Proposed Rulemaking To Establish Light-Duty Vehicle Greenhouse Gas Emission Standards and Corporate Average Fuel Economy Standards

Note: EPA no longer updates this information, but it may be useful as a reference or resource.
PDF Version (50 pp, 1138K, About PDF)

[Federal Register: September 28, 2009 (Volume 74, Number 186)]
[Proposed Rules]
[Page 49503-49552]
From the Federal Register Online via GPO Access [wais.access.gpo.gov]
[DOCID:fr28se09-25]

Proposed Rulemaking To Establish Light-Duty Vehicle Greenhouse
Gas Emission Standards and Corporate Average Fuel Economy Standards

[[Continued from page 49502]]

[[Page 49503]]

differences in how agencies apply technologies to vehicles in their
respective models, we report the ranges for the effectiveness values
used in each model. While the agencies believe that the ideal estimates
for the final rule would be based on tear down studies or BOM approach
and subjected to a transparent peer-reviewed process, NHTSA and EPA are
confident that the thorough review conducted, led to the best available
conclusion regarding technology costs and effectiveness estimates for
the current rulemaking and resulted in excellent consistency between
the agencies' respective analyses for developing the CAFE and
CO₂ standards.
    The agencies note that the effectiveness values estimated for the
technologies considered in the modeling analyses may represent average
values, and do not reflect the potentially-limitless spectrum of
possible values that could result from adding the technology to
different vehicles. For example, while the agencies have estimated an
effectiveness of 0.5 percent for low friction lubricants, each vehicle
could have a unique effectiveness estimate depending on the baseline
vehicle's oil viscosity rating. Similarly, the reduction in rolling
resistance (and thus the improvement in fuel economy and the reduction
in CO₂ emissions) due to the application of low rolling
resistance tires depends not only on the unique characteristics of the
tires originally on the vehicle, but on the unique characteristics of
the tires being applied, characteristics which must be balanced between
fuel efficiency, safety, and performance. Aerodynamic drag reduction is
much the same--it can improve fuel economy and reduce CO₂
emissions, but it is also highly dependent on vehicle-specific
functional objectives. For purposes of this NPRM, NHTSA and EPA believe
that employing average values for technology effectiveness estimates,
as adjusted depending on vehicle subclass, is an appropriate way of
recognizing the potential variation in the specific benefits that
individual manufacturers (and individual vehicles) might obtain from
adding a fuel-saving technology. However, the agencies seek comment on
whether additional levels of specificity beyond that already provided
would improve the analysis for the final rule, and if so, how those
levels of specificity should be analyzed.
    Chapter 3 of the draft Joint Technical Support Document contains a
detailed description of our assessment of vehicle technology cost and
effectiveness estimates. The agencies note that the technology costs
included in this NPRM take into account only those associated with the
initial build of the vehicle. The agencies seek comment on the
additional lifetime costs, if any, associated with the implementation
of advanced technologies including warranty costs, and maintenance and
replacement costs such as replacement costs for low rolling resistance
tires, low friction lubricants, and hybrid batteries, and maintenance
on diesel aftertreatment components.

F. Joint Economic Assumptions

    The agencies' preliminary analysis of alternative CAFE and GHG
standards for the model years covered by this proposed rulemaking rely
on a range of forecast information, economic estimates, and input
parameters. This section briefly describes the agencies' preliminary
choices of specific parameter values. These proposed economic values
play a significant role in determining the benefits of both CAFE and
GHG standards.
    In reviewing these variables and the agency's estimates of their
values for purposes of this NPRM, NHTSA and EPA reconsidered previous
comments that NHTSA had received and reviewed newly available
literature. As a consequence, the agencies elected to revise some
economic assumptions and parameter estimates, while retaining others.
Some of the most important changes, which are discussed in greater
detail in the agencies' respective sections below, as well as in
Chapter 4 of the joint TSD and in Chapter VIII of NHTSA's PRIA and
Chapter 8 of EPA's DRIA, include significant revisions to the markup
factors for technology costs; reducing the rebound effect from 15 to 10
percent; and revising the value of reducing CO₂ emissions
based on recent interagency efforts to develop estimates of this value
for government-wide use. The agencies seek comment on the economic
assumptions described below.
    • Costs of fuel economy-improving technologies--These
estimates are presented in summary form above and in more detail in the
agencies' respective sections of this preamble, in Chapter 3 of the
joint TSD, and in the agencies' respective RIAs. The technology cost
estimates used in this analysis are intended to represent
manufacturers' direct costs for high-volume production of vehicles with
these technologies and sufficient experience with their application so
that all cost reductions due to ``learning curve'' effects have been
fully realized. Costs are then modified by applying near-term indirect
cost multipliers ranging from 1.11 to 1.64 to the estimates of vehicle
manufacturers' direct costs for producing or acquiring each technology
to improve fuel economy, depending on the complexity of the technology
and the time frame over which costs are estimated.
    • Potential opportunity costs of improved fuel economy--This
estimate addresses the possibility that achieving the fuel economy
improvements required by alternative CAFE or GHG standards would
require manufacturers to compromise the performance, carrying capacity,
safety, or comfort of their vehicle models. If it did so, the resulting
sacrifice in the value of these attributes to consumers would represent
an additional cost of achieving the required improvements, and thus of
manufacturers' compliance with stricter standards. Currently the
agencies assume that these vehicle attributes do not change, and
include the cost of maintaining these attributes as part of the cost
estimates for technologies. However, it is possible that the technology
cost estimates do not include adequate allowance for the necessary
efforts by manufacturers to maintain vehicle performance, carrying
capacity, and utility while improving fuel economy and reducing GHG
emissions. While, in principle, consumer vehicle demand models can
measure these effects, these models do not appear to be robust across
specifications, since authors derive a wide range of willingness-to-pay
values for fuel economy from these models, and there is not clear
guidance from the literature on whether one specification is clearly
preferred over another. Thus, the agencies seek comment on how to
estimate explicitly the changes in vehicle buyers' welfare from the
combination of higher prices for new vehicle models, increases in their
fuel economy, and any accompanying changes in vehicle attributes such
as performance, passenger- and cargo-carrying capacity, or other
dimensions of utility.
    • The on-road fuel economy ``gap''--Actual fuel economy
levels achieved by light-duty vehicles in on-road driving fall somewhat
short of their levels measured under the laboratory-like test
conditions used by NHTSA and EPA to establish compliance with the
proposed CAFE and GHG standards. The agencies use an on-road fuel
economy gap for light-duty vehicles of 20 percent lower than published
fuel economy levels. For example, if the measured CAFE fuel economy
value of a light truck is 20 mpg, the on-road fuel economy actually
achieved by a typical driver of that vehicle is expected to be 16 mpg

[[Page 49504]]

(20*.80).\92\ NHTSA previously used this estimate in its MY 2011 final
rule, and the agencies confirmed it based on independent analysis for
use in this NPRM.
---------------------------------------------------------------------------

    \92\ U.S. Environmental Protection Agency, Final Technical
Support Document, Fuel Economy Labeling of Motor Vehicle Revisions
to Improve Calculation of Fuel Economy Estimates, EPA420-R-06-017,
December 2006.
---------------------------------------------------------------------------

    • Fuel prices and the value of saving fuel--Projected future
fuel prices are a critical input into the preliminary economic analysis
of alternative standards, because they determine the value of fuel
savings both to new vehicle buyers and to society. The agencies 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, the agencies used the AEO 2009 (April 2009
release) Reference Case forecasts of inflation-adjusted (constant-
dollar) retail gasoline and diesel fuel prices, which represent the
EIA's most up-to-date estimate of the most likely course of future
prices for petroleum products.\93\
---------------------------------------------------------------------------

    \93\ Energy Information Administration, Annual Energy Outlook
2009, Revised Updated Reference Case (April 2009), Table 12.
Available at http://www.eia.doe.gov/oiaf/servicerpt/stimulus/excel/
aeostimtab_12.xls (last accessed July 26, 2009).
---------------------------------------------------------------------------

    EIA's Updated Reference Case reflects the effects of the American
Reinvestment and Recovery Act of 2009, as well as the most recent
revisions to the U.S. and global economic outlook. In addition, it also
reflects the provisions of the Energy Independence and Security Act of
2007 (EISA), including the requirement that the combined mpg level of
U.S. cars and light trucks reach 35 miles per gallon by model year
2020. Because this provision would be expected to reduce future U.S.
demand for gasoline and other fuels, there is some concern about
whether the AEO 2009 forecast of fuel prices already partly reflects
the increases in CAFE standards considered in this rule, and thus
whether it is suitable for valuing the projected reductions in fuel
use. In response to this concern, the agencies note that EIA issued a
revised version of AEO 2008 in June 2008, which modified its previous
December 2007 Early Release of AEO 2008 to reflect the effects of the
recently-passed EISA legislation.\94\ The fuel price forecasts reported
in EIA's Revised Release of AEO 2008 differed by less than one cent per
gallon over the entire forecast period (2008-230) from those previously
issued as part of its initial release of AEO 2008. Thus, the agencies
are reasonably confident that the fuel price forecasts presented in AEO
2009 and used to analyze the value of fuel savings projected to result
from this rule are not unduly affected by the CAFE provisions of EISA,
and therefore do not cause a baseline problem. Nevertheless, the
agencies request comment on the use of the AEO 2009 fuel price
forecasts, and particularly on the potential impact of the EISA-
mandated CAFE improvements on these projections.
---------------------------------------------------------------------------

    \94\ Energy Information Administration, Annual Energy Outlook
2008, Revised Early Release (June 2008), Table 12. Available at
http://www.eia.doe.gov/oiaf/archive/aeo08/excel/aeotab_12.xls (last
accessed September 12, 2009).
---------------------------------------------------------------------------

    • Consumer valuation of fuel economy and payback period--In
estimating the value of fuel economy improvements that would result
from alternative CAFE and GHG standards to potential vehicle buyers,
the agencies assume 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 discount the
value of these future fuel savings using rates of 3% and 7%. The five-
year figure represents the current average term of consumer loans to
finance the purchase of new vehicles.
    • Vehicle sales assumptions--The first step in estimating
lifetime fuel consumption by vehicles produced during a model year is
to calculate the number that are expected to be produced and sold.\95\
The agencies relied on the AEO 2009 Reference Case for forecasts of
total vehicle sales, while the baseline market forecast developed by
the agencies (see Section II.B) divided total projected sales into
sales of cars and light trucks.
---------------------------------------------------------------------------

    \95\ 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 2 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 percent 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/Pubs/809952.pdf (last accessed July 27, 2009).
---------------------------------------------------------------------------

    • Vehicle survival assumptions--We then applied updated
values of age-specific survival rates for cars and light trucks to
these adjusted forecasts of passenger car and light truck sales to
determine the number of these vehicles remaining in use during each
year of their expected lifetimes.
    • Total vehicle use--We then calculated the total number of
miles that cars and light trucks produced in each model year will be
driven during each year of their lifetimes using estimates of annual
vehicle use by age tabulated from the Federal Highway Administration's
2001 National Household Transportation Survey (NHTS),\96\ adjusted to
account for the effect on vehicle use of subsequent increases in fuel
prices. In order to insure that the resulting mileage schedules imply
reasonable estimates of future growth in total car and light truck use,
we calculated the rate of growth in annual car and light truck mileage
at each age that is necessary for total car and light truck travel to
increase at the rates forecast in the AEO 2009 Reference Case. The
growth rate in average annual car and light truck use produced by this
calculation is approximately 1.1 percent per year.\97\ This rate was
applied to the mileage figures derived from the 2001 NHTS to estimate
annual mileage during each year of the expected lifetimes of MY 2012-
2016 cars and light trucks.\98\
---------------------------------------------------------------------------

    \96\ For a description of the Survey, see http://nhts.ornl.gov/
quickStart.shtml (last accessed July 27, 2009).
    \97\ It was not possible to estimate separate growth rates in
average annual use for cars and light trucks, because of the
significant reclassification of light truck models as passenger cars
discussed previously.
    \98\ While the adjustment for future fuel prices reduces average
mileage at each age from the values derived from the 2001 NHTS, the
adjustment for expected future growth in average vehicle use
increases it. The net effect of these two adjustments is to increase
expected lifetime mileage by about 18 percent for passenger cars and
about 16 percent for light trucks.
---------------------------------------------------------------------------

    • Accounting for the rebound effect of higher fuel economy--
The rebound effect refers to the fraction of fuel savings expected to
result from an increase in vehicle fuel economy--particularly an
increase required by the adoption of higher CAFE and GHG standards--
that is offset by additional vehicle use. The increase in vehicle use
occurs because higher fuel economy reduces the fuel cost of driving,
typically the largest single component of the monetary cost of
operating a vehicle, and vehicle owners respond to this reduction in
operating costs by driving slightly more. For purposes of this NPRM,
the agencies have elected to use a 10 percent rebound effect in their
analyses of fuel savings and other benefits from higher standards.
    • 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

[[Page 49505]]

additional travel provides benefits to drivers and their passengers by
improving their access to social and economic opportunities away from
home. The benefits from increased vehicle use include both the fuel
expenses associated with this additional travel, and the consumer
surplus it provides. We estimate the economic value of the consumer
surplus provided by added driving using the conventional approximation,
which is one half of the product of the decline in vehicle operating
costs per vehicle-mile and the resulting increase in the annual number
of miles driven. Because it depends on the extent of improvement in
fuel economy, the value of benefits from increased vehicle use changes
by model year and varies among alternative standards.
    • The value of increased driving range--By reducing the
frequency with which drivers typically refuel their vehicles, and by
extending the upper limit of the range they can travel before requiring
refueling, improving fuel economy and reducing GHG emissions thus
provides some additional benefits to their owners. No direct estimates
of the value of extended vehicle range are readily available, so the
agencies' analysis calculates the reduction in the annual number of
required refueling cycles that results from improved fuel economy, and
applies DOT-recommended values of travel time savings to convert the
resulting time savings to their economic value.\99\ The agencies invite
comment on the assumptions used in this analysis. Please see the
Chapter 4 of the draft Joint TSD for details.
---------------------------------------------------------------------------

    \99\ Department of Transportation, Guidance Memorandum, ``The
Value of Saving Travel Time: Departmental Guidance for Conducting
Economic Evaluations,'' Apr. 9, 1997. 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).
---------------------------------------------------------------------------

    • Added costs from congestion, crashes and noise--Although
it provides some benefits to drivers, increased vehicle use associated
with the rebound effect also contributes to increased traffic
congestion, motor vehicle accidents, and highway noise. Depending on
how the additional travel is distributed over the day and on where it
takes place, additional vehicle use can contribute to traffic
congestion and delays by increasing traffic volumes on facilities that
are already heavily traveled during peak periods. These added delays
impose higher costs on drivers and other vehicle occupants in the form
of increased travel time and operating expenses, increased costs
associated with traffic accidents, and increased traffic noise. The
agencies rely on estimates of congestion, accident, and noise costs
caused by automobiles and light trucks developed by the Federal Highway
Administration to estimate the increased external costs caused by added
driving due to the rebound effect.\100\
---------------------------------------------------------------------------

    \100\ These estimates were developed by FHWA for use in its 1997
Federal Highway Cost Allocation Study; http://www.fhwa.dot.gov/
policy/hcas/final/index.htm (last accessed July 29, 2009).
---------------------------------------------------------------------------

    • Petroleum consumption and import externalities--U.S.
consumption and imports of petroleum products also impose costs on the
domestic economy that are not reflected in the market price for crude
petroleum, or in the prices paid by consumers of petroleum products
such as gasoline. In economics literature on this subject, these costs
include (1) higher prices for petroleum products resulting from the
effect of U.S. oil import demand on the world oil price (``monopsony
costs''); (2) the 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.\101\ Reducing U.S. imports of crude petroleum or refined
fuels can reduce the magnitude of these external costs. Any reduction
in their total value that results from lower fuel consumption and
petroleum imports represents an economic benefit of setting more
stringent standards over and above the dollar value of fuel savings
itself. The agencies do not include a value for monopsony costs in
order to be consistent with their use of a global value for the social
cost of carbon. Based on a recently-updated ORNL study, we estimate
that each gallon of fuel saved that results in a reduction in U.S.
petroleum imports (either crude petroleum or refined fuel) will reduce
the expected costs of oil supply disruptions to the U.S. economy by
$0.169 (2007$). The agencies do not include savings in budgetary
outlays to support U.S. military activities among the benefits of
higher fuel economy and the resulting fuel savings. Each gallon of fuel
saved as a consequence of higher standards is anticipated to reduce
total U.S. imports of crude petroleum or refined fuel by 0.95 gallons.\102\
---------------------------------------------------------------------------

    \101\ 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.
    \102\ Each gallon of fuel saved is assumed to reduce imports of
refined fuel by 0.5 gallons, and the volume of fuel refined
domestically by 0.5 gallons. Domestic fuel refining is assumed to
utilize 90% imported crude petroleum and 10% domestically-produced
crude petroleum as feedstocks. Together, these assumptions imply
that each gallon of fuel saved will reduce imports of refined fuel
and crude petroleum by 0.50 gallons + 0.50 gallons*90% = 0.50
gallons + 0.45 gallons = 0.95 gallons.
---------------------------------------------------------------------------

    • Air pollutant emissions
    [cir] Impacts on criteria air pollutant emissions--While reductions
in domestic fuel refining and distribution that result from lower fuel
consumption will reduce U.S. emissions of criteria pollutants,
additional vehicle use associated with the rebound effect will increase
emissions of these pollutants. Thus the net effect of stricter
standards on emissions of each criteria pollutant depends on the
relative magnitudes of reduced emissions from fuel refining and
distribution, and increases in emissions resulting from added vehicle
use. Criteria air pollutants emitted by vehicles and during fuel
production include carbon monoxide (CO), hydrocarbon compounds (usually
referred to as ``volatile organic compounds,'' or VOC), nitrogen oxides
(NO_X), fine particulate matter (PM_2.5), and
sulfur oxides (SO_X). It is assumed that the emission rates
(per mile) stay constant for future year vehicles.
    [cir] EPA and NHTSA estimate the economic value of the human health
benefits associated with reducing exposure to PM_2.5 using a
``benefit-per-ton'' method. These PM_2.5-related benefit-per-
ton estimates provide the total monetized benefits to human health (the
sum of reductions in premature mortality and premature morbidity) that
result from eliminating one ton of directly emitted PM_2.5,
or one ton of a pollutant that contributes to secondarily-formed
PM_2.5 (such as NOX, SOX, and VOCs), from a specified source.
Chapter 4.2.9 of the Technical Support Document that accompanies this
proposal includes a description of these values.
    Reductions in GHG emissions--Emissions of carbon dioxide and other
greenhouse gases (GHGs) occur throughout the process of producing and
distributing transportation fuels, as well as from fuel combustion
itself. By reducing the volume of fuel consumed by passenger cars and
light trucks, higher standards will thus reduce GHG emissions generated
by fuel use, as well as throughout the fuel supply cycle. The agencies
estimated the increases of GHGs other than CO_2, including

[[Page 49506]]

methane and nitrous oxide, from additional vehicle use by multiplying
the increase in total miles driven by cars and light trucks of each
model year and age by emission rates per vehicle-mile for these GHGs.
These emission rates, which differ between cars and light trucks as
well as between gasoline and diesel vehicles, were estimated by EPA
using its recently-developed Motor Vehicle Emission Simulator (Draft
MOVES 2009).\103\ Increases in emissions of non-CO₂ GHGs are
converted to equivalent increases in CO₂ emissions using
estimates of the Global Warming Potential (GWP) of methane and nitrous oxide.
---------------------------------------------------------------------------

    \103\ The MOVES model assumes that the per-mile rates at which
cars and light trucks emit these GHGs are determined by the
efficiency of fuel combustion during engine operation and chemical
reactions that occur during catalytic after-treatment of engine
exhaust, and are thus independent of vehicles' fuel consumption
rates. Thus MOVES' emission factors for these GHGs, which are
expressed per mile of vehicle travel, are assumed to be unaffected
by changes in fuel economy.
---------------------------------------------------------------------------

     • Economic value of reductions in CO₂ emissions--EPA
and NHTSA assigned a dollar value to reductions in CO₂
emissions using the marginal dollar value (i.e., cost) of climate-
related damages resulting from carbon emissions, also referred to as
``social cost of carbon'' (SCC). The SCC is intended to measure the
monetary value society places on impacts resulting from increased GHGs,
such as property damage from sea level rise, forced migration due to
dry land loss, and mortality changes associated with vector-borne
diseases. Published estimates of the SCC vary widely as a result of
uncertainties about future economic growth, climate sensitivity to GHG
emissions, procedures used to model the economic impacts of climate
change, and the choice of discount rates. EPA and NHTSA's coordinated
proposals present a set of interim SCC values reflecting a Federal
interagency group's interpretation of the relevant climate economics
literature. Sections III.H and IV.C.3 provide more detail about SCC.
    • Discounting future benefits and costs--Discounting future
fuel savings and other benefits is intended to account for the
reduction in their value to society when they are deferred until some
future date, rather than received immediately. The discount rate
expresses the percent decline in the value of these benefits--as viewed
from today's perspective--for each year they are deferred into the
future. In evaluating the non-climate related benefits of the proposed
standards, the agencies have employed discount rates of both 3 percent
and 7 percent.
    For the reader's reference, Table II.F.1-1 below summarizes the
values used to calculate the impacts of each proposed standard. The
values presented in this table are summaries of the inputs used for the
models; specific values used in the agencies' respective analyses may
be aggregated, expanded, or have other relevant adjustments. See the
respective RIAs for details. The agencies seek comment on the economic
assumptions presented in the table and discussed below.
    In addition, the agencies have conducted a range of sensitivities
and present them in their respective RIAs. For example, NHTSA has
conducted a sensitivity analysis on several assumptions including (1)
forecasts of future fuel prices, (2) the discount rate applied to
future benefits and costs, (3) the magnitude of the rebound effect, (4)
the value to the U.S. economy of reducing carbon dioxide emissions, (5)
the monopsony effect, and (6) the reduction in external economic costs
resulting from lower U.S. oil imports. This information is provided in
NHTSA's PRIA. The agencies will consider additional sensitivities for
the final rule as appropriate, including sensitivities on the markup
factors applied to direct manufacturing costs to account for indirect
costs (i.e., the Indirect Cost Markups (ICMs) which are discussed in
Sections III and IV), and the learning curve estimates used in this analysis.

    Table II.F.1-1--Economic Values for Benefits Computations (2007$)
------------------------------------------------------------------------

------------------------------------------------------------------------
Fuel Economy Rebound Effect.............................             10%
``Gap'' between test and on-road MPG....................             20%
Value of refueling time per ($ per vehicle-hour)........           24.64
Annual growth in average vehicle use....................            1.1%
Fuel Prices (2012-50 average, $/gallon):
    Retail gasoline price...............................            3.77
    Pre-tax gasoline price..............................            3.40
Economic Benefits from Reducing Oil Imports ($/gallon):
    ``Monopsony'' Component.............................            0.00
    Price Shock Component...............................            0.17
    Military Security Component.........................            0.00
    Total Economic Costs ($/gallon).....................            0.17
Emission Damage Costs (2020, $/ton or $/metric ton):
    Carbon monoxide.....................................               0
    Volatile organic compounds (VOC)....................           1,283
    Nitrogen oxides (NOX)--vehicle use..................           5,116
    Nitrogen oxides (NOX)--fuel production and                     5,339
     distribution.......................................
    Particulate matter (PM2.5)--vehicle use.............         238,432
    Particulate matter (PM2.5)--fuel production and              292,180
     distribution.......................................
    Sulfur dioxide (SO2)................................          30,896
                                                                       5
                                                                      10
                                                                      20
                                                                      34
    Carbon dioxide (CO2)................................              56
    Annual Increase in CO2 Damage Cost..................              3%
External Costs from Additional Automobile Use ($/vehicle-
 mile):
    Congestion..........................................           0.054
    Accidents...........................................           0.023
    Noise...............................................           0.001
    Total External Costs................................           0.078
External Costs from Additional Light Truck Use ($/        ..............
 vehicle-mile):

[[Page 49507]]

    Congestion..........................................           0.048
    Accidents...........................................           0.026
    Noise...............................................           0.001
    Total External Costs................................           0.075
Discount Rates Applied to Future Benefits...............          3%, 7%
------------------------------------------------------------------------

III. EPA Proposal for Greenhouse Gas Vehicle Standards

A. Executive Overview of EPA Proposal

1. Introduction
    The Environmental Protection Agency (EPA) is proposing to establish
greenhouse gas emissions standards for the largest sources of
transportation greenhouse gases--light-duty vehicles, light-duty
trucks, and medium-duty passenger vehicles (hereafter light vehicles).
These vehicle categories, which include cars, sport utility vehicles,
minivans, and pickup trucks used for personal transportation, are
responsible for almost 60% of all U.S. transportation related
greenhouse gas emissions. This action represents the first-ever
proposal by EPA to regulate vehicle greenhouse gas emissions under the
Clean Air Act (CAA) and would establish standards for model years 2012
and later light vehicles sold in the U.S.
    EPA is proposing three separate standards. The first and most
important is a set of fleet-wide average carbon dioxide
(CO₂) emission standards for cars and trucks. These
standards are based on CO₂ emissions-footprint curves, where
each vehicle has a different CO₂ emissions compliance target
depending on its footprint value. Vehicle CO₂ emissions
would be measured over the EPA city and highway tests. The proposed
standard allows for credits based on demonstrated improvements in
vehicle air conditioner systems, including both efficiency and
refrigerant leakage improvement, which are not captured by the EPA
tests. The EPA projects that the average light vehicle tailpipe
CO₂ level in model year 2011 will be 326 grams per mile
while the average vehicle tailpipe CO₂ emissions compliance
level for the proposed model year 2016 standard will be 250 grams per
mile, an average reduction of 23 percent from today's CO₂ levels.
    EPA is also proposing standards that will cap tailpipe nitrous
oxide (N₂O) and methane (CH₄) emissions at 0.010
and 0.030 grams per mile, respectively. Even after adjusting for the
higher relative global warming potencies of these two compounds,
nitrous oxide and methane emissions represent less than one percent of
overall vehicle greenhouse gas emissions from new vehicles.
Accordingly, the goal of these two proposed standards is to limit any
potential increases in the future and not to force reductions relative
to today's low levels.
    This proposal represents the second-phase of EPA's response to the
Supreme Court's 2007 decision in Massachusetts v. EPA \104\ which found
that greenhouse gases were air pollutants for purposes of the Clean Air
Act. The Court held that the Administrator must determine whether or
not emissions from new motor vehicles cause or contribute to air
pollution which may reasonably be anticipated to endanger public health
or welfare, or whether the science is too uncertain to make a reasoned
decision. The Court further ruled that, in make these decisions, the
EPA Administrator is required to follow the language of section 202(a)
of the CAA. The Court remanded the case back to the Agency for
reconsideration in light of its finding.
---------------------------------------------------------------------------

    \104\ 549 U.S. 497 (2007). For further information on
Massachusetts v. EPA see the July 30, 2008 Advance Notice of
Proposed Rulemaking, ``Regulating Greenhouse Gas Emissions under the
Clean Air Act'', 73 FR 44354 at 44397. There is a comprehensive
discussion of the litigation's history, the Supreme Court's
findings, and subsequent actions undertaken by the Bush Administration
and the EPA from 2007-2008 in response to the Supreme Court remand.
---------------------------------------------------------------------------

    The Administrator responded to the Court's remand by issuing two
proposed findings under section 202(a) of the Clean Air Act.\105\
First, the Administrator proposed to find that the science supports a
positive endangerment finding that a mix of certain greenhouse gases in
the atmosphere endangers the public health and welfare of current and
future generations. This is referred to as the endangerment finding.
Second, the Administrator proposed to find that the emissions of four
of these gases--carbon dioxide, methane, nitrous oxide, and
hydrofluorocarbons--from new motor vehicles and new motor vehicle
engines contribute to the atmospheric concentrations of these key
greenhouse gases and hence to the threat of climate change. This is
referred to as the cause and contribute finding. Finalizing this
proposed light vehicle regulations is contingent upon EPA finalizing
both the endangerment finding and cause or contribute finding. Sections
III.B.1 through III.B.4 below provide more details on the legal and
scientific bases for this proposal.
---------------------------------------------------------------------------

    \105\ 74 FR 18886, April 24, 2009.
---------------------------------------------------------------------------

    As discussed in Section I, this GHG proposal is part of a joint
National Program such that a large majority of the projected benefits
are achieved jointly with NHTSA's proposed CAFE rule which is described
in detail in Section IV of this preamble. EPA's proposal projects total
carbon dioxide emissions savings of nearly 950 million metric tons, and
oil savings of 1.8 billion barrels over the lifetimes of the vehicles
sold in model years 2012-2016. EPA projects net societal benefits of
$192 billion at a 3 percent discount rate for these same vehicles, or
$136 billion at a 7 percent discount rate (both values assume a $20/ton
SCC value). Accordingly, these proposed light vehicle greenhouse gas
emissions standards would make an important ``first step'' contribution
as part of the National Program toward meeting long-term greenhouse gas
emissions and import oil reduction goals, while providing important
economic benefits as well.
2. Why is EPA Proposing this Rule?
    This proposal addresses only light vehicles. EPA is addressing
light vehicles as a first step in control of greenhouse gas emissions
under the Clean Air Act for four reasons. First, light vehicles are
responsible for almost 60% of all mobile source greenhouse gas
emissions, a share three times larger than any other mobile source
subsector, and represent about one-sixth of all U.S. greenhouse gas
emissions. Second, technology exists that can be readily and cost-
effectively applied to these vehicles to reduce greenhouse gas
emissions in the near term. Third, EPA already has an existing testing
and compliance program for these vehicles, refined since the mid-1970s
for emissions certification and fuel economy compliance, which would
require only minor modifications to accommodate greenhouse gas
emissions regulations. Finally, this proposal is an important first
step in responding to the Supreme Court's ruling in Massachusetts vs.
EPA. In addition, EPA is currently evaluating controls for motor
vehicles other than those covered

[[Page 49508]]

by this proposal, and is reviewing seven petitions submitted by various
States and organizations requesting that EPA use its Clean Air Act
authorities to take action to reduce greenhouse gas emissions from
aircraft (under Sec.  231(a)(2)), ocean-going vessels (under Sec. 
213(a)(4)), and other nonroad engines and vehicle sources (also under
Sec.  213(a)(4)).
a. Light Vehicle Emissions Contribute to Greenhouse Gases and the
Threat of Climate Change
    Greenhouse gases are gases in the atmosphere that effectively trap
some of the Earth's heat that would otherwise escape to space.
Greenhouse gases are both naturally occurring and anthropogenic. The
primary greenhouse gases of concern are directly emitted by human
activities and include carbon dioxide, methane, nitrous oxide,
hydrofluorocarbons, perfluorocarbons, and sulfur hexafluoride.
    These gases, once emitted, remain in the atmosphere for decades to
centuries. Thus, they become well mixed globally in the atmosphere and
their concentrations accumulate when emissions exceed the rate at which
natural processes remove greenhouse gases from the atmosphere. The
heating effect caused by the human-induced buildup of greenhouse gases
in the atmosphere is very likely\106\ the cause of most of the observed
global warming over the last 50 years. The key effects of climate
change observed to date and projected to occur in the future include,
but are not limited to, more frequent and intense heat waves, more
severe wildfires, degraded air quality, heavier and more frequent
downpours and flooding, increased drought, greater sea level rise, more
intense storms, harm to water resources, continued ocean acidification,
harm to agriculture, and harm to wildlife and ecosystems. A detailed
explanation of observed and projected changes in greenhouse gases and
climate change and its impact on health, society, and the environment
is included in EPA's technical support document for the recently
released Proposed Endangerment and Cause or Contribute Findings for
Greenhouse Gases Under Section 202(a) of the Clean Air Act.\107\
---------------------------------------------------------------------------

    \106\ According to Intergovernmental Panel on Climate Change
(IPCC) terminology, ``very likely'' conveys a 90 to 99 percent
probability of occurrence. ``Virtually certain'' conveys a greater
than 99 percent probability, ``likely'' conveys a 66 to 90 percent
probability, and ``about as likely as not'' conveys a 33 to 66
percent probability.
    \107\ 74 FR18886, April 24, 2009. Both the Federal Register
Notice and the Technical Support Document for this rulemaking are
found in the public docket for this rulemaking. Docket is EPA-OAR-2009-0171.
---------------------------------------------------------------------------

    Transportation sources represent a large and growing share of
United States greenhouse gases and include automobiles, highway heavy
duty trucks, airplanes, railroads, marine vessels and a variety of
other sources. In 2006, all transportation sources emitted 31.5% of all
U.S. greenhouse gases, and were the fastest-growing source of
greenhouse gases in the U.S., accounting for 47% of the net increase in
total U.S. greenhouse gas emissions from 1990-2006.\108\ The only
sector with larger greenhouse gas emissions was electricity generation
which emitted 33.7% of all U.S. greenhouse gases.
---------------------------------------------------------------------------

    \108\ Inventory of U.S. Greenhouse Gases and Sinks: 1990-2006.
---------------------------------------------------------------------------

    Light vehicles emit four greenhouse gases: carbon dioxide, methane,
nitrous oxide and hydrofluorocarbons. Carbon dioxide (CO₂)
is the end product of fossil fuel combustion. During combustion, the
carbon stored in the fuels is oxidized and emitted as CO₂
and smaller amounts of other carbon compounds.\109\ Methane
(CH₄) emissions are a function of the methane content of the
motor fuel, the amount of hydrocarbons passing uncombusted through the
engine, and any post-combustion control of hydrocarbon emissions (such
as catalytic converters).\110\ Nitrous oxide (N₂O) (and
nitrogen oxide (NO_X)) emissions from vehicles and their
engines are closely related to air-fuel ratios, combustion
temperatures, and the use of pollution control equipment. For example,
some types of catalytic converters installed to reduce motor vehicle
NO_X, carbon monoxide (CO) and hydrocarbon emissions can
promote the formation of N₂O.\111\ Hydrofluorocarbons (HFC)
emissions are progressively replacing chlorofluorocarbons (CFC) and
hydrochlorofluorocarbons (HCFC) in these vehicles' cooling and
refrigeration systems as CFCs and HCFCs are being phased out under the
Montreal Protocol and Title VI of the CAA. There are multiple emissions
pathways for HFCs with emissions occurring during charging of cooling
and refrigeration systems, during operations, and during
decommissioning and disposal.\112\
---------------------------------------------------------------------------

    \109\ Mobile source carbon dioxide emissions in 2006 equaled 26
percent of total U.S. CO₂ emissions.
    \110\ In 2006, methane emissions equaled 0.32 percent of total
U.S. methane emissions Nitrous oxide is a product of the reaction
that occurs between nitrogen and oxygen during fuel combustion.
    \111\ In 2006, nitrous oxide emissions for these sources
accounted for 8 percent of total U.S. nitrous oxide emissions.
    \112\ In 2006 HFC from these source categories equaled 56
percent of total U.S. HFC emissions, making it the single largest
source category of U.S. HFC emissions.
---------------------------------------------------------------------------

b. Basis for Action Under Clean Air Act
    Section 202(a)(1) of the Clean Air Act (CAA) states that ``the
Administrator shall by regulation prescribe (and from time to time
revise) * * * standards applicable to the emission of any air pollutant
from any class or classes of new motor vehicles * * *, which in his
judgment cause, or contribute to, air pollution which may reasonably be
anticipated to endanger public health or welfare.'' As noted above, the
Administrator has proposed to find that the air pollution of elevated
levels of greenhouse gas concentrations may reasonably be anticipated
to endanger public health and welfare.\113\ The Administrator has
proposed to define the air pollution to be the elevated concentrations
of the mix of six GHGs: carbon dioxide (CO₂), methane
(CH₄), nitrous oxide (N₂O), hydrofluorocarbons
(HFCs), perfluorocarbons (PFCs), and sulfur hexafluoride
(SF₆). The Administrator has further proposed to find under
CAA section 202(a) that CO_2, methane, N₂O and HFC
emissions from new motor vehicles and engines contribute to this air
pollution. This preamble describes proposed standards that would
control emissions of CO₂, HFCs, nitrous oxide, and methane.
Standards for these GHGs would only be finalized if EPA determines that
the criteria have been met for endangerment by the air pollution, and
that emissions of GHGs from new motor vehicles or engines ``cause or
contribute'' to that air pollution. In that case, section 202(a) would
authorize EPA to issue standards applicable to emissions of those
pollutants. For further discussion of EPA's authority under section
202(a), see Section I.C.2 of the proposal.
---------------------------------------------------------------------------

    \113\ 74 FR18886, April 24, 2009.
---------------------------------------------------------------------------

    There are a variety of other CAA Title II provisions that are
relevant to standards established under section 202(a). As noted above,
the standards are applicable to motor vehicles for their useful life.
EPA has the discretion in determining what standard applies over the
useful life. For example, EPA may set a single standard that applies
both when the vehicles are new and throughout the useful life, or where
appropriate may set a standard that varies during the term of useful
life, such as a standard that is more stringent in the early years of
the useful life and less stringent in the later years.

