufacturers with greater regulatory certainty, because there are no 
category boundaries that could be redefined in future rulemaking.
    The considerations in favor of NHTSA's decision to base the MY 
2008-11 light truck CAFE standards on a continuous function are also 
applicable to the current rulemaking, which would set footprint-based 
fuel economy standards for both light trucks and passenger cars. Thus, 
NHTSA has tentatively decided that a continuous function is the best 
choice for applying the footprint-based standards.
    We note, however, that there are a variety of mathematical forms 
available to estimate the relationship between vehicle footprint and 
fuel economy that could be used as a continuous function. In the MY 
2008-11 light truck CAFE rule, NHTSA considered a simple linear 
(straight-line) function, a quadratic (U-shaped) function, an 
exponential (curve that continuously becomes steeper or shallower) 
function, and an unconstrained logistic (S-shaped) function. Each of 
these relationships was estimated in gallons per mile (gpm) rather than 
in miles per gallon (mpg), because the relationship between fuel 
economy measured in mpg and fuel savings is not linear.\82\ NHTSA 
plotted the optimized fleets in terms of footprint versus gpm, and once 
a shape of a function was determined in terms of gpm, the agency then 
converted the functions to mpg for the purpose of evaluating the 
potential target values. See 71 FR 17600-17607 (Apr. 6, 2006) for a 
fuller discussion of the agency's process.
---------------------------------------------------------------------------

    \82\ That is to say, an increase of one mpg in a vehicle with 
low fuel economy (e.g., 20 mpg to 21 mpg) results in higher fuel 
savings than if the change occurs in a vehicle with high fuel 
economy (e.g., 30 mpg to 31 mpg). Increasing fuel economy by equal 
increments of gallons per mile provides equal fuel savings 
regardless of the fuel economy of a vehicle. For example, increasing 
the fuel economy of a vehicle from 0.06 gpm to 0.05 gpm saves 
exactly the same amount of fuel as increasing the fuel economy of a 
vehicle from 0.03 gpm to 0.02 gpm.
---------------------------------------------------------------------------

    Ultimately, NHTSA decided in the light truck CAFE rule that none of 
those four functional forms as presented would be appropriate for the 
CAFE program because they tended toward excessively high stringency 
levels at the smaller end of the footprint range, excessively low 
stringency levels at the larger end of the footprint range, or both. 
Too high stringency levels for smaller vehicles could potentially 
result in target values beyond the technological capabilities of 
manufacturers, while too low levels for larger vehicles would reduce 
fuel savings below that of the optimized fleet. NHTSA determined that a 
constrained logistic function \83\ provided a relatively good fit to 
the data points without creating problems associated with some or all 
of the other forms, i.e., excessively high targets for small vehicles, 
excessively low targets for large vehicles, or regions in which targets 
for large vehicles exceeded those for small vehicles. The constrained 
logistic function also limited the potential for the curve to be 
disproportionately influenced by a single vehicle model located at 
either end of the range (i.e., by outliers). Because most vehicle 
models are clustered in the middle of the footprint range, models 
toward either end have a greater influence on their target value, and 
thus on the overall shape of the curve that fits the data points. The 
constrained logistic function minimizes this problem.
---------------------------------------------------------------------------

    \83\ A ``constrained'' logistic function is still S-shaped, like 
an unconstrained logistic function, but plateaus at the top and 
bottom rather than continuing to increase or decrease to infinity.
---------------------------------------------------------------------------

    NHTSA's constrained logistic function in the light truck rule was 
defined by four parameters. Two parameters established the function's 
upper and lower bounds (asymptotes), respectively. A third parameter 
specified the footprint at which the function was halfway between the 
upper and lower bounds. The last parameter established the rate or 
``steepness'' of the function's transition between the upper (at low 
footprint) and lower (at high footprint) boundaries.\84\

[[Page 24391]]

The resulting curve was an elongated reverse ``S'' shape, with fuel 
economy targets decreasing as footprint increased.
---------------------------------------------------------------------------

    \84\ NHTSA determined the values of the parameters establishing 
the upper and lower asymptotes by calculating the sales-weighted 
harmonic average values of optimized fuel economy levels for light 
trucks with footprints below 43 square feet and above 65 square 
feet, respectively. Because these ranges respectively included the 
smallest and largest models represented at that time in the light 
truck fleet, the agency determined that these two segments of the 
light truck fleet were appropriate for establishing the upper and 
lower fuel economy bounds of a continuous function.
    The remaining two parameters (i.e., the ``midpoint'' and 
``curvature'' parameters) were estimated using production-weighted 
nonlinear least-squares regression to achieve the closest fit to 
data on footprint and optimized fuel economy for all light truck 
models expected to be produced during each of the model years 2008-
2011. More precisely, these two parameters determine the range 
between the vehicle footprints where the upper and lower limits of 
fuel economy are reached, and the value of footprint for which the 
value of fuel economy is midway between its upper and lower bounds.
---------------------------------------------------------------------------

    NHTSA has tentatively concluded that a constrained logistic 
function would continue to be appropriate for setting CAFE standards 
for both passenger cars and light trucks. We have reached that 
conclusion because the concerns that prevented NHTSA from choosing 
another mathematical function in the light truck CAFE rule continue to 
be relevant to the new standards. The description below of the Volpe 
model and how it works explains in much more detail how the constrained 
logistic function has been updated for purposes of this rulemaking. 
NHTSA seeks comment on whether another mathematical function might 
result in improved standards consistent with EPCA and EISA.

V. Volpe Model/Analysis/Generic Description of Function

A. The Volpe model

1. What is the Volpe model?
    As it did for the development and analysis of the April 2006 light 
truck final rule, in developing this proposal NHTSA made significant 
use of a peer-reviewed modeling system developed by the Department of 
Transportation's Volpe National Transportation Systems Center (Volpe 
Center). The CAFE Compliance and Effects Modeling System (referred to 
herein as the Volpe model) serves two fundamental purposes: Identifying 
technologies each manufacturer could apply in order to comply with a 
specified set of CAFE standards, and calculating the costs and effects 
of manufacturers' application of technologies.
    Before working with the Volpe Center to develop and apply this 
model, NHTSA had considered other options, including other modeling 
systems. NHTSA was unable to identify any other system that could 
operate at a sufficient level of detail with respect to manufacturers' 
future products, which involve thousands of unique vehicle models using 
hundreds of unique engines and hundreds of unique transmissions. NHTSA 
was also unable to identify any other system that could simulate a 
range of different possible reforms to CAFE standards. The Volpe model 
provides these and other capabilities, and helps NHTSA examine 
potential regulatory options.
2. How does the Volpe model apply technologies to manufacturers' future 
fleets?
    The Volpe model begins with an ``initial state'' of the domestic 
vehicle market, which in this case is the market for passenger cars and 
light trucks to be sold during the period covered by the proposed rule. 
The vehicle market is defined on a model-by-model, engine-by-engine, 
and transmission-by-transmission basis, such that each defined vehicle 
model refers to a separately-defined engine and a separately-defined 
transmission.
    For the model years covered by the current proposal, the light 
vehicle (passenger car and light truck) market forecast included more 
than 3,000 vehicle models, more than 400 specific engines, and nearly 
400 specific transmissions.\85\ This level of detail in the 
representation of the vehicle market is vital to an accurate analysis 
of manufacturer-specific costs and the analysis of reformed CAFE 
standards, and is much greater than the level of detail used by many 
other models and analyses relevant to light vehicle fuel economy. 
Because CAFE standards apply to the average performance of each 
manufacturer's fleets of cars and light trucks, the impact of potential 
standards on individual manufacturers cannot be credibly estimated 
without analysis of manufacturers' planned fleets. NHTSA has used this 
level of detail in CAFE analysis throughout the history of the program. 
Furthermore, because required CAFE levels under an attribute-based CAFE 
standard depend on manufacturers' fleet composition, the stringency of 
an attribute-based standard cannot be predicted without performing 
analysis at this level of detail.
---------------------------------------------------------------------------

    \85\ The market forecast is an input to the Volpe model 
developed by NHTSA using product plan information provided to the 
agency by individual vehicle manufacturers in response to NHTSA's 
requests. The submitted product plans contain confidential business 
information (CBI), which the agency is prohibited by federal law 
from disclosing. As the agency receives new product plan information 
in response to future requests, the market forecast is updated.
---------------------------------------------------------------------------

    Examples of other models and analyses that NHTSA and Volpe Center 
staff have considered include DOE's NEMS, Oak Ridge National 
Laboratory's (ORNL) Transitional Alternative Fuels and Vehicles (TAFV) 
model, and the California Air Resources Board's (CARB) analysis 
supporting California's adopted greenhouse gas emissions standards for 
light vehicles.
    DOE's NEMS represents the light-duty fleet in terms of four 
``manufacturers'' (domestic cars, imported cars, domestic light trucks, 
and imported light trucks), twelve vehicle market classes (e.g., 
``standard pickup''), and sixteen power train/fuel combinations (e.g., 
methanol fuel-cell vehicle).\86\ Therefore, as currently structured, 
NEMS is unable to estimate manufacturer-specific implications of 
attribute-based CAFE standards.
---------------------------------------------------------------------------

    \86\ U.S. Department of Energy, ``Transportation Sector Module 
of the National Energy Modeling System: Model Documentation 2007,'' 
DOE/EIA-M070, May 2007. Available at http://tonto.eia.doe.gov/
FTPROOT/modeldoc/m070(2007).pdf (last accessed April 20, 2008). 
NEMS's Manufacturers Technology Choice Submodule (MTCS) is believed 
to have logical structures similar to those in Energy and 
Environmental Analysis, Inc.'s (EEA's) Fuel Economy Regulatory 
Analysis Model (FERAM). However, FERAM documentation and source code 
have not been made available to NHTSA or Volpe Center staff.
---------------------------------------------------------------------------

    TAFV accounts for many power train/fuel combinations, having been 
originally designed to aid understanding of possible transitions to 
alternative fueled vehicles, but it represents the light-duty fleet as 
four aggregated (i.e., industry-wide) categories of vehicles: Small 
cars, large cars, small light trucks, and large light trucks.\87\ Thus, 
again, as currently structured, TAFV is unable to estimate 
manufacturer-specific implications of attribute-based CAFE standards.
---------------------------------------------------------------------------

    \87\ Greene, David. ``TAFV Alternative Fuels and Vehicles Choice 
Model Documentation,'' ORNL//TM-2001//134, July 2001. Available at 
http://www-cta.ornl.gov/cta/Publications/Reports/ORNL--TM--2001--
134.pdf (last accessed April 20, 2008).
---------------------------------------------------------------------------

    CARB's analysis of light vehicle GHG emissions standards uses two 
levels of accounting. First, based on a report prepared for Northeast 
States Center for a Clean Air Future (NESCCAF), CARB represents the 
light-duty fleet in terms of five ``representative'' vehicles. Use of 
these ``representative'' vehicles ignores the fact that the engineering 
characteristics of individual vehicle models vary widely both among 
manufacturers and within manufacturers' individual fleets. For each of 
these five vehicles, NESCCAF's report contains the results of full 
vehicle simulation given several pre-specified technology 
``packages.''\88\ Second, to evaluate manufacturer-specific regulatory 
costs, CARB essentially reduces each manufacturer's fleet to only two 
average test weights, one for each of California's two regulatory

[[Page 24392]]

classes.\89\ Even for a flat standard such as considered by California, 
NHTSA would not base its analysis of manufacturer-level costs on this 
level of aggregation. Use of CARB's methods would not enable NHTSA to 
estimate manufacturer-specific implications of the attribute-based CAFE 
standards proposed today.\90\
---------------------------------------------------------------------------

    \88\ Northeast States Center for a Clean Air Future (NESCCAF), 
Reducing Greenhouse Gases from Light-Duty Vehicles (2004). Available 
at http://bronze.nescaum.org/committees/mobile/
rpt040923ghglightduty.pdf (last accessed April 20, 2008).
    \89\ California Environmental Protection Agency, Air Resources 
Board, Staff Report: Initial Statement of Reasons (CARB ISOR) 
(2004), at 111-114. Available at http://www.arb.ca.gov/regact/
grnhsgas/isor.pdf (last accessed April 20, 2008). We note that 
California has adopted these standards but is currently unable to 
enforce them, due to EPA's February 29, 2008, denial of California's 
request for waiver of federal preemption under Section 209 of the 
Clean Air Act. For information on EPA's decision, see http://
www.epa.gov/otaq/ca-waiver.htm. (Last accessed April 20, 2008.) 
California filed a petition in the Ninth Circuit Court of Appeals 
challenging EPA's denial of the waiver on January 2, 2008.
    \90\ Although CARB's analysis covered a wider range of model 
years than does NHTSA's analysis, this does not lessen the 
importance of a detailed representation of manufacturers' fleets.
---------------------------------------------------------------------------

    The Volpe model also uses several additional categories of data and 
estimates provided in various external input files:
    One input file specifies the characteristics of fuel-saving 
technologies to be represented, and includes, for each technology, the 
first year in which the technology is expected to be ready for 
commercial application; upper and lower estimates of the effectiveness 
and cost (retail price equivalent) of the technology; coefficients 
defining the extent to which costs are expected to decline as a result 
of ``learning effects'' (discussed below); inclusion or exclusion of 
the technology on up to three technology ``paths''; and constraints 
(``phase-in caps'') on the annual rate at which manufacturers are 
estimated to be able to increase the technology's penetration rate. 
These technology characteristics and estimates are specified separately 
for each of the following categories of vehicles: Small sport/utility 
vehicles (SUVs), midsize SUVs, large SUVs, small pickups, large 
pickups, minivans, subcompact cars, compact cars, midsize cars, and 
large cars. In addition, the input file defining technology 
characteristics can (but need not) contain specified ``synergies'' 
between technologies--that is, differences in a given technology's 
effect on fuel consumption that result from the presence of other 
technologies.
    Another input file specifies vehicular emission rates for the 
following pollutants: Carbon monoxide (CO), volatile organic compounds 
(VOCs), nitrogen oxides (NOX), particulate matter (PM), and 
sulfur dioxide (SO2). These rates are defined on a model 
year-by-model year and calendar year-by-calendar year basis, and are 
used to estimate changes in emissions that result from changes in 
vehicular travel (i.e., vehicle-miles traveled or VMT).
    A third input file specifies a variety of economic and other data 
and estimates. The model can accommodate vehicle survival (i.e., 
percent of vehicles of a given vintage that remain in service) and 
mileage accumulation (i.e., annual travel by vehicles of a given 
vintage) rates extending as many years beyond the year of sale as for 
which estimates are available and use those for estimating VMT, fuel 
consumption, and emissions. The model can also accommodate forecasts of 
price and fuel taxation rates for up to seven fuels (e.g., gasoline, 
diesel) over a similar period. The model uses pump prices (i.e., 
including taxes) to estimate the value manufacturers expect vehicle 
purchasers to place on saved fuel, because they indicate the amount by 
which the manufacturer is expected to consider itself able to increase 
the retail price of the vehicle based on the purchaser's consideration 
of the vehicle's increased fuel economy. However, the model uses pretax 
fuel prices to estimate the monetized societal benefits of reduced fuel 
consumption, because fuel taxes represent transfers of resources from 
fuel buyers to government agencies rather than real resources that are 
consumed in the process of supplying or using fuel, so their value must 
be deducted from retail fuel prices to determine the value of fuel 
savings to the U.S. economy.
    Other economic inputs include the rebound effect coefficient (i.e., 
the elasticity of VMT with respect to the per-mile cost of fuel); the 
discount rate; the ``payback period'' (i.e., the number of years 
manufacturers are estimated to assume vehicle purchasers consider when 
taking into account fuel savings); the ``gap'' between laboratory and 
actual fuel economy; the per-vehicle value of travel time (in dollars 
per hour); the economic costs (in dollars per gallon) of petroleum 
consumption; various external costs (all in dollars per mile) 
associated with changes in vehicle use; damage costs (all on a dollar 
per ton basis) for each of the above-mentioned criteria pollutants; and 
the rate at which noncompliance causes civil penalties. Section V below 
describes in much more detail how these inputs are included and used by 
the model.
    The model also accommodates input data and estimates addressing the 
properties of different fuels. These include upstream carbon dioxide 
and criteria pollutant emission rates (i.e., U.S. emissions resulting 
from the production and distribution of each fuel), density (pounds/
gallon), energy density (BTU/gallon), carbon content, shares of fuel 
savings leading to reduced domestic refining, and relative shares of 
different gasoline blends. These fuel properties and related estimates 
are used to calculate changes in domestic upstream emissions resulting 
from changes in fuel consumption.
    Coefficients defining the probability distributions to apply when 
performing sensitivity analysis (i.e., Monte Carlo simulation) are also 
specified in this input file.\91\ These coefficients determine the 
likelihood that any given value will be selected when performing this 
type of analysis (e.g., the likelihood that a rebound effect of -0.1 
will be tested). High and low fuel price forecasts are also specified 
in this input file for this purpose.
---------------------------------------------------------------------------

    \91\ The sensitivity analysis and its usefulness are explained 
more fully below.
---------------------------------------------------------------------------

    The final input file contains CAFE scenarios to be examined. The 
model accommodates a baseline (i.e., business-as-usual) scenario and 
different alternative scenarios. Effects of the alternative scenarios 
are calculated relative to results for the baseline scenario. Each 
scenario defines the coverage, structure, and stringency of CAFE 
standards for each of the covered model years.
    With all of the above input data and estimates, the modeling system 
develops an estimate of a set of technologies each manufacturer could 
apply in response to each specified CAFE scenario. Because 
manufacturers have many choices regarding how to respond to CAFE 
standards, it is impossible to predict precisely how a given 
manufacturer would respond to a given set of standards. The modeling 
system begins with the ``initial state'' (i.e., business-as-usual) of 
each manufacturer's future vehicles, and accumulates the estimated 
costs of progressive additions of fuel-saving technologies. Within a 
set of specified constraints, the system adds technologies following a 
cost-minimizing approach, because this is what NHTSA expects a 
manufacturer would do in real life. At each step, the system evaluates 
the effective cost of applying available technologies to individual 
vehicle models, engines, or transmissions, and selects the application 
of technology that produces the lowest effective cost. The effective 
cost estimated to be considered by the manufacturer is calculated by 
adding the total incurred technology costs (in retail price equivalent 
or RPE), subtracting the reduction in civil

[[Page 24393]]

penalties owed for noncompliance with the CAFE standard, subtracting 
the estimated value \92\ of the reduction in fuel costs, and dividing 
the result by the number of affected vehicles.
---------------------------------------------------------------------------

    \92\ The estimated value of the reduction in fuel costs 
represents the amount by which the manufacturer is expected to 
consider itself able to increase the retail price of the vehicle 
based on the purchaser's consideration of the vehicle's increased 
fuel economy. This calculation considers the change in the 
discounted outlays for fuel (and fuel taxes) during a ``payback 
period'' specified as an input to the model.
---------------------------------------------------------------------------

    In representing manufacturer decision-making in response to a given 
CAFE standard, the modeling system accounts for the fact that 
historically some manufacturers have been unwilling to pay penalties 
and some have been willing to do so. Thus, the system applies 
technologies until any of the following conditions are met: the 
manufacturer no longer owes civil penalties for failing to meet the 
applicable standard, the manufacturer has exhausted technologies 
expected to be available in that model year, or the manufacturer is 
estimated to be willing to pay civil penalties, and doing so is 
estimated to be less expensive than continuing to add technologies. The 
system then progresses to the next model year (if included in the 
vehicle market and scenario input files), ``carrying over'' 
technologies where vehicle models are projected to be succeeded by 
other vehicle models.\93\
---------------------------------------------------------------------------

    \93\ For example, if Honda is expected to produce the Civic in 
2012 and 2013, a version of the Civic estimated to be produced in 
2013 may carry over technologies from a version of the Civic 
produced in 2012 if the latter is identified as a ``predecessor'' of 
the former.
---------------------------------------------------------------------------

    In the modeling system, this ``compliance simulation'' is 
constrained in several ways. First, technologies are defined as being 
applicable or not applicable to each of the ten vehicle categories 
listed above. The vehicle market forecast input file may also define 
some technologies as being already present or not applicable to 
specific vehicles, engines or transmissions. For example, a 
manufacturer may have indicated it plans to use low-drag brakes on some 
specific vehicle model, or NHTSA may expect that another manufacturer 
is not likely to apply a 7- or 8-speed transmission after it installs a 
6-speed transmission on a vehicle. Second, some technologies are 
subject to specific ``engineering constraints.'' For example, 
secondary-axle disconnect can only be applied to vehicles with four-
wheel (or all-wheel) drive. Third, some technologies (e.g., conversion 
from pushrod valve actuation to overhead cam actuation) are nearly 
always applied only when the vehicle is expected to be redesigned and 
others (e.g., cylinder deactivation) are applied only when the vehicle 
is expected to be refreshed or redesigned, so the model will only apply 
them at those particular points. Fourth, once the system applies a 
given technology to a percentage of a given manufacturers' fleet 
exceeding a specified phase-in cap, the system instead applies other 
technologies. The third and fourth of these constraints are intended to 
produce results consistent with manufacturers' product planning 
practices and with limitations on how quickly technologies can 
penetrate the fleet.
    One important aspect of this compliance simulation is that it does 
not attempt to account for either CAFE credits or intentional over-
compliance. In the real world, manufacturers may earn CAFE credits by 
selling flex-fueled vehicles (FFVs) and/or by exceeding CAFE standards, 
and may, within limitations, count those credits toward compliance in 
future or prior model years. However, EPCA and EISA do not allow NHTSA 
to consider these flexibilities in setting the standards. Therefore, 
the Volpe model does not attempt to account for these flexibilities.
    Another possibility NHTSA and Volpe Center staff have considered, 
but do not yet know how to analyze, is the potential that manufacturers 
might ``pull ahead'' the implementation of some technologies in 
response to CAFE standards that they know will be steadily increasing 
over time. For example, if a manufacturer plans to redesign many 
vehicles in MY2011 and not in MY2013, but the standard for MY2013 is 
considerably higher than that for MY2011, the manufacturer might find 
it less expensive during MY2011-MY2013 (taken together) to apply more 
technology in MY2011 than is necessary for compliance with the MY2011 
standard. Under some circumstances, doing so might make sense even 
without regard to the potential to earn and bank CAFE credits.
    NHTSA and Volpe Center staff have discussed the potential to 
represent this type of response, but have thus far encountered two 
challenges. First, NHTSA is not certain that in determining the maximum 
feasible standard in a given model year, it would be appropriate to 
count on manufacturers overcomplying with standards in preceding model 
years. Second, considering other inter-model year dependencies (e.g., 
technologies that carry over between model years, phase-in caps that 
accumulate across model years, volume-based learning curves), Volpe 
Center staff currently anticipate that some iterative procedure would 
likely be necessary. Also, the agency wonders whether trying to 
represent this type of response would require make undue implicit 
assumptions regarding manufacturers' ability to predict future market 
conditions. Although NHTSA and Volpe Center staff will continue to 
explore the potential to represent inter-model year timing, it is not 
yet clear that it will be appropriate and feasible to do so in the near 
term.
    The agency requests comment on the appropriateness under EPCA of 
considering (in the standard-setting context) this type of anticipatory 
application of technology. The agency further requests comment on 
appropriate methodologies for projecting and representing such 
decisions by manufacturers.
3. What effects does the Volpe model estimate?
    Having completed this compliance simulation for all manufacturers 
and all model years, the system calculates the total cost of all 
applied technologies, as well as a variety of effects of changes in 
fuel economy. The system calculates year-by-year mileage accumulation, 
taking into account any increased driving estimated to result from the 
rebound effect. Based on the calculated mileage accumulation and on 
fuel economy and the estimated gap between laboratory and actual fuel 
economy, the system calculates year-by-year fuel consumption. Based on 
calculated mileage accumulation and fuel consumption, and on specified 
emission factors, the system calculates future full fuel-cycle domestic 
carbon dioxide and criteria pollutant emissions. The system calculates 
total discounted and undiscounted national societal costs of year-by-
year fuel consumption, taking into account estimated future fuel prices 
(before taxes) and the estimated economic externalities of fuel 
consumption. Based on changes in year-by-year mileage accumulation, the 
system calculates changes in consumer surplus related to additional 
travel, as well as economic externalities related to additional 
congestion, accidents, and noise stemming from additional travel. The 
system calculates the value of time saved because increases in fuel 
economy produce increases in driving range, thereby reducing the 
frequency with which some vehicles require refueling. The system 
calculates the monetary value of damages resulting from criteria 
pollutants. Finally, the system accumulates all discounted and 
undiscounted societal benefits of each scenario as compared to the 
baseline

[[Page 24394]]

scenario. For each model year, the system compares total incurred 
technology costs to the total present value of societal benefits for 
each model year, calculating net societal benefits (i.e., discounted 
societal benefits minus total incurred technology costs) and the 
benefit-cost ratio (i.e., discounted societal benefits divided by total 
incurred technology costs).
    One effect not currently estimated by the Volpe model is the market 
response to CAFE-induced changes in vehicle prices and fuel economy 
levels. NHTSA and Volpe Center staff have worked to try and develop and 
apply a market share model capable of estimating changes in sales of 
individual vehicle models. Doing so would allow estimation of the 
feedback between market shifts and CAFE requirements. For example, if 
the relative market share of vehicles with small footprints increases, 
the average required CAFE level under a footprint-based standard will 
also increase.
    In an early experimental version of the Volpe model, Volpe Center 
staff included a market share model using a nested multinomial logit 
specification to calculate model-by-model changes in sales volumes. 
This allowed the Volpe model to calculate the resulting changes in 
manufacturers' required CAFE levels, and to seek iteratively a solution 
at which prices, fuel economy levels, sales volumes, and required CAFE 
levels converged to stable values. Although the market share model 
appeared to operate properly (and to converge rapidly), Volpe Center 
staff suspended its development because of three challenges:
    First, Volpe Center staff were not successful in calibrating a 
logically consistent set of coefficients for the underlying multinomial 
logit model. The analysis, performed using information from a known 
(2002 model year) fleet, consistently yielded one or more coefficients 
that were either directionally incorrect (e.g., indicating that some 
attributes actually detract from value) or implausibly large (e.g., 
indicating that some attributes were of overwhelming value). Although 
Volpe Center staff tested many different specifications of the market 
share model, none produced results that appeared to merit further 
consideration.
    Second, NHTSA and Volpe Center staff are not confident that 
baseline sales prices for individual vehicle models, which would be 
required by a market share model, can be reliably predicted. Although 
NHTSA requests that manufacturers include planned MSRPs in product 
plans submitted to NHTSA, MSRPs do not include the effect of various 
sales incentives that can change actual selling prices. The 
availability and dollar value of such incentives have been observed to 
vary considerably, but not necessarily predictably.
    Finally, before applying a market share model, it would be 
necessary to estimate how manufacturers would allocate compliance costs 
among vehicle models. Although one obvious approach would be to assume 
that all costs would be passed through in the form of higher prices for 
those vehicle models with improved fuel economy, other approaches are 
perhaps equally plausible. For example, a manufacturer might shift 
compliance costs toward high-demand vehicles in order to compete better 
in certain market segments. Although the above-mentioned experimental 
version of the Volpe model included a ``cost allocation'' model that 
offered several different allocation options, NHTSA and Volpe Center 
staff never achieved confidence that these aspects of manufacturer 
decisions could be reasonably estimated.
    NHTSA and Volpe Center staff are continuing to explore options for 
including these types of effects. At the same time, EPA has contracted 
with Resources for the Future (RFF) to develop a potential market share 
model. Depending on the extent to which these efforts are successful, 
the Volpe model could at some point be modified to include cost 
allocation and market share models. NHTSA seeks comments on possible 
methodologies for incorporating market responses to CAFE-induced 
changes in vehicle price and fuel economy in the Volpe model. In 
particular, NHTSA seeks comments addressing the concerns identified 
above regarding the formulation and calibration of a market share 
model, the estimation of future vehicle prices, and the estimation of 
manufacturers' decisions regarding the allocation of compliance costs.
4. How can the Volpe model be used to calibrate and evaluate potential 
CAFE standards?
    The modeling system can also be applied in a more highly-automated 
mode whereby the optimal shape of an attribute-based CAFE standard may 
be estimated and its stringency may be set at a level that produces a 
specified total technology cost or average required CAFE level among a 
specified set of manufacturers, or that is estimated to maximize net 
societal benefits. The first step in this operating mode involves 
identifying manufacturer-by-manufacturer CAFE levels at which societal 
benefits are estimated to be maximized. The second step involves 
combining the resultant fleets and statistically fitting a constrained 
logistic curve of the following form:
[GRAPHIC] [TIFF OMITTED] TP02MY08.002

    Here, TARGET is the fuel economy target (in mpg) applicable to 
vehicles of a given footprint (FOOTPRINT, in square feet), LIMITLOWER 
and LIMITUPPER are the function's lower and upper asymptotes (also in 
mpg), e is approximately equal to 2.718,\94\ MIDPOINT is the footprint 
(in square feet) at which the inverse of the fuel economy target falls 
halfway between the inverses of the lower and upper asymptotes, and 
WIDTH is a parameter (in square feet) that determines how gradually the 
fuel economy target transitions from the upper toward the lower 
asymptote as the footprint increases. Figure V-1 below shows an example 
of a logistic target function, where LIMITLOWER = 20 mpg, LIMITUPPER = 
30 mpg, MIDPOINT = 40 square feet, and WIDTH = 5 square feet:
---------------------------------------------------------------------------

    \94\ The number e is one of the most important numbers in 
mathematics and statistics. The function has a hockey stick 
appearance when plotted. The value of e itself is a never ending 
number whose first 8 digits equal 2.7182818. NHTSA uses it here 
because it occurs in many natural processes and tends to fit data 
well. In the last light truck rulemaking, NHTSA examined several 
functional forms that did not rely on e, but they were judged not to 
provide as good a fit for the data. We are using the same conclusion 
here.

---------------------------------------------------------------------------

[[Page 24395]]

[GRAPHIC] [TIFF OMITTED] TP02MY08.003

    The lower asymptote is determined by calculating the average fuel 
economy of the largest vehicles in the ``optimized'' fleet discussed 
above, where the percentage of the fleet to consider is specified 
externally. Similarly, the upper asymptote is determined by calculating 
the average fuel economy of the smallest vehicles in the same fleet. 
Initial values of the other two coefficients of the logistic function 
are determined through a standard statistical technique (nonlinear 
least-square regression), except as discussed in sections V and VI 
below regarding the adjusting of the original curve for the passenger 
car function.
    Following this initial calibration of the target function, the 
system adjusts the lower and upper asymptotes uniformly (on a gallon 
per mile basis) until one of the following externally specified 
conditions is met: the average CAFE level required of the included 
manufacturers approximately equals an externally specified goal; net 
societal benefits (i.e., total benefits minus total costs) are 
maximized, or total benefits are as close as observed (among evaluated 
stringency levels) to total costs. Due to rounding of fuel economy and 
CAFE levels, the first condition can only be satisfied on an 
approximate basis.
    The modeling system provides another type of higher-level 
automation--the ability to perform uncertainty analysis, also referred 
to as Monte Carlo simulation. For some input parameters, such as 
technology costs, values can be tested over a specified continuous 
probability distribution. For others, such as fuel prices, discrete 
scenarios (e.g., high, low, and reference cases), each with a specified 
probability, can be tested. The system performs sensitivity analysis by 
randomly selecting values for parameters to be varied, performing the 
compliance simulation and effects calculations, repeating these results 
many times and recording results for external analysis. This operating 
mode enables the examination of the uncertainty of high-level results 
(e.g., total costs, fuel savings, or net societal benefits), as well as 
their sensitivity to variations in the model's input parameters.
5. How has the Volpe model been updated since the April 2006 light 
truck CAFE final rule?
    Several changes were made to the Volpe model between the analysis 
reported in the April 2006 light truck final rule and the analysis of 
the current NPRM. As discussed above, the set of technologies 
represented was updated, the logical sequence for progressing

[[Page 24396]]

through these technologies was changed, methods to account for 
``synergies'' (i.e., interactions) between technologies and technology 
cost reductions associated with a manufacturer's ``learning'' were 
added, the effective cost calculation used in the technology 
application algorithm was modified, and the procedure for calibrating a 
reformed standard was changed, as was the procedure for estimating the 
optimal stringency of a reformed standard.
    As discussed in Section III above, the set of technologies 
considered by the agency has evolved since the previous light truck 
CAFE rulemaking. The set of technologies now included in the Volpe 
model is shown below in Table V-1, with codes used by the model to 
refer to each technology.

           Table V-1.--Revised Technology Set for Volpe Model
------------------------------------------------------------------------
                 Technology                        Code  (for Model)
------------------------------------------------------------------------
Low Friction Lubricants.....................  LUB
Engine Friction Reduction...................  EFR
Variable Valve Timing (Intake Cam Phasing)..  VVTI
Variable Valve Timing (Coupled Cam Phasing).  VVTC
Variable Valve Timing (Dual Cam Phasing)....  VVTD
Cylinder Deactivation.......................  DISP
Variable Valve Lift & Timing (Continuous      VVLTC
 VVL).
Variable Valve Lift & Timing (Discrete VVL).  VVLTD
Cylinder Deactivation on Overhead Valve       DISPO
 (OHV).
Variable Valve Timing (CCP) on OHV..........  VVTO
Multivalve Overhead Cam with CVVL...........  DOHC
Variable Valve Lift & Timing (DVVL) on OHV..  VVLTO
Camless Valve Actuation.....................  CVA
Stoichiometric Gasoline Direct Injection      SIDI
 (GDI).
Lean Burn GDI...............................  LBDI
Turbocharging and Downsizing................  TURB
Homogeneous Charge Compression Ignition.....  HCCI
Diesel with Lean NOX Trap (LNT).............  DSLL
Diesel with Selective Catalytic Reduction     DSLS
 (SCR).
5 Speed Automatic Transmission..............  5SP
Aggressive Shift Logic......................  ASL
Early Torque Converter Lockup...............  TORQ
6 Speed Automatic Transmission..............  6SP
Automatic Manual Transmission...............  AMT
Continuously Variable Transmission..........  CVT
6 Speed Manual..............................  6MAN
Improved Accessories........................  IACC
Electronic Power Steering...................  EPS
42-Volt Electrical System...................  42V
Low Rolling Resistance Tires................  ROLL
Low Drag Brakes.............................  LDB
Secondary Axle Disconnect--Unibody..........  SAXU
Secondary Axle Disconnect--Ladder Frame.....  SAXL
Aero Drag Reduction.........................  AERO
Material Substitution (1%)..................  MS1
Material Substitution (2%)..................  MS2
Material Substitution (5%)..................  MS5
Integrated Starter/Generator (ISG) with Idle- ISGO
 Off.
IMA/ISAD/BSG Hybrid (includes engine          IHYB
 downsizing).
2-Mode Hybrid...............................  2HYB
Power Split Hybrid..........................  PHYB
Full Diesel Hybrid..........................  DHYB
------------------------------------------------------------------------

    The logical sequence for progressing between these technologies has 
also been changed. As in the previous version of the Volpe model, 
technologies are assigned to groups (e.g., engine technologies) and the 
model follows a cost-minimizing approach to selecting technologies. 
However, the model now includes some ``branch points'' at which it 
selects from two or more technologies within the same group. This 
enables a more detailed representation of some technologies that have 
multiple variants (e.g., variable valve timing) and, as relevant to the 
applicability of different technologies, more specific differentiation 
between technologies that have already been applied to vehicles (e.g., 
single versus dual overhead cam engines). This revised logical 
sequencing is expected to produce results that are more realistic in 
terms of the application of technologies to different vehicle models. 
For example, in this analysis OHV engines and OHC engines were 
considered separately, and the model was generally not allowed to apply 
multivalve OHC technology to OHV engines (except where continuous 
variable valve timing and lift is applied to OHV engines, in which case 
the model assumes conversion to DOHC valvetrain).
    Figure V-2 below shows the resultant ``decision tree'' for the 
group of engine technologies. As an example of the ``branching'' 
mentioned above, having applied cylinder deactivation and coupled cam 
phasing to an overhead valve engine, the Volpe model selects either 
discrete valve lift or an engine redesign to multivalve overhead cam 
with continuous variable valve lift. Figure V-3 shows the decision tree 
for transmission technologies, and Figure V-4 shows the decision trees 
for other technologies.
BILLING CODE 4910-59-P

[[Page 24397]]

[GRAPHIC] [TIFF OMITTED] TP02MY08.004


[[Page 24398]]


[GRAPHIC] [TIFF OMITTED] TP02MY08.005


[[Page 24399]]


[GRAPHIC] [TIFF OMITTED] TP02MY08.006

    Each time the model applies a technology to a vehicle in the fleet, 
it considers the next available technology on every available path. An 
available technology is one that is not included in the base vehicle, 
has not been applied by the model, and is not disqualified due to the 
vehicle's characteristics (discussed below). For a given path, the next 
available technology is the first available item (if no technologies on 
the path have yet been applied) or the first available item following 
the most recently applied technology on that path. An available path is 
any path that includes available technologies.
    The engine and transmission paths contain several forks where the 
model may choose among two or more same-path items along with items 
from other paths. At some of these forks, conditions on the connecting 
arrows require the model to follow a particular branch. These 
conditions are based on previously applied technologies or vehicle 
characteristics. For example, ladder frame vehicles must follow the 
left branch of the transmission technology path, while unibody vehicles 
can follow either the right or left branch. The consequence is that the 
model considers both aggressive shift logic (ASL) and CVT for unibody 
vehicles, but only ASL for ladder frame vehicles. Conditions along the 
engine technologies path are based on valvetrain design (OHV, OHC, 
SOHC, and DOHC).
    Other conditions require the model to discontinue considering 
technologies along a given path. For example, 2-Mode Hybrid and Power 
Split Hybrid drivetrains can be applied only to vehicles equipped with 
automatic transmissions. If the model has already chosen a manual 
transmission and IMA/ISAD/BSG Hybrid drivetrain (or if the base vehicle 
is equipped with these), the hybrid path becomes unavailable and the 
model must choose subsequent technologies from other paths.
a. Technology Synergies
    In some cases, the change in fuel economy achieved by applying a 
given technology depends on what other technologies are already 
present. The Volpe model has been modified to provide the ability to 
represent such ``synergies'' between technologies, as discussed above. 
These effects are specified in one of the model's input files. As shown 
below in Table V-2, which uses technology codes listed in Table V-1 
above, most of the synergies represented in the analysis of this 
proposal are negative. In other words, most of the interactions are 
such that a given technology has a smaller effect on fuel economy if 
some other technologies have already been applied. The inclusion of 
such effects in the model is

[[Page 24400]]

expected to produce more realistic estimates of the benefit of applying 
various technologies.

                      Table V-2.--``Synergies'' from Technology Input File for Volpe Model
                                                  [In percent]
----------------------------------------------------------------------------------------------------------------
                   Synergies                          Synergy values by vehicle class.  Positive values are
------------------------------------------------           synergies, negative values are dissynergies.
                                                ----------------------------------------------------------------
         Technology A            Technology B                                                          Pickup-
                                                  SUV-Small     SUV-Mid     SUV-Large     Minivan       Small
----------------------------------------------------------------------------------------------------------------
VVTI.........................  5SP.............        -0.50        -0.50        -0.50        -0.50        -0.50
VVTI.........................  ISGO............        -0.50        -0.50        -0.50        -0.50        -0.50
VVTC.........................  5SP.............        -0.50        -0.50        -0.50        -0.50        -0.50
VVTC.........................  CVT.............        -0.50        -0.50        -0.50        -0.50        -0.50
VVTC.........................  ASL.............        -0.50        -0.50        -0.50        -0.50        -0.50
DISP.........................  5SP.............        -1.00        -1.00        -1.00        -1.00        -1.00
DISP.........................  CVT.............        -1.00        -1.00        -1.00        -1.00        -1.00
DISP.........................  ASL.............        -0.50        -0.50        -0.50        -0.50        -0.50
DISP.........................  ISGO............        -0.50        -0.50        -0.50        -0.50        -0.50
VVLTC........................  5SP.............        -0.50        -0.50        -0.50        -0.50        -0.50
VVLTC........................  CVT.............        -0.50        -0.50        -0.50        -0.50        -0.50
VVLTC........................  ASL.............        -0.50        -0.50        -0.50        -0.50        -0.50
VVLTC........................  6MAN............        -0.50        -0.50        -0.50        -0.50        -0.50
VVLTD........................  CVT.............        -0.50        -0.50        -0.50        -0.50        -0.50
VVLTD........................  6SP.............        -0.50        -0.50        -0.50        -0.50        -0.50
DISPO........................  5SP.............        -1.50        -1.50        -1.50        -1.50        -1.50
DISPO........................  CVT.............        -1.00        -1.00        -1.00        -1.00        -1.00
DISPO........................  ASL.............        -0.50        -0.50        -0.50        -0.50        -0.50
DISPO........................  6SP.............        -0.50        -0.50        -0.50        -0.50        -0.50
DISPO........................  ISGO............        -1.00        -1.00        -1.00        -1.00        -1.00
VVTO.........................  CVT.............        -0.50        -0.50        -0.50        -0.50        -0.50
VVTO.........................  6MAN............         0.50         0.50         0.50         0.50         0.50
DOHC.........................  5SP.............        -1.00        -1.00        -1.00        -1.00        -1.00
DOHC.........................  CVT.............        -1.00        -1.00        -1.00        -1.00        -1.00
DOHC.........................  ASL.............        -0.50        -0.50        -0.50        -0.50        -0.50
DOHC.........................  6SP.............        -0.50        -0.50        -0.50        -0.50        -0.50
DOHC.........................  6MAN............        -0.50        -0.50        -0.50        -0.50        -0.50
DOHC.........................  ISGO............         0.50         0.50         0.50         0.50         0.50
VVLTO........................  5SP.............        -0.50        -0.50        -0.50        -0.50        -0.50
VVLTO........................  CVT.............        -0.50        -0.50        -0.50        -0.50        -0.50
VVLTO........................  6SP.............        -0.50        -0.50        -0.50        -0.50        -0.50
----------------------------------------------------------------------------------------------------------------


                                                  [In percent]
----------------------------------------------------------------------------------------------------------------
                   Synergies                      Synergy values by vehicle class Positive values are synergies,
------------------------------------------------                negative values are dissynergies.
                                                ----------------------------------------------------------------
         Technology A            Technology B                                                          Pickup-
                                                  SUV-Small     SUV-Mid     SUV-Large     Minivan       Small
----------------------------------------------------------------------------------------------------------------
CVA..........................  5SP.............        -0.50        -0.50        -0.50        -0.50        -0.50
CVA..........................  CVT.............        -1.00        -1.00        -1.00        -1.00        -1.00
CVA..........................  ASL.............        -0.50        -0.50        -0.50        -0.50        -0.50
CVA..........................  6SP.............        -0.50        -0.50        -0.50        -0.50        -0.50
CVA..........................  6MAN............         0.50         0.50         0.50         0.50         0.50
HCCI.........................  CVT.............        -0.50        -0.50        -0.50        -0.50        -0.50
HCCI.........................  6SP.............        -0.50        -0.50        -0.50        -0.50        -0.50
TURB.........................  5SP.............        -1.00        -1.00        -1.00        -1.00        -1.00
TURB.........................  CVT.............        -1.00        -1.00        -1.00        -1.00        -1.00
TURB.........................  ASL.............        -0.50        -0.50        -0.50        -0.50        -0.50
TURB.........................  6SP.............        -0.50        -0.50        -0.50        -0.50        -0.50
TURB.........................  6MAN............        -0.50        -0.50        -0.50        -0.50        -0.50
E25..........................  5SP.............         0.50         0.50         0.50         0.50         0.50
E25..........................  6MAN............         0.50         0.50         0.50         0.50         0.50
E25..........................  ISGO............        -0.50        -0.50        -0.50        -0.50        -0.50
ISGO.........................  IACC............        -0.50        -0.50        -0.50        -0.50        -0.50
ISGO.........................  EPS.............        -1.00        -1.00        -1.00        -1.00        -1.00
ISGO.........................  42V.............        -1.00        -1.00        -1.00        -1.00        -1.00
DSLT.........................  5SP.............         0.50         0.50         0.50         0.50         0.50
DSLT.........................  CVT.............         0.50         0.50         0.50         0.50         0.50
DSLT.........................  ISGO............         0.50         0.50         0.50         0.50         0.50
DSLT.........................  ASL.............         0.50         0.50         0.50         0.50         0.50
DSLH.........................  5SP.............         0.50         0.50         0.50         0.50         0.50
DSLH.........................  CVT.............        -0.50        -0.50        -0.50        -0.50        -0.50
DSLH.........................  6SP.............        -0.50        -0.50        -0.50        -0.50        -0.50


[[Continued on page 24401]]


From the Federal Register Online via GPO Access [wais.access.gpo.gov]
]                         
 
[[pp. 24401-24450]] Average Fuel Economy Standards, Passenger Cars and Light Trucks; 
Model Years 2011-2015

[[Continued from page 24400]]

[[Page 24401]]


DSLH.........................  6MAN............         0.50         0.50         0.50         0.50         0.50
DSLH.........................  ISGO............         0.50         0.50         0.50         0.50         0.50
DSLS.........................  5SP.............        -0.50        -0.50        -0.50        -0.50        -0.50
DSLS.........................  CVT.............        -2.50        -2.50        -2.50        -2.50        -2.50
DSLS.........................  6SP.............        -1.00        -1.00        -1.00        -1.00        -1.00
DSLS.........................  6MAN............        -0.50        -0.50        -0.50        -0.50        -0.50
DSLS.........................  ISGO............         0.50         0.50         0.50         0.50         0.50
----------------------------------------------------------------------------------------------------------------


                                                  [In percent]
----------------------------------------------------------------------------------------------------------------
                   Synergies                          Synergy values by vehicle class.  Positive values are
------------------------------------------------           synergies, negative values are dissynergies.
                                                ----------------------------------------------------------------
         Technology A            Technology B      Pickup-
                                                    Large      Subcompact    Compact      Midsize       Large
----------------------------------------------------------------------------------------------------------------
VVTI.........................  5SP.............        -0.50        -0.50        -0.50        -0.50        -0.50
VVTI.........................  ISGO............        -0.50        -0.50        -0.50        -0.50        -0.50
VVTC.........................  5SP.............        -0.50        -0.50        -0.50        -0.50        -0.50
VVTC.........................  CVT.............        -0.50        -0.50        -0.50        -0.50        -0.50
VVTC.........................  ASL.............        -0.50        -0.50        -0.50        -0.50        -0.50
DISP.........................  5SP.............        -1.00        -1.00        -1.00        -1.00        -1.00
DISP.........................  CVT.............        -1.00        -1.00        -1.00        -1.00        -1.00
DISP.........................  ASL.............        -0.50        -0.50        -0.50        -0.50        -0.50
DISP.........................  ISGO............        -0.50        -0.50        -0.50        -0.50        -0.50
VVLTC........................  5SP.............        -0.50        -0.50        -0.50        -0.50        -0.50
VVLTC........................  CVT.............        -0.50        -0.50        -0.50        -0.50        -0.50
VVLTC........................  ASL.............        -0.50        -0.50        -0.50        -0.50        -0.50
VVLTC........................  6MAN............        -0.50        -0.50        -0.50        -0.50        -0.50
VVLTD........................  CVT.............        -0.50        -0.50        -0.50        -0.50        -0.50
VVLTD........................  6SP.............        -0.50        -0.50        -0.50        -0.50        -0.50
DISPO........................  5SP.............        -1.50        -1.50        -1.50        -1.50        -1.50
DISPO........................  CVT.............        -1.00        -1.00        -1.00        -1.00        -1.00
DISPO........................  ASL.............        -0.50        -0.50        -0.50        -0.50        -0.50
DISPO........................  6SP.............        -0.50        -0.50        -0.50        -0.50        -0.50
DISPO........................  ISGO............        -1.00        -1.00        -1.00        -1.00        -1.00
VVTO.........................  CVT.............        -0.50        -0.50        -0.50        -0.50        -0.50
VVTO.........................  6MAN............         0.50         0.50         0.50         0.50         0.50
DOHC.........................  5SP.............        -1.00        -1.00        -1.00        -1.00        -1.00
DOHC.........................  CVT.............        -1.00        -1.00        -1.00        -1.00        -1.00
DOHC.........................  ASL.............        -0.50        -0.50        -0.50        -0.50        -0.50
DOHC.........................  6SP.............        -0.50        -0.50        -0.50        -0.50        -0.50
DOHC.........................  6MAN............        -0.50        -0.50        -0.50        -0.50        -0.50
DOHC.........................  ISGO............         0.50         0.50         0.50         0.50         0.50
VVLTO........................  5SP.............        -0.50        -0.50        -0.50        -0.50        -0.50
VVLTO........................  CVT.............        -0.50        -0.50        -0.50        -0.50        -0.50
VVLTO........................  6SP.............        -0.50        -0.50        -0.50        -0.50        -0.50
----------------------------------------------------------------------------------------------------------------


