Regulating Greenhouse Gas Emissions Under the Clean Air Act

Note: EPA no longer updates this information, but it may be useful as a reference or resource.
 
PDF Version (50 pp, 1464K, About PDF)

[Federal Register: July 30, 2008 (Volume 73, Number 147)]
[Proposed Rules]
[Page 44403-44452]
From the Federal Register Online via GPO Access [wais.access.gpo.gov]
[DOCID:fr30jy08-37]

Regulating Greenhouse Gas Emissions Under the Clean Air Act

[[Continued from page 44402]]

[[Page 44403]]

their emissions. For example, some electricity is generated with low or
no CO2 emitting energy technologies, particularly non-fossil options
such as nuclear, hydroelectric, or geothermal energy. However, over
half of the electricity in the U.S. is generated by burning coal,
accounting for 94% of all coal consumed for energy in the U.S. in 2006.
    Transportation Sector: The transportation sector includes
automobiles, airplanes, railroads and a variety of other sources.
Transportation activities (excluding international bunker fuels)
accounted for approximately 28% of all GHG emissions in 2006, primarily
through the combustion of fossil fuels.\38\ Virtually all of the energy
consumed in this end-use sector came from petroleum products. Over 60%
of the CO2 emissions resulted from gasoline consumption for personal
vehicle use.
---------------------------------------------------------------------------

    \38\ International bunker fuels are used in aviation and marine
trips between countries.
---------------------------------------------------------------------------

    Industrial Sector: The industrial sector includes a wide variety of
facilities engaged in the production and sale of goods. The largest
share of emissions from industrial facilities comes from the combustion
of fossil fuels. Emissions of CO2 and other GHGs from U.S. industry
also occur as a result of specialized manufacturing processes (e.g.,
calcination of limestone in cement manufacturing). The largest emitting
industries tend to be the most energy intensive: Iron and steel,
refining, cement, lime, chemical manufacturing, etc. Overall, 19.4% of
total U.S. GHG emissions came from the industrial sector in 2006.
    Residential and Commercial Sectors: These two sectors directly emit
GHGs primarily through operation and maintenance of buildings (i.e.,
homes, offices, universities, etc.). The residential and commercial
end-use sectors accounted for 4.8 and 5.6% of total emissions,
respectively, with CO2 emissions from consumption of natural gas and
petroleum for heating and cooking making up the largest share.
    Agriculture Sector: The agriculture sector includes all activities
related to cultivating soil, producing crops, and raising livestock.
Agricultural GHG emissions result from a variety of processes,
including: Enteric fermentation in domestic livestock, livestock manure
management, rice cultivation, agricultural soil management, and field
burning of agricultural residues. Methane and N2O are the primary GHGs
emitted by agricultural activities.\39\ In 2006, agriculture emission
sources were responsible for 6.4% of total U.S. GHG emissions.
---------------------------------------------------------------------------

    \39\ Agricultural soils also emit CO2 and sequester carbon. The
fluxes are discussed under the Land-Use, Land-Use Change and
Forestry section because of the integrated nature of methodological
approaches to the carbon cycle, and international reporting
conventions.
---------------------------------------------------------------------------

    Land Use, Land-Use Change, and Forestry: Land use is not an
economic sector per se but affects the natural carbon cycle in ways
that lead to GHG emissions and sinks. Included in this category are
emissions and sequestration of CO2 from activities such as
deforestation, afforestation, forest management and management of
agricultural soils. Emissions and sequestration depend on local
conditions, but overall land use in the U.S. was a net sink in 2006
equivalent to 12.5% of total GHG emissions.
BILLING CODE 6560-50-P

[[Page 44404]]

[GRAPHIC] [TIFF OMITTED] TP30JY08.027

C. Advancing Technology

    President Bush, the IPCC, and many other private and public groups
have spotlighted the critical importance of technology to reducing GHG
emissions and the risks of climate change. International, U.S., and
private studies have identified a broad range of potential strategies
that can reduce emissions from diverse economic sectors. Many
strategies, such as increasing energy efficiency and conservation and
employing hybrid and diesel vehicle technologies, are available today.
There is also broad consensus that for many sectors of the economy new
technologies will be

[[Page 44405]]

needed to achieve deep reductions in GHG emissions at less cost than
today's technologies alone can achieve.
    In developing potential CAA (or other) controls, one important
question is the extent to which needed technological development can be
expected to occur as a result of market forces alone (e.g., as a result
of increasing prices for oil and other fossil fuels), and the extent to
which government or other action may be needed to spur development.
There are several different pathways for technological change,
including investment in research and development (private and public),
spillovers from research and development in other sectors (e.g.,
advances in computing made hybrid vehicles possible), learning by doing
(i.e., efficiency gains through repetition), and scale economies (i.e.,
aggregate cost reductions from improved process efficiencies). As
further discussed later in this section, market-based incentives that
establish a price (directly or indirectly through a limit) for carbon
and/or other GHGs could continuously spur technological innovation that
could lower the cost of reducing emissions. However, even with such a
policy, markets tend to under-invest in development of new technologies
when investors can only capture a portion of the returns. This is
particularly true at the initial stages of research and development
when risks are high and market potential is not evident. In such cases,
policies to encourage the development and diffusion of technologies
that are complements to pollution control policies may be
warranted.\40\
---------------------------------------------------------------------------

    \40\ Economic Report of the President, February 2007.
---------------------------------------------------------------------------

    This section draws insights from IPCC and other reports on
available and needed technologies. In later sections of this notice, we
explain each potentially applicable CAA provision and consider the
extent to which that provision authorizes regulatory actions and
approaches that could spur needed technology development.
1. The Role of Existing and New Technology in Addressing Climate Change
    The 2007 IPCC report on mitigation of climate change examined the
availability of current technologies and the need for new technologies
to mitigate climate change.\41\ Among its conclusions, the IPCC states:
---------------------------------------------------------------------------

    \41\ IPCC, 2007, ``Climate Change 2007: Mitigation. Contribution
of Working Group III to the Fourth Assessment Report of the
Intergovernmental Panel on Climate Change,'' [B. Metz, O.R.
Davidson, P.R. Bosch, R. Dave, L.A. Meyers (eds)], Cambridge
University Press, Cambridge, United Kingdom and New York, NY.

     The range of stabilization levels assessed [by the
IPCC] can be achieved by deployment of a portfolio of technologies
that are currently available and those that are expected to be
commercialized in coming decades. This assumes that appropriate and
effective incentives are in place for development, acquisition,
deployment and diffusion of technologies and for addressing related
barriers.\42\
---------------------------------------------------------------------------

    \42\ Ibid, ``Summary for Policymakers,'' p. 25.

    According to one study, five groups of strategies that could
substantially reduce emissions between now and 2030 include (1)
improving energy efficiency in buildings and appliances; (2) increasing
fuel efficiency and reducing GHG emissions from vehicles and the carbon
intensity of transportation fuels; (3) industrial equipment upgrades
and process changes to improve energy efficiency; (4) increasing forest
stocks and improving soil management practices; and (5) reducing carbon
emissions from electric power production through a shift toward
renewable energy, expanded nuclear capacity, improved power plant
efficiency, and use of carbon capture and storage technology on coal-
fired generation.\43\ (Note that EPA is not rank-ordering these
technologies by their relative cost effectiveness.) As noted elsewhere
in this notice, there is federal regulatory or research and development
activity ongoing in most of these areas.
---------------------------------------------------------------------------

    \43\ See McKinsey & Company, ``Reducing U.S. Greenhouse Gas
Emissions: How Much at What Cost?'', U.S. Greenhouse Gas Abatement
Mapping Initiative, Executive Report, December 2007. This study
performed an economic assessment of potential control methods based
on a ``bottom-up'' partial equilibrium model, which does not account
for interactions among economic sectors. Bottom-up models include
many more specific technologies than ``top-down'' general
equilibrium models, which account for cross-sector interactions.
---------------------------------------------------------------------------

    Many energy efficiency technologies exist that appear to be
extremely cost-effective in reducing fuel costs compared to other
alternatives. However, they have yet to be adopted as widely as
expected because of market barriers. Such barriers include lack of
knowledge or confidence in the technology by potential users,
uncertainty in the return on investment (potentially due to uncertainty
in either input prices or output prices), concerns about effects of
energy efficiency technologies on the quality of inputs or outputs,
size of the initial capital investment (coupled with potential
liquidity constraints), and requirements for specialized human capital
investments. Some of these costs are lower in larger firms, due to the
increased availability of financial resources and human capital.\44\
Vendor and other projections of cost-savings for energy efficiency
technologies are often based on average pay-back and thus do not
reflect differences among firms that can affect the costs and benefits
of these technologies and therefore the likelihood of adoption. Over
time, as firms gain more experience with these technologies, the rate
of adoption will likely increase if significant cost-savings are
realized by early adopters.
---------------------------------------------------------------------------

    \44\ Pizer, et al., ``Technology Adoption and Aggregate Energy
Efficiency,'' December 2002, December 2002 Resources for the Future
Discussion Paper 02-52.
---------------------------------------------------------------------------

    The IPCC report on mitigation identified technologies that are
currently available and additional technologies that are expected to be
commercialized by 2030, as shown in the following table.\45\ These
include technologies and practices in the energy supply,
transportation, buildings, industry, agriculture, forest, and waste
sectors:
---------------------------------------------------------------------------

    \45\ IPCC 2007, ``Summary for Policymakers,'' p. 14. Figure III-
3

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

[[Page 44406]]

[GRAPHIC] [TIFF OMITTED] TP30JY08.028

    How much any of the mitigation strategies identified by these
studies would actually be deployed to address climate change is an open
question. It is possible that unanticipated technologies could play a
significant role in reducing emissions. The point of these studies is
to illustrate that potentially feasible technologies exist that could
be employed to mitigate GHG emissions, not to predict the precise role
they will play or to suggest sectors or methods for regulation. The
particular policies pursued by governments, including the U.S. under
the CAA or other authorities, will influence the way in which these
technologies are deployed as well as incentives for developing and
deploying new technologies.
2. Federal Climate Change Technology Program
    The U.S. government is investing in a diverse portfolio of
technologies with

[[Page 44407]]

the potential to yield substantial reductions in emissions of GHGs. The
Climate Change Technology Program (CCTP) is a multi-agency planning and
coordination entity that assists the government in carrying out the
President's National Climate Change Technology Initiative. Managed by
the Department of Energy, the program is organized around five
technology areas for which working groups were established. EPA
participates in all of the working groups and chairs the group focused
on non-CO₂ GHGs.
    The CCTP strategic plan, released in September 2006, provides
strategic direction and organizes approximately $3 billion in federal
spending for climate change-related technology research, development,
demonstration, and deployment.\46\ The plan sets six complementary
goals, including five aimed at developing technologies to:
---------------------------------------------------------------------------

    \46\ U.S. Climate Change Technology Program Strategic Plan,
September 2006; http://www.climatetechnology.gov/stratplan/final/
index.htm.
---------------------------------------------------------------------------

     Reduce emissions from energy end-use and infrastructure;
     Reduce emissions from energy supply, particularly through
development and commercialization of no- or low-emission technologies;
     Capture, store and sequester CO₂;
     Reduce emissions of non-CO₂ GHGs; and
     Enhance the measurement and monitoring of CO₂
emissions.

The first four of these goals focus on GHG emissions reduction
technologies, and the fifth addresses a key need for developing
comprehensive GHG control strategies. The sixth CCTP goal is to
strengthen the contributions of basic science to climate change
technology development.
3. Potential for CAA Regulation to Encourage Technology Development
    Past EPA efforts to reduce air pollution under the CAA demonstrate
that incentives created by regulation can help encourage technology
development and deployment. As noted in a recent EPA regulatory
analysis, the history of the CAA provides many examples in which
technological innovation and ``learning by doing'' have made it
possible to achieve greater emissions reductions than had been feasible
earlier, or have reduced the costs of emission control in relation to
original estimates.\47\ Among the examples are motor vehicle emission
controls, diesel fuel and engine standards to reduce NO_X and
particulate matter emissions, engine idle-reduction technologies,
selective catalytic reduction and ultra-low NO_X burners for
NO_X emissions, high-efficiency scrubbers for SO₂
emissions from boilers, CFC-free air conditioners and refrigerators,
low or zero VOC paints, and idle-reduction technologies for
engines.\48\
---------------------------------------------------------------------------

    \47\ See section 5.4 of Final Ozone NAAQS Regulatory Impact
Analysis, March 2008, EPA-HQ-OAR-2007-0225. The RIA is available at
http://www.epa.gov/ttn/ecas/ria.html#ria2007.
    \48\ Ibid.
---------------------------------------------------------------------------

    One of the issues raised by potential CAA regulation of GHGs is
whether the CAA can help spur needed technological development for
reducing GHG emissions and the costs of those reductions. The
regulatory authorities in the CAA vary in their potential for
encouraging new technology. As discussed later in this notice, some
provisions offer little flexibility in standard-setting criteria,
emission control methods, compliance deadlines and potential for
market-oriented regulation. Other provisions offer more potential to
encourage new technology through market incentives or to establish
standards based on anticipated advances in technology. EPA requests
comment on the extent to which various CAA provisions could be used to
help spur technological development, and on the need for federally
conducted or funded research to promote technological development.

D. Relationship to Traditional Air Pollutants and Air Pollution
Controls

    An issue for any regulation of GHGs under the CAA or other
statutory authority is how a GHG control program would and should
interact with existing air quality management programs. This section
describes the relationships between climate change and air quality and
between GHG emissions and traditional air pollution control programs.
As explained below, those relationships suggest the need for integrated
approaches to climate change mitigation and air quality protection.
Differences between GHGs and traditional air pollutants should also be
taken into account in considering how CAA authorities could be employed
for GHG regulation.
1. Connections Between Climate Change and Air Quality Issues
    Climate change affects some types of air pollution, and some
traditional air pollutants affect climate. According to the IPCC,
climate change can be expected to influence the concentration and
distribution of air pollutants through a variety of direct and indirect
processes. In its recent review of the NAAQS for ozone, EPA examined
how climate change can increase ozone levels and how ozone, itself a
GHG, can contribute to climate change. Similarly, in its reviews of the
NAAQS for particulate matter, the Agency examined the extent to which
some particles help absorb solar energy in the earth's atmosphere and
others help reflect it back to space.\49\ How EPA regulates those
pollutants under the CAA is potentially part of an overall strategy for
addressing climate change, and how GHGs are regulated is potentially an
important component of protecting air quality. For example, it is
likely to become more difficult and expensive to attain the ozone NAAQS
in a future, warmer climate.
---------------------------------------------------------------------------

    \49\ EPA did not have adequate information in these reviews for
impacts on climate change to change the Agency's decision on whether
or how to revise the standards. See, e.g., 71 FR 61144, 61209-10
(October 17, 2006) (PM NAAQS review).
---------------------------------------------------------------------------

    Most of the largest emitters of GHGs are also large emitters of
traditional air pollutants and therefore are already regulated under
the CAA. The electricity generation, transportation and industrial
sectors, the three largest contributors to GHG emissions in the U.S.,
are subject to CAA controls to help meet NAAQS, control acid rain, and
reduce exposures to toxic emissions. Some manufacturers of the GHGs
that are fluorinated gases are subject to CAA regulations for
protection of the stratospheric ozone layer.
    Many measures for controlling GHG emissions also contribute to
reductions in traditional air pollutants, and some measures for
controlling traditional air pollutants result in reductions in
GHGs.\50\ Co-benefits from reduced air pollution as a result of actions
to reduce GHG emissions can be substantial.\51\ In general, fossil fuel
combustion results in emissions not only of CO₂ but also of
many traditional air pollutants, including SO₂,
NO_X, CO and various toxic air pollutants. For many types of
sources, to the extent fossil fuel combustion is reduced, emissions of
all those pollutants are reduced as well. Some control measures reduce
GHGs and traditional air pollutants, including leak detection and fuel
switching. However, some measures for controlling traditional air
pollutants increase GHGs, and some measures for controlling GHGs may
increase traditional air pollutants. For example, controls to decrease
SO₂ emissions from industrial sources require energy to
operate and result in reduced process efficiencies and increases in
GHGs, and changing

[[Page 44408]]

the composition of transportation fuels to reduce GHGs may affect
traditional air pollutant emissions.
---------------------------------------------------------------------------

    \50\ EPA, OAP, Clean Energy-Environmental Guide to Act, http://
www.epa.gov/cleanenergy/documents/gta/guide_action_full.pdf.
    \51\ IPCC, 2007, Working Group III, Summary for Policymakers.
---------------------------------------------------------------------------

    By considering policies for addressing GHGs and traditional air
pollutants in an integrated manner, EPA and the sectors potentially
subject to GHG emission controls would also have the opportunity to
consider and pursue the most effective way of accomplishing emission
control across pollutants. For example, adoption of some air quality
controls could result in a degree of ``technology lock-in'' that
restricts the ability to implement GHG control technologies for
significant periods of time because of the investment in capital and
other resources to meet the air quality control requirements. Sections
VI and VII below discuss technologies and opportunities for controlling
GHGs in more detail from various sectors, including transportation,
electricity generation, and manufacturing. EPA requests comment on
strategies and technologies for simultaneously achieving reductions in
both traditional air pollutants and GHG emissions.
    In light of the connections between climate change and air quality,
the large overlap of GHG and traditional air pollution sources, and the
potential interactions of GHG and traditional air pollution controls,
it makes sense to consider regulation of GHGs and traditional air
pollutants in an integrated manner. Indeed, the National Academy of
Sciences recommends that development of future policies for air
pollution control be integrated with climate change considerations.\52\
GHG control measures implemented today could have immediate impacts on
air pollution and air quality. Similarly, air pollution controls
implemented today could have near term impacts on GHG emissions and
thus long term impacts on climate. Ideally, any GHG control program
under the Act, or other statutory authority would address GHGs in ways
that simultaneously reduce GHGs and traditional air pollutants as
needed to mitigate climate change and air pollution.\53\
---------------------------------------------------------------------------

    \52\ National Academy of Sciences, ``Radiative Forcing of
Climate Change: Expanding the Concept and Addressing
Uncertainties,'' October 2005.
    \53\ Integration of planning efforts related to air quality,
land use, energy efficiency, and transportation to improve air
quality and reduce GHG emissions is in line with the CAA Advisory
Committee Air Quality Management Subcommittee's Phase II
recommendations (June 2007), and the recommendations of the National
Research Council of the National Academy of Sciences in its January
2004 report, ``Air Quality Management in the United States.'' EPA
has initiated several programs to encourage integrated planning
efforts, including the Sustainable Skylines Initiative, a public-
private partnership to reduce air emissions and promote
sustainability in urban environments, and the Air Quality Management
Plan pilot program for testing a comprehensive, multipollutant
planning approach.
---------------------------------------------------------------------------

2. Issues in Applying CAA Controls to GHGs
    One important issue for regulation of GHGs under some CAA
provisions concerns the emissions thresholds established by the Act for
determining the applicability of those provisions. Several CAA
provisions require stationary sources that emit traditional air
pollutants above specific emission thresholds to comply with certain
requirements. Applying the same thresholds to GHGs could result in
numerous sources, such as space heaters in large residential and
commercial buildings, becoming newly subject to those requirements.
Currently regulated sources could become subject to additional
requirements. This would occur in part because most sources typically
emit CO₂, the predominant GHG, in much larger quantities
than traditional air pollutants. Issues related to threshold levels are
discussed in more detail in Section VII below.
    Other important issues for CAA regulation of GHGs are raised by the
different temporal and spatial scope of GHGs compared to traditional
pollutants. Air pollutants currently regulated under the CAA tend to
have local (a few kilometers) or regional (hundreds to thousands of
kilometers) impacts and relatively short atmospheric lifetimes (days to
a month). Historically, this has meant that EPA could identify and
differentiate between affected and unaffected areas and devise control
strategies appropriate for each area. Controls applied within an area
with high concentrations of traditional air pollutants generally have
been effective in achieving significant reductions in air pollution
concentrations within that area in a relatively short amount of time.
The spatial nature of traditional air pollution also has made it
appropriate to place the primary responsibility for planning controls
on state, tribal, or local governments.
    In the years since the CAA was enacted, we have learned that some
traditional air pollutants (e.g., ozone, particulates and their
precursors) are transported across regions of the country and thus have
geographically broader impacts than individual states can address on
their own. Our control strategies for those pollutants have evolved
accordingly. The Nitrogen Oxides (NO_X) SIP Call Rule and the
Clean Air Interstate Rule (CAIR) are examples of regional control
programs that significantly supplement local control measures. NSPS and
motor vehicle controls are examples of national measures that also help
improve air quality locally and regionally.
    The global nature and effect of GHG emissions raise questions
regarding the suitability of CAA provisions that are designed to
protect local and regional air quality by controlling local and
regional emission sources.\54\ As noted above, GHGs are relatively
evenly distributed throughout the global atmosphere. As a result, the
geographic location of emission sources and reductions are generally
not important to mitigating global climate change. Instead, total GHG
emissions in the U.S. and elsewhere in the world over time determine
cumulative global GHG concentrations, which in turn determine the
extent of climate change. As a result, it will be the total emission
reductions achieved by the U.S. and the other countries of the world
that will determine the extent of climate change mitigation. The global
nature of GHGs suggests that the programmatic and analytical tools used
to address local and regional pollutants under the CAA (e.g., SIPs,
monitoring networks, and models) would need to be adapted to inventory,
analyze, control effectively and evaluate progress in achieving GHG
reductions.
---------------------------------------------------------------------------

    \54\ It should be noted that international transport of ozone
and particulate matter precursors contributes to NAAQS nonattainment
in some areas of the U.S. Nevertheless, most traditional air
pollution problems are largely the result of local and regional
emission sources, while for GHGs, worldwide emissions determine the
extent of the problem.
---------------------------------------------------------------------------

    EPA seeks information about how differences in pollutant
characteristics should inform regulation of these pollutants under the
CAA. EPA also requests comment on the types of effective programs at
all levels (local, regional, national and international) that may be
feasible to design and implement under existing CAA authorities.

E. Relationship to Other Environmental Media

    An effective GHG control program may require application of many
technologies and approaches that may in turn result in increased
discharges to water, generation of solid materials that require
appropriate disposal, or have other impacts to the environment that may
not be addressed under the CAA. Examples of these impacts include the
potential for groundwater contamination from geological

[[Page 44409]]

sequestration of CO₂, the generation of spent sorbent
material from carbon capture systems, or the depletion of water
resources and increased nutrient runoff into surface waters from
increased production of bioenergy feedstocks. EPA and other regulatory
agencies at the tribal, state, and local level may need to respond to
such impacts to prevent or minimize their impact to the environment and
public health under authorities other than the CAA.
    Since the nature and extent of these impacts would depend upon the
technologies and approaches that are implemented under a GHG control
program, an important consideration in designing GHG controls is
minimizing or mitigating such impacts EPA seeks comment on how
different regulatory approaches to GHG control under the CAA could
result in environmental impacts to water or land that could require
response under the CAA or EPA's other legislative authorities.

F. Other Key Policy and Economic Considerations for Selecting
Regulatory Approaches

    This section identifies general policy considerations relevant to
developing potential regulatory approaches for controlling GHG
emissions. In developing approaches under the CAA, EPA must first
consider the Act's provisions as well as the Agency's previous
interpretation of the provisions and relevant and controlling court
opinions. Provisions of the CAA vary in terms of the degree of
flexibility afforded EPA in designing implementing regulations under
the Act. To the extent particular provisions permit, EPA believes the
following considerations should guide its choice among available
regulatory approaches. This section also discusses three selected
issues in greater depth because of their importance to designing
effective GHG controls: advantages of market-oriented regulatory
approaches, economy-wide and sector-based regulation under the CAA, and
emissions leakage and international competitiveness. In discussing
these and other policy and economic considerations, EPA is not directly
or indirectly implying that it possesses the requisite statutory
authority in all areas.
1. Overview of Policy and Economic Considerations
    The following considerations are useful in developing potential
regulatory approaches to the extent permissible under the CAA. These
considerations are also generally applicable to the design of GHG
control legislation. EPA is in the process of evaluating the CAA
options described later in this notice in light of these
considerations.
    Effectiveness of health and environmental risk reduction: How much
would the approach reduce negative health and environmental impacts (or
the risk of such impacts), relative to other potential approaches?
    Certainty and transparency of results: How do the potential
regulatory approaches balance the trade-off between certainty of
emission reductions and costs? To what extent can compliance
flexibility be provided for regulated entities while maintaining
adequate accountability for emission reductions?
    Cost-effectiveness and economic efficiency considerations: To what
extent does the approach allow for achieving health and environmental
goals, determined in a broader policy process, in a manner that imposes
the least cost? How do the societal benefits compare to the societal
costs? To what extent are there non-monetizable or unquantifiable
benefits and costs? Given the uncertainties associated with climate
change, to what extent can economic efficiency be judged?
    Equity considerations (i.e., distributional effects): Does the
approach by itself or in combination with other programs result in a
socially acceptable apportionment of the burden of emission reduction
across groups in our society? Does the approach provide adequate
protection for those who will experience the adverse effects of
emissions, including future generations?
    Policy flexibility over time: Does the approach allow for updating
of environmental goals and mechanisms for meeting those goals as new
information on the costs and benefits of GHG emission reductions
becomes available?
    Incentives for innovation and technology development: Does the
approach provide incentives for development and deployment of new,
cleaner technologies in the United States and transfer abroad? Does the
approach create incentives for individual regulated entities to achieve
greater-than-required emissions reductions?
    Competitiveness/emissions shifts: Can the approach be designed to
reduce potential adverse impacts and consequent shifts in production
and emissions to other sectors or geographic areas? Can the policy be
designed to minimize the shifting, or ``leakage,'' of emissions to
other sectors or other countries, which would offset emission reduction
benefits of the policy? To what extent can the approach consider the
degree and nature of action taken by other countries?
    Administrative feasibility: How complex and resource-intensive
would the approach be for federal, state, and local governments and for
regulated entities? Do personnel in the public and private sectors have
sufficient expertise, or can they build sufficient expertise, to
successfully implement the approach?
    Enforceability: Is the approach enforceable in practice? Do
available regulatory options differ regarding whether the government or
the regulated entity bears the burden of demonstrating compliance?
    Unintended consequences: Does the approach result in unintended
consequences or unintended effects for other regulations? Does the
approach allow for consideration of, and provide tools to address, any
perverse incentives?
    Suitability of tool for the job: Overall, is the approach well-
suited to the environmental problem, or the best-suited among imperfect
alternatives? For example, does the regulatory approach fit the
characteristics of the pollutant in question (e.g., the global and
long-lived nature of GHGs, high volume of CO₂ emissions)?
2. Market-Oriented Regulatory Approaches for GHGs
    EPA believes that market-oriented regulatory approaches, when well-
suited to the environmental problem, offer important advantages over
non-market-oriented approaches. A number of theoretical and empirical
studies have shown these advantages.\55\ In general, market-oriented
approaches include ways of putting a price on emissions through a fixed
price (e.g., a tax) or exchangeable quantity-based instrument (e.g., a
cap-and-trade program), while non-market-oriented approaches set
performance standards limiting the rate at which individual entities
can emit, or prescribe what abatement behaviors or technologies they
should use.\56\ The primary regulatory advantage of a market-oriented
approach is that it can achieve a particular emissions target at a
lower

[[Page 44410]]

social cost than a non-market-oriented \57\ approach (Baumol and Oates,
1971; Tietenberg, 1973).\58\ This is because market-oriented approaches
leave the method for reducing pollution to the emitter, and emitters
have an incentive to find the least cost way of achieving the
regulatory requirement. Efficient market-oriented regulatory systems
provide a common emissions price for all emitters that contribute to a
particular harm, either through the tax on emissions or the price of an
exchangeable right to emit. As a result, the total abatement required
by the policy can theoretically be distributed across all emitters in
such a way that the marginal cost of control is equal for all emitters
and the cost of reducing emissions is minimized.\59\ Non-market-
oriented policies offer emitters fewer choices on how to reduce
emissions, which can lead to higher costs than are necessary to achieve
the overall environmental objective (i.e. emission level).
---------------------------------------------------------------------------

    \55\ See EPA (2000), Baumol and Oates (1988), Tietenberg (2006)
and Burtraw et al. (2005) for a detailed description of the
advantages of market-oriented policies, such as the Title IV sulfur
dioxide trading program, over non-market-oriented approaches.
    \56\ Performance standards provide a source flexibility to use
any emission reduction method that meets the performance standard;
they can be coupled with market-oriented approaches such as
emissions trading to promote lower costs and technology innovation,
as described later in this section.
    \57\ Many studies use the term ``command-and-control'' to refer
to non-market-oriented approaches. Here we use the term ``non-
marketed-oriented'' because the term ``command and control'' may be
misleading when used to refer to performance-based emission limits
that allow the regulated entity to choose the control technology or
strategy for compliance.
    \58\ It is important to note that judgments about the
appropriate mitigation approach also may consider important societal
values not fully captured in economic analysis, such as political,
legal, and ethical considerations. For example, different regulatory
forms may result in different distributions of costs and benefits
across individuals and firms. This is a particularly sensitive issue
with policies that raise energy costs, which are known to be
regressive. However, these issues are not discussed at length here.
    \59\ For a standard textbook treatment supporting this finding
see Tietenberg (2006) or Callan and Thomas (2007).
---------------------------------------------------------------------------

    As noted previously, it is especially important that any GHG
emission reduction policy encourage the innovation, development and
diffusion of technologies to provide a steady decline in the costs of
emission reductions. Another advantage of market-oriented approaches is
that they generally provide a greater incentive to develop new ways to
reduce pollution than non-market-oriented approaches (Malueg 1989;
Milliman and Prince 1989; Jung et al., 1996). Polluters not only have
an incentive to find the least cost way of adhering to a standard but
they also have an incentive to continually reduce emissions beyond what
is needed to comply with the standard. For every unit of emissions
reduced under a market-oriented policy, the emitter either has a lower
tax burden or can sell an emissions permit (or buy one less emissions
permit). Also, there are more opportunities under a market-oriented
approach for developers of new control technologies to work directly
with polluters to find less expensive ways to reduce emissions, and
polluters are faced with less compliance risk if a new pollution
control technique does not work as expected. This is because they can
either pay for their unanticipated emissions through the tax or by
purchasing emission rights instead of being subject to enforcement
action (Hahn, 1989).
    There are a number of examples of CAA rules in which market-
oriented approaches have been used for groups of mobile or stationary
sources. Usually this has taken the form of emissions trading within a
sector or subsector of a source category, although there are some
examples of broader trading programs. Differences in implications of
sector-specific and economy-wide market-oriented systems are discussed
in subsection below.
    The cost advantage of market-oriented policies can be extended when
emitters are allowed to achieve a particular environmental objective
across multiple pollutants that affect environment quality in the same
way but differ in the magnitude of that effect (e.g., different GHGs
have different global warming potentials). Either a cap-and-trade or a
tax approach could be designed so that the effective price per unit of
emissions is higher for those pollutants that have a greater
detrimental effect. Under a cap, the quantity of emissions reductions
is fixed but not the price; under a tax, the price is fixed but not the
emissions reductions. Some current legislative proposals include
flexible multiple-pollutant market-oriented policies for the control of
GHG emissions.
    Market-oriented approaches are relatively well-suited to
controlling GHG emissions. Since emissions of the major GHGs are
globally well-mixed, a unit of GHG emissions generally has the same
effect on global climate regardless of where it occurs. Also, while
policies can control the flow of GHG emissions, what is of ultimate
concern is the concentration of cumulative GHGs in the atmosphere.
Providing flexibility on the method, location and precise timing of GHG
reduction would not significantly affect the global climate protection
benefits of a GHG control program (assuming effective enforcement
mechanisms), but could substantially reduce the cost and encourage
technology innovation.\60\ However, it should be noted that for GHG
control strategies that also reduce emissions of traditional
pollutants, the timing and location of those controls could
significantly affect air quality in local or regional areas. There is
the potential for positive air quality effects from strategies that
reduce both GHGs and traditional pollutants, and for adverse air
quality effects that may be avoidable through complementary measures to
address air quality. For example, when the acid rain control program
was instituted, existing sulfur dioxide control programs were left in
place to ensure that trading under the acid rain program did not
undermine achievement of local air quality objectives.
---------------------------------------------------------------------------

    \60\ We say ``precise'' timing because the qualifier is
important: The IPCC and others have noted that lower GHG
stabilization targets would require steeper and earlier emission
reductions, whereas stabilization targets that allow for more
warming (with higher associated risks and impacts) would require
less steep and later emission reductions.
---------------------------------------------------------------------------

    As noted previously, broad-based market-oriented approaches include
emissions taxes and cap-and-trade programs with and without cost
containment mechanisms. While economists disagree on which of these
approaches--emissions taxes or cap-and-trade programs--may be
particularly well-suited to the task of mitigating GHG emissions, they
do agree that attributes such as flexibility, cost control, and broad
incentives for minimizing abatement costs and developing new
technologies are important policy design considerations.\61\ For a
description of various market-oriented approaches, see section VII.G.
---------------------------------------------------------------------------