[[Page 49509]]

    The standards established under CAA section 202(a) are implemented
and enforced through various mechanisms. Manufacturers are required to
obtain an EPA certificate of conformity with the section 202
regulations before they may sell or introduce their new motor vehicle
into commerce, according to CAA section 206(a). The introduction into
commerce of vehicles without a certificate of conformity is a
prohibited act under CAA section 203 that may subject a manufacturer to
civil penalties and injunctive actions (see CAA sections 204 and 205).
Under CAA section 206(b), EPA may conduct testing of new production
vehicles to determine compliance with the standards. For in-use
vehicles, if EPA determines that a substantial number of vehicles do
not conform to the applicable regulations then the manufacturer must
submit and implement a remedial plan to address the problem (see CAA
section 207(c)). There are also emissions-based warranties that the
manufacturer must implement under CAA section 207(a).
c. EPA's Greenhouse Gas Proposal Under Section 202(a) Concerning
Endangerment and Cause or Contribute Findings
    EPA's Administrator recently signed a proposed action with two
distinct findings regarding greenhouse gases under section 202(a) of
the Clean Air Act. This action is called the Proposed Endangerment and
Cause or Contribute Findings for Greenhouse Gases under the Clean Air
Act (Endangerment Proposal).\114\ The Administrator proposed an
affirmative endangerment finding that the current and projected
concentrations of a mix of six key greenhouse gases--carbon dioxide
(CO₂), methane (CH₄), nitrous oxide
(N₂O), hydrofluorocarbons (HFCs), perfluorocarbons (PFCs),
and sulfur hexafluoride(SF₆)--in the atmosphere threaten the
public health and welfare of current and future generations. She also
proposed to find that the combined emissions of four of the gases--
carbon dioxide, methane, nitrous oxide and hydrofluorocarbons from new
motor vehicles and motor vehicle engines--contribute to the atmospheric
concentrations of these greenhouse gases and therefore to the climate
change problem.
---------------------------------------------------------------------------

    \114\ 74 FR 18886 (April 24, 2009).
---------------------------------------------------------------------------

    Specifically, the Administrator proposed, after a thorough
examination of the scientific evidence on the causes and impact of
current and future climate change, to find that the science
compellingly supports a positive finding that atmospheric
concentrations of these greenhouse gases result in air pollution which
may reasonably be anticipated to endanger both public health and
welfare. In her proposed finding, the Administrator relied heavily upon
the major findings and conclusions from the recent assessments of the
U.S. Climate Change Science Program and the U.N. Intergovernmental
Panel on Climate Change.\115\ The Administrator proposed a positive
endangerment finding after considering both observed and projected
future effects of climate change, key uncertainties, and the full range
of risks and impacts to public health and welfare occurring within the
United States. In addition, the proposed finding noted that the
evidence concerning risks and impacts occurring outside the U.S.
provided further support for the proposed finding.
---------------------------------------------------------------------------

    \115\ The U.S. Climate Change Science Program (CCSP) is now
called the U.S. Global Change Research Program (GCRP).
---------------------------------------------------------------------------

    The key scientific findings supporting the proposed endangerment
finding are that:

--Concentrations of greenhouse gases are at unprecedented levels
compared to recent and distant past. These high concentrations are the
unambiguous result of anthropogenic emissions and are very likely the
cause of the observed increase in average temperatures and other
climatic changes.
--The effects of climate change observed to date and projected to occur
in the future include more frequent and intense heat waves, more severe
wildfires, degraded air quality, heavier downpours and flooding,
increasing drought, greater sea level rise, more intense storms, harm
to water resources, harm to agriculture, and harm to wildlife and
ecosystems. These impacts are effects on public health and welfare
within the meaning of the Clean Air Act.
    With regard to new motor vehicles and engines, the Administrator
also proposed a finding that the combined emissions of four greenhouse
gases--carbon dioxide, methane, nitrous oxide and hydrofluorocarbons--
from new motor vehicles and engines contributes to this air pollution,
i.e., the atmospheric concentrations of the mix of six greenhouse gases
which create the threat of climate change and its impacts. Key facts
supporting the proposed cause and contribute finding for on-highway
vehicles regulated under section 202(a) of the Clean Air Act are that
these sources are responsible for 24% of total U.S. greenhouse gas
emissions, and more than 4% of total global greenhouse gas
emissions.\116\ The Administrator also considered whether emissions of
each greenhouse gas individually, as a separate air pollutant, would
contribute to this air pollution.
---------------------------------------------------------------------------

    \116\ This figure includes the greenhouse gas contributions of
light vehicles, heavy duty vehicles, and remaining on-highway mobile sources.
---------------------------------------------------------------------------

    If the Administrator makes affirmative findings under section
202(a) on both endangerment and cause or contribute, then EPA is to
issue standards ``applicable to emission'' of the air pollutant or
pollutants that EPA finds causes or contributes to the air pollution
that endangers public health and welfare. The Endangerment Proposal
invited public comment on whether the air pollutant should be
considered the group of GHGs, or whether each GHG should be treated as
a separate air pollutant. Either way, the emissions standards proposed
today would satisfy the requirements of section 202(a) as the
Administrator has significant discretion in how to structure the
standards that apply to the emission of the air pollutant or air
pollutants at issue. For example, under either approach EPA would have
the discretion under section 202(a) to adopt separate standards for
each GHG, a single composite standard covering various gases, or any
combination of these. In this rulemaking EPA is proposing separate
standards for nitrous oxide and methane, and a CO₂ standard
that provides for credits based on reductions of HFCs, as the
appropriate way to issue standards applicable to emissions of these GHGs.
3. What is EPA Proposing?
a. Proposed Light-Duty Vehicle, Light-Duty Truck, and Medium-Duty
Passenger Vehicle Greenhouse Gas Emission Standards and Projected
Compliance Levels
    The CO₂ emissions standards are by far the most
important of the three standards and are the primary focus of this
summary. EPA is proposing an attribute-based approach for the
CO₂ fleet-wide standard (one for cars and one for trucks),
based on vehicle footprint as the attribute. These curves establish
different CO₂ emissions targets for each unique car and
truck footprint. Generally, the larger the vehicle footprint, the
higher the corresponding vehicle CO₂ emissions target. Table
III.A.3-1 shows the greenhouse gas standards for light vehicles that
EPA is proposing for model years (MY) 2012 and later:

[[Page 49510]]

                   Table III.A.3-1--Proposed Industry-Wide Greenhouse Gas Emissions Standards
----------------------------------------------------------------------------------------------------------------
  Standard/covered  pollutants     Form of  standard  Level of  standard        Credits           Test cycles
----------------------------------------------------------------------------------------------------------------
CO2 Standard \117\: Tailpipe CO2  Fleetwide average   See footprint--CO2  CO2-e credits       EPA 2-cycle (FTP
                                   footprint CO2-      curves in Figure    \118\.              and HFET test
                                   curves for cars     I.C-1 for cars                          cycles), with
                                   and trucks.         and Figure I.C-2                        separate
                                                       for trucks.                             mechanisms for A/
                                                                                               C credits.\119\
N2O Standard: Tailpipe N2O......  Cap per vehicle...  0.010 g/mi........  None..............  EPA FTP test.
CH4 Standard: Tailpipe CH4......  Cap per vehicle...  0.030 g/mi........  None..............  EPA FTP test.
----------------------------------------------------------------------------------------------------------------

    One important flexibility associated with the proposed
CO₂ standard is the proposed option for manufacturers to
obtain credits associated with improvements in their air conditioning
systems. As will be discussed in greater detail in later sections, EPA
is establishing test procedures and design criteria by which
manufacturers can demonstrate improvements in both air conditioner
efficiency (which reduces vehicle tailpipe CO₂ by reducing
the load on the engine) and air conditioner refrigerants (using lower
global warming potency refrigerants and/or improving system design to
reduce GHG emissions associated with leaks). Neither of these
strategies to reduce GHG emissions from air conditioners would be
reflected in the EPA FTP or HFET tests. These improvements would be
translated to a g/mi CO₂-equivalent credit that can be
subtracted from the manufacturer's tailpipe CO₂ compliance
value. EPA expects a high percentage of manufacturers to take advantage
of this flexibility to earn air conditioning-related credits for
MY2012-2016 vehicles such that the average credit earned is about 11
grams per mile CO₂-equivalent in 2016.
---------------------------------------------------------------------------

    \117\ While over 99 percent of the carbon in automotive fuels is
converted to CO₂ in a properly functioning engine,
compliance with the CO₂ standard will also account for
the very small levels of carbon associated with vehicle tailpipe
hydrocarbon (HC) and carbon monoxide (CO) emissions, converted to
CO₂ on a mass basis, as discussed further in section x.
    \118\ CO₂-e refers to CO₂-equivalent, and
is a metric that allows non-CO₂ greenhouse gases (such as
hydrofluorocarbons used as automotive air conditioning refrigerants)
to be expressed as an equivalent mass (i.e., corrected for relative
global warming potency) of CO₂ emissions.
    \119\ FTP is the Federal Test Procedure which uses what is
commonly referred to as the ``city'' driving schedule, and HFET is
the Highway Fuel Economy Test which uses the ``highway'' driving
schedule. Compliance with the CO₂ standard will be based
on the same 2-cycle values that are currently used for CAFE
standards compliance; EPA projects that fleet-wide in-use or real
world CO₂ emissions are approximately 25 percent higher,
on average, than 2-cycle CO₂ values.
---------------------------------------------------------------------------

    A second flexibility being proposed is CO₂ credits for
flexible and dual fuel vehicles, similar to the CAFE credits for such
vehicles which allow manufacturers to gain up to 1.2 mpg in their
overall CAFE ratings. The Energy Independence and Security Act of 2007
(EISA) mandated a phase-out of these flexible fuel vehicle CAFE credits
beginning in 2015, and ending after 2019. EPA is proposing to allow
comparable CO₂ credits for flexible fuel vehicles through MY
2015, but for MY 2016 and beyond, EPA is proposing to treat flexible
and dual fuel vehicles on a CO₂-performance basis,
calculating the overall CO₂ emissions for flexible and dual
fuel vehicles based on a fuel use-weighted average of the
CO₂ levels on gasoline and on a manufacturer's demonstrated
actual usage of the alternative fuel in its vehicle fleet.
    Table III.A.3-2 summarizes EPA projections of industry-wide 2-cycle
CO₂ emissions and fuel economy levels that would be achieved
by manufacturer compliance with the proposed GHG standards for MY2012-2016.
    For MY2011, Table III.A.3-2 uses the projected NHTSA compliance
values for its MY2011 CAFE standards of 30.2 mpg for cars and 24.1 mpg
for trucks, converted to an equivalent combined car and truck
CO₂ level of 325 grams per mile.\120\ EPA believes this is a
reasonable estimate with which to compare the proposed MY2012-2016
CO₂ emission standards. Identifying the proper MY2011
estimate is complicated for many reasons, among them being the turmoil
in the current automotive market for consumers and manufacturers,
uncertain and volatile oil and gasoline prices, the ability of
manufacturers to use flexible fuel vehicle credits to meet MY2011 CAFE
standards, and the fact that most manufacturers have been surpassing
CAFE standards (particularly the car standard) in recent years. Taking
all of these considerations into account, EPA believes that the MY2011
projected CAFE compliance values, converted to CO₂ emissions
levels, represent a reasonable estimate.
---------------------------------------------------------------------------

    \120\ 74 FR 14196.
---------------------------------------------------------------------------

    Table III.A.3-2 shows projected industry-wide average
CO₂ emissions values. The Projected CO₂ Emissions
for the Footprint-Based Standard column shows the CO₂ g/mi
level corresponding with the footprint standard that must be met. It is
based on the proposed CO₂-footprint curves and projected
footprint values, and will decrease each year to 250 grams per mile (g/
mi) in MY2016. For MY2012-2015, the emissions impact of the projected
utilization of flexible fuel vehicle (FFV) credits and the temporary
lead-time allowance alternative standard (TLAAS, discussed below) are
shown in the next two columns. Neither of these programs is proposed to
be available in MY2016. The Projected CO₂ Emissions column
gives the CO₂ emissions levels projected to be achieved
given use of the flexible fuel credits and temporary lead-time
allowance program. This column shows that, relative to the MY 2011
estimate, EPA projects that MY2016 CO₂ emissions will be
reduced by 23 percent over five years. The Projected A/C Credit column
represents the industry wide average air conditioner credit
manufacturers are expected to earn on an equivalent CO₂ gram
per mile basis in a given model year. In MY2016, the projected A/C
credit of 10.6 g/mi represents 14 percent of the 75 g/mi CO₂
emissions reductions associated with the proposed standards. The
Projected 2-cycle CO₂ Emissions column shows the projected
CO₂ emissions as measured over the EPA 2-cycle tests, which
would allow compliance with the standard assuming utilization of the
projected FFV, TLAAS, and A/C credits.

[[Page 49511]]

                Table III.A.3-2--Projected Fleetwide CO[ihel2] Emissions Values (grams per mile)
----------------------------------------------------------------------------------------------------------------
                                     Projected
                                     CO[ihel2]
                                     emissions                 Projected    Projected                 Projected
            Model year                for the     Projected      TLAAS      CO[ihel2]   Projected A/   2-cycle
                                     footprint-   FFV credit     credit     emissions     C credit    CO[ihel2]
                                       based                                                          emissions
                                      standard
----------------------------------------------------------------------------------------------------------------
2011..............................  ...........  ...........  ...........        (325)  ...........        (325)
2012..............................          295            6          0.3          302          3.1          305
2013..............................          286          5.7          0.2          291          5.0          296
2014..............................          276          5.4          0.2          281          7.5          289
2015..............................          263          4.1          0.1          267         10.0          277
2016..............................          250            0            0          250         10.6          261
----------------------------------------------------------------------------------------------------------------

    EPA is also proposing a series of flexibilities for compliance with
the CO₂ standard which are not expected to significantly
affect the projected compliance and achieved values shown above, but
which should significantly reduce the costs of achieving those
reductions. These flexibilities include the ability to earn: annual
credits for a manufacturer's over-compliance with its unique fleet-wide
average standard, early credits from MY2009-2011, credits for early
introduction of advanced technology vehicles, credit for ``off-cycle''
CO₂ reductions not reflected in CO₂/fuel economy
tests, as well as the carry-forward and carry-backward of credits, the
ability to transfer credits between a manufacturer's car and truck
fleets, and a temporary lead-time allowance alternative standard
(included in the tables above) that will permit manufacturers with less
than 400,000 vehicles produced in MY 2009 to designate a fraction of
their vehicles to meet a 25% higher CO₂ standard for MY
2012-2015. All of these proposed flexibilities are discussed in greater
detail in later sections.
    EPA is also proposing caps on the tailpipe emissions of nitrous
oxide (N₂O) and methane (CH₄)--0.010 g/mi for
N₂O and 0.030 g/mi for CH₄--over the EPA FTP
test. While N₂O and CH₄ can be potent greenhouse
gases on a relative mass basis, their emission levels from modern
vehicle designs are extremely low and represent only about 1% of total
light vehicle GHG emissions. These cap standards are designed to ensure
that N₂O and CH₄ emissions levels do not rise in
the future, rather than to force reductions in the already low
emissions levels. Accordingly, these standards are not designed to
require automakers to make any changes in current vehicle designs, and
thus EPA is not projecting any environmental or economic impacts
associated with these proposed standards.
    EPA has attempted to build on existing practice wherever possible
in designing a compliance program for the proposed GHG standards. In
particular, the program structure proposed will streamline the
compliance process for both manufacturers and EPA by enabling
manufacturers to use a single data set to satisfy both the new GHG and
CAFE testing and reporting requirements. Timing of certification,
model-level testing, and other compliance activities also follow
current practices established under the Tier 2 and CAFE programs.
b. Environmental and Economic Benefits and Costs of EPA's Proposed Standards
    In Table III.A.3-3 EPA presents estimated annual net benefits for
the indicated calendar years. The table also shows the net present
values of those benefits for the calendar years 2012-2050 using both a
3% and a 7% discount rate. As discussed previously, EPA recognizes that
much of these same costs and benefits are also attributed to the
proposed CAFE standard contained in this joint proposal.

                               Table III.A.3-3--Projected Quantifiable Benefits and Costs for Proposed CO[ihel2] Standard
                                                 [(In million 2007 $s) [Note: B = unquantified benefits]
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                               2020            2030            2040            2050           NPV, 3%         NPV, 7%
--------------------------------------------------------------------------------------------------------------------------------------------------------
Quantified Annual Costs \a\.............................        -$25,100        -$72,500       -$105,700       -$146,100     -$1,287,600       -$529,500
--------------------------------------------------------------------------------------------------------------------------------------------------------
Benefits from Reduced GHG Emissions at each assumed SCC value:
--------------------------------------------------------------------------------------------------------------------------------------------------------
    SCC 5%..............................................           1,200           3,300           5,700           9,500          69,200          28,600
    SCC 5% Newell-Pizer.................................           2,500           6,600          11,000          19,000         138,400          57,100
    SCC from 3% and 5%..................................           4,700          12,000          22,000          36,000         263,000         108,500
    SCC 3%..............................................           8,200          22,000          38,000          63,000         456,900         188,500
    SCC 3% Newell-Pizer.................................          14,000          36,000          63,000         100,000         761,400         314,200
--------------------------------------------------------------------------------------------------------------------------------------------------------
Other Quantified Externalities
--------------------------------------------------------------------------------------------------------------------------------------------------------
PM[ihel2].[ihel5] Related Benefits \b\ \c\ \d\..........           1,400           3,000           4,600           6,700          59,800          26,300
Energy Security Impacts (price shock)...................           2,300           4,800           6,200           7,800          85,800          38,800
Reduced Refueling.......................................           2,500           4,900           6,400           8,000          89,600          41,000
Value of Increased Driving \e\..........................           4,900          10,000          13,600          18,000         184,700          82,700
Accidents, Noise, Congestion............................          -2,400          -4,900          -6,300          -7,900         -88,200         -40,200
--------------------------------------------------------------------------------------------------------------------------------------------------------
Quantified Net Benefits at each assumed SCC value:
--------------------------------------------------------------------------------------------------------------------------------------------------------
    SCC 5%..............................................          35,000          93,600         135,900         188,200       1,688,500         706,700
    SCC 5% Newell-Pizer.................................          36,300          96,900         141,200         197,700       1,757,700         735,200
    SCC from 3% and 5%..................................          38,500         102,300         152,200         214,700       1,882,300         786,600

[[Page 49512]]


    SCC 3%..............................................          42,000         112,300         168,200         241,700       2,076,200         866,600
    SCC 3% Newell-Pizer.................................          47,800         126,300         193,200         278,700       2,380,700         992,300
--------------------------------------------------------------------------------------------------------------------------------------------------------
\a\ Quantified annual costs are negative because fuel savings are included as negative costs (i.e., positive savings). Since the fuel savings outweigh
  the vehicle technology costs, the costs of as presented here are actually negative (i.e., they represent savings).
\b\ Note that the co-pollutant impacts associated with the standards presented here do not include the full complement of endpoints that, if quantified
  and monetized, would change the total monetized estimate of rule-related impacts. Instead, the co-pollutant benefits are based on benefit-per-ton
  values that reflect only human health impacts associated with reductions in PM[ihel2].[ihel5] exposure. Ideally, human health and environmental
  benefits would be based on changes in ambient PM[ihel2].[ihel5] and ozone as determined by full-scale air quality modeling. However, EPA was unable to
  conduct a full-scale air quality modeling analysis in time for the proposal. EPA does intend to more fully capture the co-pollutant benefits for the
  analysis of the final standards.
\c\ The PM[ihel2].[ihel5]-related benefits (derived from benefit-per-ton values) presented in this table are based on an estimate of premature mortality
  derived from the ACS study (Pope et al., 2002). If the benefit-per-ton estimates were based on the Six Cities study (Laden et al., 2006), the values
  would be approximately 145% (nearly two-and-a-half times) larger.
\d\ The PM[ihel2].[ihel5]-related benefits (derived from benefit-per-ton values) presented in this table assume a 3% discount rate in the valuation of
  premature mortality to account for a twenty-year segmented cessation lag. If a 7% discount rate had been used, the values would be approximately 9% lower.
\e\ Calculated using pre-tax fuel prices.

4. Basis for the Proposed GHG Standards Under Section 202(a)
    EPA statutory authority under section 202(a)(1) of the Clean Air
Act (CAA) is discussed in more detail in Section I.C.2. The following
is a summary of the basis for the proposed standards under section 202(a),
which is discussed in more detail in the following portions of Section III.
    With respect to CO₂ and HFCs, EPA is proposing
attribute-based light-duty car and truck standards that achieve large
and important emissions reductions of GHGs. EPA has evaluated the
technological feasibility of the proposed standards, and the
information and analysis performed by EPA indicates that these
standards are feasible in the lead time provided. EPA and NHTSA have
carefully evaluated the effectiveness of individual technologies as
well as the interactions when technologies are combined. EPA's
projection of the technology that would be used to comply with the
proposed standards indicates that manufacturers will be able to meet
the proposed standards by employing a wide variety of technology that
is already commercially available and can be incorporated into their
vehicle at the time of redesign. In addition to the use of the
manufacturers' redesign cycle, EPA's analysis also takes into account
certain flexibilities that will facilitate compliance especially in the
early years of the program when potential lead time constraints are
most challenging. These flexibilities include averaging, banking, and
trading of various types of credits. For the industry as a whole, EPA's
projections indicate that the proposed standards can be met using
technology that will be available in the lead-time provided.
    To account for additional lead-time concerns for various
manufacturers of typically higher performance vehicles, EPA is
proposing a Temporary Lead-time Allowance that will further facilitate
compliance for limited volumes of such vehicles in the program's
initial years. For a few very small volume manufacturers, EPA projects
that manufacturers will likely comply using a combination of credits
and technology.
    EPA has also carefully considered the cost to manufacturers of
meeting the standards, estimating piece costs for all candidate
technologies, direct manufacturing costs, cost markups to account for
manufacturers' indirect costs, and manufacturer cost reductions
attributable to learning. In estimating manufacturer costs, EPA took
into account manufacturers' own standard practices such as making major
changes to model technology packages during a planned redesign cycle.
EPA then projected the average cost across the industry to employ this
technology, as well as manufacturer-by-manufacturer costs. EPA
considers the per vehicle costs estimated from this analysis to be well
within a reasonable range in light of the emissions reductions and
benefits received. EPA projects, for example, that the fuel savings
over the life of the vehicles will more than offset the increase in
cost associated with the technology used to meet the standards.
    EPA has also evaluated the impacts of these standards with respect
to reductions in GHGs and reductions in oil usage. For the lifetime of
the model year 2012-2016 vehicles we estimate GHG reductions of
approximately 950 million metric tons CO₂ eq. and fuel
reductions of 1.8 billion barrels of oil. These are important and
significant reductions that would be achieved by the proposed
standards. EPA has also analyzed a variety of other impacts of the
standards, ranging from the standards' effects on emissions of non-GHG
pollutants, impacts on noise, energy, safety and congestion. EPA has
also quantified the cost and benefits of the proposed standards, to the
extent practicable. Our analysis to date indicates that the overall
quantified benefits of the proposed standards far outweigh the
projected costs. Utilizing a 3% discount rate and a $20 per ton social
cost of carbon we estimate the total net social benefits over the life
of the model year 2012-2016 vehicles is $192 billion, and the net
present value of the net social benefits of the standards through the
year 2050 is $1.9 trillion dollars. These values are estimated at $136
billion and $787 billion, respectively, using a 7% discount rate and
the $20 per ton SCC value.
    Under section 202(a) EPA is called upon to set standards that
provide adequate lead-time for the development and application of
technology to meet the standards. EPA's proposed standards satisfy this
requirement, as discussed above. In setting the standards, EPA is
called upon to weigh and balance various factors, and to exercise
judgment in setting standards that are a reasonable balance of the
relevant factors. In this case, EPA has considered many factors, such
as cost, impacts on emissions (both GHG and non-GHG), impacts on oil
conservation, impacts on noise, energy, safety, and other factors, and
has where practicable quantified the costs and benefits of the rule. In
summary, given the technical feasibility of the standard, the moderate
cost per vehicle in light of the savings in fuel costs over the life
time of the vehicle, the very significant reductions

[[Page 49513]]

in emissions and in oil usage, and the significantly greater quantified
benefits compared to quantified costs, EPA is confident that the
proposed standards are an appropriate and reasonable balance of the
factors to consider under section 202(a). See Husqvarna AB v. EPA, 254
F.3d 195, 200 (D.C. Cir. 2001) (great discretion to balance statutory
factors in considering level of technology-based standard, and
statutory requirement ``to [give appropriate] consideration to the cost
of applying * * * technology'' does not mandate a specific method of
cost analysis); see also Hercules Inc. v. EPA, 598 F.2d 91, 106 (D.C.
Cir. 1978) (``In reviewing a numerical standard we must ask whether the
agency's numbers are within a zone of reasonableness, not whether its
numbers are precisely right''); Permian Basin Area Rate Cases, 390 U.S.
747, 797 (1968) (same); Federal Power Commission v. Conway Corp., 426
U.S. 271, 278 (1976) (same); Exxon Mobil Gas Marketing Co. v. FERC, 297
F.3d 1071, 1084 (D.C. Cir. 2002) (same).
    EPA recognizes that the vast majority of technology which we are
considering for purposes of setting standards under section 202(a) is
commercially available and already being utilized to a limited extent
across the fleet. The vast majority of the emission reductions which
would result from this proposed rule would result from the increased
use of these technologies. EPA also recognizes that this proposed rule
would enhance the development and limited use of more advanced
technologies, such as PHEVs and EVs. In this technological context,
there is no clear cut line that indicates that only one projection of
technology penetration could potentially be considered feasible for
purposes of section 202(a), or only one standard that could potentially
be considered a reasonable balancing of the factors relevant under
section 202(a). EPA has therefore evaluated two sets of alternative
standards, one more stringent than the proposed standards and one less
stringent.
    The alternatives are 4% per year increase in standards which would
be less stringent than our proposal and a 6% per year increase in the
standards which would be more stringent than our proposal. EPA is not
proposing either of these. As discussed in Section III.D.7, the 4% per
year compared to the proposal forgoes CO₂ reductions which
can be achieved at reasonable costs and are achievable by the industry
within the rule's timeframe. The 6% per year alternative requires a
significant increase in the projected required technology which may not
be achievable in this timeframe due to the limited available lead time
and the current difficult financial condition of the automotive
industry. (See Section III.D.7 for a detailed discussion of why EPA is
not proposing either of the alternatives.) EPA thus believes that it is
appropriate to propose the CO₂ standards discussed above.
EPA invites comment on all aspects of this judgment, as well as comment
on the alternative standards.
    EPA is also proposing standards for N₂O and
CH₄. EPA has designed these standards to act as emission
rate (i.e., gram per mile) caps and to avoid future increases in light
duty vehicle emissions. As discussed in Section III.B.6, N₂O
and CH₄ emissions are already generally well controlled by
current emissions standards, and EPA has not identified clear
technological steps available to manufacturers today that would
significantly reduce current emission levels for the vast majority of
vehicles manufactured today (i.e., stoichiometric gasoline vehicles).
However, for both N₂O and CH₄, some vehicle
technologies (and, for CH₄, use of natural gas fuel) could
potentially increase emissions of these GHGs in the future, and EPA
believes it is important that this be avoided. EPA expects that, almost
universally across current car and truck designs, manufacturers will be
able to meet the ``cap'' standards with little if any technological
improvements or cost. EPA has designed the level of the N₂O
and CH₄ standards with the intent that manufacturers would
be able to meet them without the need for technological improvement; in
other words, these emission standards are designed to be ``anti-
backsliding'' standards.

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

    EPA is proposing new emission standards to control greenhouse gases
(GHGs) from light-duty vehicles. First, EPA is proposing emission
standards for carbon dioxide (CO₂) on a gram per mile (g/
mile) basis that would apply to a manufacturer's fleet of cars, and a
separate standard that would apply to a manufacturer's fleet of trucks.
CO₂ is the primary pollutant resulting from the combustion
of vehicular fuels, and the amount of CO₂ emitted is
directly correlated to the amount of fuel consumed. Second, EPA is
providing auto manufacturers with the opportunity to earn credits
toward the fleet-wide average CO₂ standards for improvements
to air conditioning systems, including both hydrofluorocarbon (HFC)
refrigerant losses (i.e., system leakage) and indirect CO₂
emissions related to the increased load on the engine. Third, EPA is
proposing separate emissions standards for two other GHG pollutants:
Methane (CH₄) and nitrous oxide (N₂O).
CH₄ and N₂O emissions relate closely to the
design and efficient use of emission control hardware (i.e., catalytic
converters). The standards for CH₄ and N₂O would
be set as a cap that would limit emissions increases and prevent
backsliding from current emission levels. The proposed standards
described below would apply to passenger cars, light-duty trucks, and
medium-duty passenger vehicles (MDPVs). As an overall group, they are
referred to in this preamble as light vehicles or simply as vehicles.
In this preamble section passenger cars may be referred to simply as
``cars'', and light-duty trucks and MDPVs as ``light trucks'' or
``trucks.'' \121\
---------------------------------------------------------------------------

    \121\ As described in Section III.B.2., EPA is proposing for
purposes of GHG emissions standards to use the same vehicle category
definitions as are used in the CAFE program.
---------------------------------------------------------------------------

    EPA is establishing a system of averaging, banking, and trading of
credits integral to the fleet averaging approach, based on manufacturer
fleet average CO₂ performance, as discussed in Section
III.B.4. This approach is similar to averaging, banking, and trading
(ABT) programs EPA has established in other programs and is also
similar to provisions in the CAFE program. In addition to traditional
ABT credits based on the fleet emissions average, EPA is also proposing
to include A/C credits as an aspect of the standards, as mentioned
above. EPA is also proposing several additional credit provisions that
apply only in the initial model years of the program. These include
flex fuel vehicle credits, credits based on the use of advanced
technologies, and generation of credits prior to model year 2012. The
proposed A/C credits and additional credit opportunities are described
in Section III.C. These credit programs would provide flexibility to
manufacturers, which may be especially important during the early
transition years of the program. EPA is also proposing to allow a
manufacturer to carry a deficit into the future for a limited number of
model years. A parallel provision, referred to as credit carry-back, is
proposed as part of the CAFE program.
1. What Fleet-Wide Emissions Levels Correspond to the CO₂ Standards?
    The proposed attribute-based CO₂ standards, if made
final, are projected to achieve a national fleet-wide average, covering
both light cars and trucks, of

[[Page 49514]]

250 grams/mile of CO₂ in model year (MY) 2016. This includes
CO₂-equivalent emission reductions from A/C improvements,
reflected as credits in the standard. The standards would begin with MY
2012, with a generally linear increase in stringency from MY 2012
through MY 2016. EPA is proposing separate standards for cars and light
trucks. The tables in this section below provide overall fleet average
levels that are projected for both cars and light trucks over the
phase-in period which is estimated to correspond with the proposed
standards. The actual fleet-wide average g/mi level that will be
achieved in any year for cars and trucks will depend on the actual
production for that year, as well as the use of the various credit and
averaging, banking, and trading provisions. For example, in any year,
manufacturers may generate credits from cars and use them for
compliance with the truck standard. Such transfer of credits between
cars and trucks is not reflected in the table below. In Section III.F,
the year-by-year estimate of emissions reductions that are projected to
be achieved by the proposed standards are discussed.
    In general, the proposed schedule of standards acts as a phase-in
to the MY 2016 standards, and reflects consideration of the appropriate
lead-time for each manufacturer to implement the requisite emission
reductions technology across its product line.\122\ Note that 2016 is
the final model year in which standards become more stringent. The 2016
CO₂ standards would remain in place for 2017 and later model
years, until revised by EPA in a future rulemaking.
---------------------------------------------------------------------------

    \122\ See CAA section 202(a)(2).
---------------------------------------------------------------------------

    EPA estimates that, on a combined fleet-wide national basis, the
proposed 2016 MY standards would achieve a level of 250 g/mile
CO₂, including CO₂-equivalent credits from A/C
related reductions. The derivation of the 250 g/mile estimate is
described in Section III.B.2.
    EPA has estimated the overall fleet-wide CO₂-equivalent
emission levels that correspond with the proposed attribute-based
standards, based on the projections of the composition of each
manufacturer's fleet in each year of the program. Tables III.B.1-1 and
III.B.1-2 provide these estimates for each manufacturer.\123\
---------------------------------------------------------------------------

    \123\ These levels do not include the effect of flexible fuel
credits, transfer of credits between cars and trucks, temporary lead
time allowance, or any other credits.