                                                  [In percent]
----------------------------------------------------------------------------------------------------------------
                   Synergies                          Synergy values by vehicle class.  Positive values are
------------------------------------------------           synergies, negative values are dissynergies.
                                                ----------------------------------------------------------------
         Technology A            Technology B      Pickup-
                                                    Large      Subcompact    Compact      Midsize       Large
----------------------------------------------------------------------------------------------------------------
CVA..........................  5SP.............        -0.50        -0.50        -0.50        -0.50        -0.50
CVA..........................  CVT.............        -1.00        -1.00        -1.00        -1.00        -1.00
CVA..........................  ASL.............        -0.50        -0.50        -0.50        -0.50        -0.50
CVA..........................  6SP.............        -0.50        -0.50        -0.50        -0.50        -0.50
CVA..........................  6MAN............         0.50         0.50         0.50         0.50         0.50
HCCI.........................  CVT.............        -0.50        -0.50        -0.50        -0.50        -0.50
HCCI.........................  6SP.............        -0.50        -0.50        -0.50        -0.50        -0.50
TURB.........................  5SP.............        -1.00        -1.00        -1.00        -1.00        -1.00
TURB.........................  CVT.............        -1.00        -1.00        -1.00        -1.00        -1.00
TURB.........................  ASL.............        -0.50        -0.50        -0.50        -0.50        -0.50
TURB.........................  6SP.............        -0.50        -0.50        -0.50        -0.50        -0.50
TURB.........................  6MAN............        -0.50        -0.50        -0.50        -0.50        -0.50
E25..........................  5SP.............         0.50         0.50         0.50         0.50         0.50
E25..........................  6MAN............         0.50         0.50         0.50         0.50         0.50

[[Page 24402]]


E25..........................  ISGO............        -0.50        -0.50        -0.50        -0.50        -0.50
ISGO.........................  IACC............        -0.50        -0.50        -0.50        -0.50        -0.50
ISGO.........................  EPS.............        -1.00        -1.00        -1.00        -1.00        -1.00
ISGO.........................  42V.............        -1.00        -1.00        -1.00        -1.00        -1.00
DSLT.........................  5SP.............         0.50         0.50         0.50         0.50         0.50
DSLT.........................  CVT.............         0.50         0.50         0.50         0.50         0.50
DSLT.........................  ISGO............         0.50         0.50         0.50         0.50         0.50
DSLT.........................  ASL.............         0.50         0.00         0.00         0.50         0.50
DSLH.........................  5SP.............         0.50         0.50         0.50         0.50         0.50
DSLH.........................  CVT.............        -0.50        -0.50        -0.50        -0.50        -0.50
DSLH.........................  6SP.............        -0.50        -0.50        -0.50        -0.50        -0.50
DSLH.........................  6MAN............         0.50         0.50         0.50         0.50         0.50
DSLH.........................  ISGO............         0.50         0.50         0.50         0.50         0.50
DSLS.........................  5SP.............        -0.50        -0.50        -0.50        -0.50        -0.50
DSLS.........................  CVT.............        -2.50        -2.50        -2.50        -2.50        -2.50
DSLS.........................  6SP.............        -1.00        -1.00        -1.00        -1.00        -1.00
DSLS.........................  6MAN............        -0.50        -0.50        -0.50        -0.50        -0.50
DSLS.........................  ISGO............         0.50         0.50         0.50         0.50         0.50
----------------------------------------------------------------------------------------------------------------

b. Technology learning curves

    The Volpe model has also been modified to provide the ability to 
account for cost reductions a manufacturer may realize through learning 
achieved from experience in actually applying a given technology. Thus, 
for some of the technologies, we have included a learning factor. 
Stated another way, the ``learning curve'' describes the reduction in 
unit production costs as a function of accumulated production volume 
and small redesigns that reduce costs.
    As explained above, a typical learning curve can be described by 
three parameters: (1) The initial production volume before cost 
reductions begin to be realized; (2) the rate at which cost reductions 
occur with increases in cumulative production beyond this initial 
volume (usually referred to as the ``learning rate''); and (3) the 
production volume after which costs reach a ``floor,'' and further cost 
reductions no longer occur. Over the region where costs decline with 
accumulating production volume, an experience curve can be expressed as 
C(Q) = aQ-\b\, where a is a constant coefficient, Q 
represents cumulative production, and b is a coefficient corresponding 
to the assumed learning rate. In turn, the learning rate L, which is 
usually expressed as the percent by which average unit cost declines 
with a doubling of cumulative production, and is related to the value 
of the coefficient b by L = 100*(1 - 2-\b\).\95\
---------------------------------------------------------------------------

    \95\ See, e.g., Robert H. Williams, ``Toward Cost Buydown via 
Learning-by-Doing for Environmental Energy Technologies,'' paper 
presented at Workshop on Learning-by-Doing in Energy Technologies, 
Resources for the Future, Washington, DC, June 17-18, 2003, pp. 1-2. 
Another common but equivalent formulation of the relationship 
between L and b is (1-L) = 2\-b\, where (1-L) is referred to as the 
progress ratio; see Richard P. Rumelt, ``Note on Strategic Cost 
Dynamics,'' POL 2001-1.1, Anderson School of Business, University of 
California, Los Angeles, California, 2001, pp. 4-5.
---------------------------------------------------------------------------

    The new learning curves are described in greater detail above in 
Section III. We seek comment on the assumptions used to develop the new 
proposed learning curves.
c. Calibration of reformed CAFE standards
    The procedure used by the Volpe model to develop (i.e., calibrate) 
the initial shape of a reformed standard was also modified. In the 
version of the model used to analyze NHTSA's April 2006 light truck 
final rule, the asymptotes for the constrained logistic function 
defining fuel economy targets were assigned based on the set of 
vehicles that would have been assigned to the lowest and highest bins 
defined in that rule's 2005 NPRM. The Volpe model has been modified to 
accept specified percentages (in terms of either models or sales) of 
the fleet to include when assigning asymptotes.
    The procedure used by the Volpe model to estimate the ``optimized'' 
stringency of a reformed standard was also modified. In the version of 
the model used to analyze the 2006 light truck final rule, the shape of 
the function (i.e., the constrained logistic function) defining fuel 
economy targets was recalibrated every model year and then shifted up 
and down to estimate the stringency at which marginal costs begin to 
exceed marginal benefits or, equivalently, the point at which net 
societal benefits are maximized. However, analysis conducted by the 
agency to prepare for the current rulemaking revealed several 
opportunities to refine the procedure described above before applying 
it to an action that spans several model years. The first refinement is 
a method for gradually transforming the shape of the continuous 
function between model years and guarding against erratic fluctuations 
in the shape (though not necessarily the stringency) of the continuous 
function. The second is the implementation of several anti-backsliding 
measures that prevents the average required CAFE level from falling 
between model years and prevents the continuous function for a given 
model from crossing or falling below that of the preceding model year. 
The third, applied to passenger cars only, is an option to specify a 
fixed relationship between the function's midpoint and width 
coefficients. These refinements are discussed in greater detail in 
Section V.B below.
6. What manufacturer information does the Volpe model use?
    For purposes of determining and analyzing CAFE standards, NHTSA has 
historically made significant use of detailed product plan information 
provided to the agency by individual manufacturers, supplementing this 
information where appropriate with information from other sources, such 
as data submitted to the agency in relation to CAFE compliance. Such 
information is considered confidential business

[[Page 24403]]

information (CBI) under federal law. Although NHTSA shares the 
information with other agencies (Volpe, EPA, and DOE) involved in CAFE 
activities, neither NHTSA nor any other agency may release the 
information to the public.
    Consistent with this practice, the Volpe model uses detailed 
representations of (i.e., model-by-model, linked to specific engines 
and transmissions) the fleets manufacturers are expected to produce for 
sale in the U.S. In preparation for today's action, the agency issued 
in the spring of 2006 a request that manufacturers provide updated 
product plans for passenger cars and light trucks.
    NHTSA received product plan information from Chrysler, Ford, GM, 
Honda, Nissan, Mitsubishi, Porsche and Toyota. The agency did not 
receive any product plan information from BMW, Ferrari, Hyundai, 
Mercedes or VW.
    Chrysler, Ford, GM, Honda, Nissan, Mitsubishi, Porsche and Toyota 
provided information covering multiple model years. However, only 
Chrysler and Mitsubishi provided us with product plans that showed 
differing production quantities, vehicle introductions, vehicle 
redesigns/refreshes changes, without any carryover production 
quantities, from MY 2007 to MY 2015. The agency incorporated their 
product plan information as part of the input file to the model without 
the need to project or carryover any vehicle production data.
    For the other companies that provided data, the agency carried over 
production quantities for their vehicles, allowing for growth, starting 
with the year after their product plan data showed changes in 
production quantities or showed the introduction or redesign/refresh of 
vehicles. Product plan information was provided until MY 2013 for Ford 
and Toyota, thus the first year that we started to carry over 
production quantities for those companies was MY 2014. Product plan 
information was provided until MY 2012 for GM and Nissan, thus the 
first year that we started to carry over production quantities for 
those companies was MY 2013. Product plan information was provided by 
Honda until MY 2008. Honda asked the agency to carry over those plans 
and also provided data for the last redesign of a vehicle and asked us 
to carry them forward.
    Product plan information was provided until MY 2008 for Porsche, 
thus the first year that we started to carry over production quantities 
for Porsche was MY 2009.
    For Hyundai, given that it is one of the largest 7 manufacturers, 
the agency used the mid-year 2007 data contained in the agency's CAFE 
database to establish the baseline models and production quantities for 
their vehicles. For the other manufacturers, because of the time 
constraint the agency was under to meet the statutory deadline, we used 
the 2005 information from our database, which is the latest information 
used in the current analysis. To the extent possible, because, the CAFE 
database does not capture all of the product plan data that we request 
from companies, we supplemented the CAFE database information with 
information on public Web sites, from commercial information sources 
and for Hyundai, from the MY 2008-2011 light truck rule.
    In all cases, manufacturers' respective sales volumes were 
normalized to produce passenger car and light truck fleets that 
reflected manufacturers' MY2006 market shares and to reflect passenger 
car and light truck fleets of projected aggregate volume consistent 
with forecasts in the EIA's 2007 Annual Energy Outlook. The agency 
requests comment on whether alternative methods should be used to 
estimate manufacturers' market shares and the overall sizes of the 
future passenger car and light truck fleets.
    In a companion notice, the agency is requesting updated product 
plan information from all companies, and as in previous fuel economy 
rulemakings, we will be using those plans for the final rule. These 
plans will impact the standards for the final rule. To that end, the 
agency is requesting that these plans be as detailed and as accurate as 
possible.
7. What economic information does the Volpe model use?
    NHTSA's preliminary analysis of alternative CAFE standards for the 
model years covered by this proposed rulemaking relies on a range of 
information, economic estimates, and input parameters. This section 
describes this information and each assumption and specific parameter 
values, and discusses the rationale for tentatively choosing each one. 
Like the product plan information, these economic assumptions play a 
role in the determination of the level of the standards, with some 
having greater impacts than others. The cost of technologies and as 
discussed below, the price of gasoline and discount rate used for 
discounting future benefits have the greatest influence over the level 
of the standards. The agency seeks comment on the economic assumptions 
presented below. On the first question, based on the comparisons of the 
side cases to the base case that Jim did on Friday, the order of impact 
for the economic assumptions is: (1) Technology cost and effectiveness; 
(2) fuel prices; (3) discount rate; (4) oil import externalities; (5) 
rebound effect; (6) criteria air pollutant damage costs; (7) carbon 
costs. This reflects the base case assumptions, and could change 
slightly if we used different assumptions to start, but 1st through 3rd 
should stay the same.
    For the reader's reference, Table V-3 below summarizes the values 
used to calculate the impacts of each scenario:

      Table V-3.--Economic Values for Benefits Computations (2006$)
------------------------------------------------------------------------

------------------------------------------------------------------------
Rebound Effect (VMT Elasticity w/respect to Fuel Cost per          -0.15
 Mile).....................................................
Discount Rate Applied to Future Benefits...................           7%
Payback Period (years).....................................          5.0
``Gap'' between Test and On-Road mpg.......................          20%
Value of Travel Time per Vehicle ($/hour)..................       $24.00
Economic Costs of Oil Imports ($/gallon)
    ``Monopsony'' Component................................       $0.176
    Price Shock Component..................................       $0.109
    Military Security Component............................          $--
  Total Economic Costs ($/gallon)..........................       $0.285
    Total Economic Costs ($/BBL)...........................       $11.97
External Costs from Additional Automobile Use Due to
 ``Rebound'' Effect ($/vehicle-mile)
    Congestion.............................................       $0.047
    Accidents..............................................       $0.025
    Noise..................................................       $0.001
External Costs from Additional Light Truck Use Due to
 ``Rebound'' Effect ($/vehicle-mile)

[[Page 24404]]


    Congestion.............................................       $0.052
    Accidents..............................................       $0.023
    Noise..................................................       $0.001
Emission Damage Costs
    Carbon Monoxide ($/ton)................................          $--
    Volatile Organic Compounds ($/ton).....................       $1,700
    Nitrogen Oxides ($/ton)................................       $3,900
    Particulate Matter ($/ton).............................     $164,000
    Sulfur Dioxide ($/ton).................................      $16,000
    Carbon Dioxide ($/metric ton)..........................        $7.00
        Annual Increase in CO\2\ Damage Cost...............         2.4%
------------------------------------------------------------------------

a. Costs of Fuel Economy Technologies
    We developed detailed estimates of the costs of applying fuel 
economy-improving technologies to vehicle models for use in analyzing 
the impacts of alternative standards considered in this rulemaking. The 
estimates were based on those reported by the 2002 NAS Report analyzing 
costs for increasing fuel economy, but were modified for purposes of 
this analysis as a result of extensive consultations among engineers 
from NHTSA, EPA, and the Volpe Center. As part of this process, the 
agency also developed varying cost estimates for applying certain fuel 
economy technologies to vehicles of different sizes and body styles. We 
may adjust these cost estimates based on comments received to this 
NPRM.
    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. However, NHTSA recognizes that 
manufacturers' actual costs for applying these technologies to specific 
vehicle models are likely to include additional outlays for 
accompanying design or engineering changes to each model, development 
and testing of prototype versions, recalibrating engine operating 
parameters, and integrating the technology with other attributes of the 
vehicle. Manufacturers may also incur additional corporate overhead, 
marketing, or distribution and selling expenses as a consequence of 
their efforts to improve the fuel economy of individual vehicle models 
and their overall product lines.
    In order to account for these additional costs, NHTSA applies an 
indirect cost multiplier of 1.5 to the estimate of the vehicle 
manufacturers' direct costs for producing or acquiring each fuel 
economy-improving/CO2 emission-reducing technology. 
Historically, NHTSA has used an almost identical multiplier, 1.51, for 
the markup from variable costs or direct manufacturing costs to 
consumer costs. This markup takes into account fixed costs, burden, 
manufacturer's profit, and dealers' profit. NHTSA's methodology for 
determining this markup was recently peer reviewed.\96\
---------------------------------------------------------------------------

    \96\ See Docket No. NHTSA-2007-27454, Item 4.
---------------------------------------------------------------------------

    This estimate was confirmed by Argonne National Laboratory in a 
recent review of vehicle manufacturers' indirect costs. The Argonne 
study was specifically intended to improve the accuracy of future cost 
estimates for production of vehicles that achieve high fuel economy/low 
CO2 emissions by employing many of the same advanced 
technologies considered in our analysis.\97\ Thus, we believe that its 
recommendation that a multiplier of 1.5 be applied to direct 
manufacturing costs to reflect manufacturers' increased indirect costs 
for deploying advanced fuel economy technologies is appropriate for use 
in the analysis for this rulemaking.
---------------------------------------------------------------------------

    \97\ Vyas, Anant, Dan Santini, and Roy Cuenca, Comparison of 
Indirect Cost Multipliers for Vehicle Manufacturing, Center for 
Transportation Research, Argonne National Laboratory, April 2000. 
Available at http://www.transportation.anl.gov/pdfs/TA/57.pdf (last 
accessed April 20, 2008).
---------------------------------------------------------------------------

b. Potential Opportunity Costs of Improved Fuel Economy
    An important concern is whether achieving the fuel economy 
improvements required by alternative CAFE 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 in fuel 
economy, and thus of manufacturers' compliance with stricter CAFE 
standards. While exact dollar values of these attributes to consumers 
are difficult to infer from vehicle purchase prices, changing vehicle 
attributes can affect the utility that vehicles provide to their 
owners, and thus their value to potential buyers.
    NHTSA has approached this potential problem by developing tentative 
cost estimates for fuel economy-improving technologies that include any 
additional manufacturing costs that would be necessary to maintain the 
product plan levels of performance, comfort, capacity, or safety of any 
light-duty vehicle model to which those technologies are applied. In 
doing so, we primarily followed the precedent established by the 2002 
NAS Report, although we updated its assumptions as necessary for the 
purposes of the current rulemaking. The NAS study estimated ``constant 
performance and utility'' costs for fuel economy technologies, and 
NHTSA has used these as the basis for their further efforts to develop 
the technology costs employed in analyzing manufacturer's costs for 
complying with alternative light truck standards.
    NHTSA acknowledges the difficulty of estimating technology costs 
that include costs for the accompanying changes in vehicle design that 
are necessary to maintain performance, capacity, and utility. However, 
we believe that our tentative cost estimates for fuel economy/
CO2 emission-reduction technologies should be generally 
sufficient to prevent significant reductions in consumer welfare 
provided by vehicle models to which manufacturers apply those 
technologies. Nevertheless, we seek comments on alternative ways to 
deal with these issues.
c. 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 EPA to establish its published 
fuel economy ratings for different models. In analyzing the fuel 
savings from alternative CAFE standards, NHTSA has previously adjusted 
the actual fuel economy performance of each light truck model downward 
from its rated value to reflect the expected size of this on-road fuel

[[Page 24405]]

economy ``gap.'' On December 27, 2006, EPA adopted changes to its 
regulations on fuel economy labeling, which were intended to bring 
vehicles' rated fuel economy levels closer to their actual on-road fuel 
economy levels.\98\
---------------------------------------------------------------------------

    \98\ 71 FR 77871 (Dec. 27, 2006).
---------------------------------------------------------------------------

    In its Final Rule, EPA estimated that actual on-road fuel economy 
for light-duty vehicles averages 20 percent lower than published fuel 
economy levels. For example, if the overall EPA fuel economy rating 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 has employed EPA's revised estimate of this on-road fuel economy 
gap in its analysis of the fuel savings resulting from alternative CAFE 
standards proposed in this rulemaking.
d. Fuel Prices and the Value of Saving Fuel
    Projected future fuel prices are a critical input into the 
preliminary economic analysis of alternative CAFE standards, because 
they determine the value of fuel savings both to new vehicle buyers and 
to society. NHTSA relied on the most recent fuel price projections from 
the U.S. Energy Information Administration's (EIA) Annual Energy 
Outlook (AEO) for this analysis. Specifically, we used the AEO 2008 
Early Release 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.\99\ Federal government agencies generally use EIA's 
projections in their assessments of future energy-related policies.
---------------------------------------------------------------------------

    \99\ Energy Information Administration, Annual Energy Outlook 
2008, Early Release, Reference Case Table 12. Available at http://
www.eia.doe.gov/oiaf/aeo/pdf/aeotab_12.pdf (last accessed April 20, 
2008). EIA says that it will release the complete version of AEO 
2008--including the High and Low Price and other side cases--at the 
end of April. The agency will use those figures for the final rule.
---------------------------------------------------------------------------

    The retail fuel price forecasts presented in AEO 2008 span the 
period from 2008 through 2030. Measured in constant 2006 dollars, the 
Reference Case forecast of retail gasoline prices during calendar year 
2020 is $2.36 per gallon, rising gradually to $2.51 by the year 2030 
(these values include federal, state and local taxes). However, valuing 
fuel savings over the 36-year maximum lifetime of light trucks assumed 
in this analysis requires fuel price forecasts that extend through 
2050, the last year during which a significant number of MY 2015 
vehicles will remain in service.\100\ To obtain fuel price forecasts 
for the years 2031 through 2050, the agency assumes that retail fuel 
prices forecast in the Reference Case for 2030 will remain constant (in 
2006 dollars) through 2050.
---------------------------------------------------------------------------

    \100\ The agency defines the maximum lifetime of vehicles as the 
highest age at which more than 2 percent of those originally 
produced during a model year remain in service. In the case of 
light-duty trucks, for example, this age has typically been 36 years 
for recent model years.
---------------------------------------------------------------------------

    The value of fuel savings resulting from improved fuel economy/
reduced CO2 emissions to buyers of light-duty vehicles is 
determined by the retail price of fuel, which includes federal, state, 
and any local taxes imposed on fuel sales. Total taxes on gasoline 
averaged $0.47 per gallon during 2006, while those levied on diesel 
averaged $0.53. State fuel taxes are weighted by sales. Because fuel 
taxes represent transfers of resources from fuel buyers to government 
agencies, however, rather than real resources that are consumed in the 
process of supplying or using fuel, their value must be deducted from 
retail fuel prices to determine the value of fuel savings resulting 
from more stringent CAFE standards to the U.S. economy as a whole.
    In estimating the economy-wide or ``social'' value of fuel savings 
of increasing CAFE/reducing CO2 emissions levels, NHTSA 
assumes that current fuel taxes will remain constant in real or 
inflation-adjusted terms over the lifetimes of the vehicles proposed to 
be regulated. In effect, this assumes that the average value per gallon 
of taxes on gasoline and diesel fuel levied by all levels of government 
will rise at the rate of inflation over that period. This value is 
deducted from each future year's forecast of retail gasoline and diesel 
prices reported in AEO 2008 to determine the social value of each 
gallon of fuel saved during that year as a result of improved fuel 
economy/reduced CO2 emissions. Subtracting fuel taxes 
results in a projected value for saving gasoline of $1.83 per gallon 
during 2020, rising to $2.02 per gallon by the year 2030.
    In conducting the preliminary uncertainty analysis of benefits and 
costs from alternative CAFE standards, as required by OMB, NHTSA also 
considered higher and lower forecasts of future fuel prices. The 
results of the sensitivity runs can be found in the PRIA. EIA includes 
``High Price Case'' and ``Low Price Case'' in AEO analyses that reflect 
uncertainties regarding future levels of oil production, but those 
cases are not meant to be probabilistic, and simply illustrate the 
range of uncertainty that exists. Because AEO 2008 Early Release 
included only a Reference Case of forecast of fuel prices and did not 
include the High and Low Price cases, the agency estimated high and low 
fuel prices corresponding to the AEO 2008 Reference Case forecast by 
assuming that high and low price forecasts would bear the same 
relationship to the Reference Case forecast as reported in AEO 
2007.\101\ These alternative scenarios project retail gasoline prices 
that range from a low of $1.94 per gallon to a high of $3.26 per gallon 
during 2020, and from $2.03 to $3.70 per gallon during 2030. In 
conjunction with our assumption that fuel taxes will remain constant in 
real or inflation-adjusted terms over this period, these forecasts 
imply social values of saving fuel ranging from $1.47 to $2.79 per 
gallon during 2020, and from $1.56 to $3.23 per gallon in 2030.
---------------------------------------------------------------------------

    \101\ Energy Information Administration, Annual Energy Outlook 
2007, High Price Case, Table 12, http://www.eia.doe.gov/oiaf/aeo/
pdf/aeohptab_12.pdf (last accessed April 20, 2008) and Energy 
Information Administration, Annual Energy Outlook 2007 Low Price 
Case, Table 12, http://www.eia.doe.gov/oiaf/aeo/pdf/aeolptab_12.pdf 
(last accessed April 20, 2008).
---------------------------------------------------------------------------

    EIA is widely-recognized as an impartial and authoritative source 
of analysis and forecasts of U.S. energy production, consumption, and 
prices. The agency has published annual forecasts of energy prices and 
consumption levels for the U.S. economy since 1982 in its Annual Energy 
Outlook (AEO). These forecasts have been widely relied upon by federal 
agencies for use in regulatory analysis and for other purposes. Since 
1994, EIA's annual forecasts have been based upon the agency's National 
Energy Modeling System (NEMS), which includes detailed representation 
of supply pathways, sources of demand, and their interaction to 
determine prices for different forms of energy.
    From 1982 through 1993, EIA's forecasts of world oil prices--the 
primary determinant of prices for gasoline, diesel, and other 
transportation fuels derived from petroleum--consistently overestimated 
actual prices during future years, often very significantly. Of the 
total of 119 forecasts of future world oil prices for the years 1985 
through 2005 that EIA reported in its 1982-1993 editions of AEO, 109 
overestimated the subsequent actual values for those years, on average 
exceeding their corresponding actual values by 75 percent.
    Since that time, however, EIA's forecasts of future world oil 
prices show a more mixed record for accuracy. The 1994-2005 editions of 
AEO reported 91 separate forecasts of world oil prices for the years 
1995-2005, of which 33 have subsequently proven too high while the

[[Page 24406]]

remaining 58 have underestimated actual prices. The average absolute 
error (i.e., regardless of its direction) of these forecasts has been 
21 percent, but over- and underestimates have tended to offset one 
another, so that on average EIA's more recent forecasts have 
underestimated actual world oil prices by 7 percent. Although both its 
overestimates and underestimates of future world oil prices for recent 
years have often been large, the most recent editions of AEO have 
significantly underestimated petroleum prices during those years for 
which actual prices are now available.
    However, NHTSA does not regard EIA's recent tendency to 
underestimate future prices for petroleum and refined products or the 
high level of current fuel prices as adequate justification to employ 
forecasts that differ from the Reference Case forecast presented in 
EIA's Annual Energy Outlook 2008 Revised Early Release. This is 
particularly the case because this forecast has been revised upward 
significantly since the initial release of AEO 2008, which in turn 
represented a major upward revision from EIA's fuel price forecast 
reported previously in AEO 2007. NHTSA also notes that retail gasoline 
prices across the U.S. have averaged $2.94 per gallon (expressed in 
2005 dollars) for the first three months of 2008, slightly below EIA's 
recently revised forecast that gasoline prices will average $2.98 per 
gallon (also in 2005 dollars) throughout 2008.
    Comparing different forecasts of world oil prices also shows that 
EIA's Reference Case forecast reported in Annual Energy Outlook 2007 
(AEO 2007) was actually the highest of all six publicly-available 
forecasts of world oil prices over the 2010-30 time horizon.\102\ 
Because world petroleum prices are the primary determinant of retail 
prices for refined petroleum products such as transportation fuels, 
this suggests that the Reference Case forecast of U.S. fuel prices 
reported in AEO 2007 is likely to be the highest of those projected by 
major forecasting services. Further, as indicated above, EIA's most 
recent fuel price forecasts have been revised significantly upward from 
those previously projected in AEO 2007.
---------------------------------------------------------------------------

    \102\ See http://www.eia.doe.gov/oiaf/archive/aeo07/pdf/
forecast.pdf, Table 19, p. 106.
---------------------------------------------------------------------------

e. Consumer Valuation of Fuel Economy and Payback Period
    In estimating the value of fuel economy improvements that would 
result from alternative CAFE standards to potential vehicle buyers, 
NHTSA assumes that buyers value the resulting fuel savings 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 behave as if 
they do not discount the value of these future fuel savings. The five-
year figure represents the current average term of consumer loans to 
finance the purchase of new vehicles. We recognize that the period over 
which individual buyers finance new vehicle purchases may not 
correspond to the time horizons they apply in valuing fuel savings from 
higher fuel economy. However, NHTSA believes that five years represents 
a reasonable estimate of the average period over which buyers who 
finance their purchases of new vehicle receive--and thus must 
recognize--the monetary value of future fuel savings resulting from 
higher fuel economy.
    The value of fuel savings over the first five years of a vehicle 
model's lifetime that would result under each alternative fuel economy 
standard is calculated using the projections of retail fuel prices 
described above. It is then deducted from the technology costs incurred 
by its manufacturer to produce the improvement in that model's fuel 
economy estimated for each alternative standard, to determine the 
increase in the ``effective price'' to buyers of that vehicle model. 
The Volpe model uses these estimates of effective costs for increasing 
the fuel economy of each vehicle model to identify the order in which 
manufacturers would be likely to select models for the application of 
fuel economy-improving technologies in order to comply with stricter 
standards. The average value of the resulting increase in effective 
cost from each manufacturer's simulated compliance strategy is also 
used to estimate the impact of alternative standards on its total sales 
for future model years.
    However, it is important to recognize that NHTSA estimates the 
aggregate value to the U.S. economy of fuel savings resulting from 
alternative standards--or their ``social'' value--over the entire 
expected lifetimes of vehicles manufactured under those standards, 
rather than over this shorter ``payback period'' we assume for their 
buyers. This is discussed directly below in section f on ``Vehicle 
survival and use assumptions.'' As indicated previously, the maximum 
vehicle lifetimes used to analyze the effects of alternative fuel 
economy standards are estimated to be 25 years for automobiles and 36 
years for light trucks.
f. Vehicle Survival and Use Assumptions
    NHTSA's preliminary analysis of fuel/CO2 emissions 
savings and related benefits from adopting alternative standards for MY 
2011-2015 passenger cars and light trucks is based on estimates of the 
resulting changes in fuel use over their entire lifetimes in the U.S. 
vehicle fleet. The first step in estimating lifetime fuel consumption 
by vehicles produced during a model year is to calculate the number 
that is expected to remain in service during each future year after 
they are produced and sold.\103\ This number is calculated by 
multiplying the number of vehicles originally produced during a model 
year by the proportion expected to remain in service at the age they 
will have reached during each subsequent year, often referred to as a 
``survival rate.''
---------------------------------------------------------------------------

    \103\ Vehicles are defined to be of age 1 during the calendar 
year corresponding to the model year in which they are produced; 
thus for example, model year 2000 vehicles are considered to be of 
age 1 during calendar year 2000, age 1 during calendar year 2001, 
and to reach their maximum age of 26 years during calendar year 
2025. NHTSA considers the maximum lifetime of vehicles to be the age 
after which less than 2% of the vehicles originally produced during 
a model year remain in service. Applying these conventions to 
vehicle registration data indicates that passenger cars have a 
maximum age of 26 years, while light trucks have a maximum lifetime 
of 36 years. See Lu, S., NHTSA, Regulatory Analysis and Evaluation 
Division, ``Vehicle Survivability and Travel Mileage Schedules,'' 
DOT HS 809 952, 8-11 (January 2006). Available at http://www-
nrd.nhtsa.dot.gov/pdf/nrd-30/NCSA/Rpts/2006/809952.pdf (last 
accessed April 20, 2008).
---------------------------------------------------------------------------

    The agency relies on projections of the number of passenger cars 
and light trucks that will be produced during future years reported by 
the EIA in its AEO Reference Case forecast.\104\ It uses updated values 
of age-specific survival rates for cars and light trucks estimated from 
yearly registration data for vehicles produced during recent model 
years, to ensure that forecasts of the number of vehicles in use 
reflect recent increases in the durability and expected life spans of 
cars and light trucks.\105\
---------------------------------------------------------------------------

    \104\ The most recent edition is Energy Information 
Administration, Annual Energy Outlook 2008: Early Release. Available 
at http://www.eia.doe.gov/oiaf/aeo/index.html (last accessed April 
20, 2008).
    \105\ Lu, S., NHTSA, Regulatory Analysis and Evaluation 
Division, ``Vehicle Survivability and Travel Mileage Schedules,'' 
DOT HS 809 952, 8-11 (January 2006). Available at http://www-
nrd.nhtsa.dot.gov/pdf/nrd-30/NCSA/Rpts/2006/809952.pdf (last 
accessed April 20, 2008). These updated survival rates suggest that 
the expected lifetimes of recent-model passenger cars and light 
trucks are 13.8 and 14.5 years.
---------------------------------------------------------------------------

    The next step in estimating fuel use is to calculate the total 
number of miles that the cars and light trucks produced in each model 
year affected by the proposed CAFE standards will be driven during each 
year of their lifetimes. To

[[Page 24407]]

estimate total miles driven, the number of cars and light trucks 
projected to remain in use during each future year (calculated as 
described above) is multiplied by the average number of miles they are 
expected to be driven at the age they will have reached in that year. 
The agency estimated the average number of miles driven annually by 
cars and light trucks of each age using data from the Federal Highway 
Administration's 2001 National Household Transportation Survey 
(NHTS).\106\
---------------------------------------------------------------------------

    \106\ For a description of the Survey, see http://nhts.ornl.gov/
quickStart.shtml (last accessed April 20, 2008).
---------------------------------------------------------------------------

    Finally, fuel consumption during each year of a model year's 
lifetime is estimated by dividing the total number of miles its 
surviving vehicles are driven by the fuel economy they are expected to 
achieve under each alternative CAFE standard. Each model year's total 
lifetime fuel consumption is the sum of fuel use by the cars or light 
trucks produced during that model year that are projected to remain in 
use during each year of their maximum life spans. In turn, the savings 
in a model year's lifetime fuel use that will result from each 
alternative CAFE standard is the difference between its lifetime fuel 
use at the fuel economy level it attains under the Baseline 
alternative, and its lifetime fuel use at the higher fuel economy level 
it is projected to achieve under that alternative standard.
    To illustrate these calculations, the most recent edition of the 
AEO projections that 8.52 million light trucks will be produced during 
2012, and the agency's updated survival rates show that slightly more 
than half of these --50.1 percent, or 4.27 million--are projected to 
remain in service during the year 2027, when they will have reached an 
age of 14 years. At that age, light trucks achieving the fuel economy 
level required under the Baseline alternative are driven an average of 
about 10,400 miles, so model year 2012 light trucks will be driven a 
total of 44.4 billion miles (= 4.27 million surviving vehicles x 10,400 
miles per vehicle) during 2027. Summing the results of similar 
calculations for each year of their 36-year maximum lifetime, model 
year 2012 light trucks will be driven a total of 1,502 billion miles 
under the Baseline alternative. Under that alternative, they are 
projected to achieve a test fuel economy level of 23.8 mpg, which 
corresponds to actual on-road fuel economy of 19.0 mpg (= 23.8 mpg x 80 
percent). Thus their lifetime fuel use under the Baseline alternative 
is projected to be 79.0 billion gallons (= 1,502 billion miles divided 
by 19.0 miles per gallon).
g. Growth in Total Vehicle Use
    By assuming that the annual number of miles driven by cars and 
light trucks at any age will remain constant over the future, NHTSA's 
procedure for estimating the number of miles driven by cars and light 
trucks over their lifetimes in effect assumes that all future growth in 
total vehicle-miles driven stems from increases in the number of 
vehicles in service, rather than from increases in the average number 
of miles they are driven each year. Similarly, because the survival 
rates used to estimate the number of cars and light trucks remaining in 
service to various ages are assumed to remain fixed for future model 
years, growth in the total number of cars and light trucks in use is 
effectively assumed to result only from increasing sales of new 
vehicles. In order to determine the validity of these assumptions, the 
agency conducted a detailed analysis of the causes of recent growth in 
car and light truck use.
    From 1985 through 2005, the total number of miles driven (usually 
referred to as vehicle-miles traveled, or VMT) by passenger cars 
increased 35 percent, equivalent to a compound annual growth rate of 
1.5 percent.\107\ During that time, the total number of passenger cars 
registered for in the U.S. grew by about 0.3 percent annually, almost 
exclusively as a result of increasing sales of new cars.\108\ Thus 
growth in the average number of miles automobiles are driven each year 
accounted for the remaining 1.2 percent (= 1.5 percent--0.3 percent) 
annual growth in total automobile use.\109\
---------------------------------------------------------------------------

    \107\ Calculated from data reported in FHWA, Highway Statistics, 
Summary to 1995, Table vm201at http://www.fhwa.dot.gov/ohim/
summary95/vm201a.xlw, (last accessed April 20, 2008).and annual 
editions 1996-2005, Table VM-1 at http://www.fhwa.dot.gov/policy/
ohpi/hss/hsspubs.htm (last accessed April 20, 2008).
    \108\ A slight increase in the fraction of new passenger cars 
remaining in service beyond age 10 has accounted for a small share 
of growth in the U.S. automobile fleet. The fraction of new 
automobiles remaining in service to various ages was computed from 
R.L. Polk vehicle registration data for 1977 through 2005 by the 
agency's Center for Statistical Analysis.
    \109\ See supra note [2 above here]
---------------------------------------------------------------------------

    Over this same period, total VMT by light trucks increased much 
faster, growing at an annual rate of 5.1 percent. In contrast to the 
causes of growth in automobile use, however, nearly all growth in light 
truck use over these two decades was attributable to rapid increases in 
the number of light trucks in use.\110\ In turn, growth in the size of 
the nation's light truck fleet has resulted almost exclusively from 
rising sales of new light trucks, since the fraction of new light 
trucks remaining in service to various ages has remained stable or even 
declined slightly over the past two decades.\111\
---------------------------------------------------------------------------

    \110\ FHWA data show that growth in total miles driven by ``Two-
axle, four-tire trucks,'' a category that includes most or all light 
trucks used as passenger vehicles, averaged 5.1% annually from 1985 
through 2005. However, the number of miles light trucks are driven 
each year averaged 11,114 during 2005, almost unchanged from the 
average figure of 11,016 miles during 1985. Id.
    \111\ Unpublished analysis of R.L. Polk vehicle registration 
data conducted by NHTSA Center for Statistical Analysis, 2005.
---------------------------------------------------------------------------

    On the basis of this analysis, the agency tentatively concludes 
that its projections of future growth in light truck VMT account fully 
for the primary cause of its recent growth, which has been the rapid 
increase in sales of new light trucks during recent model years. 
However, the assumption that average annual use of passenger cars will 
remain fixed over the future appears to ignore an important source of 
recent growth in their total use, the gradual increase in the average 
number of miles they are driven. To the extent that this factor 
continues to represent a significant source of growth in future 
passenger car use, the agency's analysis is likely to underestimate the 
reductions in fuel use and related environmental impacts resulting from 
stricter CAFE standards for passenger cars.\112\ The agency plans to 
account explicitly for potential future growth in average annual use of 
both cars and light trucks in the analysis accompanying its Final Rule 
establishing CAFE standards for model years 2011-15.
---------------------------------------------------------------------------

    \112\ Assuming that average annual miles driven per automobile 
will continue to increase over the future would increase the 
agency's estimates of total lifetime mileage for MY 2011-18 
passenger cars. Their estimated lifetime fuel use would also 
increase under each alternative standard considered in this 
analysis, but in inverse relation to their fuel economy. Thus 
lifetime fuel use will increase by more under the No Increase 
alternative than under any of the alternatives that would increase 
passenger car CAFE standards, and by progressively less for the 
alternatives that impose stricter standards. Taking account of this 
factor would thus increase the agency's estimates of fuel savings 
for those alternatives, and omitting it will cause the agency's 
analysis to underestimate those fuel savings.
---------------------------------------------------------------------------

h. Accounting for the Rebound Effect of Higher Fuel Economy
    The rebound effect refers to the tendency for owners to increase 
the number of miles they drive a vehicle in response to an increase in 
its fuel economy, as would result from more stringent fuel economy 
standards. The rebound effect occurs because an increase in a vehicle's 
fuel economy reduces its owner's fuel cost for driving each mile, which 
is typically the largest

[[Page 24408]]

single component of the cost of operating a vehicle. Even with the 
vehicle's higher fuel economy, this additional driving uses some fuel, 
so the rebound effect will reduce the net fuel savings that result when 
the fuel economy standards require manufacturers to increase fuel 
economy. The rebound effect is usually expressed as the percentage by 
which annual vehicle use increases when average fuel cost per mile 
driven decreases in response to a change in the marginal cost of 
driving an extra mile, due either an increase in fuel economy or a 
reduction in the price of fuel.
    The magnitude of the rebound effect is one of the determinants of 
the actual fuel savings that are likely to result from adopting 
stricter standards, and thus an important parameter affecting NHTSA's 
evaluation of alternative standards for future model years. The rebound 
effect can be measured directly by estimating the elasticity of vehicle 
use with respect to fuel economy itself, or indirectly by the 
elasticity of vehicle use with respect to fuel cost per mile 
driven.\113\ When expressed as a positive percentage, either of these 
parameters gives the fraction of fuel savings that would otherwise 
result from adopting stricter standards, but is offset by the increase 
in fuel consumption that results when vehicles with increased fuel 
economy are driven more.
---------------------------------------------------------------------------

    \113\ Fuel cost per mile is equal to the price of fuel in 
dollars per gallon divided by fuel economy in miles per gallon, so 
this figure declines when a vehicle's fuel economy increases.
---------------------------------------------------------------------------

    Research on the magnitude of the rebound effect in light-duty 
vehicle use dates to the early 1980s, and almost unanimously concludes 
that a statistically significant rebound effect occurs when vehicle 
fuel efficiency improves.\114\ The most common approach to estimating 
its magnitude has been to analyze statistically household survey data 
on vehicle use, fuel consumption, fuel prices (often obtained from 
external sources), and other determinants of household travel demand to 
isolate the response of vehicle use to higher fuel economy. Other 
studies have relied on econometric analysis of annual U.S. data on 
vehicle use, fuel economy, fuel prices, and other variables to identify 
the response of total or average vehicle use to changes in fleet-wide 
average fuel economy and its effect of fuel cost per mile driven. Two 
recent studies analyzed yearly variation in vehicle ownership and use, 
fuel prices, and fuel economy among individual states over an extended 
time period in order to measure the response of vehicle use to changing 
fuel economy.\115\
---------------------------------------------------------------------------

    \114\ Some studies estimate that the long-run rebound effect is 
significantly larger than the immediate response to increased fuel 
efficiency. Although their estimates of the adjustment period 
required for the rebound effect to reach its long-run magnitude 
vary, this long-run effect is most appropriate for evaluating the 
fuel savings and emissions reductions resulting from stricter 
standards that would apply to future model years.
    \115\ In effect, these studies treat U.S. states as a data 
``panel'' by applying appropriate estimation procedures to data 
consisting of each year's average values of these variables for the 
separate states.
---------------------------------------------------------------------------

    An important distinction among studies of the rebound effect is 
whether they assume that the effect is constant, or varies over time in 
response to the absolute levels of fuel costs, personal income, or 
household vehicle ownership. Most studies using aggregate annual data 
for the U.S. assume a constant rebound effect, although some of these 
studies test whether the effect can vary as changes in retail fuel 
prices or average fuel economy alter fuel cost per mile driven. Many 
studies using household survey data estimate significantly different 
rebound effects for households owning varying numbers of vehicles, 
although they arrive at differing conclusions about whether the rebound 
effect is larger among households that own more vehicles. One recent 
study using state-level data concludes that the rebound effect varies 
directly in response to changes in personal income and the degree of 
urbanization of U.S. cities, as well as fuel costs.
    In order to arrive at a preliminary estimate of the rebound effect 
for use in assessing the fuel savings, emissions reductions, and other 
impacts of alternative standards, NHTSA reviewed 22 studies of the 
rebound effect conducted from 1983 through 2005. We then conducted a 
detailed analysis of the 66 separate estimates of the long-run rebound 
effect reported in these studies, which is summarized in the table 
below.\116\ As the table indicates, these 66 estimates of the long-run 
rebound effect range from as low as 7 percent to as high as 75 percent, 
with a mean value of 23 percent.
---------------------------------------------------------------------------

    \116\ In some cases, NHTSA derived estimates of the overall 
rebound effect from more detailed results reported in the studies. 
For example, where studies estimated different rebound effects for 
households owning different numbers of vehicles but did not report 
an overall value, we computed a weighted average of the reported 
values using the distribution of households among vehicle ownership 
categories.
---------------------------------------------------------------------------

    Limiting the sample to 50 estimates reported in the 17 published 
studies of the rebound effect yields the same range but a slightly 
higher mean (24 percent), while focusing on the authors' preferred 
estimates from published studies narrows this range and lowers its 
average only slightly. The median estimate of the rebound effect in all 
three samples, which is generally regarded as a more reliable indicator 
of their central tendency than the average because it is less 
influenced by unusually small and large estimates, is 22 percent. As 
Table V-4 indicates, approximately two-thirds of all estimates 
reviewed, of all published estimates, and of authors' preferred 
estimates fall in the range of 10-30 percent.

                                 Table V-4.--Summary of Rebound Effect Estimates
----------------------------------------------------------------------------------------------------------------
                                                Number of          Range                   Distribution
       Category of estimates         Number of            ------------------------------------------------------
                                      studies   estimates     Low        High      Median      Mean    Std. Dev.
----------------------------------------------------------------------------------------------------------------
All Estimates......................         22         66         7%        75%        22%        23%        14%
Published Estimates................         17         50         7%        75%        22%        24%        14%
Authors' Preferred Estimates.......         17         17         9%        75%        22%        22%        15%
 U.S. Time-Series Estimates........          7         34         7%        45%        14%        18%         9%
Household Survey Estimates.........         13         23         9%        75%        31%        31%        16%
Pooled U.S. State Estimates........          2          9         8%        58%        22%        25%        14%
Constant Rebound Effect (1)........         15         37         7%        75%        20%        23%        16%
Variable Rebound Effect: (1).......
Reported Estimates.................         10         29        10%        45%        23%        23%        10%
Updated to 2006 (2)................         10         29         6%        46%        16%        19%       12%
----------------------------------------------------------------------------------------------------------------
(1) Three studies estimate both constant and variable rebound effects.

[[Page 24409]]


(2) Reported estimates updated to reflect 2006 values of vehicle use, fuel prices, fleet fuel efficiency,
  household income, and household vehicle ownership.