    \61\ These approaches also raise the issue of the potential use
of revenues from collecting a tax or auctioning allowances to emit
GHGs at levels that do not exceed the cap. See Chapter 4 of U.S. EPA
(2000), ``Guidelines for Preparing Economic Analyses,'' EPA 240-R-
00-003.
---------------------------------------------------------------------------

3. Legal Authority for Market-Oriented Approaches Under the Clean Air
Act
    The ability of each CAA regulatory authority potentially applicable
to GHGs to support market-oriented regulatory approaches is discussed
in sections VI and VII of this notice. To summarize, some CAA
provisions permit or require market-oriented approaches, and others do
not. Trading programs within sectors or subsectors have been
successfully implemented for a variety of mobile and stationary source
categories under the Act, including the Acid Rain Control Program (58
FR 3590 (Jan. 11, 1993)) and a variety of on-road and non-road vehicle
and fuel rules. Multi-sector trading programs, though not economy-wide,
have been successfully implemented under section 110(a)(2)(D) for
nitrogen oxides (i.e. the NO_X SIP Call Rule) and under Title
VI for ozone-depleting substances, and may be

[[Page 44411]]

possible among stationary source sectors under section 111. An economy-
wide system might be legally possible under CAA section 615 (if the
two-part test unique to that section were met) or if a NAAQS were
established for GHGs. However, any economy-wide program under either
provision would not stand alone; it would be accompanied by source-
specific or sector-based requirements as a result of other CAA
provisions (e.g., PSD permitting under section 165).
    The CAA does not include a broad grant of authority for EPA to
impose taxes, fees or other monetary charges specifically for GHGs and,
therefore, additional legislative authority may be required if EPA were
to administer such charges (which we will refer to collectively as
fees). EPA may promulgate regulations that impose fees only if the
specific statutory provision at issue authorizes such fees, whether
directly or through a grant of regulatory authority that is written
broadly enough to encompass them. For example, CAA section 110(a)(2)(A)
allows for the use of ``economic incentives such as fees, marketable
permits, and auctioning allowances.'' Under this provision, some states
intend to auction allowances under CAIR (70 FR 25162 (May 12, 2005))
and some have under the NO_X SIP Call Rule (63 FR 57356 (Oct.
27, 1998)). By the same token, states have authority to impose
emissions fees as economic incentives as part of their SIPs and collect
the revenues. Similarly, section 110(a)(2)(A) authorizes EPA to impose
fees as economic incentives as part of a Federal Implementation Plan
(FIP) under section 110(c), although EPA has never done so.\62\
---------------------------------------------------------------------------

    \62\ Any such revenues from a FIP would be deposited in the
Federal Treasury under the Miscellaneous Receipts Act, and not
retained and disbursed by EPA.
---------------------------------------------------------------------------

    Section 111 authorizes EPA to promulgate ``standards of
performance,'' which are defined as ``standard[s] for emissions of air
pollutants.'' EPA has taken the position that this term authorizes a
cap-and-trade program under certain circumstances. A fee program
differs from a cap and trade because it does not establish an overall
emission limitation, and we have not taken a position on whether, given
this limitation, a fee program fits the definition of a ``standard of
performance.'' Even so, under section 111 costs may be considered when
establishing NSPS regulations, and a fee may balance the consideration
of assuring emissions are reduced but not at an unacceptably high cost.
Also, there may be advantages of including an emission fee feature into
a cap-and-trade program (i.e., as a price ceiling). The use of a price
ceiling that is not expected to be triggered except in the case of
unexpectedly high (or low) control costs may be viewed differently
under the auspices of the CAA than a stand-alone emissions fee.
    We request comment on what CAA provisions, if any, would authorize
emissions fees to control GHG emissions, and whether there are other
approaches that could be taken under the CAA that would approximate a
fee. Furthermore, we request comments on the use of emission fee
programs under other sections of the Act. We also seek comment on
whether sector-specific programs, or inter-sector programs where
emission fees on a CO₂ equivalent basis are harmonized,
might be more appropriate as possible regulatory mechanisms under the
Act.
4. Economy-Wide and Sector-Based Regulation in a Clean Air Act Context
    Several legislative cap-and-trade proposals for reducing GHG
emissions are designed to be nearly economy wide, meaning that they
attempt to reduce GHG emissions in most economic sectors through a
single regulatory system. By contrast, many CAA authorities are
designed for regulations that apply to a sector, subsector or source
category, although broader trading opportunities exist under some
authorities. This section discusses the relative merits of economy-wide
systems and sector-based market-oriented approaches. These
considerations may also be relevant in considering the use of CAA
provisions in tandem with any climate change legislation.
i. Economy-Wide Approach
    Economic theory suggests that establishing a single price for GHG
emissions across all emitters through an economy-wide, multiple GHG,
market-oriented policy would promote optimal economic efficiency in
pursuing GHG reductions. According to the economics literature,
economy-wide GHG trading or GHG emissions taxes could offer
significantly greater cost savings than a sector-by-sector approach for
GHGs because the broader the universe of sources covered by a single
market-oriented approach (within a sector, across sectors, and across
regions), the greater the potential for finding lower-cost ways to
achieve the emissions target. If sources of pollution are
compartmentalized into different sector-specific or pollutant-specific
approaches, including the relatively flexible cap-and-trade approaches,
each class of polluter may still face a different price for their
contribution to the environmental harm, and therefore some trading
opportunities that reduce pollution control costs will be unrealized
(Burtraw and Evans, 2008).\63\ Taking a sector-by-sector approach to
controlling GHG emissions is likely to result in higher costs to the
economy. For example, limiting a market-oriented GHG policy to the
electricity and transportation sectors could double the welfare cost of
achieving a five percent reduction in carbon emissions compared to when
the industrial sector is also included.\64\
---------------------------------------------------------------------------

    \63\ With traditional pollutants there are geographic issues to
consider.
    \64\ William Pizer, Dallas Burtraw, Winston Harrington, Richard
Newell, and James Sanchirico (2006), ``Modeling Economywide versus
Sectoral Climate Policies Using Combined Aggregate-Sectoral
Models,'' The Energy Journal, Vol. 27, No. 3: 135-168.
---------------------------------------------------------------------------

    A second factor that favors making the scope of a market-oriented
system as broad as possible is that the incentive for development,
deployment and diffusion of new technologies would be spread across the
economy. In contrast to an approach targeting a few key sectors, an
economy-wide approach would affect a greater number of diverse GHG-
emitting activities, and would influence a larger number of individual
economic decisions, potentially leading to innovation in parts of the
economy not addressed by a sector-by-sector approach.
    As stated at the outset of this section, there are, first and most
important, CAA authority issues as well as other policy and practical
considerations in addition to economic efficiency that must be weighed
in evaluating potential CAA approaches to GHG regulation. An economy-
wide, market-oriented environmental regulation has never been
implemented before in the U.S. The European Union, after encountering
difficulties in early years of implementation, recently adopted major
revisions to its broad multi-sector cap-and-trade system; this
illustrates that some time and adjustments may be needed for such a
program to achieve its intended effect. Although EPA has successfully
designed and implemented market-oriented systems of narrower scope, a
single economy-side system would involve new design and implementation
challenges, should the CAA make possible such a system. For example --
     Administrative costs may be a concern, because more
sources and sectors would have to be subject to

[[Page 44412]]

reporting and measurement, monitoring, and verification requirements.
     Some sources and sectors are more amenable to market-
oriented approaches than others. The feasibility and cost of accurate
monitoring and compliance assurance needed for trading programs
(whether economy-wide or sector-based) varies among sectors and source
size. As a result, there are potential tradeoffs between trading
program scope and level of assurance that required emissions reductions
will be achieved.
     To broaden the scope of cap-and-trade systems, covered
sources could be allowed to purchase GHG emission reductions
``offsets'' from non-covered sources. However, offsets raise additional
accountability issues, including how to balance cost efficiency against
certainty of emissions reductions, how to quantify resulting emissions
reductions, and how to ensure that the activities generating the
offsets are conducted and maintained over time.
     Allocating allowances or auction revenues for an economy-
wide GHG trading system would be very challenging for an executive
branch agency because of high monetary stakes and divergent stakeholder
views on how to distribute the allowances or revenues to promote
various objectives. For example, many economists believe that
auctioning allowances under a cap-and-trade system and using the
proceeds to reduce taxes that distort economic incentives would be
economically efficient, but regulated entities typically favor free
allowance allocations to offset their compliance costs.65 66
---------------------------------------------------------------------------

    \65\ Many economists also suggest that an emissions tax with
proceeds used to decrease distortionary taxes would be economically
efficient; however, the CAA does not authorize such a program.
    \66\ Bovenberg and Goulder (2001) find that freely allocating
20% of allowances to fossil fuel suppliers is enough to keep profits
from falling. When all allowances are freely allocated, profits are
found to be higher than in the absence of the carbon cap-and-trade
policy. Free allocation of allowances or an approach that exempts
particular sectors also raises the specter of ``rent-seeking,'' the
notion that sectors or particular source categories will lobby to
gain preferential treatment and, in essence, be subject to less
regulatory oversight than other sectors or competitors.
---------------------------------------------------------------------------

ii. Sector-Based and Multi-Sector Trading Under the Clean Air Act
    As mentioned above, EPA has implemented multi-sector, sector and
subsector-based cap-and-trade approaches in a number of CAA programs,
including the Acid Rain (SO₂) Program, the NO_X
SIP Call Rule, the Clean Air Interstate Rule (CAIR), and the
stratospheric ozone-depleting substances (ODS) phase-out rule. In the
case of the acid rain and ODS rules, the CAA itself called for federal
controls. By contrast, the NO_X SIP Call rule and CAIR were
established by EPA through regulations under CAA section 110(a)(2)(d)
to help states attain various NAAQS. The two rules and EPA's
accompanying model rules enable states to adopt compatible cap-and-
trade programs that form regional interstate trading programs. The
power sector and a few major industrial source categories are included
in the trading system for the NO_X SIP Call, and the trading
system for CAIR focuses on the electricity generation sector.
    In addition to creating cap-and-trade systems, EPA has often
incorporated market-oriented emissions trading elements into the more
traditional performance standard approach for mobile and stationary
sources. Coupling market-oriented provisions with performance standards
provides some of the cost advantages and market flexibility of market-
oriented solutions while also directly incentivizing technology
innovation within the particular sector, as discussed below. For
example, performance standards for mobile sources under Title II have
for many years been coupled with averaging, banking and trading
provisions within a subsector. In general, averaging allows covered
parties to meet their emissions obligation on a fleet- or unit-wide
basis rather than requiring each vehicle or unit to directly comply.
Banking provides direct incentives for additional reductions by giving
credit for over-compliance; these credits can be used toward future
compliance obligations and, as such, allow manufacturers to put
technology improvements in place when they are ready for market, rather
than being forced to adhere to a strict regulatory schedule that may or
may not conform to industry or company developments. Allowing trading
of excess emission reductions with other covered parties provides an
incentive for reducing emissions beyond what is required.
    Based on our experience with these programs, EPA believes that
sector and multi-sector trading programs for GHGs--relative to non-
market regulatory approaches--could offer substantial compliance
flexibility, cost savings and incentives for innovation to regulated
entities. In addition, as discussed below, in some sectors there may be
a need to more directly incentivize technology development because of
market barriers that a sector-specific program might help to overcome.
To the extent sector-based approaches could provide for control of
multiple pollutants (e.g., traditional pollutants and GHGs), they could
provide additional cost savings relative to multiple single-pollutant,
sector-based regulations. Another consideration is that it may be
simpler and thus faster to move forward with cap-and-trade programs for
sectors already involved in, and thus familiar with, cap-and-trade
programs. This raises the question of whether it would make sense to
phase in an economy-wide system over time.
    Sector and multi-sector approaches would not offer the relative
economic efficiency of the economy-wide model for the reasons explained
above. To the extent the program sets more stringent requirements for
new sources than for existing source, a sector or multi-sector approach
could also pose the vintage issues discussed below. It is also
important to keep in mind that the economic efficiency of any CAA cap-
and-trade approach for GHGs, sector- or economy-wide, could be reduced
to a significant extent by the application of other GHG control
requirements (e.g., PSD permitting) to the sources covered by the cap-
and-trade program, if the result were to restrict compliance options.
iii. Combining Economy-Wide and Sector-Based Approaches
    It is worth noting that market-oriented approaches may not
incentivize the most cost-effective reductions when information
problems, infrastructure issues, technological issues or other factors
pose barriers that impeded the market response to price incentives. In
such instances, there may be economic arguments for combining an
economy-wide approach with complementary sector-based requirements
unless these problems can be directly addressed, for instance by
providing the information needed or directly subsidizing the creation
of needed infrastructure.
    For instance, given the relative inelasticity of demand for
transportation, even a relative high permit price for carbon may not
substantially change consumer vehicle purchases or travel demand,
although recent reports indicate that the current price of gasoline and
diesel are inducing an increasing number of consumers to choose more
fuel efficient vehicles and drive less. Some have expressed concern
that this relatively inelastic demand may be related to undervaluation
by consumers of fuel economy when making vehicle purchasing decisions.
If consumers adequately value fuel economy, fuel saving technologies
will come online as a result of market forces. However, if

[[Page 44413]]

consumers undervalue fuel economy, vehicle or engine manufacturers may
need a more direct incentive for making improvements or the technology
innovation potential may well be delayed or not fully realized. Beyond
this consumer valuation issue, questions have been raised as to whether
a carbon price alone (especially if the impact is initially to raise
gasoline prices by pennies a gallon) will provide adequate incentives
for vehicle manufacturers to invest now in breakthrough technologies
with the capability to achieve significantly deeper emissions
reductions in the future, and for fuel providers to make substantial
investments in a new or enhanced delivery infrastructure for large-
scale deployment of lower carbon fuels.\67\
---------------------------------------------------------------------------

    \67\ See Kopp and Pizer, ``Assessing U.S. Climate Policy
Options,'' Chapter 12, RFF Press: Washington, DC (2007).
---------------------------------------------------------------------------

    EPA requests comment on how to balance the different policy and
economic considerations involved in selecting potential regulatory
approaches under the CAA, and on how the potential enactment of
legislation should affect EPA's deliberations on how to use CAA
authorities.
5. Other Selected Policy Design Issues
    Another policy and legal issue in regulatory design is whether
requirements should differentiate between new and existing sources.
Because it is generally more costly to retrofit pollution control
equipment than to incorporate it into the construction or manufacture
of a new source, environmental regulations, including under the CAA,
frequently apply stricter standards to new or refurbished sources than
to ``grandfathered'' sources that pre-date the regulation. New sources
achieve high-percentage reductions and over time existing high-emitting
sources are replaced with much cleaner ones. For example, emissions
from the U.S. auto fleet have been dramatically reduced over time
through new vehicle standards. However, some suggest that stricter
pollution control requirements for new or refurbished sources may
retard replacement of older sources, discouraging technology
investment, innovation and diffusion while encouraging older and less
efficient sources to remain in operation longer, thereby reducing the
environmental effectiveness and cost-effectiveness of the regulation.
Others believe that economic factors other than differences in new and
existing source requirements (e.g., capital outlay, power prices and
fuel costs) have the most impact on rate of return, and that
differences in regulatory stringency generally do not drive business
decisions on when to build new capacity.
    A 2002 EPA report on new source review requirements found that NSR
``appears to have little incremental impact on construction of new
electricity generation,'' but also found that ``there were credible
examples of cases in which uncertainty over the [NSR] exemption for
routine activities has resulted in delay or cancellation of projects
[at existing plants]'' that would have increased energy capacity,
improved energy efficiency and reduced air pollution.\68\ To the extent
that a gap in new and existing source requirements affects business
decisions, regulating existing as well as new sources can diminish or
eliminate that gap. In the power sector, the gap has narrowed over
time, in part as a result of CAA national and regional cap-and-trade
systems that do not discriminate between new and existing facilities
(i.e., both new and old power plants must hold allowances to cover
their NO_X and SO₂ emissions). Another
consideration is that equity issues can arise when applying retroactive
requirements to existing sources. For GHGs, EPA requests comment on the
concept of a market-oriented approach that does not differentiate
between new and existing source controls and, by avoiding different
marginal costs of control at new and existing sources, would promote
more cost-effective emissions reductions. In addition, EPA requests
comment on whether GHG regulations should differentiate between new and
existing sources for various sectors, and whether there are
circumstances in which requirements for stringent controls on new
sources would have policy benefits despite the existence of a cap-and-
trade system that also would apply to those sources.
---------------------------------------------------------------------------

    \68\ ``New Source Review: Report to the President, June 2002,''
U.S. EPA, pp. 30-31.
---------------------------------------------------------------------------

    Another possible design consideration for a GHG program is whether
and how lifecycle approaches to controlling GHG emissions could or
should be used. Lifecycle (LC) analysis and requirements have been
proposed for determining and regulating the entire stream of direct and
indirect emissions attributable to a regulated source. Indirect
emissions are emissions from the production, transportation, and
processing of the inputs that go into producing that good. Section VI.D
describes possible CAA approaches for reducing GHG emissions from
transportation fuels through lifecycle analysis and includes a brief
discussion of a potential lifecycle approach to reducing fuel-related
GHG emissions. In that context, displacing petroleum-based fuels with
renewable or alternative fuels can reduce fuel-related GHGs to the
extent the renewable or alternative fuels are produced in ways that
result in lower GHG emissions than the production of an equivalent
amount of fossil-based fuels. Tailpipe GHG emissions typically do not
vary significantly across conventional and alternative or renewable
fuels.
    EPA recognizes that other programs, such as stationary source or
area source programs described in this notice, could potentially
address at least some of the indirect GHG emissions from producing
fuels. We note that the technology and fuel changes that may result
from an economy-wide cap-and-trade approach would likely be different
from the technology and fuel changes that may result from a lifecycle
approach.
    EPA asks for comment on how a lifecycle approach for fuels could be
integrated with other stationary source approaches and whether there
are potentially overlapping incentives or disincentives. EPA also asks
for comments on whether a lifecycle approach to reducing GHG emissions
may be appropriate for other sectors and types of sources, and what the
implications for regulating other sectors would be if a lifecycle
approach is taken for fuels.
6. ``Emissions Leakage'' and International Competitiveness
    A frequently raised concern with domestic GHG regulation
unaccompanied by comparable policies abroad is that it might result in
emissions leakage or adversely affect the international competitiveness
of certain U.S. industries. The concern is that if domestic firms faced
significantly higher costs due to regulation, and foreign firms
remained unregulated, this could result in price changes that shift
emissions, and possibly some production capacity, from the U.S. to
other countries. Emissions leakage also could occur without being
caused by a competitiveness issue: for instance, if a U.S. GHG policy
raised the domestic price of petroleum-based fuels and led to reduced
U.S. demand for those fuels, the resulting world price decline could
spur increased use of petroleum-based fuels abroad, leading to
increased GHG emissions abroad that offset U.S. reductions.
    The extent to which international competitiveness is a potential
concern varies substantially by sector. This issue is mainly raised for
industries with high energy use and substantial potential

[[Page 44414]]

foreign competition. Even for vulnerable sectors, the concern would
depend on the actual extent which a program would raise costs for an
energy intensive firm facing international competition, and on whether
policies to address the competitiveness issue were adopted (either as
part of the rule or in another venue).
    Leakage also could occur within the U.S. if emissions in one sector
or region are controlled, but other sources are not. In this case, the
market effects could lead to increased activity in unregulated sectors
or regions, offsetting some of the policy's emissions reductions. In
turn, this would raise the cost of achieving the environmental
objective. The more uniform the price signal for an additional unit
reduction in GHG emissions across sectors, states, and countries, the
less potential there is for leakage to occur.
    A recent report has identified and evaluated five conceptual
options for addressing competitiveness concerns in a legislative
context; some options might also be available in a regulatory
context.\69\ The first option, weaker program targets, would affect the
entire climate protection policy. Four other options also could
somewhat decrease environmental stringency but would allow for the
targeting of industries or sectors particularly vulnerable to adverse
economic impacts:
---------------------------------------------------------------------------

    \69\ Morgenstern, Richard D., ``Issue Brief 8: Addressing
Competitiveness Concerns in the Context of a Mandatory Policy for
Reducing U.S. Greenhouse Gas Emissions,'' in Assessing U.S. Climate
Policy Options: A report summarizing work at RFF [Resources for the
Future] as part of the inter-industry U.S. Climate Policy Forum,
November 2007, Raymond J. Kopp and William A. Pizer, eds.
---------------------------------------------------------------------------

     Exemptions
     Non-market regulations to avoid direct energy price
increases on an energy-intensive industry
     Distribution of free allowances to compensate adversely
affected industries in a cap-and-trade system
     Trade-related policies such as import tariffs on carbon or
energy content, export subsidies, or requirements for importers to
submit allowances to cover the carbon content of certain products.

Significantly, the report noted that identifying the industries most
likely to be adversely affected by domestic GHG regulation, and
estimating the degree of impact, is complex in terms of data and
analytical tools needed.
    We request comment on the extent to which CAA authorities described
in this notice could be used to minimize competitiveness concerns and
leakage of emissions to other sectors or countries, and which
approaches should be preferred.

G. Analytical Challenges for Economic Analysis of Potential Regulation

    In the event that EPA pursues GHG emission reduction policies under
the CAA or as a result of legislative action, we are required by
Executive Order 12866 to analyze and take into account to the extent
permitted by law the costs and benefits of the various policy options
considered. Economic evaluation of GHG mitigation is particularly
challenging due to the temporal and spatial dimensions of the problem
discussed previously: GHG emissions have extremely long-run and global
climate implications. Furthermore, changes to the domestic economy are
likely to affect the global economy. In this section, we discuss a few
overarching analytical challenges that follow from these points. Many
of the issues discussed are also relevant when valuing changes in GHGs
associated with non-climate policies.
1. Time Horizon and International Considerations in General
    As discussed earlier in this section, changes in GHG emissions
today will affect environmental, ecological, and economic conditions
for decades to centuries into the future. In addition, changes in U.S.
GHG emissions that result from U.S. domestic policy will affect climate
change everywhere in the world, as will changes in the GHG emissions of
other countries. U.S. domestic policy could trigger emissions changes
across the U.S. economy and across regions globally, as production and
competitiveness change among economic activities. Similarly,
differences in the potential impacts of climate change across the world
can also affect competitiveness and production. Capturing these effects
requires long-run, global analysis in addition to traditional domestic
and sub-national analyses.
2. Analysis of Benefits and Costs Over a Long Time Period
    Since changes in emissions today will affect future generations in
the U.S. and internationally, costs and benefits of GHG mitigation
options need to be estimated over multiple generations. Typically,
federal agencies discount future costs or benefits back to the present
using a discount rate, where the discount rate represents how society
trades-off current consumption for future consumption. With the
benefits of GHG emissions reductions distributed over a very long time
horizon, benefit and cost estimations are likely to be very sensitive
to the discount rate. For policies that affect a single generation of
people, the analytic approach used by EPA is to use discount rates of
three and seven percent at a minimum.\70\ According to the Office of
Management and Budget (OMB), a three percent rate is consistent with
what a typical consumer might expect in the way of a risk free market
return (e.g., government bonds). A seven percent rate is an estimate of
the average before-tax rate of return to private capital in the U.S.
economy. A key challenge facing EPA is the appropriate discount rate
over the longer timeframe relevant for GHGs.
---------------------------------------------------------------------------

    \70\ EPA (U.S. Environmental Protection Agency), 2000.
Guidelines for Preparing Economic Analyses. EPA 240-R-00-003. See
also OMB (U.S. Office of Management and Budget), 2003. Circular A-4.
September 17, 2003.
---------------------------------------------------------------------------

    There are reasons to consider even lower discount rates in
discounting the costs of benefits of policy that affect climate change.
First, changes in GHG emissions--both increases and reductions--are
essentially long-run investments in changes in climate and the
potential impacts from climate change. When considering climate change
investments, they should be compared to similar alternative investments
(via the discount rate). Investments in climate change are investments
in infrastructure and technologies associated with mitigation; however,
they yield returns in terms of avoided impacts over a period of one
hundred years and longer. Furthermore, there is a potential for
significant impacts from climate change, where the exact timing and
magnitude of these impacts are unknown. These factors imply a highly
uncertain investment environment that spans multiple generations.
    When there are important benefits or costs that affect multiple
generations of the population, EPA and OMB allow for low but positive
discount rates (e.g., 0.5-3% noted by U.S. EPA, 1-3% by OMB).\71\ In
this multi-generation context, the three percent discount rate is
consistent with observed interest rates from long-term investments
available to current generations (net of risk premiums) as well as
current estimates of the impacts of climate change that reflect
potential impacts on consumers. In addition, rates of three percent or
lower are consistent with long-run uncertainty in economic growth and
interest rates, considerations of issues associated with the transfer
of wealth between generations, and the risk of

[[Page 44415]]

high impact climate damages. Given the uncertain environment, analysis
could also consider evaluating uncertainty in the discount rate (e.g.,
Newell and Pizer, 2001, 2003).\72\ EPA solicits comment on the
considerations raised and discounting alternatives for handling both
benefits and costs for this long term, inter-generational context.
---------------------------------------------------------------------------

    \71\ OMB (2003). EPA (2000). These documents are the guidance
used when preparing economic analyses for all EPA rulemakings.
    \72\ Newell, R. and W. Pizer, 2001. Discounting the benefits of
climate change mitigation: How much do uncertain rates increase
valuations? PEW Center on Global Climate Change, Washington, DC.
Newell, R. and W. Pizer, 2003. Discounting the distant future: how
much do uncertain rates increase valuations? Journal of
Environmental Economics and Management 46: 52-71.
---------------------------------------------------------------------------

3. Uncertainty in Benefits and Costs
    The long time horizon over which benefits and costs of climate
change policy would accrue and the global relationships they involve
raise additional challenges for estimation. The exact benefits and
costs of virtually every environmental regulation is at least somewhat
uncertain, because estimating benefits and costs involves projections
of future economic activity and the future effects and costs of
reducing the environmental harm. In almost every case, some of the
future effects and costs are not entirely known or able to be
quantified or monetized. In the case of climate change, the uncertainly
inherent in most economic analyses of environmental regulations is
magnified by the long-term and global scale of the problem and the
resulting uncertainties regarding socio-economic futures, corresponding
GHG emissions, climate responses to emissions changes, the bio-physical
and economic impacts associated with changes in climate, and the costs
of reducing GHG emissions. For example, uncertainties about the amount
of temperature rise for a given amount of GHG emissions and rates of
economic and population growth over the next 50 or 100 years will
result in a large range of estimates of potential benefits and costs.
Lack of information with regard to some important benefit categories
and the potential for large impacts as a result of climate exceeding
known but uncertain thresholds compound this uncertainty. Likewise,
there are uncertainties regarding the pace and form of future
technological innovation and economic growth that affect estimates of
both costs and benefits. These difficulties in predicting the future
can be addressed to some extent by evaluating alternative scenarios. In
uncertain situations such as that associated with climate, EPA
typically recommends that analysis consider a range of benefit and cost
estimates, and the potential implications of non-monetized and non-
quantified benefits.
    Given the substantial uncertainties in quantifying many aspects of
climate change mitigation and impacts, it is difficult to apply
economic efficiency criteria, or even positive net benefit
criteria.\73\ Identifying an efficient policy requires knowing the
marginal benefit and marginal cost curves for GHG emissions reductions.
If the marginal benefits are greater than the marginal costs, then
additional emissions reductions are merited (i.e., they are efficient
and provide a net benefit). However, the curves are not precise lines;
instead they are wide and partially unknown bands. Similarly, estimates
of total benefits and costs can be expressed only as ranges. As a
result, it is difficult to both identify the efficient policy and
assess net benefits.
---------------------------------------------------------------------------

    \73\ IPCC WGI. (2007). Climate Change 2007--The Physical Science
Basis Contribution of Working Group I to the Fourth Assessment
Report of the IPCC, http://www.ipcc.ch/. IPCC WGII. (2007). Climate
Change 2007--Impacts, Adaptation and Vulnerability Contribution of
Working Group II to the Fourth Assessment Report of the IPCC, http:/
/www.ipcc.ch/. IPCC WGIII (2007). Climate Change 2007--Mitigation
Contribution of Working Group III to the Fourth Assessment Report of
the IPCC, http://www.ipcc.ch/. U.S. Congressional Budget Office
(2005). Uncertainty in Analyzing Climate Change: Policy
Implications. The Congress of the United States, January 2005.
---------------------------------------------------------------------------

    In situations with large uncertainties, the economic literature
suggests a risk management framework as being appropriate for guiding
policy (Manne and Richels, 1992; IPCC WGIII, 2007).\74\ In this
framework, the policymaker selects a target level of risk and seeks the
lowest cost approach for reaching that goal. In addition, the decision-
making process is an iterative one of acting, learning, and acting
again (as opposed to there being a single decision point). In this
context, the explicit or implicit value of changes in risk is
important. Furthermore, some have expressed concern in the economics
literature that standard deterministic approaches (i.e., approaches
that imply there is only one known and single realization of the world)
do not appropriately characterize the uncertainty and risk related to
climate change and may lead to a substantial underestimation of the
benefits from taking action (Weitzman, 2007a, 2007b).\75\ Formal
uncertainty analysis may be one approach for at least partially
addressing this concern. EPA solicits comment on how to handle
uncertainty in benefits and costs calculations and application, given
the quantified and unquantified uncertainties.
---------------------------------------------------------------------------

    \74\ Manne, A. and R. Richels (1992). ``Buying Greenhouse
Insurance--the Economic Costs of Carbon Dioxide Emission Limits'',
MIT Press book, Cambridge, MA, 1992. IPCC WGIII (2007).
    \75\ Weitzman, M., 2007a, ``The Stern Review of the Economics of
Climate Change,'' Journal of Economic Literature. Weitzman, M.,
2007b, ``Structural Uncertainty and the Statistical Life in the
Economics of Catastrophic Climate Change,'' Working paper
econweb.fas.harvard.edu/faculty/weitzman/papers/
ValStatLifeClimate.pdf.
---------------------------------------------------------------------------

4. Benefits Estimation Specific Issues--Scope, Estimates, State-of-the-
art
    Another important issue in economic analysis of climate change
policies is valuing domestic and international benefits. U.S. GHG
reductions are likely to yield both domestic and global benefits.
Typically, because the benefits and costs of most environmental
regulations are predominantly domestic, EPA focuses on benefits that
accrue to the U.S. population when quantifying the impacts of domestic
regulation. However, OMB's guidance for economic analysis of federal
regulations specifically allows for consideration of international
effects.\76\
---------------------------------------------------------------------------

    \76\ OMB (2003), page 15.
---------------------------------------------------------------------------

    GHGs are global pollutants. Economic principles suggest that the
full costs to society of emissions should be considered in order to
identify the policy that maximizes the net benefits to society, i.e.,
achieves an efficient outcome (Nordhaus, 2006).\77\ Estimates of global
benefits capture more of the full value to society than domestic
estimates and can therefore help guide policies towards higher global
net benefits for GHG reductions.\78\ Furthermore, international effects
of climate change may also affect domestic benefits directly and
indirectly to the extent U.S. citizens value international impacts
(e.g., for tourism reasons, concerns for the existence of ecosystems,
and/or concern for others); U.S. international interests are affected
(e.g., risks to U.S. national security, or the U.S. economy from
potential disruptions in other nations); and/or domestic mitigation
decisions affect the level of mitigation and emissions changes in
general in other countries (i.e, the benefits realized in the U.S. will
depend on emissions changes in the U.S. and internationally). The
economics literature also suggests that policies based on direct
domestic benefits will result in little appreciable

[[Page 44416]]

reduction in global GHGs (e.g., Nordhaus, 1995).\79\
---------------------------------------------------------------------------

    \77\ Nordhaus, W., 2006, ``Paul Samuelson and Global Public
Goods,'' in M. Szenberg, L. Ramrattan, and A. Gottesman (eds),
Samuelsonian Economics, Oxford.
    \78\ Both the United Kingdom and the European Commission
following these economic principles in consideration of the global
social cost of carbon (SCC) for valuing the benefits of GHG emission
reductions in regulatory impact assessments and cost-benefit
analyses (Watkiss et al, 2006).
    \79\ Nordhaus, William D. (1995). ``Locational Competition and
the Environment: Should Countries Harmonize Their Environmental
Policies?'' in Locational Competition in the World Economy,
Symposium 1994, ed., Horst Siebert, J. C. B. Mohr (Paul Siebeck),
Tuebingen, 1995.
---------------------------------------------------------------------------

    These economic principles suggest that global benefits should also
be considered when evaluating alternative GHG reduction policies.\80\
In the literature, there are a variety of global marginal benefits
estimates (see the Tol, 2005, and Tol, 2007, meta analyses).\81\ A
marginal benefit is the estimated monetary benefit for each additional
unit of carbon dioxide emissions reduced in a particular year.\82\
---------------------------------------------------------------------------