 Table III.B.1-1--Estimated Fleet CO2-Equivalent Levels Corresponding to
                     the Proposed Standards for Cars
------------------------------------------------------------------------
                                                Model year
          Manufacturer           ---------------------------------------
                                   2012    2013    2014    2015    2016
------------------------------------------------------------------------
BMW.............................     265     257     249     238     227
Chrysler........................     266     259     251     242     231
Daimler.........................     270     263     257     245     234
Ford............................     266     259     251     239     228
General Motors..................     266     258     250     239     228
Honda...........................     259     251     244     232     221
Hyundai.........................     260     252     244     233     221
Kia.............................     262     253     246     235     223
Mazda...........................     258     250     243     231     220
Mitsubishi......................     255     247     240     228     217
Nissan..........................     263     255     247     236     225
Porsche.........................     242     234     227     215     204
Subaru..........................     252     244     237     225     214
Suzuki..........................     244     236     229     217     206
Tata............................     286     278     271     259     248
Toyota..........................     257     250     242     231     220
Volkswagen......................     254     246     239     228     217
------------------------------------------------------------------------


 Table III.B.1-2--Estimated Fleet CO2-Equivalent Levels Corresponding to
                 the Proposed Standards for Light Trucks
------------------------------------------------------------------------
                                                Model year
          Manufacturer           ---------------------------------------
                                   2012    2013    2014    2015    2016
------------------------------------------------------------------------
BMW.............................     334     324     313     298     283
Chrysler........................     349     339     329     315     300
Daimler.........................     346     334     323     308     293
Ford............................     363     352     343     329     314
General Motors..................     372     361     351     337     322
Honda...........................     333     322     311     295     280
Hyundai.........................     330     320     308     293     278
Kia.............................     341     330     319     303     288
Mazda...........................     321     311     300     286     271
Mitsubishi......................     320     310     299     284     269
Nissan..........................     352     341     332     318     303
Porsche.........................     338     327     316     301     286
Subaru..........................     319     308     297     282     267
Suzuki..........................     324     313     301     286     271
Tata............................     326     316     305     289     275
Toyota..........................     342     332     320     305     291

[[Page 49515]]

Volkswagen......................     344     333     322     307     292
------------------------------------------------------------------------

    These estimates were aggregated based on projected production
volumes into the fleet-wide averages for cars and trucks (Table
III.B.1-3).\124\
---------------------------------------------------------------------------

    \124\ Due to rounding during calculations, the estimated fleet-
wide CO₂-equivalent levels may vary by plus or minus 1 gram.

       Table III.B.1-3--Estimated Fleet-wide CO2-Equivalent Levels
                 Corresponding to the Proposed Standards
------------------------------------------------------------------------
                                               Cars           Trucks
------------------------------------------------------------------------
               Model year                   CO2 (g/mi)      CO2 (g/mi)
------------------------------------------------------------------------
2012....................................             261             352
2013....................................             254             341
2014....................................             245             331
2015....................................             234             317
2016 and later..........................             224             303
------------------------------------------------------------------------

    As shown in Table III.B.1-3, fleet-wide CO₂-equivalent
emission levels for cars under the proposed approach are projected to
decrease from 261 to 224 grams per mile between MY 2012 and MY 2016.
Similarly, fleet-wide CO₂-equivalent emission levels for
trucks are projected to decrease from 352 to 303 grams per mile. These
numbers do not include the effects of other flexibilities and credits
in the program. The estimated achieved values can be found in Chapter 5
of the Draft Regulatory Impact Analysis (DRIA).
    EPA has also estimated the average fleet-wide levels for the
combined car and truck fleets. These levels are provided in Table
III.B.1-4. As shown, the overall fleet average CO₂ level is
expected to be 250 g/mile in 2016.

  Table III.B.1-4--Estimated Fleet-wide Combined CO2-Equivalent Levels
                 Corresponding to the Proposed Standards
------------------------------------------------------------------------
                                                           Combined car
---------------------------------------------------------    and truck
                                                         ---------------
                       Model year                           CO2 (g/mi)
------------------------------------------------------------------------
2012....................................................             295
2013....................................................             286
2014....................................................             276
2015....................................................             263
2016....................................................             250
------------------------------------------------------------------------

    As noted above, EPA is proposing standards that would result in
increasingly stringent levels of CO₂ control from MY 2012
though MY 2016--applying the CO₂ footprint curves applicable
in each model year to the vehicles expected to be sold in each model
year produces fleet-wide annual reductions in CO₂ emissions.
As explained in Section III.D below and the relevant support documents,
EPA believes that the proposed level of improvement achieves important
CO₂ emissions reductions through the application of feasible
control technology at reasonable cost, considering the needed lead time
for this program. EPA further believes that the proposed averaging,
banking and trading provisions, as well as other credit-generating
mechanisms, allow manufacturers further flexibilities which reduce the
cost of the proposed CO₂ standards and help to provide
adequate lead time. EPA believes this approach is justified under
section 202(a) of the Clean Air Act.
    EPA has analyzed the feasibility under the CAA of achieving the
proposed CO₂ standards, based on projections of what actions
manufacturers are expected to take to reduce emissions. The results of
the analysis are discussed in detail in Section III.D below and in the
DRIA. EPA also presents the estimated costs and benefits of the
proposed car and truck CO₂ standards in Section III.H. In
developing the proposal, EPA has evaluated the kinds of technologies
that could be utilized by the automobile industry, as well as the
associated costs for the industry and fuel savings for the consumer,
the magnitude of the GHG reductions that may be achieved, and other
factors relevant under the CAA.
    With respect to the lead time and cost of incorporating technology
improvements that reduce GHG emissions, EPA and NHTSA place important
weight on the fact that during MYs 2012-2016 manufacturers are expected
to redesign and upgrade their light-duty vehicle products (and in some
cases introduce entirely new vehicles not on the market today). Over
these five model years there would be an opportunity for manufacturers
to evaluate almost every one of their vehicle model platforms and add
technology in a cost-effective way to control GHG emissions and improve
fuel economy. This includes redesign of the air conditioner systems in
ways that will further reduce GHG emissions. The time-frame and levels
for the proposed standards, as well as the ability to average, bank and
trade credits and carry a deficit forward for a limited time, are
expected to provide manufacturers the time needed to incorporate
technology that will achieve GHG reductions, and to do this as part of
the normal vehicle redesign process. This is an important aspect of the
proposal, as it would avoid the much higher costs that would occur if
manufacturers needed to add or change technology at times other than
these scheduled redesigns. This time period would also provide
manufacturers the opportunity to plan for compliance using a multi-year
time frame, again in accord with their normal business practice.
    Consistent with the requirement of CAA section 202(a)(1) that
standards be applicable to vehicles ``for their useful life,'' EPA is
proposing CO₂ vehicle standards that would apply for the
useful life of the vehicle. Under section 202(i) of the Act, which
authorized the Tier 2 standards, EPA established a useful life period
of 10 years or 120,000 miles, whichever first occurs, for all Tier 2
light-duty vehicles and light-duty trucks.\125\ Tier 2 refers to EPA's
standards for criteria pollutants such as NO_X, HC, and CO.
EPA is proposing new CO₂ standards for the same group of
vehicles, and therefore the Tier 2 useful life would apply for
CO₂ standards as well. The in-use emission standard will be
10% higher than the certification standard, to address issues of
production variability and test-to-test variability. The in-use
standard is discussed in Section III.E.
---------------------------------------------------------------------------

    \125\ See 65 FR 6698 (February 10, 2000).
---------------------------------------------------------------------------

    EPA is proposing to measure CO₂ for certification and
compliance purposes using the same test procedures currently used by
EPA for measuring fuel economy. These procedures are the Federal Test
Procedure (FTP or ``city'' test) and the Highway Fuel Economy

[[Page 49516]]

Test (HFET or ``highway'' test).\126\ This corresponds with the data
used to develop the footprint-based CO₂ standards, since the
data on control technology efficiency was also developed in reference
to these test procedures. Although EPA recently updated the test
procedures used for fuel economy labeling, to better reflect the actual
in-use fuel economy achieved by vehicles, EPA is not proposing to use
these test procedures for the CO₂ standards proposed here,
given the lack of data on control technology effectiveness under these
procedures.\127\ As stated in Section I, EPA and NHTSA invite comments
on potential amendments to the CAFE and GHG test procedures, including
but not limited to air conditioner-related emissions, that could be
implemented beginning in MY 2017.
---------------------------------------------------------------------------

    \126\ EPA established the FTP for emissions measurement in the
early 1970s. In 1976, in response to the Energy Policy and
Conservation Act (EPCA) statute, EPA extended the use of the FTP to
fuel economy measurement and added the HFET.\126\ The provisions in
the 1976 regulation, effective with the 1977 model year, established
procedures to calculate fuel economy values both for labeling and
for CAFE purposes.
    \127\ See 71 FR 77872, December 27, 2006.
---------------------------------------------------------------------------

    EPA proposes to include hydrocarbons (HC) and carbon monoxide (CO)
in its CO₂ emissions calculations on a CO₂-
equivalent basis. It is well accepted that HC and CO are typically
oxidized to CO₂ in the atmosphere in a relatively short
period of time and so are effectively part of the CO₂
emitted by a vehicle. In terms of standard stringency, accounting for
the carbon content of tailpipe HC and CO emissions and expressing it as
CO₂-equivalent emissions would add less than one percent to
the overall CO₂-equivalent emissions level. This will also
ensure consistency with CAFE calculations since HC and CO are included
in the ``carbon balance'' methodology that EPA uses to determine fuel
usage as part of calculating vehicle fuel economy levels.
2. What Are the CO₂ Attribute-Based Standards?
    EPA proposes to use the same vehicle category definitions that are
used in the CAFE program for the 2011 model year standards.\128\ The
CAFE vehicle category definitions differ slightly from the EPA
definitions for cars and light trucks used for the Tier 2 program, as
well as other EPA vehicle programs. Specifically, NHTSA's
reconsideration of the CAFE program statutory language has resulted in
many two-wheel drive SUVs under 6000 pounds gross vehicle weight being
reclassified as cars under the CAFE program. The proposed approach of
using CAFE definitions allows EPA's proposed CO₂ standards
and the proposed CAFE standards to be harmonized across all vehicles.
In other words, vehicles would be subject to either car standards or
truck standards under both programs, and not car standards under one
program and trucks standards under the other.
---------------------------------------------------------------------------

    \128\ See 49 CFR part 523.
---------------------------------------------------------------------------

    EPA is proposing separate car and truck standards, that is,
vehicles defined as cars have one set of footprint-based curves for MY
2012-2016 and vehicles defined as trucks have a different set for MY
2012-2016. In general, for a given footprint the CO₂ g/mi
target for trucks is less stringent then for a car with the same footprint.
    EPA is not proposing a single fleet standard where all cars and
trucks are measured against the same footprint curve for several
reasons. First, some vehicles classified as trucks (such as pick-up
trucks) have certain attributes not common on cars which attributes
contribute to higher CO₂ emissions--notably high load
carrying capability and/or high towing capability. Due to these
differences, it is reasonable to separate the light-duty vehicle fleet
into two groups. Second, EPA would like to harmonize key program design
elements of the GHG standards with NHTSA's CAFE program where it is
reasonable to do so. NHTSA is required by statute to set separate
standards for passenger cars and for non-passenger cars.
    Finally, most of the advantages of a single standard for all light
duty vehicles are also present in the two-fleet standards proposed
here. Because EPA is proposing to allow unlimited credit transfer
between a manufacturer's car and truck fleets, the two fleets can
essentially be viewed as a single fleet when manufacturers consider
compliance strategies. Manufacturers can thus choose on which vehicles
within their fleet to focus GHG reducing technology and then use credit
transfers as needed to demonstrate compliance, just as they would if
there was a single fleet standard. The one benefit of a single light-
duty fleet not captured by a two-fleet approach is that a single fleet
prevents potential ``gaming'' of the car and truck definitions to try
and design vehicles which are more similar to passenger cars but which
may meet the regulatory definition of trucks. Although this is of
concern to EPA, we do not believe at this time that concern is
sufficient to outweigh the other reasons for proposing separate car and
truck fleet standards. EPA requests comment on this approach.
    For model years 2012 and later, EPA is proposing a series of
CO₂ standards that are described mathematically by a family
of piecewise linear functions (with respect to vehicle footprint). The
form of the function is as follows:

CO₂ = a, if x <= l
CO₂ = cx + d, if l < x <= h
CO₂ = b, if x > h

Where:
CO₂ = the CO₂ target value for a given footprint (in g/mi)
a = the minimum CO₂ target value (in g/mi)
b = the maximum CO₂ target value (in g/mi)
c = the slope of the linear function (in g/mi per sq ft)
d = is the zero-offset for the line (in g/mi CO₂)
x = footprint of the vehicle model (in square feet, rounded to the
nearest tenth)
l & h are the lower and higher footprint limits, constraints, or the
boundary (``kinks'') between the flat regions and the intermediate sloped line.

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

                                                       Table III.B.2-1--Parameter Values for Cars
                                                             [For CO2 gram per mile targets]
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                                                               Lower           Upper
                       Model year                                a               b               c               d          constraint      constraint
--------------------------------------------------------------------------------------------------------------------------------------------------------
2012....................................................             242             313            4.72            48.8              41              56
2013....................................................             234             305            4.72            40.8              41              56
2014....................................................             227             297            4.72            33.2              41              56
2015....................................................             215             286            4.72            22.0              41              56
2016 and later..........................................             204             275            4.72            10.9              41              56
--------------------------------------------------------------------------------------------------------------------------------------------------------


[[Page 49517]]

                                                      Table III.B.2-2--Parameter Values for Trucks
                                                             [For CO2 gram per mile targets]
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                                                               Lower           Upper
                       Model year                                a               b               c               d          constraint      constraint
--------------------------------------------------------------------------------------------------------------------------------------------------------
2012....................................................             298             399            4.04           132.6              41              66
2013....................................................             287             388            4.04           121.6              41              66
2014....................................................             276             377            4.04           110.3              41              66
2015....................................................             261             362            4.04            95.2              41              66
2016 and later..........................................             246             347            4.04            80.4              41              66
--------------------------------------------------------------------------------------------------------------------------------------------------------

    The equations can be shown graphically for each vehicle category,
as shown in Figures III.B.2-1 and III.B.2-2. These standards (or
functions) decrease from 2012-2016 with a vertical shift. A more
detailed description of the development of the attribute based standard
can be found in Chapter 2 of the Draft Joint TSD. More background
discussion on other alternative attributes and curves EPA explored can
be found in the EPA DRIA. EPA recognizes that the CAA does not mandate
that EPA use an attribute based standard, as compared to NHTSA's
obligations under EPCA. The EPA believes that proposing a footprint-
based program will harmonize EPA's proposed program and the proposed
CAFE program as a single national program, resulting in reduced
compliance complexity for manufacturers. EPA's reasons for proposing to
use an attribute based standard are discussed in more detail in the
Joint TSD. Comments are requested on this proposal to use the
attribute-based approach for regulating tailpipe CO₂ emissions.

BILLING CODE 4910-59-P

[[Page 49518]]
[GRAPHIC] [TIFF OMITTED] TP28SE09.010
[[Page 49519]]
[GRAPHIC] [TIFF OMITTED] TP28SE09.011

BILLING CODE 4910-59-C
3. Overview of How EPA's Proposed CO₂ Standards Would Be
Implemented for Individual Manufacturers
    This section provides a brief overview of how EPA proposes to
implement the CO₂ standards. Section III.E explains EPA's
proposed approach for certification and compliance in detail. EPA is
proposing two kinds of standards--fleet average standards determined by
a manufacturer's fleet profile of various models, and in-use standards
that would apply to the various models that make up the manufacturer's
fleet. Although this is similar in concept to the current light-duty
vehicle Tier 2 program, there are

[[Page 49520]]

important differences. In explaining EPA's proposal for the
CO₂ standards, it is useful to summarize how the Tier 2 program works.
    Under Tier 2, manufacturers select a test vehicle prior to
certification and test the vehicle and/or its emissions hardware to
determine both its emissions performance when new and the emissions
performance expected at the end of its useful life. Based on this
testing, the vehicle is assigned to one of several specified bins of
emissions levels, identified in the Tier 2 rule, and this bin level
becomes the emissions standard for the test group the test vehicle
represents. All of the vehicles in the group must meet the emissions
level for that bin throughout their useful life. The emissions level
assigned to the bin is also used in calculating the manufacturer's
fleet average emissions performance.
    Since compliance with the Tier 2 fleet average depends on actual
test group sales volumes and bin levels, it is not possible to
determine compliance at the time the manufacturer applies for and
receives a certificate of conformity for a test group. Instead, at
certification, the manufacturer demonstrates that the vehicles in the
test group are expected to comply throughout their useful life with the
emissions bin assigned to that test group, and makes a good faith
demonstration that its fleet is expected to comply with the Tier 2
average when the model year is over. EPA issues a certificate for the
vehicles covered by the test group based on this demonstration, and
includes a condition in the certificate that if the manufacturer does
not comply with the fleet average then production vehicles from that
test group will be treated as not covered by the certificate to the
extent needed to bring the manufacturer's fleet average into compliance
with Tier 2.
    EPA proposes to retain the Tier 2 approach of requiring
manufacturers to demonstrate in good faith at the time of certification
that models in a test group will meet applicable standards throughout
useful life. EPA also proposes to retain the practice of conditioning
certificates upon attainment of the fleet average standard. However,
there are several important differences between a Tier 2 type of
program and the CO₂ standards program EPA is proposing.
These differences and resulting modifications to certification are
summarized below and are described in detail in Section III.E.
    EPA is proposing to certify test groups as it does for Tier 2, with
the CO₂ emission results for the test vehicle as the initial
or default standard for all of the models in the test group. However,
manufacturers would later substitute test data for individual models in
that test group, based on the model level fuel economy testing that
typically occurs through the course of the model year. This model level
data would then be used to assign a distinct certification level for
that model, instead of the initial test group level. These model level
results would then be used to calculate the fleet average after the end
of production.\129\ The option to substitute model level test data for
the test group data is at the manufacturer's discretion, except they
are required as under the CAFE test protocols to test, at a minimum,
enough models to represent 90 percent of their production. The test
group level would continue to apply for any model that is not covered
by model level testing. A related difference is that the fleet average
calculation for Tier 2 is based on test group bin levels and test group
sales whereas under this proposal the CO₂ fleet level would
be based on a combination of test group and model-level emissions and
model-level production. For the new CO₂ standards, EPA is
proposing to use production rather than sales in calculating the fleet
average in order to more closely conform with CAFE, which is a
production-based program. EPA does not expect any significant
environmental effect because there is little difference between
production and sales, and this will reduce the complexity of the
program for manufacturers.
---------------------------------------------------------------------------

    \129\ The final in-use vehicle standards for each model would
also be based on the model-level fuel economy testing. As discussed
in Section III.E.4, an in-use adjustment factor would be applied to
the model level results to determine the in-use standard that would
apply during the useful life of the vehicle.
---------------------------------------------------------------------------

4. Averaging, Banking, and Trading Provisions for CO₂ Standards
    As explained above, a fleet average CO₂ program for
passenger cars and light trucks is proposed. EPA has implemented
similar averaging programs for a range of motor vehicle types and
pollutants, from the Tier 2 fleet average for NO_X to
motorcycle hydrocarbon (HC) plus oxides of nitrogen (NO_X)
emissions to NO_X and particulate matter (PM) emissions from
heavy-duty engines.\130\ The proposed program would operate much like
EPA's existing averaging programs in that manufacturers would calculate
production-weighted fleet average emissions at the end of the model
year and compare their fleet average with a fleet average standard to
determine compliance. As in other EPA averaging programs, the Agency is
also proposing a comprehensive program for averaging, banking, and
trading of credits which together will help manufacturers in planning
and implementing the orderly phase-in of emissions control technology
in their production, using their typical redesign schedules.
---------------------------------------------------------------------------

    \130\ For example, see the Tier 2 light-duty vehicle emission
standards program (65 FR 6698, February 10, 2000), the 2010 and
later model year motorcycle emissions program (69 FR 2398, January
15, 2004), and the 2007 and later model year heavy-duty engine and
vehicle standards program (66 FR 5001, January 18, 2001).
---------------------------------------------------------------------------

    Averaging, Banking, and Trading (ABT) of emissions credits has been
an important part of many mobile source programs under CAA Title II,
both for fuels programs as well as for engine and vehicle programs. ABT
is important because it can help to address many issues of
technological feasibility and lead-time, as well as considerations of
cost. ABT is an integral part of the standard setting itself, and is
not just an add-on to help reduce costs. In many cases, ABT resolves
issues of lead-time or technical feasibility, allowing EPA to set a
standard that is either numerically more stringent or goes into effect
earlier than could have been justified otherwise. This provides
important environmental benefits at the same time it increases
flexibility and reduces costs for the regulated industry.
    This section discusses generation of credits by achieving a fleet
average CO₂ level that is lower than the manufacturer's
CO₂ fleet average standard. EPA is proposing a variety of
additional ways credits may be generated by manufacturers. Section
III.C describes these additional opportunities to generate credits in
detail. EPA is proposing that credits could be earned through A/C
system improvements beyond a specified baseline. Credits can also be
generated by producing alternative fuel vehicles, by producing advanced
technology vehicles including electric vehicles, plug-in hybrids, and
fuel cell vehicles, and by using technologies that improve off-cycle
emissions. In addition, EPA is proposing that early credits could be
generated prior to the proposed program's MY 2012 start date. The
credits would be used in calculating the fleet averages at the end of
the model year, with the exception of early credits which would be
tracked separately. These proposed credit generating opportunities are
described below in Section III.C.
    As explained earlier, manufacturers would determine the fleet
average standard that would apply to their car fleet and the standard
for their truck fleet from the applicable attribute-based curve. A
manufacturer's credit or debit

[[Page 49521]]

balance would be determined by comparing their fleet average with the
manufacturer's CO₂ standard for that model year. The
standard would be calculated from footprint values on the attribute
curve and actual production levels of vehicles at each footprint. A
manufacturer would generate credits if its car or truck fleet achieves
a fleet average CO₂ level lower than its standard and would
generate debits if its fleet average CO₂ level is above that
standard. At the end of the model year, each manufacturer would
calculate a production-weighted fleet average for each averaging set,
cars and trucks. A manufacturer's car or truck fleet that achieves a
fleet average CO₂ level lower than its standard would
generate credits, and if its fleet average CO₂ level is
above that standard its fleet would generate debits.
    EPA is proposing to account for the difference in expected lifetime
vehicle miles traveled (VMT) between cars and trucks in order to
preserve CO₂ reductions when credits are transferred between
cars and trucks. As directed by EISA, NHTSA accomplishes this in the
CAFE program by using an adjustment factor that is applied to credits
when they are transferred between car and truck compliance categories.
The CAFE adjustment factor accounts for two different influences that
can cause the transfer of car and truck credits (expressed in tenths of
a mpg), if left unadjusted, to potentially negate fuel reductions.
First, mpg is not linear with fuel consumption, i.e., a 1 mpg
improvement above a standard will imply a different amount of actual
fuel consumed depending on the level of the standard. Second, NHTSA's
conversion corrects for the fact that the typical lifetime miles for
cars is less than that for trucks, meaning that credits earned for cars
and trucks are not necessarily equal. NHTSA's adjustment factor
essentially converts credits into vehicle lifetime gallons to ensure
preservation of fuel savings and the transfer credits on an equal
basis, and then converts back to the statutorily required credit units
of tenths of a mile per gallon. To convert to gallons NHTSA's
conversion must take into account the expected lifetime mileage for
cars and trucks. Because EPA is proposing standards that are expressed
on a CO₂ gram per mile basis, which is linear with fuel
consumption, EPA's credit calculations do not need to account for the
first issue noted above. However, EPA is proposing to account for the
second issue by expressing credits when they are generated in total
lifetime megagrams (metric tons), rather than through the use of
conversion factors that would apply at certain times. In this way
credits could be freely exchanged between car and truck compliance
categories without adjustment. Additional detail regarding this
approach, including a discussion of the vehicle lifetime mileage
estimates for cars and trucks can be found in Section III.E.5. A
discussion of the estimated vehicle lifetime miles traveled can be
found in Chapter 4 of the draft Joint Technical Support Document. EPA
requests comment on the proposed approach.
    A manufacturer that generates credits in a given year and vehicle
category could use those credits in essentially four ways, although
with some limitations. These provisions are very similar to those of
other EPA averaging, banking, and trading programs. These provisions
have the potential to reduce costs and compliance burden, and support
the feasibility of the standards being proposed in terms of lead time
and orderly redesign by a manufacturer, thus promoting and not reducing
the environmental benefits of the program.
    First, the manufacturer would have to offset any deficit that had
accrued in that averaging set in a prior model year and had been
carried over to the current model year. In such a case, the
manufacturer would be obligated to use any current model year credits
to offset that deficit. This is referred to in the CAFE program as
credit carry-back. EPA's proposed deficit carry-forward, or credit
carry-back provisions are described further, below.
    Second, after satisfying any needs to offset pre-existing deficits
within a vehicle category, remaining credits could be banked, or saved
for use in future years. EPA is proposing that credits generated in
this program be available to the manufacturer for use in any of the
five years after the year in which they were generated, consistent with
the CAFE program under EISA. This is also referred to as a credit
carry-forward provision. For other new emission control programs, EPA
has sometimes initially restricted credit life to allow time for the
Agency to assess whether the credit program is functioning as intended.
When EPA first offered averaging and banking provisions in its light-
duty emissions control program (the National Low Emission Vehicle
Program), credit life was restricted to three years. The same is true
of EPA's early averaging and banking program for heavy-duty engines. As
these programs matured and were subsequently revised, EPA became
confident that the programs were functioning as intended and that the
standards were sufficiently stringent to remove the restrictions on
credit life.
    EPA is therefore acting consistently with our past practice in
proposing to reasonably restrict credit life in this new program. The
Agency believes, subject to consideration of public comment, that a
credit life of five years represents an appropriate balance between
promoting orderly redesign and upgrade of the emissions control
technology in the manufacturer's fleet and the policy goal of
preventing large numbers of credits accumulated early in the program
from interfering with the incentive to develop and transition to other
more advanced emissions control technologies. As discussed below in
Section III.C, EPA is proposing that any early credits generated by a
manufacturer, beginning as soon as MY 2009, would also be subject to
the five-year credit carry-forward restriction based on the year in
which they are generated. This would limit the effect of the early
credits on the long-term emissions reductions anticipated to result
from the proposed new standards.
    Third, EPA is proposing to allow manufacturers to transfer credits
between the two averaging sets, passenger cars and trucks, within a
manufacturer. For example, credits accrued by over-compliance with a
manufacturer's car fleet average standard could be used to offset
debits accrued due to that manufacturer's not meeting the truck fleet
average standard in a given year. EPA believes that such cross-category
use of credits by a manufacturer would provide important additional
flexibility in the transition to emissions control technology without
affecting overall emission reductions.
    Finally, accumulated credits could be traded to another vehicle
manufacturer. As with intra-company credit use, such inter-company
credit trading would provide flexibility in the transition to emissions
control technology without affecting overall emission reductions.
Trading credits to another vehicle manufacturer would be a
straightforward process between the two manufacturers, but could also
involve third parties that could serve as credit brokers. Brokers would
not own the credits at any time. These sorts of exchanges are typically
allowed under EPA's current emission credit programs, e.g., the Tier 2
light-duty vehicle NO_X fleet average standard and the heavy-
duty engine NO_X fleet average standards, although
manufacturers have seldom made such exchanges. EPA seeks comment on
enhanced reporting requirements or other methods that could help EPA
assess validity of

[[Page 49522]]

credits, especially those obtained from third-party credit brokers
    If a manufacturer had a deficit at the end of a model year--that
is, its fleet average level failed to meet the required fleet average
standard--EPA proposes that the manufacturer could carry that deficit
forward (also referred to credit carry-back) for a total of three model
years after the model year in which that deficit was generated. As
noted above, such a deficit carry-forward could only occur after the
manufacturer applied any banked credits or credits from another
averaging set. If a deficit still remained after the manufacturer had
applied all available credits, and the manufacturer did not obtain
credits elsewhere, the deficit could be carried over for up to three
model years. No deficit could be carried into the fourth model year
after the model year in which the deficit occurred. Any deficit from
the first model year that remained after the third model year would
thus constitute a violation of the condition on the certificate, which
would constitute a violation of the Clean Air Act and would be subject
to enforcement action.
    In the Tier 2 rulemaking proposal, EPA proposed to allow deficits
to be carried forward for one year. In their comments on that proposal,
manufacturers argued persuasively that by the time they can tabulate
their average emissions for a particular model year, the next model
year is likely to be well underway and it is too late to make
calibration, marketing, or production mix changes to adjust that year's
credit generation. Based on those comments, in the Tier 2 final rule
EPA finalized provisions that allowed the deficit to be carried forward
for a total of three years. EPA continues to believe that three years
is an appropriate amount of time that gives the manufacturers adequate
time to respond to a deficit situation but does not create a lengthy
period of prolonged non-compliance with the fleet average
standards.\131\ Subsequent EPA emission control programs that
incorporate ABT provisions (e.g., the Mobile Source Air Toxics rule)
have provided this three-year deficit carry-forward provision for this
reason.\132\
---------------------------------------------------------------------------

    \131\ See 65 FR 6745 (February 10, 2000).
    \132\ See 71 FR 8427 (February 26, 2007).
---------------------------------------------------------------------------

    The proposed averaging, banking, and trading provisions are
generally consistent with those included in the CAFE program, with a
few notable exceptions. As with EPA's proposed approach, CAFE allows
five year carry-forward of credits and three year carry-back. Transfers
of credits across a manufacturer's car and truck averaging sets are
also allowed, but with limits established by EISA on the use of
transferred credits. The amount of transferred credits that can be used
in a year is limited, and transferred credits may not be used to meet
the CAFE minimum domestic passenger car standard. CAFE allows credit
trading, but again, traded credits cannot be used to meet the minimum
domestic passenger car standard. EPA is not proposing these constraints
on the use of transferred credits.
    Additional details regarding the averaging, banking, and trading
provisions and how EPA proposes to implement these provisions can be
found in Section III.E.
5. CO₂ Optional Temporary Lead-time Allowance Alternative Standards
    EPA is proposing a limited and narrowly prescribed option, called
the Temporary Lead-time Allowance Alternative Standards (TLAAS), to
provide additional lead time for a certain subset of manufacturers.
This option is designed to address two different situations where we
project that more lead time is needed, based on the level of emissions
control technology and emissions control performance currently
exhibited by certain vehicles. One situation involves manufacturers who
have traditionally paid CAFE fines instead of complying with the CAFE
fleet average, and as a result at least part of their vehicle
production currently has significantly higher CO₂ and lower
fuel economy levels than the industry average. More lead time is needed
in the program's initial years to upgrade these vehicles to meet the
aggressive CO₂ emissions performance levels required by the
proposal. The other situation involves manufacturers who have a limited
line of vehicles and are unable to take advantage of averaging of
emissions performance across a full line of production. For example,
some smaller volume manufacturers focus on high performance vehicles
with higher CO₂ emissions, above the CO₂
emissions target for that vehicle footprint, but do not have other
types of vehicles in their production mix with which to average. Often,
these manufacturers also pay fines under the CAFE program rather than
meeting the applicable CAFE standard. Because voluntary non-compliance
is impermissible for the GHG standards proposed under the CAA, both of
these types of manufacturers need additional lead time to upgrade
vehicles and meet the proposed standards. EPA is proposing an optional,
temporary alternative standard, which is only slightly less stringent,
and limited to the first four model years (2012--2015) of the National
Program, so that these manufacturers can have sufficient lead time to
meet the tougher MY 2016 GHG standards, while preserving consumer
choice of vehicles during this time.
    In MY 2016, the TLAAS option ends, and all manufacturers,
regardless of size, and domestic sales volume, must comply with the
same CO₂ standards, while under the CAFE program companies
would continue to be allowed to pay civil penalties in lieu of
complying with the CAFE standards. However, because companies must meet
both the CAFE standards and the EPA CO₂ standards, the
National Program will have the practical impact of providing a level
playing field for all companies beginning in MY 2016--a situation which
has never existed under the CAFE program. This option thereby results
in more fuel savings and CO₂ reductions than would be the
case under the CAFE program.
    EPA projects that the environmental impact of the proposed TLAAS
program will be very small. If all companies eligible to use the TLAAS
use it to the maximum extent allowed, total GHG emissions from the
proposal will increase by less than 0.4% over the lifetime of the MY
2012-2016 vehicles. EPA believes the impact will be even smaller, as we
do not expect all of the eligible companies to use this option, and we
do not expect all companies who do use the program will use it to the
maximum extent allowed, as we have included provisions which discourage
companies from using the TLAAS any longer than it is needed.
    EPA has structured the TLAAS option to provide more lead time in
these kinds of situations, but to limit the program so that it would
only be used in situations where these kinds of lead time concerns
arise. Based on historic data on sales, EPA is using a specific
historic U.S. sales volume as the best way to identify the subset of
production that falls into this situation. Under the TLAAS, these
manufacturers would be allowed to produce up to but no more than
100,000 vehicles that would be subject to a somewhat less stringent
CO₂ standard. This 100,000 volume is not an annual limit,
but is an absolute limit for the total number of vehicles which can use
the TLAAS program over the model years 2012-2015. Any additional
production would be subject to the same standards as any other
manufacturer. In addition, EPA is imposing a variety of restrictions on
the use of the TLAAS program, discussed in more detail below, to ensure
that only manufacturers who need more lead-time