    The type of data used and authors' assumption about whether the 
rebound effect varies over time have important effects on its estimated 
magnitude. The 34 estimates derived from analysis of U.S. annual time-
series data produce a median estimate of 14 percent for the long-run 
rebound effect, while the median of 23 estimates based on household 
survey data is more than twice as large (31 percent), and the median of 
9 estimates based on pooled state data matches that of the entire 
sample (22 percent). The 37 estimates assuming a constant rebound 
effect produce a median of 20 percent, while the 29 originally reported 
estimates of a variable rebound effect have a slightly higher median 
value (23 percent).
    In selecting a single value for the rebound effect to use in 
analyzing alternative standards for future model years, NHTSA 
tentatively attaches greater significance to studies that allow the 
rebound effect to vary in response to changes in the various factors 
that have been found to affect its magnitude. However, it is also 
important to update authors' originally-reported estimates of variable 
rebound effects to reflect current conditions. Recalculating the 29 
original estimates of variable rebound effects to reflect current 
(2006) values for retail fuel prices, average fuel economy, personal 
income, and household vehicle ownership reduces their median estimate 
to 16 percent.\117\ NHTSA also tentatively attaches greater 
significance to the recent study by Small and Van Dender (2005), which 
finds that the rebound effect tends to decline as average fuel economy, 
personal income, and suburbanization of U.S. cities increase, but--in 
accordance with previous studies--rises with increasing fuel 
prices.\118\
---------------------------------------------------------------------------

    \117\ As an illustration, Small and Van Dender (2005) allow the 
rebound effect to vary over time in response to changes in real per 
capita income as well as average fuel cost per mile driven. While 
their estimate for the entire interval (1966-2001) they analyze is 
22 percent, updating this estimate using 2006 values of these 
variables reduces the rebound effect to approximately 10 percent. 
Similarly, updating Greene's 1992 original estimate of a 15 percent 
rebound effect to reflect 2006 fuel prices and average fuel economy 
reduces it to 6 percent. See David L. Greene, ``Vehicle Use and Fuel 
Economy: How Big is the Rebound Effect?'' The Energy Journal, 13:1 
(1992), 117-143. In contrast, the distribution of households among 
vehicle ownership categories in the data samples used by Hensher et 
al. (1990) and Greene et al. (1999) are nearly identical to the most 
recent estimates for the U.S., so updating their original estimates 
to current U.S. conditions changes them very little. See David A. 
Hensher, Frank W. Milthorpe, and Nariida C. Smith, ``The Demand for 
Vehicle Use in the Urban Household Sector: Theory and Empirical 
Evidence,'' Journal of Transport Economics and Policy, 24:2 (1990), 
119-137; and David L. Greene, James R. Kahn, and Robert C. Gibson, 
``Fuel Economy Rebound Effect for Household Vehicles,'' The Energy 
Journal, 20:3 (1999), 1-21.
    \118\ In the most recent light truck CAFE rulemaking, NHTSA 
chose not to preference the Small and Van Dender study over other 
published estimates of the value of the rebound effect, stating that 
since it ``remains an unpublished working paper that has not been 
subjected to formal peer review, ``the agency does not yet consider 
the estimates it provides to have the same credibility as the 
published and widely-cited estimates it relied upon.'' See 71 FR 
17633 (Apr. 6, 2006). The study has subsequently been published and 
peer-reviewed, so NHTSA is now prepared to ``consider it in 
developing its own estimate of the rebound effect for use in 
subsequent CAFE rulemakings.''
---------------------------------------------------------------------------

    Considering the empirical evidence on the rebound effect as a 
whole, but according greater importance to the updated estimates from 
studies allowing the rebound effect to vary--particularly the Small and 
Van Dender study--NHTSA has selected a rebound effect of 15 percent to 
evaluate the fuel savings and other effects of alternative standards 
for the time period covered by this rulemaking. However, we do not 
believe that evidence of the rebound effect's dependence on fuel prices 
or household income is sufficiently convincing to justify allowing its 
future value to vary in response to forecast changes in these 
variables. A range extending from 10 percent to at least 20 percent--
and perhaps as high as 25 percent--appears to be appropriate for the 
required analysis of the uncertainty surrounding these estimates. While 
the agency selected 15 percent, it also ran sensitivity analyses at 10 
and 20 percent. The results are shown in the PRIA.
i. 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. As 
evidenced by their decisions to make more frequent or longer trips when 
improved fuel economy reduces their costs for driving, the benefits 
from this additional travel exceed the costs drivers and passengers 
incur in making more frequent or longer trips.
    The amount by which the benefits from this additional travel exceed 
its costs (for fuel and other operating expenses) measures the net 
benefits that drivers receive from the additional travel, usually 
referred to as increased consumer surplus. NHTSA's analysis estimates 
the economic value of the increased 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. The 
magnitude of these benefits represents a small fraction of the total 
benefits from the alternative fuel economy standards considered.
j. 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. Because 
drivers do not take these added costs into account in deciding when and 
where to travel, they must be accounted for separately as a cost of the 
added driving associated with the rebound effect.
    Increased vehicle use due to the rebound effect may also increase 
the costs associated with traffic accidents. Drivers may take account 
of the potential costs they (and their passengers) face from the 
possibility of being involved in an accident when they decide to make 
additional trips. However, they probably do not consider all of the 
potential costs they impose on occupants of other vehicles and on 
pedestrians when accidents occur, so any increase in these ``external'' 
accident costs must be considered as another cost of additional 
rebound-effect driving. Like increased delay costs, any increase in 
these external accident costs caused by added driving is likely to 
depend on the traffic conditions under which it takes place, since 
accidents are more frequent in heavier traffic (although their severity 
may be reduced by the slower speeds at which heavier traffic typically 
moves).
    Finally, added vehicle use from the rebound effect may also 
increase traffic noise. Noise generated by vehicles

[[Page 24410]]

causes inconvenience, irritation, and potentially even discomfort to 
occupants of other vehicles, to pedestrians and other bystanders, and 
to residents or occupants of surrounding property. Because these 
effects are unlikely to be taken into account by the drivers whose 
vehicles contribute to traffic noise, they represent additional 
externalities associated with motor vehicle use. Although there is 
considerable uncertainty in measuring their value, any increase in the 
economic costs of traffic noise resulting from added vehicle use must 
be included together with other increased external costs from the 
rebound effect.
    NHTSA relies 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.\119\ These estimates are intended to 
measure the increases in costs from added congestion, property damages 
and injuries in traffic accidents, and noise levels caused by 
automobiles and light trucks that are borne by persons other than their 
drivers (or ``marginal'' external costs). Updated to 2006 dollars, 
FHWA's ``Middle'' estimates for marginal congestion, accident, and 
noise costs caused by automobile use amount to 5.2 cents, 2.3 cents, 
and 0.1 cents per vehicle-mile (for a total of 7.6 cents per mile), 
while those for pickup trucks and vans are 4.7 cents, 2.5 cents, and 
0.1 cents per vehicle-mile (for a total of 7.3 cents per mile).\120\, 
\121\ These costs are multiplied by the annual increases in automobile 
and light truck use from the rebound effect to yield the estimated 
increases in congestion, accident, and noise externality costs during 
each future year.
---------------------------------------------------------------------------

    \119\ These estimates were developed by FHWA for use in its 1997 
Federal Highway Cost Allocation Study; see http://www.fhwa.dot.gov/
policy/hcas/final/index.htm (last accessed April 20, 2008).
    \120\ See Federal Highway Administration, 1997 Federal Highway 
Cost Allocation Study, http://www.fhwa.dot.gov/policy/hcas/final/
index.htm, Tables V-22, V-23, and V-24 (last accessed April 20, 
2008).
    \121\ The Federal Highway Administration's estimates of these 
costs agree closely with some other recent estimates. For example, 
recent published research conducted by Resources for the Future 
(RFF) estimates marginal congestion and external accident costs for 
increased light-duty vehicle use in the U.S. to be 3.5 and 3.0 cents 
per vehicle-mile in year-2002 dollars. See Ian W.H. Parry and 
Kenneth A. Small, ``Does Britain or the U.S. Have the Right Gasoline 
Tax?'' Discussion Paper 02-12, Resources for the Future, 19 and 
Table 1 (March 2002). Available at http://www.rff.org/rff/Documents/
RFF-DP-02-12.pdf (last accessed April 20, 2008).
---------------------------------------------------------------------------

k. 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; (2) 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.\122\ Higher U.S. imports of crude oil or refined petroleum 
products increase the magnitude of these external economic costs, thus 
increasing the true economic cost of supplying transportation fuels 
above the resource costs of producing them. Conversely, reducing U.S. 
imports of crude petroleum or refined fuels or reducing fuel 
consumption can reduce these external costs. Any reduction in their 
total value that results from improved light truck fuel economy 
represents an economic benefit of setting more stringent CAFE standards 
in addition to the value of fuel savings and emissions reductions 
itself.
---------------------------------------------------------------------------

    \122\ See, e.g., Bohi, Douglas R. and W. David Montgomery 
(1982). Oil Prices, Energy Security, and Import Policy Washington, 
DC: Resources for the Future, Johns Hopkins University Press; Bohi, 
D. R., and M. A. Toman (1993). ``Energy and Security: Externalities 
and Policies,'' Energy Policy 21:1093-1109; and Toman, M. A. (1993). 
``The Economics of Energy Security: Theory, Evidence, Policy,'' in 
A. V. Kneese and J. L. Sweeney, eds. (1993). Handbook of Natural 
Resource and Energy Economics, Vol. III. Amsterdam: North-Holland, 
pp. 1167-1218.
---------------------------------------------------------------------------

    Increased U.S. oil imports can impose higher costs on all 
purchasers of petroleum products, because the U.S. is a sufficiently 
large purchaser of foreign oil supplies that changes in U.S. demand can 
affect the world price. The effect of U.S. petroleum imports on world 
oil prices is determined by the degree of OPEC monopoly power over 
global oil supplies, and the degree of monopsony power over world oil 
demand exerted by the U.S. The combination of these two factors means 
that increases in domestic demand for petroleum products that are met 
through higher oil imports can cause the price of oil in the world 
market to rise, which imposes economic costs on all other purchasers in 
the global petroleum market in excess of the higher prices paid by U.S. 
consumers.\123\ Conversely, reducing U.S. oil imports can lower the 
world petroleum price, and thus generate benefits to other oil 
purchasers by reducing these ``monopsony costs.''
---------------------------------------------------------------------------

    \123\ For example, if the U.S. imports 10 million barrels of 
petroleum per day at a world oil price of $20 per barrel, its total 
daily import bill is $200 million. If increasing imports to 11 
million barrels per day causes the world oil price to rise to $21 
per barrel, the daily U.S. import bill rises to $231 million. The 
resulting increase of $31 million per day ($231 million minus $200 
million) is attributable to increasing daily imports by only 1 
million barrels. This means that the incremental cost of importing 
each additional barrel is $31, or $10 more than the newly-increased 
world price of $21 per barrel. This additional $10 per barrel 
represents a cost imposed on all other purchasers in the global 
petroleum market by U.S. buyers, in excess of the price they pay to 
obtain those additional imports.
---------------------------------------------------------------------------

    Although the degree of current OPEC monopoly power is subject to 
debate, the consensus appears to be that OPEC remains able to exercise 
some degree of control over the response of world oil supplies to 
variation in world oil prices, so that the world oil market does not 
behave completely competitively.\124\ The extent of U.S. monopsony 
power is determined by a complex set of factors including the relative 
importance of U.S. imports in the world oil market, and the sensitivity 
of petroleum supply and demand to its world price among other 
participants in the international oil market. Most evidence appears to 
suggest that variation in U.S. demand for imported petroleum continues 
to exert some influence on world oil prices, although this influence 
appears to be limited.\125\
---------------------------------------------------------------------------

    \124\ For a summary see Leiby, Paul N., Donald W. Jones, T. 
Randall Curlee, and Russell Lee, Oil Imports: An Assessment of 
Benefits and Costs, ORNL-6851, Oak Ridge National Laboratory, 
November 1, 1997, 17. Available at http://pzl1.ed.ornl.gov/
ORNL6851.pdf (last accessed April 20, 2008).
    \125\ Id. 18-19.
---------------------------------------------------------------------------

    The second component of external economic costs imposed by U.S. 
petroleum imports arises partly because an increase in oil prices 
triggered by a disruption in the supply of imported oil reduces the 
level of output that the U.S. economy can produce. The reduction in 
potential U.S. economic output depends on the extent and duration of 
the increases in petroleum product prices that result from a disruption 
in the supply of imported oil, as well as on whether and how rapidly 
these prices return to pre-disruption levels. Even if prices for 
imported oil return completely to their original levels, however, 
economic output will be at least temporarily reduced from the level 
that would have been possible without a disruption in oil supplies.
    Because supply disruptions and resulting price increases tend to 
occur

[[Page 24411]]

suddenly rather than gradually, they can also impose costs on 
businesses and households for adjusting their use of petroleum products 
more rapidly than if the same price increase had occurred gradually 
over time. These adjustments impose costs because they temporarily 
reduce economic output even below the level that would ultimately be 
reached once the U.S. economy completely adapted to higher petroleum 
prices. The additional costs to businesses and households reflect their 
inability to adjust prices, output levels, and their use of energy and 
other resources quickly and smoothly in response to rapid changes in 
prices for petroleum products.
    Since future disruptions in foreign oil supplies are an uncertain 
prospect, each of these disruption costs must be adjusted by the 
probability that the supply of imported oil to the U.S. will actually 
be disrupted. The ``expected value'' of these costs-- the product of 
the probability that an oil import disruption will occur and the costs 
of reduced economic output and abrupt adjustment to sharply higher 
petroleum prices--is the appropriate measure of their magnitude. Any 
reduction in these expected disruption costs resulting from a measure 
that lowers U.S. oil imports represents an additional economic benefit 
beyond the direct value of savings from reduced purchases of petroleum 
products.
    While the vulnerability of the U.S. economy to oil price shocks is 
widely thought to depend on total petroleum consumption rather than on 
the level of oil imports, variation in imports is still likely to have 
some effect on the magnitude of price increases resulting from a 
disruption of import supply. In addition, changing the quantity of 
petroleum imported into the U.S. may also affect the probability that 
such a disruption will occur. If either the size of the likely price 
increase or the probability that U.S. oil supplies will be disrupted is 
affected by oil imports, the expected value of the costs from a supply 
disruption will also depend on the level of imports.
    Businesses and households use a variety of market mechanisms, 
including oil futures markets, energy conservation measures, and 
technologies that permit rapid fuel switching to ``insure'' against 
higher petroleum prices and reduce their costs for adjusting to sudden 
price increases. While the availability of these market mechanisms has 
likely reduced the potential costs of disruptions to the supply of 
imported oil, consumers of petroleum products are unlikely to take 
account of costs they impose on others, so these costs are probably not 
reflected in the price of imported oil. Thus changes in oil import 
levels probably continue to affect the expected cost to the U.S. 
economy from potential oil supply disruptions, although this component 
of oil import costs is likely to be significantly smaller than 
estimated by studies conducted in the wake of the oil supply 
disruptions during the 1970s.
    The third component of the external economic costs of importing oil 
into the U.S. includes government outlays for maintaining a military 
presence to secure the supply of oil imports from potentially unstable 
regions of the world and to protect against their interruption. Some 
analysts also include outlays for maintaining the U.S. Strategic 
Petroleum Reserve (SPR), which is intended to cushion the U.S. economy 
against the consequences of disruption in the supply of imported oil, 
as additional costs of protecting the U.S. economy from oil supply 
disruptions.
    NHTSA believes that while costs for U.S. military security may vary 
over time in response to long-term changes in the actual level of oil 
imports into the U.S., these costs are unlikely to decline in response 
to any reduction in U.S. oil imports resulting from raising future CAFE 
standards for passenger cars and light trucks. U.S. military activities 
in regions that represent vital sources of oil imports also serve a 
broader range of security and foreign policy objectives than simply 
protecting oil supplies, and as a consequence are unlikely to vary 
significantly in response to changes in the level of oil imports 
prompted by higher standards.
    Similarly, while the optimal size of the SPR from the standpoint of 
its potential influence on domestic oil prices during a supply 
disruption may be related to the level of U.S. oil consumption and 
imports, its actual size has not appeared to vary in response to recent 
changes in oil imports. Thus while the budgetary costs for maintaining 
the Reserve are similar to other external costs in that they are not 
likely to be reflected in the market price for imported oil, these 
costs do not appear to have varied in response to changes in oil import 
levels.
    In analyzing benefits from its recent actions to increase light 
truck CAFE standards for model years 2005-07 and 2008-11, NHTSA relied 
on a 1997 study by Oak Ridge National Laboratory (ORNL) to estimate the 
value of reduced economic externalities from petroleum consumption and 
imports.\126\ More recently, ORNL updated its estimates of the value of 
these externalities, using the analytic framework developed in its 
original 1997 study in conjunction with recent estimates of the 
variables and parameters that determine their value.\127\ These include 
world oil prices, current and anticipated future levels of OPEC 
petroleum production, U.S. oil import levels, the estimated 
responsiveness of oil supplies and demands to prices in different 
regions of the world, and the likelihood of oil supply disruptions. 
ORNL prepared its updated estimates of oil import externalities for use 
by EPA in evaluating the benefits of reductions in U.S. oil consumption 
and imports expected to result from its Renewable Fuel Standard Rule of 
2007 (RFS).\128\
---------------------------------------------------------------------------

    \126\ Leiby, Paul N., Donald W. Jones, T. Randall Curlee, and 
Russell Lee, Oil Imports: An Assessment of Benefits and Costs, ORNL-
6851, Oak Ridge National Laboratory, November 1, 1997. Available at 
http://pzl1.ed.ornl.gov/ORNL6851.pdf (last accessed April 20, 2008).
    \127\ Leiby, Paul N. ``Estimating the Energy Security Benefits 
of Reduced U.S. Oil Imports,'' Oak Ridge National Laboratory, ORNL/
TM-2007/028, Revised July 23, 2007. Available at http://
pzl1.ed.ornl.gov/energysecurity.html (click on link below ``Oil 
Imports Costs and Benefits'') (last accessed April 20, 2008).
    \128\ 72 FR 23899 (May 1, 2007).
---------------------------------------------------------------------------

    The updated ORNL study was subjected to a detailed peer review by 
experts selected by EPA, and its estimates of the value of oil import 
externalities were subsequently revised to reflect their comments and 
recommendations.\129\ Specifically, reviewers recommended that ORNL 
increase its estimates of the sensitivity of oil supply by non-OPEC 
producers and oil demand by nations other than the U.S. to changes in 
the world oil price, as well as reduce its estimate of the sensitivity 
of U.S. gross domestic product (GDP) to potential sudden increases in 
world oil prices.
---------------------------------------------------------------------------

    \129\ Peer Review Report Summary: Estimating the Energy Security 
Benefits of Reduced U.S. Oil Imports, ICF, Inc., September 2007.
---------------------------------------------------------------------------

    After making the revisions recommended by peer reviewers, ORNL's 
updated estimates of the monopsony cost associated with U.S. oil 
imports range from $5.22 to $9.68 per barrel, with a most likely 
estimate of $7.41 per barrel. These estimates imply that each gallon of 
fuel saved as a result of adopting higher CAFE standards will reduce 
the monopsony costs of U.S. oil imports by $0.124 to $0.230 per gallon, 
with the actual value most likely to be $0.176 per gallon saved. ORNL's 
updated and revised estimates of the increase in the expected costs 
associated with oil supply disruptions to the U.S. and the resulting 
rapid increase in prices for petroleum products amount to $4.54 to 
$5.84 per barrel, although its

[[Page 24412]]

most likely estimate of $4.59 per barrel is very close to the lower end 
of this range. According to these estimates, each gallon of fuel saved 
will reduce the expected costs disruptions to the U.S. economy by 
$0.108 to $0.139, with the actual value most likely to be $0.109 per 
gallon.
    The updated and revised ORNL estimates suggest that the combined 
reduction in monopsony costs and expected costs to the U.S. economy 
from oil supply disruptions resulting from lower fuel consumption total 
$0.232 to $0.370 per gallon, with a most likely estimate of $0.286 per 
gallon. This represents the additional economic benefit likely to 
result from each gallon of fuel saved by higher CAFE standards, beyond 
the savings in resource costs for producing and distributing each 
gallon of fuel saved. NHTSA employs this midpoint estimate in its 
analysis of the benefits from fuel savings projected to result from 
alternative CAFE standards for model years 2011-15. It also analyzes 
the effect on these benefits estimates from variation in this value 
over the range from $0.232 to $0.370 per gallon of fuel saved.
    NHTSA's analysis of benefits from alternative CAFE standards does 
not include cost savings from either reduced outlays for U.S. military 
operations or maintaining a smaller SPR among the external benefits of 
reducing gasoline consumption and petroleum imports by means of 
tightening future standards. This view concurs with that of both the 
original ORNL study of economic costs from U.S. oil imports and its 
recent update, which conclude that savings in government outlays for 
these purposes are unlikely to result from reductions in consumption of 
petroleum products and oil imports on the scale of those likely to 
result from the alternative increases in CAFE standards considered for 
model years 2011-15.
l. Air Pollutant Emissions
(i) 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 from higher fuel economy will increase emissions of these 
pollutants. Thus the net effect of stricter CAFE standards on emissions 
of each criteria pollutant depends on the relative magnitudes of its 
reduced emissions in fuel refining and distribution, and increases in 
its emissions from vehicle use. Because the relationship between 
emissions rates (emissions per gallon refined of fuel or mile driven) 
in fuel refining and vehicle use is different for each criteria 
pollutant, the net effect of fuel savings from the proposed standards 
on total emissions of each pollutant is likely to differ. 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 
(NOX), fine particulate matter (PM2.5), and sulfur oxides 
(SOX).
    The increase in emissions of these pollutants from additional 
vehicle use due to the rebound effect is estimated by multiplying the 
increase in total miles driven by vehicles of each model year and age 
by age-specific emission rates per vehicle-mile for each pollutant. 
NHTSA developed these emission rates using EPA's MOBILE6.2 motor 
vehicle emissions factor model.\130\ Emissions of these pollutants also 
occur during crude oil extraction and transportation, fuel refining, 
and fuel storage and distribution. The reduction in total emissions 
from each of these sources thus depends on the extent to which fuel 
savings result in lower imports of refined fuel, or in reduced domestic 
fuel refining. To a lesser extent, they also depend on whether any 
reduction in domestic gasoline refining is translated into reduced 
imports of crude oil or reduced domestic extraction of petroleum.
---------------------------------------------------------------------------

    \130\ U.S. Environmental Protection Agency, MOBILE6 Vehicle 
Emission Modeling Software, available at http://www.epa.gov/otaq/
m6.htm#m60 (last accessed April 20, 2008).
---------------------------------------------------------------------------

    Based on analysis of changes in U.S. gasoline imports and domestic 
gasoline consumption forecast in AEO's 2008 Early Release, NHTSA 
tentatively estimates that 50 percent of fuel savings resulting from 
higher CAFE standards will result in reduced imports of refined 
gasoline, while the remaining 50 percent will reduce domestic fuel 
refining.\131\ The reduction in domestic refining is assumed to leave 
its sources of crude petroleum unchanged from the mix of 90 percent 
imports and 10 percent domestic production projected by AEO.
---------------------------------------------------------------------------

    \131\ Estimates of the response of gasoline imports and domestic 
refining to fuel savings from stricter standards are variable and 
highly uncertain, but our preliminary analysis indicates that under 
any reasonable assumption about these responses, the magnitude of 
the net change in criteria pollutant emissions (accounting for both 
the rebound effect and changes in refining emissions) is extremely 
low relative to their current total.
---------------------------------------------------------------------------

    NHTSA proposes to estimate reductions in criteria pollutant 
emissions from gasoline refining and distribution using emission rates 
obtained from Argonne National Laboratories' Greenhouse Gases and 
Regulated Emissions in Transportation (GREET) model.\132\ The GREET 
model provides separate estimates of air pollutant emissions that occur 
in four phases of fuel production and distribution: crude oil 
extraction, crude oil transportation and storage, fuel refining, and 
fuel distribution and storage.\133\ We tentatively assume that 
reductions in imports of refined fuel would reduce criteria pollutant 
emissions during fuel storage and distribution only. Reductions in 
domestic fuel refining using imported crude oil as a feedstock are 
tentatively assumed to reduce emissions during crude oil transportation 
and storage, as well as during gasoline refining, distribution, and 
storage, because less of each of these activities would be occurring. 
Similarly, reduced domestic fuel refining using domestically-produced 
crude oil is tentatively assumed to reduce emissions during all phases 
of gasoline production and distribution.\134\
---------------------------------------------------------------------------

    \132\ Argonne National Laboratories, The Greenhouse Gas and 
Regulated Emissions from Transportation (GREET) Model, Version 1.8, 
June 2007, available at http://www.transportation.anl.gov/software/
GREET/index.html (last accessed April 20, 2008).
    \133\ Emissions that occur during vehicle refueling at retail 
gasoline stations (primarily evaporative emissions of volatile 
organic compounds, or VOCs) are already accounted for in the 
``tailpipe'' emission factors used to estimate the emissions 
generated by increased light truck use. GREET estimates emissions in 
each phase of gasoline production and distribution in mass per unit 
of gasoline energy content; these factors are then converted to mass 
per gallon of gasoline using the average energy content of gasoline.
    \134\ In effect, this assumes that the distances crude oil 
travels to U.S. refineries are approximately the same regardless of 
whether it travels from domestic oilfields or import terminals, and 
that the distances that gasoline travels from refineries to retail 
stations are approximately the same as those from import terminals 
to gasoline stations.
---------------------------------------------------------------------------

    The net changes in emissions of each criteria pollutant are 
calculated by adding the increases in their emissions that result from 
increased vehicle use and the reductions that result from lower 
domestic fuel refining and distribution. The net change in emissions of 
each criteria pollutant is converted to an economic value using 
estimates of the economic costs per ton emitted (which result primarily 
from damages to human health) developed by EPA and submitted to the 
federal Office of Management and Budget for review. For certain 
criteria pollutants, EPA estimates different per-ton costs for 
emissions from vehicle use than for emissions of the same pollutant 
during fuel production, reflecting differences in their typical 
geographic distributions,

[[Page 24413]]

contributions to ambient pollution levels, and resulting population 
exposure.
(ii) Reductions in CO2 Emissions
    Fuel savings from stricter CAFE standards also result in lower 
emissions of carbon dioxide (CO2), the main greenhouse gas emitted as a 
result of refining, distribution, and use of transportation fuels.\135\ 
Lower fuel consumption reduces carbon dioxide emissions directly, 
because the primary source of transportation-related CO2 
emissions is fuel combustion in internal combustion engines. NHTSA 
tentatively estimates reductions in carbon dioxide emissions resulting 
from fuel savings by assuming that the entire carbon content of 
gasoline, diesel, and other fuels is converted to carbon dioxide during 
the combustion process.\136\
---------------------------------------------------------------------------

    \135\ For purposes of this rulemaking, NHTSA estimated emissions 
of vehicular CO2 emissions, but did not estimate vehicular emissions 
of methane, nitrous oxide, and hydroflourocarbons. Methane and 
nitrous oxide account for less than 3 percent of the tailpipe GHG 
emissions from passenger cars and light trucks, and CO2 
emissions accounted for the remaining 97 percent. Of the total 
(including non-tailpipe) GHG emissions from passenger cars and light 
trucks, tailpipe CO2 represents about 93.1 percent, 
tailpipe methane and nitrous oxide represent about 2.4 percent, and 
hydroflourocarbons (i.e., air conditioner leaks) represent about 4.5 
percent. Calculated from U.S CO2. EPA, Inventory of U.S> 
Greenhouse Gas Emissions and Sinks 1990-2006, EPA430-R-08-05, April 
15, 2008. Available at http://www.epa.gov/climatechange/emissions/
downloads/08_CR.pdf, Table 215. (Last accessed April 20, 2008.)
    \136\ This assumption results in a slight overestimate of carbon 
dioxide emissions, since a small fraction of the carbon content of 
gasoline is emitted in the forms of carbon monoxide and unburned 
hydrocarbons. However, the magnitude of this overestimate is likely 
to be extremely small. This approach is consistent with the 
recommendation of the Intergovernmental Panel on Climate Change for 
``Tier 1'' national greenhouse gas emissions inventories. Cf. 
Intergovernmental Panel on Climate Change, 2006 Guidelines for 
National Greenhouse Gas Inventories, Volume 2, Energy, p. 3.16.
---------------------------------------------------------------------------

    Reduced fuel consumption also reduces carbon dioxide emissions that 
result from the use of carbon-based energy sources during fuel 
production and distribution.\137\ NHTSA currently estimates the 
reductions in CO2 emissions during each phase of fuel 
production and distribution using CO2 emission rates 
obtained from the GREET model, using the previous assumptions about how 
fuel savings are reflected in reductions in each phase. The total 
reduction in CO2 emissions from the improvement in fuel 
economy under each alternative CAFE standard is the sum of the 
reductions in emissions from reduced fuel use and from lower fuel 
production and distribution.
---------------------------------------------------------------------------

    \137\ NHTSA did not, for purposes of this proposed rulemaking, 
attempt to estimate changes in ``upstream'' emissions of greenhouse 
gases (GHGs) other than CO2. This was because carbon 
dioxide from final combustion itself accounts for nearly 97 percent 
of the total CO2-equivalent emissions from petroleum 
production and use, even with other GHGs that result from those 
activities (principally methane and nitrous oxide) weighted by their 
higher global warming potentials (GWPs) relative to CO2. 
Calculated from U.S. EPA, Inventory of U.S. Greenhouse Gas Emissions 
and Sinks 1990-2006, EPA430-R-08-05, April 15, 2008. Available at 
http://epa.gov/climatechange/emissions/downloads/08_CR.pdf, Tables 
3-3, 3-39, and 3-41. (Last accessed April 20, 2008.)
---------------------------------------------------------------------------

    NHTSA has not attempted to estimate changes in emissions of other 
greenhouse gases, in particular methane, nitrous oxide, and 
hydrofluorocarbons. The agency invites comment on the importance and 
potential implications of doing so under NEPA.
(iii) Economic value of reductions in CO2 emissions
    NHTSA has taken the economic benefits of reducing CO2 
emission into account in this rulemaking, both in developing proposed 
CAFE standards and in assessing the economic benefits of each 
alternative that was considered. As noted above, the Ninth Circuit 
found in CBD that NHTSA had been arbitrary and capricious in deciding 
not to monetize the benefit of reducing CO2 emissions, 
saying that the agency had not substantiated the conclusion in its 
April 2006 final rule that the appropriate course was not to monetize 
(i.e., quantify the value of) carbon emissions reduction at all.
    To this end, NHTSA reviewed published estimates of the ``social 
cost of carbon emissions'' (SCC). The SCC refers to the marginal cost 
of additional damages caused by the increase in expected climate 
impacts resulting from the emission of each additional metric ton of 
carbon, which is emitted in the form of CO2.\138\ It is 
typically estimated as the net present value of the impact over some 
time period (100 years or longer) of one additional ton of carbon 
emitted into the atmosphere. Because accumulated concentrations of 
greenhouse gases in the atmosphere and the projected impacts on global 
climate are increasing over time, the economic damages resulting from 
each additional ton of CO2 emissions in future years are 
believed to be greater as a result. Thus estimates of the SCC are 
typically reported for a specific year, and these estimates are 
generally larger for emissions in more distant future years.
---------------------------------------------------------------------------

    \138\ Carbon itself accounts for 12/44, or about 27%, of the 
mass of carbon dioxide (12/44 is the ratio of the molecular weight 
of carbon to that of carbon dioxide). Thus each ton of carbon 
emitted is associated with 44/12, or 3.67, tons of carbon dioxide 
emissions. Estimates of the SCC are typically reported in dollars 
per ton of carbon, and must be divided by 3.67 to determine their 
equivalent value per ton of carbon dioxide emissions.
---------------------------------------------------------------------------

    There is substantial variation among different authors' estimates 
of the SCC, much of which can be traced to differences in their 
underlying assumptions about several variables. These include the 
sensitivity of global temperatures and other climate attributes to 
increasing atmospheric concentrations of greenhouse gases, discount 
rates applied to future economic damages from climate change, whether 
damages sustained by developing regions of the globe should be weighted 
more heavily than damages to developed nations, how long climate 
changes persist once they occur, and the economic valuation of specific 
climate impacts.\139\
---------------------------------------------------------------------------

    \139\ For a discussion of these factors, see Yohe, G.W., R.D. 
Lasco, Q.K. Ahmad, N.W. Arnell, S.J. Cohen, C. Hope, A.C. Janetos 
and R.T. Perez, 2007: Perspectives on climate change and 
sustainability. Climate Change 2007: Impacts, Adaptation and 
Vulnerability. Contribution of Working Group II to the Fourth 
Assessment Report of the Intergovernmental Panel on Climate Change, 
M.L. Parry, O.F. Canziani, J.P. Palutikof, P.J. van der Linden and 
C.E. Hanson, Eds., Cambridge University Press, Cambridge, UK, pp. 
821-824.
---------------------------------------------------------------------------

    Taken as a whole, recent estimates of the SCC may underestimate the 
true damage costs of carbon emissions because they often exclude 
damages caused by extreme weather events or climate response scenarios 
with low probabilities but potentially extreme impacts, and may 
underestimate the climate impacts and damages that could result from 
multiple stresses on the global climatic system. At the same time, 
however, many studies fail to consider potentially beneficial impacts 
of climate change, and do not adequately account for how future 
development patterns and adaptations could reduce potential impacts 
from climate change or the economic damages they cause.
    Given the uncertainty surrounding estimates of the SCC, the use of 
any single study may not be advisable since its estimate of the SCC 
will depend on many assumptions made by its authors. The Working Group 
II's contribution to the Fourth Assessment Report of the United Nations 
Intergovernmental Panel on Climate Change (IPCC)\140\ notes that:
---------------------------------------------------------------------------

    \140\ Climate Change 2007--Impacts, Adaptation and 
Vulnerability, Contribution of Working Group II to the Fourth 
Assessment Report of the IPCC, 17. Available at http://www.ipcc-
wg2.org (last accessed ).

    The large ranges of SCC are due in the large part to differences 
in assumptions regarding climate sensitivity, response lags, the 
treatment of risk and equity, economic and non-economic impacts, the 
---------------------------------------------------------------------------
inclusion of potentially catastrophic losses, and discount rates.


[[Page 24414]]


    Although the IPCC does not recommend a single estimate of the SCC, 
it does cite the Tol (2005) study on four separate occasions (pages 17, 
65, 813, 822) as the only available survey of the peer-reviewed 
literature that has itself been subjected to peer review. Tol developed 
a probability function using the SCC estimates of the peer reviewed 
literature and found estimates ranging from less than zero to over $200 
per metric ton of carbon. In an effort to resolve some of the 
uncertainty in reported estimates of climate damage costs from carbon 
emissions, Tol (2005) reviewed and summarized one hundred and three 
estimates of the SCC from 28 published studies. He concluded that when 
only peer-reviewed studies published in recognized journals are 
considered, ``* * * climate change impacts may be very uncertain but is 
unlikely that the marginal damage costs of carbon dioxide emissions 
exceed $50 per [metric] ton carbon [about $14 per metric ton of 
CO2].'' \141\ He also concluded that the costs may be less 
than $14.
---------------------------------------------------------------------------

    \141\ Tol, Richard. The marginal damage costs of carbon dioxide 
emissions: an assessment of the uncertainties. Energy Policy 33 
(2005) 2064-2074, 2072. The summary SCC estimates reported by Tol 
are assumed to be denominated in U.S. dollars of the year of 
publication, 2005.
---------------------------------------------------------------------------

    Because of the number of assumptions required by each study, the 
wide range of uncertainty surrounding these assumptions, and their 
critical influence on the resulting estimates of climate damage costs, 
some studies have undoubtedly produced estimates of the SCC that are 
unrealistically high, while others are likely to have estimated values 
that are improbably low. Using a value for the SCC that reflects the 
central tendency of estimates drawn from many studies reduces the 
chances of relying on a single estimate that subsequently proves to be 
biased.
    It is important to note that estimates of the SCC almost invariably 
include the value of worldwide damages from potential climate impacts 
caused by carbon dioxide emissions, and are not confined to damages 
likely to be suffered within the U.S. In contrast, the other estimates 
of costs and benefits of increasing fuel economy included in this 
proposal include only the economic values of impacts that occur within 
the U.S. For example, the economic value of reducing criteria air 
pollutant emissions from overseas oil refineries is not counted as a 
benefit resulting from this rule, because any reduction in damages to 
health and property caused by overseas emissions are unlikely to be 
experienced within the U.S.
    In contrast, the reduced value of transfer payments from U.S. oil 
purchasers to foreign oil suppliers that results when lower U.S. oil 
demand reduces the world price of petroleum (the reduced ``monopsony 
effect'') is counted as a benefit of reducing fuel use.\142\ If the 
agency's analysis was conducted from a worldwide rather than a U.S. 
perspective, however, the benefit from reducing air pollution overseas 
would be included, while reduced payments from U.S. oil consumers to 
foreign suppliers would not.
---------------------------------------------------------------------------

    \142\ The reduction in payments from U.S. oil purchasers to 
domestic petroleum producers is not included as a benefit, since it 
represents a transfer that occurs entirely within the U.S. economy.
---------------------------------------------------------------------------

    In order to be consistent with NHTSA's use of exclusively domestic 
costs and benefits in prior CAFE rulemakings, the appropriate value to 
be placed on changes climate damages caused by carbon emissions should 
be one that reflects the change in damages to the United States alone. 
Accordingly, NHTSA notes that the value for the benefits of reducing 
CO2 emissions might be restricted to the fraction of those 
benefits that are likely to be experienced within the United States.
    Although no estimates of benefits to the U.S. itself that are 
likely to result from reducing CO2 emissions are currently 
available, NHTSA expects that if such values were developed, the agency 
would employ those rather than global benefit estimates in its 
analysis. NHTSA also anticipates that if such values were developed, 
they would be lower than comparable global values, since the U.S. is 
likely to sustain only a fraction of total global damages resulting 
from climate change.
    In the meantime, the agency has elected to use the IPCC estimate of 
$43 per metric ton of carbon as an upper bound on the benefits 
resulting from reducing each metric ton of U.S. emissions.\143\ This 
corresponds to approximately $12 per metric ton of CO2 when 
expressed in 2006 dollars. This estimate is based on the 2005 Tol 
study.\144\ The Tol study is cited repeatedly as an authoritative 
survey in various IPCC reports, which are widely accepted as 
representing the general consensus in the scientific community on 
climate change science. Since the IPCC estimate includes the worldwide 
costs of potential damages from carbon dioxide emissions, NHTSA has 
elected to employ it as an upper bound on the estimated value of the 
reduction in U.S. domestic damage costs that is likely to result from 
lower CO2 emissions.\145\
---------------------------------------------------------------------------

    \143\ The estimate of $43 per ton of carbon emissions is 
reported by Tol (p. 2070) as the mean of the ``best'' estimates 
reported in peer-reviewed studies (see fn. 144). It thus differs 
from the mean of all estimates reported in the peer-reviewed studies 
surveyed by Tol. The $43 per ton value is also attributed to Tol by 
IPCC Working Group II (2007), p. 822.
    \144\ Tol's more recent (2007) and inclusive survey has been 
published online with peer-review comments. The agency has elected 
not to rely on the estimates it reports, but will consider doing so 
in its analysis of the final rule if the survey has been published, 
and will also consider any other newly-published evidence.
    \145\ For purposes of comparison, we note that in the rulemaking 
to establish CAFE standards for MY 2008-11 light trucks, NRDC 
recommended a value of $10 to $25 per ton of CO2 
emissions reduced by fuel savings and both Environmental Defense and 
Union of Concerned Scientists recommended a value of $50 per ton of 
carbon (equivalent to about $14 per ton of CO2 
emissions).
---------------------------------------------------------------------------

    The IPCC Working Group II Fourth Assessment Report (2007, p. 822) 
further suggests that the SCC of carbon is growing at an annual 2.4 
percent growth rate, based on estimated increases in damages from 
future emissions reported in published studies. NHTSA has also elected 
to apply this growth rate to Tol's original 2005 estimate. Thus by 
2011, the agency estimates that the upper bound on the benefits of 
reducing CO2 emissions will have reached about $14 per 
metric ton of CO2, and will continue to increase by 2.4 
percent annually thereafter.
    In setting a lower bound, the agency agrees with the IPCC Working 
Group II (2007) report that ``significant warming across the globe and 
the locations of significant observed changes in many systems 
consistent with warming is very unlikely to be due solely to natural 
variability of temperatures or natural variability of the systems'' 
(pp. 9). Although this finding suggests that the global value of 
economic benefits from reducing carbon dioxide emissions is unlikely to 
be zero, it does not necessarily rule out low or zero values for the 
benefit to the U.S. itself from reducing emissions.
    For most of the analysis it performed to develop this proposal, 
NHTSA required a single estimate for the value of reducing 
CO2 emissions. The agency thus elected to use the midpoint 
of the range from $0 to $14 (or $7.00) per metric ton of CO2 
as the initial value for the year 2011, and assumed that this value 
would grow at 2.4 percent annually thereafter. This estimate is 
employed for the analyses conducted using the Volpe CAFE model to 
support development of the proposed standards. The agency also 
conducted sensitivity analyses of the benefits from reducing 
CO2 emissions using both the upper ($14 per metric ton) and 
lower ($0 per metric ton) bounds of this range.
    NHTSA seeks comment on its tentative conclusions for the value of

[[Page 24415]]

the SCC, the use of a domestic versus global value for the economic 
benefit of reducing CO2 emissions, the rate at which the 
value of the SCC grows over time, the desirability of and procedures 
for incorporating benefits from reducing emissions of greenhouse gases 
other than CO2, and any other aspects of developing a 
reliable SCC value for purposes of establishing CAFE standards.
m. The Value of Increased Driving Range
    Improving vehicles' fuel economy may also increase their driving 
range before they require refueling. 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 thus provides some additional benefits to their 
owners. (Alternatively, if manufacturers respond to improved fuel 
economy by reducing the size of fuel tanks to maintain a constant 
driving range, the resulting cost saving will presumably be reflected 
in lower vehicle sales prices.)
    No direct estimates of the value of extended vehicle range are 
readily available, so NHTSA's 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.\146\ As 
an illustration of how the value of extended refueling range is 
estimated, a typical small light truck model has an average fuel tank 
size of approximately 20 gallons. Assuming that drivers typically 
refuel when their tanks are 20 percent full (i.e., 4 gallons in 
reserve), increasing this model's actual on-road fuel economy from 24 
to 25 mpg would extend its driving range from 384 miles (= 16 gallons x 
24 mpg) to 400 miles (= 16 gallons x 25 mpg). Assuming that it is 
driven 12,000 miles/year, this reduces the number of times it needs to 
be refueled each year from 31.3 (= 12,000 miles per year/384 miles per 
refueling) to 30.0 (= 12,000 miles per year/400 miles per refueling), 
or by 1.3 refuelings per year.
---------------------------------------------------------------------------

    \146\ See Department of Transportation, Guidance Memorandum, 
``The Value of Saving Travel Time: Departmental Guidance for 
Conducting Economic Evaluations,'' Apr. 9, 1997. Available at http:/
/ostpxweb.dot.gov/policy/Data/VOT97guid.pdf (last accessed October 
20, 2007); update available at http://ostpxweb.dot.gov/policy/Data/
VOTrevision1--2-11-03.pdf (last accessed October 20, 2007).
---------------------------------------------------------------------------

    Weighted by the nationwide mix of urban (about 2/3) and rural 
(about 1/3) driving and average vehicle occupancy for all driving trips 
(1.6 persons), the DOT-recommended value of travel time per vehicle-
hour is $24.00 (in 2006 dollars).\147\ Assuming that locating a station 
and filling up requires ten minutes, the annual value of time saved as 
a result of less frequent refueling amounts to $5.20 (calculated as 10/
60 x 1.3 x $24.00). This calculation is repeated for each future 
calendar year that vehicles of each model year affected by the 
alternative CAFE standards proposed in this rule would remain in 
service. Like fuel savings and other benefits, however, the value of 
this benefit declines over a model year's lifetime, because a smaller 
number of vehicles originally produced during that model year remain in 
service each year, and those remaining in service are driven fewer 
miles.
n. Discounting Future Benefits and Costs
---------------------------------------------------------------------------

    \147\ The hourly wage rate during 2006 is estimated to be 
$24.00. Personal travel (94.4 percent of urban travel) is valued at 
50 percent of the hourly wage rate. Business travel (5.6 percent or 
urban travel) is valued at 100 percent of the hourly wage rate. For 
intercity travel, personal travel (87 percent) is valued at 70 
percent of the wage rate, while business travel (13 percent) is 
valued at 100 percent of the wage rate. The resulting values of 
travel time are $12.67 for urban travel and $17.66 for intercity 
travel, and must be multiplied by vehicle occupancy (1.6) to obtain 
the estimate value of time per vehicle hour.
---------------------------------------------------------------------------

    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. NHTSA uses a rate of 7 percent per year to 
discount the value of future fuel savings and other benefits to analyze 
the potential impacts of alternative CAFE standards. However, the 
agency also performed an alternative analysis of benefits from 
alternative increases in CAFE standards using a 3 percent discount 
rate, and seeks comment on whether the standards should be set using a 
3 percent rate instead of a 7 percent rate.
    There are several reasons that NHTSA relies primarily on 7 percent 
as the appropriate rate for discounting future benefits from increased 
CAFE standards. First, OMB Circular A-4 indicates that this rate 
reflects the economy-wide opportunity cost of capital.\148\ It also 
states that this ``is the appropriate discount rate whenever the main 
effect of a regulation is to displace or alter the use of capital in 
the private sector.''\149\ We believe that a substantial portion of the 
cost of this regulation may come at the expense of other investments 
the auto manufacturers might otherwise make. Several large 
manufacturers are resource-constrained with respect to their 
engineering and product-development capabilities. As a result, other 
uses of these resources will be foregone while they are required to be 
applied to technologies that improve fuel economy.
---------------------------------------------------------------------------

    \148\ Office of Management and Budget, Circular A-4, 
``Regulatory Analysis,'' September 17, 2003, 33. Available at http:/
/www.whitehouse.gov/omb/circulars/a004/a-4.pdf (last accessed Feb. 
14, 2008).
    \149\ Id.
---------------------------------------------------------------------------

    Second, 7 percent also appears to be an appropriate rate to the 
extent that the costs of the regulation come at the expense of 
consumption as opposed to investment. NHTSA believes that financing 
rates on vehicle loans represent an appropriate discount rate, because 
they reflect the opportunity costs faced by consumers when buying 
vehicles with greater fuel economy and a higher purchase price. Most 
new and used vehicle purchases are financed, and because most of the 
benefits from higher fuel economy standards accrue to vehicle 
purchasers in the form of fuel savings, the appropriate discount rate 
is the interest rate buyers pay on loans to finance their vehicle 
purchases.\150\
---------------------------------------------------------------------------

    \150\ Some empirical evidence also demonstrates that used car 
purchasers are willing to pay higher prices for greater fuel 
economy; see, e.g., James A. Kahn, ``Gasoline Price Expectations and 
the Used Automobile Market: A Rational Expectations Asset Price 
Approach,'' Quarterly Journal of Economics, Vol. 101 (May 1986), 
323-339.
---------------------------------------------------------------------------

    According to the Federal Reserve, the interest rate on new car 
loans made through commercial banks has closely tracked the rate on 10-
year treasury notes, but exceeded it by about 3 percent.\151\ The 
official Administration forecast is that real (or inflation-adjusted) 
interest rates on 10-year treasury notes will average about 3 percent 
through 2016, implying that 6 percent is a reasonable forecast for the 
real interest rate on new car loans.\152\ In turn, the interest rate on 
used car loans

[[Page 24416]]

made through automobile financing companies has closely tracked the 
rate on new car loans made through commercial banks, but exceeded it by 
about 3 percent.\153\ (We consider rates on loans that finance used car 
purchases, because some of the fuel savings resulting from improved 
fuel economy accrue to used car buyers.) Given the 6 percent estimate 
for new car loans, a reasonable forecast for used car loans is thus 9 
percent.
---------------------------------------------------------------------------

    \151\ See Federal Reserve Bank, Statistical Release H.15, 
Selected Interest Rates (Weekly) (click on ``Historical Data,'' then 
``Treasury constant maturities,'' then ``10-year, monthly''), 
available at http://www.federalreserve.gov/Releases/H15/data/
Monthly/H15_TCMNOM_Y10.txt (last accessed February 13, 2008); and 
Federal Reserve Bank, Statistical Release G.19, Consumer Credit, 
(click on ``Historical Data,'' then ``Terms of Credit'') available 
at http://www.federalreserve.gov/releases/g19/hist/cc_hist_tc.html 
(last accessed February 13, 2008).
    \152\ See The White House, Joint Press Release of the Council of 
Economic Advisors, the Department of the Treasury, and the Office of 
Management and Budget, November 29, 2007, available at http://
www.whitehouse.gov/news/releases/2007/11/20071129-4.html (last 
accessed February 13, 2008).
    \153\ See supra [2 above here] and Federal Reserve Bank, 
Statistical Release G.20, Finance Companies, (click on ``Historical 
Data,'' then ``Terms of Credit'') available at http://
www.federalreserve.gov/releases/g20/hist/fc_hist_tc.html (last 
accessed February 13, 2008).
---------------------------------------------------------------------------

    Because the benefits of fuel economy accrue to both new and used 
car owners, a discount rate between 6 percent and 9 percent is thus 
appropriate for evaluating future benefits resulting from more 
stringent fuel economy standards. Assuming that new car buyers discount 
fuel savings at 6 percent for 5 years (the average duration of a new 
car loan) \154\ and that used car buyers discount fuel savings at 9 
percent for 5 years (the average duration of a used car loan), \155\ 
the single constant discount rate that yields equivalent present value 
fuel savings is very close to 7 percent.
---------------------------------------------------------------------------

    \154\ Id.
    \155\ Id.
---------------------------------------------------------------------------

    However, NHTSA also seeks comment on whether a discount rate of 3 
percent would be more appropriate for this proposed rulemaking. OMB 
Circular A-4 also states that when regulation primarily and directly 
affects private consumption (e.g., through higher consumer prices for 
goods and services), instead of primarily affecting the allocation of 
capital, a lower discount rate may be appropriate. The alternative 
discount rate that is most appropriate in this case is the social rate 
of time preference, which refers to the rate at which society discounts 
future consumption to determine its value at the present time. The rate 
that savers are willing to accept to defer consumption into the future 
when there is no risk that borrowers will fail to pay them back offers 
one possible measure of the social rate of time preference. As noted 
above, the real rate of return on long-term government debt, which has 
averaged around 3 percent over the last 30 years, provides a reasonable 
estimate of this value.
    In the context of CAFE standards for motor vehicles, the 
appropriate discount rate depends on one's view of how the costs and 
benefits of more stringent standards are distributed between vehicle 
manufacturers and consumers. Given that the discount rate plays a 
significant role in determining the level of the standards under a 
``social optimization'' context, NHTSA conducted an analysis of what 
the standards and associated costs and benefits would be if the future 
benefits were discounted at 3 percent. The results of this analysis can 
be found in the PRIA. We estimated that following the same methods and 
criteria discussed below, but applying a 3 percent discount rate rather 
than a 7 percent discount rate, would suggest standards reaching about 
33.6 mpg (average required fuel economy among both passenger cars and 
light trucks) in MY2015, 2 mpg higher than the 31.6 mpg average 
resulting from the standards we are proposing based on a 7 percent 
discount rate. The more stringent standards during MY2011-MY2015 would 
reduce CO2 emissions by 672 million metric tons (mmt), or 29 percent 
more than the 521 mmt achieved by the proposed standards. On the other 
hand, we estimated that standards increasing at this pace would require 
about $85b in technology outlays during MY2011-MY2015, or 89 percent 
more than the $45b in technology outlays associated with the standards 
proposed today.
    Thus, although our proposed standards are based on a 7 percent 
discount rate, NHTSA seeks comment on whether it should set standards 
based on discount rate assumptions of 3 percent, instead of 7 percent.
o. Accounting for Uncertainty in Benefits and Costs
    In analyzing the uncertainty surrounding its estimates of benefits 
and costs from alternative CAFE standards, NHTSA has considered 
alternative estimates of those assumptions and parameters likely to 
have the largest effect. These include the projected costs of fuel 
economy-improving technologies and their expected effectiveness in 
reducing vehicle fuel consumption, forecasts of future fuel prices, the 
magnitude of the rebound effect, the reduction in external economic 
costs resulting from lower U.S. oil imports, the value to the U.S. 
economy of reducing carbon dioxide emissions, and the discount rate 
applied to future benefits and costs. The range for each of these 
variables employed in the uncertainty analysis is presented in the 
section of this document discussing each variable.
    The uncertainty analysis was conducted by assuming independent 
normal probability distributions for each of these variables, using the 
low and high estimates for each variable as the values below which 5 
percent and 95 percent of observed values are believed to fall. Each 
trial of the uncertainty analysis employed a set of values randomly 
drawn from each of these probability distributions, assuming that the 
value of each variable is independent of the others. Benefits and costs 
of each alternative standard were estimated using each combination of 
variables. A total of 1,000 trials were used to establish the likely 
probability distributions of estimated benefits and costs for each 
alternative standard.