    \80\ Recently, the National Highway Traffic Safety
Administration (NHTSA) proposed a new rulemaking for average fuel
economy standards for passenger cars and light trucks that is based
on domestic marginal benefit estimates for carbon dioxide
reductions. See section V.A.7.l.(iii) ``Economic value of reductions
in CO₂ emissions'' (p. 24413) of Vol. 73 of the Federal
Registry. Department of Transportation, National Highway Traffic
Safety Administration, 49 CFR Parts 523, 531, 533, 534, 536 and 537
[Docket No. NHTSA-2008 -0089], RIN 2127-AK29, Average Fuel Economy
Standards: Passenger Cars and Light Trucks, Model Years 2011-2015,
http://www.regulations.gov/fdmspublic/component/
main?main=DocumentDetail&;o=0900006480541adc.
    \81\ Tol, Richard, 2005. The marginal damage costs of carbon
dioxide emissions: an assessment of the uncertainties. Energy Policy
33: 2064-2074. Tol, Richard, 2007. The Social Cost of Carbon:
Trends, Outliers and Catastrophes. Economics Discussion Papers
Discussion Paper 2007-44, September 19, 2007. Tol (2007) has been
published on-line with peer review comments (www.economics-
ejournal.org/economics/discussionpapers/2007-44).
    \82\ This is sometimes referred to as the social cost of carbon,
which specifically is defined as the net present value of the change
in climate change impacts over the atmospheric life of the
greenhouse gas and the resulting climate inertia associated with one
additional net global metric ton of carbon emitted to the atmosphere
at a particular point in time.
---------------------------------------------------------------------------

    Based on the characteristics of GHGs and the economic principles
that follow, EPA developed ranges of global and U.S. marginal benefits
estimates. The estimates were developed as part of the work evaluating
potential GHG emission reductions from motor vehicles and their fuels
under Executive Order 13432. However, it is important to note at the
outset that the estimates are incomplete since current methods are only
able to reflect a partial accounting of the climate change impacts
identified by the IPCC (discussed more below). Also, as noted above,
domestic estimates omit potential impacts on the United States (e.g.,
economic or national security impacts) resulting from climate change
impacts in other countries. The global estimates were developed from a
survey analysis of the peer reviewed literature (i.e. meta analysis).
U.S. estimates, and a consistent set of global estimates, were
developed from a single model and are highly preliminary, under
evaluation, and likely to be revised.
    The range of estimates is wide due to the uncertainties described
above relating to socio-economic futures, climate responsiveness,
impacts modeling, as well as the choice of discount rate. For instance,
for 2007 emission reductions and a 2% discount rate the global meta
analysis estimates range from $-3 to $159/tCO₂, while the
U.S. estimates range from $0 to $16/tCO₂. For 2007 emission
reductions and a 3% discount rate, the global meta-estimates range from
$-4 to $106/tCO₂, and the U.S. estimates range from $0 to
$5/tCO₂.\83\ The global meta analysis mean values for 2007
emission reductions are $68 and $40/tCO₂ for discount rates
of 2% and 3% respectively (in 2006 real dollars) while the domestic
mean value from a single model are $4 and $1/tCO₂ for the
same discount rates. The estimates for future year emission changes
will be higher as future marginal emissions increases are expected to
produce larger incremental damages as physical and economic systems
become more stressed as the magnitude of climate change increases.\84\
---------------------------------------------------------------------------

    \83\ See the Technical Support Document on Benefits of Reducing
GHG Emissions for global estimates consistent with the U.S.
estimates in the text and for a comparison to the Tol (2005) meta
analysis peer reviewed estimates. Tol (2005) estimates were cited in
NHTSA's proposed rule and by the 9th U.S. Circuit Court (Center for
Biodiversity v. NHTSA, F. 3d. 9th Cir., Nov. 15, 2007).
    \84\ Note that, except for illustrative purposes, marginal
benefits estimates in the peer reviewed literature do not use
consumption discount rates as high as 7%.
---------------------------------------------------------------------------

    The current state-of-the-art for estimating benefits is also
important to consider when evaluating policies. There are significant
partially unquantified and omitted impact categories not captured in
the estimates provided above. The IPCC WGII (2007) concluded that
current estimates are ``very likely'' to be underestimated because they
do not include significant impacts that have yet to be monetized.\85\
Current estimates do not capture many of the main reasons for concern
about climate change, including non-market damages (e.g., species
existence value and the value of having the option for future use), the
effects of climate variability, risks of potential extreme weather
(e.g., droughts, heavy rains and wind), socially contingent effects
(such as violent conflict or humanitarian crisis), and potential long-
term catastrophic events. Underestimation is even more likely when one
considers that the current trajectory for GHG emissions is higher than
typically modeled, which when combined with current regional population
and income trajectories that are more asymmetric than typically
modeled, imply greater climate change and vulnerability to climate
change.
    Finally, with projected increasing changes in climate, some types
of potential climate change impacts may occur suddenly or begin to
increase at a much faster rate, rather than increasing gradually or
smoothly. In this case, there are likely to be jumps in the functioning
of species and ecosystems, the frequency and intensity of extreme
conditions (e.g., heavy rains, forest fires), and the occurrence of
catastrophic events (e.g., collapse of the West Antarctic Ice Sheet).
As a result, different approaches are necessary for quantifying the
benefits of ``small'' (incremental) versus ``large'' (non-incremental)
reductions in global GHGs. Marginal benefits estimates, like those
presented above, can be useful for estimating benefits for small
changes in emissions. However, for large changes in emissions, a more
comprehensive assessment of impacts would be needed to capture changes
in economic and biophysical dynamics and feedbacks in response to the
policy. Even small reductions in global GHG emissions are expected to
reduce climate change risks, including catastrophic risks.
---------------------------------------------------------------------------

    \85\ IPCC WGII, 2007. In the IPCC report, ``very likely'' was
defined as a greater than 90% likelihood based on expert judgment.
---------------------------------------------------------------------------

    EPA solicits comment on the appropriateness of using U.S. and
global values in quantifying the benefits of GHG reductions and the
appropriate application of benefits estimates given the state of the
art and overall uncertainties. We also seek comment on our estimates of
the global and U.S. marginal benefits of GHG emissions reductions that
EPA has developed, including the scientific and economic foundations,
the methods employed in developing the estimates, the discount rates
considered, current and proposed future consideration of uncertainty in
the estimates, marginal benefits estimates for non-CO₂ GHG
emissions reductions, and potential opportunities for improving the
estimates. We are also interested in comments on methods for
quantifying benefits for non-incremental reductions in global GHG
emissions.
5. Energy Security
    In recent actions, both EPA and NHTSA have considered other
benefits of a regulatory program that, though not directly
environmental, can result from compliance with the program and may

[[Page 44417]]

be quantified.\86\ One of these potential benefits, related to the
transportation sector, is increased energy security due to reduced oil
imports. It is clear that both financial and strategic risks can result
within the U.S. economy if there is a sudden disruption in the supply
or a spike in the costs of petroleum. Conversely, actions that promote
development of lower carbon fuels that can substitute for petroleum or
technologies that more efficiently combust petroleum during operation
can result in reduced U.S. oil imports, and can therefore reduce these
financial and strategic risks. This reduction in risks is a measure of
improved energy security and represents a benefit to the U.S. As the
Agency evaluates potential actions to reduce GHGs from the U.S.
economy, it intends to also consider the energy security impacts
associated with these actions.
---------------------------------------------------------------------------

    \86\ The EPA has worked with Oak Ridge National Laboratory to
develop a methodology that quantifies energy security benefits
associated with the reduction of imported oil. This methodology was
used to support the EPA's 2007 Renewable Fuels Standards Rulemaking
and NHTSA's 2008 proposed Average Fuel Economy Standards for
Passenger Cars and Light Trucks Rulemaking for Model Years 2001--
2015.
---------------------------------------------------------------------------

6. Interactions With Other Policies
    Climate change and GHG mitigation policies will likely affect most
biophysical and economic systems, and will therefore affect policies
related to these systems. For example, as previously mentioned, climate
change will affect air quality and GHG mitigation will affect criteria
pollutant emissions. These effects will need to be evaluated, both in
the context of economic costs and benefits, as well as policy design in
order to exploit synergies and avoid inefficiencies across policies.
Non-climate policies, whether focused on traditional air pollutants,
energy, transportation, or other areas, can also affect baselines and
mitigation opportunities for climate policies. For instance, energy
policies can change baseline GHG emissions and the development path of
particular energy technologies, potentially affecting the GHG
mitigation objectives of climate policies as well as changing the
relative costs of mitigation technologies. EPA seeks comment on
important policy interactions.
7. Integrating Economic and Noneconomic Considerations
    While economics can answer questions about the cost effectiveness
and efficiency of policies, judgments about the appropriate mitigation
policy, potential climate change impacts, and even the discount rate
can be informed by economics and science but also involve important
policy, legal, and ethical questions. The ultimate choice of a global
climate stabilization target may be a policy choice that incorporates
both economic and non-economic factors, while the choice of specific
implementation strategies may be based on effectiveness criteria.
Furthermore, other quantitative analyses are generally used to support
the development of regulations. Distributional analyses, environmental
justice analyses, and other analyses can be informative. For example,
to the extent that climate change affects the distribution of wealth or
the distribution of environmental damages, then climate change
mitigation policies may have significant distributional impacts, which
may in some cases be more important than overall efficiency or net
benefits. EPA seeks comment on how to adequately inform economic
choices, as well as the broader policy choices, associated with GHG
mitigation policies.

IV. Clean Air Act Authorities and Programs

    In developing a response to the Massachusetts decision, EPA
conducted a thorough review of the CAA to identify and assess all of
the Act's provisions that might be applied to GHG emissions. Although
the Massachusetts decision addresses only CAA section 202(a)(1), which
authorizes new motor vehicle emission standards, the Act contains a
number of provisions that could conceivably be applied to GHGs
emissions. EPA's review of these provisions and their interconnections
indicated that a decision to regulate GHGs under section 202(a) or
another CAA provision could or would lead to regulation under other CAA
provisions. This section of the notice provides an overview of the CAA
and examines the various interconnections among CAA provisions that
could lead to broad regulation of GHG emission sources under the Act.

A. Overview of the Clean Air Act

    The CAA provides broad authority to combat air pollution. Cars,
trucks, construction equipment, airplanes, and ships, as well as a
broad range of electric generation, industrial, commercial and other
facilities, are subject to various CAA programs. Implementation of the
Act over the past four decades has resulted in significant reductions
in air pollution at the same time the nation's economy has grown.
    As more fully examined in Section VII of this notice, the CAA
provides three main pathways for regulating stationary sources of air
pollutants. They include, in order of their appearance in the Act,
national ambient air quality standards (NAAQS) and state plans for
implementing those standards (SIPs); performance standards for new and
existing stationary sources; and hazardous air pollutant standards for
stationary sources. In addition, the Prevention of Significant
Deterioration (PSD) program requires preconstruction permitting and
emission controls for certain new and modified major stationary
sources, and the Title V program requires operating permits for all
major stationary sources.
    Section 108 of the CAA authorizes EPA to list air pollutants that
are emitted by many sources and that cause or contribute to air
pollution problems such as ozone (smog) and particulate matter (soot).
For every pollutant listed, EPA is required by section 109 to set NAAQS
that are ``requisite'' to protect public health and welfare. EPA may
not consider the costs of meeting the NAAQS in setting the standards.
Under section 110, every state develops and implements plans for
meeting the NAAQS by applying enforceable emission control measures to
sources within the state. The Act's requirements for SIPs are more
detailed and stringent for areas not meeting the standards
(nonattainment areas) than for areas meeting the standards (attainment
areas). Costs may be considered in implementing the standards. States
are aided in their efforts to meet the NAAQS by federal emissions
standards for mobile sources and major categories of stationary sources
issued under other sections of the Act.
    Under CAA section 111, EPA establishes emissions performance
standards for new stationary sources and modifications of existing
sources for categories of sources that contribute significantly to
harmful air pollution. These new source performance standards (NSPS)
reduce emissions of air pollutants addressed by NAAQS, but can be
issued regardless of whether there is a NAAQS for the pollutants being
regulated. NSPS requirements for new sources help ensure that when
large sources of air pollutants are built or modified, they apply
available emission control technologies and strategies.
    When EPA establishes a NSPS for a pollutant, section 111(d) calls
upon states to issue a standard for existing sources in the regulated
source category except in two circumstances. First, section 111(d)
prohibits regulation of a NAAQS pollutant. Second, ``where a source
category is being regulated under section 112, a section 111(d)
standard of performance cannot be established to

[[Page 44418]]

address any HAP listed under section 112(b) that may be emitted from
that particular source category.''\87\ In effect, existing source NSPS
provides a ``regulatory safety net'' for pollutants not otherwise
subject to major regulatory programs under the CAA. Section 111
provides EPA and states with significant discretion concerning the
sources to be regulated and the stringency of the standards, and allows
consideration of costs in setting NSPS.
---------------------------------------------------------------------------

    \87\ See 70 FR 15994, 16029-32 (Mar. 29, 2005).
---------------------------------------------------------------------------

    CAA section 112 provides EPA with authority to list and issue
national emissions standards for hazardous air pollutants (HAPs) from
stationary sources. HAPs are broadly defined as pollutants that
present, or may present, a threat of adverse human or environmental
effects. HAPs include substances which are, or may reasonably be
anticipated to be, carcinogenic, mutagenic, neurotoxic or acutely or
chronically toxic. Section 112 contains low emissions thresholds for
regulation in view of its focus on toxic pollutants, and requires
regulation of all major sources of HAPs. Section 112 also provides for
``maximum achievable control technology'' (MACT) standards for major
sources, limiting consideration of cost.
    The PSD program under Part C of Title I of the Act is triggered by
regulation of a pollutant under any other section of the Act except for
sections 112 and 211(o). As mentioned previously in this notice, under
this program, new major stationary sources and modifications at
existing major stationary sources undergo a preconstruction permitting
process and install best available control technology (BACT) for each
regulated pollutant. These basic requirements apply regardless of
whether a NAAQS exists for the pollutant; additional PSD requirements
apply in the event of a NAAQS. The PSD program's control requirements
help prevent large new and modified sources of air pollutants from
significantly degrading the air quality in clean air areas. A similar
program, called ``new source review,'' ensures that new or modified
large sources in areas not meeting the NAAQS do not make it more
difficult for the areas to eventually attain the air quality standards.
    Title II of the CAA provides comprehensive authority for regulating
mobile sources of air pollutants. As more fully described in Section VI
of this notice, Title II authorizes EPA to address all categories of
mobile sources and take an integrated approach to regulation by
considering the unique aspects of each category, including passenger
vehicles, trucks and nonroad vehicles, as well as the fuels that power
them. Title II requires EPA to consider technological feasibility,
costs, safety and other factors in setting standards, and gives EPA
discretion to set technology-forcing standards as appropriate. In
addition, section 211(o) of the Act establishes the renewable fuel
standard (RFS) program, which was recently strengthened by EISA to
require substantial increases in the use of renewable fuels, including
renewable fuels with significantly lower lifecycle GHG emissions than
the fossil fuel-based fuels they replace.\88\ The CAA's mobile source
authorities work in tandem with the Act's stationary source authorities
to help protect public health and the environment from air pollution.
---------------------------------------------------------------------------

    \88\ As explained further below, EISA provides that regulation
of renewable fuels based on lifecycle GHG emissions does not trigger
any other regulation of GHGs under the CAA.
---------------------------------------------------------------------------

    Title VI of the CAA authorizes EPA to take various actions to
protect stratospheric ozone, a layer of ozone high in the atmosphere
that helps protect the Earth from harmful UVB radiation. As discussed
in Section VIII of this notice, section 615 provides broad authority to
regulate any substance, practice, process or activity that may
reasonably be anticipated to affect the stratosphere and that effect
may reasonably be anticipated to endanger public health or welfare.

B. Interconnections Among Clean Air Act Provisions

    The provisions of the CAA are interconnected in multiple ways such
that a decision to regulate one source category of GHGs could or would
lead to regulation of other source categories of GHGs. As described in
detail below, there are several provisions in the CAA that contain
similar endangerment language. An endangerment finding for GHGs under
one provision of the Act could thus have ramifications under other
provisions of the Act. In addition, CAA standards applicable to GHGs
for one category of sources could trigger PSD requirements for other
categories of sources that emit GHGs. How a term is interpreted for one
part of the Act could also affect other provisions using the same term.
    These CAA interconnections are by design. As described above, the
Act combats air pollutants in several ways that reflect the nature and
effects of the particular air pollutant being addressed. The Act's
approaches are in many cases complementary and reinforcing, ensuring
that air pollutants emitted by various types of emission sources are
reduced in a manner and to an extent that reflects the relative
contribution of particular categories of sources. The CAA's authorities
are intended to work together to achieve air quality that protects
public health and welfare.
    For GHGs, the CAA's interconnections mean that careful attention
needs to be paid to the consequences and specifics of decisions
regarding endangerment and regulation of any particular category of GHG
sources under the Act. In the case of traditional air pollutants, EPA
and States have generally regulated pollutants incrementally over time,
adding source categories or program elements as evolving circumstances
make appropriate. In light of the broad variety and large number of GHG
sources, any decision to regulate under the Act could lead, relatively
quickly, to more comprehensive regulation of GHG sources under the Act.
A key issue to consider in examining the Act's provisions and their
interconnections is the extent to which EPA may choose among and/or
tailor the CAA's authorities to implement a regulatory program that
makes sense for GHGs, given the unique challenges and opportunities
that regulating them would present.
    This section of the notice explores these interconnections, and
later sections explain how each CAA provision might apply to GHGs.
1. Similar Endangerment Language Is Found in Numerous Sections of the
Clean Air Act
    The Supreme Court's decision in Massachusetts v. EPA requires EPA
to address whether GHG emissions from new motor vehicles meet the
endangerment test of CAA section 202(a)(1). That section states:

    [t]he Administrator shall by regulation prescribe (and from time
to time revise) * * * standards applicable to the emissions of any
air pollutant from any class or classes of new motor vehicles or new
motor vehicle engines, which in his judgment cause, or contribute
to, air pollution which may reasonably be anticipated to endanger
public health or welfare.

CAA section 202(a)(1). If the Administrator makes a positive
endangerment determination for GHG emissions from new motor vehicles,
he must regulate those GHG emissions under section 202(a) of the Act.
    Similar endangerment language is found in numerous sections of the
CAA, including sections 108, 111, 112, 115, 211, 213, 231 and 615. For
example, CAA section 108(a)(1) (regarding listing pollutants to be
regulated by NAAQS)

[[Page 44419]]

states, ``[T]he Administrator shall * * * publish, and shall from time
to time thereafter revise, a list which includes each air pollutant (A)
emissions of which, in his judgment, cause or contribute to air
pollution which may reasonably be anticipated to endanger public health
or welfare * * *'' CAA section 111(b)(1)(A) (regarding listing source
categories to be regulated by NSPS) states: ``[The Administrator] shall
include a category of sources in such list if in his judgment it
causes, or contributes significantly to, air pollution which may
reasonably be anticipated to endanger public health or welfare.''\89\
---------------------------------------------------------------------------

    \89\ Other CAA endangerment provisions read as follows:
    CAA section 115 (regarding international air pollution) states:
``Whenever the Administrator, upon receipt of reports, surveys or
studies from any duly constituted international agency has reason to
believe that any air pollutant or pollutants emitted in the United
States cause or contribute to air pollution which may reasonably be
anticipated to endanger public health or welfare in a foreign
country or whenever the Secretary of State requests him to do so
with respect to such pollution which the Secretary of State alleges
is of such a nature, the Administrator shall give formal
notification thereof to the Governor of the State in which such
emissions originate.''
    CAA section 211(c)(1) (regarding regulating fuels and fuel
additives) states: ``The Administrator may, * * * [regulate fuels or
fuel additives] (A) if in the judgment of the Administrator any
emission product of such fuel or fuel additive causes, or
contributes, to air pollution which may reasonably be anticipated to
endanger public health or welfare, (B) * * *''
    CAA section 213(a)(4) (regarding regulating nonroad engines)
states: ``If the Administrator determines that any emissions not
referred to in paragraph 2 [regarding CO, NO_X and VOC
emissions] from new nonroad engines or vehicles significantly
contribute to air pollution which may reasonably be anticipated to
endanger public health or welfare, the Administrator may promulgate
* * * standards applicable to emissions from those classes or
categories of new nonroad engines and new nonroad vehicles (other
than locomotives) which in the Administrator's judgment cause, or
contribute to, such air pollution, * * *''.
    CAA section 231 (regarding setting aircraft standards) states:
``The Administrator shall * * * issue proposed emissions standards
applicable to the emission of any air pollutant from any class or
classes of aircraft engines which in his judgment causes, or
contributes to, air pollution which may reasonably be anticipated to
endanger public health or welfare.''
    CAA section 615 (regarding protection of stratospheric ozone)
states: ``If, in the Administrator's judgment, any substance,
practice, process, or activity may reasonably be anticipated to
affect the stratosphere, especially ozone in the stratosphere, and
such effect may reasonably be anticipated to endanger public health
or welfare, the Administrator shall promptly promulgate regulations
respecting the control of such substance, practice, process, or
activity * * *''
---------------------------------------------------------------------------

    While no two endangerment tests are precisely the same, they
generally call on the Administrator of EPA to exercise his or her
judgment regarding whether a particular air pollutant or source
category causes or contributes to air pollution which may reasonably be
anticipated to endanger public health or welfare. For provisions
containing endangerment language, a positive finding of endangerment is
a prerequisite for regulation under that provision.\90\ The precise
effect of a positive or negative finding depends on the specific terms
of the provision under which it is made. For some provisions, a
positive endangerment finding triggers an obligation to regulate (e.g.,
section 202(a)(1)), while for other provisions, a positive finding
allows the Agency to regulate in its discretion (e.g., section 213). In
some cases, other criteria must also be met to authorize or require
regulation (e.g., section 108). Each of these sections is discussed in
more detail later in this notice.
---------------------------------------------------------------------------

    \90\ As defined by the CAA, ``air pollutant'' includes virtually
any substance or material emitted into the ambient air. Given the
breadth of that term, many CAA provisions require the Administrator
to determine whether a particular air pollutant causes or
contributes to an air pollution problem as a prerequisite to
regulating emissions of that pollutant.
---------------------------------------------------------------------------

2. Potential Impact Cross the Clean Air Act From a Positive or Negative
Endangerment Finding or Regulation of GHGs Under the Act
a. Potential Impact on Sections Containing Similar Endangerment
Language
    One important issue is whether a positive or negative endangerment
finding under one section of the CAA (e.g., under section 202(a) in
response to the ICTA petition remand) would necessarily or
automatically lead to similar findings under other provisions of the
Act containing similar language. Even though CAA endangerment tests
vary to some extent, an endangerment finding under one provision could
have some bearing on whether endangerment could or should be found
under other CAA provisions, depending on their terms and the facts at
issue. EPA request comment on the extent to which an endangerment
finding under any section of the CAA would lead EPA to make a similar
endangerment finding under another provision.
    In discussing the implications of making a positive endangerment
finding under any CAA section, we use the actual elements of the
endangerment test in section 202(a) for new motor vehicles as an
example. The section 202(a) endangerment test asks two distinct
questions--
    (1) whether the air pollution at issue may reasonably be
anticipated to endanger public health or welfare, and
    (2) whether emissions from new motor vehicles cause or contribute
to that air pollution. The first question is generic and looks at
whether the type of air pollution at issue endangers public health or
welfare. The second question is specific to motor vehicles, and
considers the contribution of motor vehicle emissions to the particular
air pollution problem. EPA must answer both questions in the
affirmative for the Agency to regulate under section 202(a) of the Act.
    A finding of endangerment under one section of the Act would not by
itself constitute a complete finding of endangerment under any other
section of the CAA. How much of a precedent an endangerment finding
under one CAA provision would be for other CAA provisions would depend
on the basis for the finding, the statutory tests for making findings,
and the facts. For example, the two-part endangerment test in section
202(a) (motor vehicles) is similar to that in sections 211(c)(1)
(highway and nonroad fuels) and 231(a)(2) (aircraft). An affirmative
finding under section 202(a) on the first part of the test--whether the
air pollution at issue endangers public health or welfare--would appear
to satisfy the first part of the test for the other two provisions as
well. However, an affirmative finding on the second part of the test,
regarding the contribution of the particular source category to that
air pollution, would not satisfy the test for the other provisions,
which apply to different source categories. Still, a finding that a
particular source category's emissions cause or contribute to the air
pollution problem would likely establish some precedent for what
constitutes a sufficient contribution for purposes of making a positive
endangerment finding for other source categories.
    Other similarities and differences among endangerment tests are
also relevant. While the first part of the test in sections 213(a)(4)
(nonroad engines and vehicles) and 111(b) (NSPS) is similar to that in
other sections (i.e., whether the air pollution at issue endangers
public health or welfare), the second part of the test in sections
213(a)(4) and 111(b) requires a finding of ``significant''
contribution. In addition, the test under section 111(b) applies to
source categories, not to a particular air pollutant.\91\ Sections 112
and 615 have somewhat different tests.
---------------------------------------------------------------------------

    \91\ As discussed below, EPA has already listed a very wide
variety of source categories under section 111(b)(1)(A).
---------------------------------------------------------------------------

    The extent to which an endangerment finding would set precedent
would also depend on the pollutants at issue. For example, the ICTA
petition to regulate motor vehicles under section 202(a)

[[Page 44420]]

addresses CO₂, CH₄ , N₂O, and HFCs,
while the petitions to regulate GHGs from other mobile source
categories collectively address water vapor, NO_X and black
carbon, as well as CO₂, CH₄, and N₂O.
As further discussed below, the differences in the GHGs emitted by
different types of sources may be relevant to the issue of how to
define ``air pollutant'' for purposes of applying the endangerment
tests.
    In addition, some CAA sections require EPA to act following a
positive endangerment finding, while others do not. In the case of
section 202(a)(1), if we make a positive endangerment finding, we are
required to issue standards applicable to motor vehicle emissions of
the GHGs covered by the finding. Section 231(a) (aircraft) uses similar
mandatory language, while sections 211(c)(1) (highway and nonroad fuel)
and 213(a)(4) (nonroad engines and vehicles) authorize but do not
require the issuance of regulations. Section 108 (NAAQS pollutants)
requires that EPA list a pollutant under that section if a positive
endangerment finding is made and two other criteria are met.
    In sum, a positive or negative endangerment finding for GHG
emissions under one provision of the Act could have a significant and
direct impact on decisions under other CAA sections containing similar
endangerment language. EPA requests comment on the interconnections
between the CAA endangerment tests and the impact that a finding under
one provision of the Act would have for other CAA provisions.
b. Potential Impact on PSD Program
    Another important issue is the potential for a decision to regulate
GHGs for mobile or stationary sources to automatically trigger
additional permitting requirements for stationary sources under the PSD
program. As explained previously and in detail in Section VII of this
notice, the main element of the PSD program under Part C of Title I of
the Act is the requirement that a PSD permit be obtained prior to
construction of any new major source or any major modification at an
existing major source. Such a permit must contain emissions limitations
based on BACT for each pollutant subject to regulation under the Act.
EPA does not interpret the PSD program provisions to apply to GHG at
this time, but any requirement to control CO₂ or other GHGs
promulgated by EPA under other provisions of the CAA would make parts
of the PSD program applicable to any additional air pollutant(s) that
EPA regulates in this manner.
    The PSD program applies to each air pollutant (other than a HAP)
that is ``subject to regulation under the Act'' within the meaning of
sections 165(a)(4) and 169(3) of the Clean Air Act and EPA's
regulations.\92\ As a practical matter, the identification of
pollutants subject to the PSD program is driven by the BACT requirement
because this requirement applies to the broadest range of pollutants.
Under EPA's PSD program regulations, BACT is required for ``each
regulated NSR pollutant.'' 40 CFR 52.21(j)(2)-(3). EPA has defined this
term to include pollutants that are regulated under a NAAQS or NSPS, a
class I or II substance under Title VI of the Act, or ``[a]ny pollutant
otherwise subject to regulation under the Act.'' See 52.21(b)(50).\93\
Similarly, the determination of whether a source is a major source
subject to PSD is based on whether the source emits more than 100 or
250 tons per year (depending on the type of source) of one or more
regulated pollutants.\94\
---------------------------------------------------------------------------

    \92\ Section 112(b)(6) precludes listed HAPs from the PSD
program. Section 210(b) of EISA provides that nothing in section
211(o) of the Act, or regulations issued pursuant to that
subsection, ``shall affect or be construed to affect the regulatory
status of carbon dioxide or any other greenhouse gas, or to expand
or limit regulatory authority regarding carbon dioxide or any other
greenhouse gas, for purposes of other provisions (including section
165) of this Act.''
    \93\ This definition reflects EPA's interpretation of the phrase
``each pollutant subject to regulation under the Act'' that is used
in the provisions in the Clean Air Act that establish the BACT
requirement. Since this statutory language (as implemented in the
definition of ``regulated NSR pollutant'') can apply to additional
pollutants that are not also subject to a NAAQS, the scope of the
BACT requirement determines the overall range of pollutants that are
subject to the PSD permitting program.
    \94\ Under the relevant regulations, a major stationary source
is determined by its emissions of ``any regulated NSR pollutant.''
See 40 CFR 52.21(b)(1)(i). Thus, the emissions that are considered
in identifying a major source are determined on the basis of the
same definition that controls the applicability of the BACT.
---------------------------------------------------------------------------

    EPA has historically interpreted the phrase ``subject to regulation
under the Act'' to describe air pollutants subject to CAA statutory
provisions or regulations that require actual control of emissions of
that pollutant.\95\ PSD permits have not been required to contain BACT
emissions limit for GHGs because GHGs (and CO₂ in
particular) have not been subject to any CAA provisions or EPA
regulations issued under the Act that require actual control of
emissions.\96\ Although CAA section 211(o) now targets GHG emissions,
EISA provides that neither it nor implementing regulations affect the
regulatory status of GHGs under the CAA. In the absence of statutory or
regulatory requirements to control GHG emissions under the Act, a
stationary source need not consider those emissions when determining
its major source status.
---------------------------------------------------------------------------

    \95\ 43 FR 26388, 26397 (June 19, 1978); Gerald E. Emison,
Director, Office of Air Quality Planning and Standards,
Implementation of North County Resource Recovery PSD Remand (Sept.
22, 1987) (footnote on the first page).
    \96\ See briefs filed before the Environmental Appeal Board on
behalf of specific EPA offices in challenges to the PSD permits for
Deseret Power Electric Cooperative (PSD Appeal No. 07-03) and
Christian County Generation LLC (PSD Appeal No. 07-01), as well as
the Response to Public Comments on Draft Air Pollution Control
Prevention of Significant Deterioration (PSD) Permit to Construct
[for Deseret Power Electric Cooperative], Permit No. PSD-OU-0002-
04.00 (August 30, 2007), at 5-6, available at http://www.epa.gov/
region8/air/permitting/deseret.html. EPA has not previously
interpreted the BACT requirement to apply to air pollutants that are
only subject to requirements to monitor and report emissions. See,
67 FR 80186, 80240 (Dec. 31, 2002); 61FR 38250, 38310 (July 31,
1996); In Re Kawaihae Cogeneration Project 7 E.A.D. 107, 132 (EAB
1997); Inter-power of New York, 5 E.A.D. 130, 151 (EAB 1994);
Memorandum from Jonathan Z. Cannon, General Counsel to Carol M.
Browner, Administrator, entitled EPA's Authority to Regulate
Pollutants Emitted by Electric Power Generation Sources (April 10,
1998) (emphasis added); Memorandum from Lydia N. Wegman, Deputy
Director, Office of Air Quality Planning and Standards, entitled
Definition of Regulated Air Pollutant for Purposes of Title V, at 5
(April 26, 1993).
---------------------------------------------------------------------------

    The Supreme Court's conclusion that GHGs are ``air pollutants''
under the CAA did not automatically make these pollutants subject to
the PSD program. A substance may be an ``air pollutant'' under the Act
without being regulated under the Act. The Supreme Court directed the
EPA Administrator to determine whether GHG emissions from motor
vehicles meet the endangerment test of CAA section 202(a). A positive
finding of endangerment would require the Administrator to then set
standards applicable to GHG emissions from motor vehicles under the
Act. The positive finding itself would not constitute a regulation
requiring actual control of emissions. GHGs would become regulated
pollutants under the Act if and when EPA subjects GHGs to control
requirements under a CAA provision other than sections 112 and 211(o).
c. Definition of ``Air Pollutant''
    Another way in which a decision to regulate GHGs under one section
of the Act could impact other sections of the Act involves how the term
``air pollutant'' is defined as part of the endangerment analysis. As
described above, many of the Act's endangerment tests require a two-
part analysis: Whether the air pollution at issue may reasonably be
anticipated to endanger public health or welfare, and whether emissions
of particular air pollutants cause or contribute to that air pollution.