[[Page 49523]]

for the kinds of reasons noted above are likely to use the program.
Finally, the program is temporary and expires at the end of MY 2015. A
more complete discussion of the program is provided below. EPA believes
the proposed program reasonably addresses a real world lead time
constraint, and does it in a way that balances the need for more lead
time with the need to minimize any resulting loss in potential
emissions reductions. EPA invites comment as to whether its proposal is
the best way to balance these concerns.
    EPA proposes to establish a TLAAS for a specified subset of
manufacturers. There are two types of companies who would make use of
TLAAS--those manufacturers who have paid CAFE fines in recent years,
and who need additional lead-time to incorporate the needed technology;
and those companies who are not full-line manufacturers, who have a
smaller range of models and vehicle types, who may need additional
lead-time as well. This alternative standard would apply to
manufacturers with total U.S. sales of less than 400,000 vehicles per
year, using 2009 model year final sales numbers to determine
eligibility for these alternative standards. EPA reviewed the sales
volumes of manufacturers over the last few years, and determined that
manufacturers below this level typically fit the characteristics
discussed above, and manufacturers above this level did not. Thus, EPA
chose this level because it functionally identifies the group of
manufacturers described above, recognizing that there is nothing
intrinsic in the sales volume itself that warrants this allowance. EPA
was not able to identify any other objective criteria that would more
appropriately identify the manufacturers and vehicle fleets described above.
    EPA is proposing that manufacturers qualifying for TLAAS would be
allowed to meet slightly less stringent standards for a limited number
of vehicles for model years 2012-2015. Specifically, an eligible
manufacturer could have a total of up to 100,000 units of cars and
trucks combined over model years 2012-2015, and during those model
years those vehicles would be subject to a standard 1.25 times the
standard that would otherwise apply to those vehicles under the primary
program. In other words, the footprint curves upon which the individual
manufacturer standards for the TLAAS fleets are based would be less
stringent by a factor of 1.25 for up to 100,000 of an eligible
manufacturer's vehicles for model years 2012-2015. As noted, this
approach seeks to balance the need to provide additional lead-time
without reducing the environmental benefits of the proposed program.
EPA believes that 100,000 units over four model years achieves an
appropriate balance as the emissions impact is quite small, but does
provide companies with some flexibility during MY 2012-2015. For
example, for a manufacturer producing 400,000 vehicles per year, this
would be a total of up to 100,000 vehicles out of a total production of
up to 1.6 million vehicles over the four year period, or about 6
percent of total production.
    Manufacturers with no U.S. sales in model year 2009 would not
qualify for the TLAAS program. Manufacturers meeting the cut-point of
400,000 for MY 2009 but with U.S. directed production above 400,000 in
any subsequent model years would remain eligible for the TLAAS program.
Also, the total sales number applies at the corporate level, so if a
corporation owns several vehicle brands the aggregate sales for the
corporation would be used. These provisions would help prevent gaming
of the provisions through corporate restructuring. Corporate ownership
or control relationships would be based on determinations made under
CAFE for model year 2009. In other words, corporations grouped together
for purposes of meeting CAFE standards, would be grouped together for
determining whether or not they are eligible under the 400,000 vehicle cut point.
    EPA derived the 100,000 maximum unit set aside number based on a
gradual phase-out schedule shown in Table III.B.5-1, below. However,
individual manufacturers' situations will vary significantly and so EPA
believes a flexible approach that allows manufacturers to use the
allowance as they see fit during these model years would be most
appropriate. As another example, an eligible manufacturer could also
choose to apply the TLAAS program to an average of 25,000 vehicles per
year, over the four-year period. Therefore, EPA is proposing that a
total of 100,000 vehicles of an eligible manufacturer, with any
combination of cars or trucks, could be subject to the alternative
standard over the four year period without restrictions.

                            Table III.B.5-1--TLAAS Example Vehicle Production Volumes
----------------------------------------------------------------------------------------------------------------
             Model year                      2012               2013               2014               2015
----------------------------------------------------------------------------------------------------------------
Sales Volume........................             40,000             30,000             20,000             10,000
----------------------------------------------------------------------------------------------------------------

    The TLAAS vehicles would be separate car and truck fleets for that
model year and would be subject to the less stringent footprint-based
standards of 1.25 times the primary fleet average that would otherwise
apply. The manufacturer would determine what vehicles are assigned to
these separate averaging sets for each model year. EPA is proposing
that credits from the primary fleet average program can be transferred
and used in the TLAAS program. Credits within the TLAAS program may
also be transferred between the TLAAS car and truck averaging sets for
use through 2015 when the TLAAS would end. However, credits generated
under TLAAS would not be allowed to be transferred or traded to the
primary program. Therefore, any unused credits under TLAAS would expire
after model year 2015. EPA believes that this is necessary to limit the
program to situations where it is needed and to prevent the allowance
from being inappropriately transferred to the long-term primary program.
    EPA is concerned that some manufacturers would be able to place
relatively clean vehicles in the TLAAS to maximize TLAAS credits if
credit use was unrestricted. However, any credits generated from the
primary program that are not needed for compliance in the primary
program, should be used to offset the TLAAS vehicles. EPA is thus
proposing to restrict the use of banking and trading between companies
of credits in the primary program in years in which the TLAAS is being
used. For example, manufacturers using the TLAAS in MY 2012 could not
bank credits in the primary program during MY 2012 for use in MY 2013
and later. No such restriction would be in place for years when the
TLAAS is not being used. EPA also believes this provision is necessary
to prevent credits from being earned simply by removing some high-
emitting vehicles from the primary fleet. Absent this restriction,
manufacturers would be able to choose to use the TLAAS for these
vehicles and also be

[[Page 49524]]

able to earn credits under the primary program that could be banked or
traded under the primary program without restriction. EPA is proposing
two additional restrictions regarding the use of the TLAAS by requiring
that for any of the 2012-2015 model years for which an eligible
manufacturer would like to use the TLAAS, the manufacturer must use two
of the available flexibilities in the GHG program first in order to try
and show compliance with the primary standard before accessing the
TLAAS. Specifically, before using the TLAAS the manufacturer must: (1)
use any banked emission credits from a previous model year; and, (2)
use any available credits from the companies' car or truck fleet for
the specific model year (i.e., use credit transfer from cars to trucks
or from trucks to cars, that is, before using the TLAAS for either the
car fleet or the truck fleet, make use of any available credit
transfers first). EPA is requesting comments on all aspects of the
proposed TLAAS program including comments on other provisions that
might be needed to ensure that the TLAAS program is being used as
intended and to ensure no gaming occurs.
    Finally, EPA recognizes that there will be a wide range of
companies within the eligible manufacturers with sales less than
400,000 vehicles in model year 2009. Some of these companies, while
having relatively small U.S. sales volumes, are large global automotive
firms, including companies such as Mercedes and Volkswagen. Other
companies are significantly smaller niche firms, with sales volumes
closer to 10,000 vehicles per year worldwide; an example of this type
of firm is Aston Martin. EPA anticipates that there are a small number
of such smaller volume manufacturers, which have claimed that they may
face greater challenges in meeting the proposed standards due to their
limited product lines across which to average. EPA requests comment on
whether the proposed TLAAS program, as described above, provides
sufficient lead-time for these smaller firms to incorporate the
technology needed to comply with the proposed GHG standards.
6. Proposed Nitrous Oxide and Methane Standards
    In addition to fleet-average CO₂ standards, EPA is
proposing separate per-vehicle standards for nitrous oxide
(N₂O) and methane (CH₄) emissions. Standards are
being proposed that would cap vehicle N₂O and CH₄
emissions at current levels. Our intention is to set emissions
standards that act to cap emissions to ensure that future vehicles do
not increase their N₂O and CH₄ emissions above
levels that would be allowed under the proposal.
    EPA considered an approach of expressing each of these standards in
common terms of CO₂-equivalent emissions and combining them
into a single standard along with CO₂ and HFC emissions.
California's ``Pavley'' program adopted such a CO₂-
equivalent emissions standards approach to GHG emissions in their
program.\133\ However, these pollutants are largely independent of one
another in terms of how they are generated by the vehicle and how they
are tested for during implementation. Potential control technologies
and strategies for each pollutant also differ. Moreover, an approach
that provided for averaging of these pollutants could undermine the
stringency of the CO₂ standards, as at this time we are
proposing standards which ``cap'' N₂O and CH₄
emissions, rather then proposing a level which is either at the
industry fleet-wide average or which would result in reductions from
these pollutants. It is possible that once EPA begins to receive more
detailed information on the N₂O and CH₄
performance of the new vehicle fleet as a result of this proposed rule
(if it were to be finalized as proposed) that for a future action for
model years 2017 and later EPA could consider a CO₂-
equivalent standard which would not result in any increases in GHG
emissions due to the current lack of detailed data on N₂O
and CH₄ emissions performance. In addition, EPA seeks
comment on whether a CO₂-equivalent emissions standard
should be considered for model years 2012 through 2016, and whether
there are advantages or disadvantages to such an approach, including
potential impacts on harmonization with CAFE standards.
---------------------------------------------------------------------------

    \133\ California Environmental Protection Agency Air Resources
Board, Staff Report: Initial Statement of Reasons for Proposed
Rulemaking Public Hearing To Consider Adoption Of Regulations To
Control Greenhouse Gas Emissions From Motor Vehicles, August 6, 2004.
---------------------------------------------------------------------------

    Almost universally across current car and truck designs, both
gasoline- and diesel-fueled, these emissions are relatively low, and
our intent is to not require manufacturers to make technological
improvements in order to reduce N₂O and CH₄ at
this time. However, it is important that future vehicle technologies or
fuels do not result in increases in these emissions, and this is the
intent of the proposed ``cap'' standards.
    EPA requests comments on our approach to regulating N₂O
and CH₄ emissions including the appropriateness of ``cap''
standards as opposed to ``technology-forcing'' standards, the technical
bases for the proposed N₂O and CH₄ standards, the
proposed test procedures, and timing. Specifically, EPA seeks comment
on the appropriateness of the proposed levels of the N₂O and
CH₄ standards to accomplish our stated intent. In addition,
EPA seeks comment on any additional emissions data on N₂O
and CH₄ from current technology vehicles.
a. Nitrous Oxide (N₂O) Exhaust Emission Standard
    N₂O is a global warming gas with a high global warming
potential.\134\ It accounts for about 2.7% of the current greenhouse
gas emissions from cars and light trucks. EPA is proposing a per-
vehicle N₂O emission standard of 0.010 g/mi, measured over
the traditional FTP vehicle laboratory test cycles. The standard would
become effective in model year 2012 for all light-duty cars and trucks.
Averaging between vehicles would not be allowed. The standard is
designed to prevent increases in N₂O emissions from current
levels, i.e. a no-backsliding standard.
---------------------------------------------------------------------------

    \134\ N₂O has a GWP of 310 according to the IPCC
Second Assessment Report (SAR).
---------------------------------------------------------------------------

    N₂O is emitted from gasoline and diesel vehicles mainly
during specific catalyst temperature conditions conducive to
N₂O formation. Specifically, N₂O can be generated
during periods of emission hardware warm-up when rising catalyst
temperatures pass through the temperature window when N₂O
formation potential is possible. For current Tier 2 compatible gasoline
engines with conventional three-way catalyst technology, N₂O
is not generally produced in significant amounts because the time the
catalyst spends at the critical temperatures during warm-up is short.
This is largely due to the need to quickly reach the higher
temperatures necessary for high catalyst efficiency to achieve emission
compliance of criteria pollutants. N₂O is a more significant
concern with diesel vehicles, and potentially future gasoline lean-burn
engines, equipped with advanced catalytic NO_X emissions
control systems. These systems can but need not be designed in a way
that emphasizes efficient NO_X control while allowing the
formation of significant quantities of N₂O. Excess oxygen
present in the exhaust during lean-burn conditions in diesel or lean-
burn gasoline engines equipped with these advanced systems can favor
N₂O formation if catalyst temperatures are not carefully
controlled. Without

[[Page 49525]]

specific attention to controlling N₂O emissions in the
development of such new NO_X control systems, vehicles could
have N₂O emissions many times greater than are emitted by
current gasoline vehicles.
    EPA is proposing an N₂O emission standard that EPA
believes would be met by current-technology gasoline vehicles at
essentially no cost. As noted, N₂O formation in current
catalyst systems occurs, but the emission levels are low, because the
time the catalyst spends at the critical temperatures during warm-up
when N₂O can form is short. At the same time, EPA believes
that the proposed standard would ensure that the design of advanced
NO_X control systems, especially for future diesel and lean-
burn gasoline vehicles, would control N₂O emission levels.
While current NO_X control approaches used on current Tier 2
diesel vehicles do not tend to form N₂O emissions, EPA
believes that the proposed standards would discourage any new emission
control designs that achieve criteria emissions compliance at the cost
of increased N₂O emissions. Thus, the proposed standard
would cap N₂O emission levels, with the expectation that
current gasoline and diesel vehicle control approaches that comply with
the Tier 2 vehicle emission standards for NO_X would not
increase their emission levels, and that the cap would ensure that
future vehicle designs would appropriately control their emissions of
N₂O. The proposed N₂O level is approximately two
times the average N₂O level of current gasoline passenger
cars and light-duty trucks that meet the Tier 2 NO_X
standards.\135\ Manufacturers typically use design targets for
NO_X emission levels of about 50% of the standard, to account
for in-use emissions deterioration and normal testing and production
variability, and manufacturers are expected to utilize a similar
approach for N₂O emission compliance. EPA is not proposing a
more stringent standard for current gasoline and diesel vehicles
because the stringent Tier 2 program and the associated NO_X
fleet average requirement already result in significant N₂O
control, and does not expect current N₂O levels to rise for
these vehicles. EPA requests comment on this technical assessment of
current and potential future N₂O formation in cars and trucks.
---------------------------------------------------------------------------

    \135\ Memo to docket ``Deriving the standard from EPA's MOVES
model emission factors, '' December 2007.
---------------------------------------------------------------------------

    While EPA believes that manufacturers will likely be able to
acquire and install N₂O analytical equipment, the agency
also recognizes that some companies may face challenges. Given the
short lead-time for this rule, EPA proposes that manufacturers be able
to apply for a certificate of conformity with the N₂O
standard for model year 2012 based on a compliance statement based on
good engineering judgment. For 2013 and later model years,
manufacturers would need to submit measurements of N₂O for
compliance purposes.
    Diesel cars and light trucks with advanced emission control
technology are in the early stages of development and
commercialization. As this segment of the vehicle market develops, the
proposed N₂O standard would require manufacturers to
incorporate control strategies that minimize N₂O formation.
Available approaches include using electronic controls to limit
catalyst conditions that might favor N₂O formation and
consider different catalyst formulations. While some of these
approaches may have modest associated costs, EPA believes that they
will be small compared to the overall costs of the advanced
NO_X control technologies already required to meet Tier 2 standards.
    Vehicle emissions regulations do not currently require testing for
N₂O, and most test facilities do not have equipment for its
measurement. Manufacturers without this capability would need to
acquire and install appropriate measurement equipment. However, EPA is
proposing four N₂O measurement methods, all of which are
commercially available today. EPA expects that most manufacturers would
use photo-acoustic measurement equipment, which the Agency estimates
would result in a one-time cost of about $50,000-$60,000 for each test
cell that would need to be upgraded.
    Overall, EPA believes that manufacturers of cars and light trucks,
both gasoline and diesel, would meet the proposed standard without
implementing any significantly new technologies, and there are not
expected to be any significant costs associated with this proposed standard.
b. Methane (CH₄) Exhaust Emission Standard
    CH₄ (or methane) is greenhouse gas with a high global
warming potential.\136\ It accounts for about 0.2% of the greenhouse
gases from cars and light trucks.
---------------------------------------------------------------------------

    \136\ CH₄ has a GWP of 21 according to the IPCC
Second Assessment Report (SAR).
---------------------------------------------------------------------------

    EPA is proposing a CH₄ emission standard of 0.030 g/mi
as measured on the FTP, to apply beginning with model year 2012 for
both cars and trucks. EPA believes that this level for the standard
would be met by current gasoline and diesel vehicles, and would prevent
large increases in future CH₄ emissions in the event that
alternative fueled vehicles with high methane emissions, like some past
dedicated compressed natural gas (CNG) vehicles, become a significant
part of the vehicle fleet. Currently EPA does not have separate
CH₄ standards because unlike other hydrocarbons it does not
contribute significantly to ozone formation,\137\ However
CH₄ emissions levels in the gasoline and diesel car and
light truck fleet have nevertheless generally been controlled by the
Tier 2 non-methane organic gases (NMOG) emission standards. However,
without an emission standard for CH₄, future emission levels
of CH₄ cannot be guaranteed to remain at current levels as
vehicle technologies and fuels evolve.
---------------------------------------------------------------------------

    \137\ But see Ford Motor Co. v. EPA, 604 F. 2d 685 (D.C. Cir. 1979)
(permissible for EPA to regulate CH₄ under CAA section 202 (b)).
---------------------------------------------------------------------------

    The proposed standard would cap CH₄ emission levels,
with the expectation that current gasoline vehicles meeting the Tier 2
emission standards would not increase their levels, and that it would
ensure that emissions would be addressed if in the future there are
increases in the use of natural gas or any other alternative fuel. The
level of the standard would generally be achievable through normal
emission control methods already required to meet Tier 2 program
emission standards for NMOG and EPA is therefore not attributing any
cost to this part of this proposal. Since CH₄ is produced in
gasoline and diesel engines similar to other hydrocarbon components,
controls targeted at reducing overall NMOG levels generally also work
at reducing CH₄ emissions. Therefore, for gasoline and
diesel vehicles, the Tier 2 NMOG standards will generally prevent
increases in CH₄ emissions levels from today. CH₄
from Tier 2 light-duty vehicles is relatively low compared to other
GHGs largely due to the high effectiveness of previous National Low
Emission Vehicle (NLEV) and current Tier 2 programs in controlling
overall HC emissions.
    The level of the proposed standard is approximately two times the
average Tier 2 gasoline passenger cars and light-duty trucks
level.\138\ As with N₂O, this proposed level recognizes that
manufacturers typically set emission design targets at about 50% of the
standard. Thus, EPA believes the proposed standard would be met by

[[Page 49526]]

current gasoline vehicles. Similarly, since current diesel vehicles
generally have even lower CH₄ emissions than gasoline
vehicles, EPA believes that diesels would also meet the proposed
standard. However, EPA also believes that to set a CH₄
emission standard more stringent than the proposed standard could
effectively make the Tier 2 NMOG standard more stringent.
---------------------------------------------------------------------------

    \138\ Memo to docket ``Deriving the standard from EPA's MOVES
model emission factors, '' December 2007.
---------------------------------------------------------------------------

    In recent model years, a small number of cars and light trucks were
sold that were designed for dedicated use of compressed natural gas
(CNG) that met Tier 2 emission standards. While emission control
designs on these recent dedicated CNG-fueled vehicles demonstrate
CH₄ control as effective as gasoline or diesel equivalent
vehicles, CNG-fueled vehicles have historically produced significantly
higher CH₄ emissions than gasoline or diesel vehicles. This
is because their CNG fuel is essentially methane and any unburned fuel
that escapes combustion and not oxidized by the catalyst is emitted as
methane. However, even if these vehicles meet the Tier 2 NMOG standard
and appear to have effective CH₄ control by nature of the
NMOG controls, Tier 2 standards do not require CH₄ control.
While the proposed CH₄ cap standard should not require any
different emission control designs beyond what is already required to
meet Tier 2 NMOG standards on a dedicated CNG vehicle, the cap will
ensure that systems maintain the current level of CH₄
control. EPA is not proposing more stringent CH₄ standards
because the same controls that are used to meet Tier 2 NMOG standards
should result in effective CH₄ control. Increased
CH₄ stringency beyond proposed levels could inadvertently
result in increased Tier 2 NMOG stringency absent an emission control
technology unique to CH₄. Since CH₄ is already
measured under the current Tier 2 regulations (so that it may be
subtracted to calculate non-methane hydrocarbons), the proposed
standard would not result in additional testing costs. EPA requests
comment on whether the proposed cap standard would result in any
significant technological challenges for makers of CNG vehicles.
7. Small Entity Deferment
    EPA is proposing to defer setting GHG emissions standards for small
entities meeting the Small Business Administration (SBA) criteria of a
small business as described in 13 CFR 121.201. EPA would instead
consider appropriate GHG standards for these entities as part of a
future regulatory action. This includes small entities in three
distinct categories of businesses for light-duty vehicles: small volume
manufacturers, independent commercial importers (ICIs), and alternative
fuel vehicle converters. EPA has identified about 13 entities that fit
the Small Business Administration (SBA) criterion of a small business.
EPA estimates there are 2 small volume manufacturers, 8 ICIs, and 3
alternative fuel vehicle converters currently in the light-duty vehicle
market. EPA estimates that these small entities comprise less than 0.1
percent of the total light-duty vehicle sales in the U.S., and
therefore the proposed deferment will have a negligible impact on the
GHG emissions reductions from the proposed standards. Further detail is
provided in Section III.I.3, below.
    To ensure that EPA is aware of which companies would be deferred,
EPA is proposing that such entities submit a declaration to EPA
containing a detailed written description of how that manufacturer
qualifies as a small entity under the provisions of 13 CFR 121.201.
Because such entities are not automatically exempted from other EPA
regulations for light-duty vehicles and light-duty trucks, absent such
a declaration, EPA would assume that the entity was subject to the
greenhouse gas control requirements in this GHG proposal. The
declaration would need to be submitted at time of vehicle emissions
certification under the EPA Tier 2 program. Small entities are
currently covered by a number of EPA motor vehicle emission
regulations, and they routinely submit information and data on an
annual basis as part of their compliance responsibilities. EPA expects
that the additional paperwork burden associated with completing and
submitting a small entity declaration to gain deferral from the
proposed GHG standards would be negligible and easily done in the
context of other routine submittals to EPA. However, EPA has accounted
for this cost with a nominal estimate included in the Information
Collection Request completed under the Paperwork Reduction Act.
Additional information can be found in the Paperwork Reduction Act
discussion in Section III.I.2.

C. Additional Credit Opportunities for CO₂ Fleet Average Program

    The standards being proposed represent a significant multi-year
challenge for manufacturers, especially in the early years of the
program. Section III.B.4 described EPA proposals for how manufacturers
could generate credits by achieving fleet average CO₂
emissions below the fleet average standard, and also how manufacturers
could use credits to comply with standards. As described in Section
III.B.4, credits could be carried forward five years, carried back
three years, transferred between vehicle categories, and traded between
manufacturers. The credits provisions proposed below would provide
manufacturers with additional ways to earn credits starting in MY 2012.
EPA is also proposing early credits provisions for the 2009-2011 model
years, as described below in Section III.C.5.
    The provisions proposed below would provide additional flexibility,
especially in the early years of the program. This flexibility helps to
address issues of lead-time or technical feasibility for various
manufacturers and in several cases provides an incentive for promotion
of technology pathways that warrant further development, whether or not
they are an important or central technology on which critical features
of this program are premised. EPA is proposing a variety of credit
opportunities because manufacturers are not likely to be in a position
to use every credit provision. EPA expects that manufacturers are
likely to select the credit opportunities that best fit their future
plans. EPA believes it is critical that manufacturers have options to
ease the transition to the final MY 2016 standards. At the same time,
EPA believes these credit programs must be designed in a way to ensure
that they achieve emission reductions that achieve real-world
reductions over the full useful life of the vehicle (or, in the case of
FFV credits and Advanced Technology credits, to incentivize the
introduction of those vehicle technologies) and are verifiable. In
addition, EPA wants to ensure these credit programs do not provide an
opportunity for manufacturers to earn ``windfall'' credits. EPA seeks
comments on how to best ensure these objectives are achieved in the
design of the credit programs. EPA requests comment on all aspects of
these proposed credits provisions.
1. Air Conditioning Related Credits
    EPA proposes that manufacturers be able to generate and use credits
for improved air conditioner (A/C) systems in complying with the
CO₂ fleetwide average standards described above. EPA expects
that most manufacturers will choose to utilize the A/C provisions as
part of its compliance demonstration (and for this reason cost of
compliance with A/C related emission reductions are assumed in the cost
analysis). The A/C provisions are structured as credits, unlike the
CO₂ standards for which manufacturers will demonstrate

[[Page 49527]]

compliance using 2-cycle tests (see Sections III.B and III.E.). Those
tests do not measure either A/C leakage or tailpipe CO₂
emissions attributable to A/C load (see Section III.C.1.b below
describing proposed alternative test procedures for assessing tailpipe
CO₂ emission attributable to A/C engine load). Thus, it is a
manufacturer's option to include A/C GHG emission reductions as an
aspect of its compliance demonstration. Since this is an elective
alternative, EPA is referring to the A/C part of the proposal as a credit.
    EPA estimates that direct A/C GHG emissions--emissions due to the
leakage of the hydrofluorocarbon refrigerant in common use today--
account for 4.3% of CO₂-equivalent GHGs from light-duty cars
and trucks. This includes the direct leakage of refrigerant as well as
the subsequent leakage associated with maintenance and servicing, and
with disposal at the end of the vehicle's life. The emissions that are
impacted by leakage reductions are the direct leakage and the
maintenance and servicing. Together these are equivalent to
CO₂ emissions of approximately 13.6 g/mi per vehicle (this
is 14.9 g/mi if end of life emissions are also included). EPA also
estimates that indirect GHG emissions (additional CO₂
emitted due to the load of the A/C system on the engine) account for
another 3.9% of light-duty GHGs.\139\ This is equivalent to
CO₂ emissions of approximately 14.2 g/mi per vehicle. The
derivation of these figures can be found in the EPA DRIA.
---------------------------------------------------------------------------

    \139\ See Chapter 2, section 2.2.1.2 of the DRIA.
---------------------------------------------------------------------------

    EPA believes that it is important to address A/C direct and
indirect emissions because the technologies that manufacturers will
employ to reduce vehicle exhaust CO₂ will have little or no
impact on A/C related emissions. Without addressing A/C-related
emissions, as vehicles become more efficient, the A/C related
contribution will become a much larger portion of the overall vehicle
GHG emissions.
    Over 95% of the new cars and light trucks in the United States are
equipped with A/C systems and, as noted, there are two mechanisms by
which A/C systems contribute to the emissions of greenhouse gases:
through leakage of refrigerant into the atmosphere and through the
consumption of fuel to provide power to the A/C system. With leakage,
it is the high global warming potential (GWP) of the current automotive
refrigerant--R134a, with a GWP of 1430--that results in the
CO₂-equivalent impact of 13.6 g/mi.\140\ Due to the high GWP
of this HFC, a small leakage of the refrigerant has a much greater
global warming impact than a similar amount of emissions of
CO₂ or other mobile source GHGs. Manufacturers can choose to
reduce A/C leakage emissions by using leak-tight components. Also,
manufacturers can largely eliminate the global warming impact of
leakage emissions by adopting systems that use an alternative, low-GWP
refrigerant.\141\ The A/C system also contributes to increased
CO₂ emissions through the additional work required to
operate the compressor, fans, and blowers. This additional work
typically is provided through the engine's crankshaft, and delivered
via belt drive to the alternator (which provides electric energy for
powering the fans and blowers) and A/C compressor (which pressurizes
the refrigerant during A/C operation). The additional fuel used to
supply the power through the crankshaft necessary to operate the A/C
system is converted into CO₂ by the engine during
combustion. This incremental CO₂ produced from A/C operation
can thus be reduced by increasing the overall efficiency of the
vehicle's A/C system, which in turn will reduce the additional load on
the engine from A/C operation.\142\
---------------------------------------------------------------------------

    \140\ The global warming potentials (GWP) used in the NPRM
analysis are consistent with Intergovernmental Panel on Climate
Change (IPCC) Fourth Assessment Report (AR4). At this time, the IPCC
Second Assessment Report (SAR) global warming potential values have
been agreed upon as the official U.S. framework for addressing
climate change. The IPCC SAR GWP values are used in the official
U.S. greenhouse gas inventory submission to the climate change
framework. When inventories are recalculated for the final rule,
changes in GWP used may lead to adjustments.
    \141\ Refrigerant emissions during maintenance and at the end of
the vehicle's life (as well as emissions during the initial charging
of the system with refrigerant) are also addressed by the CAA Title
VI stratospheric ozone program, as described below.
    \142\ We will not be addressing changes to the weight of the A/C
system, since the issue of CO₂ emissions from the fuel
consumption of normal (non-A/C) operation, including basic vehicle
weight, is inherently addressed with the primary CO₂
standards (See III.B above).
---------------------------------------------------------------------------

    Manufacturers can make very feasible improvements to their A/C
systems to address A/C system leakage and efficiency. EPA proposes two
separate credit approaches to address leakage reductions and efficiency
improvements independently. A proposed leakage reduction credit would
take into account the various technologies that could be used to reduce
the GHG impact of refrigerant leakage, including the use of an
alternative refrigerant with a lower GWP. A proposed efficiency
improvement credit would account for the various types of hardware and
control of that hardware available to increase the A/C system
efficiency. Manufacturers would be required to attest the durability of
the leakage reduction and the efficiency improvement technologies over
the full useful life of the vehicle.
    EPA believes that both reducing A/C system leakage and increasing
efficiency are highly cost-effective and technologically feasible. EPA
expects most manufacturers will choose to use these A/C credit
provisions, although some may not find it necessary to do so.
a. A/C Leakage Credits
    The refrigerant used in vehicle A/C systems can get into the
atmosphere by many different means. These refrigerant emissions occur
from the slow leakage over time that all closed high pressure systems
will experience. Refrigerant loss occurs from permeation through hoses
and leakage at connectors and other parts where the containment of the
system is compromised. The rate of leakage can increase due to
deterioration of parts and connections as well. In addition, there are
emissions that occur during accidents and maintenance and servicing
events. Finally, there are end-of-life emissions if, at the time of
vehicle scrappage, refrigerant is not fully recovered.
    Because the process of refrigerant leakage has similar root causes
as those that cause fuel evaporative emissions from the fuel system,
some of the control technologies are similar (including hose materials
and connections). There are however, some fundamental differences
between the systems that require a different approach. The most notable
difference is that A/C systems are completely closed systems, whereas
the fuel system is not. Fuel systems are meant to be refilled as liquid
fuel is consumed by the engine, while the A/C system ideally should
never require ``recharging'' of the contained refrigerant. Thus it is
critical that the A/C system leakages be kept to an absolute minimum.
These emissions are typically too low to accurately measure in most
current SHED chambers designed for fuel evaporative emissions
measurement, especially for systems that are new or early in life.
Therefore, if leakage emissions were to be measured directly, new
measurement facilities would need to be built by the OEM manufacturers
and very accurate new test procedures would need to be developed.
Especially because there are indications that much of the industry is
moving toward alternative refrigerants (post-2016 for most
manufacturers), EPA is not proposing such a direct measurement approach
to addressing refrigerant leakage.