B. How Has NHTSA Used the Volpe Model To Select the Proposed Standards?

1. Establishing a Continuous Function Standard
    NHTSA's analysis supporting determination of the proposed 
continuous function standard builds on the analysis that supported the 
determination of the standards in NHTSA's 2006 light truck final rule. 
That process involved three steps.\156\
---------------------------------------------------------------------------

    \156\ See 71 FR 17596-97 (Apr. 6, 2006) for a more complete 
discussion of this process.
---------------------------------------------------------------------------

    In ``phase one,'' NHTSA added fuel saving technologies to each 
manufacturer's fleet, model by model, for a model year until the net 
benefit from doing so reached its maximum value (i.e., until the 
incremental cost of improving its fuel economy further just equals the 
incremental value of fuel savings and other benefits from doing so). 
This was done for each of the seven largest manufacturers. Data points 
representing each vehicle's size and ``optimized'' fuel economy from 
the light truck fleets of those manufacturers were then combined into a 
single data set.
    In ``phase two,'' a preliminary continuous function was 
statistically fitted through these data points, subject to constraints 
at the upper and lower ends of the footprint range.
    Once a preliminary continuous function was statistically fitted to 
the data for a model year, ``phase three'' was performed. In that 
phase, the level of the function was adjusted to maximize net benefits, 
that is, the preliminary continuous function was raised or lowered 
until industry-wide (limited to the seven largest manufacturers) 
benefits were maximized.
    For NHTSA's 2006 light truck rulemaking, the optimization procedure 
was applied in its entirety only for MY 2011. The levels of the 
functions for MYs 2008-2010 were set at levels producing incremental 
costs approximately equivalent to those produced by the alternative 
Unreformed

[[Page 24417]]

CAFE standards promulgated for those model years in the same 
rulemaking.
    Analysis conducted by NHTSA to prepare for the current proposed 
rulemaking revealed several opportunities to refine the procedure 
described above before applying it to this action, which spans several 
model years. The resultant procedure is described below.
2. Calibration of Initial Continuous Function Standards
    For the optimized standards, the first step in the current 
procedure involves all three phases described above. Separately, for 
each of the seven largest manufacturers, the agency determined the 
level of additional technology that would maximize net benefits. The 
agency then combined the resultant fleets and used standard statistical 
analysis procedures to specify a continuous function (i.e., a function 
without abrupt changes) with asymptotes \157\ set at the average fuel 
economy levels of the smallest and largest vehicles in this 
``optimized'' fleet.\158\
---------------------------------------------------------------------------

    \157\ Some functions are not bounded. For example, a line that 
is not flat will increase in one direction without limit and will, 
in the other direction, decrease without limit. The continuous 
function applied by the agency is of a form with upper and lower 
boundaries. Even as vehicle footprint declines or increases, the 
function's value (in mpg or grams/mile) will never exceed or fall 
below a specific value. These upper and lower limits are called 
asymptotes.
    \158\ Consistent with EPCA, the passenger car and light truck 
fleets were analyzed separately. For passenger cars, the agency 
determined the asymptotes of the continuous function by calculating 
the average fuel economy of the smallest 8 percent and the largest 5 
percent of the fleet. For light trucks, the agency considered the 
smallest 11 percent and the largest 10 percent of the fleet. These 
cohorts were determined by identifying gaps in the distribution of 
vehicles according to footprint.
---------------------------------------------------------------------------

    In the 2006 light truck final rule, NHTSA created an attribute-
based fuel economy standard based upon a continuous function using a 
logistic curve. The 2006 rulemaking, and its antecedent advanced notice 
of proposed rulemaking, contain an extended discussion of alternative 
approaches, including a bin-based system and different potential 
curves. As discussed below, that final rule explains NHTSA's decision 
to promulgate a standard based on a logistic (``S shaped'') curve with 
constrained asymptotes (upper and lower limits).
    Although we did not explicitly discuss it in the MY 2008-2011 light 
truck rulemaking, NHTSA now wishes to explain that any continuous 
function with lower asymptotes, as was promulgated in the last 
rulemaking and is proposed in this rulemaking, provides an absolute 
lower fuel economy level which guards against manufacturers having an 
unlimited economic incentive to upsize their vehicles in order to lower 
their fuel economy requirement. As vehicle footprint continues to 
increase, decreases in the corresponding fuel economy target become 
progressively smaller, such that the target approaches but never 
reaches the value of the lower asymptote. Because the required level of 
CAFE is the harmonic average of targets applicable to a manufacturer's 
vehicle models, the value of the standard can approach but will never 
fall to the value of this lower asymptote, no matter how far the 
manufacturer's product mix shifts toward larger vehicles. This will 
limit any loss of fuel savings due to manufacturer decisions to upsize 
their vehicles.
    In a perfect world, NHTSA would develop the continuous functions 
for setting passenger car and light truck standards by letting the 
vehicle attribute (footprint) completely control the shape of the 
curves used for the functions in a way that provides the clearest 
observed relationship between this attribute and its fuel economy. But, 
NHTSA must balance many real world practical and public policy aspects 
in order to ensure that the standards are achieving the purpose set 
forth by EPCA and EISA. In developing the Agency's last light truck 
rule, the curve used to fit the data (attribute versus fuel economy) 
was a sales-weighted least-squares logistic curve. During this 
rulemaking, as NHTSA continued to look for ways to improve its standard 
setting methodology, consideration was given to other methods that 
could be used to develop the continuous functions. One such method that 
NHTSA explored and is using in this proposal is unweighted analysis of 
the data using the Mean Absolute Deviation (MAD) statistical procedure. 
Unweighted regression involves counting each vehicle model once, rather 
than as many times as vehicles included in that model are to be 
produced. MAD involves weighting deviations from predicted values based 
on their absolute rather than squared magnitude. As discussed below, 
NHTSA has tentatively concluded that, compared to sales-weighted least-
squares analysis, unweighted MAD is better suited to data with wide 
disparities in weight (i.e., sales volumes) and with many outliers.
    In establishing footprint-based CAFE standards, the agency does not 
have the sole objective of seeking to reflect a clear engineering 
relationship between footprint and fuel economy. Attributes other than 
footprint would be more closely correlated with fuel economy. The 
agency's objective is to make CAFE regulations more consistent with 
public policy goals, in particular (1) a rebalancing of requirements 
such that full-line manufacturers are not disproportionately burdened 
and (2) the establishment of an incentive that discourages 
manufacturers from responding to CAFE standards in ways that could 
compromise occupant protection and highway safety. While it is helpful 
that the attribute--in this case footprint--has an observed 
relationship to fuel economy, it is not necessary that this 
relationship be isolated from accompanying relationships (e.g., between 
weight and fuel economy) that can be better related to estimable 
physical processes. Similarly, it is more important that the functional 
form for the attribute-based standard yield desirable outcomes than 
that it singly seek a clear foundation in estimable physical processes.
    In general, public policy considerations and available vehicle data 
combine to suggest that the fuel economy standard should be generally 
downward sloping (on a fuel economy basis) with respect to NHTSA's 
chosen attribute, vehicle footprint. The arguments that favor an 
attribute-based system (maintaining consumer choice, protecting safety, 
more equitable distribution of costs, reducing the cost of regulation) 
all argue for a downward sloping curve. Larger vehicles should, in 
principle, have higher drag, weigh more, and therefore have greater 
inertia than otherwise identical smaller vehicles. Hence, all other 
factors remaining equal, larger vehicles should have lower fuel economy 
than smaller vehicles. Therefore, the selection of vehicle footprint as 
the reference attribute should produce downward sloping curves. Also, 
the tendency of larger vehicles to have lower fuel economy than smaller 
vehicles should provide some disincentive to shift to larger vehicles 
rather than adding technology; although doing so would tend to reduce 
the required CAFE level, it would also tend to reduce the achieved CAFE 
level.
    However, vehicle data, by itself, does not necessarily define what 
functional form that the curve ought to take. In the 2006 light truck 
rulemaking, NHTSA considered linear, quadratic, exponential, 
unconstrained logistic, and constrained logistic functions as possible 
alternatives. For light trucks, the various approaches produced broadly 
similar standards through the most commonly used vehicle sizes, but 
drastically different standards at the high and low ends of the range.
     Linear functions produced very high fuel economy standards 
for the

[[Page 24418]]

smallest vehicles, and low standards for the largest vehicles.
     The quadratic function generated a minimum at about 75 
square feet, and then perversely turned upward for vehicles with larger 
footprints. The standard for very small vehicles was unreasonably high.
     The exponential and unconstrained logistic functions 
produced unreasonably high standards for small vehicles, but flattened 
out for larger vehicles.
     The constrained logistic function provided a broadly 
linear downward-sloping through the most commonly used vehicle sizes, 
along with basically flat standards for very large and very small 
vehicles.
    On this basis, NHTSA believed that, while the data did not dictate 
a particular functional form, public policy considerations made the 
constrained logistic function particularly attractive. The 
considerations include:
     A relatively flat standard for larger vehicles acts as a 
de facto `backstop' for the standard in the event that future market 
conditions encourage manufacturers to build very large vehicles. 
Nothing prevents manufacturers from building larger vehicles. With a 
logistic curve, however, vehicles upsizing beyond some limit face a 
flat standard that is increasingly difficult to meet.
     A constrained logistic curve doesn't impose unachievable 
fuel economy standards on vehicles that have unusually small 
footprints, thus continuing to keep manufacturing fuel-efficient small 
vehicles available as a compliance option.
     A curve fitted without upper and lower constraints could 
reach very high fuel economy levels for small vehicles and very low 
fuel economy vehicles for large vehicles. While such a curve might 
produce similar required CAFE levels for the industry as a whole, it 
could have a particular adverse impact on manufacturers that specialize 
in very small vehicles, for example, two-seater sports cars. By the 
same token, it could require little or nothing of manufacturers 
specializing in very large vehicles.
     The transition from the `flat' portions of the curve to 
the `slope' portions of the curve is smooth and gradual, reducing the 
incentive for manufacturers to achieve compliance through marginal 
changes in vehicle size.
     The inflection points are set by the data and can 
potentially vary from year to year, rather than being chosen by NHTSA.
    On the other hand, a constrained logistic curve shares with other 
functional forms a risk of an excessively steep or excessively flat 
slope. The slope of the compliance curve may be considered as `too 
steep' for public policy purposes when manufacturers can achieve 
appreciable reductions in compliance costs by marginally increasing the 
size of a vehicle's footprint--e.g., the cost of compliance from 
upsizing is lower than other cost-effective compliance methods open to 
manufacturers.
    A slope is `too flat' for public policy purposes when it negates 
the advantages of an attribute-based system: Where the standard doesn't 
meaningfully vary with respect to changes in the underlying attribute, 
it cannot be said to be an attribute-based system within the meaning of 
the statute.
    NHTSA chose footprint as the best attribute for an attribute-based 
standard in part because we believed changing a vehicle's footprint 
would involve significant costs for manufacturers, probably requiring a 
redesign of the vehicle.
    While ``too steep'' or ``too flat'' inevitably cannot be defined 
with precision, they need to be kept in mind.
    For the proposed standards, the agency defined the continuous 
function using the following formula:
[GRAPHIC] [TIFF OMITTED] TP02MY08.007

Where:

T = the fuel economy target (in mpg)
a = the maximum fuel economy target (in mpg)
b = the minimum fuel economy target (in mpg)
c = the footprint value (in square feet) at which the fuel economy 
target is midway between a and b \159\
---------------------------------------------------------------------------

    \159\ That is, the midpoint.
---------------------------------------------------------------------------

d = the parameter (in square feet) defining the rate at which the 
value of targets decline from the largest to smallest values
e = 2.718\160\
---------------------------------------------------------------------------

    \160\ For the purpose of the Reformed CAFE standard, we are 
carrying e out to only three decimal places.
---------------------------------------------------------------------------

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

    NHTSA invites comment regarding the relative importance of the 
curve as a means of (1) providing a basis for describing the observed 
relationship between footprint and fuel economy, (2) providing a basis 
for describing a theoretical physical relationship (assuming one can be 
defined) between footprint and fuel economy, and (3) providing socially 
desirable incentives to manufacturers. The agency further invites 
comment on functional forms that would be consistent with each of these 
purposes.
    As for analysis of the light truck rule promulgated in 2006, NHTSA 
constrained this function by determining the maximum and minimum 
targets (a and b) and then holding those targets constant while using 
statistical techniques to fit the other two coefficients (c and d) in 
this equation.
    In the current analysis for passenger cars, the upper and lower 
asymptotes are based on the smallest three percent and largest four 
percent, respectively, of the fleet. These reflect footprint values 
defining distinct cohorts outside the bulk of the fleet, and correspond 
to footprint values of less than 39.5 square feet (i.e., up to the 
approximate size of a Honda Fit) and greater than 52.5 square feet 
(i.e., at least as great as the approximate size of a Toyota Avalon), 
respectively:

[[Page 24419]]

[GRAPHIC] [TIFF OMITTED] TP02MY08.008

    For light trucks, the upper asymptote (i.e., the highest mpg value 
of the continuous function defining fuel economy targets) is based on 
the smallest (in terms of footprint) eleven percent of the fleet, and 
the lower asymptote is based on the largest six percent of the fleet. 
These cohorts correspond to footprint values of less than 44.5 square 
feet (i.e., up to the approximate size of a Honda CR-V) and greater 
than 72.5 square feet (i.e., comprised primarily of extended vans and 
long-bed pickup trucks), respectively:

[[Page 24420]]

[GRAPHIC] [TIFF OMITTED] TP02MY08.009

    NHTSA invites comment on the identification of vehicle cohorts for 
purposes of establishing upper and lower limits (asymptotes) bounding 
the attribute-based standard. After updating its baseline market 
forecast in consideration of new product plan information from 
manufacturers, the agency plans to reevaluate these cohorts for both 
passenger cars and light trucks before promulgating a final rule, and 
notes that changes in approach could lead to changes in stringency.
    Given the above asymptotes, fitting the above functional form to 
the ``optimized'' passenger car fleet resulted in the following initial 
continuous functions:

[[Page 24421]]

[GRAPHIC] [TIFF OMITTED] TP02MY08.010

    For each model year, NHTSA then raised or lowered the resultant 
continuous function until net benefits were maximized for the seven 
largest manufacturers (in total). Without subsequent recalibrations 
discussed below, this produced the following continuous functions for 
passenger cars:

[[Page 24422]]

[GRAPHIC] [TIFF OMITTED] TP02MY08.011

    The agency followed the same procedures for setting light truck 
standards and doing so resulted in the following continuous functions:

[[Page 24423]]

[GRAPHIC] [TIFF OMITTED] TP02MY08.012

    In fitting the continuous function, NHTSA considered a range of 
statistical estimation techniques. In the 2006 light truck rulemaking, 
NHTSA estimated the parameters of the logistic function using fuel 
consumption (measured in gallons per mile) for each vehicle produced in 
a particular model year, weighted by sales.
    For this rulemaking, we observed that estimated fuel consumption 
functions for passenger cars were significantly affected by several 
outliers--a small number of popular vehicles that had significantly 
higher fuel economy than the fleet as a whole and, even more so, than 
vehicles of similar footprint. For passenger cars, the function, as 
estimated by weighted ordinary least squares, was exceptionally steep 
within the range considered. This observation, in turn, led NHTSA to 
consider alternative approaches to statistically fitting the continuous 
function.
    Among the options considered by NHTSA were the following: dropping 
the outlying vehicles from the estimation process, weighted and 
unweighted ordinary least squares, and weighted and unweighted mean 
absolute deviation (MAD). MAD is a statistical procedure that has been 
demonstrated to produce more efficient parameter estimates in the 
presence of significant outliers.\161\ As examples, the following two 
charts show the MY2015 passenger car and light truck fleets after the 
application of technologies to each manufacturer's fleet. These charts 
reveal numerous outliers for the passenger car fleet and, to a lesser 
extent, the light truck fleet:
---------------------------------------------------------------------------

    \161\ In the case of a dataset not drawn from a sample with a 
Gaussian, or normal, distribution, there is often a need to employ 
robust estimation methods rather than rely on least-squares approach 
to curve fitting. The least-squares approach has, as an underlying 
assumption, that the data are drawn from a normal distribution, and 
hence fits a curve using a sum-of-squares method to minimize errors. 
This approach will, in a sample drawn from a non-normal 
distribution, give excessive weight to outliers by making their 
presence felt in proportion to the square of their distance from the 
fitted curve, and, hence, distort the resulting fit. With outliers 
in the sample, the typical solution is to use a robust method such 
as a minimum absolute deviation, rather than a squared term, to 
estimate the fit (see, e.g., ``AI Access: Your Access to Data 
Modeling,'' at http://www.aiaccess.net/English/Glossaries/GlosMod/
e_gm_O_Pa.htm#Outlier). The effect on the estimation is to let 
the presence of each observation be felt more uniformly, resulting 
in a curve more representative of the data (see, e.g., Peter 
Kennedy, A Guide to Econometrics, 3rd edition, 1992, MIT Press, 
Cambridge, MA).

---------------------------------------------------------------------------

[[Page 24424]]

[GRAPHIC] [TIFF OMITTED] TP02MY08.013


[[Page 24425]]


[GRAPHIC] [TIFF OMITTED] TP02MY08.014

    NHTSA requests comment on the best method for statistically fitting 
the continuous function.
    There are good theoretical arguments for using an unweighted 
(rather than weighted) analysis. Although the purpose of the attribute-
based standard is to discourage downsizing (because of safety 
implications) and more equitably distribute compliance burdens among 
manufacturers, we strive to develop the curves based on the observed 
physical relationship between vehicle size (i.e., footprint) and fuel 
economy. The curve developed using unweighted sales data better 
reflects this relationship.
    However, the process by which we select the stringency (as distinct 
from the form) of the standard must consider sales volumes because the 
standards are based on sales-weighted average performance. Therefore, 
even if we use unweighted analysis develop the form of the standard, we 
would continue to evaluate the standard's stringency (and, therefore, 
its costs and benefits) based on sales-weighted average calculations 
done on a manufacturer-by-manufacturer basis.
    There is already precedent for using unweighted data to produce 
curves that are descriptive of engineering relationships. In NHTSA's 
Preliminary Regulatory Impact Analysis for FMVSS 216 roof crush 
standards, a series of force-versus-deflection curves were produced for 
individual vehicle models and then averaged together. In that case, the 
agency was seeking observed relationships that reflect engineering 
possibilities, rather than a profile of the existing sales fleet.
    In terms of relative emphasis on different vehicle models, the 
distinction between unweighted and weighted analysis is profound in the 
light vehicle market, in part because of the way ``models'' are defined 
for purposes of CAFE. The highest-selling passenger car model 
represents 356,000 units, and the lowest-selling model represents only 
5 units. As a group, the five lowest-selling models represent only 305 
units. Thus, weighted analysis places more than 1,000 times the 
emphasis on the highest-selling model than on the five lowest-selling 
models, and more than 70,000 times the emphasis than on the single 
lowest-selling model. The following histograms show the broader 
distributions of models and sales with respect to model-level sales 
(first for passenger cars, then for light trucks):

[[Page 24426]]

[GRAPHIC] [TIFF OMITTED] TP02MY08.015


[[Page 24427]]


[GRAPHIC] [TIFF OMITTED] TP02MY08.016

    For purposes of setting the stringency of the corporate average 
fuel economy standard, this is vital because enforcement is based on 
the sales-weighted average. However, for purposes of developing a curve 
intended to represent fuel economy levels achieved at a given 
footprint, weighted analysis effectively ignores many models.
    On the other hand, unweighted estimation is depending on the 
definition of a ``model''. Manufacturers will sometimes offer 
substantially similar vehicles with different badges (i.e., Ford 
Taurus/Mercury Sable) as two different models. The distinction between 
differing ``options packages'' on a single model and two distinct 
models is inevitably a bit blurry. When estimating fuel economy 
standards using a sales-weighted regression, this distinction is not 
material, since the estimation process will produce substantially the 
same results independently of the number of distribution of those sales 
into larger or smaller numbers of models. In unweighted estimation, 
however, dividing a particular vehicle family into a larger number of 
distinct models give that family some extra influence in the analysis. 
Nonetheless, considering that such parsing less than does sales 
weighting. NHTSA has tentatively concluded that unweighted estimation 
remains preferable to sales-weighted estimation, but invites comment on 
whether and, if so how substantially similar vehicles should be 
combined for purposes of fitting an attribute-based function when using 
unweighted estimation.
    The following charts show, for MY2015 passenger cars and light 
trucks, how the use of sales-weighted least-squares estimation compares 
to the proposed approach, which uses unweighted mean absolute 
deviation. For passenger cars, the curve resulting from proposed 
approach is somewhat shallower than the curve resulting from sales-
weighted least squares estimation. For light trucks, the curve 
resulting from proposed approach is somewhat steeper:
BILLING CODE 4910-59-P

[[Page 24428]]

[GRAPHIC] [TIFF OMITTED] TP02MY08.017

    NHTSA invites comment on the relative merits of unweighted and 
weighted estimation, as well as on the other curve fitting options 
(e.g., the use of mean absolute deviation) raised here. The agency 
plans to reevaluate curve fitting approaches for both passenger cars 
and light trucks before promulgating a final rule, and notes that 
changes in approach could lead to changes in stringency and impacts on 
different manufacturers.

[[Page 24429]]

[GRAPHIC] [TIFF OMITTED] TP02MY08.018

BILLING CODE 4910-59-C
3. Adjustments To Address Policy Considerations
    NHTSA believes that the resultant curve characteristics discussed 
above are empirically correct in that they correspond to the footprint 
and fuel economy values of the fleet obtained by adding fuel saving 
technologies to each manufacturer's fleet until the net benefit from 
doing so reached its maximum value.
    However, there are three issues (described above) which may tend to 
reduce the effectiveness of fuel economy regulation over time. These 
concerns are:
     Curve crossings;
     Excessive steepness of the passenger car curve;
     Risk of upsizing.
    In this rule, NHTSA proposes a solution to the curve crossing 
issue, requests comment on various methods of reducing the steepness of 
the passenger car, and examines the potential for upsizing generally 
under the provisions of this proposed rule.
a. Curve Crossings
    For both passenger cars and light trucks, NHTSA observed some curve 
crossings from one model year to the next (i.e., for the same 
footprint, some targets fell below the levels attained in the previous 
model year), as revealed in the above charts. The upper limit of the MY 
2012 passenger car curve falls slightly (about 0.1 mpg) below the MY 
2011 value. For light trucks, the lower asymptote in MY 2012 is 0.9 mpg 
below the lower asymptote in MY 2011. This was not observed during the 
last round of light truck rulemaking because reformed CAFE was fully 
implemented only in MY 2011. During the transition period (MYs 2008-
2010), the standards were set at levels equivalent in cost to 
unreformed CAFE. However, for this rulemaking, because the projected 
fleet composition changes between model years and the fuel economy 
target function is optimized in every model year, the initial 
continuous functions do not change monotonically (i.e., in only one 
direction--increasing) from year to

[[Page 24430]]

year at every footprint value. Given the availability of lead time and 
the importance of improving fuel economy, NHTSA has decided that, in 
the setting of the standards, we should ensure that the fuel economy 
targets do not fall from one year to the next at any footprint value.
    To address the year-to-year fluctuations in the functions, which 
may lead to these curve crossings, NHTSA recalibrated each continuous 
function to prevent it from crossing the continuous function from any 
previous model year. In doing so, the agency attempted to avoid 
continuous functions that would artificially encourage the product mix 
to approximate that of earlier years. Instead, the agency recalibrated 
by gradually shifting the initial continuous functions for each model 
year toward the initial continuous function determined above for the 
product mix for MY 2015. For both passenger cars and light trucks, the 
agency adjusted each of the four coefficients in the formula 
determining the continuous function such that regular steps were taken 
year by year between the values determined above for MY 2011 and those 
for MY 2015. For example, the inflection point (the coefficient 
determining the footprint at which the target falls halfway between its 
minimum and maximum values) defining the light truck target function 
was increased by 0.034 square feet annually from 51.9 square feet in MY 
2011 to 52.1 square feet in MY 2015.
    NHTSA also recalibrated the continuous function for each model year 
by adding, as needed, anti-backsliding constraints that prevent the 
function from either (a) yielding an industry wide average level of 
CAFE lower than that for the preceding model year, (b) for a given 
footprint, having targets that fall below the level of previous year, 
and (c) having an asymptote lower than that of the preceding model 
year. The ``decision tree'' for determining for each model year the 
need for each of these constraints is summarized below in Figure V 16.

[[Page 24431]]

[GRAPHIC] [TIFF OMITTED] TP02MY08.019

    The industry-wide average CAFE is prevented from decreasing between 
model years in order to prevent standards from falling below the level 
that was determined to be achievable for the model year before. To 
allow the industry-wide CAFE level to fall between successive model 
years would be to promulgate a standard that, notwithstanding 
maximizing net benefits, falls below what the agency has determined to 
be feasible in previous years. In a model year in which simple 
maximization of net benefits would have caused this to occur, NHTSA 
shifted the resultant curve upward (without changing the curve's shape) 
in order to produce an industry-wide CAFE equal to that of the 
preceding model year.
    Application of the decision tree shown above results in the 
following target functions for passenger cars and light trucks, 
respectively. These target functions are identical to those shown below 
in Section VI, which discusses the standards proposed today by NHTSA:
BILLING CODE 4910-59-P

[[Page 24432]]

[GRAPHIC] [TIFF OMITTED] TP02MY08.020


[[Page 24433]]


[GRAPHIC] [TIFF OMITTED] TP02MY08.021

BILLING CODE 4910-59-C
b. Steep Curves for Pasenger Cars
    NHTSA has developed a set of attribute-based curves for passenger 
cars for this proposal consistent with the methodology used in the 
2008-2011 light duty truck rule. However, unlike the relatively 
gradually sloped curve related fuel economy to footprint for trucks, 
our analysis for cars when utilizing a constained logistic curve 
produces a comparatively steep ``S''-shaped curve for passenger cars. 
This occurs primarily because--unlike trucks--current passenger car 
sales include vehicles with a wide range of fuel economy spanning a 
relatively narrow footprint range. Consequently, there is a relatively 
steep curve applied to the middle range of footprint values with a more 
rapid change of slope in the tails to flatten curve and thus satisfy 
the constrained logistic functional form.
    In this rule, NHTSA is proposing a relatively ``steep'' curve. The 
agency has considered and experimented with several methods of reducing 
the steepness of the passenger car curve. However, each of these 
approaches has created challenges that may potentially be worse than 
the problem they are trying to cure. The Agency is questioning whether 
the steep slope portion of the curve could potentially motivate vehicle 
manufacturers to reduce their compliance obligation under the standard 
by slightly increasing its footprint when they redesign their vehicles. 
We do not know the extent to which this is a real problem, but the 
agency has considered this possibility and has worked to minimize 
steepness of the slope while maintaining the scientific integrity 
behind our methodology.
    However, any attempt to ``fix'' the steepness of the passenger car 
curve appears to come at a price: First, flattening the curve by any 
particular method will move the curve away from the actual vehicle 
data. Second, flatter curves are generally place greater compliance 
burdens on full-line manufacturers than comparatively stringent (in 
terms of average require CAFE) standards. Furthermore, NHTSA believes 
that this could increase the overall costs required to achieve a given 
amount of fuel savings and societal

[[Page 24434]]

benefits, and it increases the risk that NHTSA would need to return to 
a ``least capable manufacturer'' approach in order to ensure economic 
practicability. Doing so would likely reduce stringency, and reduce 
fuel savings. In deciding on a particular approach, NHTSA must balance 
the certainty of high costs and lost fuel savings through a less 
``efficient'' standard against the risk that the steepness of the curve 
might stimulate manufacturers to evade the standard over time by 
redesigning their vehicles over time.
    In proposing the steep curve for this rule, NHTSA has tentatively 
decided that the cures that we have identified come at too high a 
price, i.e., lost stringency or undesirable side effects. However, 
NHTSA requests comment on these and other potential solutions to reduce 
the steepness of the proposed car curves for passenger cars.
    Some of the approaches considered or tested by NHTSA include:
    Linear standards. When the fuel consumption of vehicles with added 
technologies is plotted against footprint, we note a roughly linear 
relationship over the existing range of footprint values. Hence, a 
simple alternative to the current constrained logistic function would 
be to estimate a linear form of the curve with the sales data. However, 
NHTSA is concerned that such an approach may result in very low fuel 
economy standards for the largest footprint vehicles, very high fuel 
economy standards for the smallest vehicles, and loss of the inherent 
backstop properties of the constrained logistic function.
    In addition, the slope of a line estimated through a ``cloud'' of 
data may be very sensitive to the exact characteristics of vehicles 
with the largest and smallest footprints. It may turn out that small 
changes in vehicle characteristics in the tails could shift the slope 
of a linear estimate. Further, it may be impossible to materially 
adjust the slope of a linear standard in future years without accepting 
curve crossing. The following two charts compare linear regression 
results for MY2015 to the curves proposed today by NHTSA. The result 
for passenger cars illustrates the concern regarding behavior at large 
and small footprints. Over the range of footprints in which light 
trucks are expected to be offered in MY2015, the result for light 
trucks shows less difference from the proposed curve.
[GRAPHIC] [TIFF OMITTED] TP02MY08.022


[[Page 24435]]


[GRAPHIC] [TIFF OMITTED] TP02MY08.023

    Constrained linear standards. Another possible approach would be to 
retain the flattened tails proposed today but reduce the steepness of 
the middle portion by allowing it to directly reflect a linear 
relationship. This approach could be likened to a simplification or 
linearization of the constrained logistic function. The same minima and 
maxima would be used to bound the vertical extent of the linear form. 
The following two charts suggest that, at least for the MY2015 
passenger car and light truck fleets considered today, a constrained 
linear standard would, compared to the standard proposed today, likely 
result in a similar distribution of compliance burdens among 
manufacturers (because the stringency at each footprint would be 
similar):
BILLING CODE 4910-59-P

[[Page 24436]]

[GRAPHIC] [TIFF OMITTED] TP02MY08.024


[[Page 24437]]


[GRAPHIC] [TIFF OMITTED] TP02MY08.025

BILLING CODE 4910-59-C
    However, the agency remains concerned that the slope could exhibit 
greater year-to-year variation than the proposed logistic form 
(although further analysis would be required in order to address this 
concern). Also, as discussed in the preamble to the 2006 Federal 
Register notice regarding light truck CAFE standards, the agency 
remains concerned that the upper and lower ``kinks'' in the function 
could offer unexpected incentives for manufacturers to redesign 
vehicles with footprints close to the kink-point.
    Dual Attribute Approaches. A third possible solution would be to 
use additional attribute-based information to spread out the 
distribution of passenger cars across the x-axis. In effect, this 
approach uses a second attribute to normalize the footprint-fuel 
economy relationship. This second attribute might be horsepower, 
weight, or horsepower-to-weight.
    In analyzing the expected passenger car market, NHTSA observes that 
the ratio of engine horsepower to vehicle weight generally increases 
with increasing footprint. Higher power-to-weight ratios tend to imply 
lower fuel economy, as the engine is typically larger and operating 
less efficiently under driving conditions applicable to certification. 
Thus, the fuel consumption versus footprint curves for passenger cars 
reflect this relationship. For trucks, there does not appear to be a 
relationship between footprint and the power-to-weight ratio. For 
passenger cars, then, adjusting fuel consumption values to normalize 
for differences in power-to-weight ratio may produce a flatter curve 
providing less of an upsizing incentive for middle footprint values.
    NHTSA has experimented with normalizing footprint by horsepower-to-
weight ratio. The result was a nearly flat standard with respect to 
footprint across the most popular size ranges. This did not appear to 
deliver the benefits of an attribute-based system. In addition, it 
involves significant downward adjustments to the fuel economy of hybrid 
electric vehicles (such as the Toyota Prius), for which the engine is 
not the sole source of motive power. Also, it involves significant 
upward adjustments to the fuel economy of

[[Page 24438]]

vehicles with high power-to-weight ratios (such as the Chevrolet 
Corvette). Some of these upward and downward adjustments are large 
enough to suggest radical changes in the nature of the original 
vehicles. Furthermore, insofar as such normalization implies that NHTSA 
should adopt a two-attributed standard (e.g., in which the target 
depends on footprint and power-to-weight ratio), it may be challenging 
and time consuming to come up with a sufficiently precise vehicle-by-
vehicle definition of horsepower or horsepower-to-weight to be used for 
regulatory purposes.
[GRAPHIC] [TIFF OMITTED] TP02MY08.026

    Shape Based on Combined Fleet. A fourth possible solution would be 
to combine the passenger car and light truck fleet to determine the 
shape of the constrained logistic curve, and then determine the 
stringency (i.e., height) of that curve separately for each fleet. On 
one hand, this approach would base the curve's shape on the widest 
available range of information. On the other, the resultant initial 
shape for each fleet would be based on vehicles from the other fleet. 
For example, the initial shape applied to passenger cars would be 
based, in part, on large SUVs and pickup trucks, and the initial shape 
applied to light trucks would be based, in part, on subcompact cars. 
Stringency would still be determined separately for passenger cars and 
light trucks. NHTSA invites comments on the consistency of this 
approach with the requirement in EPCA to establish separate standards 
for passenger cars and light trucks.
    NHTSA performed a preliminary analysis of this approach. 
Considering the very wide range of fuel consumption levels in the 
combined fleet, NHTSA developed the asymptotes based on the average 
fuel consumption of all passenger cars and light trucks, respectively, 
rather than on the smallest passenger cars and the largest light 
trucks. The resultant MY2015 curve, shown below, is similar in 
curvature to the proposed curve for passenger cars

[[Page 24439]]

and notably steeper than the proposed curve for light trucks.
BILLING CODE 4910-59-P
[GRAPHIC] [TIFF OMITTED] TP02MY08.027


[[Page 24440]]


[GRAPHIC] [TIFF OMITTED] TP02MY08.028

    Ignoring Outliers. A fifth possible solution would be to ignore 
outliers (data points that are unique and skew the curve). Lacking an 
objective means of classifying specific vehicle models as outliers that 
should be excluded from the analysis, NHTSA explored the possibility of 
excluding all hybrid electric vehicles (HEVs). The Japanese government 
also excluded HEVs for purposes of developing Japan's light vehicle 
efficiency standards. However, doing so yields initial curves of shapes 
similar to those proposed, but displaced slightly in the direction of 
lower fuel consumption. The similarity of the shapes of these curves 
suggests that optimization against the full fleet (with HEVs) would 
produce standards whose stringency is similar to that of those proposed 
today.
BILLING CODE 4910-59-P

[[Page 24441]]

[GRAPHIC] [TIFF OMITTED] TP02MY08.029


[[Page 24442]]


[GRAPHIC] [TIFF OMITTED] TP02MY08.030

BILLING CODE 4910-59-C

    NHTSA invites comments on the importance of addressing the relative 
steepness of the proposed curves for passenger cars, and on the 
feasibility of, technical basis for, and implications of any options 
for doing so. The agency plans to reevaluate standards for both 
passenger cars and light trucks before promulgating a final rule, and 
notes that changes in approach--including measures to address the 
steepness of the passenger car curves--could lead to changes in 
stringency as well as different impacts on different manufacturers.
c. Risk of Upsizing
    The steepness of the proposed curve for passenger cars presents a 
localized risk that manufacturers will respond in ways that compromise 
expected fuel savings. That is, although the constrained logistic curve 
has a steep region, that region does not cover a wide range of 
footprints. However, any attribute-based system involves the broader 
risk that manufacturers will shift toward vehicles with the lowest fuel 
economy targets to the extent that upsizing can be accomplished 
sufficiently cheaply and without so much weight increase as to nullify 
the effect of a lower target. As mentioned above, the constrained 
logistic curve proposed by NHTSA provides an absolute floor. That is, 
even if manufacturers discontinue all but the very largest known 
passenger cars and light trucks, they would still be required to meet 
CAFE standards no lower than the lower asymptote (on an mpg basis) of 
the constrained logistic curve. Also, for domestic passenger cars, EISA 
establishes a floor or ``backstop'' equal to 92 percent of the average 
required CAFE level for passenger cars. This backstop is discussed 
below in Section VI.
    It is difficult to assess the risk that manufacturers may shift the 
mix of vehicles enough to approach the EISA floor for domestic 
passenger cars, or to approach the lower asymptotes for light trucks or 
imported passenger cars. However, considering the footprint 
distribution of vehicles (as indicated by the various histograms and 
scatter plots shown above in this section) expected to be covered by 
the proposed rule,

[[Page 24443]]

NHTSA anticipates that manufacturers would not be able to approach 
these reductions in stringency without dramatically altering product 
mix. The agency doubts that manufacturers could do so unless consumer 
preferences for larger vehicles also shift dramatically.
    NHTSA also notes that under attribute-based CAFE standards such as 
the agency is proposing today, shifts in consumer preferences could 
cause manufacturers' required CAFE levels and, therefore, achieved fuel 
savings (and perhaps costs) to increase. For example, if changes in 
fuel prices combine with demographic and/or other factors to cause 
market preferences to shift significantly toward vehicles with smaller 
footprints, manufacturers shifting (relative to current estimates) in 
that direction will face higher required CAFE levels than the agency 
has estimated.

VI. Proposed Fuel Economy Standards

A. Standards for Passenger Cars and Light Trucks

    For both passenger cars and light trucks, the agency is proposing 
CAFE standards estimated, as for the previously-promulgated reformed MY 
2008-2011 light truck standards, to maximize net benefits to society. 
However, as discussed in Section V, the agency considered and analyzed 
modified approaches to calibrating the continuous function and fitting 
the data in order to address characteristics of the data (vehicles with 
outlying fuel economy, footprint, and or sales), and to address the 
issues of backsliding, steepness of the curve, and curve crossings from 
one model year to the next. While the agency is proposing the curves 
below, we continue to be concerned about the steepness of the passenger 
car curve and about gaming potential and are seeking comments on 
different approaches to address the steepness, as discussed in Section 
V. The proposed curves below and their respective shapes are calibrated 
using unweighted mean absolute deviation (MAD) regression and 
determined through a gradual transformation of curves to guard against 
erratic fluctuations and through a series of anti-backsliding measures 
that prevents the average required CAFE level from falling between 
model years and prevents the continuous function for a given model from 
crossing or falling below that of the preceding model year. These 
refinements are discussed in greater detail in Section V of the notice.
1. Proposed Passenger Car Standards MY 2011-2015
    We have tentatively determined that the proposed standards for MY 
2011-2015 passenger cars would result in required fuel economy levels 
that are technologically feasible, economically practicable, and set by 
taking into account both the effect of other motor vehicle standards of 
the Government on fuel economy and the need of the United States to 
conserve energy. Values for the parameters defining the target 
functions defining these proposed standards for cars are as follows:

----------------------------------------------------------------------------------------------------------------
                                                                            Model year
                   Parameter                    ----------------------------------------------------------------
                                                     2011         2012         2013         2014         2015
----------------------------------------------------------------------------------------------------------------
a..............................................         38.2         40.0         40.8         41.2         41.7
b..............................................         25.9         27.4         28.7         29.9         31.2
c..............................................         45.9         45.8         45.7         45.6         45.5
d..............................................          1.6          1.5          1.5          1.4          1.4
----------------------------------------------------------------------------------------------------------------

Where, per the adjusted continuous function formula above in Section 
V:
    a = the maximum fuel economy target (in mpg)
    b = the minimum fuel economy target (in mpg)
    c = the footprint value (in square feet) at which the fuel 
economy target is midway between a and b
    d = the parameter (in square feet) defining the rate at which 
the value of targets decline from the largest to smallest values

    The resultant target functions have the following shapes:
    Based on the product plan information provided by manufacturers in 
response to the February 2007 request for information and the 
incorporation of publicly available supplemental data and information, 
NHTSA has estimated the required average fuel economy levels under the 
proposed adjusted standards for MYs 2011-2015 as follows:

[[Page 24444]]

[GRAPHIC] [TIFF OMITTED] TP02MY08.031


                           Table VI-1.--Required CAFE Levels (mpg) for Passenger Cars
----------------------------------------------------------------------------------------------------------------
                  Manufacturer                     MY 2011      MY 2012      MY 2013      MY 2014      MY 2015
----------------------------------------------------------------------------------------------------------------
BMW............................................         33.3         35.0         36.0         36.8         37.7
Chrysler.......................................         28.7         29.3         32.2         32.6         33.6
Ferrari........................................         30.4         32.0         33.1         33.9         34.9
Ford...........................................         31.0         32.7         33.7         34.5         35.5
Fuji (Subaru)..................................         36.9         38.7         39.6         40.1         40.8
General Motors.................................         30.0         31.7         32.8         33.7         34.7
Honda..........................................         32.1         33.8         34.8         35.5         36.4
Hyundai........................................         33.4         35.1         36.0         36.7         37.5
Lotus..........................................         38.1         40.0         40.8         41.2         41.7
Maserati.......................................         28.9         30.6         31.8         32.8         34.0
Mercedes.......................................         31.7         33.3         34.4         35.3         36.2
Mitsubishi.....................................         33.0         35.1         35.9         37.0         37.9
Nissan.........................................         31.2         33.2         34.2         35.0         35.9
Porsche........................................         37.6         39.4         40.3         40.7         41.3
Suzuki.........................................         37.3         39.2         40.1         40.6         41.2
Toyota.........................................         30.1         31.5         32.7         33.6         34.6
Volkswagen.....................................         35.4         37.2         38.2         38.8         39.5
                                                ----------------------------------------------------------------
    Total/Average..............................         31.2         32.8         34.0         34.8         35.7
----------------------------------------------------------------------------------------------------------------


[[Page 24445]]

2. Proposed Standards for Light Trucks MY 2011-2015
    NHTSA is proposing light truck fuel economy standards for MYs 2011 
through 2015. In taking a fresh look at what truck standard should be 
established for MY 2011, as required by EISA, NHTSA used the newer set 
of assumptions that it had developed for the purpose of this 
rulemaking. These assumptions differ from those used by the agency in 
setting the MY 2008-2011 light truck standards in early 2006, and 
result in an increase in the projected overall average fuel economy for 
MY 2011. The agency used the most up-to-date EIA projections for 
available gasoline prices. These projections are, on average, at 
approximately $0.25 per gallon higher than the projections used in the 
last light truck rulemaking. Other differences in assumptions include 
more current product plan information (i.e., spring 2007 product plans 
reflecting persistently higher fuel prices, instead of the fall 2005 
plans used in the 2006 final rule), an updated technology list and 
updated costs estimates and penetration rates for technologies, and 
updated values for externalities such as energy security and placing a 
value of carbon dioxide emission reductions.
    NHTSA is proposing ``optimized'' standards for MY 2011-2015 light 
trucks, the process for establishing which is described at length 
above, but which may be briefly described as maximizing net social 
benefits plus anti-backsliding measures. We have tentatively determined 
that the proposed light truck standards for MYs 2011-2015 represent the 
maximum feasible fuel economy level for that approach. In reaching this 
tentative conclusion, we have balanced the express statutory factors 
and other relevant considerations, such as safety and effects on 
employment, and we will also consider our NEPA analysis in the agency's 
final action.
    The proposed standards are determined by a continuous function 
specifying fuel economy targets applicable at different vehicle 
footprint sizes, the equation for which is given above in Section V 
Values for the parameters defining the target functions defining these 
proposed standards for light trucks are as follows:

----------------------------------------------------------------------------------------------------------------
                                                                            Model year
                   Parameter                    ----------------------------------------------------------------
                                                     2011         2012         2013         2014         2015
----------------------------------------------------------------------------------------------------------------
A..............................................         30.9         32.7         34.1         34.1         34.3
B..............................................         21.5         22.8         23.8         24.3         24.8
C..............................................         51.9         52.0         52.0         52.1         52.1
D..............................................          3.8          3.8          3.8          3.9          3.9
----------------------------------------------------------------------------------------------------------------

Where:
    a = the maximum fuel economy target (in mpg)
    b = the minimum fuel economy target (in mpg)
    c = the footprint value (in square feet) at which the fuel 
economy target is midway between a and b
    d = the parameter (in square feet) defining the rate at which 
the value of targets decline from the largest to smallest values

    The resultant target functions have the following shapes:

[[Page 24446]]

[GRAPHIC] [TIFF OMITTED] TP02MY08.032

    Based on the product plans provided by manufacturers in response to 
the February 2007 request for information and the incorporation of 
publicly available supplemental data and information, the agency has 
estimated the required average fuel economy levels under the proposed 
optimized standards for MYs 2011-2015 as follows:

                            Table VI-2.--Required CAFE Levels (mpg) for Light Trucks
----------------------------------------------------------------------------------------------------------------
                  Manufacturer                     MY 2011      MY 2012      MY 2013      MY 2014      MY 2015
----------------------------------------------------------------------------------------------------------------
BMW............................................         28.2         29.9         31.2         31.4         31.7
Chrysler.......................................         25.2         26.6         28.0         28.5         29.1
Ford...........................................         24.7         26.1         28.0         28.3         28.8
Fuji (Subaru)..................................         30.0         31.7         33.1         33.2         33.4
General Motors.................................         23.9         25.4         26.5         27.0         27.4
Honda..........................................         26.1         27.7         28.9         29.2         29.6
Hyundai........................................         27.5         29.1         30.4         30.6         31.0
Mercedes.......................................         28.4         30.1         31.4         31.6         31.9
Mitsubishi.....................................         29.4         30.8         32.2         32.3         32.6
Nissan.........................................         24.9         26.2         27.3         27.7         28.2
Porsche........................................         25.9         27.4         28.7         29.0         29.4
Suzuki.........................................         30.3         32.1         33.5         33.5         33.7
Toyota.........................................         24.9         26.0         27.2         27.6         28.0
Volkswagen.....................................         26.2         27.8         29.0         29.3         29.7
----------------------------------------------------------------------------------------------------------------

[[Page 24447]]


    Total/Average..............................         25.0         26.4         27.8         28.2         28.6
----------------------------------------------------------------------------------------------------------------

    We recognize that the manufacturer product plans that we used in 
developing the manufacturers' required fuel economy levels for both 
passenger cars and light trucks will be updated in some respects before 
the final rule is published. To that end, the agency is publishing a 
separate request for product plans at the same time as this NPRM to 
obtain whatever updates have been made already. Further, we note that a 
manufacturer's required fuel economy level for a model year under the 
adjusted standards would be based on its actual production numbers in 
that model year. Therefore, its official required fuel economy level 
would not be known until the end of that model year. However, because 
the targets for each vehicle footprint would be established in advance 
of the model year, a manufacturer should be able to estimate its 
required level accurately and develop a product plan that would comply 
with that level.
3. Energy and Environmental Backstop
    EISA requires each manufacturer to meet a minimum fuel economy 
standard for domestically manufactured passenger cars in addition to 
meeting the standards set by NHTSA. 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. * * *'' \162\ The 
agency must publish the projected minimum standards in the Federal 
Register when the passenger car standards for the model year in 
question are promulgated.
---------------------------------------------------------------------------

    \162\ 49 U.S.C. 32902(b)(4).
---------------------------------------------------------------------------

    NHTSA calculated 92 percent of the proposed projected passenger car 
standards as the minimum standard, which is presented below. The 
calculated minimum standards will be updated for the final rule to 
reflect any changes in the projected passenger car standards.