[[Page 44421]]

As discussed in more detail in the following sections, what GHGs might
be defined as an ``air pollutant'' and whether those GHGs are treated
individually or as a group could impact EPA's flexibility to define the
GHGs as air pollutants elsewhere in the CAA.
    For example, as noted above, how EPA defines GHGs as air pollutants
in making any positive endangerment finding could carry over into
implementation of the PSD program. If EPA defines each individual GHG
as a separate air pollutant in making a positive endangerment finding,
then each GHG would be considered individually as a ``regulated NSR
pollutant'' in the PSD program. On the other hand, if EPA defines the
group of GHGs as an air pollutant, then the PSD program would need to
treat the GHGs in the same manner--as a group. As discussed in more
detail below, there are flexibilities and considerations under various
approaches. One question is whether we could or should define GHGs as
an ``air pollutant'' one way under one section of the Act (e.g.,
section 202) and another way under another section (e.g., section 231).
See, e.g., Environmental Defense v. Duke Energy Corp., 127 S.Ct. 1423,
1432 (2007) (explaining that the general presumption that the same term
has the same meaning is not rigid and readily gives way to context).
Another question is whether having different definitions of ``air
pollutant'' would result in both definitions applying to the PSD
program, and whether that result would mean that any flexibilities
gained under one definition would be lost with the application of the
second.
    Another consideration, noted above, is that different source
categories emit different GHGs. This fact could impact the definition
of ``air pollutant'' more broadly. EPA requests comment on the issues
raised in this section, to assist the Agency as it considers the
implications of how to define a GHG ``air pollutant'' for the first
time under any section of the Act.
2. Relationships Among Various Stationary Source Programs
    As a result of other interactions among various CAA sections, a
decision to act under one part of the CAA may preclude action under
another part of the Act. These interactions reflect the Act's different
regulatory treatment of pollutants meeting different criteria, and
prevent duplicative regulation. For instance, listing a pollutant under
section 108(a), which leads to setting a NAAQS and developing SIPs for
the pollutant, generally precludes listing the same air pollutant as a
HAP under section 112(b), which leads to every major source of a listed
HAP having to comply with MACT standards for the HAP. CAA section
112(b)(2).\97\ Listing an air pollutant under section 108(a) also
preludes regulation of that air pollutant from existing sources under
section 111(d), which is intended to provide for regulation of air
pollutants not otherwise subject to the major regulatory programs under
the Act. CAA section 111(d)(1)(A).
---------------------------------------------------------------------------

    \97\ ``No air pollutant which is listed under section 108(a) may
be added to the list under this section, except that the prohibition
of this sentence shall not apply to any pollutant which
independently meets the listing criteria of this paragraph and is a
precursor to a pollutant which is listed under section 108(a) or to
any pollutant which is in a class of pollutants listed under such
section.''
---------------------------------------------------------------------------

    Similarly, regulation of a substance under Title VI precludes
listing that substance as a HAP under section 112(b) based solely on
the adverse effects on the environment of that air pollutant. CAA
section 112(b)(2). Moreover, listing an air pollutant as a HAP under
section 112(b) generally precludes regulation of that air pollutant
from existing sources under section 111(d). CAA section
111(d)(1)(A).\98\ Finally, section 112(b)(6) provides that the
provisions of the PSD program ``shall not apply to pollutants listed
under [section 112].'' CAA section 112(b)(6), 42 U.S.C. 7412(b)(6)
---------------------------------------------------------------------------

    \98\ However, see 70 FR 15994, 16029-32 (2005) (explaining EPA's
interpretation of the conflicting amendments to section 111(d)
regarding HAPs).
---------------------------------------------------------------------------

V. Endangerment Analysis and Issues

    In this section, we present our work to date on an endangerment
analysis in response to the Supreme Court's decision in Massachusetts
v. EPA. As explained previously, the Supreme Court remanded EPA's
denial of the ICTA petition and ruled that EPA must either decide
whether GHG emissions from new motor vehicles cause or contribute to
air pollution which may reasonably be anticipated to endanger public
health or welfare, or explain why scientific uncertainty is so profound
that it prevents making a reasoned judgment on such a determination.
    In response to the remand, EPA analyzed synthesis reports and
studies on how elevated concentrations of GHGs in the atmosphere, and
other factors, contribute to climate change, and how climate change is
affecting, and may affect in the future, human health and welfare,
primarily within the United States. We also analyzed direct GHG effects
on human health and welfare, i.e., those effects from elevated
concentrations of GHGs that do not occur via climate change. This
information, summarized briefly below, is contained in the Endangerment
Technical Support Document found in the docket for today's notice. In
addition, we compiled information concerning motor vehicle GHG
emissions to assess whether motor vehicles cause or contribute to
elevated concentrations of GHGs in the atmosphere. Information on motor
vehicle emissions is contained in the Section 202 Technical Support
Document, also found in the docket.
    As discussed above, making an endangerment finding under one
section of the CAA has implications for other sections of the Act. In
this ANPR, we consider, and seek comment on these implications and
other questions relevant to making an endangerment finding regarding
GHG emissions.
    This section is organized as follows. Section A discusses the legal
framework for the endangerment analysis. Section B provides information
on how ``air pollution'' could be defined for purposes of the
endangerment analysis, as well as a summary of the science regarding
GHGs and climate change and their effects on health and welfare.
Section C uses the information on emissions of GHGs from the mobile
source categories relevant to the ICTA Petition to frame a discussion
about whether GHGs as ``air pollutants'' ``cause or contribute'' to
``air pollution'' which may reasonably be anticipated to endanger
public health or welfare.

A. Legal Framework

    The endangerment language relevant to the ICTA petition is
contained in section 202(a) of the CAA. As explained previously, it is
similar to endangerment language in many other provisions of the Act
and establishes a two-part test. First, the Administrator must decide
if, in his judgment, air pollution may reasonably be anticipated to
endanger public health or welfare. Second, the Administrator must
decide whether, in his judgment, emissions of any air pollutant from
new motor vehicles or engines cause or contribute to this air
pollution.
1. Origin of Current Endangerment and Cause or Contribute Language
    The endangerment language in section 202(a) and other provisions of
the CAA share a common legislative history that sheds light on the
meaning of this language. As part of the 1977 amendments to the CAA,
Congress added or revised endangerment language in various sections of
the Act. The legislative history of those amendments, particularly the
report by the House Committee on Interstate and Foreign Commerce,
provides important information regarding Congress' intent

[[Page 44422]]

when it revised this language. See H.R. Rep. 95-294 (1977), as
reprinted in 4 A Legislative History of the Clean Air Act Amendments of
1977 at 2465 (hereinafter ``LH'').
a. Ethyl Corp. v. EPA
    In revising the endangerment language, Congress relied heavily on
the approach discussed in a federal appeals court opinion interpreting
the pre-1977 version of CAA section 211. In Ethyl Corp v. EPA, 541 F.2d
1 (D.C. Cir. 1976), the en banc (i.e. full) court reversed a 3-judge
panel decision regarding an EPA rule restricting the content of lead in
leaded gasoline.\99\ The en banc court began its opinion by stating:
---------------------------------------------------------------------------

    \99\ At the time of the 1973 rules requiring the reduction of
lead in gasoline, section 211(c)(1)(A) of the CAA stated that the
Administrator may promulgate regulations that control or prohibit
the manufacture, introduction into commerce, offering for sale, or
sale of any fuel or fuel additive for use in a motor vehicle or
motor vehicle engine (A) if any emissions product of such fuel or
fuel additive will endanger the public health or welfare * * * .
    CAA section 211(c)(1)(A) (1970) (emphasis added). The italicized
language in the above quote is the relevant language revised by the
1977 amendments.

    Man's ability to alter his environment has developed far more
rapidly than his ability to foresee with certainty the effects of
---------------------------------------------------------------------------
his alterations.

541 F.2d at 6. After reviewing the relevant facts and law, the full-
court evaluated the statutory language at issue to see what level of
``certainty [was] required by the Clean Air Act before EPA may act.''
Id.
    By a 2-1 vote, the 3-judge panel had held that the statutory
language ``will endanger'' required proof of actual harm, and that the
actual harm had to come from fuels ``in and of themselves.'' Id. at 12.
The en banc court rejected this approach, finding that the term
``endanger'' allowed the Administrator to act when harm is threatened,
and did not require proof of actual harm. Id. at 13. ``A statute
allowing for regulation in the face of danger is, necessarily, a
precautionary statute.'' Id. Optimally, the court held, regulatory
action would not only precede, but prevent, a perceived threat. Id.
    The court also rejected petitioners' argument that any threatened
harm must be ``probable'' before regulation was authorized.
Specifically, the court recognized that danger ``is set not by a fixed
probability of harm, but rather is composed of reciprocal elements of
risk and harm, or probability or severity.'' Id. at 18. Next, the court
held that EPA's evaluation of risk is necessarily an exercise of
judgment, and that the statute did not require a factual finding. Id.
at 24. Thus, ultimately, the Administrator must ``act, in part on
`factual issues,' but largely on choices of policy, on an assessment of
risks, [and] on predictions dealing with matters on the frontiers of
scientific knowledge * * * .'' Id. at 29 (citations omitted). Finally,
the en banc court agreed with EPA that even without the language in
section 202 regarding ``cause or contribute to,'' section 211
authorized EPA to consider the cumulative impact of lead from numerous
sources, not just the fuels being regulated under section 211. Id. at
29-31.
b. The 1977 Clean Air Act Amendments
    The dissent in the original Ethyl Corp decision and the en banc
opinion were of ``critical importance'' to the House Committee which
proposed the revisions to the endangerment language in the 1977
amendments to the CAA. H.R. Rep. 95-294 at 48, 4 LH at 2515. In
particular, the Committee believed the Ethyl Corp decision posed
several ``crucial policy questions'' regarding the protection of public
health and welfare.'' Id.\100\ The Committee addressed those questions
with the endangerment language that now appears in section 202(a) and
several other CAA provisions--``which in [the Administrator's] judgment
cause, or contribute to, air pollution which may reasonably be
anticipated to endanger public health or welfare.''
---------------------------------------------------------------------------

    \100\ The Supreme Court recognized that the current language in
section 202(a)(1) is ``more-protective'' than the 1970 version that
was similar to the section 211 language before the D.C. Circuit in
Ethyl Corp. 127 S.Ct. at 1447, fn 1.
---------------------------------------------------------------------------

    The Committee intended the language to serve several purposes
consistent with the en banc decision in Ethyl Corp.\101\ First, the
phrases ``in his judgment'' and ``in the judgment of the
Administrator'' call for the Administrator to make comparative
assessment of risks and projections of future possibilities, consider
uncertainties, and extrapolate from limited data. Thus, the
Administrator must balance the likelihood of effects with the severity
of the effects in reaching his judgment. The Committee emphasized that
``judgment'' is different from a factual ``finding.'' Importantly,
projections, assessments and estimates must be reasonable, and cannot
be based on a ``crystal ball inquiry.'' Moreover, procedural safeguards
apply (e.g., CAA 307(d)) to the exercise of judgment, and final
decisions are subject to judicial review. Also, the phrase ``in his
judgment'' modifies both phrases ``cause and contribute'' and ``may
reasonably be anticipated'' discussed below. H.R. Rep. 95-294 at 50-51,
4 LH at 2517-18.
---------------------------------------------------------------------------

    \101\ Specifically, the language (1) emphasizes the
precautionary or preventive purpose of the CAA; (2) authorizes the
Administrator to reasonably project into the future and weigh risks;
(3) requires the consideration of the cumulative impact of all
sources; (4) instructs that the health of susceptible individuals,
as well as healthy adults, should be part of the analysis; and (5)
indicates an awareness of the uncertainties and limitations in
information available to the Administrator. H.R. Rep. 95-294 at 49-
50, 4 LH at 2516-17. Congress also wanted to standardize this
language across the various sections of the CAA which address
emissions from both stationary and mobile sources which may
reasonably be anticipated to endanger public health or welfare. H.R.
Rep. 95-294 at 50, 4 LH at 2517; Section 401 of CAA Amendments of
1977.
---------------------------------------------------------------------------

    As the Committee further explained, the phrase ``may reasonably be
anticipated'' builds upon the precautionary and preventative goals
already provided in the use of the term ``endanger.'' Thus, the
Administrator is to assess current and future risks rather than wait
for proof of actual harm. This phrase is also intended to instruct the
Administrator to consider the limitations and difficulties inherent in
information on public health and welfare. H.R. Rep. 95-294 at 51, 4 LH
at 2518.
    Finally, the phrase ``cause or contribute'' ensures that all
sources of the contaminant which contribute to air pollution be
considered in the endangerment analysis (e.g., not a single source or
category of sources). It is also intended to require the Administrator
to consider all sources of exposure to a pollutant (e.g., food, water,
air) when determining risk. Id.
3. Additional Considerations for the ``Cause or Contribute'' Analysis
    While the legislative history sheds light on what should be
considered in making an endangerment finding, it is not clear regarding
what constitutes a sufficient ``contribution'' for purposes of making a
finding. The CAA does not define the concept ``cause or contribute''
and instead requires that the Administrator exercise his judgment when
determining whether emissions of air pollutants cause or contribute to
air pollution. As a result, the Administrator has the discretion to
interpret ``cause or contribute'' in a reasonable manner when applying
it to the circumstances before him.
    The D.C. Circuit has discussed the concept of ``contribution'' in
the context of a CAA section 213 rule for nonroad vehicles. In
Bluewater Network v. EPA, 370 F.3d 1 (2004), industry argued that
section 213(a)(3) requires a finding of a significant contribution
before EPA could regulate, but EPA argued that the CAA requires a
finding only of ``contribution.'' \102\ Id. at 13. The court

[[Page 44423]]

looked at the ``ordinary meaning of `contribute''' when upholding EPA's
reading. After referencing dictionary definitions of contribute,\103\
the court also noted that ``[s]tanding alone, the term has no inherent
connotation as to the magnitude or importance of the relevant `share'
in the effect; certainly it does not incorporate any `significance'
requirement.'' Id.\104\ The court also found relevant the fact that
section 213(a) uses the term ``significant contributor'' in some places
and the term ``contribute'' elsewhere, suggesting that the
``contribute'' language invests the Administrator with discretion to
exercise his judgment regarding what constitutes a sufficient
contribution for the purpose of making an endangerment finding. Id. at
14
---------------------------------------------------------------------------

    \102\ The relevant language in section 213(a)(3) reads ``[i]f
the Administrator makes an affirmative determination under paragraph
(2) the Administrator shall, * * * promulgate (and from time to time
revise) regulations containing standards applicable to emissions
from those classes or categories of new nonroad engines and new
nonroad vehicles (other than locomotives or engines used in
locomotives) which in the Administrator's judgment cause, or
contribute to, such air pollution.'' Notably, CAA section 213(a)(2),
which is referenced in section 213(a)(3), requires that the
``Administrator shall determine * * * whether emissions of carbon
monoxide, oxides of nitrogen, and volatile organic compounds from
new and existing nonroad engines or nonroad vehicles (other than
locomotives or engines used in locomotives) are significant
contributors to ozone or carbon monoxide concentrations in more than
1 area which has failed to attain the national ambient air quality
standards for ozone or carbon monoxide'' (emphasis added).
    \103\ Specifically, the decision noted that `` `contribute'
means simply `to have a share in any act or effect,' Webster's Third
New International Dictionary 496 (1993), or `to have a part or share
in producing,' 3 Oxford English Dictionary 849 (2d ed. 1989).'' 370
F.3d at 13.
    \104\ The court explained, ``The repeated use of the term
`significant' to modify the contribution required for all nonroad
vehicles, coupled with the omission of this modifier from the
`cause, or contribute to' finding required for individual categories
of new nonroad vehicles, indicates that Congress did not intend to
require a finding of `significant contribution' for individual
vehicle categories.'' Id.
---------------------------------------------------------------------------

    In the past the Administrator has looked at emissions of air
pollutants in various ways to determine whether they ``cause or
contribute'' to the relevant air pollution. For instance, in some
mobile source rulemakings, the Administrator has looked at the percent
of emissions from the regulated mobile source category compared to the
total mobile source inventory for that air pollutant. See, e.g., 66 FR
5001 (2001) (heavy duty engine and diesel sulfur rule). In other
instances the Administrator has looked at the percent of emissions
compared to the total nonattainment area inventory of the air pollution
at issue. See, e.g., 67 FR 68242 (2002) (snowmobile rule). EPA has
found that air pollutant emissions that amount to 1.2% of the total
inventory ``contribute.'' Bluewater Network, 370 F.3d at 15 (``For
Fairbanks, this contribution was equivalent to 1.2% of the total daily
CO inventory for 2001.'').
    We solicit comment on these prior precedents, including their
relevance to contribution findings EPA may be considering regarding GHG
emissions. Where appropriate, may the Administrator determine that
emissions at a certain level or percentage contribute to air pollution
in one instance, while also finding that the same level or percentage
of another air pollutant and involving different air pollution, and
different overall circumstances, does not contribute? When exercising
his judgment, is it appropriate for the Administrator to consider not
only the cumulative impact, but also the totality of the circumstances
(e.g., the air pollutant, the air pollution, the type of source
category, the number of sources in the source category, the number and
type of other source categories that may emit the air pollutant) when
determining whether the emissions ``justify regulation'' under the CAA?
See Ethyl Corp., 541 F.2d at 31, n62 (``Moreover, even under a
cumulative impact theory emissions must make more than a minimal
contribution to total exposure in order to justify regulation under
Sec.  211(c)(1)(A).'').

B. Is the Air Pollution at Issue Reasonably Anticipated to Endanger
Public Health or Welfare?

    This section discusses options for defining, with respect to GHGs,
the ``air pollution'' that may or may not be reasonably anticipated to
endanger public health or welfare, the first part of the two part
endangerment test. It also summarizes the state of the science on GHGs
and climate change, and relates that science to the endangerment
question. We solicit comment generally on the information and issues
discussed below.
1. What is the Air Pollution?
    As noted above, in applying the endangerment test in section 202(a)
or other sections of the Act to GHG emissions, the Administrator must
define the scope and nature of the relevant ``air pollution'' that may
or may not be reasonably anticipated to endanger public health or
welfare. The endangerment issue discussed in today's notice involves,
primarily, anthropogenic emissions of GHGs, the accumulation of GHGs in
the atmosphere, the resultant impacts including climate change, and the
risks and impacts to human health and welfare associated with those
impacts.
a. The Six Major GHGs of Concern
    The six major GHGs of concern are CO₂, CH₄,
N₂O, HFCs, PFCs, and SF₆. The IPCC focuses on
these six GHGs for both scientific assessments and emissions inventory
purposes because these are the six long-lived, well-mixed GHGs not
controlled by the Montreal Protocol on Substances that Deplete the
Ozone Layer. These six GHGs are directly emitted by human activities,
are reported annually in EPA's Inventory of U.S. Greenhouse Gas
Emissions and Sinks, and are the common focus of the climate change
research community. The ICTA petition addresses the first four of these
GHGs, and the President's Executive Orders 13423 and 13432 define GHGs
to include all six of these GHGs.
    Carbon dioxide is the most important GHG directly emitted by human
activities, and is the most significant driver of climate change. The
anthropogenic combined heating effect (referred to as forcing) of
CH₄, N₂O, HFCs, PFCs and SF₆ is about
40% as large as the CO₂ cumulative heating effect since pre-
industrial times, according to the Fourth Assessment Report of the
IPCC.
b. Emissions and Elevated Concentrations of the Six GHGs
    As mentioned previously, these six GHGs can remain in the
atmosphere for decades to centuries. Therefore, these GHGs, once
emitted, become well mixed throughout the global atmosphere regardless
of their emission origin, such that their average concentrations over
the U.S. are roughly the same as the global average. This also means
that current GHG concentrations are the cumulative result of both
historic and current emissions, and that future concentrations will be
the cumulative result of historic, current and future emissions.
    Greenhouse gases trap some of the Earth's heat that would otherwise
escape to space. The additional heating effect caused by the buildup of
anthropogenic GHGs in the atmosphere enhances the Earth's natural
greenhouse effect and causes global temperatures to increase, with
associated climatic changes (e.g., change in precipitation patterns,
rise in sea levels, and changes in the frequency and intensity of
extreme weather events). Current atmospheric concentrations of all of
these GHGs are significantly higher than pre-industrial (~1750) levels
as a result of human activities. Atmospheric concentrations of
CO₂ and other GHGs

[[Page 44424]]

are projected to continue to climb over the next several decades.
    The scientific literature that assesses the potential risks and
end-point impacts of climate change (driven by the accumulation of
atmospheric concentrations of GHGs) does not assess these impacts on a
gas-by-gas basis. Observed climate change and associated effects are
driven by the buildup of all GHGs in the atmosphere, as well as other
natural and anthropogenic factors that influence the Earth's energy
balance. Likewise, the future projections of climate change that have
been done are driven by emission scenarios of all six GHGs, as well as
other pollutants, many of which are already regulated in the U.S. and
other countries.
    For these reasons, EPA is considering defining the ``air
pollution'' related to GHGs as the elevated combined current and
projected atmospheric concentration of the six GHGs. This approach is
consistent with other provisions of the CAA and previous EPA practice
under the CAA, where separate air pollutants from different sources but
with common properties may be treated as a class (e.g., Class I and
Class II substances under Title VI of the CAA). It also addresses the
cumulative effect that the elevated concentrations of the six GHGs have
on climate, and thus on different elements of health, society and the
environment. We seek comment on this potential approach, as well as
other alternative ways to define ``air pollution.'' One alternative
would be to define air pollution as the elevated concentration of an
individual GHG; however, in this case the Administrator may still have
to consider the impact of the individual GHG in combination with the
impacts caused by the elevated concentrations of the other GHGs.
c. Other Anthropogenic Factors That Have a Climatic Warming Effect
Beyond the Six Major GHGs
    There are other GHGs and aerosols that have climatic warming
effects: water vapor, chlorofluorocarbons (CFCs),
hydrochlorofluorocarbons (HCFCs), halons, stratospheric and
tropospheric ozone (O₃), and black carbon. Each of these is
discussed here. We seek comment on whether and how they should be
considered in the definition of ``air pollution'' for purposes of an
endangerment finding.
    Water vapor is the most abundant naturally occurring GHG and
therefore makes up a significant share of the natural, background
greenhouse effect. However, water vapor emissions from human activities
have only a negligible effect on atmospheric concentrations of water
vapor. Significant changes to global atmospheric concentrations of
water vapor occur indirectly through human-induced global warming,
which then increases the amount of water vapor in the atmosphere
because a warmer atmosphere can hold more moisture. Therefore, changes
in water vapor concentrations are not an initial driver of climate
change, but rather an effect of climate change which then acts as a
positive feedback that further enhances warming. For this reason, the
IPCC does not list direct emissions of water vapor as an anthropogenic
forcing agent of climate change, but does include this water vapor
feedback mechanism in response to human-induced warming in all modeling
scenarios of future climate change. Based on this recognition that
anthropogenic emissions of water vapor are not a significant driver of
anthropogenic climate change, EPA's annual Inventory of U.S. Greenhouse
Gas Emissions and Sinks does not include water vapor, and GHG inventory
reporting guidelines under the United Nations Framework Convention on
Climate Change (UNFCCC) do not require data on water vapor emissions.
    Water vapor emissions may be an issue for concern when they are
emitted by aircraft at high altitudes, where, under certain conditions,
they can lead to the formation of condensation trails, referred to as
contrails. Similar to high-altitude, thin clouds, contrails have a
warming effect. Extensive cirrus clouds can also develop from aviation
contrails, and increases in cirrus cloud cover would also have a
warming effect. The IPCC Fourth Assessment Report estimated a very
small positive radiative forcing effect for linear contrails, with a
low degree of scientific understanding. Unlike the warming effects
associated with the six long-lived, well-mixed GHGs, the warming
effects associated with contrails or contrail-induced cirrus cloud
cover are more regional and temporal in nature. Further discussion of
aviation contrails can be found in Section VI on mobile sources. EPA
invites input and comment on the scientific and policy issues related
to consideration of water vapor's association with aviation contrails
in an endangerment analysis.
    The CFCs, HCFCs, and halons are all strong anthropogenic GHGs that
are long-lived in the atmosphere and are adding to the global
anthropogenic heating effect. Therefore, these gases share common
climatic properties with the six GHGs discussed above. The production
and consumption of these substances (and hence their anthropogenic
emissions) are being controlled and phased out, not because of their
effects on climate change, but because they deplete stratospheric
O₃, which protects against harmful ultraviolet B (UVB)
radiation. The control and phase-out of these substances in the U.S.
and globally is occurring under the Montreal Protocol on Substances
that Deplete the Ozone Layer, and in the U.S. under Title VI of the CAA
as well.\105\ Therefore, the climate change research and policy
community typically does not focus on these substances, precisely
because they are essentially already being 'taken care of' with non-
climate policy mechanisms. For example, the UNFCCC does not address
these substances, and instead defers their treatment to the Montreal
Protocol. As mentioned above, the President's Executive Orders 13423
and 13432 do not include these substances in the definition of GHGs.
For these reasons, EPA's preliminary conclusion is that we would not
include CFCs, HCFCs and halons in the definition of ``air pollution''
for purposes of an endangerment finding. We seek comment on this issue.
---------------------------------------------------------------------------

    \105\ Under the Montreal Protocol, production and consumption of
CFCs were phased out in developed countries in 1996 (with some
essential use exemptions) and are scheduled for phase-out by 2010 in
developing countries (with some essential use exemptions). For
halons the schedule was 1994 for phase out in developed countries
and 2010 for developing countries; HCFC production was frozen in
2004 in developed countries, and in 2016 production will be frozen
in developing countries; and HCFC consumption phase-out dates are
2030 for developed countries and 2040 in developing countries.
---------------------------------------------------------------------------

    The depletion of stratospheric O₃ due to CFCs, HCFCs,
and other ozone-depleting substances has resulted in a small cooling
effect on the planet.
    Increased concentrations of tropospheric O₃ are causing
a significant anthropogenic warming effect, but, unlike the long-lived
six GHGs, tropospheric O₃ has a short atmospheric lifetime
(hours to weeks), and therefore its concentrations are more variable
over space and time. For these reasons, its global heating effect and
relevance to climate change tends to entail greater uncertainty
compared to the well-mixed, long-lived GHGs. More importantly,
tropospheric ozone is already listed as a NAAQS pollutant and is
regulated through SIPs and other measures under the CAA, due to its
direct health effects including increases in respiratory infection,
medicine use by asthmatics, emergency department visits and hospital
admissions, and its potential to contribute to premature death,
especially in susceptible populations such as asthmatics,

[[Page 44425]]

children and the elderly. Tropospheric O₃ is not addressed
under the UNFCCC. For these reasons, EPA's preliminary conclusion is
that we would not include tropospheric O₃ in the definition
of ``air pollution'' for purposes of an endangerment finding because,
as with CFCs, HCFCs and halons, it is already being addressed by
regulatory actions that control precursor emissions (NO_X and
volatile organic compounds (VOCs)) from major U.S. sources. We invite
comment on this issue.
    Black carbon is an aerosol particle that results from incomplete
combustion of the carbon contained in fossil fuels, and it remains in
the atmosphere for about a week. Black carbon causes a warming effect
by absorbing incoming sunlight in the atmosphere (whereas GHGs cause
warming by trapping outgoing, infrared heat), and by darkening bright
surfaces such as snow and ice, which reduces reflectivity and increases
absorption of sunlight at the surface. Some recent research,\106\
published after the IPCC Fourth Assessment Report, has suggested that
black carbon may play a larger role in warming than previously thought.
Like other aerosols, black carbon can also alter the reflectivity and
lifetime of clouds, which in turn can have an additional climate
effect. How black carbon and other aerosols alter cloud properties is a
key source of uncertainty in climate change science. Given these
reasons, there is considerably more uncertainty associated with black
carbon's warming effect compared to the estimated warming effect of the
six long-lived GHGs.
---------------------------------------------------------------------------

    \106\ Ramathan, V, and G. Carmichael (2008) Global and regional
climate changes due to black carbon. Nature Geoscience, 1: 221-227.
---------------------------------------------------------------------------

    Black carbon is also co-emitted with organic carbon, which tends to
have a cooling effect on climate because it reflects and scatters
incoming sunlight. The ratio of black carbon to organic carbon varies
by fuel type and by combustion efficiency. Diesel vehicles, for
example, emit a much greater portion of black carbon, whereas forest
fires tend to emit much more organic carbon. The net effect of black
carbon and organic carbon on climate should therefore be considered.
Also, black carbon is a subcomponent of particulate matter (PM), which
is regulated as a NAAQS pollutant under the CAA due to its direct
health effects caused by inhalation. Diesel vehicles are estimated to
be the largest source of black carbon in the U.S., but these emissions
are expected to decline substantially over the coming decades due to
recently promulgated EPA regulations targeting PM_2.5
emissions from on-road and off-road diesel vehicles (the Highway Diesel
Rule and the Clean Air Nonroad Diesel Rule, the Locomotive and Marine
Compression Ignition Rule). Non-regulatory partnership programs such as
the National Clean Diesel Campaign and Smartway are reducing black
carbon as well. In sum, black carbon has different climate properties
compared to long-lived GHGs, and major U.S. sources of black carbon are
already being aggressively reduced through regulatory actions due to
health concerns. Nevertheless, EPA has recently received petitions
asking the Agency to reduce black carbon emissions from some mobile
source categories (see Section VI.). Therefore, EPA seeks comment on
how to treat black carbon (and co-emitted organic carbon) regarding the
definition of ``air pollution'' in the endangerment context.
2. Science Summary
    The following provides a summary of the underlying science that was
reviewed and utilized in the Endangerment Technical Support Document
for the endangerment discussion, which in turn relied heavily on the
IPCC Fourth Assessment Report. We seek comment on the best available
science for purposes of the endangerment discussion, and in particular
on the use of the more recent findings of the U.S. Climate Change
Science Program.
a. Observed Global Effects
    The global atmospheric CO2 concentration has increased about 35%
from pre-industrial levels to 2005, and almost all of the increase is
due to anthropogenic emissions. The global atmospheric concentration of
CH4 has increased by 148% since pre-industrial levels. Current
atmospheric concentrations of CO2 and CH4 far exceed the recorded
natural range of the last 650,000 years. The N2O concentration has
increased 18%. The observed concentration increase in these non-CO2
gases can also be attributed primarily to anthropogenic emissions. The
industrial fluorinated gases, HFCs, PFCs, and SF6, have relatively low
atmospheric concentrations but are increasing rapidly; these gases are
entirely anthropogenic in origin.
    Current ambient concentrations of CO2 and other GHGs remain well
below published thresholds for any direct adverse health effects, such
as respiratory or toxic effects.
    The global average net effect of the increase in atmospheric GHG
concentrations, plus other human activities (e.g., land use change and
aerosol emissions), on the global energy balance since 1750 has been
one of warming. This total net radiative forcing (a measure of the
heating effect caused by changing the Earth's energy balance) is
estimated to be +1.6 Watts per square meter (W/m\2\). The combined
radiative forcing due to the cumulative (i.e., 1750 to 2005) increase
in atmospheric concentrations of CO2, CH4, and N2O is +2.30 W/m\2\. The
rate of increase in positive radiative forcing due to these three GHGs
during the industrial era is very likely to have been unprecedented in
more than 10,000 years. The positive radiative forcing due to the
increase in CO2 concentrations is the largest (+1.66 W/m\2\). The
increase in CH4 concentrations is the second largest source of positive
radiative forcing (+0.48 W/m2). The increase in N2O has a positive
radiative forcing of +0.16 W/m\2\.
    Warming of the climate system is unequivocal, as is now evident
from observations of increases in global average air and ocean
temperatures, widespread melting of snow and ice, and rising global
average sea level. Global mean surface temperatures have risen by
0.74[deg]C (1.3[deg]F) over the last 100 years. The average rate of
warming over the last 50 years is almost double that over the last 100
years. Global mean surface temperature was higher during the last few
decades of the 20th century than during any comparable period during
the preceding four centuries.
    Most of the observed increase in global average temperatures since
the mid-20th century is very likely due to the observed increase in
anthropogenic GHG concentrations. Global observed temperatures over the
last century can be reproduced only when model simulations include both
natural and anthropogenic forcings, i.e., simulations that remove
anthropogenic forcings are unable to reproduce observed temperature
changes. Thus, the warming cannot be explained by natural variability
alone.
    Observational evidence from all continents and most oceans shows
that many natural systems are being affected by regional climate
changes, particularly temperature increases. Observations show that
changes are occurring in the amount, intensity, frequency and type of
precipitation. There is strong evidence that global sea level gradually
rose in the 20th century and is currently rising at an increased rate.
Widespread changes in extreme temperatures have been observed in the
last 50 years. Globally, cold days, cold nights, and frost have become
less frequent, while hot days, hot nights, and heat waves have become
more frequent.