[[Page 49528]]

    Instead, EPA proposes that manufacturers demonstrate improvements
in their A/C system designs and components through a design-based
method. Manufacturers implementing systems expected to result in
reduced refrigerant leakage would be eligible for credits that could
then be used to meet their CO₂ emission compliance
requirements. The proposed ``A/C Leakage Credit'' provisions would
generally assign larger credits to system designs that are expected to
result in greater leakage reduction. In addition, EPA proposes that
proportionately larger A/C Leakage Credits be available to
manufacturers that substitute a lower-GWP refrigerant for the current
R134a refrigerant.
    Our proposed method for calculating A/C Leakage Credits is based
closely on an industry-consensus leakage scoring method, described
below. This leakage scoring method is correlated to experimentally-
measured leakage rates from a number of vehicles using the different
available A/C components. Under the proposed approach, manufacturers
would choose from a menu of A/C equipment and components used in their
vehicles in order to establish leakage scores which would characterize
their A/C system leakage performance. The leakage score can be compared
to expected fleetwide leakage rates in order to quantify improvements
for a given A/C system. Credits would be generated from leakage
reduction improvements that exceeded average fleetwide leakage rates.
    EPA believes that the design-based approach would result in
estimates of likely leakage emissions reductions that would be
comparable to those that would eventually result from performance-based
testing. At the same time, comments are encouraged on all developments
that may lead to a robust, practical, performance-based test for
measuring A/C refrigerant leakage emissions.
    The cooperative industry and government Improved Mobile Air
Conditioning (IMAC) program \143\ has demonstrated that new-vehicle
leakage emissions can be reduced by 50%. This program has shown that
this level of improvement can be accomplished by reducing the number
and improving the quality of the components, fittings, seals, and hoses
of the A/C system. All of these technologies are already in commercial
use and exist on some of today's systems.
---------------------------------------------------------------------------

    \143\ Team 1-Refrigerant Leakage Reduction: Final Report to
Sponsors, SAE, 2007.
---------------------------------------------------------------------------

    EPA is proposing that a manufacturer wishing to earn A/C Leakage
Credits would compare the components of its A/C system with a set of
leakage-reduction technologies and actions that is based closely on
that being developed through IMAC and the Society of Automotive
Engineers (as SAE Surface Vehicle Standard J2727, August 2008 version).
The J2727 approach is developed from laboratory testing of a variety of
A/C related components, and EPA believes that the J2727 leakage scoring
system generally represents a reasonable correlation with average real-
world leakage in new vehicles. Like the IMAC approach, our proposed
credit approach would associate each component with a specific leakage
rate in grams per year identical to the values in J2727. A manufacturer
choosing to claim Leakage Credits would sum the leakage values for an
A/C system for a total A/C leakage score. EPA is proposing a formula
for converting the grams-per-year leakage score to a grams-per-mile
CO₂eq value, taking vehicle miles traveled (VMT) and the GWP
of the refrigerant into account. This formula is:

Credit = (MaxCredit) * [1 - (LeakScore/AvgImpact) * (GWPRefrigerant/1430)]

Where:

MaxCredit is 12.6 and 15.7 g/mi CO₂eq for cars and trucks
respectively. These become 13.8 and 17.2 for cars and trucks if
alternative refrigerants are used since they get additional credits
for end-of-life emissions reductions.
LeakScore is the leakage score of the A/C system as measured
according to methods similar to the J2727 procedure in units of g/
yr. The minimum score which is deemed feasible is fixed at 8.3 and
10.4 g/yr for cars and trucks respectively.
AvgImpact is the average impact of A/C leakage, which is 16.6 and
20.7 g/yr for cars and trucks respectively.
GWPRefrigerant is the global warming potential for direct radiative
forcing of the refrigerant as defined by EPA (or IPCC).
All of the parameters and limits of the equation are derived in the EPA DRIA.

    For systems using the current refrigerant, EPA proposes that these
emission rates could at most be feasibly reduced by half, based on the
conclusions of the IMAC study, and consideration of emission over the full
life of the vehicle. (This latter point is discussed further in the DRIA.)
    As discussed above, EPA recognizes that substituting an alternative
refrigerant (one with a significantly lower global warming potential,
GWP), would potentially be a very effective way to reduce the impact of
all forms of refrigerant emissions, including maintenance, accidents,
and vehicle scrappage. To address future GHG regulations in Europe and
California, systems using alternative refrigerants--including
HFO1234yf, with a GWP of 4--are under serious development and have been
demonstrated in prototypes by A/C component suppliers. These
alternative refrigerants have remaining cost, safety and feasibility
hurdles for commercial applications.\144\ However, the European Union
has enacted regulations phasing in alternative refrigerants with GWP
less than 150 starting in 2010, and the State of California proposed
providing credits for alternative refrigerant use in its GHG rule.
---------------------------------------------------------------------------

    \144\ Although see 71 FR 55140 (Sept. 21, 2006) (proposal
pursuant to section 612 of the CAA finding CO₂ and HFC
152a as acceptable refrigerant substitutes as replacements for CFC-
12 in motor vehicle air conditioning systems, and stating (at 55142)
that ``data [hellip] indicate that use of CO₂ and HFC
152a with risk mitigation technologies does not pose greater risks
compared to other substitutes'').
---------------------------------------------------------------------------

    Within the timeframe of 2012-2016, EPA is not expecting the use of
low-GWP refrigerants to be widespread. However, EPA believes that these
developments are promising, and have included in our proposed A/C
Leakage Credit system provisions to account for the effective
refrigerant reductions that could be expected from refrigerant
substitution. The quantity of A/C Leakage Credits that would be
available would be a function of the GWP of the alternative
refrigerant, with the largest credits being available for refrigerants
approaching a GWP of zero.\145\ For a hypothetical alternative
refrigerant with a GWP of 1, effectively eliminating leakage as a GHG
concern, our proposed credit calculation method could result in maximum
credits equal total average emissions, or credits of 13.4 and 17.8 g/mi
CO₂eq for cars and trucks, respectively. This option is also
captured in the equation above.
---------------------------------------------------------------------------

    \145\ For example, the GWP for R152a is 120, the GWP of HFO-
1234yf is 4, and the GWP of CO₂ as a refrigerant is 1.
---------------------------------------------------------------------------

    It is possible that alternative refrigerants could, without
compensating action by the manufacturer, reduce the efficiency of the
A/C system (see discussion of the A/C Efficiency Credit below.)
However, EPA believes that manufacturers will have substantial
incentives to design their systems to maintain the efficiency of the A/
C system, therefore EPA is not accounting for any potential efficiency
degradation.
    EPA requests comment on all aspects of our proposed A/C Leakage
Credit system.

[[Page 49529]]

b. A/C Efficiency Credits
    EPA is proposing that manufacturers that make improvements in their
A/C systems to increase efficiency and thus reduce CO₂
emissions due to A/C system operation be eligible for A/C Efficiency
Credits. As with A/C Leakage Credits, manufacturers could apply A/C
Efficiency Credits toward compliance with their overall CO₂ standards.
    As mentioned above, EPA estimates that the CO₂ emissions
due to A/C related loads on the engine account for approximately 3.9%
of total greenhouse gas emissions from passenger vehicles in the United
States. Usage of A/C systems is inherently higher in hotter and more
humid months and climates; however, vehicle owners may use their A/C
systems all year round in all parts of the nation. For example, people
commonly use A/C systems to cool and dehumidify the cabin air for
passenger comfort on hot humid days, but they also use the systems to
de-humidify cabin air to assist in defogging/de-icing the front
windshield and side glass in cooler weather conditions for improved
visibility. A more detailed discussion of seasonal and geographical A/C
usage rates can be found in the DRIA.
    Most of the additional load on the engine from A/C system operation
comes from the compressor, which pumps the refrigerant around the
system loop. Significant additional load on the engine may also come
from electric or hydraulic fans, which are used to move air across the
condenser, and from the electric blower, which is used to move air
across the evaporator and into the cabin. Manufacturers have several
currently-existing technology options for improving efficiency,
including more efficient compressors, fans, and motors, and systems
controls that avoid over-chilling the air (and subsequently re-heating
it to provide the desired air temperature with an associated loss of
efficiency). For vehicles equipped with automatic climate-control
systems, real-time adjustment of several aspects of the overall system
(such as engaging the full capacity of the cooling system only when it
is needed, and maximizing the use of recirculated air) can result in
improved efficiency. Table III.C.1-1 below lists some of these
technologies and their respective efficiency improvements.
    As with the A/C Leakage Credit program, EPA is interested in
performance-based standards (or credits) based on measurement
procedures whenever possible. While design-based assessments of
expected emissions can be a reasonably robust way of quantifying
emission improvements, these approaches have inherent shortcomings, as
discussed for the case of A/C leakage above. Design-based approaches
depend on the quality of the data from which they are calibrated, and
it is possible that apparently proper equipment may function less
effectively than expected. Therefore, while the proposal uses a design-
based menu approach to quantify improvements in A/C efficiency, it is
also proposed to begin requiring manufacturers to confirm that
technologies applying for Efficiency Credits are measurably improving
system efficiency.
    EPA believes that there is a more critical need for a test
procedure to quantify A/C Efficiency Credits than for Leakage Credits,
for two reasons. First, the efficiency gains for various technologies
are more difficult to quantify using a design-based program (like the
SAEJ2727-based procedure used to generate Leakage Credits). Second,
while leakage may disappear as a significant source of GHG emissions if
a shift toward alternate refrigerants develops, no parallel factor
exists in the case of efficiency improvements. EPA is thus proposing to
phase-in a performance-based test procedure over time beginning in
2014, as discussed below. In the interim, EPA proposes a design-based
``menu'' approach for estimating efficiency improvements and, thus,
quantifying A/C Efficiency Credits.
    For model years 2012 and 2013, EPA proposes that a manufacturer
wishing to generate A/C Efficiency Credits for a group of its vehicles
with similar A/C systems would compare several of its vehicle A/C-
related components and systems with a ``menu'' of efficiency-related
technology improvements (see Table III.C.1-1 below). Based on the
technologies the manufacturer chooses, an A/C Efficiency Credit value
would be established. This design-based approach would recognize the
relationships and synergies among efficiency-related technologies.
Manufacturers could receive credit based on the technologies they chose
to incorporate in their A/C systems and the associated credit value for
each technology. The total A/C Efficiency Credit would be the total of
these values, up to a maximum feasible credit of 5.7 g/mi
CO₂eq. This would be the maximum improvement from current
average efficiencies for A/C systems (see the DRIA for a full
discussion of our derivation of the proposed reductions and credit
values for individual technologies and for the maximum total credit
available). Although the total of the individual technology credit
values may exceed 5.7 g/mi CO₂eq, synergies among the
technologies mean that the values are not additive, and thus A/C
Efficiency credit could not exceed 5.7 g/mi CO₂eq.
    The EPA requests comment on adjusting the A/C efficiency credit to
account for potential decreases (or increases) in efficiency when using
an alternative refrigerant by using the change in the coefficient of
performance. The effects may include the impact of a secondary loop
system (including the incremental effect on tailpipe CO₂
emissions that the added weight of such a system would incur).

    Table III.C.1-1 Efficiency-Improving A/C Technologies and Credits
------------------------------------------------------------------------
                                        Estimated
                                     reduction in A/C    A/C Efficiency
      Technology description          CO2 emissions    credit (g/mi CO2)
                                        (percent)
------------------------------------------------------------------------
Reduced reheat, with externally-                   30                1.7
 controlled, variable-displacement
 compressor.......................
Reduced reheat, with externally-                   20                1.1
 controlled, fixed-displacement or
 pneumatic variable-displacement
 compressor.......................
Default to recirculated air                        30                1.7
 whenever ambient temperature is
 greater than 75 [deg]F...........
Blower motor and cooling fan                       15                0.9
 controls which limit waste energy
 (e.g. pulse width modulated power
 controller)......................
Electronic expansion valve........                 20                1.1
Improved evaporators and                           20                1.1
 condensers (with system analysis
 on each component indicating a
 COP improvement greater than 10%,
 when compared to previous design)
Oil Separator.....................                 10                0.6
------------------------------------------------------------------------

[[Page 49530]]

    For model years 2014 and later, EPA proposes that manufacturers
seeking to generate A/C Efficiency Credits would need to use a specific
performance test to confirm that the design changes were also improving
A/C efficiency. Manufacturers would need to perform an A/C
CO₂ Idle Test for each A/C system (family) for which it
desired to generate Efficiency Credits. Manufacturers would need to
demonstrate at least a 30% improvement over current average efficiency
levels to qualify for credits. Upon qualifying on the Idle Test, the
manufacturer would be eligible to use the menu approach above to
quantify the credits it would earn.
    The proposed A/C CO₂ Idle Test procedure, which EPA has
designed specifically to measure A/C CO₂ emissions, would be
performed while the vehicle engine is at idle. This proposed laboratory
idle test would be similar to the idle carbon monoxide (CO) test that
was once a part of EPA vehicle certification. The test would determine
the additional CO₂ generated at idle when the A/C system is
operated. The A/C CO₂ Idle Test would be run with and
without the A/C system cooling the interior cabin while the vehicle's
engine is operating at idle and with the system under complete control
of the engine and climate control system
    The proposed A/C CO₂ Idle Test is similar to that
proposed in April 2009 for the Mandatory GHG Reporting Rule, with
several improvements. These improvements include tighter restrictions
on test cell temperatures and humidity levels in order to more closely
control the loads from operation of the A/C system. EPA also made
additional refinements to the required in-vehicle blower fan settings
for manually controlled systems to more closely represent ``real
world'' usage patterns. These details can be found in the DRIA and the
regulations.
    The design of the A/C CO₂ Idle Test represents a
balancing of the need for performance tests whenever possible to ensure
the most accurate quantification of efficiency improvements, with
practical concerns for testing burden and facility requirements. EPA
believes that the proposed Idle Test adds to the robust quantification
of A/C credits that will result in real-world efficiency improvements
and reductions in A/C-related CO₂ emissions. EPA is
proposing that the Idle Test be required in order to qualify for A/C
Efficiency Credits beginning in 2014 to allow sufficient time for
manufacturers to make the necessary facilities improvements and to
establish a comfort level with the test.
    EPA also considered a more comprehensive testing approach to
quantifying A/C CO₂ emissions that could be somewhat more
technically robust, but would require more test time and test facility
improvements for many manufacturers. This approach would be to adapt an
existing test procedure, the Supplemental Federal Test Procedure (SFTP)
for A/C operation, called the SC03, in specific ways for it to function
as a tool to evaluate A/C CO₂ emissions. The potential test
method is described in some detail here, and EPA encourages comment on
how this type of test might or might not accomplish the goals of robust
performance-based testing and reasonable test burdens.
    EPA designed the SC03 test to measure criteria pollutants under
severe air conditioning conditions not represented in the FTP and
Highway Fuel Economy Tests. EPA did not specifically design the SC03 to
measure incremental reductions in CO₂ emissions from more
efficient A/C technologies. For example, due to the severity of the
SC03 test environmental conditions and the relatively short duration of
the SC03 cycle, it is difficult for the A/C system to achieve a
stabilized interior cabin condition that reflects incremental
improvements. Many potential efficiency improvements in the A/C
components and controls (i.e., automatic recirculation and heat
exchanger fan control) are specifically measured only during stabilized
conditions, and therefore become difficult or impossible to measure and
quantify during this test. In addition, SC03 testing is also somewhat
constrained and costly due to limited number of test facilities
currently capable of performing testing under the required
environmental conditions.
    One value of using the SC03 as the basis for a new test to quantify
A/C-related efficiency improvements would be the significant degree of
control of test cell ambient conditions. The load placed on an A/C
system, and thus the incremental CO₂ emissions, are highly
dependent on the ambient conditions in the test cell, especially
temperature and humidity, as well as simulated solar load. Thus, as
with the proposed Idle Test, a new SC03-based test would need to
accurately and reliably control these conditions. (This contrasts with
FTP testing for criteria pollutants, which does not require precise
control of cell conditions because test results are generally much less
sensitive to changes in cell temperature or humidity).
    However, for the purpose of quantifying A/C system efficiency
improvements, EPA believes a test cell temperature less severe than the
95[deg]F required by the SC03 would be appropriate. A cell temperature
of 85[deg]F would better align the initial cooling phase (``pull-
down'') as well as the stabilized phase of A/C operation with real-
world driving conditions.
    Another value of an SC03-based test would be the opportunity to
create operating conditions for vehicle A/C systems that in some ways
would better simulate ``real world'' operation than either the proposed
Idle Test or the current SC03. The SC03 test cycle, roughly 10 minutes
in length, has a similar average speed, maximum speed, and percentage
of time at idle as the FTP. However, since the SC03 test cycle was
designed principally to measure criteria pollutants under maximum A/C
load conditions, it is not long enough to allow temperatures in the
passenger cabin to consistently stabilize. EPA believes that once the
pull-down phase has occurred and cabin temperatures have dropped
dramatically to a suitable interior comfort level, additional test
cycle time would be needed to measure how efficiently the A/C system
operates under stabilized conditions.
    To capture the A/C operation during stabilized operation, EPA would
consider adding two phases to the SC03 test of roughly 10 minutes each.
Each additional phase would simply be repeats of the SC03 drive cycle,
with two exceptions. During the second phase, the A/C system would now
be operating at cabin temperature at or approaching a stabilized
condition. During the third phase, the A/C system would be turned off.
The purpose of the third phase would be to establish the base
CO₂ emissions with no A/C loads on the engine, which would
provide a baseline for the incremental CO₂ due to A/C use.
EPA would likely weight the CO₂ g/mi results for the first
and second phases of the test as follows: 50% for phase 1, and 50% for
phase 2. From this average CO₂ the methodology would
subtract the CO₂ result from phase 3, yielding an
incremental CO₂ (in g/mi) due to A/C use.
    EPA expects to continue working with industry, the California Air
Resources Board, and other stakeholders to move toward increasingly
robust performance tests for A/C and may include such changes in this
final rule. EPA requests comment on all aspects of our proposed A/C
Efficiency Credits program.
c. Interaction With Title VI Refrigerant Regulations
    Title VI of the Clean Air Act deals with the protection of
stratospheric ozone. Section 608 establishes a comprehensive program to
limit emissions of certain ozone-depleting

[[Page 49531]]

substances (ODS). The rules promulgated under section 608 regulate the
use and disposal of such substances during the service, repair or
disposal of appliances and industrial process refrigeration. In
addition, section 608 and the regulations promulgated under it,
prohibit knowingly venting or releasing ODS during the course of
maintaining, servicing, repairing or disposing of an appliance or
industrial process refrigeration equipment. Section 609 governs the
servicing of motor vehicle air conditioners (MVACs). The regulations
promulgated under section 609 (40 CFR part 82, subpart B) establish
standards and requirements regarding the servicing of MVACs. These
regulations include establishing standards for equipment that recovers
and recycles or only recovers refrigerant (CFC-12, HFC 134a, and for
blends only recovers) from MVACs; requiring technician training and
certification by an EPA-approved organization; establishing
recordkeeping requirements; imposing sales restrictions; and
prohibiting the venting of refrigerants. Section 612 requires EPA to
review substitutes for class I and class II ozone depleting substances
and to consider whether such substitutes will cause an adverse effect
to human health or the environment as compared with other substitutes
that are currently or potentially available. EPA promulgated
regulations for this program in 1992 and those regulations are located
at 40 CFR part 82, subpart G. When reviewing substitutes, in addition
to finding them acceptable or unacceptable, EPA may also find them
acceptable so long as the user meets certain use conditions. For
example, all motor vehicle air conditioning system must have unique
fittings and a uniquely colored label for the refrigerant being used in
the system.
    EPA views this proposed rule as complementing these Title VI
programs, and not conflicting with them. To the extent that
manufacturers choose to reduce refrigerant leakage in order to earn A/C
Leakage Credits, this would dovetail with the Title VI section 609
standards which apply to maintenance events, and to end-of-vehicle life
disposal. In fact, as noted, a benefit of the proposed A/C credit
provisions is that there should be fewer and less impactive maintenance
events for MVACs, since there will be less leakage. In addition, the
credit provisions would not conflict (or overlap) with the Title VI
section 609 standards. EPA also believes the menu of leak control
technologies proposed today would complement the section 612
requirements, because these control technologies would help ensure that
R134a (or other refrigerants) would be used in a manner that further
minimizes potential adverse effects on human health and the environment.
2. Flex Fuel and Alternative Fuel Vehicle Credits
    As described in this section, EPA is proposing credits for
flexible-fuel vehicles (FFVs) and alternative fuel vehicles starting in
the 2012 model year. FFVs are vehicles that can run both on an
alternative fuel and conventional fuel. Most FFVs are E-85 vehicles,
which can run on a mixture of up to 85 percent ethanol and gasoline.
Dedicated alternative fuel vehicles are vehicles that run exclusively
on an alternative fuel (e.g., compressed natural gas). EPCA includes an
incentive under the CAFE program for production of dual-fueled vehicles
or FFVs, and dedicated alternative fuel vehicles.\146\ EPCA's
provisions were amended by the EISA to extend the period of
availability of the FFV credits, but to begin phasing them out by
annually reducing the amount of FFV credits that can be used in
demonstrating compliance with the CAFE standards.\147\ EPCA does not
premise the availability of the FFV credits on actual use of
alternative fuel. Under EPCA, after MY 2019 no FFV credits will be
available for CAFE compliance.\148\ Under EPCA, for dedicated
alternative fuel vehicles, there are no limits or phase-out. EPA is
proposing that FFV and Alternative Fuel Vehicle Credits be calculated
as a part of the calculation of a manufacturer's overall fleet average
fuel economy and fleet average carbon-related exhaust emissions (Sec. 
600.510-12).
---------------------------------------------------------------------------

    \146\ 49 U.S.C 32905.
    \147\ See 49 U.S.C 32906. The mechanism by which EPCA provides
an incentive for production of FFVs is by specifying that their fuel
economy is determined using a special calculation procedure that
results in those vehicles being assigned a higher fuel economy level
than would otherwise occur. 49 U.S.C. section 32905(b). This is
typically referred to as an FFV credit.
    \148\ 49 U.S.C 32906.
---------------------------------------------------------------------------

    EPA is not proposing to include electric vehicles (EVs) or plug-in
hybrid electric vehicles (PHEVs) in these flex fuel and alternative
fuel provisions. These vehicles would be covered by the proposed
advanced technology vehicle credits provisions described in Section
III.C.3, so including them here would lead to a double counting of credits.
a. Model Year 2012--2015 Credits
    i. FFVs
    For the GHG program, EPA is proposing to allow FFV credits
corresponding to the amounts allowed by the amended EPCA only during
the period from MYs 2012 to 2015. (As discussed below in Section
III.E., EPA is proposing that CAFE-based FFV credits would not be
permitted as part of the early credits program.) Several manufacturers
have already taken the availability of FFV credits into account in
their near-term future planning for CAFE and this reliance indicates
that these credits need to be considered in considering adequacy of
lead time for the CO₂ standards. EPA thus believes that
allowing these credits, in the near term, would help provide adequate
lead time for manufacturers to implement the new multi-year standards,
but that for the longer term there is adequate lead time without the
use of such credits. This will also tend to harmonize the GHG and the
CAFE program during these interim years. As discussed below, EPA is
proposing for MY 2016 and later that manufacturers would not receive
FFV credits unless they reliably estimate the extent the alternative
fuel is actually being used by vehicles in order to count the alternative
fuel use in the vehicle's CO₂ emissions level determination.
    As with the CAFE program, EPA proposes to base credits on the
assumption that the vehicles would operate 50% of the time on the
alternative fuel and 50% of the time on conventional fuel, resulting in
CO₂ emissions that are based on an arithmetic average of
alternative fuel and conventional fuel CO₂ emissions.\149\
The measured CO₂ emissions on the alternative fuel would be
multiplied by a 0.15 volumetric conversion factor which is included in
the CAFE calculation as provided by EPCA. Through this mechanism a
gallon of alternative fuel is deemed to contain 0.15 gallons of fuel.
EPA is proposing to take the same approach for 2012-2015 model years.
For example, for a flexible-fuel vehicle that emitted 330 g/mi
CO₂ operating on E-85 and 350 g/mi CO₂ operating
on gasoline, the resulting CO₂ level to be used in the
manufacturer's fleet average calculation would be:
---------------------------------------------------------------------------

    \149\ 49 U.S.C 32905 (b).
    [GRAPHIC] [TIFF OMITTED] TP28SE09.012
   
    EPA understands that by using the CAFE approach--including the 0.15
factor--the CO₂ emissions value for the vehicle is
calculated to be significantly lower than it actually would be
otherwise, even if the vehicle were assumed to operate on the alternative
fuel at all times. This represents a ``credit'' being provided to FFVs.

[[Page 49532]]

    EPA notes also that the above equation and example are based on an
FFV that is an E-85 vehicle. EPCA, as amended by EISA, also establishes
the use of this approach, including the 0.15 factor, for all
alternative fuels, not just E-85.\150\ The 0.15 factor is used for B-20
(20 percent biofuel and 80 percent diesel) FFVs. EPCA also establishes
this approach, including the 0.15 factor, for gaseous-fueled FFVs such
as a vehicle able to operate on gasoline and CNG.\151\ (For natural gas
FFVs, EPCA establishes a factor of 0.823 gallons of fuel for every 100
cubic feet a natural gas used to calculate a gallons equivalent.) \152\
The EISA statute's use of the 0.15 factor in this way provides a
similar regulatory treatment across the various types of alternative
fuel vehicles. EPA also proposes to use the 0.15 factor for all FFVs in
keeping with the goal of not disrupting manufacturers' near-term
compliance planning. EPA, in any case, expects the vast majority of
FFVs to be E-85 vehicles, as is the case today.
---------------------------------------------------------------------------

    \150\ 49 U.S.C 32905 (c).
    \151\ 49 U.S.C 32905 (d).
    \152\ 49 U.S.C section 32905 (c).
---------------------------------------------------------------------------

    The FFV credit limits for CAFE are 1.2 mpg for model years 2012-
2014 and 1.0 mpg for model year 2015.\153\ In CO₂ terms,
these CAFE limits translate to declining CO₂ credit limits
over the four model years, as the CAFE standards increase in stringency
(as the CAFE standard increases numerically, the limit becomes a
smaller fraction of the standard). EPA proposes credit limits shown in
Table III.C.2-1 based on the proposed average CO₂ standards
for cars and trucks. These have been calculated by comparing the
average proposed CAFE standards with and without the FFV credits,
converted to CO₂. EPA requests comments on this proposed approach.
---------------------------------------------------------------------------

    \153\ 49 U.S.C section 32906 (a).

        Table III.C.2-1--FFV CO2 Standard Credit Limits (g/mile)
------------------------------------------------------------------------
                    Model year                         Cars      Trucks
------------------------------------------------------------------------
2012..............................................        9.8       17.9
2013..............................................        9.3       17.1
2014..............................................        8.9       16.3
2015..............................................        6.9       12.6
------------------------------------------------------------------------

    EPA also requests comments on basing the calculated CO₂
credit limit on the individual manufacturer standards calculated from
the footprint curves. For example, if a manufacturer's 2012 car
standard was 260 g/mile, the credit limit in CO₂ terms would
be 9.5 g/mile and if it were 270 g/mile the limit would be 10.2 g/mile.
This approach would be somewhat more complex and would mean that the
FFV CO₂ credit limits would vary by manufacturer as their
footprint based standards vary. However, it would more closely track
CAFE FFV credit limits.
    ii. Dedicated Alternative Fuel Vehicles
    EPA proposes to calculate CO₂ emissions from dedicated
alternative fuel vehicles for MY 2012--2015 by measuring the
CO₂ emissions over the test procedure and multiplying the
results by the 0.15 conversion factor described above. For example, for
a dedicated alternative fuel vehicle that would achieve 330 g/mi
CO₂ while operating on alcohol (ethanol or methanol), the
effective CO₂ emissions of the vehicle for use in determining
the vehicle's CO₂) emissions would be calculated as follows:

CO₂ = 330 x 0.15 = 49.5 g/mi
b. Model Years 2016 and Later
    i. FFVs
    For 2016 and later model years, EPA proposes to treat FFVs
similarly to conventional fueled vehicles in that FFV emissions would
be based on actual CO₂ results from emission testing on the
alternative fuel. The manufacturer would also be required to
demonstrate that the alternative fuel is actually being used in the
vehicles. The manufacturer would need to establish the ratio of
operation that is on the alternative fuel compared to the conventional
fuel. The ratio would be used to weight the CO₂ emissions
performance over the 2-cycle test on the two fuels. The 0.15 conversion
factor would no longer be included in the CO₂ emissions
calculation. For example, for a flexible-fuel vehicle that emitted 300
g/mi CO₂ operating on E-85 ten percent of the time and 350
g/mi CO₂ operating on gasoline ninety percent of the time,
the CO₂ emissions for the vehicles to be used in the
manufacturer's fleet average would be calculated as follows:

CO₂ = (300 x 0.10) + (350 x 0.90)= 345 g/mi
    The most complex part of this approach is to establish what data
are needed for a manufacturer to accurately demonstrate use of the
alternative fuel. One option EPA is considering is establishing a
rebuttable presumption using a ``top-down'' approach based on national
E-85 fuel use to assign credits to FFVs sold by manufacturers under
this program. For example, national E-85 volumes and national FFV sales
could be used to prorate E-85 use by manufacturer sales volumes and
FFVs already in-use. EPA would conduct an analysis of vehicle miles
travelled (VMT) by year for all FFVs using its emissions inventory
MOVES model. Using the VMT ratios and the overall E-85 sales, E-85
usage could be assigned to each vehicle. This method would account for
the VMT of new FFVs and FFVs already in the existing fleet using VMT
data in the model. The model could then be used to determine the ratio
of E-85 and gasoline for new vehicles being sold. Fluctuations in E-85
sales and FFV sales would be taken into account to adjust the credits
annually. EPA believes this is a reasonable way to apportion E-85 use
across the fleet.
    If manufacturers decided not to use EPA's assigned credits based on
the top-down analysis, they would have a second option of presenting
their own data for consideration as the basis for credits.
Manufacturers have suggested demonstrations using vehicle on-board data
gathering through the use of on-board sensors and computers.
California's program allows FFV credits based on FFV use and envisioned
manufacturers collecting fuel use data from vehicles in fleets with on-
site refueling. Any approach must reasonably ensure that no
CO₂ emissions reductions anticipated under the program are lost.
    EPA proposes that manufacturers would need to present a statistical
analysis of alternative fuel usage data collected on actual vehicle
operation. EPA is not attempting to specify how the data is collected
or the amount of data needed. However, the analysis must be based on
sound statistical methodology. Uncertainty in the analysis must be
accounted for in a way that provides reasonable certainty that the
program does not result in loss of emissions reductions. EPA requests
comment on how this demonstration could reasonably be made.
    EPA recognizes that under EPCA FFV credits are entirely phased-out
of the CAFE program by MY 2020, and apply in the prior years with
certain limitations, but without a requirement that the manufacturers
demonstrate actual use of the alternative fuel. Under this proposal EPA
would treat FFV credits the same as under EPCA for model years 2012-
2015, but would apply a different approach starting with model year
2016. Unlike EPCA, CAA section 202(a) does not mandate that EPA treat
FFVs in a specific way. Instead EPA is required to exercise its own
judgment and determine an appropriate approach that best promotes the
goals of this CAA section. Under these circumstances, EPA proposes to
treat FFVs for model years 2012-2015 the same as under EPCA, for the
lead time reasons described above. Starting

[[Page 49533]]

with model year 2016, EPA believes the appropriate approach is to
ensure that emissions reduction credits are based upon a demonstration
that emissions reductions have been achieved, to ensure the credits are
for real reductions instead of reductions that have not likely
occurred. This will promote the environmental goals of this proposal.
At the same time, the ability to generate credits upon a demonstration
of usage of the alternative fuel will provide an actual incentive to
see that such fuels are used. Under the EPCA credit provision, there is
an incentive to produce FFVs but no actual incentive to ensure that the
alternative fuels are used. GHG and energy security benefits are only
achieved if the alternative fuel is actually used, and EPA's approach
will now provide such an incentive. This approach will promote greater
use of renewable fuels, as compared to a situation where there is a
credit but no usage requirement. This is also consistent with the
agency's overall commitment to the expanded use of renewable fuels.
Therefore EPA is not proposing to phase-out the FFV program for MYs
2016 and later but instead to base the program on real-world reductions
(i.e., actual vehicle CO₂ emissions levels based on actual
use of the two fuels, without the 0.15 conversion factor specified
under EISA). Based on existing certification data, E-85 FFV
CO₂ emissions are typically about 5 percent lower on E-85
than CO₂ emissions on 100 percent gasoline. However,
currently there is little incentive to optimize CO₂
performance for vehicles when running on E-85. EPA believes the above
approach would provide such an incentive to manufacturers and that E-85
vehicles could be optimized through engine redesign and calibration to
provide additional CO₂ reductions. EPA requests comments on
the above.
    ii. Dedicated Alternative Fuel Vehicles
    EPA proposes that for model years 2016 and later dedicated
alternative fuel vehicles, CO₂ would be measured over the 2-
cycle test in order to be included in a manufacturer's fleet average
CO₂ calculations. As noted above, this is different than
CAFE methodology which provides a methodology for calculating a
petroleum-based mpg equivalent for alternative fuel vehicles so they
can be included in CAFE. However, because CO₂ can be
measured directly from alternative fuel vehicles over the test
procedure, EPA believes this is the simplest and best approach since it
is consistent with all other vehicle testing under the proposed
CO₂ program.
3. Advanced Technology Vehicle Credits for Electric Vehicles, Plug-in
Hybrids, and Fuel Cells
    EPA is proposing additional credit opportunities to encourage the
early commercialization of advanced vehicle powertrains, including
electric vehicles (EVs), plug-in hybrid electric vehicles (PHEVs), and
fuel cell vehicles. These technologies have the potential for more
significant reductions of GHG emissions than any technology currently
in commercial use, and EPA believes that encouraging early introduction
of such technologies will help to enable their wider use in the future,
promoting the technology-based emission reduction goals of section
202(a)(1) of the Clean Air Act.
    EPA proposes that these advanced technology credits would take the
form of a multiplier that would be applied to the number of vehicles
sold such that they would count as more than one vehicle in the
manufacturer's fleet average. These advanced technology vehicles would
then count more heavily when calculating fleet average CO₂
levels. The multiplier would not be applied when calculating the
manufacturer's foot-print-based standard, only when calculating the
manufacturer's fleet average levels. EPA proposes to use a multiplier
in the range of 1.2 to 2.0 for all EVs, PHEVs, and fuel cell vehicles
produced from MY 2012 through MY 2016. EPA proposes that starting in MY
2017, the multiplier would no longer be used. As described in Section
III.C.5, EPA is also proposing to allow early advanced technology
vehicle credits to be generated for model years 2009-2011. EPA requests
comment on the level of the multiplier and whether it should be the
same value for each of these three technologies. Further, if EPA
determines that a multiplier of 2.0, or another level near the higher
end of this range, is appropriate for the final rule, EPA requests
comment on whether the multiplier should be phased down over time, such
as: 2.0 for MY 2009 through MY 2012, 1.8 in MY 2013, 1.6 in MY 2014,
1.4 in MY 2015, and 1.2 in MY 2016 (i.e., the multiplier could phase-
down by 0.2 per year). In addition, EPA requests comment on whether or
not it would be appropriate to differentiate between EVs and PHEVs for
advanced technology credits. Under such an approach, PHEVs could be
provided a lesser multiplier compare to EVs. Also, the PHEV multiplier
could be prorated based on the equivalent electric range (i.e., the
extent to which the PHEV operates on average as an EV) of the vehicle
in order to incentivize battery technology development. This approach
would give more credits to ``stronger'' PHEV technology.
    EPA has provided this type of credit previously, in the Tier 2
program. This approach provides an incentive for manufacturers to prove
out ultra-clean technology during the early years of the program. In
Tier 2, early credits for Tier 2 vehicles certified to the very
cleanest bins (equivalent to California's standards for super ultra low
emissions vehicles (SULEVs) and zero emissions vehicles (ZEVs)) had a
multiplier of 1.5 or 2.0.\154\ The multiplier range of 1.2 to 2.0 being
proposed for GHGs is consistent with the Tier 2 approach. EPA believes
it is appropriate to provide incentives to manufacturers to produce
vehicles with very low emissions levels and that these incentives may
help pave the way for greater and/or more cost effective emission
reductions from future vehicles. EPA would like to finalize an approach
which appropriately balances the benefits of encouraging advanced
technologies with the overall environmental reductions of the proposed
standards as a whole.
---------------------------------------------------------------------------