------------------------------------------------------------------------
                                                               Minimum
                         Model year                            standard
------------------------------------------------------------------------
2011.......................................................         28.7
2012.......................................................         30.2
2013.......................................................         31.3
2014.......................................................         32.0
2015.......................................................         32.9
------------------------------------------------------------------------

    The agency would like to note that EISA requires the minimum 
domestic passenger car standard to be the greater of 27.5 mpg or the 
calculated 92 percent, the calculated minimum standard. In all five 
model years, the percentage-based value exceeded 27.5 mpg. We also note 
that the minimum standards apply only to domestically manufactured 
passenger cars, not to non-domestically manufactured passenger cars or 
to light trucks.
    In CBD, the Ninth Circuit agreed with the agency that EPCA, as it 
was then written, did not explicitly require the adoption of a 
backstop, i.e., a minimum CAFE standard that is fixed. A fixed minimum 
standard is one that does not change in response to changes in a 
manufacturer's vehicle mix.
    The Court said, however, that the issue was not whether the 
adoption was expressly required, but whether it was arbitrary and 
capricious for the agency to decline to adopt a backstop. The Court 
said that Congress was silent in EPCA on this issue. The Court 
concluded that it was arbitrary and capricious for the agency to 
decline to adopt a backstop because it did not, in the view of the 
Court, address the statutory factors for determining the maximum 
feasible level of average fuel economy.
    NHTSA believes that it considered and discussed the express 
statutory factors such as technological feasibility and economic 
practicability and related factors such as safety in deciding not to 
adopt a backstop. We do not believe that further discussion is 
warranted because Congress has spoken directly on this issue since the 
Ninth Circuit's decision.
    The enactment of EISA resolved this issue. Congress expressly 
mandated that CAFE standards for automobiles be attribute-based. That 
is, they must be based on an attribute related to fuel economy, e.g., 
footprint and they must adjust in response to changes in vehicle mix. 
Taken by itself, this mandate precludes the agency from adopting a 
fixed minimum standard. The only exception to that mandate is the 
provision in which Congress mandated a fixed and flat \163\ minimum 
standard for one of the three compliance categories. It required one 
for domestic passenger cars, but not for either nondomestic passenger 
cars or light trucks.
---------------------------------------------------------------------------

    \163\ A flat standard is one that requires each manufacturer to 
achieve the same numerical level of CAFE.
---------------------------------------------------------------------------

    Given the clarity of the requirement for attribute-based standards 
and the equally clear narrow exception to that requirement, the agency 
tentatively concludes that had Congress intended backstops to be 
established for either of the other two compliance categories, it would 
have required them. Congress did not, however, do so. Absent explicit 
statutory language that provides the agency authority to set flat 
standards, the agency believes that the setting of a supplementary 
minimum flat standard for the other two compliance categories would be 
contrary to the requirement to set an attribute-based standard under 
EISA.
    Regardless, the agency notes that the curve of an attribute-based 
standard has features that limit backsliding. Some of these features, 
which are fully described in Section V.B of the notice, were added as 
the agency refined and modified the Volpe model for the purpose of this 
rulemaking. Others, such as the lower asymptote, which serves as a 
backstop, are inherent in the logistic function. We believe that these 
features help address the concern that has been expressed regarding the 
possibility of vehicle upsizing without compromising the benefits of 
reform. In addition, the agency notes that the 35 mpg requirement in 
and of itself serves as a backstop. The agency must set the standards 
high enough to ensure that the average fuel economy level of the 
combined car and light fleet is making steady progress toward and 
achieves the statutory requirement of at least 35 mpg by 2020. If the 
agency finds that this requirement might not be achieved, it will 
consider setting standards for model years 2016 through 2015 early 
enough and in any event high enough to ensure reaching the 35 mpg 
requirement.
4. Combined Fleet Performance
    The combined industry wide average fuel economy (in miles per 
gallon, or mpg) levels for both cars and light

[[Page 24448]]

trucks, if each manufacturer just met its obligations under the 
proposed ``optimized'' standards for each model year, would be as 
follows:

MY 2011: 27.8 mpg
MY 2012: 29.2 mpg
MY 2013: 30.5 mpg
MY 2014: 31.0 mpg
MY 2015: 31.6 mpg

    The annual average increase during this five year period is 
approximately 4.5 percent. Due to the uneven distribution of new model 
introductions during this period and to the fact that significant 
technological changes can be most readily made in conjunction with 
those introductions, the annual percentage increases are greater in the 
early years in this period. In order for the combined industry wide 
average fuel economy to reach at least 35 mpg by MY 2020, it would have 
to increase an average of 2.1 percent per year for MYs 2016 through 
2020.

B. Estimated Technology Utilization Under Proposed Standards

    NHTSA anticipates that manufacturers will significantly increase 
the use of fuel-saving technologies in response to the standards we are 
proposing for passenger cars. Although it is impossible to predict 
exactly how manufacturers will respond, the Volpe model provides 
estimates of technologies manufacturers could apply in order to comply 
with the proposed standards. The preliminary Regulatory Impact Analysis 
(PRIA) presents estimated increases in the industry-wide utilization of 
each technology included in agency's analysis. Tables VI-3 and VI-4 
show rates at which the seven largest manufacturers' product plans 
indicated plans to use some selected technologies, as well as rates at 
which the Volpe model estimated that the same technologies might 
penetrate these manufacturers' passenger car fleet in response to the 
baseline and proposed standards.
    The average penetration rate is the percentage of the entire fleet 
to which the technology is applied. For example, tables VI-3 and VI-4 
show that these manufacturers could apply hybrid powertrains to 15 
percent of the entire passenger car fleet in MY 2015, as opposed to the 
5 percent shown in their product plans. However, not all manufacturers 
begin with the same technology penetration rates, and not all 
manufacturers are affected equally by the proposed standards. The next 
column shows the maximum penetration rate among the seven manufacturers 
with a significant market share (Chrysler, Ford, GM, Honda, Hyundai, 
Nissan, and Toyota). For example, the Volpe model estimated that one of 
these manufacturers would apply hybrid powertrains to 19 percent of its 
passenger car fleet to comply with the proposed MY 2015 standard.
    As tables VI-3 and VI-4 demonstrate, the Volpe model estimated that 
manufacturers might need to apply significant numbers of advanced 
engines, advanced transmissions, and hybrid powertrains in order to 
comply with the proposed standards. (Most of the hybrids are integrated 
starter generators, although significant numbers of IMA and power-split 
hybrids also penetrate the fleet.) For example, the Volpe model 
estimated that one of the seven largest light truck manufacturers could 
be including diesel engines in 45 percent of its light trucks by MY2015 
in response to the proposed standards.

                                    Table VI.--3. Estimated Technology Penetration Rates in MY2015 for Passenger Cars
                                                                      [In percent]
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                             Average among seven largest manufacturers       Maximum among seven largest manufacturers
                                                         -----------------------------------------------------------------------------------------------
                       Technology                                                              Under                                           Under
                                                           Product plan      Adjusted        proposed      Product plan      Adjusted        proposed
                                                                             baseline        standard                        baseline        standard
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                     Passenger Cars
--------------------------------------------------------------------------------------------------------------------------------------------------------
Automatically Shifted Manual Transmission...............              10              10              39              59              59              86
Spark Ignited Direct Injection..........................              22              22              30              76              76              82
Turbocharging & Engine Downsizing.......................               5               5              17              11              11              51
Diesel Engine...........................................               0               0               2               0               0               5
Hybrid Electric Vehicles................................               5               5              15              14              14              19
--------------------------------------------------------------------------------------------------------------------------------------------------------


                                     Table VI.--4. Estimated Technology Penetration Rates in MY2015 for Light Trucks
                                                                      [In percent]
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                             Maximum among seven largest manufacturers       Maximum among seven largest manufacturers
                                                         -----------------------------------------------------------------------------------------------
                       Technology                                            Adjusted     Under proposed                     Adjusted     Under proposed
                                                           Product plan      baseline         standard     Product plan      baseline         standard
--------------------------------------------------------------------------------------------------------------------------------------------------------
Automatically Shifted Manual Transmission...............              10              14              55              41              41              72
Spark Ignited Direct Injection..........................              23              24              40              46              46              73
Turbocharing & Engine Downsizing........................               9              11              31              32              32              44
Diesel Engine...........................................               3               6              10               7              29              45
Hybrid Electric Vehicles................................               2               6              25               5              13              32
--------------------------------------------------------------------------------------------------------------------------------------------------------

    The agency uses Volpe model analysis of technology application 
rates as a way of determining the economic practicability and 
technological feasibility of the proposed standards, but we note that 
manufacturers may always comply with the standards by applying 
different technologies in different orders and at different rates.

[[Page 24449]]

Insofar as our conclusion of what the maximum feasible standards would 
be is predicated on our analysis, however, the agency requests comment 
on the feasibility of these rates of increase in the penetration of 
these advanced technologies, and for other technologies discussed in 
the PRIA.

C. Benefits and Costs of Proposed Standards

1. Benefits
    We estimate that the proposed standards for passenger cars would 
save approximately 19 billion gallons of fuel and prevent 178 billion 
metric tons of tailpipe CO2 emissions over the lifetime of 
the passenger cars sold during those model years, compared to the fuel 
savings and emissions reductions that would occur if the standards 
remained at the adjusted baseline (i.e., the higher of manufacturer's 
plans and the manufacturer's required level of average fuel economy for 
MY 2010).\164\
---------------------------------------------------------------------------

    \164\ See supra text accompanying note 103.
---------------------------------------------------------------------------

    We estimate that the value of the total benefits of the proposed 
passenger car standards would be approximately $31 billion \165\ over 
the lifetime of the 5 model years combined. This estimate of societal 
benefits includes direct impacts from lower fuel consumption as well as 
externalities, and also reflects offsetting societal costs resulting 
from the rebound effect. Direct benefits to consumers, including fuel 
savings, account for 85 percent ($29.5 billion) of the roughly $35 
billion in gross \166\ consumer benefits resulting from increased 
passenger car CAFE. Petroleum market externalities account for roughly 
10 percent ($3.6 billion). Environmental externalities, i.e., reduction 
of air pollutants accounts for roughly 5 percent ($1.8 billion). Over 
half of this $1.8 billion figure is the result of greenhouse gas 
(primarily CO2) reduction ($1.0 billion). Increased 
congestion, noise and accidents from increased driving will offset 
roughly $3.8 billion of the $35 billion in consumer benefits, leaving 
net consumer benefits of $31 billion.
---------------------------------------------------------------------------

    \165\ The $31 billion estimate is based on a 7% discount rate 
for valuing future impacts. NHTSA estimated benefits using both 7% 
and 3% discount rates. Under a 3% rate, total consumer benefits for 
passenger car CAFE improvements total $36 billion.
    \166\ Gross consumer benefits are benefits measured prior to 
accounting for the negative impacts of the rebound effect. They 
include fuel savings, consumer surplus from additional driving, 
reduced refueling time, reduced criteria pollutants, and reduced 
greenhouse gas production. Negative impacts from the rebound effect 
include added congestion, noise, and crash costs due to additional 
driving.
---------------------------------------------------------------------------

    The following table sets out the relative dollar value of the 
various benefits of this rulemaking on a per gallon saved basis:

    Table VI-5.--Economic Benefits and Costs per Gallon of Fuel Saved
                             [Undiscounted]
------------------------------------------------------------------------
                                                          Value  (2006 $
            Category                     Variable           per gallon)
------------------------------------------------------------------------
Benefits.......................  Savings in Fuel                   $1.99
                                  Production Cost.
                                 Reduction in Oil Import             .28
                                  Externalities.
                                 Value of Additional                 .24
                                  Rebound-Effect Driving.
                                 Reduction in Criteria               .16
                                  Pollutant Emissions.
                                 Value of Reduced                    .12
                                  Refueling Time.
                                 Reduction in CO2              \167\ .02
                                  Emissions.
                                                         ---------------
                                 Gross Benefits.........            2.81
Costs..........................  Externalities from                 0.30
                                  Additional Rebound-
                                  Effect Driving.
                                                         ---------------
Net Benefits...................  Net Benefits...........            2.51
------------------------------------------------------------------------

    We estimate that the proposed standards for light trucks would save 
approximately 36 billion gallons of fuel and prevent 343 million metric 
tons of tailpipe CO2 emissions over the lifetime of the 
light trucks sold during those model years, compared to the fuel 
savings and emissions reductions that would occur if the standards 
remained at the adjusted baseline.
---------------------------------------------------------------------------

    \167\ Based on a value of $7.00 per ton of carbon dioxide.
---------------------------------------------------------------------------

    We estimate that the value of the total benefits of the proposed 
light truck standards would be approximately $57 billion \168\ over the 
lifetime of the 5 model years of light trucks combined. This estimate 
of societal benefits includes direct impacts from lower fuel 
consumption as well as externalities and also reflects offsetting 
societal costs resulting from the rebound effect. Direct benefits to 
consumers, including fuel savings, account for 84 percent ($52.7 
billion) of the roughly $63 billion in gross consumer benefits 
resulting from increased light truck CAFE. Petroleum market 
externalities account for roughly 10 percent ($6.5 billion). 
Environmental externalities, i.e., reduction of air pollutants accounts 
for roughly 6 percent ($3.5 billion). Over half of this figure is the 
result of greenhouse gas (primarily CO2) reduction ($1.9 
billion). Increased congestion, noise and accidents from increased 
driving will offset roughly $5.4 billion of the $63 billion in consumer 
benefits, leaving net consumer benefits of $57 billion.
---------------------------------------------------------------------------

    \168\ The $57 billion estimate is based on a 7% discount rate 
for valuing future impacts. NHTSA estimated benefits using both 7% 
and 3% discount rates. Under a 3% rate, total consumer benefits for 
light truck CAFE improvements are $72 billion.
---------------------------------------------------------------------------

2. Costs
    The total costs for manufacturers just complying with the standards 
for MY 2011-2015 passenger cars would be approximately $16 billion, 
compared to the costs they would incur if the standards remained at the 
adjusted baseline. The resulting vehicle price increases to buyers of 
MY 2015 passenger cars would be recovered or paid back \169\ in 
additional fuel savings in an average of 56 months, assuming fuel 
prices ranging from $2.26 per gallon in 2016 to $2.51 per gallon in 
2030.\170\
---------------------------------------------------------------------------

    \169\ See Section V.A.7 below for discussion of payback period.
    \170\ The fuel prices (shown here in 2006 dollars) used to 
calculate the length of the payback period are those projected 
(Annual Energy Outlook 2008, revised early release) by the Energy 
Information Administration over the life of the MY 2011-2015 light 
trucks, not current fuel prices.
---------------------------------------------------------------------------

    The total costs for manufacturers just complying with the standards 
for MY 2011-2015 light trucks would be approximately $31 billion, 
compared to the costs they would incur if the standards remained at the 
adjusted

[[Page 24450]]

baseline. The resulting vehicle price increases to buyers of MY 2015 
light trucks would be paid back in additional fuel savings in an 
average of 50 months, assuming fuel prices ranging from $2.26 to $2.51 
per gallon.

Comparison of Estimated Benefits to Estimated Costs

    The table below compares the incremental benefits and costs for the 
car and light truck CAFE standards, in millions of dollars.

                                                               Table VI-6.--Passenger Cars
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                            Model year                                         Total
                                                         -----------------------------------------------------------------------------------------------
                                                               2011            2012            2013            2014            2015          2011-2015
--------------------------------------------------------------------------------------------------------------------------------------------------------
Benefits................................................           2,596           4,933           6,148           7,889           9,420          30,986
Costs...................................................           1,884           2,373           2,879           3,798           4,862          15,796
Net Benefits............................................             712           2,560           3,269           4,091           4,558          15,190
--------------------------------------------------------------------------------------------------------------------------------------------------------


                                                                Table VI-7.--Light Trucks
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                            Model Year                                         Total
                                                         -----------------------------------------------------------------------------------------------
                                                               2011            2012            2013            2014            2015          2011-2015
--------------------------------------------------------------------------------------------------------------------------------------------------------
Benefits................................................           3,909           8,779          13,560          14,915          16,192          57,355
Costs...................................................           1,649           4,986           7,394           8,160           8,761          30,949
Net Benefits............................................           2,260           3,793           6,166           6,755           7,431          26,406
--------------------------------------------------------------------------------------------------------------------------------------------------------

    The average annual per vehicle cost increases are shown in the 
PRIA.

D. Flexibility Mechanisms

    The agency's benefit and cost estimates do not reflect the 
availability and use of flexibility mechanisms, such as compliance 
credits and credit trading because EPCA prohibits NHTSA from 
considering the effects of those mechanisms in setting CAFE standards. 
EPCA has precluded consideration of the FFV adjustments ever since it 
was amended to provide for those adjustments. The prohibition against 
considering compliance credits was added by EISA.
    The benefit and compliance cost estimates used by the agency in 
determining the maximum feasible level of the CAFE standards assume 
that manufacturers will rely solely on the installation of fuel economy 
technology to achieve compliance with the proposed standards. In 
reality, however, manufacturers are likely to rely to some extent on 
three flexibility mechanisms provided by EPCA and will thereby reduce 
the cost of complying with the proposed standards. First, some 
manufacturers will rely on a combination of technology and compliance 
credits that they earn (including credits transferred from one 
compliance category to another) as their compliance strategy. Second, 
they may also supplement their technological efforts by relying on the 
special fuel economy adjustment procedures provided by EPCA as an 
incentive for manufacturers to produce flexible fuel vehicles (FFV). 
Third, the agency is instituting a credit trading program that, if 
taken advantage of, would further provide flexibility.
    The agency believes that manufacturers are likely to take advantage 
of these flexibility mechanisms, thereby reducing benefits and costs 
meaningfully, but does not have any reliable basis for predicting which 
manufacturers might use compliance credits, how they might use them or 
the extent to which they might do so.
    With respect to earned credits through over-compliance NHTSA notes 
that while the manufacturers have relatively few light truck credits, 
several manufacturers already have a substantial amount of banked 
passenger car credits earned under the long term 27.5 mpg flat or 
nonattributed-based standard for those automobiles. Further, they will 
earn significant additional passenger car credits through MY 2010, the 
last year before the passenger car standards are increased and the 
first year in which those standards will be attribute-based. These pre-
MY 2011 passenger car credits can be carried forward into the MY 2011-
2015 period.
    While manufacturers might use credits to a significant extent, 
thereby reducing benefits and costs to a meaningful level, the agency 
believes it important to note that the potential effect of these 
flexibility mechanisms is largely limited to MY 2011-2015. The earning 
of credits will become more difficult in MY 2011. MY 2011 is the first 
year in which all manufacturers will be required to comply with 
attribute-based CAFE standards for passenger cars and light trucks. The 
earning of compliance credits will be more challenging under attribute-
based standards since each manufacturer's legal obligation to improve 
CAFE will be based, in part, on that manufacturer's own product mix. 
Further, the standards will significantly increase every year. On the 
other hand, credits earned in MY 2011 or thereafter can be transferred 
across fleets to a limited extent, adding additional flexibility to the 
system.
    With respect to overcompliance through production of FFV vehicles, 
EISA also extended the FFV adjustment through 2019. Manufacturers can 
build enough FFV vehicles to raise the CAFE of their fleets. FFVs are 
assigned high fuel economy values using a formula specified in the 
Alternative Motor Fuels Act (AMFA). For example, a Ford Taurus has a 
fuel economy of 26.39 mpg--if it is converted to a FFV, its fuel 
economy increases to 44.88 mpg. Converting a vehicle into an FFV is 
more cost-effective than converting it, for example, into a diesel, 
which is more costly and achieves lower fuel economy. However, the 
maximum extent to which the adjustments can be used to raise the CAFE 
of a manufacturer's fleet is 1.2 mpg in MY 2011-2014. In MY 2015, the 
cap begins to decline. The cap continues to decline each year 
thereafter by 0.2 mpg until it reaches 0 mpg in MY 2020 and beyond.
    Given that there will be considerably less opportunity to use 
credits in lieu of installing fuel saving technologies after MY 2015, 
the manufacturers may elect to apply technology early in the MY 2011-
2020 period when redesign


[[Continued on page 24451]]


From the Federal Register Online via GPO Access [wais.access.gpo.gov]
]                         
 
[[pp. 24451-24487]] Average Fuel Economy Standards, Passenger Cars and Light Trucks; 
Model Years 2011-2015

[[Continued from page 24450]]

[[Page 24451]]

opportunities arise rather than relying on credits or FFV adjustments, 
but then face being limited compliance options in later years. The 
declining influence of the flexibility mechanisms during this period 
guarantees that the standards for that year will be met almost entirely 
through the use of technology, thus helping to ensure the 35.0 mpg goal 
of EISA will be achieved.
    Finally, with respect to cost reduction through reliance on credit 
trading, credits earned in MY 2011 or thereafter can be traded. There 
is a study in which the Congressional Budget Office estimated that 
credit trading would cut the costs of achieving a combined 27.5 mpg 
standard by 16 percent.\171\ This study assumed that manufacturer 
compliance costs varied widely and that manufacturers were willing to 
engage in trading. While some manufacturers have expressed reluctance 
to trade with competitors, we believe that the credit trading program 
has the potential to reduce compliance costs meaningfully without any 
impact on overall fuel savings.
---------------------------------------------------------------------------

    \171\ ``The Economic Costs of Fuel Economy Standards Versus a 
Gasoline Tax'', Report from the Congressional Budget Office, 
December, 2003.
---------------------------------------------------------------------------

E. Consistency of Proposed Passenger Car and Light Truck Standards With 
EPCA Statutory Factors

    As explained above, EPCA requires the agency to set fuel economy 
standards for each model year and for each fleet separately at the 
maximum feasible level for that model year and fleet. In determining 
the ``maximum feasible'' level of average fuel economy, the agency 
considers the four statutory factors: Technological feasibility, 
economic practicability, the effect of other motor vehicle standards of 
the Government on fuel economy, and the need of the United States to 
conserve energy, along with additional relevant factors such as safety. 
In determining how to weigh these considerations, we are mindful of 
EPCA's overarching purpose of energy conservation. NHTSA's NEPA 
analysis for this rulemaking (see Section XIII.B of this document) also 
will inform the agency's final action.
    The section above proposes footprint-based CAFE standards for MY 
2011-2015 passenger cars and light truck. The agency has considered 
this set of standards in light of both the relevant factors and EPCA's 
overarching purpose of energy conservation, and seeks comment on 
whether the public agrees that the agency's analysis is sound or should 
have considered the factors differently or considered additional 
factors.
    We have tentatively determined that the proposed passenger car and 
light truck standards are at the maximum feasible level for passenger 
car and light truck manufacturers for MY 2011-2015. As discussed above, 
the standards are basically determined by following the same procedure 
as for setting the optimized light truck standards for 2008-2011.
1. Technological Feasibility
    We tentatively conclude that the proposed standards are 
technologically feasible. Whether a technology may be feasibly applied 
in a given model year is not simply a function of whether the 
technology will exist in that model year, but also whether the data 
sources reviewed by the agency indicate that the technology is mature 
enough to be applied in that year, whether it will conflict with other 
technologies being applied, and so on. The Volpe model maximizes net 
benefits by applying fuel-saving technologies to vehicle models in a 
cost-effective manner, which generally prevents it from applying 
technologies to vehicles before manufacturers would be ready to do so. 
Thus, we tentatively conclude that standards that maximize net benefits 
based on Volpe model analysis are technologically feasible.
    We described above how we tentatively conclude that the additional 
measures used to set the optimized standards do not take the standards 
out of the realm of technological feasibility, because if targets are 
feasible in one year, they will continue to be feasible.
2. Economic Practicability
    NHTSA has historically assessed whether a potential CAFE standard 
is economically practicable in terms of whether the standard is one 
``within the financial capability of the industry, but not so stringent 
as to threaten substantial economic hardship for the industry.'' See, 
e.g., Public Citizen v. NHTSA, 848 F.2d 256, 264 (DC Cir. 1988). We 
tentatively conclude that the proposed standards are economically 
feasible. Making appropriate assumptions about key factors such as 
leadtime and using them in the Volpe model provides a benchmark for 
assessing the economic practicability of a proposed standard, because 
it avoids applying technologies at an infeasible rate and avoids 
application of technologies whose benefits are insufficient to justify 
their costs when the agency determines a manufacturer's capability. In 
other words, this approach ensures that each identified private 
technology investment projected by the model produces marginal benefits 
at least equal to marginal cost. The Volpe model also takes into 
account other factors closely associated with economic practicability, 
such as lead time and phase-in rates for technologies that it applies. 
By limiting the consideration of technologies to those that will be 
available and limiting their rate of application using these 
assumptions, the cost-benefit analysis assumes that manufactures will 
make improvements that are cost-justified.
    In addition to carefully making these assumptions and using cost-
benefit analysis, the agency also performs sales and employment impacts 
analysis on individual manufacturers. The sales analysis looks at a 
purchasing decision from the eyes of a knowledgeable and rational 
consumer, comparing the estimated cost increases versus the payback in 
fuel savings over 5 years (the average new vehicle loan) for each 
manufacturer. This relationship depends on the cost-effectiveness of 
technologies available to each manufacturer. Overall, based on a 7 
percent discount rate for future fuel savings, we expect there would be 
no significant sales or job losses for these proposed standards. 
Therefore, we tentatively conclude that the proposed standards are 
economically practicable.
3. Effect of Other Motor Vehicle Standards of the Government on Fuel 
Economy
    We tentatively conclude that the proposed standards for passenger 
cars and light trucks account for the effect of other motor vehicle 
standards of the Government on fuel economy. This statutory factor 
constitutes an express recognition that fuel economy standards should 
not be set without due consideration given to the effects of efforts to 
address other regulatory concerns, such as motor vehicle safety and 
pollutant emissions. The primary influence of many of these regulations 
is the addition of weight to the vehicle, with the commensurate 
reduction in fuel economy. Manufacturers incorporate this information 
in their product plans, which are accounted for as part of the Volpe 
model analysis used to set the standards. Because the addition of 
weight to the vehicle is only relevant if it occurs within the 
timeframe of the regulations (i.e., MY 2011-2015), we consider the 
Federal Motor Vehicle Safety Standards set by NHTSA and the Federal 
Motor Vehicle Emissions Standards set by EPA which become effective 
during the timeframe.

[[Page 24452]]

Federal Motor Vehicle Safety Standards

    NHTSA has completed a preliminary evaluation of the impact of the 
Federal motor vehicle safety standards (FMVSSs) using MY 2010 vehicles 
as a baseline for passenger cars. We have issued or proposed to issue a 
number of FMVSSs that become effective between the baselines and MY 
2015. These have been analyzed for their potential impact on vehicle 
weights for vehicles manufactured in these years: The fuel economy 
impact, if any, of these new requirements will take the form of 
increased vehicle weight resulting from the design changes needed to 
meet the new FMVSSs.
    The average test weight (curb weight plus 300 pounds) of the 
passenger car fleet is currently 3,570 lbs. During the time period 
addressed by this rulemaking, the average test weight is the passenger 
car fleet is projected to be between 3,608 and 3,635 lbs. The average 
test weight of Chrysler's passenger car fleet is currently 3,928 lbs. 
The average test weight of Chrysler's passenger car fleet is projected 
to be between 3,844 and 3,993 lbs in the future. For Ford, the average 
test weight of the passenger car fleet is currently 3,660 lbs, and is 
projected to be between 3,649 and 3,677 lbs. For GM, the average test 
weight of the passenger car fleet is currently 3,649 lbs, and is 
projected to be between 3,768 and 3,855 lbs. For Toyota, the average 
test weight of the passenger car fleet is currently 3,330 lbs, and is 
projected to be between 3,416 and 3,451 lbs.
    The average test weight (curb weight plus 300 pounds) of the light 
truck fleet is 4,727 pounds, and during the time period addressed by 
this rulemaking, the average test weight of the light truck fleet is 
projected to be between 4,824 and 4,924 lbs. The average test weight of 
Chrysler's light truck fleet is currently 4,673 lbs, while during the 
time period addressed by this rulemaking, the average test weight of 
Chrysler's light truck fleet is projected to be between 4,830 and 4,906 
lbs. For Ford, the light truck fleet's average test weight is currently 
4,887 lbs, while during the time period addressed by this rulemaking, 
the average test weight is projected to be between 4,619 and 4,941 lbs. 
For GM, the light truck fleet's average test weight is currently 5,024 
lbs, while during the time period addressed by this rulemaking, the 
average test weight is projected to be between 5,324 and 5,415 lbs. For 
Toyota, the light truck fleet's average test weight is currently 4,567 
lbs, while during the time period addressed by this rulemaking, the 
average test weight is projected to be between 4,535 and 4,583 lbs.
    Thus, overall, the four largest manufacturers of light-duty 
vehicles expect the average weight of their vehicles to remain mostly 
unchanged, with slight weight increases projected during the time 
period addressed by this rulemaking. The changes in weight include all 
factors, such as changes in the fleet mix of vehicles, required safety 
improvements, voluntary safety improvements, and other changes for 
marketing purposes. These changes in weight over the model years in 
question would have a negligible impact on fuel economy of their 
vehicles.

Weight Impacts of Required Safety Standards (Final Rules)

    NHTSA has issued two final rules on safety standards that become 
effective for passenger cars and light trucks between MY 2011 and MY 
2015. These have been analyzed for their potential impact on passenger 
car and light truck weights, using MY 2010 as a baseline.

1. FMVSS No. 126, Electronic Stability Control
2. FMVSS No. 214, Side Impact Oblique Pole Test

FMVSS No. 126, Electronic Stability Control:

    The phase-in schedule for vehicle manufacturers is:

----------------------------------------------------------------------------------------------------------------
               Model year                         Production beginning date                   Requirement
----------------------------------------------------------------------------------------------------------------
2009...................................  September 1, 2008..........................  55% with carryover credit.
2010...................................  September 1, 2009..........................  75% with carryover credit.
2011...................................  September 1, 2010..........................  95% with carryover credit.
2012...................................  September 1, 2011..........................  All light vehicles.
----------------------------------------------------------------------------------------------------------------

    The final rule requires 75 percent of all light vehicles to meet 
the ESC requirement for MY 2010, 95 percent of all light vehicles to 
meet the ESC requirements by MY 2011, and all light vehicles to meet 
the requirements by MY 2012. Thus, in MY 2010, manufacturers must add 
ESC to 20 percent of vehicles; in MY 2011, to an additional 20 percent 
of vehicles; and in MY 2012, to another 5 percent of vehicles.
    The agency's analysis of weight impacts found that ABS adds 10.7 
lbs. and ESC adds 1.8 lbs. per vehicle for a total of 12.5 lbs. Based 
on manufacturers' plans for voluntary installation of ESC, 85 percent 
of passenger cars in MY 2010 would have ABS and 52 percent would have 
ESC. Thus, the total added weight in MY 2011 for passenger cars would 
be about 2.5 lbs. (0.15 x 10.7 + 0.48 x 1.8), and in MY 2012 would be 
about 0.6 lbs. For light trucks, manufacturers' plans indicate that 99 
percent of all light trucks would have ABS by MY 2011 and that 52 
percent would have ESC by that time. Thus for light trucks, the 
incremental weight impacts of adding ESC would be slightly less than 1 
pound (0.01 x 10.7 + 0.48 x 1.8).

FMVSS No. 214, Side Impact Protection

    NHTSA recently issued a final rule to incorporate a dynamic pole 
test into FMVSS No. 214, ``Side Impact Protection.'' \172\ The rule 
will lead to the installation of new technologies, such as side curtain 
air bags and torso side air bags, which are capable of improving head 
and thorax protection to occupants of vehicles and that crash into 
poles and trees and vehicles that are laterally struck by a higher 
vehicle. The phase-in requirements for the side impact test are as 
shown below: \173\
---------------------------------------------------------------------------

    \172\ 72 FR 51907 (Sept. 11, 2007).
    \173\ Id. 51971-72.

------------------------------------------------------------------------
                                  Percent of each manufacturer's light
        Phase-in date            vehicles that must  comply during the
                                           production period
------------------------------------------------------------------------
September 1, 2009 to August    20 percent (excluding vehicles GVWR >
 31, 2010.                      8,500 lbs.).
September 1, 2010 to August    50 percent of vehicles (excluding
 31, 2011.                      vehicles GVWR > 8,500 lbs.).
September 1, 2011 to August    75 percent of vehicles (excluding
 31, 2012.                      vehicles GVWR > 8,500 lbs.).

[[Page 24453]]


September 1, 2012 to August    All vehicles including limited line
 31, 2013.                      vehicles, except vehicles with GVWR >
                                8,500 lbs., alterers, and multi-stage
                                manufacturers.
On or after September 1, 2013  All vehicles, including vehicles with
                                GVWR > 8,500 lbs., alterers and multi-
                                stage manufacturers.
------------------------------------------------------------------------

    Based on manufacturers' plans to provide window curtains and torso 
bags voluntarily, we estimate that 90 percent of passenger cars and 
light trucks would have window curtains and 72 percent would have torso 
bags for MY 2010. A very similar percentage is estimated for MY 2011. A 
teardown study of 5 thorax air bags resulted in an average weight 
increase per vehicle of 4.77 pounds (2.17 kg).\174\ A second study 
performed teardowns of 5 window curtain systems.\175\ One of the window 
curtain systems was very heavy (23.45 pounds). The other four window 
curtain systems had an average weight increase per vehicle of 6.78 
pounds (3.08 kg), a figure which is assumed to be average for all 
vehicles in the future.
---------------------------------------------------------------------------

    \174\ Khadilkar, et al. ``Teardown Cost Estimates of Automotive 
Equipment Manufactured to Comply with Motor Vehicle Standard--FMVSS 
214(D)--Side Impact Protection, Side Air Bag Features'', April 2003, 
DOT HS 809 809.
    \175\ Ludtke & Associates, ``Perform Cost and Weight Analysis, 
Head Protection Air Bag Systems, FMVSS 201'', page 4-3 to 4-5, DOT 
HS 809 842.
---------------------------------------------------------------------------

    Assuming in the future that the typical system used to comply with 
the requirements of FMVSS No. 214 will be thorax bags with a window 
curtain, the average weight increase would be 2 pounds (0.10 x 6.78 + 
0.28 x 4.77). However, there is the potential that some light trucks 
might need to add structure to meet the test. The agency has no 
estimate of this potential weight impact for structure.

Weight Impacts of Proposed/Planned Standards

Proposed FMVSS No. 216, Roof Crush

    On August 23, 2005, NHTSA proposed amending the roof crush standard 
to increase the roof crush standard from 1.5 times the vehicle weight 
to 2.5 times the vehicle weight.\176\ The NPRM proposed to extend the 
standard to vehicles with a GVWR of 10,000 pounds or less, thus 
including many light trucks that had not been required to meet the 
standard in the past. The proposed effective date was the first 
September 1 occurring three years after publication of the final rule. 
Thus, it is still possible that the final rule could be effective with 
MY 2011. In the PRIA, the average light truck weight was estimated to 
increase by 6.1 pounds for a 2.5 strength to weight ratio. Based on 
comments on the NPRM, the agency believes that this weight estimate is 
likely to increase. However, the agency does not yet have an estimate 
for the final rule.
---------------------------------------------------------------------------

    \176\ 70 FR 49223 (Aug. 23, 2005). The PRIA for this NPRM is 
available at Docket No. NHTSA-2005-22143-2.
---------------------------------------------------------------------------

Planned NHTSA Initiative on Ejection Mitigation

    The agency is planning on issuing a proposal on ejection 
mitigation. The likely result of the planned proposal is for window 
curtain side air bags to be made larger and for a rollover sensor to be 
installed. The likely result will be an increase in weight of at least 
1 pound; however, this analysis is not completed. In addition, advanced 
glazing is one alternative that manufacturers might pursue for specific 
window applications (possibly for fixed windows for third row 
applications) or more broadly. Advanced glazing is likely to have 
weight implications. Again, the agency has not made an estimate of the 
likelihood that advanced glazing might be used or its weight 
implications.

Summary--Overview of Anticipated Weight Increases

    The following table summarizes estimates made by NHTSA regarding 
the weight added in MY 2010 or later to institute the above discussed 
standards or likely rulemakings. In summary, NHTSA estimates that 
weight additions required by final rules and likely NHTSA regulations 
effective in MY 2011 and beyond for passenger cars, compared to the MY 
2010 fleet, will increase passenger car weight by an average of 12.2 
pounds or more (5.5 kg or more). The agency estimates that weight 
additions required by final rules and likely NHTSA regulations 
effective in MY 2011 and beyond for light trucks, compared to the MY 
2010 fleet, will increase light truck weight by an average of 10.1 
pounds or more.

Table VI-8.--Minimum Weight Additions Due to Final Rules or Likely NHTSA
             Regulations Compared to MY 2010 Baseline Fleet
------------------------------------------------------------------------
                                           Added weight    Added weight
              Standard no.                   in pounds     in kilograms
------------------------------------------------------------------------
126.....................................             3.1             1.4
214.....................................             2.0             0.9
216.....................................           6.1-?           2.8-?
Ejection Mitigation.....................           1.0-?           0.4-?
                                         -------------------------------
    Total...............................          12.2-?           5.5-?
------------------------------------------------------------------------

    Based on NHTSA's weight-versus-fuel-economy algorithms, a 3-4 pound 
increase in weight equates to a loss of 0.01 mpg in fuel economy. Thus, 
the agency's estimate of the safety/weight effects is 0.025 to 0.04 mpg 
or more for already issued or likely future safety standards.

Federal Motor Vehicle Emissions Standards

    EPA's Fuel Economy Labeling Rule employs a new vehicle-specific, 5-
cycle approach to calculating fuel economy

[[Page 24454]]

labels which incorporates estimates of the fuel efficiency of each 
vehicle during high speed, aggressive driving, air conditioning 
operation and cold temperatures into each vehicle's fuel economy 
label.\177\ The rule became effective January 26, 2007, and will take 
effect starting with MY 2008.
---------------------------------------------------------------------------

    \177\ See 71 FR 77872 (December 26, 2006).
---------------------------------------------------------------------------

    The new testing procedures will combine measured fuel economy over 
the two current fuel economy tests, the FTP and HFET, as well as that 
over the US06, SC03 and cold FTP tests into estimates of city and 
highway fuel economy for labeling purposes. The test results from each 
cycle will be weighted to represent the contribution of each cycle's 
attributes to onroad driving and fuel consumption. The labeling rule 
does not alter the FTP and HFET driving cycles, the measurement 
techniques, or the calculation methods used to determine CAFE.
    The EPA Labeling Rule will not impact CAFE standards or test 
procedures or other USG regulations.\178\ Rather, the changes to 
existing test procedures will allow for the collection of appropriate 
fuel economy data to ensure that existing test procedures better 
represent real-world conditions.\179\ Further, the labeling rule does 
not have a direct effect upon a vehicle's weight, nor on the fuel 
economy level that a vehicle can achieve. Instead, the labeling rule 
serves to provide consumers with a more accurate estimate of fuel 
economy based on more comprehensive factors reflecting real-world 
driving use.
---------------------------------------------------------------------------

    \178\ Id. section I.F.
    \179\ Id. sections II, IV.
---------------------------------------------------------------------------

    There are two groups of State emissions standards do not qualify 
under 49 U.S.C. 32902(f), and therefore are not considered. One is 
consists of State standards that cannot be adopted and enforced by any 
State because there has been no waiver granted by the EPA under the 
preemption waiver provision in the Clean Air Act.\180\ The other 
consists of State emissions standards that are expressly or impliedly 
preempted under EPCA, regardless of whether or not they have received 
such a waiver. Preempted standards include, for example:

    \180\ 42 U.S.C. 7543 (a).
---------------------------------------------------------------------------

    (1) A fuel economy standard; and
    (2) A law or regulation that has essentially all of the effects 
of a fuel economy standard, but is not labeled as one (i.e., a State 
tailpipe CO2 standard).

4. Need of the U.S. To Conserve Energy
    Congress' requirement to set standards at the maximum feasible 
level and inclusion of the need of the nation to conserve energy as a 
factor to consider in setting CAFE standards ensures that standard 
setting decisions are made with this purpose and all of the associated 
benefits in mind. As discussed above, ``the need of the United States 
to conserve energy'' means ``the consumer cost, national balance of 
payments, environmental, and foreign policy implications of our need 
for large quantities of petroleum, especially imported petroleum.'' 
Environmental implications principally include reductions in emissions 
of criteria pollutants and carbon dioxide.
    The need to conserve energy is, from several different standpoints, 
more crucial today as it was at the time of EPCA's enactment in the 
late 1970s. U.S. energy consumption has been outstripping U.S. energy 
production at an increasing rate. Crude oil prices are currently around 
$100 per barrel, despite having averaged about $13 per barrel as 
recently as 1998, and gasoline prices have doubled in this period.\181\ 
Net petroleum imports now account for 60 percent of U.S. domestic 
petroleum consumption.\182\ World crude oil production continues to be 
highly concentrated, exacerbating the risks of supply disruptions and 
their negative effects on both the U.S. and global economies. Figure 
VI-3 below shows the increase of crude oil imports and the decline of 
U.S. oil production since 1920.
---------------------------------------------------------------------------

    \181\ Energy Information Administration, Annual Energy Review 
2006, Table 5.21, p. 171. Available at http://www.eia.doe.gov/emeu/
aer/pdf/pages/sec5_51.pdf (last accessed Nov. 29, 2007).
    \182\ Energy Information Administration, Annual Energy Review 
2006, Table 5.1, p. 125. Available at http://www.eia.doe.gov/emeu/
aer/pdf/pages/sec5_5.pdf (last accessed Nov. 29, 2007).

---------------------------------------------------------------------------

[[Page 24455]]

[GRAPHIC] [TIFF OMITTED] TP02MY08.034

    The need to conserve energy is also more crucial today because of 
growing greenhouse gas emissions from petroleum consumption by motor 
vehicles and growing concerns about the effects of those emissions. 
Since 1999, the transportation sector has led all U.S. end-use sectors 
in emissions of carbon dioxide. Transportation sector CO2 
emissions in 2006 were 407.5 million metric tons higher than in 1990, 
an increase that represents 46.4 percent of the growth in unadjusted 
energy related carbon dioxide emissions from all sectors over the 
period. Petroleum consumption, which is directly related to fuel 
economy, is the largest source of carbon dioxide emissions in the 
transportation sector.\183\ Moreover, transportation sector emissions 
from gasoline and diesel fuel combustion generally parallel total 
vehicle miles traveled. The need of the nation to conserve energy also 
encompasses all of these issues, insofar as carbon dioxide emissions 
from passenger cars and light trucks decrease as fuel economy improves 
and more energy is conserved.\184\
---------------------------------------------------------------------------

    \183\ However, increases in ethanol fuel consumption have 
mitigated the growth in transportation-related emissions somewhat 
(emissions from energy inputs to ethanol production plants are 
counted in the industrial sector).
    \184\ The above statistics are derived from Energy Information 
Administration, ``Emissions of Greenhouse Gases Report,'' Report 
 DOE/EIA-0573 (2006), released November 28, 2007. Available 
at http://www.eia.doe.gov/oiaf/1605/ggrpt/carbon.html (last accessed 
Feb. 3, 2008).
---------------------------------------------------------------------------

    The need of the nation to reduce energy consumption would be 
properly reflected in the buying decisions of vehicle purchasers, if:
     Vehicle buyers behave as if they have unbiased 
expectations of their future driving patterns and fuel prices; and
     The public social, economic, security, and environmental 
impacts of petroleum consumption are fully identified, quantified and 
reflected in current and future gasoline prices; and
     Vehicle buyers behave as if they account for the impact of 
fuel economy

[[Page 24456]]

on their future driving costs in their purchasing decisions.

Basic economic theory suggests that the price of vehicles should 
reflect the value that the consumer places on the fuel economy 
attribute of his or her vehicle. It is not clear that consumers have 
the information or inclination to value the impact of fuel economy in 
their vehicle purchasing decisions. Consumers generally have no direct 
incentive to value benefits that are not included in the price of 
fuel--for example, benefits such as energy security and limiting global 
climate change. These are the market failures which EPCA requires NHTSA 
to address.

    By accounting for the need of the nation to conserve energy in 
setting CAFE standards, NHTSA helps to mitigate the risks posed by 
petroleum consumption. In its analysis, NHTSA quantifies the need of 
the nation to conserve energy by calculating how much fuel economy a 
vehicle buyer ought to purchase, or rather, how much a vehicle buyer 
ought to value fuel economy, based both on fuel prices and potentially 
estimable externalities (including energy security, the benefits of 
mitigating a ton of CO2 emissions, criteria pollutant 
emissions, noise, safety, and others).
    The Volpe model uses values for these effects in helping to 
determine each model year's CAFE standards. Thus, each model year's 
CAFE standards are set based on an attempt to quantify the need of the 
United States to conserve energy, balanced against the other factors 
considered in the Volpe model, such as the technology inputs that help 
the model establish economically practicable and technologically 
feasible standards.
    Also, as Congress intended, by accounting for the need of the 
nation to conserve energy in setting CAFE standards, NHTSA fulfills 
EPCA's overall goal of improving energy conservation. Factors that 
increase the need of the nation to conserve energy, such as rising oil 
prices or environmental concerns, may be reflected in more stringent, 
but still demonstrably economically practicable fuel economy standards. 
Balancing the EPCA factors against each other, and considering NHTSA's 
NEPA analysis for this rulemaking (see Section XIII.B. of this 
document), NHTSA may decide to set higher CAFE standards, and achieve 
more fuel savings and CO2 emissions reduction, by expressly 
including the quantifiable values of the factors that affect the need 
of the nation to conserve energy.
    These standards will enhance the normal market response to higher 
fuel prices, and will reduce light duty vehicle fuel consumption and 
CO2 tailpipe emissions over the next several decades, 
responding to the need of the nation to conserve energy, as EPCA 
intended. More specifically, the proposed standards will save 55 
billion gallons of fuel and 521 million metric tons of CO2 
over the lifetime of the regulated vehicles. NHTSA will evaluate the 
potential environmental impacts associated with such CO2 
emissions reductions and other environmental impacts of the proposed 
standards through the NEPA process.

F. Other Considerations in Setting Standards Under EPCA

    As explained above, EPCA requires NHTSA to balance the four factors 
of technological feasibility, economic practicability, the effect of 
other motor vehicle standards of the Government on fuel economy, and 
the need of the United States to conserve energy in setting CAFE 
standards for passenger cars and light trucks. As discussed above, EPCA 
also prohibits NHTSA from considering certain factors (e.g., credits) 
in setting CAFE standards. The next section highlights some of the 
issues that NHTSA may (and does) and may not take into account in 
setting CAFE standards under EPCA.
1. Safety
    NHTSA has historically included the potential for adverse safety 
consequences when deciding upon a maximum feasible level, and has been 
upheld by courts in doing so.\185\ Currently, we account for safety in 
the model as we develop the standards: Because downweighting is a 
common compliance strategy, and because the agency believes that 
downweighting of lighter vehicles makes them less safe, our model does 
not rely on weight reductions to achieve the standards for vehicles 
under 5,000 pounds GVWR,\186\ and then only up to 5 percent. As 
explained above, the overarching principle that emerges from the 
enumerated factors and the court-sanctioned practice of considering 
safety and links them together is that CAFE standards should be set at 
a level that will achieve the greatest amount of fuel savings without 
leading to adverse economic or other societal consequences.
---------------------------------------------------------------------------

    \185\ See, e.g., Competitive Enterprise Institute v. NHTSA (CEI 
I), 901 F.2d 107, 120 at n. 11 (DC Cir. 1990) (``NHTSA has always 
examined the safety consequences of the CAFE standards in its 
overall consideration of relevant factors since its earliest 
rulemaking under the CAFE program.'')
    \186\ Kahane study, supra note 78.
---------------------------------------------------------------------------

2. Alternative Fuel Vehicle Incentives
    49 U.S.C. 32902(h) expressly prohibits NHTSA from considering the 
fuel economy of ``dedicated'' automobiles in setting CAFE standards. 
Dedicated automobiles are those that operate only on an alternative 
fuel, like all-electric or natural gas vehicles.\187\ Dedicated 
vehicles often achieve higher mile per gallon (or equivalent) ratings 
than regular gasoline vehicles, so this prohibition prevents NHTSA from 
raising CAFE standards by averaging these vehicles into our 
determination of a manufacturer's maximum feasible fuel economy level.
---------------------------------------------------------------------------

    \187\ 49 U.S.C. 32901(a)(7).
---------------------------------------------------------------------------

    Section 32902(h) also directs NHTSA to ignore the fuel economy 
incentives for dual-fueled (e.g., E85-capable) automobiles in setting 
CAFE standards. Sec.  32905(b) and (d) use special calculations for 
determining the fuel economy of dual-fueled automobiles that give those 
vehicles higher fuel economy ratings than identical regular 
automobiles. Through MY 2014, manufacturers may use this ``dual-fuel'' 
incentive to raise their average fuel economy up to 1.2 miles a gallon 
higher than it would otherwise be; after MY 2014, Congress has set a 
schedule by which the dual-fuel incentive diminishes ratably until it 
is extinguished after MY 2019.\188\ Although manufacturers may use this 
additional credit for their CAFE compliance, NHTSA may not consider it 
in setting standards. As above, this prohibition prevents NHTSA from 
raising CAFE standards by averaging these vehicles into our 
determination of a manufacturer's maximum feasible fuel economy level.
---------------------------------------------------------------------------

    \188\ 49 U.S.C. 32906(a).
---------------------------------------------------------------------------

3. Manufacturer Credits
    Section 32903 was recently revised by EISA, and allows 
manufacturers to earn credits for exceeding CAFE standards in a given 
year and to apply them to CAFE compliance for up to three model years 
before and five model years after the year in which they were earned. 
However, section 32903(a) states expressly that fuel economy standards 
must be ``determined * * * without regard to credits under this 
section.'' Thus, NHTSA may not raise CAFE standards because 
manufacturers have enough credits to meet the higher standards, nor may 
NHTSA lower standards because manufacturers do not have enough credits 
to meet existing standards.