[[Page 44426]]

    The Endangerment Technical Support Document provides evidence that
the U.S. and the rest of the world are experiencing effects from
climate change now.
b. Observed U.S. Effects
    U.S. temperatures also warmed during the 20th and into the 21st
century. U.S. temperatures are now approximately 1.0 [deg]F warmer than
at the start of the 20th century, with an increased rate of warming
over the past 30 years. The past nine years have all been among the 25
warmest years on record for the contiguous U.S., a streak which is
unprecedented in the historical record. Like the average global
temperature increase, the observed temperature increase for North
America has been attributed to the global buildup of anthropogenic GHG
concentrations in the atmosphere.
    Widespread changes in extreme temperatures have been observed in
the last 50 years across all world regions including the U.S. Cold
days, cold nights, and frost have become less frequent, while hot days,
hot nights, and heat waves have become more frequent.
    Total annual precipitation has increased over the U.S. on average
over the last century (about 6%), and there is evidence of an increase
in heavy precipitation events. Nearly all of the Atlantic Ocean shows
sea level rise during the past decade with highest rate in areas that
include the U.S. east coast.
    Observations show that climate change is currently impacting the
nation's ecosystems and services in significant ways.
c. Projected Effects
    The Endangerment Technical Support Document, the IPCC Fourth
Assessment Report, and a report under the U.S. Climate Change Science
Program, provide projections of future ambient concentrations of GHGs,
future climate change, and future anticipated effects from climate
change under various scenarios. This section summarizes some of the key
global projections, such as changes in global temperature, as well as
those particular to North America and the United States.
    Overall risk to human health, society and the environment increases
with increases in both the rate and magnitude of climate change.
Climate warming may increase the possibility of large, abrupt, and
worrisome regional or global climatic events (e.g., disintegration of
the Greenland Ice Sheet or collapse of the West Antarctic Ice Sheet).
The majority of the climate change impacts literature assesses the
potential effects on health, society and the environment due to
projected changes in average conditions (e.g., temperature increase,
precipitation change, sea level rise) and do not take into account how
the frequency and severity of extreme events due to climate change may
cause certain additional impacts. Likewise, impact studies typically do
not account for large, abrupt climatic events, and generally consider
rates of warming that would result from climate sensitivities \107\
within the most likely range, not at the tails of the distribution. To
weigh the full range of risks and impacts, it is important to consider
these possible extreme outcomes, including those that are of low
probability.
---------------------------------------------------------------------------

    \107\ ``Climate sensitivity'' is a term used to describe how
much long-term global warming occurs if global atmospheric
concentrations of CO₂ are doubled compared to their pre-
industrial levels. The IPCC Fourth Assessment Report states that
climate sensitivity is very likely greater than 1.5[deg]C (2.7
[deg]F) and likely to lie in the range of 2 [deg]C to 4.5 [deg]C
(3.6 [deg]F to 8.1 [deg]F), with a most likely value of about 3
[deg]C (5.4 [deg]F), and that a climate sensitivity higher than 4.5
[deg]C cannot be ruled out.
---------------------------------------------------------------------------

i. Global Effects
    The majority of future reference-case scenarios (assuming no
explicit GHG mitigation actions beyond those already enacted) project
an increase of global GHG emissions over the century, with climbing GHG
concentrations and associated increases in radiative forcing and
average global temperatures.
    Projected ambient concentrations of CO2 and other GHGs remain well
below published thresholds for any direct adverse health effects, such
as respiration or toxic effects.
    Through about 2030, the global warming rate is affected little by
different future scenario assumptions or different model sensitivities,
because there is already some degree of commitment to future warming
given past and present GHG emissions. By mid-century, the choice of
scenario becomes more important for the magnitude of the projected
warming because only about a third of that warming is projected to be
due to climate change that is already committed. By the end of the
century, projected average global warming (compared to average
temperature around 1990) varies significantly by emissions scenario,
with IPCC's best estimates ranging from 1.8 to 4.0 [deg]C (3.2 to 7.2
[deg]F), with a fuller likely range of 1.1 to 6.4 [deg]C (2.0 to 11.5
[deg]F), which takes into account a wider range of future emission
scenarios and a wider range of uncertainties.\108\
---------------------------------------------------------------------------

    \108\ The IPCC scenarios are also described in the Technical
Support Document and include a range of future global emission
scenarios and a range of climate sensitivities (which measure how
much global warming occurs for a given increase in global
CO₂ concentrations).
---------------------------------------------------------------------------

    The IPCC identifies the most vulnerable world regions as the
Arctic, because of high rates of projected warming on natural systems;
Africa, especially the sub-Saharan region, because of current low
adaptive capacity; small islands, due to high exposure of population
and infrastructure to risk of sea-level rise and increased storm surge;
and Asian mega deltas, due to large populations and high exposure to
sea level rise, storm surge, and river flooding. Climate change impacts
in certain regions of the world may exacerbate problems that raise
humanitarian and national security issues for the U.S. Climate change
has been described as a potential threat multiplier regarding national
security issues.
ii. United States Effects
    Projected global warming is anticipated to lead to effects in the
U.S. For instance, all of the U.S. is very likely to warm during this
century, and most areas of the U.S. are expected to warm by more than
the global average. The U.S, along with the rest of the world, is
projected to see an increase in the intensity of precipitation events
and the risk of flooding, greater runoff and erosion, and thus the
potential for adverse water quality effects.
    Severe heat waves are projected to intensify in magnitude,
frequency, and duration over the portions of the U.S. where these
events already occur, with likely increases in mortality and morbidity,
especially among the elderly, young, and frail. Warmer temperatures can
also lead to fewer cold-related deaths. It is currently not possible to
quantify the balance between decreased cold-related deaths and
increased heat-related deaths attributable to climate change over time.
    The IPCC projects with virtual certainty (i.e., greater than 99%
likelihood) declining air quality in cities due to warmer days and
nights, and fewer cold days and nights, and/or more frequent hot days
and nights over most land areas, including the U.S. Climate change is
expected to lead to increases in regional ozone pollution, with
associated risks for respiratory infection, aggravation of asthma, and
potential premature death, especially for people in susceptible groups.
Climate change effects on ambient PM are currently less certain.
    Additional human health concerns include a change in the range of
vector-

[[Page 44427]]

borne diseases, and a likely trend towards more intense hurricanes
(even though any single hurricane event cannot be attributed to climate
change) and other extreme weather events. For many of these issues,
sensitive populations, such as the elderly, young, asthmatics, the
frail and the poor, are most vulnerable.
    Moderate climate change in the early decades of the century is
projected to increase aggregate yields of rainfed agriculture in the
United States by 5-20%. However, as temperatures continue to rise,
grain and oilseed crops will increasingly experience failure,
especially if climate variability increases and precipitation lessens
or becomes more variable. How climatic variability and extreme weather
events will continue to change under a changing climate is a key
uncertainty, and these events also have the potential to offset the
benefits of CO2 fertilization and a longer growing season.
    Climate change is projected to constrain over-allocated water
resources in the U.S., increasing competition among agricultural,
municipal, industrial, and ecological uses. Rising temperatures will
diminish snowpack and increase evaporation, affecting seasonal
availability of water.
    Disturbances like wildfire and insect outbreaks are increasing and
are likely to intensify in a warmer future with drier soils and longer
growing seasons. Overall forest growth in the U.S. will likely increase
by 10-20% as a result of extended growing seasons and elevated
CO₂ over the next century, but with important spatial and
temporal variation. Although recent climate trends have increased
vegetation growth in parts of the United States, continuing increases
in disturbances are likely to limit carbon storage, facilitate invasive
species, and disrupt ecosystem services.
    The U.S. will be affected by global sea level rise, which is
expected to increase between 0.18 and 0.59 meters by the end of the
century relative to around 1990. These numbers represent the lowest and
highest projections of the 5 to 95% ranges for all scenarios considered
collectively and include neither uncertainty in carbon cycle feedbacks
nor rapid dynamical changes in ice sheet flow. U.S. coastal communities
and habitats will be increasingly stressed by climate change
interacting with development and pollution. Sea level is already rising
along much of the coast, and the rate of change is expected to increase
in the future, exacerbating the impacts of progressive inundation,
storm-surge flooding, and shoreline erosion.
    Climate change is likely to affect U.S. energy use (e.g., heating
and cooling requirements), and energy production (e.g., effects on
hydropower), physical infrastructures (including coastal roads,
railways, transit systems and runways) and institutional
infrastructures. Climate change will likely interact with and possibly
exacerbate ongoing environmental change and environmental pressures in
some settlements, particularly in Alaska where indigenous communities
are facing major environmental and cultural impacts.
3. Endangerment Discussion Regarding Air Pollution
    The Administrator must exercise his judgment in evaluating whether
the first part of the endangerment test is met, i.e., whether air
pollution (e.g., the elevated concentrations of GHGs) is reasonably
anticipated to endanger public health or welfare. As discussed above,
in exercising his judgment it is appropriate for the Administrator to
make comparative assessments of risk and projections of future
possibilities, consider uncertainties, and extrapolate from limited
data. The precautionary nature of the statutory language also means
that the Administrator should act to prevent harm rather than wait for
proof of actual harm.
    The scientific record shows there is compelling and robust evidence
that observed climate change can be attributed to the heating effect
caused by global anthropogenic GHG emissions. The evidence goes beyond
increases in global average temperature to include observed changes in
precipitation patterns, sea level rise, extreme hot and cold days, sea
ice, glaciers, ecosystem functioning and wildlife patterns. Global
warming trends over the last 50 years stand out as significant compared
to estimated global average temperatures for at least the last few
centuries. Some degree of future warming is now unavoidable given the
current buildup of atmospheric concentrations of GHGs, as the result of
past and present GHG emissions. Based on the scientific evidence, it is
reasonable to conclude that future climate change will result from
current and future emissions of GHGs. Future warming over the course of
the 21st century, even under scenarios of low emissions growth, is very
likely to be greater than observed warming over the past century.
    The range of potential impacts that can result from climate change
spans many elements of the global environment, and all regions of the
U.S. will be affected in some way. The U.S. has a long and populous
coastline. Sea level rise will continue and exacerbate storm-surge
flooding and shoreline erosion. In areas where heat waves already
occur, they are expected to become more intense, more frequent, and
longer lasting. Wildfires and the wildfire season are already
increasing and climate change is expected to continue to worsen
conditions that facilitate wildfires. Where water resources are already
scarce and over-allocated in the western U.S., climate change is
expected to put additional strain on these water management issues for
municipal, agricultural, energy and industrial uses. Climate change
also introduces an additional stress on ecosystems which are already
affected by development, habitat fragmentation, and broken ecological
dynamics. There is a wide range in the magnitude of these estimated
impacts, with there being more confidence in the occurrence of some
effects and less confidence in the occurrence of others.
    In addition to the effects from changes in climate, there are some
additional welfare effects that occur directly from the anthropogenic
GHG emissions themselves. For example, ocean acidification occurs
through elevated concentrations of CO₂, and crop and other
vegetation growth can be enhanced through elevated CO₂
concentrations as well.
    Current and projected levels of ambient concentrations of the six
GHGs are not expected to cause any direct adverse health effects, such
as respiratory or toxic effects, which would occur as a result of the
elevated GHG concentrations themselves rather than through the effects
of climate change. However, there are indirect human health risks
(e.g., heat-related mortality, exacerbated air quality, extreme events)
and benefits (e.g., less cold-related mortality) that occur due to
climate change. We seek comment on how these human health impacts
should be characterized under the CAA for purposes of an endangerment
analysis.
    Some elements of human health, society and the environment may
benefit from climate change (e.g., short-term increases in agricultural
yields, less cold-related mortality). We seek comment on how the
potential for some benefits should be viewed against the full weight of
evidence showing numerous risks and the potential for adverse impacts.
    Quantifying the exact nature and timing of impacts due to climate
change over the next few decades and beyond, and across all vulnerable
elements of U.S. health, society and the environment, is currently not
possible. However, the full weight of evidence as

[[Page 44428]]

summarized above and as documented in the Endangerment Technical
Support Document points towards the robust conclusion that expected
rates of climate change (driven by past, present and plausible future
GHG emissions) pose a number of serious risks to the U.S., even if the
exact nature of the risks is difficult to quantify with confidence. The
uncertainties in this context can also mean that future rates of
climate change are being underestimated, and that the potential for
associated and difficult-to-predict-and-quantify extreme events is not
adequately incorporated into impact assessments. The scientific
literature states that risk increases with increases in both the rate
and magnitude of climate change. We solicit comment on how these
uncertainties should be considered.
    We seek comment on whether, in light of the precautionary nature of
the statutory language, the Administrator needs to find that current
levels of GHG concentrations endanger public health or welfare now. As
noted above, the fact that GHGs remain in the atmosphere for decades to
centuries means that future concentrations are dependent not only on
tomorrow's emissions, but also on today's emissions. Should the
Administrator consider both current and projected future elevated
concentrations of GHGs, as well as the totality of the observed and
projected effects that result from current and projected
concentrations? Or should the Administrator focus on future projected
elevated concentrations of GHGs and their projected effects in the
United States because they are larger and of greater concern than
current GHG concentrations and observed effects?
    In sum, EPA invites comment on all issues relevant to making an
endangerment finding, including the scientific basis supporting a
finding that there is or is not endangerment under the CAA, as well as
the potential scope of the finding (i.e., public health, welfare, or
both).

C. Illustration for the ``Cause or Contribute'' Part of the
Endangerment Discussion: Do emissions of air pollutants from motor
vehicles or fuels cause or contribute to the air pollution that may
reasonably be anticipated to endanger public health or welfare in the
United States?

 1. What Is/Are the Air pollutant(s)?
a. Background and Context
    If the Administrator, in his judgment, finds that GHG ``air
pollution'' may reasonably be anticipated to endanger public health or
welfare, he must then define ``air pollutant(s)'' for purposes of
making the ``cause or contribute'' determination. The question is
whether the ``air pollutants'' to be evaluated for ``cause or
contribute'' should be the individual GHGs, or whether the ``air
pollutant'' is one or more classes of GHGs as a group.
    We recognize that the alternative definitions could have important
implications for how GHGs are treated under other provisions of the
Act. The Administrator seeks comment on these options, and is
particularly interested in views regarding the implications for the
potential future regulation of GHGs under other parts of the Act.
b. Defining ``Air Pollutant'' as Each Individual Greenhouse Gas
    Under this approach, the Administrator could define ``air
pollutant'' as each individual GHG rather than as GHGs as a collective
whole for the purposes of assessing ``cause or contribute.'' The
Administrator would evaluate each individual GHG to determine if it
causes, or contributes to, the elevated combined level of GHG
concentrations.
    This approach enables an evaluation of the unique characteristics
and properties of each GHG (e.g., radiative forcing, lifetimes, etc.),
as well as current and projected emissions. This facilitates a
customized approach accounting for these factors. This approach also is
consistent with the approach taken in several federal GHG programs
which target reductions of individual greenhouse gases. For example,
EPA manages a variety of partnership programs aimed at reducing
emissions of specific sources of methane and the fluorinated gases
(HFCs, PFCs and SF₆).
c. Defining ``Air Pollutants'' Collectively as a Class of Greenhouse
Gases
    Under this approach, the Administrator could define the ``air
pollutant'' as (a) the collective group of the six GHGs discussed above
(CO₂, CH₄, N₂O, HFCs, PFCs, and
SF₆), (b) the collective group of the specific GHGs that are
emitted from the relevant source category at issue in the endangerment
finding (e.g., for section 202 sources it would be CO₂,
CH₄, N₂O, and HFCs), or (c) other reasonable
groupings.
    There are several federal and state climate programs, such as EPA's
Climate Leaders program, DOE's 1605b program, and Multi-state Climate
Registry, that encourage firms to report (and reduce) emissions of all
six GHGs, recognizing that the non-CO₂ GHG emissions are a
significant part of the atmospheric buildup of GHG concentrations and
thus radiative forcing. In addition, the President's recent 2007
Executive Orders (13423 and 13432) and his 2002-2012 intensity goal
both encompass the collective emissions of all six GHGs. Consideration
of a class of gases collectively takes into account the multiple
effects of mitigation options and technologies on each gas, thus
enabling a more coordinated approach in addressing emissions from a
source. For example, collection and combustion of fugitive methane will
lead to net increases in CO₂ and possibly nitrous oxide
emissions, but this is nevertheless desirable from an overall
mitigation perspective given the lower total radiative forcing.
2. Discussion of ``Cause or Contribute''
    Once the ``air pollutant(s)'' is defined, the Administrator must
look at the emissions of the air pollutant from the relevant source
category in determining whether those emissions cause or contribute to
the air pollution he has determined may reasonably be anticipated to
endanger public health or welfare. There arguably are many possible
ways of assessing ``cause and contribute'' and different approaches
have been used in previous endangerment determinations under the CAA.
For example, EPA could consider how emissions from the relevant source
category would compare as a share of the following:
     Total global aggregated emissions of the 6 GHGs discussed
in the definition of ``air pollution'';
     Total aggregated U.S. emissions of the 6 GHGs;
     Total global emissions of the individual GHG in question;
     Total U.S. emissions of the individual GHG in question;
and
     Total global atmospheric concentrations of the GHG in
question.
    In the past, the smallest level or amount of emissions that the
Administrator determined ``contributed'' to the air pollution at issue
was just less than 1% (67 FR 68242 (2002)). We solicit comment on other
factors that may be relevant to a contribution determination for GHG
emissions. For example, given the global nature of the air pollution
being addressed in this rulemaking, one might expect that the
percentage contribution of specific GHGs and sectors would be much
smaller than for previous rulemakings when the nature of the air
pollution at issue was regional or local. On an absolute basis, a small
U.S. GHG source on a global scale may have emissions at the same level
as one of the largest sources in a single small to medium size country,
and given the

[[Page 44429]]

large size of the global denominator, even sectors with significant
emissions could be very small in percentage terms.
    In addition, EPA notes that the EPA promotes the reduction of
particular GHG emissions through a variety of voluntary programs (e.g.,
EPA's domestic CH4 partnership programs and the international Methane
to Markets Partnership (launched in 2004)). EPA requests comment on how
these and other efforts to encourage the voluntary reductions in even
small amounts of GHG emissions are relevant to decisions about what
level of ``contribution'' merits mandatory regulations.
    Below we use the section 202 source category to illustrate these
and other various ways to consider and compare source category GHG
emissions for the ``cause or contribute'' analysis. In keeping with the
discussion above regarding possible definitions of ``air pollutant,''
we provide the information on an individual GHG and collective GHG
basis. In addition, we raise various policy considerations that could
be relevant to a ``cause or contribute'' determination. EPA invites
comment on the various approaches, data, and policy considerations
discussed below.
a. Overview of Section 202 Source Categories
    The relevant mobile sources under section 202(a)(1) of the Clean
Air Act are ``any class or classes of new motor vehicles or new motor
vehicle engines, * * * '' CAA section 202(a)(1). To support this
illustrative assessment, EPA analyzed historical GHG emissions data for
motor vehicles and motor vehicle engines in the United States from 1990
to 2006.\109\
---------------------------------------------------------------------------

    \109\ The source of the emissions data is the Inventory of U.S.
Greenhouse Gas Emissions and Sinks: 1990-2006 (USEPA #430-R-08-005)
(hereinafter ``U.S. Inventory''). See the Emissions Technical
Support Document for a discussion on the correspondence between
Section 202 source categories and IPCC source categories. The most
recent year for which official EPA estimates are available is 2006.
---------------------------------------------------------------------------

    The motor vehicles and motor vehicle engines (hereinafter ``section
202 source categories'') addressed include passenger cars, light-duty
trucks, motorcycles, buses, medium/heavy-duty trucks, and cooling.\110\
Of the six primary GHGs, four are associated with section 202 source
categories: CO₂, CH₄, N₂O, and HFCs.
---------------------------------------------------------------------------

    \110\ Greenhouse gas emissions result from the use of HFCs in
cooling systems designed for passenger comfort, as well as auxiliary
systems for refrigeration.
---------------------------------------------------------------------------

    A summary of the section 202 emissions information is presented
here, and a more detailed description along with data tables is
contained in the Emissions Technical Support Document. All annual
emissions data are considered on a CO₂ equivalent basis.
b. Carbon Dioxide Emissions From Section 202 Sources
    CO₂ is emitted from motor vehicles and motor vehicle
engines during the fossil fuel combustion process. During combustion,
the carbon stored in the fuels is oxidized and emitted as
CO₂ and smaller amounts of other carbon compounds.\111\
---------------------------------------------------------------------------

    \111\ Detailed CO₂ emissions data from section 202
source categories are presented in the Emissions Technical Support
Document. Other carbon compounds emitted such as CO, and non-methane
volatile organic compounds oxidize in the atmosphere to form
CO₂ in a period of hours to days.
---------------------------------------------------------------------------

    CO₂ is the dominant GHG emitted from motor vehicles and
motor vehicle engines, and the dominant GHG emitted in the U.S. and
globally.\112\ CO₂ emissions from section 202 sources grew
by 32% between 1990 and 2006, largely due to increased CO₂
emissions from light-duty trucks (61% since 1990) and medium/heavy-duty
trucks (76%). Emissions of CO₂ from section 202 sources, and
U.S. and global emissions are presented below in Table V-1.
---------------------------------------------------------------------------

    \112\ EPA typically uses current motor vehicle fleet emissions
information when making a contribution analysis under section 202.
We solicit comment on how or whether the reductions in
CO₂ emissions expected by implementation of EISA, or any
other projected change in emissions from factors such as growth in
the fleet or vehicle miles traveled, would impact a contribution
analysis for CO₂.

          Table V-1--Section 202 CO2, U.S. and Global Emissions
------------------------------------------------------------------------
                                                            Sec 202 CO2
             U.S. Emissions                    2006            share
                                                             (percent)
------------------------------------------------------------------------
Section 202 CO2.........................         1,564.6
All U.S. CO2............................          5983.1            26.2
U.S. emissions of Sec 202 GHG...........         1,665.4            93.9
All U.S. GHG emissions..................         7,054.2           22.2%
------------------------------------------------------------------------



                                                            Sec 202 CO2
                                                             share (in
            Global Emissions                   2000            2000)
                                                             (percent)
------------------------------------------------------------------------
All global CO2 emissions................        30,689.5             4.8
Global transport GHG emissions..........         5,315.2            27.5
All global GHG emissions................        36,727.9             4.0
------------------------------------------------------------------------



                                                           Share of U.S.
        Other Sources of U.S. CO2              2006        CO2 emissions
                                                             (percent)
------------------------------------------------------------------------
Electricity Sector CO2..................          2360.3            39.4
Industrial Sector CO2...................           984.1            16.4
------------------------------------------------------------------------

    Arguably, based on these data, if the Administrator did not find
that, for purposes of section 202, that CO₂ emissions from
section 202 source categories contribute to the elevated combined level
of GHG concentrations, it is unlikely that he would find that the other
GHGs emitted by section 202 source categories contribute.
c. Methane Emissions From Section 202 Source Categories
    Methane (CH4) emissions from motor vehicles are a function of the
CH4 content of the motor fuel, the amount of

[[Page 44430]]

hydrocarbons passing uncombusted through the engine, and any post-
combustion control of hydrocarbon emissions (such as catalytic
converters). Methane emissions from these source categories decreased
by 58% between 1990 and 2006, largely due to decreased CH4 emissions
from passenger cars and light-duty trucks.\113\ Emissions of
CH₄ from section 202 sources, and U.S. and global emissions
are presented below in Table V-2.
---------------------------------------------------------------------------

    \113\ Detailed methane emissions data for section 202 source
categories are presented in the Emissions Technical Support
Document.

          Table V-2--Section 202 CH4, U.S. and Global Emissions
------------------------------------------------------------------------
                                                            Sec 202 CH4
             U.S. Emissions                    2006            share
                                                             (percent)
------------------------------------------------------------------------
Section 202 CH4.........................            1.80
All U.S. CH4............................           555.3            0.32
U.S. emissions of Sec 202 GHG...........        1,665.40            0.11
All U.S. GHG emissions..................        7,054.20            0.03
------------------------------------------------------------------------



                                                            Sec 202 CH4
                                                             share (in
            Global Emissions                   2000            2000)
                                                             (percent)
------------------------------------------------------------------------
All global CH4 emissions................        5,854.90            0.05
Global transport GHG emissions..........        5,315.20            0.05
All global GHG emissions................       36,727.90            0.01
------------------------------------------------------------------------



                                                           Share of U.S.
        Other Sources of U.S. CH4              2006        CH4 emissions
                                                             (percent)
------------------------------------------------------------------------
Landfill CH4 emissions..................           125.7            22.6
Natural Gas CH4 emissions...............           102.4            18.4
------------------------------------------------------------------------

    EPA also notes that the EPA promotes the reduction of
CH₄ and other non-CO₂ GHG emissions, as
manifested in its domestic CH₄ partnership programs and the
international Methane to Markets Partnership (launched in 2004), which
are not focused on the transportation sector. EPA requests comment on
how these and other efforts to encourage the voluntary reductions in
even small amounts of GHG emissions are relevant to decisions about
what level of ``contribution'' merits mandatory regulations.
d. Nitrous Oxide Emissions From Section 202 Source Categories
    Nitrous oxide (N₂O) is a product of the reaction that
occurs between nitrogen and oxygen during fuel combustion.
N₂O (and nitrogen oxide (NO_X)) emissions from
motor vehicles and motor vehicle engines are closely related to fuel
characteristics, air-fuel mixes, combustion temperatures, and the use
of pollution control equipment.
    Nitrous oxide emissions from section 202 sources decreased by 27%
between 1990 and 2006, largely due to decreased emissions from
passenger cars and light-duty trucks.\114\ Earlier generation control
technologies initially resulted in higher N₂O emissions,
causing a 24% increase in N₂O emissions from motor vehicles
between 1990 and 1995. Improvements in later-generation emission
control technologies have reduced N₂O output, resulting in a
41% decrease in N₂O emissions from 1995 to 2006. Emissions
of N₂O from section 202 sources, and U.S. and global
emissions are presented below in Table V-3.
---------------------------------------------------------------------------

    \114\ Detailed nitrous oxide emissions data for section 202
source categories are presented in the Emissions Technical Support
Document.

          Table V-3--Section 202 N2O, U.S. and Global Emissions
------------------------------------------------------------------------
                                                            Sec 202 N2O
             U.S. Emissions                    2006            share
                                                             (percent)
------------------------------------------------------------------------
Section 202 N2O.........................            29.5
All U.S. N2O............................           367.9             8.0
U.S. emissions of Sec 202 GHG...........          1665.4             1.8
All U.S. GHG emissions..................          7054.2             0.4
------------------------------------------------------------------------



                                                            Sec 202 N2O
                                                             share (in
            Global Emissions                   2000            2000)
                                                             (percent)
------------------------------------------------------------------------
All global N2O emissions................         3,113.8             1.6
Global transport GHG emissions..........         5,315.2             0.9
All global GHG emissions................        36,727.9             0.1
------------------------------------------------------------------------


[[Page 44431]]



                                                           Share of U.S.
        Other Sources of U.S. N2O              2006        N2O emissions
                                                             (percent)
------------------------------------------------------------------------
Agricultural Soil N2O emissions.........           265.0            72.0
Nitric Acid N2O emissions...............            15.6             4.3
------------------------------------------------------------------------

    Past experience has shown that substantial emissions reductions can
be made by small N₂O sources. For example, the
N₂O emissions from adipic acid production is smaller than
that of Section 202 sources, and this sector reduced its emission by
over 60 percent from 1990 to 2006 as a result of voluntary adoption of
N₂O abatement technology by the three major U.S. adipic acid
plants.\115\
---------------------------------------------------------------------------

    \115\ Inventory of U.S. Greenhouse Gas Emissions and Sinks:
1990-2006 (USEPA #430-R-08-005), p.2-22.
---------------------------------------------------------------------------

e. Hydrofluorocarbons Emissions From Section 202 Source Categories
    Hydrofluorocarbons (a term which encompasses a group of eleven
related compounds) are progressively replacing CFCs and HCFCs in
section 202 cooling and refrigeration systems as they are being phased
out under the Montreal Protocol and Title VI of the CAA.\116\
---------------------------------------------------------------------------

    \116\ 2006 IPCC Guidelines, Volume 3, Chapter 7. Page 43.
---------------------------------------------------------------------------

    Hydrofluorocarbons were not used in motor vehicles or refrigerated
rail and marine transport in the U.S. in 1990, but by 2006 emissions
had increased to 70 Tg CO₂e.\117\ Emissions of HFC from
section 202 sources, and U.S. and global emissions are presented below
in Table V-4.
---------------------------------------------------------------------------

    \117\ Detailed HFC emissions data for section 202 source
categories are presented in Tables in the Emissions Technical
Support Document.

          Table V-4--Section 202 HFC, U.S. and Global Emissions
------------------------------------------------------------------------
                                                            Sec 202 HFC
             U.S. Emissions                    2006            share
                                                             (percent)
------------------------------------------------------------------------
Section 202 HFC.........................            69.5
All U.S. HFC............................           124.5            55.8
U.S. emissions of Sec 202 GHG...........          1665.4             4.2
All U.S. GHG emissions..................          7054.2             1.0
------------------------------------------------------------------------



                                                            Sec 202 HFC
                                                             share (in
            Global Emissions                   2000            2000)
                                                             (percent)
------------------------------------------------------------------------
All global HFC emissions................           259.2            20.3
Global transport GHG emissions..........         5,315.2             1.0
All global GHG emissions................        36,727.9             0.1
------------------------------------------------------------------------



                                                           Share of U.S.
        Other Sources of U.S. HFC              2006        HFC emissions
                                                             (percent)
------------------------------------------------------------------------
HCFC-22 Production......................            13.8            11.1
Other ODS Substitutes...................            41.2            33.1
------------------------------------------------------------------------

    EPA notes that section 202 HFC emissions are the largest source of
HFC emissions in the United States, that these emissions increased by
274% from 1995 to 2006, and that section 202 sources are also the
largest source of emissions of high GWP gases (i.e., HFCs, PFCs or
SF₆) in the U.S. Thus, a decision not to set standards for
HFCs under section 202 could be viewed as precedential with respect to
the likelihood of future regulatory actions for any of these three
gases.
f. Perfluorocarbons and Sulfur Hexafluoride
    Perfluorocarbons (PFCs) and sulfur hexafluoride (SF₆)
are not emitted from motor vehicles or motor vehicle engines in the
United States.
g. Total GHG Emissions From Section 202 Source Categories
    We note if ``air pollutant'' were defined as the collective group
of four to six GHGs, the emissions of a single component (e.g.,
CO₂) could theoretically support a positive contribution
finding. We also solicit comment on whether the fact that total GHG
emissions from section 202 source categories are approximately 4.3% of
total global GHG emissions would mean that adopting this definition of
``air pollutant'' would make it unnecessary to assess the individual
GHG emissions levels less than that amount. Table V-5 below presents
the contribution of individual GHGs to total GHG emissions from section
202 sources, and from all sources in the U.S.

              Table V-5--Contribution of Individual gases in 2006 to Section 202 and U.S. Total GHG
                                                  (In percent)
----------------------------------------------------------------------------------------------------------------
                                                   CO2        CH4        N2O        HFC        PFC        SF6
----------------------------------------------------------------------------------------------------------------
Section 202...................................       93.9        0.1        1.8        4.2

[[Page 44432]]


U.S. Total....................................       84.8        7.9        5.2        1.8        0.1        0.2
----------------------------------------------------------------------------------------------------------------

    Emissions of GHG from section 202 sources, and U.S. and global
emissions are presented below in Table V-6.

          Table V-6--Section 202 GHG, U.S. and Global Emissions
------------------------------------------------------------------------
                                                            Sec 202 GHG
             U.S. Emissions                    2006            share
                                                             (percent)
------------------------------------------------------------------------
Section 202 GHG.........................          1665.4
All U.S. GHG emissions..................          7054.2            23.6
------------------------------------------------------------------------



                                                            Sec 202 GHG
                                                             share (in
            Global Emissions                   2000            2000)
                                                             (percent)
------------------------------------------------------------------------
Global transport GHG emissions..........         5,315.2            29.5
All global GHG emissions................        36,727.9             4.3
------------------------------------------------------------------------



                                                           Share of U.S.
        Other Sources of U.S. GHG              2006       GHG  emissions
                                                             (percent)
------------------------------------------------------------------------
Electricity Sector emissions............          2377.8            33.7
Industrial Sector emissions.............          1371.5            19.4
------------------------------------------------------------------------

h. Summary of Requests for Comment
    EPA is seeking comment on the approach outlined above in the
context of section 202 source categories, regarding how ``air
pollutant'' should be defined, and contribution analyzed. Specifically,
EPA is interested in comments regarding the data and comparisons
underlying the above example contained in Emissions Technical Support
Document. We also welcome comment on prior precedents for assessing
contributions, as well as the potential precedential impact of a
positive section 202 contribution findings for other potential sources
of these and other GHGs. We also welcome comment on the relationship of
these proposals to existing U.S. climate change emissions reduction
programs and the magnitude of reductions sought under these programs.