    \154\ See 65 FR 6746, February 10, 2000.
---------------------------------------------------------------------------

    As with other vehicles, CO₂ for these vehicles would be
determined as part of vehicle certification, based on emissions over
the 2-cycle test procedures, to be included in the fleet average
CO₂ levels.
    For electric vehicles, EPA proposes that manufacturers would
include them in the average with CO₂ emissions of zero
grams/mile both for early credits, and for the MY 2012-2016 time frame.
Similarly, EPA proposes to include as zero grams/mile of CO₂
the electric portion of PHEVs (i.e., when PHEVs are operating as
electric vehicles) and fuel cell vehicles. EPA recognizes that for each
EV that is sold, in reality the total emissions off-set relative to the
typical gasoline or diesel powered vehicle is not zero, as there is a
corresponding increase in upstream CO₂ emissions due to an
increase in the requirements for electric utility generation. However,
for the time frame of this proposed rule, EPA is also interested in
promoting very advanced technologies such as EVs which offer the future
promise of significant reductions in GHG emissions, in particular when
coupled with a broader context which would include reductions from the
electricity generation. For the California Paley 1 program, California
assigned EVs a CO₂ performance value of 130 g/mile, which
was intended to represent the average CO₂ emissions required
to charge an EV using representative CO₂ values for the
California electric utility grid. For this

[[Page 49534]]

proposal, EPA is assigning an EV a value of zero g/mile, which should
be viewed as an interim solution for how to account for the emission
reduction potential of this type of vehicle, and may not be the
appropriate long-term approach. EPA requests comment on this proposal
and whether alternative approaches to address EV emissions should be
considered, including approaches for considering the lifecycle
emissions from such advanced vehicle technologies.
    The criteria and definitions for what vehicles qualify for the
multiplier are provided in Section III.E. As described in Section
III.E, EPA is proposing definitions for EVs, PHEVs, and fuel cell
vehicles to ensure that only credible advanced technology vehicles are
provided credits.
    EPA requests comments on the proposed approach for advanced
technology vehicle credits.
4. Off-Cycle Technology Credits
    EPA is proposing an optional credit opportunity intended to apply
to new and innovative technologies that reduce vehicle CO₂
emissions, but for which the CO₂ reduction benefits are not
captured over the 2-cycle test procedure used to determine compliance
with the fleet average standards (i.e., ``off-cycle''). Eligible
innovative technologies would be those that are relatively newly
introduced in one or more vehicle models, but that are not yet
implemented in widespread use in the light-duty fleet. EPA will not
approve credits for technologies that are not innovative or novel
approaches to reducing greenhouse gas emissions. Further, any credits
for these off-cycle technologies must be based on real-world GHG
reductions not captured on the current 2-cycle tests and verifiable
test methods, and represent average U.S. driving conditions.
    Similar to the technologies used to reduce A/C system indirect
CO₂ emissions such as compressor efficiency improvements,
eligible technologies would not be active during the 2-cycle test and
therefore the associated improvements in CO₂ emissions would
not be captured. EPA will not consider technologies to be eligible for
these credits if the technology has a significant impact on
CO₂ emissions over the FTP and HFET tests. Because these
technologies are not nearly so well developed and understood, EPA is
not prepared to require their utilization to meet the CO₂
standards. However, EPA is aware of some emerging and innovative
technologies and concepts in various stages of development with
CO₂ reduction potential that might not be adequately
captured on the FTP or HFET, and that some of these technologies might
merit some additional CO₂ credit for the manufacturer.
Examples include solar panels on hybrids or electric vehicles, adaptive
cruise control, and active aerodynamics. EPA believes it would be
appropriate to provide an incentive to encourage the introduction of
these types of technologies and that a credit mechanism is an effective
way to do this. This optional credit opportunity would be available
through the 2016 model year.
    EPA is proposing that manufacturers quantify CO₂
reductions associated with the use of the off-cycle technologies such
that the credits could be applied on a g/mile equivalent basis, as is
proposed for A/C system improvements. Credits would have to be based on
real additional reductions of CO₂ emissions and would need
to be quantifiable and verifiable with a repeatable methodology. Such
submissions of data should be submitted to EPA subject to public
scrutiny. EPA proposes that the technologies upon which the credits are
based would be subject to full useful life compliance provisions, as
with other emissions controls. Unless the manufacturer can demonstrate
that the technology would not be subject to in-use deterioration over
the useful life of the vehicle, the manufacturer would have to account
for deterioration in the estimation of the credits in order to ensure
that the credits are based on real in-use emissions reductions over the
life of the vehicle.
    As discussed below, EPA is proposing a two-tiered process for
demonstrating the CO₂ reductions of an innovative and novel
technology with benefits not captured by the FTP and HFET test
procedures. First, a manufacturer would determine whether the benefit
of the technology could be captured using the 5-cycle methodology
currently used to determine fuel economy label values. EPA established
the 5-cycle test methods to better represent real-world factors
impacting fuel economy, including higher speeds and more aggressive
driving, colder temperature operation, and the use of air conditioning.
If this determination is affirmative, the manufacturer would follow the
protocol laid out below and in the proposed regulations. If the
manufacturer finds that the technology is such that the benefit is not
adequately captured using the 5-cycle approach, then the manufacturer
would have to develop a robust methodology, subject to EPA approval, to
demonstrate the benefit and determine the appropriate CO₂
gram per mile credit.
a. Technology Demonstration Using EPA 5-Cycle Methodology
    As noted above, the CO₂ reduction benefit of some
innovative technologies could be demonstrated using the 5-cycle
approach currently used for EPA's fuel economy labeling program. The 5-
cycle methodology was finalized in EPA's 2006 fuel economy labeling
rule,\155\ which provides a more accurate fuel economy label estimate
to consumers starting with 2008 model year vehicles. In addition to the
FTP and HFET test procedures, the 5-cycle approach folds in the test
results from three additional test procedures to determine fuel
economy. The additional test cycles include cold temperature operation,
high temperature, high humidity and solar loading, and aggressive and
high-speed driving; thus these tests could be used to demonstrate the
benefit of a technology that reduces CO₂ over these types of
driving and environmental conditions. Using the test results from these
additional test cycles collectively with the 2-cycle data provides a
more precise estimate of the average fuel economy and CO₂
emissions of a vehicle for both the city and highway independently. A
significant benefit of using the 5-cycle methodology to measure and
quantify the CO₂ reductions is that the test cycles are
properly weighted for the expected average U.S. operation, meaning that
the test results could be used without further adjustments.
---------------------------------------------------------------------------

    \155\ Fuel Economy Labeling of Motor Vehicles: Revisions to
Improve Calculation of Fuel Economy Estimates; Final Rule (71 FR
77872, December 27, 2006).
---------------------------------------------------------------------------

    The use of these supplemental cycles may provide a method by which
technologies not demonstrated on the baseline 2-cycles can be
quantified. The cold temperature FTP can capture new technologies that
improve the CO₂ performance of vehicles during colder
weather operation. These improvements may be related to warm-up of the
engine or other operation during the colder temperature. An example of
such a new, innovative technology is a waste heat capture device that
provides heat to the cabin interior, enabling additional engine-off
operation during colder weather not previously enabled due to heating
and defrosting requirements. The additional engine-off time would
result in additional CO₂ reductions that otherwise would not
have been realized without the heat capture technology.
    While A/C credits for efficiency improvements will largely be
captured in the A/C credits proposal through the credit menu of known
efficiency improving components and controls,

[[Page 49535]]

certain new technologies may be able to use the high temperatures,
humidity, and solar load of the SC03 test cycle to accurately measure
their impact. An example of a new technology may be a refrigerant
storage device that accumulates pressurized refrigerant during driving
operation or uses recovered vehicle kinetic energy during deceleration
to pressurize the refrigerant. Much like the waste heat capture device
used in cold weather, this device would also allow additional engine-
off operation while maintaining appropriate vehicle interior occupant
comfort levels. SC03 test data measuring the relative impact of
innovative A/C-related technologies could be applied to the 5-cycle
equation to quantify the CO₂ reductions of the technology.
Another example is glazed windows. This reflects sunlight away from the
cabin so that the energy required to stabilize the cabin air to a
comfortable level is decreased. The impact of these windows may be
measureable on an SC03 test (with and without the window option).
    The US06 cycle may be used to capture innovative technologies
designed to reduce CO₂ emissions during higher speed and
more aggressive acceleration conditions, but not reflected on the 2-
cycle tests. An example of this is an active aerodynamic technology.
This technology recognizes the benefits of reduced aerodynamic drag at
higher speeds and makes changes to the vehicle at those speeds. The
changes may include active front or grill air deflection devices
designed to redirect frontal airflow. Certain active suspension devices
designed primarily to reduce aerodynamic drag by lowering the vehicle
at higher speeds may also be measured on the US06 cycle. To properly
measure these technologies on the US06, the vehicle would require
unique load coefficients with and without the technologies. The
different load coefficient (properly weighted for the US06 cycle) could
effectively result in reduced vehicle loads at the higher speeds when
the technologies are active. Similar to the previously discussed
cycles, the results from the US06 test with and without the technology
could then use the 5-cycle methodology to quantify CO₂ reductions.
    If the 5-cycle procedures can be used to demonstrate the innovative
technology, then the process would be relatively simple. The
manufacturer would simply test vehicles with and without the technology
installed or operating and compare results. All 5-cycles would be
tested with the technology enabled and disabled, and the test results
would be used to calculate a combined city/highway CO₂ value
with the technology and without the technology. These values would be
compared to determine the amount of the credit; the combined city/
highway CO₂ value with the technology operating would be
subtracted from the combined city/highway CO₂ value without
the technology operating to determine the gram per mile CO₂
credit. It is likely that multiple tests of each of the five test
procedures would need to be performed in order to achieve the necessary
strong degree of statistical significance of the credit determination
results. This would have to be done for each model type for which a
credit was being sought, unless the manufacturer could demonstrate that
the impact of the technology was independent of the vehicle
configuration on which it was installed. In this case, EPA may consider
allowing the test to be performed on an engine family basis or other
grouping. At the end of the model year, the manufacturer would
determine the number of vehicles produced subject to each credit amount
and report that to EPA in the final model year report. The gram per
mile credit value determined with the 5-cycle comparison testing would
be multiplied by the total production of vehicles subject to that value
to determine the total number of credits.
b. Alternative Off-Cycle Credit Methodologies
    In cases where the benefit of a technological approach to reducing
CO₂ emissions can not be adequately represented using
existing test cycles, EPA will work with and advise manufacturers in
developing test procedures and analytical approaches to estimate the
effectiveness of the technology for the purpose of generating credits.
Clearly the first step should be a thorough assessment of whether the
5-cycle approach can be used, but if the manufacturer finds that the 5-
cycle process is fundamentally inadequate for the specific technology
being considered by the manufacturer, then an alternative approach may
be developed and submitted to EPA for approval. The demonstration
program should be robust, verifiable, and capable of demonstrating the
real-world emissions benefit of the technology with strong statistical
significance.
    The CO₂ benefit of some technologies may be able to be
demonstrated with a modeling approach, using engineering principles. An
example would be where a roof solar panel is used to charge the on-
board vehicle battery. The amount of potential electrical power that
the panel could supply could be modeled for average U.S. conditions and
the units of electrical power translated to equivalent fuel energy or
annualized CO₂ emission rate reduction from the captured
solar energy. The CO₂ reductions from other technologies may
be more challenging to quantify, especially if they are interactive
with the driver, geographic location, environmental condition, or other
aspect related to operation on actual roads. In these cases,
manufacturers might have to design extensive on-road test programs. Any
such on-road testing programs would need to be statistically robust and
based on average U.S. driving conditions, factoring in differences in
geography, climate, and driving behavior across the U.S.
    Whether the approach involves on-road testing, modeling, or some
other analytical approach, the manufacturer would be required to
present a proposed methodology to EPA. EPA would approve the
methodology and credits only if certain criteria were met. Baseline
emissions and control emissions would need to be clearly demonstrated
over a wide range of real world driving conditions and over a
sufficient number of vehicles to address issues of uncertainty with the
data. Data would need to be on a vehicle model-specific basis unless a
manufacturer demonstrated model specific data was not necessary.
Approval of the approach to determining a CO₂ benefit would
not imply approval of the results of the program or methodology; when
the testing, modeling, or analyses are complete the results would
likewise be subject to EPA review and approval. EPA believes that
manufacturers could work together to develop testing, modeling, or
analytical methods for certain technologies, similar to the SAE
approach used for A/C refrigerant leakage credits.
    EPA requests comments on the proposed approach for off-cycle
emissions credits, including comments on how best to structure the
program. EPA particularly requests comments on how the case-by-case
approach to assessing off-cycle innovative technology credits could
best be designed, including ways to ensure the verification of real-
world emissions benefits and to ensure transparency in the process of
reviewing manufacturer's proposed test methods.
5. Early Credit Options
    EPA is proposing to allow manufacturers to generate early credits
in model years 2009-2011. As described below, credits could be
generated through early additional fleet average CO₂
reductions, early A/C system improvements, early advanced

[[Page 49536]]

technology vehicle credits, and early off-cycle credits. As with other
credits, early credits would be subject to a five year carry-forward
limit based on the model year in which they are generated. Early
credits could also be transferred between vehicle categories (e.g.,
between the car and truck fleet) or traded among manufacturers without
limits. The agencies note that CAFE credits earned in MYs prior to MY
2011 will still be available to manufacturers for use in the CAFE
program in accordance with applicable regulations.
    EPA is not proposing certification, compliance, or in-use
requirements for vehicles generating early credits. MY 2009 would be
complete and MY 2010 would be well underway by the time the rule is
promulgated. This would make certification, compliance, and in-use
requirements unworkable. As discussed below, manufacturers would be
required to submit an early credits report to EPA for approval no later
than the time they submit their final CAFE report for MY 2011. This
report would need to include details on all early credits the
manufacturer generates, why the credits are bona fide, how they are
quantified, and how they can be verified.
    As a general principle, EPA believes these early credit programs
must be designed in a way to ensure that they are capturing real-world
reductions. In addition, EPA wants to ensure these credit programs do
not provide an opportunity for manufacturers to earn ``windfall''
credits that do not result in actual, surplus CO₂ emission
reductions. EPA seeks comments on how to best ensure these objectives
are achieved in the design of the early credit program options.
a. Credits Based on Early Fleet Average CO₂ Reductions
    EPA is proposing opportunities for early credit generation in MYs
2009-2011 through over-compliance with a fleet average CO₂
baseline established by EPA. EPA is proposing four pathways for doing
so. Manufacturers would select one of the four paths for credit
generation for the entire three year period and could not switch
between pathways for different model years. For two pathways, the
baseline would be set by EPA to be equivalent to the California
standards for the relevant model year. Generally, manufacturers that
over-comply with those CARB standards would earn credits. Two
additional pathways, described below, would include credits based on
over-compliance with CAFE standards in States that have not adopted the
California standards.
    Pathway 1 would be to earn credits by over-complying with the
California equivalent baseline over the manufacturer's fleet of
vehicles sold nationwide. Pathway 2 would be for manufacturers to
generate credits against the baseline only for the fleet of vehicles
sold in California and the CAA section 177 States.\156\ This approach
would include any CAA 177 States as of the date of promulgation of the
Final Rule in this proceeding. Manufacturers would be required to
include both cars and trucks in the program. Under Pathways 1 and 2,
EPA proposes that manufacturers would be required to cover any deficits
incurred against the baseline levels established by EPA during the
three year period 2009-2011 before credits could be carried forward
into the 2012 model year. For example, a deficit in 2011 would have to
be subtracted from the sum of credits earned in 2009 and 2010 before
any credits could be applied to 2012 (or later) model year fleets. EPA
is proposing this provision to help ensure the early credits generated
under this program are consistent with the credits available under the
California program during these model years.
---------------------------------------------------------------------------

    \156\ CAA 177 States refers to States that have adopted the
California GHG standards. At present, there are thirteen CAA 177
States including New York, Massachusetts, Maryland, Vermont, Maine,
Connecticut, Arizona, New Jersey, New Mexico, Oregon, Pennsylvania,
Rhode Island, Washington, and Washington, DC.
---------------------------------------------------------------------------

    Table III.C.5-1 provides the California equivalent baselines EPA
proposes to use as the basis for CO₂ credit generation under
the California-based pathways. These are the California GHG standards
for the model years shown, with a 2.0 g/mile adjustment to account for
the exclusion of N₂O and CH₄, which are included
in the California GHG standards, but not included in the credits
program. Manufacturers would generate CO₂ credits by
achieving fleet average CO₂ levels below these baselines. As
shown in the table, the California-based early credit pathways are
based on the California vehicle categories. Also, the California-based
baseline levels are not footprint-based, but universal levels that all
manufacturers would use. Manufacturers would need to achieve fleet
levels below those shown in the table in order to earn credits.

        Table III.C.5-1--California Equivalent Baselines CO2 Emissions Levels for Early Credit Generation
----------------------------------------------------------------------------------------------------------------
                                                                                         Light trucks with a LVW
                                                                   Passenger cars and     of 3,751 or more and a
                          Model year                              light trucks with an   GVWR of up to 8,500 lbs
                                                                   LVW of 0-3,750 lbs        plus medium-duty
                                                                                            passenger vehicles
----------------------------------------------------------------------------------------------------------------
2009..........................................................                      321                      437
2010..........................................................                      299                      418
2011..........................................................                      265                      388
----------------------------------------------------------------------------------------------------------------

    EPA proposes that manufacturers using Pathways 1 or 2 above would
use year end car and truck sales in each category. Although production
data is used for the program starting in 2012, EPA is proposing to use
sales data for the early credits program in order to apportion vehicles
by State. This is described further below. Manufacturers would
calculate actual fleet average emissions over the appropriate vehicle
fleet, either for vehicles sold nationwide for Pathway 1, or California
plus 177 States sales for Pathway 2. Early CO₂ credits would
be based on the difference between the baseline shown in the table
above and the actual fleet average emissions level achieved. Any early
A/C credits generated by the manufacturer, described below in Section
III.C.5.b, would be included in the fleet average level determination.
In model year 2009, the California CO₂ standards for cars
(321 g/mi CO₂) are only slightly more stringent than the
2009 CAFE car standard of 27.5 mpg, which is approximately equivalent
to 323 g/mi CO₂, and the California light-truck standard
(437 g/mi CO₂) is less stringent than the equivalent CAFE
standard, recognizing that there are some differences between the way
the California program and the CAFE

[[Page 49537]]

program categorize vehicles. Under the proposed option, manufacturers
would have to show that they over comply over the entire three model
year time period, not just the 2009 model year, to generate early
credits under either Pathways 1, 2 or 3. A manufacturer cannot use
credits generated in model year 2009 unless they offset any debits from
model years 2010 and 2011. EPA expects that the requirement to over
comply over the entire time period covering these three model years
should mean that the credits that are generated are real and are in
excess of what would have otherwise occurred. However, because of the
circumstances involving the 2009 model year, in particular for
companies with significant truck sales, there is some concern that
under Pathways 1, 2, and 3, there is a potential for a large number of
credits generated in 2009 against the California standard, in
particular for a number of companies who have significantly over-
achieved on CAFE in recent model years. EPA wants to avoid a situation
where, contrary to expectation, some part of the early credits
generated by a manufacturer are in fact not excess, where companies
could trade such credits to other manufacturers, risking a delay in the
addition of new technology across the industry from the 2012 and later
EPA CO₂ standards. For this reason, EPA requests comment on
the merits of prohibiting the trading of model year 2009 generated
early credits between firms.
    In addition, for Pathways 1 and 2, EPA proposes that manufacturers
may also include alternative compliance credits earned per the
California alternative compliance program.\157\ These alternative
compliance credits are based on the demonstrated use of alternative
fuels in flex fuel vehicles. As with the California program, the
credits would be available beginning in MY 2010. Therefore, these early
alternative compliance credits would be available under EPA's program
for the 2010 and 2011 model years. FFVs would otherwise be included in
the early credit fleet average based on their emissions on the
conventional fuel. This would not apply to EVs and PHEVs. The emissions
of EVs and PHEVs would be determined as described in Section III.E.
Manufacturers could choose to either include their EVs and PHEVs in one
of the four pathways described in this section or under the early
advanced technology emissions credits described below, but not both due
to issues of credit double counting.
---------------------------------------------------------------------------

    \157\ See Section 6.6.E, California Environmental Protection
Agency Air Resources Board, Staff Report: Initial Statement of
Reasons For Proposed Rulemaking, Public Hearing to Consider Adoption
of Regulations to Control Greenhouse Gas Emissions From Motor
Vehicles, August 6, 2004.
---------------------------------------------------------------------------

    EPA is also proposing two additional early credit pathways
manufacturers could select. Pathways 3 and 4 incorporate credits based
on over-compliance with CAFE standards for vehicles sold outside of
California and CAA 177 States in MY 2009-2011. Pathway 3 would allow
manufacturers to earn credits as under Pathway 2, plus earn CAFE-based
credits in other States. Credits would not be generated for cars sold
in California and CAA 177 States unless vehicle fleets in those States
are performing better than the standards which otherwise would apply in
those States, i.e. the baselines shown in Table III.C.5-1 above.
    Pathway 4 would be for manufacturers choosing to forego California-
based early credits entirely and earn only CAFE-based credits outside
of California and CAA 177 States. EPA proposes that manufacturers would
not be able to include FFV credits under the CAFE-based early credit
pathways since those credits do not automatically reflect actual
reductions in CO₂ emissions.
    The proposed baselines for CAFE-based early pathways are provided
in Table III.C.5-2 below. They are based on the CAFE standards for the
2009-2011 model years. For CAFE standards in 2009-2011 model years that
are footprint-based, the baseline would vary by manufacturer.
Footprint-based standards are in effect for the 2011 model year CAFE
standards.\158\ Additionally, for Reform CAFE truck standards,
footprint standards are optional for the 2009-2010 model years. Where
CAFE footprint-based standards are in effect, manufacturers would
calculate a baseline using the footprints and sales of vehicles outside
of California and CAA 177 States. The actual fleet CO₂
performance calculation would also only include the vehicles sold
outside of California and CAA 177 States, and as mentioned above, may
not include FFV credits.
---------------------------------------------------------------------------

    \158\ 74 FR 14196, March 30, 2009.

   Table III.C.5-2--CAFE Equivalent Baselines CO2 Emissions Levels for
                         Early Credit Generation
------------------------------------------------------------------------
           Model year                    Cars               Trucks
------------------------------------------------------------------------
2009............................  323...............  381.*
2010............................  323...............  376.*
2011............................  Footprint-based     Footprint-based
                                   standard.           standard.
------------------------------------------------------------------------
* Would be footprint-based standard for manufacturers selecting
  footprint option under CAFE.

    For the CAFE-based pathways, EPA proposes to use the NHTSA car and
truck definitions that are in place for the model year in which credits
are being generated. EPA understands that the NHTSA definitions change
starting in the 2011 model year, and would therefore change part way
through the early credits program. EPA further recognizes that MDPVs
are not part of the CAFE program until the 2011 model year, and
therefore would not be part of the early credits calculations for 2009-
2010 under the CAFE-based pathways.
    Pathways 2 through 4 involve splitting the vehicle fleet into two
groups, vehicles sold in California and CAA 177 States and vehicles
sold outside of these States. This approach would require a clear
accounting of location of vehicle sales by the manufacturer. EPA
believes it will be reasonable for manufacturers to accurately track
sales by State, based on its experience with the National Low Emissions
Vehicle (NLEV) Program. NLEV required manufacturers to meet separate
fleet average standards for vehicles sold in two different regions of
the country.\159\ As with NLEV, the determination would be based on
where the completed vehicles are delivered as a point of first sale,
which in most cases would be the dealer.\160\
---------------------------------------------------------------------------

    \159\ 62 FR 31211, June 6, 1997.
    \160\ 62 FR 31212, June 6, 1997.
---------------------------------------------------------------------------

    As noted above, EPA proposes that manufacturers choosing to
generate early credits would select one of the four pathways for the
entire early credits program and would not be able to switch among
them. EPA proposes that manufacturers would submit their early credits
report when they submit their final CAFE report for MY 2011 (which is
required to be submitted no

[[Page 49538]]

later than 90 days after the end of the model year). Manufacturers
would have until then to decide which pathway to select. This would
give manufacturers enough time to determine which pathway works best
for them. This timing may be necessary in cases where manufacturers
earn credits in MY 2011 and need time to assess data and prepare an
early credits submittal for final EPA approval.
    The table below provides a summary of the four fleet average-based
CO₂ early credit pathways EPA is proposing. As noted above,
EPA is concerned with potential ``windfall'' credits and is seeking
comments on how to best ensure the objective of achieving surplus,
real-world reductions is achieved in the design of the credit programs.
In addition, EPA requests comments on the merits of each of these
pathways. Specifically, EPA requests comment on whether or not any of
the pathways could be eliminated to simplify the program without
diminishing its overall flexibility. For example, Pathway 2 may not be
particularly useful to manufacturers if the California/177 State and
overall national fleets are projected to be similar during these model
years. EPA also requests comment on proposed program implementation
structure and provisions.

   Table III.C.5-3--Summary of Proposed Early Fleet Average CO2 Credit
                                Pathways
------------------------------------------------------------------------

------------------------------------------------------------------------
Common Elements...................  --Manufacturers would select a
                                     pathway. Once selected, may not
                                     switch among pathways.
                                    --All credits subject to 5 year
                                     carry-forward restrictions.
                                    --For Pathways 2-4, vehicles
                                     apportioned by State based on point
                                     of first sale.
Pathway 1: California-based         --Manufacturers earn credits based
 Credits for National Fleet..        on fleet average emissions compared
                                     with California equivalent baseline
                                     set by EPA.
                                    --Based on nationwide CO2 sales-
                                     weighted fleet average.
                                    --Based on use of California vehicle
                                     categories.
                                    --FFV alternative compliance credits
                                     per California program may be
                                     included.
                                    --Once in the program, manufacturers
                                     must make up any deficits that are
                                     incurred prior to 2012 in order to
                                     carry credits forward to 2012 and
                                     later.
Pathway 2: California-based         --Same as Pathway 1, but
 Credits for vehicles sold in        manufacturers only includes
 California plus CAA 177 States.     vehicles sold in California and CAA
                                     177 States in the fleet average
                                     calculation.
Pathway 3: Pathway 2 plus CAFE-     --Manufacturer earns credits as
 based Credits outside of            provided by Pathway 2: California-
 California plus CAA 177 States.     based credits for vehicles sold in
                                     California plus CAA 177 States,
                                     plus:
                                    --CAFE-based credits allowed for
                                     vehicles sold outside of California
                                     and CAA 177 States.
                                    --For CAFE-based credits,
                                     manufacturers earn credits based on
                                     fleet average emissions compared
                                     with baseline set by EPA.
                                    --CAFE-based credits based on NHTSA
                                     car and truck definitions.
                                    --FFV credits not allowed to be
                                     included for CAFE-based credits.
Pathway 4: Only CAFE-based Credits  --Manufacturer elects to only earn
 outside of California plus CAA      CAFE-based credits for vehicles
 177 States.                         sold outside of California and CAA
                                     177 States. Earns no California and
                                     177 State credits.
                                    --For CAFE-based credits,
                                     manufacturers earn credits based on
                                     fleet average emissions compared
                                     with baseline set by EPA.
                                    --CAFE-based credits based on NHTSA
                                     car and truck definitions.
                                    --FFV credits not allowed to be
                                     included for CAFE-based credits.
------------------------------------------------------------------------

b. Early A/C Credits
    EPA proposes that manufacturers could earn early A/C credits in MYs
2009-2011 using the same A/C system design-based EPA provisions being
proposed for MYs commencing in 2012, as described in Section III.C.1,
above. Manufacturers would be able to earn early A/C CO₂-
equivalent credits by demonstrating improved A/C system performance,
for both direct and indirect emissions. To earn credits for vehicles
sold in California and CAA 177 States, the vehicles would need to be
included in one of the California-based early credit pathways described
above in III.C.5.a. EPA is proposing this constraint in order to avoid
credit double counting with the California program in place in those
States, which also allows A/C system credits in this time frame.
Manufacturers would fold the A/C credits into the fleet average
CO₂ calculations under the California-based pathway. For
example, the MY 2009 California-based program car baseline would be 321
g/mile (see Table III.C.5-1). If a manufacturer under Pathway 1 had a
MY 2009 car fleet average CO₂ level of 320 g/mile and then
earned an additional 9 g/mile CO₂-equivalent A/C credit, the
manufacturers would earn a total of 10 g/mile of credit. Vehicles sold
outside of California and 177 States would be eligible for the early A/
C credits whether or not the manufacturers participate in other aspects
of the early credits program.
c. Early Advanced Technology Vehicle Credits
    EPA is proposing to allow early advanced technology vehicle credits
for sales of EVs, PHEVs, and fuel cell vehicles. To avoid double-
counting, manufacturers would not be allowed to generate advanced
technology credits for vehicles they choose to include in Pathways 1
through 4 described in III.C.5.a, above. EPA proposes to use a similar
methodology to that proposed for MYs 2012 and later, as described in
Section III.C.3, above. EPA proposes to use a multiplier in the range
of 1.2 to 2.0 for all eligible vehicles (i.e., EVs, PHEVs, and fuel
cells). Manufacturers, however, would track the number of these
vehicles sold in the model years 2009--2011, and the emissions level of
the vehicles, rather than a CO₂ credit. When a manufacturer
chooses to use the vehicle credits to comply with 2012 or later
standards, the vehicle counts including the multiplier would be folded
into the CO₂ fleet average. For example, if a manufacturer
sells 1,000 EVs in MY 2011, and if the final multiplier level were 2.0,
the manufacturer would apply the multiplier of 2.0 and then be able to
include 2,000 vehicles at 0 g/mile in their MY 2012 fleet to decrease
the fleet average for that model year. As with other early credits,
these early advanced technology vehicle credits would be tracked by
model year (2009, 2010, or 2011) and would be subject to 5 year carry-
forward restrictions. Again,

[[Page 49539]]

manufacturers would not be allowed to include the EVs, PHEVs, or fuel
cell vehicles in the early credit pathways discussed above in Section
III.C.5.a, otherwise the vehicles would be double counted. As discussed
in Section III.C.3, EPA is requesting comment on a multiplier in the
range of 1.2 to 2.0, including a potential phase-down in the multiplier
by model year 2016, if a multiplier near the higher end of this range
is determined for the final rule. This request for comment also extends
to the potential for early advance technology vehicle credits. EPA is
also requesting comment on the appropriate gram/mile metric for EVs and
fuel cellvehicles, as well as for the EV-only contribution for a PHEV.
d. Early Off-Cycle Credits
    EPA's proposed off-cycle innovative technology credit provisions
are provided in Section III.C.4. EPA requests comment on beginning
these credits in the 2009-2011 time frame, provided manufacturers are
able to make the necessary demonstrations outlined in Section III.C.4, above.