[[Page 24457]]

G. Environmental Impacts of the Proposed Standards

    As noted above, environmental concerns are among the issues bearing 
on the need of the nation to conserve energy. They are also relevant 
under the National Environmental Policy Act (NEPA), 42 U.S.C. 4321-
4347. Requiring improvements in fuel economy will necessarily reduce 
CO2 emissions, because the less fuel a vehicle burns, the 
less CO2 it emits. Reductions in CO2 emissions, 
in turn, may slow or mitigate climate change and associated 
environmental impacts. Increased fuel economy also may affect other 
aspects of the environment, such as emissions of criteria air 
pollutants and air quality.\189\ In order to inform its consideration 
of the proposed standards, NHTSA has initiated an environmental review 
of the proposed standards and reasonable alternatives pursuant to NEPA. 
On March 28, 2008, NHTSA published a notice of intent to prepare an 
environmental impact statement and requested scoping comments (73 FR 
16615). NHTSA is publishing a supplemental notice of public scoping and 
request for scoping comments that invites Federal, State, and local 
agencies, Indian tribes, and the public to participate in the scoping 
process and to help identify the environmental issues and reasonable 
alternatives to be examined in the EIS. The scoping notice also 
provides information about the proposed standards, the alternatives 
NHTSA expects to consider in its NEPA analysis, and the scoping 
process.
---------------------------------------------------------------------------

    \189\ Because CO2 accounts for such a large fraction 
of total greenhouse gases (GHG) emitted during fuel production and 
use--more than 95%, even after accounting for the higher global 
warming potentials of other GHG--NHTSA's analysis of the GHG impacts 
of increasing CAFE standards focuses on reductions in CO2 
emissions resulting from the savings in fuel use that accompany 
higher fuel economy.
---------------------------------------------------------------------------

    As discussed in the scoping notice, in preparing an EIS for this 
rulemaking, NHTSA expects to consider potential environmental impacts 
of the proposed standards and reasonable alternatives, including 
impacts associated with CO2 emissions and climate change. 
NHTSA expects that its NEPA analysis will include: direct impacts 
related to fuel and energy use and emissions of CO2 and air 
pollutants; indirect impacts related to emissions and climate change, 
such as impacts on air quality and temperature and resulting impacts on 
natural resources and on the human environment; and other indirect 
impacts. NHTSA's NEPA analysis will inform its decisions on the 
proposed standards, consistent with NEPA and EPCA.

H. Balancing the Factors to Determine Maximum Feasible CAFE Levels

    While the agency carefully considered alternative stringencies as 
discussed in section X, it tentatively concludes that in stopping at 
the point that maximizes net benefits, it has achieved the best 
balancing of all of the statutory requirements, including the 35 mpg 
requirement. In striking that balance, the agency was mindful of the 
growing need of the nation to conserve energy for reasons that include 
increasing energy independence and security and protecting the 
environment. It was mindful also that this is the first rulemaking in 
which the agency has simultaneously proposed to raise both passenger 
car and light truck standards, and that it was doing so in the context 
of statutory requirements for significant annual increases over an 
extended period of years.
    Among the steps it took in its analysis and balancing were the 
following:
     First, the agency pushed many of the manufacturers in 
their application of technology. NHTSA is proposing standards that it 
estimates will entail risk that some manufacturers will exhaust 
available technologies in some model years. However, the agency has 
tentatively concluded that the additional risk is outweighed by the 
significant increase in estimated net benefits to society.
     Second, as observed in the technology penetration table 
above, the agency believes that more and more advanced, but expensive 
fuel economy technologies will penetrate the fleet by 2015. However, 
the agency was careful to ensure that those technologies are applied in 
an economically and technologically feasible manner by focusing on 
linking certain expensive technologies to redesign and refresh dates 
and by phasing in technologies over time as it is difficult for 
companies to implement many of the technologies on 100 percent of their 
vehicles all at once. Sections III and V describe in fuller detail how 
the agency addressed these issues in its modeling.
     Third, in assessing costs and benefits, the agency took 
into account the private and social benefits, including environmental 
and energy security benefits (e.g., it monetized important 
externalities, such as energy security and CO2) and ensured 
that for every dollar of investment the country gets at least 1 dollar 
of benefits.
     Fourth, in setting attribute based standards as required 
by EISA, the agency will minimize safety implications and preserve 
consumer choice. Further, through its choice of footprint as an 
attribute, the agency minimized the risk of upsizing as it is more 
difficult to change the footprint than to simply add weight to the 
vehicle.
     Fifth, the agency evaluated the costs and benefits 
described above and ensured that the standards were achievable without 
the industry's being economically harmed through significant sales 
losses.
     Sixth, the agency weighed those costs and benefits vis-a-
vis the need of the nation to conserve energy for reasons that include 
increasing energy independence and security and protecting the 
environment and compared the results for a wide variety of alternatives 
as discussed in Chapter X.
     NHTSA tentatively concludes that it has exercised sound 
judgment and discretion in considering degrees of technology 
utilization and degrees of risk, and has appropriately balanced these 
considerations against estimates of the resultant costs and benefits to 
society, thereby arriving at standards that represent the maximum 
feasible standards as required by EPCA. The agency invites comment 
regarding whether it has struck a proper balance and, if not, how it 
should do so.

VII. Standards for Commercial Medium- and Heavy-Duty On-Highway 
Vehicles and ``Work Trucks''

    NHTSA is not promulgating standards for commercial medium- and 
heavy-duty on-highway vehicles or ``work trucks'' \190\ as part of this 
proposed rule. EISA added a new provision to 49 U.S.C. 32902 requiring 
DOT, in consultation with the Department of Energy and the EPA, to 
examine the fuel efficiency of commercial medium- and heavy-duty on-
highway vehicles and work trucks, and determine the appropriate test 
procedures and methodologies for measuring the fuel efficiency of these 
vehicles, as well as the appropriate metric for measuring and 
expressing their fuel efficiency performance and the range of factors 
that affect their fuel efficiency. This study would need to be 
performed within 1 year of the publication of the NAS study required by 
section 108 of EISA.\191\
---------------------------------------------------------------------------

    \190\ ``Work trucks'' are vehicles rated between 8,500 and 
10,000 lbs GVWR and which are not medium-duty passenger vehicles. 49 
U.S.C. 32901(a)(19).
    \191\ 49 U.S.C. 32902(k)(1).
---------------------------------------------------------------------------

    Within 2 years of the completion of the study, DOT would need to 
undertake rulemaking to ``determine''

[[Page 24458]]

* * * how to implement a commercial medium- and heavy-duty on-highway 
vehicle and work truck fuel efficiency improvement program designed to 
achieve the maximum feasible improvement, and * * * adopt and implement 
appropriate test methods, measurement metrics, fuel economy standards, 
and compliance and enforcement protocols that are appropriate, cost-
effective, and technologically feasible'' for these vehicles.\192\ EISA 
also requires a four-year lead time for fuel economy standards 
promulgated under this section, and would allow separate standards to 
be prescribed for different classes of vehicles.\193\
---------------------------------------------------------------------------

    \192\ 49 U.S.C. 32902(k)(2).
    \193\ 49 U.S.C. 32902(k)(2) and (3).
---------------------------------------------------------------------------

VIII. Vehicle Classification

A. Origins of the Regulatory Definitions

    NHTSA developed the regulatory definitions for passenger cars and 
light trucks based on our interpretation of EPCA's language and of 
Congress' intent as evidenced through the legislative history. The 
statutory language is clear that some vehicles must be passenger 
automobiles and some must be non-passenger automobiles. Passenger 
automobiles were defined as ``any automobile (other than an automobile 
capable of off-highway operation) which the Secretary [i.e., NHTSA] 
decides by rule is manufactured primarily for use in the transportation 
of not more than 10 individuals.'' EPCA Sec.  501(2), 89 Stat. 901.
    Thus, under EPCA, there are two general groups of automobiles that 
qualify as non-passenger automobiles: (1) Those defined by NHTSA in its 
regulations as other than passenger automobiles due to their having not 
been manufactured ``primarily'' for transporting up to ten individuals; 
and (2) those expressly excluded from the passenger category by statute 
due to their capability for off-highway operation regardless of whether 
they were manufactured primarily for passenger transportation. NHTSA's 
classification rule directly tracks those two broad groups of non-
passenger automobiles in subsections (a) and (b), respectively, of 49 
CFR 523.5.
    EPCA also defined vehicle ``capable of off-highway operation'' as 
one that NHTSA decides by regulation:

has a ``significant feature'' (other than 4-wheel drive) which is 
designed to equip such automobile for off-highway operation, and 
either (i) is a 4-wheel drive automobile or (ii) is rated at more 
than 6,000 pounds gross vehicle weight.''

    Thus under the statute, any vehicle that has a ``significant 
feature'' and also is either 4-wheel drive or over 6,000 lbs GVWR can 
never be a passenger vehicle. Generally speaking, the ``significant 
feature'' that NHTSA's regulation focuses on relates to high ground 
clearance. EPCA does not prohibit us from choosing other or additional 
significant features, but Congress has had multiple opportunities to 
disagree with our interpretation and has not done so.
    In its final rule establishing its vehicle classification 
regulation, NHTSA noted the ambiguity of the statutory definitions of 
``automobile'' and ``passenger automobile'' and considered at length 
the legislative history of those definitions.\194\ The agency concluded 
that ``* * * both houses of Congress had expressed an intent that 
vehicles classed by EPA as light duty vehicles be subject to average 
fuel economy standards separate from the standards imposed on passenger 
cars.''\195\ The agency thus found it necessary to analyze what 
Congress meant by ``primarily.''
---------------------------------------------------------------------------

    \194\ 42 FR 38362, 38365-67; July 28, 1977.
    \195\ Id. 38366.
---------------------------------------------------------------------------

    In establishing 49 CFR part 523 in the 1970s, we determined that 
Congress intended ``primarily'' to mean ``chiefly'' [or firstly, in the 
first place], not ``substantially'' [or largely, in large part],\196\ 
for two main reasons. First, if ``primarily'' meant ``substantially'' 
or ``in large part,'' ``then almost every automobile would be a 
passenger automobile, since a substantial function of almost all 
automobiles is to transport at least two persons. The only non-
passenger automobiles under this interpretation would be those 
specifically excluded by the definition * * * ''\197\ Because Congress 
gave NHTSA authority to develop the definitions by regulation, it did 
not make sense to read ``primarily'' as limiting the category of non-
passenger automobiles to just those specifically excluded by the 
precise language of the statute.
---------------------------------------------------------------------------

    \196\ We stated that ``the word `primarily' has two ordinary, 
everyday meanings in legal usage--`chiefly' and `substantially.' '' 
See Board of Governors of the Federal Reserve System v. Agnew, 329 
U.S. 441, 446 (1947).
    \197\ 42 FR 38362, 38365 (Jul. 28, 1977).
---------------------------------------------------------------------------

    And second, we concluded that considering ``primarily'' ``against a 
legislative backdrop of other statutes using the identical phrase, and 
the remedial purposes of this Act,'' justified a broad interpretation 
of ``non-passenger automobile.''\198\ The remedial purposes of EPCA--to 
improve fuel efficiency and increase fuel savings--do not require all 
vehicles to be classified as passenger automobiles. Since non-passenger 
automobile CAFE standards must still be set at the maximum feasible 
level, fuel economy of all vehicles would be improved regardless of how 
the vehicles were classified.\199\ Additionally, interpreting ``non-
passenger automobile'' broadly was determined to be consistent with the 
Vehicle Safety Act \200\ and EPA emissions regulations promulgated 
under the Clean Air Act. A broad interpretation of ``non-passenger 
automobile'' served to ``minimize the possibility of inconsistent 
regulatory requirements.''\201\ And finally, analyzing the legislative 
history, NHTSA concluded that ``By using existing terms with existing 
applications [such as ``light duty truck'' as used by EPA], Congress 
gave a clear indication of the types of automobiles that were intended 
to be treated separately from passenger 
automobiles.''202 203
---------------------------------------------------------------------------

    \198\ Id. at 38365-66.
    \199\ Id. at 38366.
    \200\ The Vehicle Safety Act distinguished between ``passenger 
cars'' and ``trucks.''
    \201\ 42 FR 38362, 38366.
    \202\ Id.
    \203\ We note that the 2003 ANPRM that preceded the 2006 CAFE 
rule incorrectly summarized the agency's review of the legislative 
history in the late 1970s. The 2003 ANPRM erroneously stated that 
Congress intended that passenger automobiles be defined as those 
used primarily for the transport of individuals. 68 FR 74926 (Dec. 
29, 2003)
---------------------------------------------------------------------------

     Thus, as NHTSA developed the regulatory definitions, we kept these 
indications from Congress in mind, which resulted in four basic types 
of non-passenger automobiles:

    (1) Automobiles designed primarily to transport more than 10 
persons.
    As a practical matter, this category basically encompasses large 
passenger vans.
    (2) Automobiles designed primarily for purposes of 
transportation of property.

    NHTSA has included in this category both vehicles with open beds 
like pickup trucks, and vehicles which provide greater cargo-carrying 
than passenger-carrying volume. As we stated in the 1977 final rule, 
pickup trucks are not ``manufactured chiefly to transport individuals, 
since well over half of the available space on those automobiles 
consists of the cargo bed, which is exclusively cargo-carrying area. 
Further, this type of automobile is designed to carry heavy loads.'' 
\204\ Regarding vehicles which provide greater cargo-carrying than 
passenger-carrying volume, we stated that ``Since more of the space 
inside the vehicle has been dedicated to transporting cargo, and such 
vehicles are typically designed to carry heavy loads, this agency

[[Page 24459]]

concludes that the chief consideration in designing the vehicle was the 
ability to transport property.'' This included, for example, cargo vans 
and multistop vehicles.
---------------------------------------------------------------------------

    \204\ Id. at 38367.

    (3) Automobiles which are derivatives of automobiles designed 
---------------------------------------------------------------------------
primarily for the transportation of property.

    This could include vehicles in which the cargo-carrying area has 
been converted to provide temporary living quarters, because they would 
typically be a derivative of a cargo van or a pickup truck. 
Additionally, these could include a passenger van with seating 
positions for less than 10 people. Such a vehicle would be basically a 
cargo van with readily removable seats, so removing the seats would 
create more cargo-carrying than passenger-carrying volume. These 
vehicles would be distinguished from station wagons, which have seats 
that can fold down to create a flat cargo space, but are not 
``derivatives,'' in that their parent vehicle is not a non-passenger 
automobile, and do not have the same chassis, springs, or suspension 
system as a non-passenger automobile.

    (4) Automobiles which are capable of off-highway operation.

    NHTSA generally defines ``capable of off-highway operation'' as 
meeting the high ground clearance characteristics of Sec.  523.5(b)(2) 
and either having 4-wheel drive or being rated at more than 6,000 
pounds gross vehicle weight, or both. We note that a vehicle is 
considered as having 4-wheel drive only if it is manufactured with 4-
wheel drive. The fact that the same model is available in 4-wheel drive 
would not be sufficient to classify a 2-wheel drive vehicle as one that 
``has'' 4-wheel drive under Sec.  523.5(b)(1)(i).

B. The Rationale for the Regulatory Definitions in Light of the Current 
Automobile Market

    The categories listed above make up the various criteria which 
allow classification of a vehicle as a light truck under Part 523. 
However, as the 2002 NAS Report noted, the national vehicle market has 
evolved, and the fleets have changed. Until the passage of the Energy 
Independence and Security Act of 2007, Congress had provided no further 
insight since EPCA's enactment into how new types of vehicles that have 
developed since the 1970s should be classified. NHTSA had to classify 
these vehicles based on the words of the statute and on its own 
interpretation of what Congress appears to have wanted. The following 
section identifies the main vehicle types currently classified as light 
trucks, and explains the agency's reasoning for each.
    Pickup trucks were among the original automobiles identified by 
Congress in EPCA's legislative history as vehicles that would not be 
passenger automobiles.\205\ As mentioned earlier, we originally 
identified automobiles ``which can transport property on an open bed'' 
as ones ``not manufactured chiefly to transport individuals, since well 
over half of the available space on those automobiles consists of the 
cargo bed, which is exclusively cargo carrying area.'' \206\ We stated 
further that ``this type of automobile is designed to carry heavy 
loads,'' and is therefore properly a non-passenger automobile or light 
truck.
---------------------------------------------------------------------------

    \205\ EPA included pickup trucks as ``light duty trucks,'' and 
the Senate bill which became EPCA used EPA's definition of light 
duty trucks as examples of vehicles that would be non-passenger 
automobiles. 42 FR 38362, 38366 (Jul. 28, 1977).
    \206\ Id. 38367.
---------------------------------------------------------------------------

    NHTSA recognizes that pickup trucks have evolved since the 1970s, 
and that some now come with extended cabs for extra passenger room and 
smaller open beds. These features, however, do not change the fact that 
pickup trucks are designed to carry loads. Moreover, even with an 
extended cab and a smaller open bed, the fact that the open bed is 
still present indicates to us that the vehicle was manufactured chiefly 
for transporting cargo. If the manufacturer intended the vehicle's 
first purpose to be the carrying of passengers, it could have enclosed 
the entire vehicle. Thus, as 49 CFR 523.5(a)(3) indicates, a pickup 
truck with an open bed is to be classified as a light truck regardless 
of any other features it may possess.
    Sport utility vehicles (SUVs), which possess a substantial market 
share today, had not yet developed when EPCA was enacted or when NHTSA 
first promulgated Part 523, although their forebears like the AMC Jeep 
and other off-road and military style vehicles were known at the time. 
These vehicles originally tended to be classified as light trucks 
because they were capable of off-highway operation, and possessed 
either the necessary high ground clearance characteristics or 4-wheel 
drive or both. They may also be greater than 6,000 pounds GVWR, and/or 
manufactured to permit expanded use of the automobile for cargo-
carrying or other nonpassenger-carrying purposes.
    Part of the overall popularity of SUVs is due to the great variety 
of forms in which they are available. For example, consumer demand has 
led manufacturers to offer smaller SUVs (i.e., less than 6,000 pounds 
GVWR) with features such as the high ground clearance that many drivers 
enjoy. These vehicles may come with two or even three rows of seats as 
standard. If these smaller vehicles actually have 4-wheel drive and the 
requisite number of clearance characteristics, they would properly be 
classified as light trucks under Sec.  523.5(b) without regard to 
functional considerations such as cargo volume.
    However, if these lighter vehicles (i.e., under 6,000 pounds) have 
2-wheel drive, they would not qualify as light trucks under Sec.  
523.5(b) despite having the clearance characteristics. Such vehicles 
may nevertheless be classified as light trucks if they meet one or more 
of the functional criteria in Sec.  523.5(a). For example, if a vehicle 
has three standard rows of seats, it should be classified in accordance 
with Sec.  523.5(a)(5)(ii), on the same basis as many minivans are 
currently classified--that it provides a certain minimum potential 
cargo-carrying capacity that NHTSA has believed is consistent with what 
Congress had in mind when it originally considered the distinction 
between passenger and non-passenger automobiles. Alternatively, a 2-
wheel drive automobile may properly be classified as a light truck 
under Sec.  523.5(a)(4) if it provides ``greater cargo-carrying than 
passenger-carrying volume'' as discussed in one of NHTSA's longstanding 
interpretations.\207\
---------------------------------------------------------------------------

    \207\ In 1981, General Motors asked NHTSA whether a 2-wheel 
drive utility vehicle would be properly classified as a light truck 
as long as the cargo-carrying volume exceeded the passenger-carrying 
volume. We agreed in a letter of interpretation responding to GM 
that ``two-wheel drive utility vehicles which are truck derivatives 
and which, in base form, have greater cargo-carrying volume than 
passenger-carrying volume should be classified as light trucks for 
fuel economy purposes.'' (Emphasis added.) This letter of 
interpretation indicates that in order to be properly classified as 
a light truck under Sec.  523.5(a)(4), a 2-wheel drive SUV must have 
greater cargo-carrying volume than passenger-carrying volume ``in 
base form.'' Base form means the version of the vehicle sold as 
``standard,'' without optional equipment installed, and does not 
include a version that would meet the cargo volume criterion only if 
``delete options'' were exercised to remove standard equipment. For 
example, a base vehicle that comes equipped with a standard second-
row seat would not be classified as a light truck merely because the 
purchaser has an option to delete the second-row seat.
---------------------------------------------------------------------------

    Minivans are another general category of vehicles that essentially 
developed after the enactment of EPCA and the promulgation of Part 523 
are minivans. Minivans are classified as light trucks under the ``flat 
floor'' provision of Sec.  523.5(a)(5), because their seats may be 
easily removed or folded down to create a large flat level surface for 
cargo-carrying. The flat floor provision was originally based on the 
agency's

[[Page 24460]]

determination that passenger vans with removable seats and a flat load 
floor were derived from cargo vans, and should therefore be classified 
as light trucks.\208\
---------------------------------------------------------------------------

    \208\ 42 FR 38362, 38367 (Jul. 28, 1977).
---------------------------------------------------------------------------

    In the preamble to the final rule establishing the MY 1983-1985 
light truck fuel economy standards, in response to a comment by 
Chrysler, we explained that the regulations classified ``large 
passenger vans as light trucks based on the ability of passenger van 
users to readily remove the rear seats to produce a flat, floor level 
cargo-carrying space.'' \209\ Manufacturers generally responded to 
NHTSA's statement by building compact passenger vans--i.e., minivans--
with readily removable rear seats in order to qualify as light trucks 
under the flat floor provision. In short, because minivans often have 
removable seats and a flat floor, they have traditionally been 
classified as light trucks for fuel economy purposes. EPA also 
classifies minivans as light duty trucks for emissions purposes, as 
derivatives of light trucks.
---------------------------------------------------------------------------

    \209\ 45 FR 81593, 81599 (Dec. 11, 1980).
---------------------------------------------------------------------------

    In recent years, many minivans have been designed with seats that 
fold down flat or into the floor pan, rather than being completely 
removable. In the 2006 light truck CAFE final rule, NHTSA revised Sec.  
523.5(a)(5) to allow these minivans to continue to qualify for 
classification as light trucks, requiring ``vehicles equipped with at 
least 3 rows of seats'' to be able to create a ``flat, leveled cargo 
surface'' instead of a ``flat, floor level, surface.'' We believe that 
this is consistent with Congress' intent that vehicles manufactured 
with the capacity to permit expanded use of the automobile for cargo-
carrying or other nonpassenger-carrying purposes be classified as light 
trucks. Minivans have this capacity just as passenger vans do. In order 
to distinguish them from other vehicles like station wagons that also 
arguably have this capacity, we require vehicles to have three rows of 
seats in order to qualify as light trucks on this basis. This helps to 
guarantee a certain amount of potential cargo-carrying volume, since 
manufacturers will not be able to fit an additional row of seats in a 
vehicle under a certain size. Congress did not specify how much cargo 
volume was necessary for a vehicle to be classified as a light truck. 
We believe that this requirement for light truck classification is both 
consistent with Congress' intent that light trucks permit expanded use 
for cargo-carrying purposes, and accommodates the evolution of this 
section of the modern vehicle fleet.
    The latest vehicle type growing rapidly in the U.S. market today is 
the ``crossover'' vehicle. Crossover vehicles are generally designed on 
passenger car-like platforms (unibody construction), but are also 
designed with the functionality of SUVs and minivans. Crossover 
vehicles blur the typical divisions between passenger cars, SUVs and 
minivans (higher ground clearance, two or three rows of seats, and 
varying amounts of cargo space). These vehicles can come in any shape 
or size, they may or may not look like traditional passenger cars, SUVs 
or minivans, and they may be available in a variety of drive 
configurations (2WD, 4WD, AWD, or some combination). As more and more 
of these vehicles become available it will become more difficult to 
categorize them into one particular vehicle category. The majority of 
existing crossover vehicles have been categorized by vehicle 
manufacturers as light trucks under section 523.5(b) if they are off-
highway capable, or under section 523.5(a) due to their functional 
characteristics. NHTSA plans to continue to allow these vehicles to be 
classified as light trucks as long as they continue to meet the light 
truck classification requirements as specified in part 523. As with 
SUVs, when determining off road capability, a vehicle ``has'' 4-wheel 
drive (or AWD) if it is actually equipped with it; a 2-wheel drive 
vehicle is counted as a 2-wheel drive vehicle regardless of whether the 
same model is available in 4-wheel drive. Furthermore, when evaluating 
the functional capabilities against the requirements of section 
523.5(a), vehicles should be classified by model, including all 
vehicles of a particular model. When the light truck determination is 
made based upon the functional characteristics requirements of section 
523.5(a), the base or standard vehicle (vehicle with no options) is 
used to classify the associated model. For example, if a vehicle model 
does not come standard with a third row of seats, but can be purchased 
with an optional third row seat, the vehicle, and all the vehicles 
within that model line, cannot be classified as a light truck under 
523.5(a)(5), which requires vehicles to be equipped, as standard 
equipment, with at least 3 rows of seats and able to create a ``flat, 
leveled cargo'' surface.

C. NHTSA Is Not Proposing To Change the Regulatory Definitions at This 
Time

    As explained above, NHTSA's regulations defining vehicle 
classifications for fuel economy purposes (49 CFR part 523) are based 
on the underlying statute. We continue to believe that they are valid, 
as discussed above. In addition, EISA Congress specifically addressed 
the vehicle classification issue. It redefined ``automobile,'' added a 
definition of ``commercial medium- and heavy-duty on-highway vehicle,'' 
defined non-passenger automobile and defined ``work truck.'' 
Significantly, it did not change other definitions and its new 
definition of ``non-passenger automobile,'' which is most relevant in 
this context, in no way contradicted how NHTSA has long construed that 
term. In enacting EISA, Congress demonstrated its full awareness of how 
NHTSA classifies vehicles for fuel economy purposes and chose not to 
alter those classifications. That strongly suggests Congressional 
approval of the agency's 30-year approach to vehicle classification.
    Accordingly, other than by incorporating EISA's new and revised 
definitions, we are not proposing to change the agency's regulations 
defining vehicle classification. Congress has indicated no need for us 
to do so and such changes would not help achieve Congress' objectives.
    Moreover, Congress has given clear direction that overall 
objectives must be obtained regardless of vehicle classification. The 
EISA adds a significant requirement to EPCA--the combined car and light 
truck fleet must achieve at least 35 mpg in the 2020 model year. Thus, 
regardless of whether the entire fleet is classified as cars or light 
trucks, or any proportion of each, the result must still be a fleet 
performance of at least 35 mpg in 2020. This suggests that Congress did 
not want to spend additional time on the subject of whether vehicles 
are cars or light trucks. Instead, Congress focused on mandating fuel 
economy performance, regardless of classifications.
    With respect to the impact on fuel savings, our tentative 
conclusion is that moving large numbers of vehicles from the light 
truck to the passenger car category would not increase fuel savings or 
stringency of the standards. Under a Reformed attribute-based CAFE 
system, passenger car and light truck CAFE standards will simply be 
reoptimized if vehicles are moved from one category to another. To the 
extent that some relatively fuel-efficient vehicles are moved out of 
the light truck category, the optimization for the remainder of the 
group would likely result in lower standards, because there would now 
be fewer higher performers in the light truck category. However, when 
these trucks are moved into the car category, they are likely to be 
less fuel-efficient than similarly sized cars. Thus,

[[Page 24461]]

including those vehicles could well drag down the optimized targets for 
the car category. Preliminary analyses have suggested that this is what 
happens, but the agency specifically requests comments on this and any 
supporting data for the commenter's position. Further, since EISA now 
permits manufacturers to transfer CAFE credits earned for their 
passenger car fleet to their light truck fleet and vice versa, it makes 
even less difference how a vehicle is classified, because the benefit a 
manufacturer gets for exceeding a standard may be applied anywhere. If 
there is no fuel savings benefit to be gained from revising the 
regulatory definitions, NHTSA does not see how doing so would 
facilitate achieving EPCA's overarching goal of improving fuel savings. 
Although NHTSA does not propose to change the vehicle classification 
standards, the agency does intend to apply those definitions strictly 
and in accordance with agency interpretations, as set out above, and 
the standards presented in the final rule will reflect this. NHTSA 
seeks comment on its reading of the statute with regard to vehicle 
classification and its decision not to change its definitions.

IX. Enforcement

A. Overview

    NHTSA's enforcement under the CAFE program essentially consists of 
gauging a manufacturer's compliance in each model year with the 
passenger car and light truck standards against their credit status. If 
a manufacturer's average miles per gallon for a given fleet falls below 
the relevant standard, and the manufacturer cannot make up the 
difference by using credits earned previously or anticipated to be 
earned for over-compliance, the manufacturer is subject to penalties. 
The penalty, as adjusted for inflation by law,\210\ is $5.50 for each 
tenth of a mpg that a manufacturer's average fuel economy falls short 
of the standard for a given model year multiplied by the total volume 
of those vehicles in the affected fleet (i.e., import or domestic 
passenger car, or light truck), manufactured for that model year. NHTSA 
has collected $735,422,635.50 to date in CAFE penalties, the largest 
ever being paid by DaimlerChrysler for its MY 2006 import passenger car 
fleet, $30,257,920.00. For their MY 2006 fleets, six manufacturers paid 
CAFE fines for not meeting an applicable standard--Ferrari, Maserati, 
BMW, Porsche, Volkswagen, and DaimlerChrysler--for a total of 
$43,170,896.50.
---------------------------------------------------------------------------

    \210\ Federal Civil Penalties Inflation Adjustment Act of 1990, 
28 U.S.C. 2461 note, as amended by the Debt Collection Improvement 
Act of 1996, Pub. L. 104-134, 110 Stat. 1320, Sec.  31001(s).
---------------------------------------------------------------------------

    EPCA authorizes increasing the civil penalty up to $10.00, 
exclusive of inflationary adjustments, if NHTSA decides that the 
increase in the penalty--

    (i) Will result in, or substantially further, substantial energy 
conservation for automobiles in model years in which the increased 
penalty may be imposed; and
    (ii) Will not have a substantial deleterious impact on the 
economy of the United States, a State, or a region of a State.\211\
---------------------------------------------------------------------------

    \211\ 49 U.S.C. 32912(c).

    The agency requests comment on whether it should initiate a 
proceeding to consider raising the civil penalty. Paying civil 
penalties represents a substantial less expensive alternative to 
installing fuel saving technology in order to achieve compliance with 
the CAFE standards or buying credits from another manufacturer. (See 
discussion of credit trading below.)
    Manufacturers can earn CAFE credits to offset deficiencies in their 
CAFE performances under 49 U.S.C. 32903. Specifically, when the average 
fuel economy of either the domestic or imported passenger car or light 
truck fleet for a particular model year exceeds the established 
standard for that category of vehicles, the manufacturer earns credits. 
The amount of credit a manufacturer earns is determined by multiplying 
the tenths of a mile per gallon that the manufacturer exceeded the CAFE 
standard in that model year by the number of vehicles in that category 
it manufactured in that model year. Credits are discussed at much 
greater length in the section below.
    NHTSA begins to determine CAFE compliance by considering pre- and 
mid-model year reports submitted by manufacturers pursuant to 49 CFR 
part 537, Automotive Fuel Economy Reports. The reports for the current 
model year are submitted to NHTSA every December and July. Although the 
reports are used for NHTSA's reference only, they help the agency, and 
the manufacturers who prepare them, anticipate potential compliance 
issues as early as possible, and help manufacturers plan compliance 
strategies.
    NHTSA makes its ultimate determination of manufacturers' CAFE 
compliance based on EPA's official calculations, which are in turn 
based on final model year data submitted by manufacturers to EPA 
pursuant to 40 CFR 600.512, Model Year Report, no later than 90 days 
after the end of the calendar year. EPA then verifies the data 
submitted by manufacturers and issues final CAFE reports to 
manufacturers and to NHTSA between April and October of each year (for 
the previous model year). NHTSA identifies the manufacturers' fleets 
that have failed to meet the applicable CAFE fleet standards, and 
issues enforcement letters to manufacturers not meeting one or more of 
the standards. Letters are generally issued within one to two weeks of 
receipt of EPA's final CAFE reports.
    For the enforcement letters, NHTSA calculates a cumulative credit 
status for each of a manufacturer's vehicle categories according to 49 
U.S.C. 32903. If sufficient credits are available, NHTSA determines a 
carry-forward credit allocation plan. If the manufacturer does not have 
enough credits to offset the shortfall, NHTSA requests payment of a 
corresponding civil penalty unless the manufacturer submits a carry-
back credit allocation plan. We note that any penalties paid are paid 
to the U.S. Treasury and not to NHTSA itself.
    After enforcement letters are sent, NHTSA continues to monitor 
civil penalty payments that are due within 60 days from the date of 
receipt of the letter by the vehicle manufacturer, and takes further 
action if the manufacturer is delinquent in payment. NHTSA also 
monitors receipt of carry-back plans from manufacturers who choose this 
compliance alternative. Plans are required within 60 days from the date 
of receipt of the enforcement letter by the vehicle manufacturer.

B. CAFE Credits

    The ability to earn and apply credits has existed since EPCA's 
original enactment,\212\ but the issue of the ability to trade credits, 
i.e., to sell credits to other manufacturers or buy credits from them, 
was first raised in the 2002 NAS Report. NAS found that
---------------------------------------------------------------------------

    \212\ The credit provision (currently codified at 49 U.S.C. 
32903) was originally section 508 of EPCA's Public Law version.

changing the current CAFE system to one featuring tradable fuel 
economy credits and a ``cap'' on the price of these credits appears 
to be particularly attractive. It would provide incentives for all 
manufacturers, including those that exceed the fuel economy targets, 
to continually increase fuel economy, while allowing manufacturers 
flexibility to meet consumer preferences.\213\
---------------------------------------------------------------------------

    \213\ NAS, Finding 11, 113.

After receiving the 2002 NAS Report, Secretary of Transportation Mineta 
wrote to Congress asking for authority to implement all of NAS'' 
recommendations.
    While waiting for that express authority, NHTSA raised the issue of

[[Page 24462]]

credit trading in both its 2002 Request for Comments \214\ and its 2003 
ANPRM.\215\ The initial response to the idea was mixed: environmental 
and consumer groups expressed concern that vehicle manufacturers would 
use a credit trading system in lieu of increasing fuel economy to meet 
the CAFE standards, while vehicle manufacturers generally supported the 
prospect of increased flexibility in the CAFE program.\216\ However, 
without clear authority to implement a credit trading program, NHTSA 
was unable to take further action at the time.
---------------------------------------------------------------------------

    \214\ 67 FR 5767, 5772 (Feb. 7, 2002).
    \215\ 68 FR 74908, 74915-16 (Dec. 29, 2003).
    \216\ Id.
---------------------------------------------------------------------------

    NHTSA raised the issue of credit transfer, i.e., the application of 
credits earned by manufacturer in one compliance category to another 
compliance category, in its 2005 NPRM \217\ and 2006 final rule for the 
MY 2008-11 light truck standards, but concluded that it would interfere 
with the transition to Reformed CAFE by making it more difficult for 
manufacturers to determine their compliance obligations.\218\ The 2006 
final rule also stated that the agency would not adopt a credit trading 
program, again on the basis that its authority to do so was 
unclear.\219\ However, NHTSA submitted several draft bills to Congress 
during this time period and after, most recently in February 2007. In 
an address to the Senate Committee on Commerce, Science, and 
Transportation on March 6, 2007 regarding the February 2007 bill, 
Administrator Nason stated that credit trading was a ``natural 
extension'' of the existing EPCA credit framework, and that trading 
would be ``purely voluntary, and [that] we believe[d] it will help 
lower the industry's cost of complying with CAFE.'' \220\
---------------------------------------------------------------------------

    \217\ 70 FR 51414, 51439-40 (Aug. 30, 2005).
    \218\ 71 FR 17566, 17616 (Apr. 6, 2006).
    \219\ Id. 17653-54.
    \220\ Transcript available at http://commerce.senate.gov/public/
index.cfm?FuseAction=Hearings.TestimonyHearing--ID=1827--Witness--
ID=2362 (last accessed Feb. 2, 2008).
---------------------------------------------------------------------------

    EISA provided express authority for both credit trading and 
transferring and made other changes as well to EPCA regarding credits:
     Authorizing the establishment of a credit trading program;
     Requiring the establishment of a credit transferring 
program; and
     Extending the carry-forward period from 3 to 5 years.
    NHTSA has developed a proposal for a new Part 536 setting up these 
two credit programs. We believe that our proposal is consistent with 
Congress' intent. The agency seeks comment generally on the following 
three topics with respect to the proposed Part 536: (1) Whether the 
agency has correctly interpreted Congress' intent; (2) whether there 
are any ways to improve the proposed credit trading and transferring 
system consistent with EISA and Congress' intent that the agency might 
have overlooked; and (3) whether any of the aspects of the programs 
proposed by the agency are either inconsistent with EISA and Congress' 
intent or the rest of the CAFE regulations, or are otherwise 
unworkable. The following section describes the proposed credit trading 
and transfer programs, as well as several other related ideas that the 
agency is considering.
1. Credit Trading
    EPCA, as amended by EISA, states
    The Secretary of Transportation [by delegation, the 
Administrator of NHTSA] may establish by regulation a fuel economy 
credit trading program to allow manufacturers whose automobiles 
exceed the average fuel economy standards prescribed under section 
32902 to earn credits to be sold to manufacturers whose automobiles 
fail to achieve the prescribed standards such that the total oil 
savings associated with manufacturers that exceed the prescribed 
standards are preserved when trading credits to manufacturers that 
fail to achieve the prescribed standards.\221\
---------------------------------------------------------------------------

    \221\ 49 U.S.C. 32903(f)(1).

EISA also prevents traded credits from being used by a manufacturer to 
meet the minimum fuel economy standard for domestically-manufactured 
passenger cars.\222\
---------------------------------------------------------------------------

    \222\ 49 U.S.C. 32903(f)(2).
---------------------------------------------------------------------------

    Proposed new part 536 would permit credit trading, beginning with 
credits earned in MY 2011. Although only manufacturers may earn credits 
and apply them toward compliance, NHTSA would allow credits to be 
purchased and traded by both manufacturers and non-manufacturers in 
order to facilitate greater flexibility in the credit market.
    NHTSA proposes that credit trading be conducted as follows: If a 
credit holder wishes to trade credits to another party, the current 
credit holder and the receiving party must jointly issue an instruction 
to NHTSA, identifying the specific credits to be traded by quantity, 
vintage (model year of origin), compliance category of origin (domestic 
passenger cars, imported passenger cars, or light trucks), and 
originating manufacturer. These identification requirements are 
intended to help ensure accurate calculation for preserving total oil 
savings. If the credit recipient is not already an account holder, it 
must provide sufficient information for NHTSA to establish an account 
for them. Once an account has been established or identified, NHTSA 
will complete the trade by debiting the transferor's account and 
crediting the recipient's account. NHTSA will track the quantity, 
vintage, compliance category, and originator of all credits held or 
traded by all account-holders.
    Manufacturers need not restrict their use of traded credits to the 
compliance category from which the credits were earned. However, if a 
manufacturer wishes to transfer a credit received by trade to another 
compliance category, it must instruct NHTSA of its intention so that 
NHTSA can apply an adjustment factor in order to preserve ``total oil 
savings,'' as required by EISA.\223\ EISA requires total oil savings to 
be preserved because one credit is not necessarily equal to another, as 
Congress realized. For example, the fuel savings lost if the average 
fuel economy of a manufacturer falls one-tenth of a mpg below the level 
of a relatively low standard are greater than the fuel savings gained 
by raising the average fuel economy of a manufacturer one-tenth of a 
mpg above the level of a relatively high CAFE standard.
---------------------------------------------------------------------------

    \223\ 49 U.S.C. 32903(f)(1).
---------------------------------------------------------------------------

    Table IX-1 shows a simple numerical example of this on an 
individual vehicle level. Vehicle A has a fuel economy of 30 mpg and is 
driven 150,000 miles over its lifetime, consuming 5,000 gallons of 
fuel. Increasing the fuel economy of vehicle A by one mpg lowers the 
lifetime fuel consumption by 161 gallons to 4,839 gallons. Vehicle B 
has a fuel economy of 15 mpg and is driven 150,000 miles over its 
lifetime, consuming 10,000 gallons of fuel. Increasing the fuel economy 
of vehicle B by one mpg lowers the lifetime fuel consumption by 625 
gallons to 9,375 gallons. Both vehicles' fuel economy rises by the same 
amount, one mpg, but much more fuel is saved by vehicle B because it 
uses much more gas per mile than does vehicle A.

[[Page 24463]]



    Table IX-I.--Comparison of Fuel Savings at Different Fuel Economy
                                Baselines
------------------------------------------------------------------------
                                             Vehicle A       Vehicle B
------------------------------------------------------------------------
Lifetime Miles Driven...................         150,000         150,000
Initial Fuel Economy....................              30              15
Initial Lifetime Fuel Consumption.......           5,000          10,000
Final Fuel Economy......................              31              16
Final Lifetime Fuel Consumption.........           4,839           9,375
Savings.................................             161             625
------------------------------------------------------------------------

    To preserve total oil savings in credit trading, NHTSA would apply 
an adjustment factor to traded credits. More specifically, the agency 
would multiply the value of each credit (with a nominal value of 0.1 
mpg per vehicle) by an adjustment factor calculated by the following 
formula:
[GRAPHIC] [TIFF OMITTED] TP02MY08.033

    Where:

A = adjustment factor applied to traded credits by multiplying mpg 
for a particular credit;
VMTe = lifetime vehicle miles traveled for the compliance 
category in which the credit was earned (152,000 miles for domestic 
and imported passenger cars; 179,000 miles for light trucks);
VMTu = lifetime vehicle miles traveled for the compliance 
category in which the credit is used for compliance (152,000 miles 
for domestic and imported passenger cars; 179,000 miles for light 
trucks);
MPGe = fuel economy standard for the originating 
manufacturer, compliance category, and model year in which the 
credit was earned;
MPGu = fuel economy standard for the manufacturer, 
compliance category, and model year in which the credit will be 
used.

    The effect of applying this formula would be to increase the value 
of credits that were earned for exceeding a relatively low CAFE 
standard and are to be applied to a compliance category with a 
relatively high CAFE standard and decrease the value of credits that 
were earned for exceeding a relatively high CAFE standard and are to be 
applied to a compliance category with a relatively low CAFE standard. 
NHTSA is proposing to use the fuel economy standard in the formula 
rather than the actual fuel economy or some average of the two, 
primarily because we believe it will be more predictable for credit 
holders and traders. However, we seek comment on those two 
alternatives, since they may be more precise in their ability to 
account for fuel savings.
    Congress also restricted the use of credit trading in EISA by 
providing that manufacturers must comply with the minimum domestic 
passenger car standard specified in 49 U.S.C. 32902(b)(4) without the 
aid of credits obtained through trading. The minimum standard equals 
the greater of 27.5 mph or 92 percent of the projected average fuel 
economy level for all passenger cars for the model year in question. 49 
U.S.C. 32903(f)(2) states that trading and transferring of credits to 
the domestic passenger car compliance category are limited to the 
extent that the fuel economy of such automobiles shall comply with the 
minimum standard without regard to trading or transferring of credits 
from other compliance categories. Thus, our proposed credit trading 
regulation prevents the use of traded credits to comply with the 
minimum domestic passenger car standard.
    In developing this regulation, NHTSA has proposed additional 
restrictions on the use of credits as necessary for consistency with 
Congress' intent in EISA. For example, a credit that has been traded 
and is then traded back to the originating manufacturer is deemed never 
to have been traded, to avoid manufacturers gaining value from the same 
credit twice.
2. Credit Transferring
    If a credit holding manufacturer wishes to transfer credits that it 
has earned, it need simply instruct NHTSA which credits to transfer to 
which alternate compliance category, identifying the quantity, vintage, 
and original compliance category in which the credits were earned. 
NHTSA will then transfer the credits. As explained above, if a credit 
holding manufacturer wishes to transfer credits that it has received by 
trade, it must similarly instruct NHTSA. NHTSA will apply an adjustment 
factor to the traded credits to ensure, pursuant to EISA, that total 
oil (fuel) savings are preserved.
    Credit transfers are limited by EISA both in the extent to which 
they may increase a manufacturer's average fuel economy in a compliance 
category, and when they may be begun to be used. Section 32903(g)(3) 
states that a manufacturer's average fuel economy in a compliance 
category cannot be increased through the use of transferred credits by 
more than 1 mpg in MYs 2011-2013, more than 1.5 mpg in MYs 2014-2017, 
or more than 2 mpg in MYs 2018 and after. Section 32903(g)(5) also 
states that credits can only be transferred if they are earned after MY 
2010. Our proposed credit transferring regulation reflects these 
limitations.
    Congress also restricted the use of credit transferring in EISA by 
providing that manufacturers must comply with the minimum domestic 
passenger car standard without the aid of credits obtained through 
transfer. 49 U.S.C. 32903(g)(4) states that transferring of credits to 
the domestic passenger car compliance category is limited to the extent 
that the fuel economy of such automobiles shall comply with the minimum 
standard without regard to transferring of credits from other 
compliance categories. Thus, our proposed credit transferring 
regulation prevents the use of transferred credits to comply with the 
minimum domestic passenger car standard.
    NHTSA is proposing to denominate credits in miles per gallon (mpg), 
not in gallons. NHTSA requests comments, however, on whether 
transferred credits

[[Page 24464]]

should be denominated in gallons, because doing so would ensure that no 
transfers result in any loss of fuel savings or in a missed opportunity 
to reduce CO2 emissions.\224\ The risk of fuel savings loss 
can be illustrated by the following example. Suppose there were a 
manufacturer that produces the same number of automobiles in two 
different compliance categories. Each of the two categories is required 
to meet the same level of CAFE. If the manufacturer exceeds the 
standard for one category by one mile per gallon and falls short of the 
other standard by the same amount, the additional fuel saved by the 
automobiles subject to the first standard would be less than the 
additional fuel consumed by the automobiles subject to the second 
standard. The risk is even greater if the example is changed so that 
the standards are different and the manufacturer exceeds the higher 
standard and falls short of the lower standard.
---------------------------------------------------------------------------

    \224\ NHTSA previously addressed this issue in the 2006 final 
rule establishing CAFE standards for MY 2008-2011 light trucks. See 
71 FR 17566, 17616.
---------------------------------------------------------------------------

3. Credit Carry-Forward/Carry-Back
    Credit lifespan has always been dictated by statute. A manufacturer 
may only use credits for a certain number of model years before and 
after the year in which it was earned. Congress intended credits to 
provide manufacturers greater compliance flexibility, but did not wish 
that flexibility to be so great as to obviate the need to continue 
improving fleet fuel economy. Before EISA's enactment, EPCA permitted 
credits to be used for 3 model years before and after the model year in 
which a credit was earned; EISA extended the ``carry-forward'' time to 
5 model years. Because EISA was enacted in the middle of model year 
2008,\225\ NHTSA concluded that the best interpretation of this change 
in lifespan was to apply it only to vehicles manufactured in or after 
MY 2009; the alternative of finding some way to prorate the change in 
lifespan presents considerable administrative difficulties, especially 
since credits are denominated by year of origin, not month and year of 
origin. Thus, credits earned for MYs 2008 and earlier will continue to 
have a 3-year carry-forward/carry-back lifespan; credits earned in MY 
2009 or thereafter will have a 5-year carry-forward and a 3-year carry-
back lifespan.
---------------------------------------------------------------------------

    \225\ EISA's effective date was December 20, 2007; the 2008 
model year began on October 1, 2007.
---------------------------------------------------------------------------

C. Extension and Phasing Out of Flexible-Fuel Incentive Program

    EPCA encourages manufacturers to build alternative-fueled and dual-
fueled vehicles. This is accomplished by using a special, statutorily 
specified calculation procedure for determining the fuel economy of 
these vehicles. The specially calculated fuel economy figure is based 
on the assumption that the vehicle operates on the alternative fuel a 
significant portion of the time. This approach gives such vehicles a 
much-higher fuel economy level compared to similar gasoline-fueled 
vehicles. These vehicles can then be factored into a manufacturer's 
general fleet fuel economy calculation, thus raising the average fuel 
economy level of the fleet. EPCA limited the extent to which a 
manufacturer could raise its fuel economy level due to the incentive to 
1.2 mpg per compliance category.
    Prior to the enactment of EISA, this incentive was only available 
through MY 2010. EISA extended the incentive, but also provided for 
phasing it out between MYs 2015 and 2019, by progressively reducing the 
amount by which fleet fuel economy could be raised due to the 
incentive.\226\ Thus, the maximum fuel economy increase which may be 
attributed to the incentive is as follows for:
---------------------------------------------------------------------------

    \226\ 49 U.S.C. 32906.