VI. Mobile Source Authorities, Petitions, and Potential Regulation

A. Mobile Sources and Title II of the Clean Air Act

    Title II of the CAA provides EPA's statutory authority for mobile
source air pollution control. Mobile sources include cars and light
trucks, heavy trucks and buses, nonroad recreational vehicles (such as
dirt bikes and snowmobiles), farm and construction machines, lawn and
garden equipment, marine engines, aircraft, and locomotives. The Title
II program has led to the development and widespread commercialization
of emission control technologies throughout the various categories of
mobile sources. Overall, the new technologies sparked by EPA regulation
over four decades have reduced the rate of emission of regulated
pollutants from personal vehicles by 98% or more, and are key
components of today's high-tech cars and SUVs. EPA's heavy-duty,
nonroad, and transportation fuels regulatory programs have likewise
promoted both pollution reduction and cost-effective technological
innovation.
    In this section, we consider how Title II authorities could be used
to reduce GHG emissions from mobile sources and the fuels that power
them. The existing mobile source emissions control program provides one
possible model for how EPA could use Title II of the CAA to achieve
long-term reductions in mobile source GHG emissions. The approach would
be to set increasingly stringent performance standards that
manufacturers would be required to meet over 10, 20 or 30 years using
flexible compliance mechanisms like emissions averaging, trading and
banking to increase the economic effectiveness of emission reductions
over less flexible approaches. These performance standards would
reflect EPA's evaluation of available and developing technologies,
including the potential for technology innovation, that could provide
sustained long-term GHG emissions reductions while allowing mobile
sources to satisfy the full range of consumer and business needs.
    Another approach we explore is the extent to which CAA authorities
could be used to establish a cap-and-trade system for reducing mobile
source-related GHG emissions that could provide even greater
flexibility to manufacturers in finding least cost emission reductions
available within the sector. With respect to cars and light trucks, we
also present and discuss an alternative approach to standard-setting,
focused on technology already in the market today in evaluating near
term standards, that EPA began developing in 2007 as part of an inter-
agency effort in response to the Massachusetts decision and the
President's May 2007 directive. This approach took into consideration
and used as a starting point the President's 20-in-10 goals for vehicle
standards. Congress subsequently

[[Page 44433]]

addressed many of the 20-in-10 goals through its action in passing EISA
in December 2007.
    EPA seeks public comment on how a Title II regulatory program could
serve as an approach for addressing GHG emissions from mobile sources.
In addition, EPA invites comments on the following specific questions:
     What are the implications for developing Title II programs
in view of the global and long-lived nature of GHGs?
     What factors should be considered in developing a long-
term, i.e, 2050, GHG emissions target for the transportation sector?
     Should the transportation sector make GHG emission
reductions proportional to the sector's share of total U.S. GHG
emissions or should other approaches be taken to determining the
relative contribution of the transportation sector to GHG emission
reductions?
     What are the merits and challenges of different regulatory
timeframes such as 5 years, 10-15 years, 30-40 years?
     Should Title II GHG standards be based on environmental
need, current projections of future technology feasibility, and/or
current projections of future net societal benefits?
     Could Title II accommodate a mobile sources cap-and-trade
program and/or could Title II regulations complement a broader cap-and-
trade program?
     Should trading between mobile sources and sources in other
sectors be allowed?
     Is it necessary or would it be helpful to have new
legislation to complement Title II (such as legislation to provide
incentives for the development and commercialization of low-GHG mobile
source technologies)?
     How best can EPA fulfill its CAA obligations under Title
II yet avoid inconsistency with NHTSA's regulatory approach under EPCA?

EPA also invites comments on whether there are specific limitations of
a Title II program that would best be addressed by new legislation.
1. Clean Air Act Title II Authorities
    In this section we review the Title II provisions that could be
applied to GHG emissions from various categories of motor vehicles and
fuels. For each provision, we describe the relevant category of mobile
sources, the terms of any required ``endangerment'' finding, and the
applicable standard-setting criteria. We also identify the full range
of factors EPA may consider, including costs and safety, and discuss
the extent to which standards may be technology-forcing.
a. CAA Section 202(a)
    Section 202(a)(1) provides broad authority to regulate new ``motor
vehicles,'' which are on-road vehicles. While other provisions of Title
II address specific model years and emissions of motor vehicles,
section 202(a)(1) provides the authority that EPA would use to regulate
GHGs from new on-road vehicles. The ICTA petition sought motor vehicle
GHG emission standards under this section of the Act.
    As previously discussed, section 202(a)(1) makes a positive
endangerment finding a prerequisite for setting emission standards for
new motor vehicles. Any such standards ``shall be applicable to such
vehicles * * * for their useful life.'' Emission standards under CAA
section 202(a)(1) are technology-based, i.e. the levels chosen must be
premised on a finding of technological feasibility. They may also be
technology-forcing to the extent EPA finds that technological advances
are achievable in the available lead time and that the reductions such
advances would obtain are needed and appropriate. However, EPA also has
the discretion to consider and weigh various additional factors, such
as the cost of compliance (see section 202(a)(2)), lead time necessary
for compliance (section 202(a)(2)), safety (see NRDC v. EPA, 655 F. 2d
318, 336 n. 31 (D.C. Cir. 1981)) and other impacts on consumers, and
energy impacts. Also see George E. Warren Corp. v. EPA, 159 F.3d 616,
623-624 (D.C. Cir. 1998). CAA section 202(a)(1) does not specify the
weight to apply to each factor, and EPA accordingly has significant
discretion in choosing an appropriate balance among the factors. See
EPA's interpretation of a similar provision, CAA section 231, at 70 FR
69664, 69676 (Nov. 17, 2005), upheld in NACAA v. EPA, 489 F.3d 1221,
1230 (2007).
b. CAA Section 213
    CAA section 213 provides broad authority to regulate emissions of
non-road vehicles and engines, which are a wide array of mobile sources
including ocean-going vessels, locomotives, construction equipment,
farm tractors, forklifts, harbor crafts, and lawn and garden equipment.
    CAA section 213(a)(4) authorizes EPA to establish standards to
control pollutants, other than NO_X, volatile organic
compounds and CO, which are addressed in section 213(a)(3), if EPA
determines that emissions from nonroad engines and vehicles as a whole
contribute significantly to air pollution ``which may reasonably be
anticipated to endanger public health or welfare''. Once this
determination is made, CAA section 213(a)(4) provides that EPA ``may''
promulgate standards it deems ``appropriate'' for ``those classes or
categories of new nonroad engines and new nonroad vehicles (other than
locomotives or engines used in locomotives), which in the
Administrator's judgment, cause or contribute to, such air pollution,
taking into account costs, noise, safety, and energy factors associated
with the application of available technology to those vehicles and
engines.'' As with section 202(a)(1), this provision authorizes EPA to
set technology-forcing standards to the extent appropriate considering
all the relevant factors.
    CAA section 213(a)(5) authorizes EPA to adopt standards for new
locomotives and new locomotive engines. These standards must achieve
the greatest degree of emissions reduction achievable through the
application of available technology, giving appropriate consideration
to the cost of applying such technology, lead time, noise, energy and
safety. Section 213(a)(5) does not require that EPA review the
contribution of locomotive emissions to air pollution which may
reasonably be expected to endanger public health or welfare before
setting emission standards, although in the past, EPA has provided such
information in its rulemakings.
c. CAA Section 231
    CAA section 231(a) provides broad authority for EPA to establish
emission standards applicable to the ``emission of any air pollutant
from any class or classes of aircraft engines, which in the
Administrator's judgment, causes, or contributes to, air pollution
which may reasonably be anticipated to endanger public health or
welfare.'' NACAA v. EPA, 489 F.3d 1221, 1229 (D.C. Cir. 2007). As with
sections 202(a) and 213(a)(4), this provision authorizes, but does not
require, EPA to set technology-forcing standards to the extent
appropriate considering all the relevant factors, including noise,
safety, cost and necessary lead time for the development and
application of requisite technology.
    Unlike the motor vehicle and non-road programs, however, EPA does
not directly enforce its standards regulating aircraft engine
emissions. Under CAA section 232, the Federal Aviation Administration
(FAA) is required to prescribe regulations to insure compliance with
EPA's standards. Moreover, FAA has authority to regulate aviation
fuels, under Federal Aviation

[[Page 44434]]

Act section 44714. However, under the Federal Aviation Act, the FAA
prescribes standards for the composition or chemical or physical
properties of an aircraft fuel or fuel additive to control or eliminate
aircraft emissions the EPA ``decides under section 231 of the CAA
endanger the public health or welfare[.]''
d. CAA Section 211
    Section 211(c) authorizes regulation of vehicle fuels and fuel
additives (excluding aircraft fuel) as appropriate to protect public
health and welfare, and section 211(o) establishes requirements for the
addition of renewable fuels to the nation's vehicle fuel supply.\118\
In relevant parts, section 211(c) states that, ``[t]he Administrator
may * * * by regulation, control or prohibit the manufacture,
introduction into commerce, offering for sale, or sale of any fuel or
fuel additive for use in a motor vehicle, motor vehicle engine, or
nonroad engine or nonroad vehicle'' if, in the judgment of the
Administrator, any fuel or fuel additive or any emission product of
such fuel or fuel additive causes, or contributes, to air pollution or
water pollution (including any degradation in the quality of
groundwater) which may reasonably be anticipated to endanger the public
health or welfare, * * *'' Similar to other CAA mobile source
provisions, section 211(c)(1) involves an endangerment finding that
includes considering the contribution to air pollution made by the fuel
or fuel additive.
---------------------------------------------------------------------------

    \118\ EPA's authority to regulate fuels under CAA section 211
does not exend to aircraft engine fuel. Instead, under the Federal
Aviatiion Act, the FAA prescribes standads for the compositiion or
chemical or physical properties of an aircraft fuel or additive to
control or eliminate aircraft emissions the EPA ``decides under
section 231 of the Clean Air Act endanger the public health or
welfare[.]'' 49 U.S.C. 44714.
---------------------------------------------------------------------------

    The Energy Policy Act of 2005 also added section 211(o) to
establish the volume-based Renewable Fuels Standard program. Section
211(o) was amended by the Energy Independence and Security Act of 2007.
    Section VI.D of this notice provides more information and
discussion about the CAA section 211 authorities.
2. EPA's Existing Mobile Source Emissions Control Program
    In this notice, EPA is examining whether and how the regulatory
mechanisms employed under Title II to reduce conventional emissions
could also prove effective for reducing GHG emissions. Under Title II,
mobile source standards are technology-based, taking such factors as
cost and lead time into consideration. Various Title II provisions
authorize or require EPA to set standards that are technology forcing,
such as standards for certain pollutants for heavy-duty or nonroad
engines.\119\ Title II also provides for comprehensive regulation of
mobile sources so that emissions of air pollutants from all categories
of mobile sources may be addressed as needed to protect public health
and the environment.
---------------------------------------------------------------------------

    \119\ Technology-forcing standards are based upon performance of
technology that EPA determines will be available (considering
technical feasibility, cost, safety, and other relevant factors)
when the standard takes effect, as opposed to standards based upon
technology which is already available. Technology-forcing standards
further Congress' goal of having EPA project future advances in
pollution control technology, rather than being limited by
technology which already exists. NRDC v. Thomas, 805 F. 2d 410, 428
n. 30 (D.C. Cir. 1981). Technology-forcing standards are performance
standards and do not require the development or use of a specific
technology.
---------------------------------------------------------------------------

    Pursuant to Title II, EPA has taken a comprehensive, integrated
approach to mobile source emission control that has produced benefits
well in excess of the costs of regulation. In developing the Title II
program, the Agency's historic, initial focus was on personal vehicles
since that category represented the largest source of mobile source
emissions. Over time, EPA has established stringent emissions standards
for large truck and other heavy-duty engines, nonroad engines, and
marine and locomotive engines, as well. The Agency's initial focus on
personal vehicles has resulted in significant control of emissions from
these vehicles, and also led to technology transfer to the other mobile
source categories that made possible the stringent standards for these
other categories.
    As a result of Title II requirements, new cars and SUVs sold today
have emissions levels of hydrocarbons, oxides of nitrogen, and carbon
monoxide that are 98-99% lower than new vehicles sold in the 1960s, on
a per mile basis. Similarly, standards established for heavy-duty
highway and nonroad sources require emissions rate reductions on the
order of 90% or more for particulate matter and oxides of nitrogen.
Overall ambient levels of automotive-related pollutants are lower now
than in 1970, even as economic growth and vehicle miles traveled have
nearly tripled. These programs have resulted in millions of tons of
pollution reduction and major reductions in pollution-related deaths
(estimated in the tens of thousands per year) and illnesses. The net
societal benefits of the mobile source programs are large. In its
annual reports on federal regulations, the Office of Management and
Budget reports that many of EPA's mobile source emissions standards
typically have projected benefit-to-cost ratios of 5:1 to 10:1 or more.
Follow-up studies show that long-term compliance costs to the industry
are typically lower than the cost projected by EPA at the time of
regulation, which result in even more favorable real world benefit-to-
cost ratios. Title II emission standards have also stimulated the
development of a much broader set of advanced automotive technologies,
such as on-board computers and fuel injection systems, which are at the
core of today's automotive designs and have yielded not only lower
emissions, but improved vehicle performance, reliability, and
durability.
    EPA requests comment on whether and how the approach it has taken
under Title II could effectively be employed to reduce mobile source
emissions of GHGs. In particular, EPA seeks comment and information on
ways to use Title II authorities that would promote development and
transfer of GHG control technologies for and among the various mobile
source categories. The Agency is also interested in receiving
information on the extent to which GHG-reducing technologies developed
for the U.S. could usefully and profitably be exported around the
world. Finally, EPA requests comments on how the Agency could implement
its independent obligations under the CAA in a manner to avoid
inconsistency with NHTSA CAFE rulemakings, in keeping with the Supreme
Court's observation in the Massachusetts decision (``there is no reason
to think the two agencies cannot both administer their obligations yet
avoid inconsistencies'').
3. Mobile Sources and GHGs
    The domestic transportation sector emits 28% of total U.S. GHG
emissions based on the standard accounting methodology used by EPA in
compiling the inventory of U.S. GHG emissions pursuant to the United
Nations Framework Convention on Climate Change (Figure VI-1).
BILLING CODE 6560-50-P

[[Page 44435]]

[GRAPHIC] [TIFF OMITTED] TP30JY08.029

    The only economic sector with higher GHG emissions is electricity
generation which accounts for 34% of total U.S. GHG emissions. However,
the inventory accounting methodology attributes to other sectors two
sources of emissions that EPA has the authority to regulate under Title
II of the CAA. First, the methodology includes upstream transportation
fuel emissions (associated with extraction, shipping, refining, and
distribution, some of which occur outside of the U.S.) in the emissions
of the industry sector, not the transportation sector. However,
reducing transportation fuel consumption would automatically and
proportionally reduce upstream transportation fuel-related GHG
emissions as well. Second, nonroad mobile sources (such as
construction, farm, and lawn and garden equipment) are also included in
the industry sector contribution. All of these emissions can be
addressed under CAA Title II authority, at least with respect to
domestic usage. Including these upstream transportation fuel (some of
which occur outside of U.S. boundaries) and nonroad equipment GHG
emissions in the mobile sources inventory would raise the contribution
from mobile sources and the fuels utilized by mobile sources to
approximately 36% of total U.S. GHG emissions. Since, based on 2004
data, the U.S. emits about 23% of global GHG emissions, under the
traditional accounting methodology the U.S. transportation sector
contributes about 6% of the total global inventory. If upstream
transportation fuel emissions and nonroad equipment emissions are also
included, U.S. mobile sources are responsible for about 8% of total
global GHG emissions.
    Personal vehicles (cars, sport utility vehicles, minivans, and
smaller pickup trucks) emit 54% of total U.S. transportation sector GHG
emissions (including nonroad mobile sources), with heavy-duty vehicles
the second largest contributor at 18%, aviation at 11%, nonroad sources
at 8%, marine at 5%, rail at 3%, and pipelines at 1% (Figure VI-2).
CO₂ is responsible for about 95% of transportation GHG
emissions, with air conditioner refrigerant HFCs accounting for 3%,
vehicle tailpipe nitrous oxide emissions for 2%, and vehicle tailpipe
methane emissions for less than 1% (Figure VI-3).

[[Page 44436]]

[GRAPHIC] [TIFF OMITTED] TP30JY08.030

[GRAPHIC] [TIFF OMITTED] TP30JY08.031

    As noted previously, global climate change is a long-term problem.
Climate experts such as the IPCC often use 2050 as a key reference
point for future projections. Long-term projections of U.S. mobile
source GHG emissions show that there is likely to be a major increase
in transportation GHG emissions in the future.
    Prior to the passage of EISA, U.S. transportation GHG emissions
(including upstream fuel emissions) were projected to grow
significantly, from about 2800 million metric tons in 2005 to about
4800 million metric tons in 2050 (see Figure VI-4, top curve). The fuel
economy and renewable fuels provisions of EISA (Figure VI.A.2.-4,
second curve from top) provide significant near-term mobile source GHG
emissions reductions relative to the non-EISA baseline case. However,
addressing climate change requires setting long-term goals. President
Bush has proposed a new goal of stopping the growth of GHG emissions by
2025, and the IPCC has modeled several long-term climate mitigation
targets for 2050.

[[Page 44437]]

    Using Title II authority, mobile sources could achieve additional
GHG emission reductions based on a variety of criteria including the
amount of reduction needed, technological feasibility and cost
effectiveness. While EISA's fuel economy and renewable fuel
requirements will contribute to mobile source GHG emission reductions,
its fuel economy standards affect only CO₂ emissions and do
not apply to the full range of mobile source categories. EISA also
specifies that fuel economy standards be set for no more than five
years at a time, effectively limiting the extent to which those
standards can take into account advancing technologies. Moreover, its
renewable fuel provisions are limited in the extent to which they
provide for GHG emission reductions, although EISA does mandate the use
of renewable fuels that meet different lifecycle GHG emission reduction
requirements.
    Under Title II, EPA has broad authority to potentially address all
GHGs from all categories of mobile sources. In addition, Title II does
not restrict EPA to specific timeframes for action. If circumstances
warrant, EPA could set longer term standards and promote technological
advances by basing standards on the performance of technologies not yet
available but which are projected to be available at the time the
standard takes effect. Title II also provides authority to potentially
require GHG emission reductions from transportation fuels.
Consequently, the CAA authorizes EPA to consider what GHG emissions
reductions might be available and appropriate to require from the
mobile source sector, consistent with the Act.
    EPA has not determined what level of GHG emission reduction would
be appropriate from the mobile source sector in the event a positive
endangerment finding is made, although this ANPR includes some
discussion of possible reductions. Any such determination is
necessarily the province of future rulemaking activity. Without
prejudging this important issue, and for illustrative purposes only,
the final three curves in Figure VI-4 illustrate the additional
reductions mobile sources would have to achieve if mobile sources were
to make a proportional contribution to meeting the President's climate
goal, the IPCC 450 CO₂ ppm stabilization scenario, and an
economy-wide GHG emissions cap based on a 70% reduction in 2005
emissions by 2050.\120\ As the figure illustrates, EISA provides about
25%, 15% and 10% of the transportation GHG emissions reductions that
would be needed for mobile sources to make a proportional contribution
to meeting the President's climate goal by 2050 (Figure VI-4, third
curve), the IPCC 450 CO₂ ppm stabilization scenario in 2050
(Figure VI-4, fourth curve), and a 70% reduction in 2005 levels in 2050
(Figure VI-4, bottom curve), respectively.\121\ These curves shed light
on the possible additional role the transportation sector could play in
achieving reductions, but do not address whether such reductions would
be cost effective compared to other sectors. Title II regulation of GHG
emissions could conceivably achieve greater emissions reductions so
that mobile sources would make a larger contribution to meeting these
targets. EPA requests comment on the usefulness of the information
provided in Figure VI-4 and on approaches for determining what
additional mobile source GHG emissions reductions would be appropriate.
As described later in this section, our assessment of available and
developing mobile source technologies for reducing GHG emissions
indicates that mobile sources could feasibly achieve significant
additional reductions.
---------------------------------------------------------------------------

    \120\ Prior to the passage of EISA, an EPA analysis projected
that, absent additional regulatory approaches, transportation would
provide about one-tenth of the GHG emission reductions that would be
required to comply with an emissions cap based on a 70% reduction
from 2005 levels in 2050, even though transportation is responsible
for 28% of the official U.S. GHG emissions inventory.
    \121\ Calculation of the GHG emission reductions that EISA's
fuel economy provisions will achieve include standards that result
in an industry-wide fleet average fuel economy of 35 miles per
gallon by 2020.

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

[[Page 44438]]

[GRAPHIC] [TIFF OMITTED] TP30JY08.032

4. Potential Approaches for Using Clean Air Act Title II To Reduce
Mobile Source GHG Emissions
    The regulatory approach and principles that guided development of
our current mobile source emissions control program may prove useful in
considering a possible mobile source GHG emissions control strategy
under Title II of the CAA. As explained above, under Title II, EPA
could potentially apply its historical approach for regulating
traditional tailpipe emissions to long-term mobile source GHG emissions
control, with the aim of providing strong incentives for technological
innovation. The Agency invites public comment on the principles and
underlying legal authority it has applied in the past and other
possible principles for establishing GHG emissions standards under
Title II, including--
     Coverage of all key vehicle, engine, and equipment sub-
sectors in the entire transportation sector so that GHG emission
standards are set not only for cars and light trucks, but for heavy-
duty vehicles, non-road engines and equipment, including locomotive and
marine engines, and aircraft as well. This broader regulatory coverage
would provide more comprehensive mobile source GHG emissions reductions
and market incentives to seek the most cost-effective solutions within
the sector.
     Coverage of all GHGs emitted by the transportation sector
by setting emissions standards that address every GHG for which the
Agency makes the appropriate cause or contribute endangerment finding.
     Inclusion of transportation fuels in the program by
considering vehicles and fuels as a system, rather than as isolated
components.
     Addressing transportation fuels by setting GHG standards
that account for the complete lifecycle of GHG emissions, including
upstream GHG emissions associated with transportation fuel
production.\122\
---------------------------------------------------------------------------

    \122\ EPA invites comment on how such an approach would interact
with GHG regulations under other parts of the CAA or with a possible
economy-wide approach.
---------------------------------------------------------------------------

     Identifying long-term U.S. mobile source GHG emissions
targets based on scientific assessments of environmental need, and
basing the stringency of standards for individual mobile source sub-
sectors on technology feasibility, cost and fuel savings, taking into
account the relationship of mobile source reductions to reductions in
other sectors under any economy-wide program.
     Allowing for staggered rulemakings for various sub-sectors
and fuels, rather than regulating all mobile source entities at one
time. EPA seeks comment on its CAA authority in this area, as well as
on an approach to base the timing of the staggered rulemakings on
factors such as the contribution of the mobile source sub-sector to the
overall GHG emissions inventory and the lead time necessary for the
commercialization of innovative technology.
     Use of Title II statutory authority to adopt technology-
forcing standards, when appropriate, in conjunction with periodic
reviews of technology and other key analytical inputs as a ``reality
check'' to determine whether mid-course corrections in GHG emissions
standards are needed.
     Use of our statutory authority to increase the rate of
emissions reduction targets over time while allowing sufficient time
for entrepreneurs and engineers to develop cost-effective technological
solutions and minimize the risk of early retirement of capital
investments.
     Establishment of a flexible compliance program that would
allow averaging, banking and borrowing, and credit trading. Existing
Title II programs generally allow credit trading only within individual
mobile source sub-sector programs. EPA solicits comments on whether the
global nature of climate

[[Page 44439]]

change supports allowing credit trading between obligated parties
across all mobile source sub-sectors and whether this would allow the
sector as a whole to seek the lowest-cost solutions.
     Design of enforcement programs to ensure real world
emissions reductions over the life of vehicles, engines, and equipment.
     Providing sufficient flexibility so that mobile source GHG
emissions control programs can complement and harmonize with existing
regulatory programs for certain pollutants.
    In developing potential approaches to design of a Title II program,
it is critical for EPA to understand the full ramifications of advanced
technologies. Accordingly, EPA seeks public comment on potential GHG
reducing technologies and their impacts, including availability,
practicality, emissions reduction potential, cost, performance,
reliability, and durability. EPA also seeks comment on how best to
balance factors such as the need to send effective long-term signals
that stimulate technology innovation, the imprecision of predictions of
future technology innovation, and the importance of lead time to allow
orderly investment cycles.
    While advanced technology for reducing GHGs would likely increase
the initial cost of vehicles and equipment to consumers and businesses,
it would also increase efficiency and reduce fuel costs. In many cases,
there is the potential for the efficiency advantages of low-GHG
technologies to offset or more than offset the higher initial
technology cost over the lifetime of the vehicle or equipment. EPA
recognizes that not all consumers may understand or value changes to
vehicles that reduce GHG emissions by increasing fuel efficiency, even
though these changes lower fuel costs (see discussion in Section
VI.C.2). One analytic issue that has policy implications is the most
appropriate method for treating future consumer fuel savings when
calculating cost effectiveness for a mobile sources GHG control
strategy. Some analyses that consider the decisions made by automakers
in isolation from the market and consumers exclude future fuel savings
entirely. A second approach, used in models trying to predict future
consumer behavior based on past experience, counts only those future
fuel savings which consumers implicitly value in their new vehicle
purchase decisions. A third method, reflecting a societal-wide
accounting of benefits, includes all future fuel savings over vehicle
lifetimes, whether overtly valued by new vehicle purchasers or not. EPA
seeks comments on what could be done under Title II, or under any new
legislation to complement Title II, to establish economic incentives
that send long-term market signals to consumers and manufacturers that
would help spark development of and investment in the necessary
technology innovation.
    An effective mobile source emissions compliance and enforcement
program is fundamental to ensuring that the environmental benefits of
the emission standards are achieved. We request comments on all aspects
of the compliance approaches discussed in this notice and any other
approaches to a compliance program for mobile source GHG emissions
control. Topics to address could include, but are not limited to,
methods for classifying, grouping and testing vehicles for
certification, useful life and component durability demonstration, in-
use testing, warranty and tampering, prohibited acts, and flexibilities
for manufacturers.
    Historically, EPA's programs to reduce criteria pollutants have
typically included provisions to allow the generation, averaging,
banking, and trading of emission credits within a vehicle or engine
category. For example, there are averaging, banking, and trading (ABT)
programs for light-duty vehicles, heavy-duty engines, and nonroad
engines, among others. In these programs, manufacturers with vehicles
or engines designed to over-comply with the standards can generate
credits. These credits can then be used by that manufacturer or sold to
other manufacturers in order to allow similar vehicles or engines with
emissions above the standards to be certified and sold.
    However, for a variety of reasons, we have in most cases not
provided for trading of emission credits from one mobile source
category to another. For example, credits generated in the light-duty
vehicle program cannot be used for heavy-duty engines to comply, or
credits generated for lawn and garden equipment cannot be used for
larger gasoline engines to comply. These limitations are generally
grounded in characteristics of required pollutants that do not
necessarily apply in the case of GHG emissions. For instance, in the
case of hydrocarbon emissions, because our programs are meant, in part,
to reduce the pollutant in areas where it most contributes to ozone
formation, we have not allowed farm tractors in rural areas to generate
credits that would allow urban passenger cars to be sold with little or
no emission control. Similarly, for problems like carbon monoxide ``hot
spots'' or direct, personal exposure to diesel PM, it has been
important to ensure a certain minimum degree of control from each
vehicle or engine, rather than allowing the very localized benefits to
be ``traded away.''
    Given the global nature of the major GHGs, we request comment on
whether new provisions could be used to allow broad trading of
CO₂-equivalent emission credits among the full range of
mobile sources, and if so, how they could be designed, including
highway and nonroad vehicles and engines as well as mobile source
fuels.
    EPA has also considered the potential of GHG emissions leakage to
other domestic economic sectors, or to other countries, should EPA
adopt Title II standards for motor vehicle GHG emissions and GHG
emissions from transportation fuels. As discussed in more detail later
in this section, there are transportation fuels (such as grid
electricity) that do not result in tailpipe GHG emissions, but that do
result in GHG emissions when the fuel is produced. Greater use of such
fuels in transportation would reduce GHG emissions covered by Title II,
but would increase GHG emissions covered by Title I, requiring
coordination among the CAA programs to ensure the desired level of
overall GHG control. In addition, GHG emissions from potential land use
changes caused by transportation fuel changes could cause GHG emissions
leakage unless accounted for in any transportation fuels GHG program.
Finally, since transportation fuels can be fungible commodities, if
other countries do not adopt similar GHG control programs, it is
possible that lower-lifecycle GHG fuels will be concentrated in the
U.S. market, while higher-lifecycle GHG fuels will be concentrated in
unregulated markets. For example, sugar cane-based ethanol, if it were
determined to have more favorable upstream GHG emissions, could shift
from the Brazilian to the U.S. market, and corn-based ethanol, if it
were determined to have less favorable upstream GHG emissions, could
shift from the U.S. to the Brazilian market. This shifting could ease
compliance with U.S. transportation fuel GHG regulations, but could
actually increase global GHG emissions due to the GHG emissions that
would result from transporting both types of ethanol fuels over greater
distances. EPA seeks comments on all possible GHG emissions leakage
issues associated with mobile source GHG regulation, and in particular
on whether the theoretical concern with fungible transportation fuels
is likely to be realized.
    While the preceding discussion has focused on using the existing
CAA Title

[[Page 44440]]

II model for regulating mobile source GHG emissions, there are other
alternative regulatory approaches on which EPA invites comments. In
particular, long-term mobile source GHG emissions reductions from
vehicles and equipment might be achieved by establishing GHG emissions
caps on vehicle, engine, and/or equipment manufacturers to the extent
authorized by the CAA. EPA's existing regulatory program uses
performance standards that are rate-based, meaning that they require
manufacturers to meet a certain gram/mile average for their fleet, as
in the Tier 2 light-duty vehicle program. Manufacturers produce
vehicles with varying rates of emissions performance, and through
averaging, banking, and trading demonstrate compliance with this
performance standard on a sales-weighted average basis. While a
manufacturer must take its fleet mix of higher-emitting and lower-
emitting models into account in demonstrating compliance, the sales-
weighted average is independent of overall sales as long as the fleet
mix does not change. As a result, a manufacturer's fleet may emit more
or less total pollution depending on its total sales, so long as the
sales-weighted average emissions of its vehicles do not exceed the
standard.
    In a cap-and-trade program, the standard set by EPA would not be an
average, sales-weighted rate of emissions, but rather a cap on overall
emissions from a manufacturer's production. Under such a program, the
emissions attributable to a manufacturer's fleet could not grow with
sales unless the manufacturer obtained (e.g., through trading)
additional allowances to cover higher emissions. Presumably, EPA could
assign a VMT or usage value to be used by manufacturers, and
manufacturers would demonstrate compliance by combining the rate of
performance of their vehicles, their sales volume, and the assigned VMT
or usage value to determine overall emissions.
    EPA could set standards under an emissions cap-and-trade program by
assessing the same kind of factors as we have in the past: Availability
and effectiveness of technology, cost, safety, energy factors, etc.
Setting an appropriate emissions cap would be more complex, and EPA
would need to demonstrate that the cap is appropriate, given that
changes in sales levels (both industry-wide and for individual
manufacturers) must be accounted for in the standard-setting process.
An emissions cap approach also raises difficult issues of how allowable
emissions under the cap would be allocated among the manufacturers,
including new entrants.
    EPA invites comment on all issues involving this emissions cap-and-
trade approach, including comment on relevant technical and policy
issues, and on EPA's authority to adopt such an approach under Title
II.
    A third possible model for regulating mobile source GHG emissions
would combine elements of these approaches. This type of hybrid
approach would include, as one element, either rate-based GHG emissions
performance standards similar to the existing mobile source program for
conventional pollutants or GHG emissions caps for key vehicle, engine,
and/or equipment manufacturers, both of which would be promulgated
under Title II of the CAA. The second element of this hybrid approach
would be an upstream emissions cap on fuel refiners for all life-cycle
GHG emissions associated with transportation fuels, including both
upstream fuel production GHG emissions and downstream vehicle GHG
emissions, to the extent authorized under the CAA or future climate
change legislation. For a discussion of issues associated with
including direct mobile source obligations in combination with an
economy-wide approach, see section III.F.3.
    An important interrelationship between stationary sources and
mobile sources would develop if grid electricity becomes a more
prevalent transportation fuel in the future. There is considerable
interest, both by consumers and automakers, in the possible development
and commercialization of plug-in hybrid electric vehicles (PHEVs) that
would use electricity from the grid as one of two sources of energy for
vehicle propulsion. Use of grid electricity would yield zero vehicle
tailpipe GHG emissions, providing automakers with a major incentive to
consider PHEVs, which may be appropriate given that vehicle cost is the
single biggest market barrier to PHEV commercialization. But it would
also result in a net increase in demand for electricity, which could
add to the challenge of reducing GHG emissions from the power sector.
Any evaluation of the overall merits of using grid electricity as a
transportation fuel could not be done in isolation, but would require a
coordinated assessment and approach involving both mobile sources under
CAA Title II and stationary sources under CAA Title I. Linking efforts
under Titles I and II would allow for needed coordination regarding any
type of future transportation fuel that would have zero vehicle
tailpipe GHG emissions but significant fuel production GHG emissions.
    EPA seeks comment on all aspects, including the advantages and
disadvantages, of using Title II regulations to complement an economy-
wide cap-and-trade GHG emissions program.
    EPA also seeks public comment on the available authority for, and
the merits of, allowing credit trading between mobile sources and non-
mobile source sectors. One of the potential limitations of allowing
credit trading only within the transportation sector is that it would
not permit firms to take advantage of emission reduction opportunities
available elsewhere in the economy. In particular, EPA requests comment
on the advantages and disadvantages of allowing trading across sectors,
and how to ensure that credit trading would have environmental
integrity and that credits are real and permanent.
    Finally, EPA seeks public comment on two remaining issues: (1) How
a CAA Title II mobile source GHG emissions control program and NHTSA's
corporate average fuel economy program for cars and light-duty trucks
could best be coordinated; and (2) whether and how Title II, or other
provisions in the CAA, could be used to promote lower vehicle miles
traveled and equipment activity.