D. Feasibility of the Proposed CO₂ Standards

    This proposal is based on the need to obtain significant GHG
emissions reductions from the transportation sector, and the
recognition that there are cost-effective technologies to achieve such
reductions in the 2012-2016 time frame. As in many prior mobile source
rulemakings, the decision on what standard to set is largely based on
the effectiveness of the emissions control technology, the cost and
other impacts of implementing the technology, and the lead time needed
for manufacturers to employ the control technology. The standards
derived from assessing these issues are also evaluated in terms of the
need for reductions of greenhouse gases, the degree of reductions
achieved by the standards, and the impacts of the standards in terms of
costs, quantified benefits, and other impacts of the standards. The
availability of technology to achieve reductions and the cost and other
aspects of this technology are therefore a central focus of this rulemaking.
    EPA is taking the same basic approach in this rulemaking, although
the technological problems and solutions involved in this rulemaking
differ in some ways from prior mobile source rulemakings. Here, the
focus of the emissions control technology is on reducing CO₂
and other greenhouse gases. Vehicles combust fuel to perform two basic
functions: (1) Transport the vehicle, its passengers and its contents,
and (2) operate various accessories during the operation of the vehicle
such as the air conditioner. Technology can reduce CO₂
emissions by either making more efficient use of the energy that is
produced through combustion of the fuel or reducing the energy needed
to perform either of these functions.
    This focus on efficiency calls for looking at the vehicle as an
entire system. In addition to fuel delivery, combustion, and
aftertreatment technology, any aspect of the vehicle that affects the
need to produce energy must also be considered. For example, the
efficiency of the transmission system, which takes the energy produced
by the engine and transmits it to the wheels, and the resistance of the
tires to rolling both have major impacts on the amount of fuel that is
combusted while operating the vehicle. The braking system, the
aerodynamics of the vehicle, and the efficiency of accessories, such as
the air conditioner, all affect how much fuel is combusted.
    In evaluating vehicle efficiency, we have excluded fundamental
changes in vehicles' size and utility. For example, we did not evaluate
converting minivans and SUVs to station wagons, converting vehicles
with four wheel drive to two wheel drive, or reducing headroom in order
to lower the roofline and reduce aerodynamic drag. We have limited our
assessment of technical feasibility and resultant vehicle cost to
technologies which maintain vehicle utility as much as possible.
Manufacturers may decide to alter the utility of the vehicles which
they sell in response to this rule. Assessing the societal cost of such
changes is very difficult as it involves assessing consumer preference
for a wide range of vehicle features.
    This need to focus on the efficient use of energy by the vehicle as
a system leads to a broad focus on a wide variety of technologies that
affect almost all the systems in the design of a vehicle. As discussed
below, there are many technologies that are currently available which
can reduce vehicle energy consumption. These technologies are already
being commercially utilized to a limited degree in the current light-
duty fleet. These technologies include hybrid technologies that use
higher efficiency electric motors as the power source in combination
with or instead of internal combustion engines. While already
commercialized, hybrid technology continues to be developed and offers
the potential for even greater efficiency improvements. Finally, there
are other advanced technologies under development, such as lean burn
gasoline engines, which offer the potential of improved energy
generation through improvements in the basic combustion process. In
addition, the available technologies are not limited to powertrain
improvements but also include mass reduction, electrical system
efficiencies, and aerodynamic improvements.
    The large number of possible technologies to consider and the
breadth of vehicle systems that are affected mean that consideration of
the manufacturer's design and production process plays a major role in
developing the proposed standards. Vehicle manufacturers typically
develop many different models by basing them on a limited number of
vehicle platforms. The platform typically consists of a common vehicle
architecture and structural components. This allows for efficient use
of design and manufacturing resources. Given the very large investment
put into designing and producing each vehicle model, manufacturers
typically plan on a major redesign for the models approximately every 5
years. At the redesign stage, the manufacturer will upgrade or add all
of the technology and make most other changes supporting the
manufacturer's plans for the next several years, including plans
related to emissions, fuel economy, and safety regulations.
    This redesign often involves a package of changes designed to work
together to meet the various requirements and plans for the model for
several model years after the redesign. This often involves significant
engineering, development, manufacturing, and marketing resources to
create a new product with multiple new features. In order to leverage
this significant upfront investment, manufacturers plan vehicle
redesigns with several model years of production in mind. Vehicle
models are not completely static between redesigns as limited changes
are often incorporated for each model year. This interim process is
called a refresh of the vehicle and generally does not allow for major
technology changes although more minor ones can be done (e.g., small
aerodynamic improvements, valve timing improvements, etc). More major
technology upgrades that affect multiple systems of the vehicle thus
occur at the vehicle redesign stage and not in the time period between
redesigns.
    As discussed below, there are a wide variety of CO₂
reducing technologies involving several different systems in the
vehicle that are available for consideration. Many can involve major
changes to the vehicle, such as changes to the engine block and
cylinder heads, redesign of the transmission and its

[[Page 49540]]

packaging in the vehicle, changes in vehicle shape to improve
aerodynamic efficiency and the application of aluminum in body panels
to reduce mass. Logically, the incorporation of emissions control
technologies would be during the periodic redesign process. This
approach would allow manufacturers to develop appropriate packages of
technology upgrades that combine technologies in ways that work
together and fit with the overall goals of the redesign. It also allows
the manufacturer to fit the process of upgrading emissions control
technology into its multi-year planning process, and it avoids the
large increase in resources and costs that would occur if technology
had to be added outside of the redesign process.
    This proposed rule affects five years of vehicle production, model
years 2012-2016. Given the now-typical five year redesign cycle, nearly
all of a manufacturer's vehicles will be redesigned over this period.
However, this assumes that a manufacturer has sufficient lead time to
redesign the first model year affected by this proposed rule with the
requirements of this proposed rule in mind. In fact, the lead time
available for model year 2012 is relatively short. The time between a
likely final rule and the start of 2013 model year production is likely
to be just over two years. At the same time, manufacturer product plans
indicate that they are planning on introducing many of the technologies
EPA projects could be used to show compliance with the proposed
CO₂ standards in both 2012 and 2013. In order to account for
the relatively short lead time available prior to the 2012 and 2013
model years, albeit mitigated by their existing plans, EPA has factored
this reality into how the availability is modeled for much of the
technology being considered for model years 2012-2016 as a whole. If
the technology to control greenhouse gas emissions is efficiently
folded into this redesign process, then EPA projects that 85 percent of
each manufacturer's sales will be able to be redesigned with many of
the CO₂ emission reducing technologies by the 2016 model
year, and as discussed below, to reduce emissions of HFCs from the air
conditioner.
    In determining the level of this first ever GHG emissions standard
under the CAA for light-duty vehicles, EPA proposes to use an approach
that accounts for and builds on this redesign process. This provides
the opportunity for several control technologies to be incorporated
into the vehicle during redesign, achieving significant emissions
reductions from the model at one time. This is in contrast to what
would be a much more costly approach of trying to achieve small
increments of reductions over multiple years by adding technology to
the vehicle piece by piece outside of the redesign process.
    As described below, the vast majority of technology required by
this proposal is commercially available and already being employed to a
limited extent across the fleet. The vast majority of the emission
reductions which would result from this proposed rule would result from
the increased use of these technologies. EPA also believes that this
proposed rule would encourage the development and limited use of more
advanced technologies, such as PHEVs and EVs.
    In developing the proposed standard, EPA built on the technical
work performed by the State of California during its development of its
statewide GHG program. EPA began by evaluating a nationwide CAA
standard for MY 2016 that would require the levels of technology
upgrade, across the country, which California standards would require
for the subset of vehicles sold in California under Pavley 1. In
essence, EPA evaluated the stringency of the California Pavley 1
program but for a national standard. As mentioned above, and as
described in detail in Section II.C of this preamble and Chapter 3 of
the Joint TSD, one of the important technical documents included in EPA
and NHTSA's assessment of vehicle technology effectiveness and costs
was the 2004 NESCCAF report which was the technical foundation for
California's Pavley 1 standard. However, in order to evaluate the
impact of standards with similar stringency on a national basis to the
California program EPA chose not to evaluate the specific California
standards for several reasons. First, California's standards are
universal standards (one for cars and one for trucks), while EPA is
proposing attribute-based standards using vehicle footprint. Second,
California's definitions of what vehicles are classified as cars and
which are classified as trucks are different from those used by NHTSA
for CAFE purposes and different from EPA's proposed classifications in
this notice (which harmonizes with the CAFE definitions). In addition,
there has been progress in the refinement of the estimation of the
effectiveness and cost estimation for technologies which can be applied
to cars and trucks since the California analysis in 2004 which could
lead to different relative stringencies between cars and trucks than
what California determined for its Pavley 1 program. There have also
been improvements in the fuel economy and CO₂ performance of
the actual new vehicle fleet since California's 2004 analysis which EPA
wanted to reflect in our current assessment. For these reasons, EPA
developed an assessment of an equivalent national new vehicle fleet-
wide CO₂ performance standards for model year 2016 which
would result in the new vehicle fleet in the State of California having
CO₂ performance equal to the performance from the California
Pavley 1 standards. This assessment is documented in Chapter 3.1 of the
DRIA. The results of this assessment predicts that a national light-
duty vehicle fleet which adopts technology that achieves performance of
250 g/mile CO₂ for model year 2016 would result in vehicles
sold in California that would achieve the CO₂ performance
equivalent to the Pavley 1 standards.
    EPA then analyzed a level of 250 g/mi CO₂ in 2016 using
the OMEGA model, and the car and truck footprint curves relative
stringency discussed in Section II to determine what technology would
be needed to achieve a fleet wide average of 250 g/mi CO₂.
As discussed later in this section we believe this level of technology
application to the light-duty vehicle fleet can be achieved in this
time frame, that such standards will produce significant reductions in
GHG emissions, and that the costs for both the industry and the costs
to the consumer are reasonable. EPA also developed standards for the
model years 2012 through 2015 that lead up to the 2016 level.
    EPA's independent technical assessment of the technical feasibility
of the proposed MY2012-2016 standards is described below. EPA has also
evaluated a set of alternative standards for these model years, one
that is more stringent than the proposed standards and one that is less
stringent. The technical feasibility of these alternative standards is
discussed at the end of this section.
    Evaluating the feasibility of these standards primarily includes
identifying available technologies and assessing their effectiveness,
cost, and impact on relevant aspects of vehicle performance and
utility. The wide number of technologies which are available and likely
to be used in combination requires a more sophisticated assessment of
their combined cost and effectiveness. An important factor is also the
degree that these technologies are already being used in the current
vehicle fleet and thus, unavailable for use to improve energy
efficiency beyond current levels. Finally, the challenge for
manufacturers to design the technology

[[Page 49541]]

into their products, and the appropriate lead time needed to employ the
technology over the product line of the industry must be considered.
    Applying these technologies efficiently to the wide range of
vehicles produced by various manufacturers is a challenging task. In
order to assist in this task, EPA has developed a computerized model
called the Optimization Model for reducing Emissions of Greenhouse
gases from Automobiles (OMEGA) model. Broadly, the model starts with a
description of the future vehicle fleet, including manufacturer, sales,
base CO₂ emissions, footprint and the extent to which
emission control technologies are already employed. For the purpose of
this analysis, over 200 vehicle platforms were used to capture the
important differences in vehicle and engine design and utility of
future vehicle sales of roughly 16 million units in the 2016 timeframe.
The model is then provided with a list of technologies which are
applicable to various types of vehicles, along with their cost and
effectiveness and the percentage of vehicle sales which can receive
each technology during the redesign cycle of interest. The model
combines this information with economic parameters, such as fuel prices
and a discount rate, to project how various manufacturers would apply
the available technology in order to meet various levels of emission
control. The result is a description of which technologies are added to
each vehicle platform, along with the resulting cost. While OMEGA can
apply technologies which reduce CO₂ emissions and HFC
refrigerant emissions associated with air conditioner use, this task is
currently handled outside of the OMEGA model. The model can be set to
account for various types of compliance flexibilities, such as FFV credits.
    EPA invites comment on all aspects of this feasibility assessment.
Both the OMEGA model and its inputs have been placed in the docket to
this proposed rule and available for review.
    The remainder of this section describes the technical feasibility
analysis in greater detail. Section III.D.1 describes the development
of our projection of the MY 2012-2016 fleet in the absence of this
proposed rule. Section III.D.2 describes our estimates of the
effectiveness and cost of the control technologies available for
application in the 2012-2016 timeframe. Section III.D.3 combines these
technologies into packages likely to be applied at the same time by a
manufacturer. In this section, the overall effectiveness of the
technology packages vis-[agrave]-vis their effectiveness when combined
individually is described. Section III.D.4 describes the process which
manufacturers typically use to apply new technology to their vehicles.
Section III.D.5 describes EPA's OMEGA model and its approach to
estimating how manufacturers would add technology to their vehicles in
order to comply with CO₂ emission standards. Section III.D.6
presents the results of the OMEGA modeling, namely the level of
technology added to manufacturers' vehicles and its cost. Section
III.D.7 discusses the feasibility of the alternative 4-percent-per-year
and 6-percent-per-year standards. Further detail on all of these issues
can be found in EPA and NHTSA's draft Joint Technical Support Document
as well as EPA's draft Regulatory Impact Analysis.
1. How Did EPA Develop a Reference Vehicle Fleet for Evaluating Further
CO₂ Reductions?
    In order to calculate the impacts of this proposed regulation, it
is necessary to project the GHG emissions characteristics of the future
vehicle fleet absent this proposed regulation. This is called the
``reference'' fleet. EPA developed this reference fleet by determining
the characteristics of a specific model year (in this case, 2008) of
vehicles, called the baseline fleet, and then projecting what changes
if any would be made to these vehicles to comply with the MY2011 CAFE
standards. Thus, the MY 2008 fleet is our ``baseline fleet,'' and the
projection of the baseline to MY 2011-2016 is called the ``reference fleet.''
    EPA used 2008 model year vehicles as the basis for its baseline
fleet. 2008 model year is the most recent model year for which data is
publicly available. Sources of data for the baseline include the EPA
vehicle certification data, Ward's Automotive Group data,
Motortrend.com, Edmunds.com, manufacturer product plans, and other
sources to a lesser extent (such as articles about specific vehicles)
revealed from Internet search engine research. EPA then projects this
fleet out to the 2016 MY, taking into account factors such as changes
in overall sales volume. Section II.B describes the development of the
EPA reference fleet, and further details can be found in Section II.B
of this preamble and Chapter 1 of the Draft Joint TSD.
    The light-duty vehicle market is currently in a state of flux due
to the volatility in fuel prices over the past several years and the
current economic downturn. These factors have changed the relative
sales of the various types of light-duty vehicles marketed, as well as
total sales volumes. EPA and NHTSA desire to account for these changes
to the degree possible in our forecast of the make-up of the future
vehicle fleet. EPA wants to include improvements in fuel economy
associated with the existing CAFE program. It is possible that
manufacturers could increase fuel economy beyond the level of the 2011
MY CAFE standards for marketing purposes. However, it is difficult to
separate fuel economy improvements in those years for marketing
purposes from those designed to facilitate compliance with anticipated
CAFE or CO₂ emission standards. Thus, EPA limits fuel
economy improvements in the reference fleet to those projected to
result from the existing CAFE standards. The addition of technology to
the baseline fleet so that it complies with the MY 2011 CAFE standards
is described later in Section III.D.4, as this uses the same
methodology used to project compliance with the proposed CO₂
emission standards. In summary, the reference fleet represents vehicle
characteristics and sales in the 2012 and later model years absent this
proposed rule. Technology is then added to these vehicles in order to
reduce CO₂ emissions to achieve compliance with the proposed
CO₂ standards. EPA did not factor in any changes to vehicle
characteristics or sales in projecting manufacturers' compliance with
this proposal.
    After the reference fleet is created, the next step aggregates
vehicle sales by a combination of manufacturer, vehicle platform, and
engine design. As discussed in Section III.D.4 below, manufacturers
implement major design changes at vehicle redesign and tend to
implement these changes across a vehicle platform. Because the cost of
modifying the engine depends on the valve train design (such as SOHC,
DOHC, etc.), the number of cylinders and in some cases head design, the
vehicle sales are broken down beyond the platform level to reflect relevant
engine differences. The vehicle groupings are shown in Table III.D.1-1.

[[Page 49542]]

                 Table III.D.1-1--Vehicle Groupings \a\
------------------------------------------------------------------------
                                  Vehicle        Vehicle        Vehicle
      Vehicle description           type       description        type
------------------------------------------------------------------------
Large SUV (Car) V8+ OHV........         13  Subcompact Auto            1
                                             I4.
Large SUV (Car) V6 4v..........         16  Large Pickup V8+          19
                                             DOHC.
Large SUV (Car) V6 OHV.........         12  Large Pickup V8+          14
                                             SOHC 3v.
Large SUV (Car) V6 2v SOHC.....          9  Large Pickup V8+          13
                                             OHV.
Large SUV (Car) I4 and I5......          7  Large Pickup V8+          10
                                             SOHC.
Midsize SUV (Car) V6 2v SOHC...          8  Large Pickup V6           18
                                             DOHC.
Midsize SUV (Car) V6 S/DOHC 4v.          5  Large Pickup V6           12
                                             OHV.
Midsize SUV (Car) I4...........          7  Large Pickup V6           11
                                             SOHC 2v.
Small SUV (Car) V6 OHV.........         12  Large Pickup I4 S/         7
                                             DOHC.
Small SUV (Car) V6 S/DOHC......          4  Small Pickup V6           12
                                             OHV.
Small SUV (Car) I4.............          3  Small Pickup V6            8
                                             2v SOHC.
Large Auto V8+ OHV.............         13  Small Pickup I4..          7
Large Auto V8+ SOHC............         10  Large SUV V8+             17
                                             DOHC.
Large Auto V8+ DOHC, 4v SOHC...          6  Large SUV V8+             14
                                             SOHC 3v.
Large Auto V6 OHV..............         12  Large SUV V8+ OHV         13
Large Auto V6 SOHC 2/3v........          5  Large SUV V8+             10
                                             SOHC.
Midsize Auto V8+ OHV...........         13  Large SUV V6 S/           16
                                             DOHC 4v.
Midsize Auto V8+ SOHC..........         10  Large SUV V6 OHV.         12
Midsize Auto V7+ DOHC, 4v SOHC.          6  Large SUV V6 SOHC          9
                                             2v.
Midsize Auto V6 OHV............         12  Large SUV I4/....          7
Midsize Auto V6 2v SOHC........          8  Midsize SUV V6            12
                                             OHV.
Midsize Auto V6 S/DOHC 4v......          5  Midsize SUV V6 2v          8
                                             SOHC.
Midsize Auto I4................          3  Midsize SUV V6 S/          5
                                             DOHC 4v.
Compact Auto V7+ S/DOHC........          6  Midsize SUV I4 S/          7
                                             DOHC.
Compact Auto V6 OHV............         12  Small SUV V6 OHV.         12
Compact Auto V6 S/DOHC 4v......          4  Minivan V6 S/DOHC         16
Compact Auto I5................          7  Minivan V6 OHV...         12
Compact Auto I4................          2  Minivan I4.......          7
Subcompact Auto V8+ OHV........         13  Cargo Van V8+ OHV         13
Subcompact Auto V8+ S/DOHC.....          6  Cargo Van V8+             10
                                             SOHC.
Subcompact Auto V6 2v SOHC.....          8  Cargo Van V6 OHV.         12
Subcompact Auto I5/V6 S/DOHC 4v          4  .................  .........
------------------------------------------------------------------------
\a\ I4 = 4 cylinder engine, I5 = 5 cylinder engine, V6, V7, and V8 = 6,
  7, and 8 cylinder engines, respectively, DOHC = Double overhead cam,
  SOHC = Single overhead cam, OHV = Overhead valve, v = number of valves
  per cylinder, ``/'' = and, ``+'' = or larger.

    As mentioned above, the second factor which needs to be considered
in developing a reference fleet against which to evaluate the impacts
of this proposed rule is the impact of the 2011 MY CAFE standards,
which were published earlier this year. Since the vehicles which
comprise the above reference fleet are those sold in the 2008 MY, when
coupled with our sales projections, they do not necessarily meet the
2011 MY CAFE standards.
    The levels of the 2011 MY CAFE standards are straightforward to
apply to future sales fleets, as is the potential fine-paying
flexibility afforded by the CAFE program (i.e., $55 per mpg of
shortfall). However, projecting some of the compliance flexibilities
afforded by EISA and the CAFE program are less clear. Two of these
compliance flexibilities are relevant to EPA's analysis: (1) The credit
for FFVs, and (2) the limit on the transferring of credits between car
and truck fleets. The FFV credit is limited to 1.2 mpg in 2011 and EISA
gradually reduces this credit, to 1.0 mpg in 2015 and eventually to
zero in 2020. In contrast, the limit on car truck transfer is limited
to 1.0 mpg in 2011, and EISA increases this to 1.5 mpg beginning in
2015 and then to 2.0 mpg beginning in 2020. The question here is
whether to hold the 2011 MY CAFE provisions constant in the future or
incorporate the changes in the FFV credit and car-truck credit trading
limits contained in EISA.
    EPA decided to hold the 2011 MY limits on FFV credit and car-truck
credit trading constant in projecting the fuel economy and
CO₂ emission levels of vehicles in our reference case. This
approach treats the changes in the FFV credit and car-truck credit
trading provisions consistently with the other EISA-mandated changes in
the CAFE standards themselves. All EISA provisions relevant to 2011 MY
vehicles are reflected in our reference case fleet, while all post-2011
MY provisions are not. Practically, relative to the alternative, this
increases both the cost and benefit of the proposed standards. In our
analysis of this proposed rule, any quantified benefits from the
presence of FFVs in the fleet are not considered. Thus, the only impact
of the FFV credit is to reduce onroad fuel economy. By assuming that
the FFV credit stays at 1.2 mpg in the future absent this rule, the
assumed level of onroad fuel economy that would occur absent this
proposal is reduced. As this proposal eliminates the FFV credit
starting in 2016, the net result is to increase the projected level of
fuel savings from our proposed standards. Similarly, the higher level
of FFV credit reduces projected compliance cost for manufacturers to
meet the 2011 MY standards in our reference case. This increases the
projected cost of meeting the proposed 2012 and later standards.
    As just implied, EPA needs to project the technology (and resultant
costs) required for the 2008 MY vehicles to comply with the 2011 MY
CAFE standards in those cases where they do not automatically do so.
The technology and costs are projected using the same methodology that
projects compliance with the proposed 2012 and later CO₂
standards. The description of this process is described in the
following four sections.
    A more detailed description of the methodology used to develop
these sales projections can be found in the Draft Joint TSD. Detailed
sales projections by model year and manufacturer can also be found in
the TSD. EPA requests comments on both

[[Page 49543]]

the methodology used to develop the reference fleet, as well as the
characteristics of the reference fleet.
2. What Are the Effectiveness and Costs of CO₂-Reducing Technologies?
    EPA and NHTSA worked together to jointly develop information on the
effectiveness and cost of the CO₂-reducing technologies, and
fuel economy-improving technologies, other than A/C related control
technologies. This joint work is reflected in Chapter 3 of the Draft
Joint TSD and in Section II of this preamble. A summary of the
effectiveness and cost of A/C related technology is contained here. For
more detailed information on the effectiveness and cost of A/C related
technology, please refer to Section III.C of this preamble and Chapter
2 of EPA's DRIA.
    A/C improvements are an integral part of EPA's technology analysis
and have been included in this section along with the other technology
options. While discussed in Section III.C as a credit opportunity, air
conditioning-related improvements are included in Table III.D.2-
1.because A/C improvements are a very cost-effective technology at
reducing CO₂ (or CO₂-equivalent) emissions. EPA
expects most manufacturers will choose to use AC improvement credit
opportunities as a strategy for meeting compliance with the
CO₂ standards. Note that the costs shown in Table III.D.2-1
do not include maintenance savings that would be expected from the new
AC systems. Further, EPA does not include AC-related maintenance
savings in our cost and benefit analysis presented in Section III.H.
EPA discusses the likely maintenance savings in Chapter 2 of the DRIA
and requests comment on that discussion because we may include
maintenance savings in the final rule and would like to have the best
information available in order to do so. The EPA approximates that the
level of the credits earned will increase from 2012 to 2016 as more
vehicles in the fleet are redesigned. The penetrations and average
levels of credit are summarized in Table III.D.2-2, though the
derivation of these numbers (and the breakdown of car vs. truck
credits) is described in the DRIA. As demonstrated in the IMAC study
(and described in Section III.C as well as the DRIA), these levels are
feasible and achievable with technologies that are available and cost-
effective today.
    These improvements are categorized as either leakage reduction,
including use of alternative refrigerants, or system efficiency
improvements. Unlike the majority of the technologies described in this
section, A/C improvements will not be demonstrated in the test cycles
used to quantify CO₂ reductions in this proposal. As
described earlier, for this analysis A/C-related CO₂
reductions are handled outside of OMEGA model and therefore their
CO₂ reduction potential is expressed in grams per mile
rather than a percentage used by the OMEGA model. See Section III.C for
the method by which potential reductions are calculated or measured.
Further discussion of the technological basis for these improvements is
included in Chapter 2 of the DRIA.

  Table III.D.2-1--Total CO2 Reduction Potential and 2016 Cost for A/C
              Related Technologies for All Vehicle Classes
                         [Costs in 2007 dollars]
------------------------------------------------------------------------
                                    CO2 reduction         Incremental
                                      potential         compliance costs
------------------------------------------------------------------------
A/C refrigerant leakage         7.5 g/mi \161\.......                $17
 reduction.
A/C efficiency improvements...  5.7 g/mi.............                 53
------------------------------------------------------------------------


 Table III.D.2-2 A/C Related Tech- nology Penetration and Credit Levels
                          Expected To Be Earned
------------------------------------------------------------------------
                                            Technology        Average
                                            penetration    credit  over
                                             (Percent)     entire  fleet
------------------------------------------------------------------------
2012....................................              25             3.1
2013....................................              40             5.0
2014....................................              60             7.5
2015....................................              80            10.0
2016....................................              85            10.6
------------------------------------------------------------------------

3. How Can Technologies Be Combined into ``Packages'' and What Is the
Cost and Effectiveness of Packages?
    Individual technologies can be used by manufacturers to achieve
incremental CO₂ reductions. However, as mentioned in Section
III.D.1, EPA believes that manufacturers are more likely to bundle
technologies into ``packages'' to capture synergistic aspects and
reflect progressively larger CO₂ reductions with additions
or changes to any given package. In addition, manufacturers would
typically apply new technologies in packages during model redesigns--
which occur once roughly every five years--rather than adding new
technologies one at a time on an annual or biennial basis. This way,
manufacturers can more efficiently make use of their redesign resources
and more effectively plan for changes necessary to meet future standards.
---------------------------------------------------------------------------

    \161\ This represents 50% improvement in leakage and thus 50% of
the A/C leakage impact potential compared to a maximum of 15 g/mi
credit that can be achieved through the incorporation of a low very
GWP refrigerant.
---------------------------------------------------------------------------

    Therefore, the approach taken here is to group technologies into
packages of increasing cost and effectiveness. EPA determined that 19
different vehicle types provided adequate representation to accurately
model the entire fleet. This was the result of analyzing the existing
light duty fleet with respect to vehicle size and powertrain
configurations. All vehicles, including cars and trucks, were first
distributed based on their relative size, starting from compact cars
and working upward to large trucks. Next, each vehicle was evaluated
for powertrain, specifically the engine size, I4, V6, and V8, and
finally by the number of valves per cylinder. Note that each of these
19 vehicle types was mapped into one of the five classes of vehicles
mentioned in Section III.D.2. While the five classes provide adequate
representation for the cost basis associated with most technology
application, they do not adequately account for all existing vehicle
attributes such as base vehicle powertrain configuration and mass
reduction. As an example, costs and effectiveness estimates for engine
friction reduction for the small car class were used to represent cost
and effectiveness for three vehicle types: Subcompact cars, compact
cars, and small multi-purpose vehicles (MPV) equipped with a 4-cylinder
engine, however the mass reduction associated for each of these vehicle
types was based on the vehicle type sales-weighted average. In another
example, a vehicle type for V8 single overhead cam 3-valve engines was
created to properly account for the incremental cost in moving to a
dual overhead cam 4-valve

[[Page 49544]]

configuration. Note also that these 19 vehicle types span the range of
vehicle footprints--smaller footprints for smaller vehicles and larger
footprints for larger vehicles--which serve as the basis for the
standards proposed in this rule. A complete list of vehicles and their
associated vehicle types is shown above in Table III.D.1-1.
    Within each of the 19 vehicle types multiple technology packages
were created in increasing technology content and, hence, increasing
effectiveness. Important to note is that the effort in creating the
packages attempted to maintain a constant utility for each package as
compared to the baseline package. As such, each package is meant to
provide equivalent driver-perceived performance to the baseline
package. The initial packages represent what a manufacturer will most
likely implement on all vehicles, including low rolling resistance
tires, low friction lubricants, engine friction reduction, aggressive
shift logic, early torque converter lock-up, improved electrical
accessories, and low drag brakes.\162\ Subsequent packages include
advanced gasoline engine and transmission technologies such as turbo/
downsizing, GDI, and dual-clutch transmission. The most technologically
advanced packages within a segment included HEV, PHEV and EV designs.
The end result being a list of several packages for each of 19
different vehicle types from which a manufacturer could choose in order
to modify its fleet such that compliance could be achieved.
---------------------------------------------------------------------------

    \162\ When making reference to low friction lubricants, the
technology being referred to is the engine changes and possible
durability testing that would be done to accommodate the low
friction lubricants, not the lubricants themselves.
---------------------------------------------------------------------------

    Before using these technology packages as inputs to the OMEGA
model, the cost and effectiveness for the package was calculated. The
first step--mentioned briefly above--was to apply the scaling class for
each technology package and vehicle type combination. The scaling class
establishes the cost and effectiveness for each technology with respect
to the vehicle size or type. The Large Car class was provided as an
example in Section III.D.2. Additional classes include Small Car,
Minivan, Small Truck, and Large Truck and each of the 19 vehicle types
was mapped into one of those five classes. In the next step, the cost
for a particular technology package, was determined as the sum of the
costs of the applied technologies. The final step, determination of
effectiveness, requires greater care due to the synergistic effects
mentioned in Section III.D.2. This step is described immediately below.
    Usually, the benefits of the engine and transmission technologies
can be combined multiplicatively. For example, if an engine technology
reduces CO₂ emissions by five percent and a transmission
technology reduces CO₂ emissions by four percent, the
benefit of applying both technologies is 8.8 percent (100%-(100%-4%) *
(100%-5%)). In some cases, however, the benefit of the transmission-
related technologies overlaps with many of the engine technologies.
This occurs because the primary goal of most of the transmission
technologies is to shift operation of the engine to more efficient
locations on the engine map. Some of the engine technologies have the
same goal, such as cylinder deactivation. In order to account for this
overlap and avoid over-estimating emissions reduction effectiveness,
EPA has developed a set of adjustment factors associated with specific
pairs of engine and transmission technologies.
    The various transmission technologies are generally mutually
exclusive. As such, the effectiveness of each transmission technology
generally supersedes each other. For example, the 9.5-14.5 percent
reduction in CO₂ emissions associated with the automated
manual transmission includes the 4.5-6.5 percent benefit of a 6-speed
automatic transmission. Exceptions are aggressive shift logic and early
torque converter lock-up. The former can be applied to any vehicle and
the latter can be applied to any vehicle with an automatic transmission.
    EPA has chosen to use an engineering approach known as the lumped-
parameter technique to determine these adjustment factors. The results
from this approach were then applied directly to the vehicle packages.
The lumped-parameter technique is well documented in the literature,
and the specific approach developed by EPA is detailed in Chapter 3 of
the Draft Joint TSD.
    Table III.D.3-1 presents several examples of the reduction in the
effectiveness of technology pairs. A complete list and detailed discussion
of these synergies is presented in Chapter 3 of the Draft Joint TSD.

   Table III.D.3-1--Reduction in Effectiveness for Selected Technology
                                  Pairs
------------------------------------------------------------------------
                                                          Reduction in
                                      Transmission          combined
        Engine technology              technology        effectiveness
                                                           (percent)
------------------------------------------------------------------------
Intake cam phasing..............  5 speed automatic..                0.5
Coupled cam phasing.............  5 speed automatic..                0.5
Coupled cam phasing.............  Aggressive shift                   0.5
                                   logic.
Cylinder deactivation...........  5 speed automatic..                1.0
Cylinder deactivation...........  Aggressive shift                   0.5
                                   logic.
------------------------------------------------------------------------

    Table III.D.3-2 presents several examples of the CO₂-
reducing technology vehicle packages used in the OMEGA model for the
large car class. Similar packages were generated for each of the 19
vehicle types and the costs and effectiveness estimates for each of
those packages are discussed in detail in Chapter 3 of the Draft Joint TSD.