------------------------------------------------------------------------
                                                                  mpg
------------------------------------------------------------------------
MYs 1993-2014................................................        1.2
MY 2015......................................................        1.0
MY 2016......................................................        0.8
MY 2017......................................................        0.6
MY 2018......................................................        0.4
MY 2019......................................................        0.2
After MY 2019................................................          0
------------------------------------------------------------------------

    NHTSA promulgated 49 CFR part 538 to implement the statutory 
alternative-fueled and dual-fueled vehicle manufacturing incentive. We 
are not now proposing to amend Part 538 to reflect the EISA changes, 
due to the already-large scope of the current rulemaking, but will do 
so in an upcoming rulemaking.

X. Regulatory Alternatives

    As noted above, in developing the proposed standards, the agency 
considered the four statutory factors underlying maximum feasibility 
(technological feasibility, economic practicability, the effect of 
other standards of the Government on fuel economy, and the need of the 
nation to conserve energy) as well as other relevant considerations 
such as safety. NHTSA assessed what fuel saving technologies would be 
available, how effective they are, and how quickly they could be 
introduced. This assessment considered technological feasibility, 
economic practicability and associated energy conservation. We also 
considered other standards to the extent captured by EPCA \227\ and 
environmental and safety concerns. This information was factored into 
the computer model used by NHTSA for applying technologies to 
particular vehicle models. The agency then balanced the factors 
relevant to standard setting. NHTSA's NEPA analysis, discussed in 
Section XIII.B. of this document, also will inform NHTSA's 
consideration of the proposed standards and reasonable alternatives in 
developing a final rule.
---------------------------------------------------------------------------

    \227\ 71 FR 17566, 17669-70; April 6, 2006.
---------------------------------------------------------------------------

    In balancing these factors, NHTSA generally observes that the 
increasing application of technologies increases fuel economy and 
associated benefits, but it also increases costs. Initial applications 
of technologies provide far more fuel savings per dollar of expenditure 
on them than applications of remaining technologies, which provide less 
incremental fuel savings at greater cost and, with progressive 
additions of technologies, eventually far greater cost. At some stage, 
the increasing application of technologies is not justified. A 
significant question is what methodology and decisionmaking criteria 
are used in the balancing to determine when to cease adding 
technologies and thus arrive at regulatory fuel economy targets.
    In developing its proposed standards, the agency used a net 
benefit-maximizing analysis that placed monetary values on relevant 
externalities (both energy security and environmental externalities, 
including the benefits of reductions in CO2 emissions) and 
produced what is called the ``optimized scenario.'' The optimized 
standards reflect levels such that, considering the seven largest 
manufacturers, net benefits (that is, total benefits minus total costs) 
are higher than at every other examined level of stringency. The agency 
also reviewed the results of the model's estimates of stringencies 
maximizing net benefits to assure that the results made sense in terms 
of balancing EPCA's statutory factors and in meeting EISA's 
requirements for improved fuel economy.
    In addition to the optimized scenario, NHTSA considered and 
analyzed five additional regulatory alternatives that do not rely upon 
marginal benefit-cost

[[Page 24465]]

analysis. In ascending order of stringency, the six alternatives are:
     Standards that fall below the optimized scenario by the 
same absolute amount by which the +25 percent alternative exceeds the 
optimized scenario (``25 percent below optimized'' alternative),
     Standards based on applying technologies until net 
benefits are maximized (optimized scenario), and
     Standards that exceed the optimized scenario by 25 percent 
of the interval between the optimized scenario and the TC = TB 
alternative (see below) (``25 percent above optimized'' alternative),
     Standards that exceed the optimized scenario by 50 percent 
of the interval between the optimized scenario and the TC = TB 
alternative (``50 percent above optimized'' alternative),
     Standards based on applying technologies until total costs 
equal total benefits (zero net benefits) (TC = TB alternative),\228\ 
and
---------------------------------------------------------------------------

    \228\ The agency considered the ``TC = TB'' alternative because 
one or more commenters in the rulemaking on standards for MY 2008-
2011 light trucks urged NHTSA to consider setting the standards on 
this basis rather than on the basis of maximizing net benefits. In 
addition, while the Ninth Circuit Court of Appeals concluded that 
EPCA neither requires nor prohibits the setting of standards at the 
level at which net benefits are maximized, the Court raised the 
possibility of tilting the balance more toward reducing energy 
consumption and CO2.
---------------------------------------------------------------------------

     Standards based on applying all feasible technologies 
without regard to cost (technology exhaustion alternative).\229\
---------------------------------------------------------------------------

    \229\ This was accomplished by determining the stringency at 
which a reformed standard would require every manufacturer to apply 
every technology estimated to be potentially available. At such 
stringencies, all but one manufacturer would be expected to fail to 
comply with the standard, and many manufacturers would owe large 
civil penalties as a result. The agency considered this alternative 
because the agency wished to explore the stringency and consequences 
of standards based solely on the potential availability of 
technologies at the individual manufacturer level.
---------------------------------------------------------------------------

    NHTSA chose these alternatives in order to consider and evaluate 
the impacts of balancing the EPCA factors differently in determining 
maximum feasibility than the agency has in prior rulemakings. In Center 
for Biological Diversity v. NHTSA, the Ninth Circuit Court recognized 
that ``EPCA gives NHTSA discretion to decide how to balance the 
statutory factors--as long as NHTSA's balancing does not undermine the 
fundamental purpose of EPCA: Energy conservation.'' 508 F.3d 508, 527 
(9th Cir. 2007). The Court also raised the possibility that NHTSA's 
current balancing of the statutory factors might be different from the 
agency's balancing in the past, given the greater importance today of 
the need of the nation to conserve energy and more advanced 
understanding of climate change. Id. at 530-31.
    Given EPCA's mandate that NHTSA consider four specific factors in 
setting CAFE standards and NEPA's instruction that agencies give effect 
to NEPA's policies as well, NHTSA recognizes that numerous alternative 
CAFE levels are theoretically conceivable and that the alternatives 
described above essentially represent only several on a continuum of 
alternatives. Along the continuum, each alternative represents a 
different way in which NHTSA conceivably could assign weight to each of 
the four EPCA factors and NEPA's policies. For the alternatives that 
fall above the optimized scenario (the +25, +50 and TC = TB 
alternatives), the agency would evaluate policies that put increasingly 
more emphasis on reducing energy consumption and CO2 
emissions, given their impact on global warming, and less on the other 
factors, including the economic impacts on the industry. Conversely, 
for the alternative that falls below the optimum scenario, the agency 
would evaluate policies that place relatively more weight on the 
economic situation of the industry and less on reducing energy 
consumption and CO2 emissions.
    The graphs below show, for passenger cars, light trucks, and the 
combined fleet, the average annual fuel economy levels for the four 
alternatives as compared to the proposed standards. Subsequent graphs 
and tables present their estimated costs, benefits, and net benefits 
(in billions of dollars). In addition, tables that are provided 
summarized the average extent to which manufacturers' CAFE levels are 
projected to fall short of CAFE standards--i.e., the average 
shortfall--under each of these alternatives. Manufacturer-specific 
shortfall is shown for the proposed and TC=TB alternative.
BILLING CODE 4910-59-P

[[Page 24466]]

[GRAPHIC] [TIFF OMITTED] TP02MY08.035


[[Page 24467]]


[GRAPHIC] [TIFF OMITTED] TP02MY08.036


[[Page 24468]]


[GRAPHIC] [TIFF OMITTED] TP02MY08.037


[[Page 24469]]


[GRAPHIC] [TIFF OMITTED] TP02MY08.038


[[Page 24470]]


[GRAPHIC] [TIFF OMITTED] TP02MY08.039


[[Page 24471]]


[GRAPHIC] [TIFF OMITTED] TP02MY08.040

BILLING CODE 4910-59-C

    For the proposal and each regulatory alternative, the Tables X-1 
and X-3 show the total net benefits in millions of dollars at a 7 
percent discount rate for the projected fleet of sales for each model 
year.

                      Table X-1.--Total Benefits Over the Vehicle's Lifetime--Present Value
                                    [Millions of 2006 dollars, discounted 7%]
----------------------------------------------------------------------------------------------------------------
                                      MY 2011         MY 2012         MY 2013         MY 2014         MY 2015
----------------------------------------------------------------------------------------------------------------
Passenger Cars:
    25% Below...................           1,156           2,104           3,235           5,197           6,799
    Optimized...................           2,596           4,933           6,148           7,889           9,420
    25% Above...................           3,755           7,280           8,454          10,638          12,083
    50% Above...................           4,274           8,825          10,213          12,576          14,495
    TC = TB.....................           5,769          10,878          12,087          14,644          16,492
    Technology Exhaust..........           5,834          11,282          12,968          15,930          18,061
Light Trucks:
    25% Below...................           3,508           7,910          12,603          12,433          12,441
    Optimized...................           3,909           8,779          13,560          14,915          16,192
    25% Above...................           4,201           9,990          14,236          16,587          19,457
    50% Above...................           4,642          10,507          15,011          17,687          20,892
    TC = TB.....................           5,027          11,453          16,330          19,515          22,367

[[Page 24472]]


    Technology Exhaust..........           5,088          11,512          19,395          22,074          24,779
Combined PC+LT:
    25% Below...................           4,664          10,014          15,838          17,630          19,240
    Optimized...................           6,505          13,712          19,708          22,804          25,612
    25% Above...................           7,956          17,270          22,690          27,225          31,540
    50% Above...................           8,916          19,331          25,224          30,263          35,387
    TC = TB.....................          10,796          22,331          28,417          34,159          38,860
    Technology Exhaust..........          10,922          22,795          32,363          38,004          42,820
----------------------------------------------------------------------------------------------------------------


                                             Table X-2.--Total Costs
                                    [Millions of 2006 dollars, discounted 7%]
----------------------------------------------------------------------------------------------------------------
                                      MY 2011         MY 2012         MY 2013         MY 2014         MY 2015
----------------------------------------------------------------------------------------------------------------
Passenger Cars:
    25% Below...................             835             818           1,253           2,153           3,209
    Optimized...................           1,884           2,373           2,879           3,798           4,862
    25% Above...................           3,387           5,653           6,445           8,240           9,084
    50% Above...................           4,010           7,885           8,986          11,207          12,981
    TC = TB.....................           5,913          10,796          12,303          15,403          17,398
    Technology Exhaust..........           6,079          12,595          14,701          18,759          21,110
Light Trucks:
    25% Below...................           1,349           4,296           6,329           6,212           6,326
    Optimized...................           1,649           4,986           7,394           8,160           8,761
    25% Above...................           2,072           7,034           9,815          11,903          14,781
    50% Above...................           2,922           8,098          11,586          14,386          17,969
    TC = TB.....................           3,788          10,525          15,196          18,762          21,364
    Technology Exhaust..........           3,933          10,670          18,275          21,051          23,479
Combined PC+LT:
    25% Below...................           2,184           5,114           7,582           8,365           9,534
    Optimized...................           3,534           7,358          10,273          11,957          13,623
    25% Above...................           5,459          12,687          16,261          20,143          23,865
    50% Above...................           6,932          15,983          20,572          25,593          30,950
    TC = TB.....................           9,702          21,321          27,499          34,164          38,761
    Technology Exhaust..........          10,013          23,266          32,976          39,810          44,589
----------------------------------------------------------------------------------------------------------------


                   Table X-3.--Net Total Benefits Over the Vehicle's Lifetime--Present Value *
                                    [Millions of 2006 dollars, discounted 7%]
----------------------------------------------------------------------------------------------------------------
                                      MY 2011         MY 2012         MY 2013         MY 2014         MY 2015
----------------------------------------------------------------------------------------------------------------
Passenger Cars:
    25% Below...................             321           1,285           1,982           3,045           3,590
    Optimized...................             711           2,560           3,269           4,092           4,558
    25% Above...................             368           1,627           2,009           2,398           2,999
    50% Above...................             264             940           1,226           1,370           1,514
    TC = TB.....................            -144              82            -216            -759            -906
    Technology Exhaust..........            -245          -1,313          -1,733          -2,829          -3,049
Light Trucks:
    25% Below...................           2,154           3,633           6,348           6,288           6,258
    Optimized...................           2,260           3,793           6,167           6,755           7,432
    25% Above...................           2,129           2,956           4,421           4,684           4,676
    50% Above...................           1,720           2,408           3,426           3,301           2,924
    TC = TB.....................           1,239             928           1,134             753           1,003
    Technology Exhaust..........           1,155             843           1,120           1,023           1,280
Combined PC+LT:
    25% Below...................           2,476           4,919           8,330           9,333           9,848
    Optimized...................           2,971           6,353           9,435          10,847          11,989
    25% Above...................           2,497           4,583           6,430           7,082           7,675
    50% Above...................           1,984           3,349           4,652           4,670           4,437
    TC = TB.....................           1,094           1,010             918              -5              98
    Technology Exhaust..........             909            -471            -613          -1,806         -1,769
----------------------------------------------------------------------------------------------------------------
* Negative values mean that costs exceed benefits.


[[Page 24473]]

    In tentatively deciding which alternative to propose, the agency 
looked at a variety of factors. The agency notes that once stringency 
levels exceed the point at which net benefits are maximized, the 
societal costs of each incremental increase in stringency exceed the 
accompanying societal benefits. If we have valued benefits 
appropriately, it does not make economic sense to mandate the spending 
of more money than society receives in return. The resources used to 
meet overly stringent CAFE standards, instead of the optimized scenario 
standards, would better be allocated to other uses such as technology 
research and development, or improvements in vehicle safety.
    The agency considered the burden placed on specific manufacturers, 
consumers and employment. As CAFE standards increase, the incremental 
benefits are approximately constant while the incremental costs 
increase rapidly. Figure X-5 above shows that as stringency is 
increased, costs rise out of proportion compared to the benefits or the 
fuel savings. Increasingly higher costs have a negative impact on sales 
and employment. Each of the alternatives that is more stringent than 
the optimized alternative negatively impact sales and employment.
    The agency also considered technological feasibility. The Volpe 
model assumes that major manufacturers will exhaust all available 
technology before paying noncompliance civil penalties, even though the 
latter is often less costly. Historically, the large manufacturers have 
never paid civil penalties. In the more stringent alternatives, the 
Volpe model predicts that increasing numbers of manufacturers will run 
out of technology to apply and, theoretically, resort to penalty 
payment. NHTSA provisionally believes that setting standards this high 
is not technologically feasible, nor does it serve the need of the 
nation to conserve fuel. Paying a CAFE penalty does not result in any 
fuel savings.
    In analyzing the ``-25 percent below optimized'' alternative, the 
agency notes that these standards are more aggressive than the 
standards that the agency has proposed since the first years of the 
program and would impose unprecedented costs on manufacturers. The 
agency also recognizes that even this pace of increase in the standards 
may burden some of the manufacturers, particularly since the agency is 
now increasing car and light truck standards simultaneously. However, 
in light of the need of the nation to conserve energy and reduce global 
warming, the agency does not believe that this alternative would be 
maximum feasible under the statute. The agency is also concerned that 
the combined fleet might not reach the 35 mpg requirement by 2020 under 
EISA.
    Underlying the differences in costs, benefits, and net benefits for 
the other alternatives are differences in the degree to which NHTSA has 
estimated that technologies might be applied in response to the 
standards corresponding to each of these alternatives. The following 
tables show estimates of the average penetration rates of some selected 
technologies in the MY2015 passenger car and light truck fleets under 
each of the alternatives discussed here:

                        Table X-4.--Estimated Average Technology Penetration (Largest Seven Manufacturers) MY2015 Passenger Cars
                                                                      [In percent]
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                 Average among seven largest manufacturers
                                                 -------------------------------------------------------------------------------------------------------
                   Technology                       Product      Adjusted    25% Below     Proposed    25% Above    50% Above                   Tech.
                                                      plan       baseline     proposed     standard     proposed     proposed     TC = TB     exhaustion
--------------------------------------------------------------------------------------------------------------------------------------------------------
Automatically Shifted Manual Transmissions......           10           10           23           39           47           55           63           69
Spark Ignited Direct Injection Engines..........           22           22           22           30           37           48           68           63
Turbocharging & Engine Downsizing...............            5            5            8           17           30           40           62           57
Diesel Engines..................................            0            0            3            2            7           13           18           21
Hybrid Electric Vehicles........................            5            5           14           15           22           28           35           38
--------------------------------------------------------------------------------------------------------------------------------------------------------


                         Table X-5.--Estimated Average Technology Penetration (Largest Seven Manufacturers) MY2015 Light Trucks
                                                                      [In percent]
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                 Average among seven largest manufacturers
                                                 -------------------------------------------------------------------------------------------------------
                   Technology                       Product      Adjusted    25% Below     Proposed    25% Above    50% Above                   Tech.
                                                      plan       baseline     proposed     standard     proposed     proposed     TC = TB     exhaustion
--------------------------------------------------------------------------------------------------------------------------------------------------------
Automatically Shifted Manual Transmissions......           10           14           42           55           58           60           59           70
Spark Ignited Direct Injection Engines..........           23           24           31           40           42           55           60           69
Turbocharging & Engine Downsizing...............            9           11           21           31           38           51           54           65
Diesel Engines..................................            3            6            8           10           20           23           26           28
Hybrid Electric Vehicles........................            2            6           15           25           29           31           30           30
--------------------------------------------------------------------------------------------------------------------------------------------------------

    As the first of the above tables indicates, the Volpe model 
estimated that, under the standards proposed today, manufacturers might 
triple the planned utilization of turbochargers and hybrid electric 
powertrains in the

[[Page 24474]]

passenger car fleet. This table also indicates that the use of 
turbochargers in passenger cars might increase by an additional factor 
of two under the ``25% above proposed'' alternative.
    Similarly, the second table indicates that manufacturers might 
triple the planned utilization of diesel engines in the light truck 
fleet, and increase the utilization of hybrid electric powertrains by 
more than an order of magnitude. This table also shows a significant 
difference between the proposed and ``25% above proposed'' alternative, 
including an additional doubling in the utilization of diesel engines.
    NHTSA has examined the extent to which each alternative would (as 
estimated by the Volpe model and using the input information discussed 
in preceding sections) cause manufacturers to exhaust technologies 
projected to be available during MY2011-MY2015. The following chart 
summarizes the frequency with which this was estimated to occur--i.e., 
the number of instances in which an individual manufacturer exhausted 
technologies and thus fell below a standard in individual model years 
divided by 35 (seven manufacturers times five model years).
[GRAPHIC] [TIFF OMITTED] TP02MY08.041

    As this analysis indicates, the ``25% below proposed'' alternative 
caused technologies to be exhausted 3 percent of the time for passenger 
cars, and 17 percent of the time for light trucks. Under the proposed 
standards, the rate

[[Page 24475]]

of technology exhaustion increased to 11 percent for passenger cars, 
but did not change for light trucks. However, under the ``25% above 
proposed'' alternative, the corresponding rates increased to 26 percent 
and 37 percent, respectively. In other words, under this alternative, 
the Volpe model estimated that, more than a quarter of the time, 
manufacturers would be unable to comply with the passenger car 
standards solely using technologies expected to be available, and that 
they would be unable to comply with the light truck standards using 
available technologies more than a third of the time. These rates were 
estimated to be considerably higher for the remaining three 
alternatives.
    These estimates of technology utilization and the exhaustion of 
available technologies indicate that all of the alternatives NHTSA has 
considered entail risk that one or more manufacturers would not be able 
to comply with both the passenger car and light truck standards in 
every model year solely by applying technology. This risk is mitigated 
somewhat by the fact that our analysis may not encompass every 
technology that will potentially be available during MY2011-MY2015. For 
example, some manufacturers have made public statements regarding hopes 
to offer ``plug-in'' HEVs before MY2015, but such vehicles are not 
represented in our analysis.\230\ Nonetheless, the agency has 
tentatively concluded that the scope of technologies it has included is 
comprehensive enough that the analysis shown above indicates that under 
some alternatives, there is considerable risk that some manufacturers 
would exhaust available technologies in some model years.
---------------------------------------------------------------------------

    \230\ If included in the new product plans that the agency is 
requesting, these vehicles will be included in our analysis for the 
final rule.
---------------------------------------------------------------------------

    In tentatively concluding that the proposed standards are the 
maximum feasible standards, NHTSA has balanced this risk against the 
other considerations it must take into account, in particular the need 
of the nation to conserve energy, which encompasses concerns regarding 
carbon dioxide emissions. The agency's analysis includes economic 
measures of these needs--that is, economic measures of the 
externalities of petroleum consumption and the damages associated with 
carbon dioxide emissions. These measures are reflected in the agency's 
estimates of the total and net benefits of each of the alternatives.
    NHTSA is proposing standards that it estimates will entail risk 
that some manufacturers will exhaust available technologies in some 
model years. However, relative to the less stringent ``25% below 
proposed'' alternative, the agency has tentatively concluded that the 
additional risk is outweighed by the significant increase in estimated 
net benefits to society, ranging from an additional $0.5b in MY2011 to 
an additional $2.1b in MY2015. Conversely, the agency has tentatively 
concluded that, relative to the proposed standards, the more than 
doubling of risk posed by the ``25% above proposed'' alternative is not 
warranted, especially considering that this alternative is estimated to 
significantly reduce net benefits, by $0.5b in MY2011 and, eventually, 
$4.3b in MY2015.
    NHTSA tentatively concludes that it has exercised reasonable 
judgment in considering degrees of technology utilization and degrees 
of risk, and has appropriately balanced these considerations against 
estimates of the resultant costs and benefits to society.
    Notwithstanding the tentative conclusions described above, NHTSA 
seeks comment on these and other regulatory alternatives to aid in 
determining what standards to adopt in the final rule.\231\ The agency 
invites comment regarding whether it has struck a proper balance and, 
if not, how it should do so. The alternatives identified by the agency 
are intended to aid public commenters in helping the agency to explore 
that issue. NHTSA's NEPA analysis also will inform its further action 
on today's proposal and may influence the final standards.
---------------------------------------------------------------------------

    \231\ In assessing the alternatives set out in this document, 
commenters may find it useful to examine the approaches being taken 
by other countries to improving fuel economy and reducing tailpipe 
CO2 emissions, e.g., Canada, http://www.tc.gc.ca/pol/en/environment/
FuelConsumption/index.html (last accessed April 20, 2008); European 
Union, http://ec.europa.eu/environment/co2/co2_home.htm (last 
accessed April 20, 2008); and Japan, http://www.eccj.or.jp/top_
runner/pdf/vehicles_gasdiesel_feb2007.pdf (last accessed April 20, 
2008).
---------------------------------------------------------------------------

    Specific sensitivity runs that vary fuel prices, the rebound 
effect, CO2 and discount rate were conducted for the proposed Optimized 
standards. These analyses have an impact on the standards, costs and 
benefits. For example, in analyzing the ``optimized alternative'', we 
estimated that following the same methods and criteria for setting the 
standards, but applying a 3 percent discount rate rather than a 7 
percent discount rate, would suggest standards reaching about 33.6 mpg 
(average required fuel economy among both passenger cars and light 
trucks) in MY2015, 2 mpg higher than the 31.6 mpg average resulting 
from the standards we are proposing based on a 7 percent discount rate. 
The more stringent standards during MY2011-MY2015 would reduce CO2 
emissions by 672 million metric tons (mmt), or 29 percent more than the 
521 mmt achieved by the proposed standards. On the other hand, we 
estimate that standards increasing at this pace would require about 
$85b in technology outlays during MY2011-MY2015, or 89 percent more 
than the $45b in technology outlays associated with the standards 
proposed today. The impact of the 3 percent rate is shown in the body 
of the PRIA along with the 6 formal alternatives. All other sensitivity 
analyses are shown in Chapter IX of the PRIA.

XI. Sensitivity and Monte Carlo Analysis

    NHTSA is proposing fuel economy standards that maximize net 
societal benefits, based on the Volpe model. That is, where the 
estimated benefits to society exceed the estimated cost of the rule by 
the highest amount. This analysis is based, among other things, on many 
underlying estimates, all of which entail uncertainty. Future fuel 
prices, the cost and effectiveness of available technologies, the 
damage cost of carbon dioxide emissions, the economic externalities of 
petroleum consumption, and other factors cannot be predicted with 
certainty.
    Recognizing these uncertainties, NHTSA has used the Volpe model to 
conduct both sensitivity analyses, by changing one factor at a time, 
and a probabilistic uncertainty analysis (a Monte Carlo analysis that 
allows simultaneous variation in these factors) to examine how key 
measures (e.g., mpg levels of the standard, total costs and total 
benefits) vary in response to changes in these factors.
    However, NHTSA has not conducted a probabilistic uncertainty 
analysis to evaluate how optimized stringency levels respond to such 
changes in these factors. The Volpe model currently does not have the 
capability to integrate Monte Carlo simulation with stringency 
optimization.
    The results of the sensitivity analyses indicate that the value of 
CO2, the value of externalities, and the value of the rebound effect 
have almost no impact on the level of the standards. Assuming a higher 
price of gasoline has the largest impact of the sensitivity analyses 
examined (raising the MY 2015 passenger car standard level by 6.7 mpg 
and the light truck level by 0.8 mpg). It appears that the light truck 
levels are not as sensitive as the passenger car levels to changes in 
the estimated benefits. This can occur because the technologies that 
have not been used

[[Page 24476]]

under the Optimized alternative, and are still available for light 
trucks, are not that close to being cost effective and it takes a 
larger increase in benefits to bring them over the cost-benefit 
threshold.
    NHTSA's sensitivity analysis found that changes in the damage cost 
of carbon dioxide emissions and the economic externalities of petroleum 
consumption had very little impact on the stringency levels of the 
proposed standards (at most 0.1 mpg per year). The agency varied 
estimated carbon dioxide damage costs over a range of $0 to $14 per 
metric ton and varied the economic externalities of petroleum 
consumption over a range of $0.120 to $0.504 per gallon.
    However, the sensitivity analysis did show significant changes in 
the stringency of the standards in response to large increases in the 
projected future cost of gasoline. By increasing the price of gasoline 
by an average of $0.88 in 2016 to $1.22 in 2020 per gallon, the 
passenger car standard that maximized net societal benefits for MY 2015 
increased from 35.7 mpg to 42.4 mpg and the light truck standard for MY 
2015 increases from 28.6 mpg to 29.4 mpg. NHTSA notes that, unlike 
carbon dioxide damage costs and the economic externalities of petroleum 
consumption, the price of gasoline is not an externality. The Volpe 
model assumes manufacturers consider fuel prices when selecting among 
available technologies.
    OMB Circular A-4 requires formal probabilistic uncertainty analysis 
of complex rules where there are large, multiple uncertainties whose 
analysis raises technical challenges or where effects cascade and where 
the impacts of the rule exceed $1 billion. The agency identified and 
quantified the major uncertainties in the preliminary regulatory impact 
analysis and estimated the probability distribution of how those 
uncertainties affect the benefits, costs, and net benefits of the 
alternatives considered in a Monte Carlo analysis. The results of that 
analysis, summarized for the combined passenger car and light truck 
fleet across both the 7 percent (typically the lower range) and 3 
percent (typically upper range) discount rates \232\ are as follows:
---------------------------------------------------------------------------

    \232\ In a few cases the upper range results were obtained from 
the 7% rate and the lower range results were obtained from the 3% 
rate. While this may seem counterintuitive, it results from the 
random selection process that is inherent in the Monte Carlo 
technique.
---------------------------------------------------------------------------

    Fuel Savings: The analysis indicates that MY 2011 vehicles (both 
passenger cars and light trucks) will experience between 3,370 million 
and 4,735 million gallons of fuel savings over their useful lifespan. 
MY 2012 vehicles will experience between 7,476 million and 9,639 
million gallons of fuel savings over their useful lifespan. MY 2013 
vehicles will experience between 10,863 million and 13,763 million 
gallons of fuel savings over their useful lifespan. MY 2014 vehicles 
will experience between 12,568 and 15,664 million gallons of fuel 
savings over their useful lifespan. MY 2015 vehicles will experience 
between 14,188 and 17,659 million gallons of fuel savings over their 
useful lifespan. Over the combined lifespan of the five model years, 
between 48.5 billion and 61.4 billion gallons of fuel will be saved.
    Total Costs: The analysis indicates that owners of MY 2011 
passenger cars and light trucks will pay between $2,447 million and 
$5,256 million in higher vehicle prices to purchase vehicles with 
improved fuel efficiency. MY 2012 owners will pay between $5,817 
million and $10,427 million more. MY 2013 owners will pay between 
$7,942 million and $15,288 million more. MY 2014 owners will pay 
between $9,338 million and $17,189 million more. MY 2015 owners will 
pay between $10,940 million and $19,842 million more. Owners of all 
five model years vehicles combined will pay between $36.5 billion and 
$67.9 billion in higher vehicle prices to purchase vehicles with 
improved fuel efficiency.
    Societal Benefits: The analysis indicates that changes to MY 2011 
passenger cars and light trucks to meet the proposed CAFE standards 
will produce overall societal benefits valued between $4,375 million 
and $13,041 million. MY 2012 vehicles will produce benefits valued 
between $9,363 million and $28,214 million. MY 2013 vehicles will 
produce benefits valued between $13,370 million and $41,027 million. MY 
2014 vehicles will produce benefits valued between $15,586 million and 
$47,087 million. MY 2015 vehicles will produce benefits valued between 
$17,486 million and $53,708 million. Over the combined lifespan of the 
five model years, societal benefits valued between $60.1 billion and 
$183.1 billion will be produced.
    Net Benefits: The uncertainty analysis indicates that the net 
impact of the higher CAFE requirements for MY 2011 passenger cars and 
light trucks will be a net benefit of between $937 million and $9,678 
million. There is at least a 99.3 percent certainty that changes made 
to MY 2011 vehicles to achieve the higher CAFE standards will produce a 
net benefit. The net impact of the higher CAFE requirements for MY 2012 
will be a net benefit of between $283 million and a net benefit of 
$21,139 million. There is at least a 99.6 percent certainty that 
changes made to MY 2012 vehicles to achieve the CAFE standards will 
produce a net benefit. The net impact of the higher CAFE requirements 
for MY 2013 will be a net benefit of between $494 million and a net 
benefit of $31,311 million. There is at least a 99.6 percent certainty 
that changes made to MY 2013 vehicles to achieve the higher CAFE 
standards will produce a net benefit. The net impact of the higher CAFE 
requirements for MY 2014 will be a net benefit of between $711 million 
and $35,746 million. There is 100 percent certainty that changes made 
to MY 2014 vehicles to achieve the CAFE standards will produce a net 
benefit. The net impact of the higher CAFE requirements for MY 2015 
will be a net benefit of between $654 million and $40,703 million. 
There is 100 percent certainty that changes made to MY 2015 vehicles to 
achieve the CAFE standards will produce a net benefit. Over all five 
model years, the higher CAFE standards will produce net benefits 
ranging from $3.1 billion to $138.6 billion. There is at least a 99.3 
percent certainty that higher CAFE standards will produce a net 
societal benefit in each of the model years covered by this final rule. 
In most years, this probability is 100 percent.

XII. Public Participation

How do I prepare and submit comments?

    Your comments must be written and in English. To ensure that your 
comments are correctly filed in the Docket, please include the docket 
number of this document in your comments. Your comments must not be 
more than 15 pages long.\233\ We established this limit to encourage 
you to write your primary comments in a concise fashion. However, you 
may attach necessary additional documents to your comments. There is no 
limit on the length of the attachments.
---------------------------------------------------------------------------

    \233\ See 49 CFR 553.21.
---------------------------------------------------------------------------

    Please submit your comments by any of the following methods:
     Federal eRulemaking Portal: Go to http://
www.regulations.gov. Follow the online instructions for submitting 
comments.
     Mail: 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.
     Hand Delivery or Courier: West Building Ground Floor, Room 
W12-140, 1200 New Jersey Avenue, SE., between

[[Page 24477]]

9 a.m. and 5 p.m. Eastern Time, Monday through Friday, except Federal 
holidays.
     Fax: (202) 493-2251.
    If you are submitting comments electronically as a PDF (Adobe) 
file, we ask that the documents submitted be scanned using the Optical 
Character Recognition (OCR) process, thus allowing the agency to search 
and copy certain portions of your submissions.\234\
---------------------------------------------------------------------------

    \234\ Optical character recognition (OCR) is the process of 
converting an image of text, such as a scanned paper document or 
electronic fax file, into computer-editable text.
---------------------------------------------------------------------------

    Please note that pursuant to the Data Quality Act, in order for 
substantive data to be relied upon and used by the agency, it must meet 
the information quality standards set forth in the OMB and DOT Data 
Quality Act guidelines. Accordingly, we encourage you to consult the 
guidelines in preparing your comments. OMB's guidelines may be accessed 
at http://www.whitehouse.gov/omb/fedreg/reproducible.html. DOT's 
guidelines may be accessed at http://dmses.dot.gov/submit/
DataQualityGuidelines.pdf.

How can I be sure that my comments were received?

    If you submit your comments by mail and wish Docket Management to 
notify you upon its receipt of your comments, enclose a self-addressed, 
stamped postcard in the envelope containing your comments. Upon 
receiving your comments, Docket Management will return the postcard by 
mail.

How do I submit confidential business information?

    If you wish to submit any information under a claim of 
confidentiality, you should submit three copies of your complete 
submission, including the information you claim to be confidential 
business information, to the Chief Counsel, NHTSA, at the address given 
above under FOR FURTHER INFORMATION CONTACT. When you send a comment 
containing information claimed to be confidential business information, 
you should include a cover letter setting forth the information 
specified in our confidential business information regulation.\235\
---------------------------------------------------------------------------

    \235\ See 49 CFR 512.
---------------------------------------------------------------------------

    In addition, you should submit a copy, from which you have deleted 
the claimed confidential business information, to the Docket by one of 
the methods set forth above.

Will the agency consider late comments?

    We will consider all comments received before the close of business 
on the comment closing date indicated above under DATES. To the extent 
possible, we will also consider comments received after that date. If 
interested persons believe that any new information the agency places 
in the docket affects their comments, they may submit comments after 
the closing date concerning how the agency should consider that 
information for the final rule. However, the agency's ability to 
consider late comments in this rulemaking will be limited as the agency 
anticipates issuing a final rule this fall.
    If a comment is received too late for us to consider in developing 
a final rule (assuming that one is issued), we will consider that 
comment as an informal suggestion for future rulemaking action.

How can I read the comments submitted by other people?

    You may read the materials placed in the docket for this document 
(e.g., the comments submitted in response to this document by other 
interested persons) at any time by going to http://www.regulations.gov. 
Follow the online instructions for accessing the dockets. You may also 
read the materials at the Docket Management Facility by going to the 
street address given above under ADDRESSES. The Docket Management 
Facility is open between 9 a.m. and 5 p.m. Eastern Time, Monday through 
Friday, except Federal holidays.

XIII. Regulatory Notices and Analyses

A. Executive Order 12866 and DOT Regulatory Policies and Procedures

    Executive Order 12866, ``Regulatory Planning and Review'' (58 FR 
51735, Oct. 4, 1993), provides for making determinations whether a 
regulatory action is ``significant'' and therefore subject to OMB 
review and to the requirements of the Executive Order. The Order 
defines a ``significant regulatory action'' as one that is likely to 
result in a rule that may:
    (1) Have an annual effect on the economy of $100 million or more or 
adversely affect in a material way the economy, a sector of the 
economy, productivity, competition, jobs, the environment, public 
health or safety, or State, local or Tribal governments or communities;
    (2) Create a serious inconsistency or otherwise interfere with an 
action taken or planned by another agency;
    (3) Materially alter the budgetary impact of entitlements, grants, 
user fees, or loan programs or the rights and obligations of recipients 
thereof; or
    (4) Raise novel legal or policy issues arising out of legal 
mandates, the President's priorities, or the principles set forth in 
the Executive Order.
    The rulemaking proposed in this NPRM will be economically 
significant if adopted. Accordingly, OMB reviewed it under Executive 
Order 12866. The rule, if adopted, would also be significant within the 
meaning of the Department of Transportation's Regulatory Policies and 
Procedures.
    The benefits and costs of this proposal are described above. 
Because the proposed rule would, if adopted, be economically 
significant under both the Department of Transportation's procedures 
and OMB guidelines, the agency has prepared a Preliminary Regulatory 
Impact Analysis (PRIA) and placed it in the docket and on the agency's 
Web site. Further, pursuant to OMB Circular A-4, we have prepared a 
formal probabilistic uncertainty analysis for this proposal. The 
circular requires such an analysis for complex rules where there are 
large, multiple uncertainties whose analysis raises technical 
challenges or where effects cascade and where the impacts of the rule 
exceed $1 billion. This proposal meets these criteria on all counts.

B. National Environmental Policy Act

    In litigation concerning NHTSA's 2006 final rule, ``Average Fuel 
Economy Standards for Light Trucks, Model Years 2008-2011,'' 71 FR 
17566, April 6, 2006 (Final Rule), the U.S. Court of Appeals for the 
Ninth Circuit ordered NHTSA to prepare an Environmental Impact 
Statement (EIS) for that rule. Center for Biological Diversity v. 
NHTSA, 508 F.3d 508, 558 (9th Cir. 2007). The Government is seeking 
rehearing on the appropriateness of that remedy, instead of a remand of 
the agency's Environmental Assessment (EA) and Finding of No 
Significant Impact (FONSI) for further consideration.
    Simultaneously, NHTSA has initiated the EIS process under the 
National Environmental Policy Act (NEPA), 42 U.S.C. 4321-4347, and 
implementing regulations issued by the Council on Environmental Quality 
(CEQ), 40 CFR part 1500, and NHTSA, 49 CFR part 520. On March 28, 2008, 
NHTSA published a notice of intent to prepare an EIS for this 
rulemaking and requested scoping comments. (73 FR 16615) NHTSA is 
publishing a supplemental notice of public scoping and request for 
scoping comments that invites Federal, State, and local agencies, 
Indian tribes, and the public to participate in the scoping process and 
to help identify the environmental issues and reasonable alternatives 
to be examined in the EIS. The scoping notice also provides information 
about the proposed standards, the alternatives

[[Page 24478]]

NHTSA expects to consider in its NEPA analysis, and the scoping 
process.

C. Regulatory Flexibility Act

    Pursuant to the Regulatory Flexibility Act (5 U.S.C. 601 et seq., 
as amended by the Small Business Regulatory Enforcement Fairness Act 
(SBREFA) of 1996), whenever an agency is required to publish a notice 
of rulemaking for any proposed or final rule, it must prepare and make 
available for public comment a regulatory flexibility analysis that 
describes the effect of the rule on small entities (i.e., small 
businesses, small organizations, and small governmental jurisdictions). 
The Small Business Administration's regulations at 13 CFR part 121 
define a small business, in part, as a business entity ``which operates 
primarily within the United States.'' 13 CFR 121.105(a). No regulatory 
flexibility analysis is required if the head of an agency certifies the 
rule will not have a significant economic impact on a substantial 
number of small entities.
    I certify that the proposed rule would not have a significant 
economic impact on a substantial number of small entities. The 
following is NHTSA's statement providing the factual basis for the 
certification (5 U.S.C. 605(b)).
    If adopted, the proposal would directly affect seventeen large 
single stage motor vehicle manufacturers.\236\ The proposal would also 
affect four small domestic single stage motor vehicle 
manufacturers.\237\ According to the Small Business Administration's 
small business size standards (see 13 CFR 121.201), a single stage 
automobile or light truck manufacturer (NAICS code 336111, Automobile 
Manufacturing; 336112, Light Truck and Utility Vehicle Manufacturing) 
must have 1,000 or fewer employees to qualify as a small business. All 
four of the vehicle manufacturers have less than 1,000 employees and 
make less than 1,000 vehicles per year. We believe that the rulemaking 
would not have a significant economic impact on the small vehicle 
manufacturers because under Part 525, passenger car manufacturer making 
less than 10,000 vehicles per year can petition NHTSA to have 
alternative standards set for those manufacturers. These manufacturers 
currently don't meet the 27.5 mpg standard and must already petition 
the agency for relief. If the standard is raised, it has no meaningful 
impact on these manufacturers, they still must go through the same 
process and petition for relief. Given that there already is a 
mechanism for handling small businesses, which is the purpose of the 
Regulatory Flexibility Act, a regulatory flexibility analysis was not 
prepared.
---------------------------------------------------------------------------

    \236\ BMW, Mercedes, Chrysler, Ferrari, Ford, Subaru, General 
Motors, Honda, Hyundai, Lotus, Maserati, Mitsubishi, Nissan, 
Porsche, Suzuki, Toyota, and Volkswagen.
    \237\ The Regulatory Flexibility Act only requires analysis of 
small domestic manufacturers. There are four passenger car 
manufacturers we know of and no light truck manufacturers: Avanti, 
Panoz, Saleen, and Shelby.
---------------------------------------------------------------------------

D. Executive Order 13132 Federalism

    Executive Order 13132 requires NHTSA to develop an accountable 
process to ensure ``meaningful and timely input by State and local 
officials in the development of regulatory policies that have 
federalism implications.'' The Order defines the term ``Policies that 
have federalism implications' to include regulations that have 
``substantial direct effects on the States, on the relationship between 
the national government and the States, or on the distribution of power 
and responsibilities among the various levels of government.'' Under 
the Order, NHTSA may not issue a regulation that has federalism 
implications, that imposes substantial direct compliance costs, and 
that is not required by statute, unless the Federal government provides 
the funds necessary to pay the direct compliance costs incurred by 
State and local governments, or NHTSA consults with State and local 
officials early in the process of developing the proposed regulation. 
The agency has complied with Order's requirements.
    The issue of preemption of State emissions standard under EPCA is 
not a new one; there is an ongoing public dialogue regarding the 
preemptive impact of CAFE standards whose beginning pre-dates this 
rulemaking. This dialogue has involved a variety of parties (i.e., the 
States, the federal government and the general public) and has taken 
place through a variety of means, including several rulemaking 
proceedings. NHTSA first addressed the issue in its rulemaking on CAFE 
standards for MY 2005-2007 light trucks \238\ and explored it at great 
length, after receiving extensive public comment, in its rulemaking for 
MY 2008-2011 light trucks.\239\ Throughout this time, NHTSA has 
consistently taken the position that state regulations regulating 
CO2 tailpipe emissions from automobiles are expressly and 
impliedly preempted.
---------------------------------------------------------------------------

    \238\ 67 FR 77015, 77025; December 16, 2002, and 68 FR 16868, 
16895; April 7, 2003.
    \239\ 70 FR 51414, 51457; August 30, 2005, and 71 FR 17566, 
17654-17670; April 6, 2006.
---------------------------------------------------------------------------

    NHTSA's position remains unchanged, notwithstanding the occurrence 
of several significant events since the issuance of the final rule for 
MY 2008-2011 light trucks in April 2006. In 2007, the Supreme Court 
ruled Massachusetts v. EPA that carbon dioxide is an ``air pollutant'' 
within the meaning of the Clean Air Act and thus potentially subject to 
regulation under that statute. Later that year, two Federal district 
courts ruled in Vermont and California that the GHG motor vehicle 
emission standards adopted by those states are not preempted under 
EPCA. Still later that year, Congress enacted EISA, amending EPCA by 
mandating substantial and sustained annual increases in the passenger 
car and light truck CAFE standards. As further amended by EISA, EPCA 
also mandates that standards be attribute-based and established and 
implemented separately for passenger cars and light trucks. As it did 
before EISA, EPCA permits manufacturers to adjust their product mix on 
a national basis in order to achieve compliance while meeting consumer 
demand.
    NHTSA has carefully considered those events and reexamined the 
detailed technological and scientific analyses and conclusions it 
presented in its 2006 final rule. The agency reaffirms those analyses 
and conclusions.
    The Supreme Court did not consider the issue of preemption under 
EPCA of state regulations regulating CO2 tailpipe emissions 
from automobiles. Instead, it addressed the relationship of EPA and 
NHTSA rulemaking.
    We respectfully disagree with the two district court rulings. We 
note that an appeal has been filed concerning the Vermont decision and 
that the appellants' briefs have already been filed. EPCA's express 
preemption provision preempts state standards ``related to'' average 
fuel economy standards. Under the relatedness test, preemption is not 
dependent on the existence or nonexistence of any inconsistency or any 
difference between those State standards and the CAFE standards. 
Likewise, it is not dependent upon a state standard or a portion of a 
state standard's being identical to or equivalent to a CAFE standard.
    The enactment of EISA has increased the conflict between state 
regulations regulating CO2 tailpipe emissions from 
automobiles and EPCA. A conflict between state and federal law arises 
when compliance with both federal and state regulations is a physical 
impossibility or when state law stands as an obstacle to the 
accomplishment and execution of the full purposes and objectives of 
Congress. Contrary to the recommendations of NAS, the judgment of 
NHTSA, and the mandate of

[[Page 24479]]

Congress, the state regulations regulating CO2 tailpipe 
emissions, which are equivalent in effect to fuel economy standards, 
are not attribute-based, thus presenting risks to safety and 
employment. Contrary also to EISA, the state regulations do not 
establish separate standards.
    In reaffirming its position, NHTSA fully appreciates the great 
importance to the environment of addressing and reducing GHG emissions. 
Given that substantially reducing CO2 tailpipe emissions 
from automobiles is unavoidably and overwhelmingly dependent upon 
substantially increasing fuel economy through installation of engine 
technologies; transmission technologies; accessory technologies; 
vehicle technologies; and hybrid technologies, increases in fuel 
economy will produce commensurate reductions in CO2 tailpipe 
emissions. And as noted above, through EISA, Congress has ensured that 
there will be substantial and sustained, long term improvements in fuel 
economy.
    Given the importance of an effective, smooth functioning national 
program to improve fuel economy and in light of the fact that district 
court considered this agency's analysis and carefully crafted position 
on preemption, NHTSA is considering taking the further step of 
summarizing that position in appendices to be added to the parts in the 
Code of Federal Regulations setting forth the passenger car and light 
truck CAFE standards. That summary is as follows:

    (a) To the extent that any state regulation regulates tailpipe 
carbon dioxide emissions from automobiles, such a regulation relates 
to average fuel economy standards within the meaning of 49 U.S.C. 
32919.
    1. Automobile fuel economy is directly and very substantially 
related to automobile tailpipe emissions of carbon dioxide.
    2. Carbon dioxide is the natural by-product of automobile fuel 
consumption.
    3. The most significant and controlling factor in making the 
measurements necessary to determine the compliance of automobiles 
with the fuel economy standards in this Part is their rate of 
tailpipe carbon dioxide emissions.
    4. Most of the technologically feasible reduction of tailpipe 
emissions of carbon dioxide is achievable only through improving 
fuel economy, thereby reducing both the consumption of fuel and the 
creation and emission of carbon dioxide.
    5. Accordingly, as a practical matter, regulating fuel economy 
controls the amount of tailpipe emissions of carbon dioxide to a 
very substantial extent, and regulating the tailpipe emissions of 
carbon dioxide controls fuel economy to a very substantial extent.
    (b) As a state regulation related to fuel economy standards, any 
state regulation regulating tailpipe carbon dioxide emissions from 
automobiles is expressly preempted under 49 U.S.C. 32919.
    (c) A state regulation regulating tailpipe carbon dioxide 
emissions from automobiles, particularly a regulation that is not 
attribute-based and does not separately regulate passenger cars and 
light trucks, conflicts with
    1. The fuel economy standards in this Part,
    2. The judgments made by the agency in establishing those 
standards, and
    3. The achievement of the objectives of the statute (49 U.S.C. 
Chapter 329) under which those standards were established, including 
objectives relating to reducing fuel consumption in a manner and to 
the extent consistent with manufacturer flexibility, consumer 
choice, and automobile safety.
    (d) Any state regulation regulating tailpipe carbon dioxide 
emissions from automobiles is impliedly preempted under 49 U.S.C. 
Chapter 329.