B. On-Highway Mobile Sources

1. Passenger Cars and Light-Duty Trucks
    In this section, we discuss and request comment on several
potential approaches for establishing light-duty vehicle GHG emission
standards under section 202(a)(1). These approaches build off of, to
varying extents, the analysis EPA undertook during 2007 to support the
development of a near-term control program for GHG emissions for
passenger cars and light duty trucks under the authorities of Title II
of the CAA.
    We begin this section with a discussion of one potential approach
for establishing GHG standards under section 202(a) of the CAA that
reflects EPA's historical approach used for traditional pollutants,
including the principles EPA has used in the past under Title II. This
approach focuses on long-term standard setting based on the technology-
forcing authority provided under Title II. Next we present and discuss
the results of alternative approaches to standard-setting which EPA
considered during 2007 in the work performed under EO 13432. This
alternative approach is based on setting near-term standards based
primarily on technology already in the market today.

[[Page 44441]]

This is followed by a discussion of the wide range of technologies
available today and technologies that we project will be available in
the future to reduce GHG emissions from light-duty vehicles. We next
include a discussion of a potential approach to reduce HFC, methane,
N₂O, and vehicle air conditioning-related CO₂
emissions. We conclude with a discussion of the key implementation
issues EPA has considered for the development of a potential light-duty
vehicle GHG control program.
    Our work to date indicates that there are significant reductions of
GHG emissions that could be achieved for passenger cars and light-duty
trucks up to 2020 and beyond that would result in large net monetized
benefits to society. For example, taking into account specific vehicle
technologies that are likely to be available in that time period and
other factors relevant to motor vehicle standard-setting under the CAA,
EPA's analysis suggests that substantial reductions can occur where the
cost-per-ton of GHG reduced is more than offset by the value of fuel
savings, and the net present value to society could be on the order of
$340 to $830 billion without considering benefits of GHG reductions
(see section VI.B.1.b).\123\
---------------------------------------------------------------------------

    \123\ These estimates do not account for the future CAFE
standards that will be established under EISA.
---------------------------------------------------------------------------

a. Traditional Approach to Setting Light-Duty Vehicle GHG Standards
    In this section we discuss and request comment on employing EPA's
traditional approach to setting mobile source emissions standards to
develop standards aimed at ensuring continued, long-term, technology-
based GHG reductions from light-duty vehicles, in light of the unique
properties of GHG emissions. We also request comment on how EPA could
otherwise use its CAA Title II authorities to provide incentives to the
market to accelerate the development and introduction of ultra clean,
low GHG emissions technologies.
    Based on our work to date, we expect that such an approach could
result in standards for the 2020 to 2025 time frame that reflect a
majority of the new light-duty fleet achieving emission reductions
based on what could be accomplished by many of the most advanced
technologies we know of today (e.g., hybrids, diesels, plug-in hybrid
vehicles, full electric vehicles, and fuel cell vehicles, all with
significant use of light-weight materials). Our analysis (presented in
section VI.B.1.b) indicates that standards below 250 g/mile
CO₂ (above 35 mpg) could be achievable in this time frame,
and the net benefit to society could be in excess of $800 billion.
These estimates, however, do not account for future CAFE standards that
will be established under EISA.
    EPA's historical approach for setting air pollutant standards for
mobile sources has been to assess the capabilities of pollution control
technologies, including advanced control technologies; whether
reductions associated with these technologies are feasible considering
cost, safety, energy, and other relevant factors; and the benefits of
these controls in light of overall public health and environmental
goals. Public health and environmental goals provide the important
context in which this technology-driven process occurs. In many cases
in the past, the goals have involved the need for emissions reductions
to attain and maintain NAAQS.
    As mentioned previously, EPA has utilized the CAA to establish
mobile source programs which apply progressively more stringent
standards over many years, often with substantial lead time to maximize
the potential for technology innovation, and where appropriate, we have
included technology reviews along the way to allow for ``mid-course
corrections,'' if needed. We have also provided incentives for
manufacturers to develop and introduce low emission technologies more
quickly than required by the standards. For example, in our most recent
highway heavy-duty engine standards for PM and NO_X, we
established technology-forcing standards via a rulemaking completed in
2000 which provided six years of lead-time for the start of the program
and nearly ten years of lead-time for the completion of the phase-in of
the standards. In addition, EPA performed periodic technology reviews
to ensure industry was on target to comply with the new standards, and
these reviews allowed EPA to adjust the program if necessary. This same
program provided early incentive emission credits for manufacturers who
introduced products complying with the standards well in advance of the
program requirements.
    Consistent with the CAA and with our existing mobile source
programs, we request comment on using the following traditional
principles for development of long-term GHG standards for light-duty
vehicles: Technology-forcing standards, sufficient lead-time (including
phase-in of standards reflecting use of more advanced technologies),
continual improvements in the rate of emissions reduction, appropriate
consideration of the costs and benefits of new standards, and the use
of flexible mechanisms such as banking and credit trading (between
sources within or outside of this sector). EPA's goal would be to
determine the appropriate level of GHG emission standards to require by
an appropriate point in the future. We would establish the future time
frame in light of the needs of the program. EPA would evaluate a broad
range of technologies in order to determine what is feasible and
appropriate in the time frame chosen, when considering the fleet as a
whole. EPA would analyze the costs and reductions associated with the
technologies, and compare those to the benefits from and the need for
such reductions. We would determine what reductions are appropriate to
require in that time frame, assuming industry started now, and then
determine what appropriate interim standards should be set to most
effectively move to this long-term result.
    In developing long-term standards, we would consider known and
projected technologies which in some cases are in the market in limited
production or which may not yet be in the market but which we project
can be, provided sufficient lead-time. We would consider how broadly
and how rapidly specific technologies could be applied across the
industry. If appropriate, EPA could include technology reviews during
the implementation of new standards to review the industry's progress
and to make adjustments as necessary. EPA would evaluate the amount of
lead-time necessary and if appropriate the phase-in period for long-
term standards. To the extent that future standards may result in
significant increases in advanced technologies such as plug-in electric
hybrid or full electric vehicles, we would consider how a Title II
program might interact with a potential Title I program to ensure that
reductions in GHG emissions due to a decrease in gasoline consumption
are not off-set by increases in GHG emissions from the electric utility
sector. We would also consider the need for flexibilities and
incentives to promote technology innovation and provide incentives for
advanced technologies to be developed and brought to the market. We
would consider the need for orderly manufacturer production planning to
ensure that capital investments are wisely used and not stranded.
Finally, EPA would evaluate the near and long-term costs and benefits
of future standards in order to ensure the appropriate relationship
between benefits and costs, e.g. ensuring that

[[Page 44442]]

benefits of any future standards exceed the costs. This could lead to
standard phase-in schedules significantly different from the two
approaches contained in our Light-duty Vehicle Technical Support
Document analysis (available in the docket for this advance notice);
which under one approach was the same incremental increase in
stringency each year (the 4% per year approach), and for the second
approach lead to large increases in stringency the first several years
followed by small changes in the later years (the model-optimized
approach).
    One critical element in this approach is the time frame over which
we should consider new GHG standards for light-duty vehicles. We
request comment on the advantages and disadvantages of establishing
standards for the 2020 or 2025 time frame, which is roughly consistent
with EPA's traditional approach to setting standards while allowing a
sufficient time period for investment and technological change, and
even longer. There are two major factors which may support a long-term
approach. First, addressing climate change will require on-going
reductions from the transportation sector for the foreseeable future.
Thus, establishing short-term goals will not provide the long-term road
map which the environmental problem requires. Second, providing a long-
term road map could have substantial benefits for the private sector.
The automotive industry itself is very capital intensive--the costs for
developing and producing a major vehicle model is on the order of
several billion dollars. A manufacturer making a major investment to
build a new engine, transmission or vehicle production plant expects to
continue to use such a facility without major additional investments
for at least 15 years, if not more. A regulatory approach which
provides a long-term road map could allow the automotive industry to
plan their future investments in an orderly manner and minimize the
potential for stranded capital investment, thus helping to ensure the
most efficient use of societal resources. A long-term regulatory
program could also provide industry with the regulatory certainty
necessary to stimulate technology development, and help ensure that the
billions of dollars invested in technology research and development are
focused on long-term needs, rather than on short-term targets alone.
    There could also be disadvantages to establishing long-term
standards. For example, uncertainties in the original analysis
underlying the long-term standards could result in overly conservative
or optimistic assumptions about emission reductions could and should be
accomplished. Long-terms standards could also reduce flexibility to
respond to more immediate market changes or other unforeseen events.
EPA has tools, such as technology reviews, that could help reduce these
risks of long-term standards. We request comment on the advantages and
disadvantages of a long-term approach to standard-setting, and any
issues it might raise for integration with an economy-wide approach to
emission reductions.
    More generally, EPA requests comment on the issues discussed in
this section, and specifically the appropriateness of a light-duty
vehicle GHG regulatory approach in which EPA would identify long-term
emissions targets (e.g., the 2020-2025 time frame or longer) based on
scientific assessments of environmental need, and developing standards
based on a technology-forcing approach with appropriate consideration
for lead-time, costs and societal benefits.
b. 2007 Approach to Setting Light-Duty Vehicle Emission Standards
i. CAA and EPCA Authority; Passage of EISA
    As indicated above in section VI.A.2, CAA section 202(a) provides
broad authority to regulate light-duty vehicles. Standards which EPA
promulgates under this authority are technology-based and applicable
for the useful life of a vehicle. EPA has discretion to consider and
weigh various additional factors, including the cost of compliance,
safety and other impacts on consumers, and energy impacts.
    NHTSA authority to set CAFE standards derives from the Energy
Policy and Conservation Act (42 U.S.C. section 6201 et seq.) as amended
by EISA. This statutory authority, enacted in December 2007, directs
NHTSA to consider four factors in determining maximum feasible fuel
economy standards--technological feasibility, economic practicability,
the effect of other standards issued by the government on fuel economy,
and the need of the nation to conserve energy. NHTSA may also take into
account other relevant considerations such as safety.
    EISA amends NHTSA's fuel economy standard-setting authority in
several ways. Specifically it replaces the statutory default standard
of 27.5 miles per gallon for passenger cars with a mandate to establish
separate passenger cars and light truck standards annually beginning in
model year 2011 to reflect the maximum feasible level. It also requires
that standards for model years 2011-2020 be set sufficiently high to
ensure that the average fuel economy of the combined industry-wide
fleet of all new passenger cars and light trucks sold in the U.S.
during MY 2020 is at least 35 miles per gallon. In addition, EISA
provides that fuel economy standards for no more than five model years
be established in a single rulemaking, and mandated the reform of CAFE
standards for passenger cars by requiring that all CAFE standards be
based on one or more vehicle attributes, among other changes.\124\ EISA
also directs NHTSA to consult with EPA and the Department of Energy on
its new CAFE regulations.
---------------------------------------------------------------------------

    \124\ For a full discussion of EISA requirements and NHTSA
interpretation of its statutory authority please see 73 FR 24352
(May 2, 2008).
---------------------------------------------------------------------------

    Pursuant to EISA's amendments to EPCA, NHTSA recently issued a
notice of proposed rulemaking for new, more stringent CAFE standards
for model years 2011-2015 for both passenger cars and light-duty
trucks. 73 FR 24352 (May 2, 2008).
    Prior to EISA's enactment, EPA and NHTSA had coordinated under EO
13432 on the development of CAA rules that would achieve large GHG
emission reductions and CAFE rules that would achieve large
improvements in fuel economy. As discussed later in this section, there
are important differences in the two agencies' relevant statutory
authorities. EPA nevertheless believes that it is important that any
future GHG regulations under CAA Title II and future fuel economy
regulations under NHTSA's statutory authority be designed to ensure
that an automaker's actions to comply with CAA standards not interfere
with or impede actions taken for meeting fuel economy standards and
vice versa. The goals of oil savings and GHG emissions reductions are
often closely correlated, but they are not the same. As the Supreme
Court pointed out in its Massachusetts decision, ``[EPA's] statutory
obligation is wholly independent of DOT's mandate to promote energy
efficiency'', and ``[t]he two obligations may overlap, but there is no
reason to think the two agencies cannot both administer their
obligations and yet avoid inconsistency.'' It is thus important for EPA
and NHTSA to maximize coordination between their programs so that both
the appropriate degree of GHG emissions reductions and oil savings are
cost-effectively achieved, given the agencies' respective statutory
authorities. EPA asks for comment on how EPA's and NHTSA's respective
statutory authorities can best be

[[Page 44443]]

coordinated under all of the alternatives presented in this section so
that inconsistency can be avoided.
ii. 2007 Approach
    In this section, we present an overview of two alternative
approaches for setting potential light-duty vehicle GHG standards based
on our work during 2007 under EO 13432. As noted previously, in
response to Massachusetts v. EPA and as required by EO 13432, prior to
EISA's passage, we coordinated with NHTSA and the Department of Energy
in developing approaches and options for a comprehensive near-term
program under the CAA to reduce GHG emissions from cars and light-duty
trucks.\125\ Results from this effort are discussed below and in a
Technical Support Document, ``Evaluating Potential GHG Reduction
Programs for Light Vehicles'' (referred to as the ``Light-duty Vehicle
TSD'' in the remainder of this notice).
---------------------------------------------------------------------------

    \125\ E.O. 13432 called on the agencies to, ``undertake such
regulatory action, to the maximum extent permitted by law and
determined by the head of the agency to be practicable, jointly with
other agencies.''
---------------------------------------------------------------------------

    The Light-duty Vehicle TSD represents EPA's assessment during 2007
of how a light-duty vehicle program for GHG emissions reduction under
the CAA might be designed and implemented in keeping with program
parameters (e.g., time frame, program structure, and analytical tools)
developed with NHTSA prior to enactment of EISA. In addition, the
Light-duty Vehicle TSD assesses the magnitude of the contribution of
light-duty vehicles to U.S. GHG emissions. It also addresses both
tailpipe CO₂ emissions as measured by EPA tests used for
purposes of determining compliance with CAFE standards, and control of
other vehicular GHG emissions. These other emissions are not accounted
for if the regulatory focus is solely on CO₂, and involve
greenhouse gases that have higher global warming potentials than
CO₂. These emissions, as well as air-conditioning-related
CO₂, are not measured by the existing EPA test procedure for
determining compliance with CAFE standards, so that there is no overlap
with control of these emissions and CAFE standards if these emissions
are controlled under the CAA. As described in the section VI.B.1.d of
this advance notice, these emissions account for 10 percent of light-
duty vehicle GHG emissions on a CO₂ equivalent basis. They
include emissions of CO₂ from air conditioning use and
emissions of HFCs from air conditioning system leaks. Technologies
exist which can reduce these emissions on the order of 40 to 75% (for
air conditioning efficiency improvements and HFC leakage control,
respectively), at an initial cost to the consumer of less than $110.
This initial cost would be more than offset by the reduced maintenance
and fuel savings due to the new technology over the life of the
vehicle. We also considered standards which would prevent future
increases in N₂O and methane.
    Based on our work in 2007 pursuant to Executive Order 13432, EPA
developed two different analytical approaches which could be pursued
under the CAA for establishing light-duty vehicle CO₂
standards. Both are attribute-based approaches, using vehicle footprint
(correlating roughly to vehicle size) as the attribute. Under either
approach, a CO₂-footprint continuous function curve is
defined that establishes different CO₂ emission targets for
each unique vehicle footprint. In general, the larger the vehicle
footprint, the higher (less stringent) the corresponding vehicle
CO₂ emission target will be. Each manufacturer would have a
different overall fleet average CO₂ emissions standard
depending on the distribution of footprint values for the vehicles it
sells. See Section VI.B.1.d and the Light-duty Vehicle TSD of this
Advance Notice for additional discussion of attribute-based standards
and other approaches (e.g., a non-attribute, or universal standard).
    One approach was based on a fixed percentage reduction per year in
CO2 emissions. We examined a 4% per year reduction in CO2 emissions,
reflecting the projected reductions envisioned by the President in his
20-in-10 plan in the 2007 State of the Union address and subsequent
legislative proposals . The other approach identified CO2 standards
which an engineering optimization model projects as resulting in
maximum net benefits for society (hereafter referred to as the ``model-
optimized'' approach). That approach uses a computer model developed by
the Department of Transportation Volpe Center called the CAFE Effects
and Compliance Model (the ``Volpe Model''). The Volpe Model was
designed by DOT as an analytical tool which could evaluate potential
changes in the stringency and structure of the CAFE program, and was
first used in DOT's 2006 rulemaking establishing CAFE standards for
model years 2008-2011 light-trucks.126 127
---------------------------------------------------------------------------

    \126\ See 66 FR 17566--Average Fuel Economy Standards for Light
Trucks Model Years 2008-2011.
    \127\ See ``CAFE Compliance and Effects Modeling System
Documentation, Draft, 1/26/07'' published by DOT, a copy of which is
available in the docket for this Advanced Notice.
---------------------------------------------------------------------------

    Using the fixed percentage reduction approach, projections
regarding technology feasibility, technology effectiveness, and lead-
time are critical as these are the most important factors in
determining whether and how the emission reductions required by a
future standard would be achieved. When using the model-optimized
approach, a larger set of inputs are critical, as each of these inputs
can have a significant impact in the model's projections as to the
future standard. These inputs include technology costs and
effectiveness, lead-time, appropriate discount rates, future fuel
prices, and the valuation of a number of externalities (e.g., criteria
air pollution improvements, GHG emission reductions, and energy
security). Although all of these factors are relevant under either
approach, there are major differences in the way this information is
used in each approach to develop and evaluate appropriate standards.
    EPA believes both of these approaches for establishing fleet-wide
average CO2 emissions standards are permissible, conceptually, under
section 202(a) of the Act. Section 202(a)(2) requires EPA to give
consideration to ``the cost of compliance'' for use of the technology
projected to be used to achieve the standards (``requisite
technology''). The model-optimized approach can be used in appropriate
circumstances to satisfy this requirement.\128\ The fixed percent per
year approach is broadly consistent with EPA's traditional means of
setting standards for mobile sources, which identifies levels of
emissions reductions that are technologically feasible at reasonable
cost with marginal emissions reduction benefits which may far outweigh
marginal program costs, without adverse impacts on safety and with
positive impacts on energy utilization, and which address a societal
need for reductions.\129\ Comparing and contrasting these approaches
with the model-optimized approach is one way to evaluate options for
appropriate standards under section 202(a). We request comment on these
approaches and whether one or the other is a more appropriate method
for EPA to consider future light-duty GHG standards under section 202
of the CAA. We also request comment on other potential approaches

[[Page 44444]]

EPA should consider, including the approach described in section
VI.B.1.a.
---------------------------------------------------------------------------

    \128\ See Husqvarna AB v. EPA, 254 F. 3d 195, 200 (D.C. Cir.
2001) (EPA reasonably chose not to use marginal cost-benefit
analysis to analyze standards [under the technology-forcing section
213 of the Act], where section 213 does not mandate a specific
method of cost analysis).
    \129\ See NRDC v. EPA, 655 F. 2d 318, 332-334 (D.C. Cir. 1981).
---------------------------------------------------------------------------

    During 2007, EPA, DOT's Volpe Center, and NHTSA expended a major
technical effort to make a series of significant enhancements to the
Volpe Model by reviewing and updating, where possible, many of the
critical inputs to the Model (e.g., cost reduction learning curves, the
number and estimated costs and effectiveness of potential CO2/mpg
control technologies), as well as making updates to the Model itself.
This technical work notably improved the Volpe Model. However, the
Volpe Model was designed specifically to analyze potential changes to
NHTSA's CAFE program, and there remained several aspects of the
analysis we conducted that did not reflect differences between EPA and
NHTSA statutory authorities, and we were not able to address these
aspects in 2007. As a result, our analysis tended to underestimate the
benefits and/or overestimate the costs of light-duty vehicle CO2
standards that could be established under the CAA. We discuss these
issues below.
    First, past NHTSA CAFE regulatory actions have generally had a
short-term focus (a 3-5 year timeframe), and NHTSA is currently
proposing more stringent CAFE standards for five model years, 2011-
2015, in keeping with its revised statutory authority, as discussed
above. In contrast, EPA's Title II authority permits EPA to set
standards over a significantly longer period of time as appropriate in
light of environmental goals, developing technologies, costs, and other
factors. A short-term focus can have a significant implication for the
technology assumptions which go into a standard-setting analysis.
    In our 2007 analysis, we assumed limited technology innovation
beyond what is known today, and did not include several commercially
available or promising technologies such as advanced lightweight
materials for all vehicle classes (several auto companies have recently
announced plans for large future reductions in vehicle weight), plug-in
hybrids, optimized ethanol vehicles, and electric vehicles. To the
extent such innovations penetrate the market over the next 10 years,
the societal benefits and/or decreased societal cost of CO2 standards
will be greater than what we projected. A short-term focus may yield a
more reliable short-term projection because it relies on available
technology and is less prone to uncertainties involved in projecting
technological developments and other variables over a longer term. The
trade-off is that such a focus may not stimulate the development of
advanced, low GHG-emitting technologies. For the auto industry,
significant technological advances have historically required many
years and large amounts of capital.
    Second, our 2007 analysis does not account for a series of
flexibilities that EPA may employ under the CAA to reduce compliance
costs, such as multi-year strategic planning, and credit trading and
banking. As mentioned previously, EPA has used many of these
flexibilities in its existing mobile source programs, and we would
attempt to include such flexibilities in any future EPA GHG standards
analysis.
    Third, under the CAA manufacturers traditionally choose to comply
instead of non-comply, since they cannot sell new vehicles unless they
receive a certificate of conformity from EPA that is based on a
demonstration of compliance. Under the penalty provisions of the CAA,
light-duty vehicle manufacturers may not pay a civil penalty or a fine
for non-compliance with the standards and still introduce their
vehicles into commerce. In our 2007 analysis, we assumed a number of
manufacturers would pay fees rather than comply with the analyzed
standards. This assumption resulted in a lower compliance cost
estimation and lower GHG benefits.
    Fourth, in our 2007 analysis, we did not reflect the difference in
carbon content between gasoline and diesel fuel. This difference has
not been germane to NHTSA's setting of CAFE standards, but it is
important to the GHG emissions reductions that different standards can
achieve. Therefore, our Light-duty Vehicle TSD analysis did not account
for the higher CO2 emissions which result from the use of a gallon of
diesel fuel compared to a gallon of gasoline (diesel fuel has a higher
carbon content than gasoline fuel), and we would address this issue in
any future EPA GHG standards analysis.
    As noted previously, our 2007 analysis relied upon the use of key
inputs concerning predictions of future technologies and fuel prices
and valuation of a number of externalities, such as the benefits of
climate change mitigation and improvements in energy security. The
information used for these key inputs can have a significant effect on
projections regarding the costs of a standard based on a fixed
percentage reduction or the level of a model-optimized standard. In the
analyses we present in this notice, we have generally taken an approach
similar to NHTSA's, although we have also used alternative values in
some cases to illustrate the impact from different, alternative values.
For example, to account for large uncertainties regarding the magnitude
of the marginal benefits of GHG emission reductions, we looked at
alternative approaches to valuing those benefits and developed a range
of values to capture the uncertainties. (See section III.G in this ANPR
for a discussion of GHG benefits issues and marginal benefits
estimates.)
    Another key, but uncertain, input is the future price of fuel.
Important for any analysis of fuel savings over a long time frame is an
adequate projection of future oil prices. Typically, EPA relies on
Annual Energy Outlook (AEO) forecasts made by the Energy Information
Agency. However, AEO forecasts in past decades have at times over-
predicted the price of oil, and more recently, with the rapid increase
in oil prices over the past several years, AEO forecasts have
consistently under-predicted near-term oil prices. In the Light-duty
Vehicle TSD analysis, we used the Energy Information Administration's
2007 AEO projections for future oil and fuel prices, which correspond
to a projected retail gasoline price of slightly more than $2 per
gallon in the 2010-2020 time period, while current gasoline fuel prices
are on the order of $3.50 to $3.80 per gallon or more. Since our
analyses are sensitive to the oil price used, this raised concerns
regarding the ability to accurately estimate fuel savings. In addition,
when using a model-optimized approach, this can have a significant
impact on the appropriate standard predicted by the model. For our
updated analysis (described in more detail below), however, we have
continued to use the AEO2007 forecasted fuel prices. The ``baseline''
for our Light-duty Vehicle TSD and updated analysis reflects
projections from the automotive manufacturers regarding future product
offerings which were developed by the manufacturers in late 2006
through the spring of 2007. The AEO2007 fuel price projections are more
representative of the fuel prices considered by the manufacturers when
they developed the baseline future product offerings used as an input
in the analysis.
    This approach has certain limitations. Given the large increases in
fuel price in the past year, most major automotive companies have since
announced major changes to their future product offerings, and these
changes are not represented in our analysis. However, the projection of
future product offerings (model mix and sales volume) is static in the
analysis we have performed, both for the baseline (projections with no
new standards) and in the control scenarios (projections

[[Page 44445]]

with the impact of new standards). Our analysis to date does not
account for a range of possible consumer and automaker responses to
higher fuel prices, higher vehicle prices and attribute-based standards
that could affect manufacturer market share, car/truck market share, or
vehicle model mix changes. EPA has initiated work with Resources for
the Future to develop a consumer choice economic model which may allow
us to examine the impact of consumer choice and varying fuel prices
when analyzing potential standard scenarios in the future, and to more
realistically estimate a future baseline. Higher fuel prices than those
predicted in AEO2007 can certainly have a large impact on the projected
costs and benefits of future light-duty GHG limits, and we will
continue to examine this issue as part of our on going work.
    We ask for comment on the relative importance of, and how best to
address, the various issues we have highlighted with our analysis of
potential light-duty vehicle GHG standards performed to date. In
particular, we seek comment on the feasibility and utility of
incorporating into the regulations themselves a mechanism for
correcting mistaken future projections or accomplishing the same
through a periodic review of the regulations.
    We now summarize the results from our 2007 analysis. Since 2007 we
have updated this analysis to address several of the issues noted
above, in order to evaluate the impact of these issues. EPA requests
comment on the two approaches we examined for setting standards, and
seeks input on alternative approaches, including the approach described
in section VI.B.1.a.
    In Table VI-1 we present weighted combined car and truck standards
we developed based on efforts to update the work we did in 2007 to
address some of the issues identified above. We show the results from
our 2007 analysis, as well as the updated results when we utilize the
same methodology for the 4% per year approach, but attempt to address a
number of the issues discussed above. As part of addressing these
issues, we have extended the time frame for our analysis to 2020, while
our Light-duty Vehicle TSD analysis was limited to 2018. Our updated
analysis results are documented in a separate technical memorandum
available in the public docket for this Advance Notice.\130\
---------------------------------------------------------------------------

    \130\ See EPA Technical Memorandum, ``Documentation of Updated
Light-duty Vehicle GHG Scenarios.''

 Table VI-1--Projected Vehicle CO2 (Gram/Mile Units) and MPG Standards (MPG Units in Square Brackets), Including
                                                 A/C CO2 Limits
----------------------------------------------------------------------------------------------------------------
                                                              Light-duty vehicle TSD analysis     Updated 2008
                                                           ------------------------------------     analysis
                           Year                                                                -----------------
                                                               4% per year     Model-Optimized     4% per year
----------------------------------------------------------------------------------------------------------------
2011......................................................        338 [26.3]        334 [26.6]        335 [26.5]
2012......................................................        323 [27.5]        317 [28.0]        321 [27.7]
2013......................................................        309 [28.8]        295 [30.1]        307 [28.9]
2014......................................................        296 [30.0]        287 [31.0]        293 [30.3]
2015......................................................        285 [31.2]        281 [31.6]        283 [31.4]
2016......................................................        274 [32.4]        275 [32.3]        272 [32.7]
2017......................................................        263 [33.8]        270 [32.9]        261 [34.0]
2018......................................................        253 [35.1]        266 [33.4]        251 [35.4]
2019......................................................               n/a               n/a        241 [36.9]
2020......................................................               n/a               n/a        232 [38.3]
----------------------------------------------------------------------------------------------------------------

    Compared to the Light-duty Vehicle TSD analysis, we have attempted
in the updated analysis to address for potential CAA purposes several,
but not all, of the noted issues, and as such we continue to believe
that the results of this analysis are conservative--that is, they tend
to overestimate the costs and/or underestimate the benefits. We have
included the following updates:

--Inclusion of plug-in hybrids as a viable technology beginning in
2012;
--Consideration of multi-year planning cycles available to
manufacturers;
--Consideration of CO₂ trading between car and truck fleets
within the same manufacturer;
--Assumption that all major manufacturers would comply with the
standards rather than paying a monetary penalty;
--Correction of the CO₂ reduction effectiveness for diesel
technology.

    Our updated analysis does not address all of the issues we
discussed previously. For example, we have not considered the
widespread use of lightweight materials, further improvements in the
CO₂ reduction effectiveness of existing technologies,
potential for cost reductions beyond our 2007 analysis, and the
potential for new technologies. We also have not addressed the
potential changes in vehicle market shifts that may occur in the future
in response to new standards, new consumer preferences, or the
potential for higher fuel prices. Recent trends in the U.S. auto
industry indicate there may be a major shift occurring in consumer
demand away from light-duty trucks and SUVs and towards smaller
passenger cars.\131\ Such potential trends are not captured in our
analysis and they could have a first-order impact on the results.
---------------------------------------------------------------------------

    \131\ See ``As Gas Costs Soar, Buyers Are Flocking to Small
Cars'', New York Times, May 2, 2008, page A1.
---------------------------------------------------------------------------

    Table VI-2 summarizes the most important societal and consumer
impacts of the standards we have analyzed.

[[Page 44446]]



      Table VI-2--Summary of Societal and Consumer Impacts From Potential Light-Duty Vehicle GHG Standards
                                          [2006 $s, AEO2007 oil prices]
----------------------------------------------------------------------------------------------------------------
                                            Light-duty vehicle TSD analysis *             Updated 2008 analysis
                                  ------------------------------------------------------------------------------
                                          4% per year              Model-Optimized             4% per year
----------------------------------------------------------------------------------------------------------------
                                                Societal Impacts
----------------------------------------------------------------------------------------------------------------
GHG Reductions (MMTCO2 equivalent  378......................  343.....................  635
 in 2040).
Fuel Savings (million bpd in       2.3......................  2.0.....................  4.2
 2040).
Net Societal Benefits in 2040      $54 + B..................  $54 + B.................  $130 + B
 (Billion $s) **.
Net Present Value of Net Benefits
 through 2040 (Billion $s): **
    3% DR........................  $320 + B.................  $390 + B................  $830 + B
    7% DR........................  $120 + B.................  $160 + B................  $340 + B
----------------------------------------------------------------------------------------------------------------
                                                Consumer Impacts
----------------------------------------------------------------------------------------------------------------
Per-Vehicle Costs:
    2015.........................  $736.....................  $672....................  $565
    2018.........................  $1,567...................  $995....................  $1,380
    2020.........................  n/a......................  n/a.....................  $1,924
Payback Period: ***
    3% DR........................  6.2 yr. (2018)...........  4.8 yr. (2018)..........  6.0 yrs. (2020)
    7% DR........................  8.9 yr. (2018)...........  6.0 yr. (2018)..........  8.7 yrs. (2020)
Lifetime Monetary Impact: ***
    3% DR........................  $2,753 (2018)............  $2,245 (2018)...........  $1,630 (2020)
    7% DR........................  $1,850 (2018)............  $1,508 (2018)...........  $437 (2020)
----------------------------------------------------------------------------------------------------------------
* The Light-duty Vehicle TSD Societal Impacts are based on new stds. for 2011-2018 for cars and 2012-2017 for
  trucks, while the updated analysis is based on new stds. for 2011-2020 for cars and trucks.
** The identified ``B'' = unquantified benefits, for example, we have not quantified the co-pollutant impacts
  (PM, ozone, and air toxics), and does not include a monetized value for the social cost of carbon. Societal
  benefits exclude all fuel taxes because they represent transfer payments. In addition, for the updated
  analysis, we have not included the increased costs nor the GHG emissions of electricity associated with the
  use of plug-in electric hybrid vehicles. We have also not quantified the costs and/or benefits associated with
  changes in consumer preferences for new vehicles.
*** The payback period and lifetime monetary impact values for Light-duty Vehicle TSD analysis is for the
  average 2018 vehicle, and 2020 for the updated analysis.