[[Page 49545]]

    Table III.D.3-2--CO2 Reducing Technology Vehicle Packages for a Large Car Effectiveness and Costs in 2016
                                             [Costs in 2007 dollars]
----------------------------------------------------------------------------------------------------------------
                                            Transmission                                    CO2        Package
         Engine  technology                  technology         Additional  technology   reduction       cost
----------------------------------------------------------------------------------------------------------------
3.3L V6.............................  4 speed automatic......  None...................          Baseline
                                                                                       -------------------------
3.0L V6 + GDI + CCP.................  6 speed automatic......  3% Mass Reduction......        17.9%       $1,022
3.0L V6 + GDI + CCP + Deac..........  6 speed automatic......  5% Mass Reduction......         20.6        1,280
3.0L V6 + GDI + CCP + Deac..........  6 speed DCT............  10% Mass Reduction              34.2        2,108
                                                                Start-Stop.
2.2L I4 + GDI + Turbo + DCP.........  6 speed DCT............  10% Mass Reduction              34.3        2,245
                                                                Start-Stop.
----------------------------------------------------------------------------------------------------------------

4. Manufacturers' Application of Technology
    Vehicle manufacturers often introduce major product changes
together, as a package. In this manner the manufacturers can optimize
their available resources, including engineering, development,
manufacturing and marketing activities to create a product with
multiple new features. In addition, manufacturers recognize that a
vehicle will need to remain competitive over its intended life, meet
future regulatory requirements, and contribute to a manufacturer's CAFE
requirements. Furthermore, automotive manufacturers are largely focused
on creating vehicle platforms to limit the development of entirely new
vehicles and to realize economies of scale with regard to variable
cost. In very limited cases, manufacturers may implement an individual
technology outside of a vehicle's redesign cycle. In following with
these industry practices, EPA has created a set of vehicle technology
packages that represent the entire light duty fleet.
    EPA has historically allowed manufacturers of new vehicles or
nonroad equipment to phase in available emission control technology
over a number of years. Examples of this are EPA's Tier 2 program for
cars and light trucks and its 2007 and later PM and NO_X
emission standards for heavy-duty vehicles. In both of these rules, the
major modifications expected from the rules were the addition of
exhaust aftertreatment control technologies. Some changes to the engine
were expected as well, but these were not expected to affect engine
size, packaging or performance. The CO₂ reduction
technologies described above potentially involve much more significant
changes to car and light truck designs. Many of the engine technologies
involve changes to the engine block and heads. The transmission
technologies could change the size and shape of the transmission and
thus, packaging. Improvements to aerodynamic drag could involve body
design and therefore, the dies used to produce body panels. Changes of
this sort potentially involve new capital investment and the
obsolescence of existing investment.
    At the same time, vehicle designs are not static, but change in
major ways periodically. The manufacturers' product plans indicate that
vehicles are usually redesigned every 5 years on average. Vehicles also
tend to receive a more modest ``refresh'' between major redesigns, as
discussed above. Because manufacturers are already changing their
tooling, equipment and designs at these times, further changes to
vehicle design at these times involve a minimum of stranded capital
equipment. Thus, the timing of any major technological changes is
projected to coincide with changes that manufacturers would already
tend to be making to their vehicles. This approach effectively avoids
the need to quantify any costs associated with discarding equipment,
tooling, emission and safety certification, etc. when CO₂-
reducing equipment is incorporated into a vehicle.
    This proposed rule affects five years of vehicle production, model
years 2012-2016. Given the now-typical five-year redesign cycle, nearly
all of a manufacturer's vehicles will be redesigned over this period.
However, this assumes that a manufacturer has sufficient lead time to
redesign the first model year affected by this proposed rule with the
requirements of this proposed rule in mind. In fact, the lead time
available for model year 2012 is relatively short. The time between a
likely final rule and the start of 2013 model year production is likely
to be just over two years. At the same time, the manufacturer product
plans indicate that they are planning on introducing many of the
technologies projected to be required by this proposed rule in both
2012 and 2013. In order to account for the relatively short lead time
available prior to the 2012 and 2013 model years, albeit mitigated by
their existing plans, EPA projects that only 85 percent of each
manufacturer's sales will be able to be redesigned with major
CO₂ emission-reducing technologies by the 2016 model year.
Less intrusive technologies can be introduced into essentially all a
manufacturer's sales. This resulted in three levels of technology
penetration caps, by manufacturer. Common technologies (e.g., low
friction lubes, aerodynamic improvements) had a penetration cap of
100%. More advanced powertrain technologies (e.g., stoichiometric GDI,
turbocharging) had a penetration cap of 85%. The most advanced
technologies considered in this analysis (e.g., diesel engines, as well
as IMA, powersplit and 2-mode hybrids) had a 15% penetration cap.
5. How Is EPA Projecting That a Manufacturer Would Decide Between
Options To Improve CO₂ Performance To Meet a Fleet Average Standard?
    There are many ways for a manufacturer to reduce CO₂-
emissions from its vehicles. A manufacturer can choose from a myriad of
CO₂ reducing technologies and can apply one or more of these
technologies to some or all of its vehicles. Thus, for a variety of
levels of CO₂ emission control, there are an almost infinite
number of technology combinations which produce the desired
CO₂ reduction. EPA has created a new vehicle model, the
Optimization Model for Emissions of Greenhouse gases from Automobiles
(OMEGA) in order to make a reasonable estimate of how manufacturers
will add technologies to vehicles in order to meet a fleet-wide
CO₂ emissions level. EPA has described OMEGA's specific
methodologies and algorithms in a memo to the docket for this
rulemaking (Docket EPA-HQ-OAR-2009-0472).
    The OMEGA model utilizes four basic sets of input data. The first
is a description of the vehicle fleet. The key pieces of data required
for each vehicle are its manufacturer, CO₂ emission level,
fuel type, projected sales and footprint. The model also requires that

[[Page 49546]]

each vehicle be assigned to one of the 19 vehicle types, which tells
the model which set of technologies can be applied to that vehicle.
(For a description of how the 19 vehicle types were created, reference
Section III.D.3.) In addition, the degree to which each vehicle already
reflects the effectiveness and cost of each available technology must
also be input. This avoids the situation, for example, where the model
might try to add a basic engine improvement to a current hybrid
vehicle. Except for this type of information, the development of the
required data regarding the reference fleet was described in Section
III.D.1 above and in Chapter 1 of the Draft Joint TSD.
    The second type of input data used by the model is a description of
the technologies available to manufacturers, primarily their cost and
effectiveness. Note that the five vehicle classes are not explicitly
used by the model, rather the costs and effectiveness associated with
each vehicle package is based on the associated class. This information
was described in Sections III.D.2 and III.D.3 above as well as Chapter
3 of the Draft Joint TSD. In all cases, the order of the technologies
or technology packages for a particular vehicle type is determined by
the model user prior to running the model. Several criteria can be used
to develop a reasonable ordering of technologies or packages. These are
described in the Draft Joint TSD.
    The third type of input data describes vehicle operational data,
such as annual scrap rates and mileage accumulation rates, and economic
data, such as fuel prices and discount rates. These estimates are
described in Section II.F above, Section III.H below and Chapter 4 of
the Draft Joint TSD.
    The fourth type of data describes the CO₂ emission
standards being modeled. These include the CO₂ emission
equivalents of the 2011 MY CAFE standards and the proposed
CO₂ standards for 2016. As described in more detail below,
the application of A/C technology is evaluated in a separate analysis
from those technologies which impact CO₂ emissions over the
2-cycle test procedure. Thus, for the percent of vehicles that are
projected to achieve A/C related reductions, the CO₂ credit
associated with the projected use of improved A/C systems is used to
adjust the proposed CO₂ standard which would be applicable
to each manufacturer to develop a target for CO₂ emissions
over the 2-cycle test which is assessed in our OMEGA modeling.
    As mentioned above for the market data input file utilized by
OMEGA, which characterizes the vehicle fleet, our modeling must and
does account for the fact that many 2008 MY vehicles are already
equipped with one or more of the technologies discussed in Section
III.D.2 above. Because of the choice to apply technologies in packages,
and 2008 vehicles are equipped with individual technologies in a wide
variety of combinations, accounting for the presence of specific
technologies in terms of their proportion of package cost and
CO₂ effectiveness requires careful, detailed analysis. The
first step in this analysis is to develop a list of individual
technologies which are either contained in each technology package, or
would supplant the addition of the relevant portion of each technology
package. An example would be a 2008 MY vehicle equipped with variable
valve timing and a 6-speed automatic transmission. The cost and
effectiveness of variable valve timing would be considered to be
already present for any technology packages which included the addition
of variable valve timing or technologies which went beyond this
technology in terms of engine related CO₂ control
efficiency. An example of a technology which supplants several
technologies would be a 2008 MY vehicle which was equipped with a
diesel engine. The effectiveness of this technology would be considered
to be present for technology packages which included improvements to a
gasoline engine, since the resultant gasoline engines have a lower
CO₂ control efficiency than the diesel engine. However, if
these packages which included improvements also included improvements
unrelated to the engine, like transmission improvements, only the
engine related portion of the package already present on the vehicle
would be considered. The transmission related portion of the package's
cost and effectiveness would be allowed to be applied in order to
comply with future CO₂ emission standards.
    The second step in this process is to determine the total cost and
CO₂ effectiveness of the technologies already present and
relevant to each available package. Determining the total cost usually
simply involves adding up the costs of the individual technologies
present. In order to determine the total effectiveness of the
technologies already present on each vehicle, the lumped parameter
model described above is used. Because the specific technologies
present on each 2008 vehicle are known, the applicable synergies and
dis-synergies can be fully accounted for.
    The third step in this process is to divide the total cost and
CO₂ effectiveness values determined in step 2 by the total
cost and CO₂ effectiveness of the relevant technology
packages. These fractions are capped at a value of 1.0 or less, since a
value of 1.0 causes the OMEGA model to not change either the cost or
CO₂ emissions of a vehicle when that technology package is added.
    As described in Section III.D.3 above, technology packages are
applied to groups of vehicles which generally represent a single
vehicle platform and which are equipped with a single engine size
(e.g., compact cars with four cylinder engine produced by Ford). These
groupings are described in Table III.D.1-1. Thus, the fourth step is to
combine the fractions of the cost and effectiveness of each technology
package already present on the individual 2008 vehicles models for each
vehicle grouping. For cost, percentages of each package already present
are combined using a simple sales-weighting procedure, since the cost
of each package is the same for each vehicle in a grouping. For
effectiveness, the individual percentages are combined by weighting
them by both sales and base CO₂ emission level. This
appropriately weights vehicle models with either higher sales or
CO₂ emissions within a grouping. Once again, this process
prevents the model from adding technology which is already present on
vehicles, and thus ensures that the model does not double count
technology effectiveness and cost associated with complying with the
2011 MY CAFE standards and the proposed CO₂ standards.
    Conceptually, the OMEGA model begins by determining the specific
CO₂ emission standard applicable for each manufacturer and
its vehicle class (i.e., car or truck). Since the proposed rule allows
for averaging across a manufacturer's cars and trucks, the model
determines the CO₂ emission standard applicable to each
manufacturer's car and truck sales from the two sets of coefficients
describing the piecewise linear standard functions for cars and trucks
in the inputs, and creates a combined car-truck standard. This combined
standard considers the difference in lifetime VMT of cars and trucks,
as indicated in the proposed regulations which would govern credit
trading between these two vehicle classes. For both the 2011 CAFE and
2016 CO₂ standards, these standards are a function of each
manufacturer's sales of cars and trucks and their footprint values.
When evaluating the 2011 MY CAFE standards, the car-truck trading was
limited to 1.2 mpg. When evaluating the proposed CO₂
standards, the OMEGA model was run only for MY 2016. OMEGA is designed
to evaluate technology addition over a complete

[[Page 49547]]

redesign cycle and 2016 represents the final year of a redesign cycle
starting with the first year of the proposed CO₂ standards,
2012. Estimates of the technology and cost for the interim model years
are developed from the model projections made for 2016. This process is
discussed in Chapter 6 of EPA's DRIA to this proposed rule. When
evaluating the 2016 standards using the OMEGA model, the proposed
CO₂ standard which manufacturers would otherwise have to
meet to account for the anticipated level of A/C credits generated was
adjusted. On an industry wide basis, the projection shows that
manufacturers would generate 11 g/mi of A/C credit in 2016. Thus, the
2016 CO₂ target for the fleet evaluated using OMEGA was 261
g/mi instead of 250 g/mi.
    The cost of the improved A/C systems required to generate the 11 g/
mi credit was estimated separately. This is consistent with our
proposed A/C credit procedures, which would grant manufacturers A/C
credits based on their total use of improved A/C systems, and not on
the increased use of such systems relative to some base model year
fleet. Some manufacturers may already be using improved A/C technology.
However, this represents a small fraction of current vehicle sales. To
the degree that such systems are already being used, EPA is over-
estimating both the cost and benefit of the addition of improved A/C
technology relative to the true reference fleet to a small degree.
    The model then works with one manufacturer at a time to add
technologies until that manufacturer meets its applicable standard. The
OMEGA model can utilize several approaches to determining the order in
which vehicles receive technologies. For this analysis, EPA used a
``manufacturer-based net cost-effectiveness factor'' to rank the
technology packages in the order in which a manufacturer would likely
apply them. Conceptually, this approach estimates the cost of adding
the technology from the manufacturer's perspective and divides it by
the mass of CO₂ the technology will reduce. One component of
the cost of adding a technology is its production cost, as discussed
above. However, it is expected that new vehicle purchasers value
improved fuel economy since it reduces the cost of operating the
vehicle. Typical vehicle purchasers are assumed to value the fuel
savings accrued over the period of time which they will own the
vehicle, which is estimated to be roughly five years. It is also
assumed that consumers discount these savings at the same rate as that
used in the rest of the analysis (3 or 7 percent). Any residual value
of the additional technology which might remain when the vehicle is
sold is not considered. The CO₂ emission reduction is the
change in CO₂ emissions multiplied by the percentage of
vehicles surviving after each year of use multiplied by the annual
miles travelled by age, again discounted to the year of vehicle purchase.
    Given this definition, the higher priority technologies are those
with the lowest manufacturer-based net cost-effectiveness value
(relatively low technology cost or high fuel savings leads to lower
values). Because the order of technology application is set for each
vehicle, the model uses the manufacturer-based net cost-effectiveness
primarily to decide which vehicle receives the next technology
addition. Initially, technology package #1 is the only one
available to any particular vehicle. However, as soon as a vehicle
receives technology package #1, the model considers the
manufacturer-based net cost-effectiveness of technology package
#2 for that vehicle and so on. In general terms, the equation
describing the calculation of manufacturer-based cost effectiveness is
as follows:
[GRAPHIC] [TIFF OMITTED] TP28SE09.013

Where:

ManufCostEff = Manufacturer-Based Cost Effectiveness (in dollars per
kilogram CO₂),
TechCost = Marked up cost of the technology (dollars),
PP = Payback period, or the number of years of vehicle use over
which consumers value fuel savings when evaluating the value of a
new vehicle at time of purchase,
dFS_i = Difference in fuel consumption due to the addition
of technology times fuel price in year i,
dCO₂ = Difference in CO₂ emissions due to the
addition of technology
VMTi = product of annual VMT for a vehicle of age i and the
percentage of vehicles of age i still on the road,
1- Gap = Ratio of onroad fuel economy to two-cycle (FTP/HFET) fuel economy

    EPA describes the technology ranking methodology and manufacturer-
based cost effectiveness metric in greater detail in a technical memo
to the Docket for this proposed rule (Docket EPA-HQ-OAR-2009-0472).
    When calculating the fuel savings, the full retail price of fuel,
including taxes is used. While taxes are not generally included when
calculating the cost or benefits of a regulation, the net cost
component of the manufacturer-based net cost-effectiveness equation is
not a measure of the social cost of this proposal, but a measure of the
private cost, (i.e., a measure of the vehicle purchaser's willingness
to pay more for a vehicle with higher fuel efficiency). Since vehicle
operators pay the full price of fuel, including taxes, they value fuel
costs or savings at this level, and the manufacturers will consider
this when choosing among the technology options.
    This definition of manufacturer-based net cost-effectiveness
ignores any change in the residual value of the vehicle due to the
additional technology when the vehicle is five years old. As discussed
in Chapter 1of the DRIA, based on historic used car pricing, applicable
sales taxes, and insurance, vehicles are worth roughly 23% of their
original cost after five years, discounted to year of vehicle purchase
at 7% per annum. It is reasonable to estimate that the added technology
to improve CO₂ level and fuel economy would retain this same
percentage of value when the vehicle is five years old. However, it is
less clear whether first purchasers, and thus, manufacturers would
consider this residual value when ranking technologies and making
vehicle purchases, respectively. For this proposal, this factor was not
included in our determination of manufacturer-based net cost-
effectiveness in the analyses performed in support of this proposed
rule. Comments are requested on the benefit of including an increase

[[Page 49548]]

in the vehicle's residual value after five years in the calculation of
effective cost.
    The values of manufacturer-based net cost-effectiveness for
specific technologies will vary from vehicle to vehicle, often
substantially. This occurs for three reasons. First, both the cost and
fuel-saving component cost, ownership fuel-savings, and lifetime
CO₂ effectiveness of a specific technology all vary by the
type of vehicle or engine to which it is being applied (e.g., small car
versus large truck, or 4-cylinder versus 8-cylinder engine). Second,
the effectiveness of a specific technology often depends on the
presence of other technologies already being used on the vehicle (i.e.,
the dis-synergies. Third, the absolute fuel savings and CO₂
reduction of a percentage an incremental reduction in fuel consumption
depends on the CO₂ level of the vehicle prior to adding the
technology. Chapter 1 of the DRIA of this proposed rule contains
further detail on the values of manufacturer-based net cost-
effectiveness for the various technology packages.
    EPA requests comment on the use of manufacturer-based net cost-
effectiveness to rank CO₂ emission reduction technologies in
the context of evaluating alternative fleet average standards for this
rule. EPA believes this manufacturer-based net cost-effectiveness
metric is appropriate for ranking technology in this proposed program
because it considers effectiveness values that may vary widely among
technology packages when determining the order of technology addition.
Comments are requested on this option and on any others thought to be
appropriate.
6. Why Are the Proposed CO₂ Standards Feasible?
    The finding that the proposed standards would be technically
feasible is based primarily on two factors. One is the level of
technology needed to meet the proposed standards. The other is the cost
of this technology. The focus is on the proposed standards for 2016, as
this is the most stringent standard and requires the most extensive use
of technology.
    With respect to the level of technology required to meet the
standards, EPA established technology penetration caps. As described in
Section III.D.4, EPA used two constraints to limit the model's
application of technology by manufacturer. The first was the
application of common fuel economy enablers such as low rolling
resistance tires and transmission logic changes. These were allowed to
be used on all vehicles and hence had no penetration cap. The second
constraint was applied to most other technologies and limited their
application to 85% with the exception of the most advanced technologies
(e.g., powersplit and 2-mode hybrids) whose application was limited to 15%.
    EPA used the OMEGA model to project the technology (and resultant
cost) required for manufacturers to meet the current 2011 MY CAFE
standards and the proposed 2016 MY CO₂ emission standards.
Both sets of standards were evaluated using the OMEGA model. The 2011
MY CAFE standards were applied to cars and trucks separately with the
transfer of credits from one category to the other allowed up to an
increase in fuel economy of 1.0 mpg. Chrysler, Ford and General Motors
are assumed to utilize FFV credits up to the maximum of 1.2 mpg for
both their car and truck sales. Nissan is assumed to utilize FFV
credits up to the maximum of 1.2 mpg for only their truck sales. The
use of any banked credits from previous model years was not considered.
The modification of the reference fleet to comply with the 2011 CAFE
standards through the application of technology by the OMEGA model is
the final step in creating the final reference fleet. This final reference
fleet forms the basis for comparison for the model year 2016 standards.
    Table III.D.6-1 shows the usage level of selected technologies in
the 2008 vehicles coupled with 2016 sales prior to projecting their
compliance with the 2011 MY CAFE standards. These technologies include
converting port fuel-injected gasoline engines to direct injection
(GDI), adding the ability to deactivate certain engine cylinders during
low load operation (Deac), adding a turbocharger and downsizing the
engine (Turbo), increasing the number of transmission speeds to 6 or,
converting automatic transmissions to dual-clutch automated manual
transmissions (Dual-Clutch Trans), adding 42 volt start-stop capability
(Start-Stop), and converting a vehicle to a intermediate or strong
hybrid design. This last category includes three current hybrid
designs: integrated motor assist (IMA), power-split (PS) and 2-mode hybrids.

                              Table III.D.6-1--Penetration of Technology in 2008 Vehicles With 2016 Sales: Cars and Trucks
                                                                   [Percent of sales]
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                                       6 Speed or  Dual clutch
                                                      GDI       GDI+ deac    GDI+ turbo     Diesel      CV trans      trans      Start-stop     Hybrid
--------------------------------------------------------------------------------------------------------------------------------------------------------
BMW.............................................          6.7          0.0          0.0          0.0         98.8          0.8          0.0          0.1
Chrysler........................................          0.0          0.0          0.0          0.0         27.9          0.0          0.0          0.0
Daimler.........................................          6.2          0.0          0.0          6.2         74.7         11.4          0.0          0.0
Ford............................................          0.6          0.0          0.0          0.0         28.1          0.0          0.0          0.0
General Motors..................................          3.3          0.0          0.0          0.0         13.7          0.0          0.1          0.1
Honda...........................................          1.2          0.0          0.0          0.0          4.2          0.0          0.0          2.1
Hyundai.........................................          0.0          0.0          0.0          0.0          4.9          0.0          0.0          0.0
Kia.............................................          0.0          0.0          0.0          0.0          0.9          0.0          0.0          0.0
Mazda...........................................         11.8          0.0          0.0          0.0         37.1          0.0          0.0          0.0
Mitsubishi......................................          0.0          0.0          0.0          0.0         76.1          0.0          0.0          0.1
Nissan..........................................         17.7          0.0          0.0          0.0         33.3          0.0          0.0          0.0
Porsche.........................................          0.0          0.0          0.0          0.0          3.9          0.0          0.0          0.0
Subaru..........................................          0.0          0.0          0.0          0.0         29.0          0.0          0.0          0.0
Suzuki..........................................          0.0          0.0          0.0          0.0        100.0          0.0          0.0          0.0
Tata............................................          0.0          0.0          0.0          0.0          0.0          0.0          0.0          0.0
Toyota..........................................          7.5          0.0          0.0          0.0         30.6          0.0          0.0         12.8
Volkswagen......................................         52.2          0.0          0.0          0.1         82.8         10.9          0.0          0.0
Overall.........................................          6.4          0.0          0.0          0.1         27.1          0.6          0.0          2.8
--------------------------------------------------------------------------------------------------------------------------------------------------------

[[Page 49549]]

    As can be seen, all of these technologies except for the direct
injection gasoline engines with either cylinder deactivation or
turbocharging and downsizing, were already being used on some 2008 MY
vehicles. High speed transmissions were the most prevalent, with some
manufacturers (e.g., BMW, Suzuki) using them on essentially all of
their vehicles. Both Daimler and VW equip many of their vehicles with
automated manual transmissions, while VW makes extensive use of direct
injection gasoline engine technology. Toyota has converted a
significant percentage of its 2008 vehicles to strong hybrid design.
    Table III.D.6-2 shows the usage level of the same technologies in
the reference case fleet after projecting their compliance with the
2011 MY CAFE standards. Except for mass reduction, the figures shown
represent the percentages of each manufacturer's sales which are
projected to be equipped with the indicated technology. For mass
reduction, the overall mass reduction projected for that manufacturer's
sales is shown. The last row in Table III.D.6-2 shows the increase in
projected technology penetration due to compliance with the 2011 MY
CAFE standards. The results of DOT's Volpe Modeling were used to
project that all manufacturers would comply with the 2011 MY standards
in 2016 without the need to pay fines, with one exception. This
exception was Porsche in the case of their car fleet. When projecting
Porsche's compliance with the 2011 MY CAFE standard for cars, the car
fleet was assumed to achieve a CO₂ emission level of 293.2
g/mi instead of the required 285.2 g/mi level (30.3 mpg instead of 31.2 mpg).

                         Table III.D.6-2--Penetration of Technology Under 2011 MY CAFE Standards in 2016 Sales: Cars and Trucks
                                                                   [Percent of sales]
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                                                                                 Mass
                                                      GDI       GDI+ deac    GDI+ turbo   6 Speed or  Dual clutch   Start-stop     Hybrid     reduction
                                                                                           CV trans      trans                                (percent)
--------------------------------------------------------------------------------------------------------------------------------------------------------
BMW.............................................          7.3         11.1          0.0         86.3         11.1         11.1          0.1          0.5
Chrysler........................................          0.0          0.0          0.0         27.9          0.0          0.0          0.0          0.0
Daimler.........................................         16.4         10.3         14.3         45.8         36.0         24.6          0.0          0.9
Ford............................................          0.6          0.0          0.0         28.1          0.0          0.0          0.0          0.0
General Motors..................................          3.3          0.0          0.0         13.7          0.0          0.1          0.1          0.0
Honda...........................................          1.2          0.0          0.0          4.2          0.0          0.0          2.1          0.0
Hyundai.........................................          0.0          0.0          0.0          4.9          0.0          0.0          0.0          0.0
Kia.............................................          0.0          0.0          0.0          0.9          0.0          0.0          0.0          0.0
Mazda...........................................         11.8          0.0          0.0         37.1          0.0          0.0          0.0          0.0
Mitsubishi......................................          0.0          2.2          0.0         76.0          2.2          2.2          0.1          0.0
Nissan..........................................         17.7          0.0          0.0         33.3          0.0          0.0          0.0          0.0
Porsche.........................................          0.0         25.0         23.2          0.0         48.2         37.1          0.0          1.2
Subaru..........................................          0.0          0.0          0.0         29.0          0.0          0.0          0.0          0.0
Suzuki..........................................          4.5          0.0          0.0        100.0          0.0          0.0          0.0          0.0
Tata............................................         14.5         60.9          0.0         14.5         60.9         60.9          0.0          2.6
Toyota..........................................          7.5          0.0          0.0         30.6          0.0          0.0         12.8          0.0
Volkswagen......................................         51.2          6.9         11.8         60.8         29.6         18.7          0.0          0.3
Overall.........................................          6.7          1.2          0.8         25.4          2.6          2.0          2.8          0.1
Increase over 2008 MY...........................          0.3          1.2          0.8         -1.7          2.0          2.0          0.0          0.0
--------------------------------------------------------------------------------------------------------------------------------------------------------

    As can be seen, the 2011 MY CAFE standards, when evaluated on an
industry wide basis, require only a modest increase in the use of these
technologies. Higher speed automatic transmission use actually
decreases due to conversion of these units to more efficient designs
such as automated manual transmissions and hybrids. However, the impact
of the 2011 MY CAFE standards is much greater on selected
manufacturers, particularly BMW, Daimler, Porsche, Tata (Jaguar/Land
Rover) and VW. All of these manufacturers are projected to increase
their use of advanced direct injection gasoline engine technology,
advanced transmission technology, and start-stop technology. It should
be noted that these manufacturers have traditionally paid fines under
the CAFE program. However, with higher fuel prices and the lead-time
available by 2016, these manufacturers would likely find it in their
best interest to improve their fuel economy levels instead of
continuing to pay fines (again with the exception of Porsche cars).
While not shown, no gasoline engines were projected to be converted to
diesel technology.
    This 2008 baseline fleet, modified to meet 2011 standards, becomes
our ``reference'' case. This is the fleet by which the control program
(or 2016 rule) will be compared. Thus, it is also the fleet that would
be assumed to exist in the absence of this rule. No air conditioning
improvements are assumed for model year 2011 vehicles. The average
CO₂ emission levels of this reference fleet vary slightly
from 2012-2016 due to small changes in the vehicle sales by market
segments and manufacturer. CO₂ emissions from cars range
from 282-284 g/mi, while those from trucks range from 382-384 g/mi.
CO₂ emissions from the combined fleet range from 316-320.
These estimates are described in greater detail in Section 5.3.2.2 of the DRIA.
    Conceptually, both EPA and NHTSA perform the same projection in
order to develop their respective reference fleets. However, because
the two agencies use two different models to modify the baseline fleet
to meet the 2011 CAFE standards, the technology added will be slightly
different. The differences, however, are small since most manufacturers
do not require a lot of additional technology to meet the 2011 standards.
    EPA then used the OMEGA model once again to project the level of
technology needed to meet the proposed 2016 CO₂ emission
standards. Using the results of the OMEGA model, every manufacturer was
projected to be able to meet the proposed 2016 standards with the
technology described above except for four: BMW, VW, Porsche and Tata
due to the OMEGA cap on technology penetration by manufacturer. For
these manufacturers, the results presented below are those with the
fully allowable

[[Page 49550]]

application of technology and not for the technology projected to
enable compliance with the proposed standards. Described below are a
number of potential feasible solutions for how these companies can
achieve compliance. The overall level of technology needed to meet the
proposed 2016 standards is shown in Table III.D.6-3. As discussed
above, all manufacturers are projected to improve the air conditioning
systems on 85% of their 2016 sales.

                               Table III.D.6-3--Penetration of Technology for Proposed 2016 CO2 Standards: Cars and Trucks
                                                                   [Percent of sales]
--------------------------------------------------------------------------------------------------------------------------------------------------------
                                                                                           6 Speed    Dual clutch                                Mass
                                                      GDI       GDI+ deac    GDI+ turbo   auto trans     trans      Start-stop     Hybrid     reduction
--------------------------------------------------------------------------------------------------------------------------------------------------------
BMW.............................................            4           35           47           15           71           71           14            5
Chrysler........................................           51           28            3           37           51           51            0            6
Daimler.........................................            3           44           39           11           73           72           13            5
Ford............................................           29           39           13           19           67           67            0            6
General Motors..................................           34           26            7           13           55           55            0            5
Honda...........................................           24            1            2           10           22           22            2            2
Hyundai.........................................           28            3           14            3           43           43            0            3
Kia.............................................           37            0            5            7           35           35            0            3
Mazda...........................................           54            2           16           31           43           43            0            4
Mitsubishi......................................           65            2            7           22           66           66            0            6
Nissan..........................................           29           26            5           34           57           56            1            5
Porsche.........................................            7           36           49           10           70           70           15            4
Subaru..........................................           46            4           14            0           64           51            0            4
Suzuki..........................................           66            5            8            9           69           69            0            4
Tata............................................            4           81            0           14           70           70           15            6
Toyota..........................................           37            2            0           30           33           16           13            2
Volkswagen......................................            9           26           58           12           72           70           15            4
Overall.........................................           30           18           10           19           49           45            4            4
Increase over 2011 CAFE.........................           24           17            9           -7           46           43            1            4
--------------------------------------------------------------------------------------------------------------------------------------------------------

    As can be seen, the overall average reduction in vehicle weight is
projected to be 4%. This reduction varies across the two vehicle
classes and vehicle base weight. For cars below 2,950 pounds curb
weight, the average reduction is 2.3% (62 pounds), while the average
was 4.4% (154 pounds) for cars above 2,950 curb weight. For trucks
below 3,850 pounds curb weight, the average reduction is 3.5% (119
pounds), while it was 4.5% (215 pounds) for trucks above 3,850 curb
weight. Splitting trucks at a higher weight, for trucks below 5,000
pounds curb weight, the average reduction is 3.3% (140 pounds), while
it was 6.7% (352 pounds) for trucks above 5,000 curb weight.
    The levels of requisite technologies differ significantly across
the various manufacturers. Therefore, several analyses were performed
to ascertain the cause. Because the baseline case fleet consists of
2008 MY vehicle designs, these analyses were focused on these vehicles,
their technology and their CO₂ emission levels.
    Comparing CO₂ emissions across manufacturers is not a
simple task. In addition to widely varying vehicle styles, designs, and
sizes, manufacturers have implemented fuel efficient technologies to
varying degrees, as indicated in Table III.D.6-1. The projected levels
of requisite technology to enable compliance with the proposed 2016
standards shown in Table III.D.6-3 account for two of the major factors
which can affect CO₂ emissions: (1) Level of technology
already being utilized and (2) vehicle size, as represented by footprint.
    For example, the fuel economy of a manufacturer's 2008 vehicles may
be relatively high because of the use of advanced technology. This is
the case with Toyota's high sales of their Prius hybrid. However, the
presence of this technology in a 2008 vehicle eliminates the ability to
significantly reduce CO₂ further through the use of this
technology. In the extreme, if a manufacturer were to hybridize a high
level of its sales in 2016, it doesn't matter whether this technology
was present in 2008 or whether it would be added in order to comply
with the standards. The final level of hybrid technology would be the
same. Thus, the level at which technology is present in 2008 vehicles
does not explain the difference in requisite technology levels shown in
Table III.D.6-3.
    Similarly, the proposed CO₂ emission standards adjust
the required CO₂ level according to a vehicle's footprint,
requiring lower absolute emission levels from smaller vehicles. Thus,
just because a manufacturer produces larger vehicles than another
manufacturer does not explain the differences seen in Table III.D.6-3.
    In order to remove these two factors from our comparison, the EPA
lumped parameter model described above was used to estimate the degree
to which technology present on each 2008 MY vehicle in our reference
fleet was improving fuel efficiency. The effect of this technology was
removed and each vehicle's CO₂ emissions were estimated as
if it utilized no additional fuel efficiency technology beyond the
baseline. The differences in vehicle size were accounted for by
determining the difference between the sales-weighted average of each
manufacturer's ``no technology'' CO₂ levels to their
required CO₂ emission level under the proposed 2016
standards. The industry-wide difference was subtracted from each
manufacturer's value to highlight which manufacturers had lower and
higher than average ``no technology'' emissions. The results are shown
in Figure III.D.6-1.
BILLING CODE 4910-59-P

[[Page 49551]]
[GRAPHIC] [TIFF OMITTED] TP28SE09.014
[[Page 49552]]

    As can be seen in Table III.D.6-3 the manufacturers projected to
require the greatest levels of technology also show the highest offsets
relative to the industry. The greatest offset shown in Figure III.D.6-1
is for Tata's trucks (Land Rover). These vehicles are estimated to have
100 g/mi greater CO₂ emissions than the average 2008 MY
truck after accounting for differences in the use of fuel saving
technology and footprint. The lowest adjustment is for Subaru's trucks,
which have 50 g/mi CO₂ lower emissions than the average truck.
    While this comparison confirms the differences in the technology
penetrations shown in Table III.D.6-3, it does not yet explain why
these differences exist. Two well known factors affecting vehicle fuel
efficiency are vehicle weight and performance. The footprint-based form
of the proposed CO₂ standard accounts for most of the
difference in vehicle weight seen in the 2008 MY fleet. However, even
at the same footprint, vehicles can have varying weights. Higher
performing vehicles also tend to have higher CO₂ emissions
over the two-cycle test procedure. So manufacturers with higher average
performance levels will tend to have higher average CO₂
emissions for any given footprint.
    The impact of these two factors on each manufacturer's ``no
technology'' CO₂ emissions was estimated. First, the ``no
technology'' CO₂ emissions levels were statistically
analyzed to determine the average impact of weight and the ratio of
horsepower to weight on CO₂ emissions. Both factors were
found to be statistically significant at the 95 percent confidence
level. Together, they explained over 80 percent of the variability in
vehicles' CO₂ emissions for cars and over 70 percent for
trucks. These relationships were then used to adjust each vehicle's
``no technology'' CO₂ emissions to the average weight for
its footprint value and to the average horsepower to weight ratio of
either the car or truck fleet. The comparison was repeated as shown in
Figure III.D.6-1. The results are shown in Figure III.D.6-2.

[[Continued on page 49553]]
Notices For
2009
2008
2007
2006
2005
2004
2003
2002
2001
2000
1999
1998
1997
1996
1995
1994
Proposed Rulemaking To Establish Light-Duty Vehicle Greenhouse Gas Emission Standards and Corporate Average Fuel Economy Standards

Local Navigation