    We have closely examined our authority and obligations under EPCA 
and that statute's express preemption provision. For those rulemaking 
actions undertaken at an agency's discretion, Section 3(a) of Executive 
Order 13132 instructs agencies to closely examine their statutory 
authority supporting any action that would limit the policymaking 
discretion of the States and assess the necessity for such action. This 
is not such a rulemaking action. NHTSA has no discretion not to issue 
the CAFE standards proposed in this document. EPCA mandates that the 
issuance of CAFE standards for passenger cars and light trucks for 
model years 2011-2015. Given that a State regulation for tailpipe 
emissions of CO2 is the functional equivalent of a CAFE 
standard, there is no way that NHTSA can tailor a fuel economy standard 
so as to avoid preemption. Further, EPCA itself precludes a State from 
adopting or enforcing a law or regulation related to fuel economy (49 
U.S.C. 32919(a)).

E. Executive Order 12988 (Civil Justice Reform)

    Pursuant to Executive Order 12988, ``Civil Justice Reform,'' \240\ 
NHTSA has considered whether this rulemaking would have any retroactive 
effect. This proposed rule does not have any retroactive effect.
---------------------------------------------------------------------------

    \240\ 61 FR 4729 (Feb. 7, 1996).
---------------------------------------------------------------------------

F. Unfunded Mandates Reform Act

    Section 202 of the Unfunded Mandates Reform Act of 1995 (UMRA) 
requires Federal agencies to prepare a written assessment of the costs, 
benefits, and other effects of a proposed or final rule that includes a 
Federal mandate likely to result in the expenditure by State, local, or 
tribal governments, in the aggregate, or by the private sector, of more 
than $100 million in any one year (adjusted for inflation with base 
year of 1995). Adjusting this amount by the implicit gross domestic 
product price deflator for 2006 results in $126 million (116.043/92.106 
= 1.26). Before promulgating a rule for which a written statement is 
needed, section 205 of UMRA generally requires NHTSA to identify and 
consider a reasonable number of regulatory alternatives and adopt the 
least costly, most cost-effective, or least burdensome alternative that 
achieves the objectives of the rule. The provisions of section 205 do 
not apply when they are inconsistent with applicable law. Moreover, 
section 205 allows NHTSA to adopt an alternative other than the least 
costly, most cost-effective, or least burdensome alternative if the 
agency publishes with the final rule an explanation why that 
alternative was not adopted.
    This proposed rule will not result in the expenditure by State, 
local, or tribal governments, in the aggregate, of more than $126 
million annually, but it will result in the expenditure of that 
magnitude by vehicle manufacturers and/or their suppliers. In 
promulgating this proposal, NHTSA considered a variety of alternative 
average fuel economy standards lower and higher than those proposed. 
NHTSA is statutorily required to set standards at the maximum feasible 
level achievable by manufacturers and has tentatively concluded that 
the proposed fuel economy standards are the maximum feasible standards 
for the passenger car fleet for MYs 2011-2015 and for the light truck 
fleet for MYs 2011-2015 in light of the statutory considerations.

G. Paperwork Reduction Act

    Under the procedures established by the Paperwork Reduction Act of 
1995, a person is not required to respond to a collection of 
information by a Federal agency unless the collection displays a valid 
OMB control number. The proposed rule would amend the reporting 
requirements under 49 CFR part 537, Automotive Fuel Economy Reports. In 
addition to the vehicle model information collected under the approved 
data collection (OMB control number 2127-0019) in Part 537, passenger 
car manufacturers would also be required to provide data on vehicle 
footprint. Manufacturers and other persons wishing to trade fuel 
economy credits would be required to provide an instruction to NHTSA on 
the credits to be traded.
    In compliance with the PRA, we announce that NHTSA is seeking 
comment on the proposed revisions to the collection.

[[Page 24480]]

    Agency: National Highway Traffic Safety Administration (NHTSA).
    Title: 49 CFR part 537, Automotive Fuel Economy (F.E.) Reports.
    Type of Request: Amend existing collection.
    OMB Clearance Number: 2127-0019.
    Form Number: This collection of information will not use any 
standard forms.
    Requested Expiration Date of Approval: Three years from the date of 
approval.

Summary of the Collection of Information

    NHTSA is proposing that manufacturers would be required to provide 
data on vehicle (including passenger car) footprint so that the agency 
could determine a manufacturer's required fuel economy level. This 
information collection would be included as part of the existing fuel 
economy reporting requirements. NHTSA is also proposing that 
manufacturers and other persons wishing to trade fuel economy credits 
would be required to provide an instruction to NHTSA on the credits to 
be traded.

Description of the Need for the Information and Proposed Use of the 
Information

    NHTSA would need the footprint information to determine a 
manufacturer's required fuel economy level and its compliance with that 
level. NHTSA would need the credit trading instruction to ensure that 
its records of a manufacturer's available credits are accurate in order 
to determine whether a manufacturer has sufficient credits available to 
offset any non-compliance with the CAFE requirements in a given year.

Description of the Likely Respondents (Including Estimated Number, and 
Proposed Frequency of Response to the Collection of Information)

    NHTSA estimates that 20 manufacturers would submit the required 
information. The frequency of reporting would not change from that 
currently authorized under collection number 2127-0019.

Estimate of the Total Annual Reporting and Recordkeeping Burden 
Resulting from the Collection of Information

    For footprint, NHTSA estimates that each passenger car manufacturer 
would incur an additional 10 burden hours per year. This estimate is 
based on the fact that data collection would involve only computer 
tabulation. Thus, each passenger car manufacturer would incur an 
additional burden of 10 hours or a total on industry of an additional 
200 hours a year (assuming there are 20 manufacturers). At an assumed 
rate of $21.23 an hour, the annual, estimated cost of collecting and 
preparing the additional passenger car footprint information is $4,246.
    For credit trading, NHTSA estimates that each instruction would 
incur an additional burden hour per year. This estimate is based on the 
fact that the data required is already available and thus the only 
burden is the actual preparation of the instruction. NHTSA estimates 
that the maximum instructions it would receive each year is 20. While 
non-manufacturers may also participate in credit trading, NHTSA does 
not believe that every manufacturer would need to, or be able to, 
participate in credit trading every year. NHTSA does not, at this time, 
have a way of estimating how many non-manufacturers may wish to 
participate in credit trading. Therefore NHTSA believes that the total 
number of manufacturers is a reasonable estimate, for a total annual 
additional burden of 20 hours a year. At an assumed rate of $21.23 an 
hour, the annual estimated cost of collecting and preparing the credit 
trading instruction is $425.
    NHTSA estimates that the recordkeeping burden resulting from the 
collection of information would be 0 hours because the information 
would be retained on each manufacturer's existing computer systems for 
each manufacturer's internal administrative purposes. There would be no 
capital or start-up costs as a result of this collection. Manufacturers 
can collect and tabulate the information by using existing equipment. 
Thus, there would be no additional costs to respondents or record 
keepers.
    NHTSA requests comment on its estimates of the total annual hour 
and cost burdens resulting from this collection of information. Please 
submit any comments to the NHTSA Docket Number referenced in the 
heading of this document, and to Ken Katz, Lead Engineer, Fuel Economy 
Division, Office of International Policy, Fuel Economy, and Consumer 
Programs, National Highway Traffic Safety Administration, 1200 New 
Jersey Avenue, SE., Washington, DC 20590. You may also contact him by 
phone at (202) 366-0846, by fax at (202) 493-2290, or by e-mail at 
ken.katz@dot.gov. Comments are due by July 1, 2008.

H. Regulation Identifier Number (RIN)

    The Department of Transportation assigns a regulation identifier 
number (RIN) to each regulatory action listed in the Unified Agenda of 
Federal Regulations. The Regulatory Information Service Center 
publishes the Unified Agenda in April and October of each year. You may 
use the RIN contained in the heading at the beginning of this document 
to find this action in the Unified Agenda.

I. Executive Order 13045

    Executive Order 13045 1A \241\ applies to any rule that: (1) Is 
determined to be economically significant as defined under E.O. 12866, 
and (2) concerns an environmental, health or safety risk that NHTSA has 
reason to believe may have a disproportionate effect on children. If 
the regulatory action meets both criteria, we must evaluate the 
environmental health or safety effects of the proposed rule on 
children, and explain why the proposed regulation is preferable to 
other potentially effective and reasonably feasible alternatives 
considered by us.
---------------------------------------------------------------------------

    \241\ 62 FR 19885 (Apr. 23, 1997).
---------------------------------------------------------------------------

    This proposed rule does not pose such a risk for children. The 
primary effects of this proposal are to conserve energy and to reduce 
tailpipe emissions of CO2, the primary greenhouse gas, by 
setting fuel economy standards for motor vehicles.

J. National Technology Transfer and Advancement Act

    Section 12(d) of the National Technology Transfer and Advancement 
Act (NTTAA) requires NHTSA to evaluate and use existing voluntary 
consensus standards in its regulatory activities unless doing so would 
be inconsistent with applicable law (e.g., the statutory provisions 
regarding NHTSA's vehicle safety authority) or otherwise impractical.
    Voluntary consensus standards are technical standards developed or 
adopted by voluntary consensus standards bodies. Technical standards 
are defined by the NTTAA as ``performance-based or design-specific 
technical specification and related management systems practices.'' 
They pertain to ``products and processes, such as size, strength, or 
technical performance of a product, process or material.''
    Examples of organizations generally regarded as voluntary consensus 
standards bodies include the American Society for Testing and Materials 
(ASTM), the Society of Automotive Engineers (SAE), and the American 
National Standards Institute (ANSI). If NHTSA does not use available 
and potentially applicable voluntary

[[Page 24481]]

consensus standards, we are required by the Act to provide Congress, 
through OMB, an explanation of the reasons for not using such 
standards.
    The document proposes to categorize passenger cars according to 
vehicle footprint (average track width X wheelbase). For purposes of 
this calculation, NHTSA proposes to base these measurements on those 
developed by the automotive industry. Determination of wheelbase would 
be consistent with L101-wheelbase, defined in SAE J1100 MAY95, Motor 
vehicle dimensions. NHTSA's proposal uses a modified version of the SAE 
definitions for track width (W101-tread-front and W102-tread-rear as 
defined in SAE J1100 MAY95). The proposed definition of track width 
reduces a manufacturer's ability to adjust a vehicle's track width 
through minor alterations.

K. Executive Order 13211

    Executive Order 13211 \242\ applies to any rule that: (1) Is 
determined to be economically significant as defined under E.O. 12866, 
and is likely to have a significant adverse effect on the supply, 
distribution, or use of energy; or (2) that is designated by the 
Administrator of the Office of Information and Regulatory Affairs as a 
significant energy action. If the regulatory action meets either 
criterion, we must evaluate the adverse energy effects of the proposed 
rule and explain why the proposed regulation is preferable to other 
potentially effective and reasonably feasible alternatives considered 
by us.
---------------------------------------------------------------------------

    \242\ 66 FR 28355 (May 18, 2001).
---------------------------------------------------------------------------

    The proposed rule seeks to establish passenger car and light truck 
fuel economy standards that will reduce the consumption of petroleum 
and will not have any adverse energy effects. Accordingly, this 
proposed rulemaking action is not designated as a significant energy 
action.

L. Department of Energy Review

    In accordance with 49 U.S.C. 32902(j)(1), we submitted this 
proposed rule to the Department of Energy for review. That Department 
did not make any comments that we have not addressed.

M. Plain Language

    Executive Order 12866 requires each agency to write all rules in 
plain language. Application of the principles of plain language 
includes consideration of the following questions:
     Have we organized the material to suit the public's needs?
     Are the requirements in the rule clearly stated?
     Does the rule contain technical language or jargon that 
isn't clear?
     Would a different format (grouping and order of sections, 
use of headings, paragraphing) make the rule easier to understand?
     Would more (but shorter) sections be better?
     Could we improve clarity by adding tables, lists, or 
diagrams?
     What else could we do to make the rule easier to 
understand?
    If you have any responses to these questions, please include them 
in your comments on this proposal.

N. Privacy Act

    Anyone is able to search the electronic form of all comments 
received into any of our dockets by the name of the individual 
submitting the comment (or signing the comment, if submitted on behalf 
of an organization, business, labor union, etc.). You may review DOT's 
complete Privacy Act statement in the Federal Register published on 
April 11, 2000 (Volume 65, Number 70; Pages 19477-78) or you may visit 
http://www.dot.gov/privacy.html.

XIV. Regulatory Text

List of Subjects in 49 CFR Parts 523, 531, 533, 534, 535, 536, and 
537

    Fuel economy and Reporting and recordkeeping requirements.
    In consideration of the foregoing, under the authority of 49 U.S.C. 
32901, 32902, 32903, and 32907, and delegation of authority at 49 CFR 
1.50, NHTSA proposes to amend 49 CFR Chapter V as follows:

PART 523--VEHICLE CLASSIFICATION

    1. Amend the authority citation for part 523 by revising to read as 
follows:

    Authority: 49 U.S.C. 32901, delegation of authority at 49 CFR 
1.50.

    2. Amend Sec.  523.2 by adding, in alphabetical order, definitions 
of ``light truck'' and ``work truck'' to read as follows:


Sec.  523.2  Definitions.

* * * * *
    Light truck means a non-passenger automobile as defined in Sec.  
523.5.
* * * * *
    Work truck means a vehicle that is rated at more than 8,500 and 
less than or equal to 10,000 pounds gross vehicle weight, and is not a 
medium-duty passenger vehicle as defined in 40 CFR 86.1803-01 as in 
effect on December 20, 2007.
    3. Amend Sec.  523.3 by revising paragraph (a) to read as follows:


Sec.  523.3  Automobile.

    (a) An automobile is any 4-wheeled vehicle that is propelled by 
fuel, or by alternative fuel, manufactured primarily for use on public 
streets, roads, and highways and rated at less than 10,000 pounds gross 
vehicle weight, except:
    (1) A vehicle operated only on a rail line;
    (2) A vehicle manufactured in different stages by 2 or more 
manufacturers, if no intermediate or final-stage manufacturer of that 
vehicle manufactures more than 10,000 multi-stage vehicles per year; or
    (3) A work truck.
* * * * *
    4. Amend Sec.  523.5 by revising the introductory text, and 
paragraphs (a) introductory text, (b) introductory text, (b)(1), and 
(b)(2) introductory text to read as follows:


Sec.  523.5  Non-passenger automobile.

    A non-passenger automobile means an automobile that is not a 
passenger automobile or a work truck and includes vehicles described in 
paragraphs (a) or (b) of this section:
    (a) An automobile designed to perform at least one of the following 
functions:
* * * * *
    (b) An automobile capable of off-highway operation, as indicated by 
the fact that it:
    (1)(i) Has 4-wheel drive or
    (ii) Is rated at more than 6,000 pounds gross vehicle weight; and
    (2) Has at least four of the following characteristics--
* * * * *

PART 531--PASSENGER AUTOMOBILE AVERAGE FUEL ECONOMY STANDARDS

    5. The authority citation for part 531 continues to read as 
follows:

    Authority: 49 U.S.C. 32902; delegation of authority at 49 CFR 
1.50.

    6. Amend Sec.  531.5 by revising paragraph (a), redesignating 
paragraph (b) as paragraph (d), and adding paragraphs (b) and (c) to 
read as follows:


Sec.  531.5  Fuel economy standards.

    (a) Except as provided in paragraph (d) of this section, each 
manufacturer of passenger automobiles shall comply with the average 
fuel economy standards in Table I, expressed in miles per gallon, in 
the model year specified as applicable:

[[Page 24482]]



                                 Table I
------------------------------------------------------------------------
                       Model year                            Standard
------------------------------------------------------------------------
1978....................................................            18.0
1979....................................................            19.0
1980....................................................            20.0
1981....................................................            22.0
1982....................................................            24.0
1983....................................................            26.0
1984....................................................            27.0
1985....................................................            27.5
1986....................................................            26.0
1987....................................................            26.0
1988....................................................            26.0
1989....................................................            26.5
1990-2010...............................................            27.5
------------------------------------------------------------------------

    (b) For each of model years 2011 through 2015, a manufacturer's 
passenger automobile fleet shall comply with the fuel economy level 
calculated for that model year according to Figure 1 and the 
appropriate values in Table II.
[GRAPHIC] [TIFF OMITTED] TP02MY08.042

Where:
    N is the total number (sum) of passenger automobiles produced by 
a manufacturer,
    Ni is the number (sum) of the ith model 
passenger automobile produced by the manufacturer, and
Ti is fuel economy target of the ith model 
passenger automobile, which is determined according to the following 
formula, rounded to the nearest hundredth:
[GRAPHIC] [TIFF OMITTED] TP02MY08.043

Where,
    Parameters a, b, c, and d are defined in Table II;
    e = 2.718; and
    x = footprint (in square feet, rounded to the nearest tenth) of 
the vehicle model.

                     Table II.--Parameters for the Passenger Automobile Fuel Economy Targets
----------------------------------------------------------------------------------------------------------------
                                                                                  Parameters
                         Model year                          ---------------------------------------------------
                                                                   a            b            c            d
----------------------------------------------------------------------------------------------------------------
2011........................................................        38.20        25.80        45.88         1.60
2012........................................................        40.00        27.40        45.79         1.54
2013........................................................        40.80        28.70        45.70         1.48
2014........................................................        41.20        29.90        45.61         1.42
2015........................................................        41.70        31.20        45.51         1.36
----------------------------------------------------------------------------------------------------------------

    (c) In addition to the requirement of paragraph (b) of this 
section, each manufacturer shall also meet the minimum standard for 
domestically manufactured passenger automobiles expressed in Table III:

                                Table III
------------------------------------------------------------------------
                                                               Minimum
                         Model year                            standard
------------------------------------------------------------------------
2011.......................................................         28.7
2012.......................................................         30.2
2013.......................................................         31.3
2014.......................................................         32.0
2015.......................................................         32.9
------------------------------------------------------------------------

* * * * *
    7. Part 531 is amended by adding the following new Appendix A at 
the end:

Appendix A to Part 531--Preemption of State Regulations Regulating 
Tailpipe Carbon Dioxide Emissions From Automobiles

    (a) To the extent that any state regulation regulates tailpipe 
carbon dioxide emissions from automobiles, such a regulation relates 
to average fuel economy standards within the meaning of 49 U.S.C. 
32919.
    1. Automobile fuel economy is directly and very substantially 
related to automobile tailpipe emissions of carbon dioxide.
    2. Carbon dioxide is the natural by-product of automobile fuel 
consumption.
    3. The most significant and controlling factor in making the 
measurements necessary to determine the compliance of automobiles 
with the fuel economy standards in this Part is their rate of 
tailpipe carbon dioxide emissions.
    4. Most of the technologically feasible reduction of tailpipe 
emissions of carbon dioxide is achievable only through improving 
fuel economy, thereby reducing both the consumption of fuel and the 
creation and emission of carbon dioxide.
    5. Accordingly, as a practical matter, regulating fuel economy 
controls the amount of tailpipe emissions of carbon dioxide to a 
very substantial extent, and regulating the tailpipe emissions of 
carbon dioxide controls fuel economy to a very substantial extent.
    (b) As a state regulation related to fuel economy standards, any 
state regulation regulating tailpipe carbon dioxide emissions from 
automobiles is expressly preempted under 49 U.S.C. 32919.
    (c) A state regulation regulating tailpipe carbon dioxide 
emissions from automobiles, particularly a regulation that is not 
attribute-based and does not separately regulate passenger cars and 
light trucks, conflicts with
    1. The fuel economy standards in this Part,
    2. The judgments made by the agency in establishing those 
standards, and
    3. The achievement of the objectives of the statute (49 U.S.C. 
Chapter 329) under which those standards were established, including 
objectives relating to reducing fuel consumption in a manner and to 
the extent consistent with manufacturer flexibility, consumer 
choice, and automobile safety.
    (d) Any state regulation regulating tailpipe carbon dioxide 
emissions from automobiles is impliedly preempted under 49 U.S.C. 
Chapter 329.

PART 533--LIGHT TRUCK FUEL ECONOMY STANDARDS

    8. The authority citation for part 533 continues to read as 
follows:

    Authority: 49 U.S.C. 32902; delegation of authority at 49 CFR 
1.50.

    9. Amend Sec.  533.5 by revising Table V of paragraph (a) and 
revising paragraph (h) to read as follows:

[[Page 24483]]

Sec.  533.5  Requirements.

    (a) * * *

                          Table V.--Parameters for the Light Truck Fuel Economy Targets
----------------------------------------------------------------------------------------------------------------
                                                                            Parameters
                   Model year                    ---------------------------------------------------------------
                                                         a               b               c               d
----------------------------------------------------------------------------------------------------------------
2008............................................           28.56           19.99           49.30            5.58
2009............................................           30.07           20.87           48.00            5.81
2010............................................           29.96           21.20           48.49            5.50
2011............................................           30.90           21.50           51.94            3.80
2012............................................           32.70           22.80           51.98            3.82
2013............................................           34.10           23.80           52.02            3.84
2014............................................           34.10           24.30           52.06            3.86
2015............................................           34.30           24.80           52.11            3.87
----------------------------------------------------------------------------------------------------------------

* * * * *
    (h) For each of model years 2011-2015, a manufacturer's light truck 
fleet shall comply with the fuel economy level calculated for that 
model year according to Figure 1 and the appropriate values in Table V.
    10. Part 533 is amended by adding the following new Appendix B at 
the end:

Appendix B to Part 533--Preemption of state regulations regulating 
tailpipe carbon dioxide emissions from automobiles

    (a) To the extent that any state regulation regulates tailpipe 
carbon dioxide emissions from automobiles, such a regulation relates 
to average fuel economy standards within the meaning of 49 U.S.C. 
32919.
    1. Automobile fuel economy is directly and very substantially 
related to automobile tailpipe emissions of carbon dioxide.
    2. Carbon dioxide is the natural by-product of automobile fuel 
consumption.
    3. The most significant and controlling factor in making the 
measurements necessary to determine the compliance of automobiles 
with the fuel economy standards in this Part is their rate of 
tailpipe carbon dioxide emissions.
    4. Most of the technologically feasible reduction of tailpipe 
emissions of carbon dioxide is achievable only through improving 
fuel economy, thereby reducing both the consumption of fuel and the 
creation and emission of carbon dioxide.
    5. Accordingly, as a practical matter, regulating fuel economy 
controls the amount of tailpipe emissions of carbon dioxide to a 
very substantial extent, and regulating the tailpipe emissions of 
carbon dioxide controls fuel economy to a very substantial extent.
    (b) As a state regulation related to fuel economy standards, any 
state regulation regulating tailpipe carbon dioxide emissions from 
automobiles is expressly preempted under 49 U.S.C. 32919.
    (c) A state regulation regulating tailpipe carbon dioxide 
emissions from automobiles, particularly a regulation that is not 
attribute-based and does not separately regulate passenger cars and 
light trucks, conflicts with
    1. The fuel economy standards in this Part,
    2. The judgments made by the agency in establishing those 
standards, and
    3. The achievement of the objectives of the statute (49 U.S.C. 
Chapter 329) under which those standards were established, including 
objectives relating to reducing fuel consumption in a manner and to 
the extent consistent with manufacturer flexibility, consumer 
choice, and automobile safety.
    (d) Any state regulation regulating tailpipe carbon dioxide 
emissions from automobiles is impliedly preempted under 49 U.S.C. 
Chapter 329.

PART 534--RIGHTS AND RESPONSIBILITIES OF MANUFACTURERS IN THE 
CONTEXT OF CHANGES IN CORPORATE RELATIONSHIPS

    11. The authority citation for part 534 continues to read as 
follows:

    Authority: 49 U.S.C. 32901; delegation of authority at 49 CFR 
1.50.

    12. Amend Sec.  534.4 by revising paragraphs (c) and (d) to read as 
follows:


Sec.  534.4  Successors and predecessors.

* * * * *
    (c) Credits earned by a predecessor before or during model year 
2008 may be used by a successor, subject to the availability of credits 
and the general three-year restriction on carrying credits forward and 
the general three-year restriction on carrying credits backward. 
Credits earned by a predecessor after model year 2008 may be used by a 
successor, subject to the availability of credits and the general five-
year restriction on carrying credits forward and the general three-year 
restriction on carrying credits backward.
    (d) Credits earned by a successor before or during model year 2008 
may be used to offset a predecessor's shortfall, subject to the 
availability of credits and the general three-year restriction on 
carrying credits forward and the general three-year restriction on 
carrying credits backward. Credits earned by a successor after model 
year 2008 may be used to offset a predecessor's shortfall, subject to 
the availability of credits and the general five-year restriction on 
carrying credits forward and the general three-year restriction on 
carrying credits backward.
    13. Amend Sec.  534.5 by revising paragraphs (c) and (d) to read as 
follows:


Sec.  534.5  Manufacturers within control relationships.

* * * * *
    (c) Credits of a manufacturer within a control relationship may be 
used by the group of manufacturers within the control relationship to 
offset shortfalls, subject to the agreement of the other manufacturers, 
the availability of the credits, and the general three-year restriction 
on carrying credits forward or backward prior to or during model year 
2008, or the general five-year restriction on carrying credits forward 
and the general three-year restriction on carrying credits backward 
after model year 2008.
    (d) If a manufacturer within a group of manufacturers is sold or 
otherwise spun off so that it is no longer within that control 
relationship, the manufacturer may use credits that were earned by the 
group of manufacturers within the control relationship while the 
manufacturer was within that relationship, subject to the agreement of 
the other manufacturers, the availability of the credits, and the 
general three-year restriction on carrying credits forward or backward 
prior to or during model year 2008, or the general five-year 
restriction on carrying credits forward and the general three-year 
restriction on carrying credits backward after model year 2008.
* * * * *

PART 535--[REMOVED]

    14. Remove Part 535.
    15. Part 536 is added to read as follows:

[[Page 24484]]

PART 536--TRANSFER AND TRADING OF FUEL ECONOMY CREDITS

Sec.
536.1 Scope.
536.2 Application.
536.3 Definitions.
536.4 Credits.
536.5 Trading infrastructure.
536.6 Treatment of credits earned prior to model year 2011.
536.7 Treatment of carryback credits.
536.8 Conditions for trading of credits.
536.9 Use of credits with regard to the domestically manufactured 
passenger automobile minimum standard.
536.10 Treatment of dual-fuel and alternative fuel vehicles--
consistency with 49 CFR Part 538.

    Authority: 49 U.S.C. 32903; delegation of authority at 49 CFR 
1.50.


Sec.  536.1  Scope.

    This part establishes regulations governing the use and application 
of CAFE credits up to three model years before and five model years 
after the model year in which the credit was earned. It also specifies 
requirements for manufacturers wishing to transfer fuel economy credits 
between their fleets and for manufacturers and other persons wishing to 
trade fuel economy credits to achieve compliance with prescribed fuel 
economy standards.


Sec.  536.2  Application.

    This part applies to all credits earned (and transferable and 
tradable) for exceeding applicable average fuel economy standards in a 
given model year for domestically manufactured passenger cars, imported 
passenger cars, and light trucks.


Sec.  536.3  Definitions.

    (a) Statutory terms. In this part, all terms defined in 49 U.S.C. 
32901(a) are used in their statutory meaning.
    (b) Other terms. As used in this part:
    Above standard fuel economy means, with respect to a compliance 
category, that the automobiles manufactured by a manufacturer in that 
compliance category in a particular model year have greater average 
fuel economy (calculated in a manner that reflects the incentives for 
alternative fuel automobiles per 49 U.S.C. 32905) than that 
manufacturer's fuel economy standard for that compliance category and 
model year.
    Adjustment factor means a factor used to adjust the value of a 
traded credit for compliance purposes to ensure that the compliance 
value of the credit reflects the total volume of oil saved when the 
credit was earned.
    Below standard fuel economy means, with respect to a compliance 
category, that the automobiles manufactured by a manufacturer in that 
compliance category in a particular model year have lower average fuel 
economy (calculated in a manner that reflects the incentives for 
alternative fuel automobiles per 49 U.S.C. 32905) than that 
manufacturer's fuel economy standard for that compliance category and 
model year.
    Compliance. (1) Compliance means a manufacturer achieves compliance 
in a particular compliance category when:
    (i) The average fuel economy of the vehicles in that category 
exceed or meet the fuel economy standard for that category, or
    (ii) The average fuel economy of the vehicles in that category do 
not meet the fuel economy standard for that category, but the 
manufacturer proffers a sufficient number of valid credits, adjusted 
for total oil savings, to cover the gap between the average fuel 
economy of the vehicles in that category and the required average fuel 
economy.
    (2) A manufacturer achieves compliance for its fleet if conditions 
(1)(i) or (1)(ii) of this definition are simultaneously met for all 
compliance categories.
    Compliance category means any of three categories of automobiles 
subject to Federal fuel economy regulations. The three compliance 
categories recognized by 49 U.S.C. 32903(g)(6) are domestically 
manufactured passenger automobiles, imported passenger automobiles, and 
non-passenger automobiles (``light trucks'').
    Credit holder (or holder) means a legal person that has valid 
possession of credits, either because they are a manufacturer who has 
earned credits by exceeding an applicable fuel economy standard, or 
because they are a designated recipient who has received credits from 
another holder. Credit holders need not be manufacturers, although all 
manufacturers may be credit holders.
    Credits (or fuel economy credits) means an earned or purchased 
allowance recognizing that the average fuel economy of a particular 
manufacturer's vehicles within a particular compliance category and 
model year exceeds that manufacturer's fuel economy standard for that 
compliance category and model year. One credit is equal to \1/10\ of a 
mile per gallon above the fuel economy standard per one vehicle within 
a compliance category. Credits are denominated according to model year 
in which they are earned (vintage), originating manufacturer, and 
compliance category.
    Expiry date means the model year after which fuel economy credits 
may no longer be used to achieve compliance with fuel economy 
regulations. Expiry Dates are calculated in terms of model years: For 
example, if a manufacturer earns credits for model year 2011, these 
credits may be used for compliance in model years 2008-2016.
    Fleet means all automobiles that are manufactured by a manufacturer 
in a particular model year and are subject to fuel economy standards 
under 49 CFR Part 531 and 533. For the purposes of this regulation, a 
manufacturer's fleet means all domestically manufactured and imported 
passenger automobiles and non-passenger automobiles (``light trucks''). 
``Work trucks'' and medium and heavy trucks are not included in this 
definition for purposes of this regulation.
    Light truck means the same as ``non-passenger automobile,'' as that 
term is defined in 49 U.S.C. 32901(a)(17), and as ``light truck,'' as 
that term is defined at 49 CFR 523.5.
    Originating manufacturer means the manufacturer that originally 
earned a particular credit. Each credit earned will be identified with 
the name of the originating manufacturer.
    Trade means the receipt by NHTSA of an instruction from a credit 
holder to place one of its credits in the account of another credit 
holder. A credit that has been traded can be identified because the 
originating manufacturer will be a different party than the current 
credit holder. If a credit has been traded to another credit holder and 
is subsequently traded back to the originating manufacture, it will be 
deemed not to have been traded for compliance purposes.
    Transfer means the application by a manufacturer of credits earned 
by that manufacturer in one compliance category or credits acquired by 
trade (and originally earned by another manufacturer in that category) 
to achieve compliance with fuel economy standards with respect to a 
different compliance category. For example, a manufacturer may purchase 
light truck credits from another manufacturer, and transfer them to 
achieve compliance in the manufacturer's domestically manufactured 
passenger car fleet.
    Vintage means, with respect to a credit, the model year in which 
the credit was earned.


Sec.  536.4  Credits.

    (a) Type and vintage. All credits are identified and distinguished 
in the accounts by originating manufacturer, compliance category, and 
model year of origin (vintage).
    (b) Application of credits. All credits earned and applied are 
calculated, per 49 U.S.C. 32903(c), in tenths of a mile per gallon by 
which the average fuel economy of vehicles in a particular

[[Page 24485]]

compliance category manufactured by a manufacturer in the model year in 
which the credits are earned exceeds the applicable average fuel 
economy standard, multiplied by the number of vehicles sold in that 
compliance category. However, credits that have been traded, defined as 
credits that are used for compliance by a manufacturer other than the 
originating manufacturer, are valued for compliance purposes using the 
adjustment factor specified in paragraph (c) of this section, pursuant 
to the ``total oil savings'' requirement of 49 U.S.C. 32903(f)(1).
    (c) Adjustment factor. Vehicle fuel economy, measured in miles per 
gallon (mpg), is adjusted to ensure constant oil savings when traded 
between manufacturers. Adjusted mpg is shown by multiplying the value 
of each credit (with a nominal value of 0.1 mpg per vehicle) by an 
adjustment factor calculated by the following formula:
[GRAPHIC] [TIFF OMITTED] TP02MY08.044

Where:

A = Adjustment Factor applied to traded credits by multiplying mpg 
for a particular credit;
VMTe = Lifetime vehicle miles traveled for the compliance category 
in which the credit was earned: 152,000 miles for domestically 
manufactured and imported passenger cars, 179,000 miles for light 
trucks;
VMTu = Lifetime vehicle miles traveled for the compliance category 
in which the credit is used for compliance: 152,000 miles for 
domestically manufactured and imported passenger cars, 179,000 miles 
for light trucks;
MPGe = Fuel economy standard for the originating manufacturer, 
compliance category, and model year in which the credit was earned;
MPGu = Fuel economy standard for the manufacturer, compliance 
category, and model year in which the credit will be used.


Sec.  536.5  Trading Infrastructure.

    (a) Accounts. NHTSA maintains ``accounts'' for each credit holder. 
The account consists of a balance of credits in each compliance 
category and vintage held by the holder.
    (b) Who may hold credits. Every manufacturer subject to fuel 
economy standards under 49 CFR parts 531 or 533 is automatically an 
account holder. If the manufacturer earns credits pursuant to this 
part, or receives credits from another party, so that the 
manufacturer's account has a non-zero balance, then the manufacturer is 
also a credit holder. Any party designated as a recipient of credits by 
a current credit holder will receive an account from NHTSA and become a 
credit holder, subject to the following conditions:
    (1) A designated recipient must provide name, address, contacting 
information, and a valid taxpayer identification number or social 
security number;
    (2) NHTSA does not grant a request to open a new account by any 
party other than a party designated as a recipient of credits by a 
credit holder;
    (3) NHTSA maintains accounts with zero balances for a period of 
time, but reserves the right to close accounts that have had zero 
balances for more than one year.
    (c) Automatic debits and credits of accounts.
    (1) Upon receipt of a verified instruction to trade credits from an 
existing credit holder, NHTSA verifies the presence of sufficient 
credits in the account of the trader, then debit the account of the 
trader and credit the account of the recipient with credits of the 
vintage, origin, and compliance category designated. If the recipient 
is not a current account holder, NHTSA establishes the account subject 
to the conditions described in paragraph (b) of this section, and 
shifts the credits to the newly-opened account.
    (2) NHTSA automatically deletes unused credits from holders' 
accounts as they reach their expiry date.
    (d) Compliance.
    (1) NHTSA assesses compliance with fuel economy standards each 
year, utilizing the certified and reported CAFE data provided by the 
Environmental Protection Agency for enforcement of the CAFE program 
pursuant to 49 U.S.C. 32904(e). Credit values are calculated based on 
the CAFE data from the EPA. If a particular compliance category within 
a manufacturer's fleet has above standard fuel economy, NHTSA adds 
credits to the manufacturer's account for that compliance category and 
vintage in the appropriate amount by which the manufacturer has 
exceeded the applicable standard.
    (2) If a manufacturer's vehicles in a particular compliance 
category have below standard fuel economy, NHTSA automatically debits 
the manufacturer's unexpired credits, earned or obtained through 
trading, within the compliance category from the manufacturer's 
account, beginning with the oldest credits held by the manufacturer.
    (3) If there are insufficient credits within the compliance 
category to enable the manufacturer to achieve compliance in that 
category, NHTSA automatically transfers any available existing surplus 
credits, including credits obtained through trading, from other 
compliance categories to the extent permitted by 49 U.S.C. 32903(g)(3) 
and this regulation, beginning with the oldest vintage of available 
surplus credits.
    (4) The value, when used for compliance, of any credits received 
via trade is adjusted, using the adjustment factor described in Sec.  
536.4(c), pursuant to 49 U.S.C. 32902(f)(1).
    (5) If a manufacturer is still unable to comply with the applicable 
standards for one or more compliance categories after NHTSA has applied 
all available credits from within and without the compliance category, 
NHTSA shall inform the manufacturer of its non-compliant status and 
their liability for fines, which may be avoided by submitting 
additional credits obtained through trading, or deferred by submitting 
a carryback plan for NHTSA's approval pursuant to 49 U.S.C. 
32903(b)(2).
    (6) NHTSA will enforce the CAFE program using the certified and 
reported CAFE values provided by the Environmental Protection Agency as 
required by 49 U.S.C. 32904(c) and (e). Credit values will be 
calculated from the CAFE numbers issued from EPA.
    (e) Reporting.
    (1) NHTSA periodically publishes the names and credit holdings of 
all credit holders. NHTSA does not publish individual transactions, nor 
respond to individual requests for updated balances from any party 
other than the account holder.

[[Page 24486]]

    (2) NHTSA issues an annual credit status letter to each party that 
is a credit holder at that time. The letter to a credit holder includes 
a credit accounting record that identifies the credit status of the 
credit holder including any activity (earned, expired, transferred, 
traded, carry-forward and carry-back credit transactions/allocations) 
that took place during the identified activity period.


Sec.  536.6  Treatment of credits earned prior to model year 2011.

    (a) Credits earned in a compliance category before and during model 
year 2008 may be applied by the manufacturer that earned them to 
carryback plans for that compliance category approved up to three model 
years prior to the year in which the credits were earned, or may be 
applied to compliance in that compliance category for up to three model 
years after the year in which the credits were earned.
    (b) Credits earned in a compliance category after model year 2008 
may be applied by the manufacturer that earned them to carryback plans 
for that compliance category approved up to three years prior to the 
year in which the credits were earned, or may be held or applied for up 
to five model years after the year in which the credits were earned.
    (c) Credits earned in a compliance category prior to model year 
2011 may not be transferred or traded by a manufacturer to another 
compliance category.


Sec.  536.7  Treatment of carryback credits.

    (a) Credits earned in a compliance category in any model year may 
be used in carryback plans approved by NHTSA, pursuant to 49 U.S.C. 
32903(b), for up to three model years prior to the year in which the 
credit was earned.
    (b) For purposes of this regulation, NHTSA will treat the use of 
future credits for compliance, as through a carryback plan, as a 
deferral of penalties for non-compliance with an applicable fuel 
economy standard.
    (c) If NHTSA receives and approves a manufacturer's carryback plan 
to earn future credits within the following three model years in order 
to comply with current regulatory obligations, NHTSA will defer levying 
fines for non-compliance until the date(s) when the manufacturer's 
approved plan indicates that credits will be earned or acquired to 
achieve compliance, and upon receiving confirmed CAFE data from EPA. If 
the manufacturer fails to acquire or earn sufficient credits by the 
plan dates, NHTSA will initiate compliance proceedings.
    (d) In the event that NHTSA fails to receive or approve a plan for 
a non-compliant manufacturer, NHTSA will levy fines pursuant to 
statute. If within three years, the non-compliant manufacturer earns or 
acquires additional credits to reduce or eliminate the non-compliance, 
NHTSA will reduce any fines owed, or repay fines to the extent that 
credits received reduce the non-compliance.
    (e) No credits from any source will be accepted in lieu of 
compliance after three model years after the non-compliance.
    (f) If a manufacturer is unable to comply in any compliance 
category in any model year, NHTSA will automatically deduct and 
extinguish any eligible credits subsequently held, earned, or acquired 
to reduce the oldest instance of non-compliance before allowing credits 
to accumulate or applying credits to achieve compliance in later years.
    (g) A carryback plan may not include the use of credits earned 
before model year 2011 that have been subsequently traded or 
transferred to another party.


Sec.  536.8  Conditions for trading of credits.

    (a) Trading of credits. If a credit holder wishes to trade credits 
to another party, the current credit holder and the receiving party 
must jointly issue an instruction to NHTSA, identifying the quantity, 
vintage, compliance category, and originator of the credits to be 
traded. If the recipient is not a current account holder, the recipient 
must provide sufficient information for NHTSA to establish an account 
for the recipient. Once an account has been established or identified 
for the recipient, NHTSA completes the trade by debiting the 
transferor's account and crediting the recipient's account. NHTSA will 
track the quantity, vintage, compliance category, and originator of all 
credits held or traded by all account-holders.
    (b) Trading between and within compliance categories. For credits 
earned in model year 2011 or thereafter, and used to satisfy compliance 
obligations for model year 2011 or thereafter:
    (1) Manufacturers may use credits originally earned by another 
manufacturer in a particular compliance category to satisfy compliance 
obligations within the same compliance category.
    (2) Once a manufacturer acquires by trade credits originally earned 
by another manufacturer in a particular compliance category, the 
manufacturer may transfer the credits to satisfy its compliance 
obligations in a different compliance category, but only to the extent 
that the CAFE increase attributable to the transferred credits does not 
exceed the limits in 49 U.S.C. 32903(g)(3). For any compliance 
category, the sum of a manufacturer's transferred credits earned by 
that manufacturer and transferred credits obtained by that manufacturer 
through trade must not exceed that limit.
    (c) Changes in corporate ownership and control. Manufacturers must 
inform NHTSA of corporate relationship changes to ensure that credit 
accounts are identified correctly and credits are assigned and 
allocated properly.
    (1) In general, if two manufacturers merge in any way, they must 
inform NHTSA how they plan to merge their credit accounts. NHTSA will 
subsequently assess corporate fuel economy and compliance status of the 
merged fleet instead of the original separate fleets.
    (2) If a manufacturer divides or divests itself of a portion of its 
automobile manufacturing business, it must inform NHTSA how it plans to 
divide the manufacturer's credit holdings into two or more accounts. 
NHTSA will subsequently distribute holdings as directed by the 
manufacturer, subject to provision for reasonably anticipated 
compliance obligations.
    (3) If a manufacturer is a successor to another manufacturer's 
business, it must inform NHTSA how it plans to allocate credits and 
resolve liabilities per 49 CFR part 534, Rights and Responsibilities of 
Manufacturers in the Context of Corporate Relationships.
    (d) No short or forward sales. NHTSA will not honor any 
instructions to trade or transfer more credits than are currently held 
in any account. NHTSA will not honor instructions to trade or transfer 
credits from any future vintage (i.e., credits not yet earned). NHTSA 
will not participate in or facilitate contingent trades.
    (e) Cancellation of credits. A credit holder may instruct NHTSA to 
cancel its currently held credits, specifying the originating 
manufacturer, vintage, and compliance category of the credits to be 
cancelled. These credits will be permanently null and void; NHTSA will 
remove the specific credits from the credit holder's account, and will 
not reissue them to any other party.
    (f) Errors or fraud in earning credits. If NHTSA determines that a 
manufacturer has been credited, through error or fraud, with earning 
credits, NHTSA will cancel those credits if possible. If the 
manufacturer credited with having earned those credits has already 
traded them when the error or

[[Page 24487]]

fraud is discovered, NHTSA will hold the receiving manufacturer 
responsible for returning the same or equivalent credits to NHTSA for 
cancellation.
    (g) Error or fraud in trading. In general, all trades are final and 
irrevocable once executed, and may only be reversed by a new, mutually-
agreed transaction. If NHTSA executes an erroneous instruction to trade 
credits from one holder to another through error or fraud, NHTSA will 
reverse the transaction if possible. If those credits have been traded 
away, the recipient holder is responsible for obtaining the same or 
equivalent credits for return to the previous holder.


Sec.  536.9  Use of credits with regard to the domestically 
manufactured passenger automobile minimum standard.

    (a) Transferred or traded credits may not be used, pursuant to 49 
U.S.C. 32903(g)(4), to meet the domestically manufactured passenger 
automobile minimum standard specified in 49 U.S.C. 32902(b)(4).
    (b) Each manufacturer is responsible for compliance with both the 
minimum standard and the attribute-based standard.
    (c) If a manufacturer's average fuel economy level for domestically 
manufactured passenger automobiles is lower than the attribute-based 
standard, but higher than the minimum standard, then the manufacturer 
may achieve compliance with the attribute-based standard by applying 
credits.
    (d) If a manufacturer's average fuel economy level for domestically 
manufactured passenger automobiles is lower than both the attribute-
based standard and the minimum standard, then the difference between 
the attribute-based standard and the minimum standard may be relieved 
by the use of credits, but the difference between the minimum standard 
and the manufacturer's actual fuel economy level may not be relieved by 
credits and will be subject to penalties.


Sec.  536.10  Treatment of dual-fuel and alternative fuel vehicles--
consistency with 49 CFR Part 538.

    (a) Statutory alternative fuel and dual-fuel vehicle calculations 
are treated as a change in the underlying fuel economy of the vehicle 
for purposes of this regulation, not as a credit that may be 
transferred or traded. Improvements in alternative fuel or dual fuel 
vehicle fuel economy as calculated pursuant to 49 U.S.C. 32905 and 
limited by 49 U.S.C. 32906 are therefore attributable only to the 
particular compliance category and model year to which the alternative 
or dual-fuel vehicle belongs.
    (b) If a manufacturer's calculated fuel economy for a particular 
compliance category, including any required calculations for 
alternative fuel and dual fuel vehicles, is higher or lower than the 
applicable fuel economy standard, manufacturers will earn credits or 
must apply credits or pay fines equal to the difference between the 
calculated fuel economy level in that compliance category and the 
applicable standard. Credits earned are the same as any other credits, 
and may be held, transferred, or traded by the manufacturer subject to 
the limitations of the statute and this regulation.
    (c) If a manufacturer builds enough alternative fuel or dual fuel 
vehicles to improve the calculated fuel economy in a particular 
compliance category by more than the limits set forth in 49 U.S.C. 
32906(a), the improvement in fuel economy for compliance purposes is 
restricted to the statutory limit. Manufacturers may not earn credits 
nor reduce the application of credits or fines for calculated 
improvements in fuel economy based on alternative or dual fuel vehicles 
beyond the statutory limit.

PART 537--AUTOMOTIVE FUEL ECONOMY REPORTS

    16. The authority citation for part 537 continues to read as 
follows:

    Authority: 49 U.S.C. 32907, delegation of authority at 49 CFR 
1.50.

    17. Amend Sec.  537.7 by revising paragraphs (b) and (c)(4)(xvi)(A) 
to read as follows:


Sec.  537.7  Pre-model year and mid-model year reports.

* * * * *
    (b) Projected average and target fuel economy. (1) State the 
projected average fuel economy for the manufacturer's automobiles 
determined in accordance with Sec.  537.9 and based upon the fuel 
economy values and projected sales figures provided under paragraph 
(c)(2) of this section.
    (2) State the projected final average fuel economy that the 
manufacturer anticipates having if changes implemented during the model 
year will cause that average to be different from the average fuel 
economy projected under paragraph (b)(1) of this section.
    (3) State the projected target fuel economy for the manufacturer's 
passenger automobiles and light trucks determined in accordance with 49 
CFR 531.5(c) and 49 CFR 533.5(h) and based upon the projected sales 
figures provided under paragraph (c)(2) of this section.
    (4) State the projected final target fuel economy that the 
manufacturer anticipates having if changes implemented during the model 
year will cause the targets to be different from the target fuel 
economy projected under paragraph (b)(3) of this section.
    (5) State whether the manufacturer believes that the projections it 
provides under paragraphs (b)(2) and (b)(4) of this section, or if it 
does not provide an average or target under those paragraphs, the 
projections it provides under paragraphs (b)(1) and (b)(3) of this 
section, sufficiently represent the manufacturer's average and target 
fuel economy for the current model year for purposes of the Act. In the 
case of a manufacturer that believes that the projections are not 
sufficiently representative for those purposes, state the specific 
nature of any reason for the insufficiency and the specific additional 
testing or derivation of fuel economy values by analytical methods 
believed by the manufacturer necessary to eliminate the insufficiency 
and any plans of the manufacturer to undertake that testing or 
derivation voluntarily and submit the resulting data to the 
Environmental Protection Agency under 40 CFR 600.509.
* * * * *
    (c) * * *
    (4) * * *
    (xvi)(A) In the case of passenger automobiles:
    (1) Interior volume index, determined in accordance with subpart D 
of 40 CFR part 600,
    (2) Body style,
    (3) Beginning model year 2010, track width as defined in 49 CFR 
523.2,
    (4) Beginning model year 2010, wheelbase as defined in 49 CFR 
523.2, and
    (5) Beginning model year 2010, footprint as defined in 49 CFR 
523.2.
* * * * *

    Issued: April 22, 2008.
Nicole R. Nason,
Administrator.
[FR Doc. 08-1186 Filed 4-23-08; 9:16 am]

BILLING CODE 4910-59-P