    Given the current uncertainty regarding the social cost of carbon,
Table VI-2 does not include a monetized value for the reduction in GHG
emissions. We present here a number of different values and indicate
what impact they would have on the net social benefits for our updated
analysis. Presentation of these values does not represent, and should
not be interpreted to represent, any determination by EPA as to what
the social cost of carbon should be for purposes of calculating
benefits pursuant to the Clean Air Act.
    We have analyzed the valuation for the social cost of carbon of $40
per metric ton (for emission changes in year 2007, in 2006 dollars,
grown at a rate of 3% per year) that reflects potential global,
including domestic, benefits of climate change mitigation. This
valuation (which is the mean value from a meta analysis of global
marginal benefits estimates for a 3% discount rate discussed in section
III.G. of this Advance Notice) would result in an increase in the 2040
monetized benefits for the 2008 updated analysis of $67 billion. Given
the nature of the investment in GHG reductions, we believe that values
associated with lower discount rates should also be considered. For
example, for a 2% discount rate for year 2007, the mean value from the
meta analysis is $68 per metric ton. This valuation would result in an
increase in the 2040 monetized benefits for the 2008 updated analysis
of $110 billion.
    As discussed in section III.G, another approach to developing a
value for the social cost of carbon is to consider only the domestic
benefits of climate change mitigation. The two approaches--use of
domestic or global estimates--are discussed in section III.G of this
notice. There is considerable uncertainty regarding the valuation of
the social cost of carbon, and in future analyses EPA would likely
utilize a range of values (see section III.G).\132\ Furthermore,
current estimates are incomplete and omit a number of impact categories
such that the IPCC has concluded that current estimates of the social
cost of carbon are very likely to underestimate the benefits of GHG
reductions.
---------------------------------------------------------------------------

    \132\ Ranges better reflect the available scientific information
and the uncertainties in marginal benefits estimates, and the fact
that there are estimates well above the means. The corresponding
ranges for the 2007 mean estimates discussed above are the
following: For the meta-analysis global marginal benefits estimates,
the range is $-4 to $106 per metric ton CO₂ based on a 3
percent discount rate, or $-3 to $159 per metric ton CO₂
based on a 2 percent discount rate. The preliminary domestic ranges
derived from a single model are $0 to $5 per metric ton
CO₂ based on a 3 percent discount rate, and $0 to $16 per
metric ton CO₂ based on a 2 percent discount rate.
---------------------------------------------------------------------------

    This Advance Notice asks for comment on the appropriate value or
range of values to use to quantify the benefits of GHG emission
reductions, including the use of a global value. While OMB Guidance
allows for consideration of international effects, it also suggests
that the Agency consider domestic benefits in regulatory analysis.
Section III.G.4 discusses very preliminary ranges for U.S. domestic
estimates with means of $1 and $4 per metric ton in 2007, depending on
the discount rate. These valuations ($1 and $4 per metric ton in 2007)
would result in an increase in the 2040 monetized benefits for the 2008
updated analysis of $1.7-6.7 billion. In its recent proposed
rulemaking, NHTSA utilized $7 per metric ton as the initial value for
U.S. CO₂ emissions in 2011.
    Table VI-2 shows the impact of addressing a number of the issues
noted

[[Page 44447]]

above. With respect to per-vehicle costs, the updated 4% per year
approach shows a $171 per vehicle lower cost in 2015 and a $187 per
vehicle lower cost in 2018 compared to our 2007 analysis, for a
slightly more stringent standard in both cases. This is primarily due
to the impact of including multi-year planning and car-truck trading
within a given manufacturer.
    The estimated CO₂ reductions in 2040 from the updated
analysis are much larger than the 2007 analysis (by nearly a factor of
2). This occurs primarily because we have addressed the diesel
CO₂ issue noted above, and because we have extended the time
frame for the analyzed standards to 2020. The estimated fuel savings
are also larger primarily due to the additional years we extended the
4% per year standard to. The estimated monetized net benefits for the
updated analysis are also significantly higher than our previous
estimates. This is a result of a combination of factors: lower
estimates for the increased vehicle costs due to multi-year planning
and within manufacturer car-truck trading; and the extension of the
analyzed standards to 2020.
    Table VI-2 also provides estimates of ``payback period'' and
``lifetime monetary impact''. The payback period is an estimate of how
long it will take for the purchaser of the average new vehicle to
break-even; that is, where the increased vehicle costs is off-set by
the fuel savings. Our updated analysis shows for the average 2020
vehicle that period of time ranges from 6.0 to 8.7 years (depending
upon the assumed discount rate). The lifetime monetary impact provides
an estimate of the costs to the consumer who owns a vehicle for the
vehicle's entire life. The lifetime monetary impact is simply the
difference between the higher initial vehicle cost increase and the
lifetime, discounted fuel savings. Our updated analysis indicates the
lifetime, discounted fuel savings will exceed the initial cost increase
substantially. As shown in the table, the positive lifetime monetary
impact ranges from about $440 to $1,630 per vehicle (depending upon the
assumed discount rate). Section VI.C.2 of the Light-duty Vehicle TSD
discusses possible explanations for why consumers do not necessarily
factor in these fuel savings in making car-buying decisions.
    Our updated analysis projects the 2020 CO₂ limit of 232
gram/mile (38.3 mpg) shown in Table VI-1, could be achieved with about
33% of the new vehicle fleet in 2020 using diesel engines and full
hybrid systems (including plug-in electric hybrid vehicles). Higher
penetrations of these and other advanced technologies (including for
example the wide-spread application of light-weight materials) could
result in a much greater GHG reductions.
    The results of our updated analysis indicate that:

    --Technology is readily available to achieve significant reductions
in light-duty vehicle GHG emissions between now and 2020 (and beyond);
    --The benefits of these new standards far outweigh their costs;
    --Owners of vehicles complying with the new standard will recoup
their increased vehicle costs within 6-9 years, and;
    --New standards would result in substantial reductions in GHGs.

    We request comment on all aspects of this analysis, the
appropriateness of the two approaches described, and the inputs and the
tools that we utilized in performing the assessment, when considering
the setting of light-duty vehicle GHG standards under the CAA. We also
request comment on the alternative approach for establishing light-duty
vehicle GHG standards described in section VI.B.1.a of this advance
notice.
c. Technologies Available To Reduce Light-Duty Vehicle GHGs
    In this section we discuss a range of technologies that can be used
to significantly reduce GHG emissions from cars and light trucks. We
discuss EPA's assessment of the availability of these technologies,
their readiness for introduction into the market, estimates of their
cost, and estimates of their GHG emission reduction potential. We
request comment on all aspects of our current assessment, including
supporting data regarding technology costs and effectiveness.
    In the past year EPA undertook a comprehensive review of
information in the literature regarding GHG-reducing technologies
available for cars and light trucks. In addition, we reviewed
confidential business information from the majority of the major
automotive companies, and we met with a large number of the automotive
companies as well as global automotive technology suppliers regarding
the costs and effectiveness of current and future GHG-reducing
technologies. EPA also worked with an internationally recognized
automotive technology firm to perform a detailed assessment of the GHG
reduction effectiveness of a number of advanced automotive
technologies.\133\
---------------------------------------------------------------------------

    \133\ See ``A Study of Potential Effectiveness of Carbon Dioxide
Reducing Vehicle Technologies'', Ricardo, Inc., EPA Report 420-R-08-
004a, June 2008.
---------------------------------------------------------------------------

    EPA recently published a Staff Technical Report describing the
results of our assessment, and we provided this report to the National
Academy of Sciences Committee on the Assessment of Technologies for
Improving Light-Duty Vehicle Fuel Economy.\134\ This Staff Technical
Report details our estimates of the costs and GHG reduction potential
of more than 40 technologies applicable to light-duty vehicles, and is
one of the key inputs to our analysis of potential future standards
presented in Section VI.B.1.b. These technologies span a large range of
effectiveness and technical availability, from technologies as simple
as reduced rolling resistance tires (offering a 1-2% reduction in
vehicle CO₂ emissions) to advanced powertrain systems like
gasoline and diesel hybrids, plug-in electric hybrids, and full
electric vehicles (offering up to a 100% reduction in vehicle
CO₂ emissions).
---------------------------------------------------------------------------

    \134\ See ``EPA Staff Technical Report: Cost and Effectiveness
Estimates of Technologies Used to Reduce Light-duty Vehicle Carbon
Dioxide Emissions'', EPA Report 420-R-08-008, March 2008.
---------------------------------------------------------------------------

    The majority of the technologies we investigated are in production
and available on vehicles today, either in the United States, Japan or
Europe. Over the past year, most of the major automotive companies or
suppliers have announced the introduction of new technologies to the
U.S. market. The following are some recent examples:

--Ford's new ``EcoBoost'' turbocharged, down-sized direct-injection
gasoline engines;
--Honda's new 2009 global gasoline hybrid and 2009 advanced diesel
powertrain;
--Toyota and General Motors plans for gasoline plug-in hybrid systems
within the next two to three years;
--General Motors breakthroughs in lower-cost advanced diesel engines;
--Nissan's 2010 introduction of a clean diesel passenger car;
--Chrysler's widespread use of dual-clutch automated manual
transmissions beginning in 2009; and,
--Mercedes' new product offerings for clean diesel applications as well
as diesel-electric hybrid technologies.

    We also evaluated the costs and potential GHG emissions reductions
from some of the advanced systems not currently in production or that
are only available in specialty niche vehicles, such as gasoline
homogeneous charge compression ignition engines, camless valve
actuation systems, hydraulic hybrid powertrains, and full electric

[[Page 44448]]

vehicles. These technologies are described in detail, along with our
estimates for costs and GHG reduction potential, in our Staff Technical
Report.
    An additional area where we see opportunities for significant
CO₂ emissions reduction is in material weight substitution.
The substitution of traditional vehicle materials (e.g., steel, glass)
with lighter materials (e.g., aluminum, plastic composites) can provide
substantial reductions in CO₂ emissions while maintaining or
enhancing vehicle size, comfort, and safety attributes. Several
companies have recently announced plans to utilize weight reduction as
a means to improve vehicle efficiency while meeting all applicable
safety standards.\135\ We request data and comment on the extent to
which material substitution should be considered as a means to reduce
GHG emissions, and information on the costs and potential scope of
material substitution over the next 5 to 20 years.
---------------------------------------------------------------------------

    \135\ See Automotive News, February 11, 2008, in which Daimler-
Benz CEO states that Mercedes-Benz will reduce the weight of all new
vehicle models by 5%, and Ford announces every model will lose
between 250 and 750 pounds.
---------------------------------------------------------------------------

    Finally, we note that in the past 30 years there has been a steady,
nearly linear increase in the performance of cars and light trucks. We
estimate that the average new vehicle sold in 2007 had a 0-60 miles/
hour acceleration time of 9.6 seconds--compared to 14.1 seconds in
1975.\136\ If this historic trend continues, by 2020 the average 0-60
acceleration for the combined new car and truck fleet will be less than
8 seconds. During the past 20 years, this increase in acceleration has
been accompanied by a gradual increase in vehicle weight. It is
generally accepted that over the past 20 years, while fuel economy for
the light-duty fleet has changed very little, the fuel efficiency has
in fact improved but has largely been used to enable increases in both
the weight and the performance of vehicles. We request comment on how
we should consider the potential for future changes in vehicle weight
and performance (e.g., acceleration time) in assessing the costs and
benefits of standards for reducing GHG emissions.
---------------------------------------------------------------------------

    \136\ See ``Light-Duty Automotive Technology and Fuel Economy
Trends: 1995-2007'', EPA Report EPA420-R-07-008, September 2007.
---------------------------------------------------------------------------

d. Potential Options for Reducing HFCs, N₂O, CH₄,
and Air Conditioning-Related CO₂
    As described above, in addition to fleet average and in-use
CO₂ standards, EPA has analyzed how new control measures
might be developed for other car and light truck emissions that have
global warming impacts: air conditioning (``A/C'')-related emissions of
HFCs and CO₂, and tailpipe emissions of nitrous oxide
(N₂O), and methane (CH₄). Under CAA section
202(a), EPA may regulate these emissions if a positive endangerment
finding is made for the relevant GHGs. Together, these emissions
account for about 10% of greenhouse gases from light-duty cars and
trucks (on a CO₂ equivalent basis). The direct HFC emissions
account for 4.3%, while the A/C CO₂ emissions are 3.1%.
N₂O and CH₄ account for 2.7% and 0.2%
respectively. With regard to air conditioning-related emissions,
significant opportunity exists to reduce HFC emissions from refrigerant
leakage and CO₂ from A/C induced engine loads, and EPA has
considered potential standards to reduce these emissions. In addition,
EPA has considered potential limits for N₂O and
CH₄ emissions that could apply to both cars and light trucks
that would limit future growth of these emissions.
i. Potential Controls for Air Conditioning-Related GHG Emissions
    Over 95% of the new cars and light trucks in the U.S. are equipped
with A/C systems. There are two mechanisms by which A/C systems
contribute to the emissions of GHGs. The first is through direct
leakage of the refrigerant (currently the HFC compound R134a) into the
air. Based on the higher GWP of HFCs, a small leakage of the
refrigerant has a greater global warming impact than a similar amount
of emissions of other mobile source GHGs. Leakage can occur slowly
through seals, gaskets, hose permeation and even small failures in the
containment of the refrigerant, or more quickly through rapid component
deterioration, vehicle accidents or during maintenance and end-of-life
vehicle scrappage (especially when refrigerant capture and recycling
programs are less efficient). The leakage emissions can be reduced
through the choice of leak-tight, durable components, or the global
warming impact of leakage emissions can be addressed through the
implementation of an alternative refrigerant. 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 already addressed by the CAA Title VI stratospheric ozone
protection program, as described in section VIII of this notice.\137\
---------------------------------------------------------------------------

    \137\ The second mechanism by which vehicle A/C systems
contribute to GHG emissions is through the consumption of excess
fuel when the A/C system is running, and from carrying around the
weight of the A/C system hardware all-year round. This excess fuel
required to run the system is converted into CO₂ by the
engine during combustion. This excess CO₂ from A/C
operation can thus be reduced by increasing the efficiency of the
overall vehicle-A/C system.
---------------------------------------------------------------------------

    EPA's analysis indicates that together, these A/C-related emissions
account for about 7.5% of the GHG emissions from cars and light trucks.
EPA considered standards designed to reduce direct leakage emissions by
75% and to reduce the incremental increase of A/C related
CO₂ emissions by 40% in model year 2015 vehicles, phasing in
starting in model year 2012. It is appropriate to separate the
discussion of these two categories of A/C-related emissions because of
the fundamental differences in the emission mechanisms and the methods
of emission control. Refrigerant leakage control is akin in many
respects to past EPA fuel evaporation control programs in that
containment of a fluid is the key control feature, while efficiency
improvements are more similar to the vehicle-based control of
CO₂ in that they would be achieved through specific hardware
and controls.
    The Memo to the Docket, ``Light-Duty Vehicle Hydrofluorocarbon,
Nitrous Oxide, Methane, and Air Conditioning-Related Carbon Dioxide
Emissions'' provides a more detailed discussion of the air
conditioning-related GHG emissions, both refrigerant leakage and
CO₂ emissions from A/C use, as well as potential test
procedure and compliance approaches that have been considered by EPA.
ii. Feasibility of Potential A/C Reduction Approaches
    EPA believes that significant reductions in A/C HFC leakage and A/C
CO₂ emissions would be readily technically feasible and
highly cost effective. The types of technologies and methods that
manufacturers could use to reduce both types of A/C emissions are
commercially available and used today in many models of U.S. cars and
light trucks. For example, materials and components that reduce leakage
as well as electronic monitoring systems have been used on various
vehicles in recent years. Regarding A/C CO₂ reduction, such
technologies as variable-displacement compressors and their controls
are also in use today. Although manufacturers might find that more
advanced technologies, like alternate refrigerants, become economically
attractive in the coming years, EPA believes that currently available
technologies and systems designs would

[[Page 44449]]

be sufficient to meet potential limits being assessed by EPA.
iii. Potential Impacts of Requiring Improved A/C Systems
(1) Emission Reductions for Improved A/C Systems
    Manufacturers producing cars and light trucks for the U.S. market
have not historically had economic or regulatory incentives or
requirements to reduce refrigerant leakage and CO₂ from A/C
systems. As a result, there is an opportunity for significant
reductions in both of these types of emissions. With potential
standards like the ones considered above, EPA has estimated that
reductions in HFC refrigerant leakage, converted to CO₂
equivalent emissions, and added to projected A/C CO₂
reductions, these limits would result in an average per-vehicle
reduction in CO₂-equivalent emissions of about 4.7%
(excluding CH₄ and N₂O from the baseline). This
reduction is equivalent to about 7.5% of light vehicle CO₂-
equivalent emissions, or about 2 tons per year.
(2) Potential Costs for Improved A/C Systems
    Although the technologies and system designs EPA expects could be
used to comply with the two A/C related standards being considered are
currently available, not all manufacturers are using them on all
vehicles. Thus, the industry would necessarily incur some costs to
apply these technologies more broadly across the car and truck fleet.
EPA estimates that the cost of meeting the full HFC leakage standard it
is considering would average about $40 per vehicle (retail price
equivalent or RPE) and that the cost of meeting the A/C CO₂
standard would be about $70 per vehicle (RPE). At the same time,
complying with such limits would result in very significant savings in
fuel costs (as system efficiency improves) and in A/C-related
maintenance costs (as more durable systems result in less frequent
repairs). In fact, EPA's analysis shows that these cost savings would
significantly exceed projected retail costs of the potential A/C
standards, more than offsetting the costs of both types of A/C system
improvements.\138\
---------------------------------------------------------------------------

    \138\ See Appendix 3.B. of the EPA Technical Memorandum
``Documentation of Updated Light-duty Vehicle GHG Scenarios'' for a
detailed discussion of these costs estimates.
---------------------------------------------------------------------------

iv. Potential Interaction With Title VI Refrigerant Regulations
    As described further in Section VIII of this notice, Title VI of
the CAA deals with the protection of stratospheric ozone. Section 608
of the Act establishes a comprehensive program to limit emissions of
certain ozone-depleting substances (ODS) from appliances and
refrigeration. 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 the
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.
    Another Title VI provision that could interact with potential Title
II motor vehicle regulation of GHGs is section 612, which requires EPA
to review substitutes for ozone depleting substances and to consider
whether such substitutes would 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 the potential program analyzed here as complementing
these Title VI programs, and not conflicting with them. The potential
standards would apply at pre-production when manufacturers demonstrate
that they are utilizing requisite equipment (or utilizing other means
designated in the potential program) to achieve the suggested 75% leak
reduction requirement. These requirements 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 a program is
that there could be fewer and less impactive maintenance events for
MVACs, since there would be less leakage. In addition, although the
suggested standards would also apply in-use, the means of enforcement
should not conflict (or overlap) with the Title VI section 609
standards. EPA also believes the menu of leak control technologies
described above would complement the section 612 requirements because
these control technologies would help ensure that 134a (or other
refrigerants) would be used in a manner that would further minimize
potential adverse effects on human health and the environment.
v. Potential Controls for Nitrous Oxide Emissions
    Nitrous oxide, or N₂O, is emitted from gasoline and
diesel car and light truck tailpipes and is generated during specific
catalyst warm-up temperature conditions conducive to N₂O
formation. While N₂O emissions from current Tier 2 vehicles
with conventional three-way catalysts are relatively low on a mass
basis (e.g., around 0.005 g/mi), N₂O does have a high GWP of
310. 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 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 has considered a ``cap'' approach to controlling N₂O
emissions would not require any new technology for current Tier 2
gasoline vehicles, but would limit any increases in N₂O
emissions that might otherwise occur with future technology vehicles.
Such an approach would have minimal feasibility, emissions, or cost
impacts.
    The Memo to the Docket, ``Light-Duty Vehicle Hydrofluorocarbon,
Nitrous Oxide, Methane, and Air Conditioning-Related Carbon Dioxide
Emissions'' has more in-depth discussion of car and light truck
N₂O emissions, as well as of potential test procedure and
compliance

[[Page 44450]]

approaches that have been considered by EPA.
vi. Potential Controls for Methane Emissions
    Methane, or CH₄, is emitted from gasoline and diesel car
and light truck tailpipes and is one of the family of hydrocarbon
compounds generated in the engine as a by-product of gasoline and
diesel fuel combustion. As such, levels of CH₄ emissions
have been somewhat controlled by the lower hydrocarbon emissions
standards that have been phased in since the early 1970s. Current
CH₄ emissions from Tier 2 gasoline vehicles are relatively
low (about 0.017 g/mi on average), and CH₄ has a global
warming potential of 23. The one technology where much higher
CH₄ emissions could be of concern would be natural gas-
fueled vehicles, since CH₄ is the primary constituent of
natural gas fuel and would be the largest component of unburned fuel
emissions.
    As with N₂O, EPA has considered a ``cap'' CH₄
emissions standard approach that would not require any new technology
for current Tier 2 gasoline vehicles, but would limit any increases in
CH₄ emissions that might otherwise occur with future natural
gas vehicles. Such an approach would have no significant feasibility,
emissions, or cost impacts.
    The Memo to the Docket, ``Light-Duty Vehicle Hydrofluorocarbon,
Nitrous Oxide, Methane, and Air Conditioning-Related Carbon Dioxide
Emissions'' has greater discussion of car and light truck
CH₄ emissions.
e. Specific Programmatic Design Issues
    As discussed above, Title II of the CAA provides the Agency with
both direction and flexibility in designing and implementing a GHG
control program. Consistent with existing motor vehicle programs, the
Agency would need to develop appropriate mechanisms to address issues
such as certification of new motor vehicles to applicable standards,
ensuring the emissions requirements are being met throughout the
designated useful life of the vehicle, and appropriate compliance
mechanisms if the requirements are not being met. Domestic and imported
vehicles and engines subject to emissions standards must obtain a
certificate of conformity in order to be sold in the U.S. marketplace.
EPA has utilized a wide range of program design tools and compliance
mechanisms to help address the large variation of market participants
yet still provide a level regulatory playing field for these parties.
As part of the design effort for a GHG program, it would be appropriate
to take into account these flexibilities as well as existing
requirements that the automobile and engine industries already face in
order to help reduce compliance costs if possible while still
maintaining our overall environmental objectives. However, given the
nature of GHG control, it would also be appropriate to determine if new
design structures and compliance measures might be more effective.
    The Light-duty Vehicle TSD includes a discussion of a wide range of
programmatic and technical issues and presents potential approaches
that would address these issues in the design of a comprehensive near-
term light-duty vehicle GHG control program. We highlight here a few of
these issues, and point the reader to the Light-duty Vehicle TSD for
additional detail. Among the issues discussed in the Light-duty Vehicle
TSD are several which could differ significantly under a different
approach. EPA specifically requests comment on these issues:

--Potential classification approaches for light-duty vehicles (e.g.,
treating cars and light trucks in a single averaging class or separate,
and the potential classification of vehicle types as either a passenger
car or a light truck);
--How any classification approaches would relate to NHTSA's regulatory
approach;
--The significant flexibilities allowed under Title II which we utilize
for existing criteria pollutant standards for light-duty vehicles,
including detailed concepts for a GHG averaging, banking, and trading
program;
--Potential light-duty GHG compliance program concepts.

    As we have considered various potential light-duty vehicle GHG
approaches, significant thought and stakeholder outreach went into
designing a potential system for determining compliance that would meet
Agency and industry needs and goals. The Light-duty Vehicle TSD
presents a compliance structure for vehicle GHG control that adheres to
CAA requirements and at the same time is compatible with the existing
CAFE program. However, this is not the only approach to compliance, as
is discussed in the Light-duty Vehicle TSD. Other compliance approaches
could also be considered, each with their own advantages. For example,
a GHG compliance program patterned after the Tier 2 light duty vehicles
emissions program offers an approach that is more similar to the
existing compliance structure for other pollutants.
    We discuss below in detail three specific issues regarding
potential future light-duty vehicle GHG programmatic issues: universal
and attribute-based standards; environmental backstop standards; and
tailpipe CO₂ test cycles.
i. Universal and Attribute-Based Vehicle GHG Standard Approaches
    A specific programmatic issue that EPA would like to highlight here
is the use of attribute-based standards for vehicle GHG standards, and
the concept of an environmental backstop to accompany an attribute-
based standard promulgated under the CAA, in order to assure that GHG
emission reductions which are feasible at reasonable cost under section
202(a) are not foregone. A CAA program for reducing GHG emissions from
light vehicles could set the average emissions standards for
manufacturers in one of two fundamental ways. A ``universal'' GHG
standard would apply a single numerical requirement to each
manufacturer, to be met on average across its entire light-duty vehicle
production. One potential consequence of the universal approach is that
the costs of compliance may fall unevenly on different manufacturers.
That is, complying with a single standard would be more difficult for
companies with current product mixes weighted relatively heavily toward
vehicles with higher compliance costs.
    The other approach EPA has considered would set individual
standards for each manufacturer, based on one or more vehicle
attributes (such as the footprint attribute approach currently used by
NHTSA). Thus, to the extent a manufacturer produced vehicles with
different attributes from the vehicles of another manufacturer; unique
standards would be set for each company. The Light-duty Vehicle TSD
discusses various vehicle attributes on which light duty vehicle
CO₂ standards could be based. EPA requests comment on the
use of an attribute-based approach, and on each of the attributes
considered in the Light-duty Vehicle TSD, as well as on a universal
standard approach. In addition, some in the industry have suggested
power-to-weight ratio may be an appropriate attribute for this purpose,
and we request comment on that attribute as well.
    A key characteristic of any attribute-based program is that
significant industry shifts in the attribute over time would increase
or decrease the average emission performance requirement for the fleet.
For example, if such a shift in attributes resulted in the unique
manufacturer standards being on

[[Page 44451]]

average less stringent than those determined to be feasible and cost-
effective in the establishment of the program, the program would fall
short of those overall emissions reductions, and conversely, market
shifts could also result in larger emissions reductions than those
determined to be feasible and cost-effective at the time the program
was established. EPA seeks comment on the universal approach as
compared to the attribute-based approach.
ii. Concepts for Light-Duty Vehicle GHG Environmental Backstops
    In order to limit the potential loss of feasible emissions control
due to a change in market attributes, EPA could consider a supplemental
``backstop'' carbon dioxide emissions standard for each year (also
referred to as an ``anti-backsliding'' provision) as a complement under
the CAA to an attribute-based standard. This would be an additional
obligation for manufacturers that would limit the maximum fleet average
carbon dioxide emissions, independent of attributes. The backstop
requirement could establish fixed minimum and feasible fleet average
CO₂ g/mile standards. The backstop would apply separately to
the domestic car, import car, and truck classes. This backstop
obligation may not apply to small volume manufacturers. While EPA will
quantitatively describe one specific backstop concept below, we are
seeking public comment on a range of alternative approaches described
qualitatively below, briefly, as well. More generally, EPA seeks
comment as to whether a backstop approach would be appropriate under
the CAA as a means of providing greater emission reduction certainty.
    A backstop could be an appropriate complement under the CAA to an
attribute-based standard. The most important factor under section
202(a) of the Act is to ensure reductions of the emissions from the
motor vehicle sector which cause or contribute to the endangerment
caused by greenhouse gas emissions. As discussed earlier, one important
feature of an attribute-based program is that collective decisions by
consumers and manufacturers could result in higher or lower industry-
wide average footprint values than projected by EPA at the time of
promulgation. Since the attribute-based curve establishes a fleet
average for a manufacturer based on the manufacturer's sales and
attribute values, the actual reductions achieved by the program could
vary as this mix varies. In the extreme, if the entire industry moved
to much higher attribute values, then the carbon dioxide emissions
reductions could be significantly less than projected by EPA as
technically feasible and cost effective.
    Under section 202(a), EPA could consider a supplemental fleet
average backstop standard that would be the same for every manufacturer
in a given year. Such a standard would ensure that a minimum level of
reductions would be achieved as the fleet mix changes over time. EPA
could base such a standard on feasible carbon dioxide emission
reductions and other important factors such as technological
feasibility, cost, energy, and safety in analyzing section 202(a)
standards. EPA recognizes that a CO₂ emissions backstop
could partially reduce the flexibility and market elements of an
attribute-based approach, but believes it could be needed to provide
for an appropriate degree of emissions reduction certainty.
    As with other structural issues such as universal versus attribute-
based approaches, EPA believes that various backstop approaches have
conceptual advantages and disadvantages with respect to relevant
criteria such as certainty of industry-wide carbon dioxide emissions
reductions, flexibility with respect to consumer choice and vehicle
offerings, varying treatment of automakers, and complexity of
explanation and implementation. Any approach would also need to address
the relevant factors, including cost (economic feasibility, cost
effectiveness, and per vehicle cost) and technological feasibility. EPA
encourages commenters to evaluate the design approaches presented
below, as well as to suggest alternative approaches, in terms of these
and other relevant criteria.
    As an illustrative example, Table VI-3 shows one set of fleet
average carbon dioxide emissions and mpg backstops, along with the
projected, average industry-wide carbon dioxide emissions and mpg
compliance levels, for the two sets of fleet average carbon dioxide
emissions standards based on the footprint attribute, analyzed in
December 2007, and discussed earlier in this advance notice: The 4% per
year and model-optimized scenarios. These carbon dioxide emissions
backstops are based on the projected fleet average carbon dioxide
emissions compliance levels for the high-volume car and light truck
manufacturers with the highest projected car and light truck footprint
levels, based on the footprint curves that were developed by EPA in
December 2007. Chrysler is the high-volume car manufacturer with the
highest projected footprint values, and General Motors has the highest
projected footprint values among the high-volume truck manufacturers.
    These backstops would be universally applied to every manufacturer,
except small volume manufacturers, and would become the effective fleet
average standard for any automaker that would otherwise have a higher
fleet average carbon dioxide emissions standard, for any of the three
respective averaging sets (import and domestic cars and trucks), based
on the footprint curve.
    The underlying rationale for this backstop approach is that the
manufacturer that is projected to sell the highest footprint vehicles,
which therefore is projected to be able to comply with the highest
fleet average carbon dioxide emissions compliance levels, should be
treated as establishing the minimum acceptable level of emissions
reductions for the industry. Similarly, no other manufacturers should
exceed the feasible, cost effective level established by that projected
highest footprint manufacturer. The approach, and underlying rationale,
is similar to the approach used by NHTSA before the 2006 truck
standards, whereby the level of a universal standard was established
based on the capabilities of the least capable large manufacturer
(Public Citizen v. NHTSA, 848 F. 2d 256, 259, D.C. Cir. 1988). Although
the backstop would not prohibit the highest footprint manufacturer from
selling higher footprint vehicles, it would prohibit any carbon dioxide
emissions ``backsliding'' that would otherwise be associated with that
increase in footprint. Average carbon dioxide emissions from other
manufacturers could increase, of course, in accordance with the
footprint curve, but in no case could the carbon dioxide emissions
level for any manufacturer increase beyond these backstop levels.
    The passenger car carbon dioxide emissions and mpg backstop levels
shown in Table VI-3 adhere to the methodology described above with one
exception. Based on Chrysler's projected footprint values, its 2011
standard for the 4% per year option would be 325 g/mi, equivalent to a
gasoline vehicle fuel economy of 27.3 mpg. Since the current car CAFE
standard, which acts as an effective fuel economy backstop, is 27.5
mpg, EPA could instead consider a 2011 backstop of 323 g/mi for the 4%
per year option, which is equivalent to a 27.5 mpg gasoline vehicle.
    In this illustrative backstop example, the carbon dioxide emissions
backstop levels would range from 8 to 22 g/mi, or 2 to 8%, higher than
the projected, average industry-wide carbon dioxide levels.

[[Page 44452]]



 Table VI-3--Illustrative Backstops for the Fleet Average Carbon Dioxide Emissions Standard (CO2 grams per mile/
                                                      mpg)
----------------------------------------------------------------------------------------------------------------
                                                                             CARS
                                             -------------------------------------------------------------------
                                                  4 percent per year option          Model-optimized option
                                             -------------------------------------------------------------------
                                                 Projected                         Projected
                                               industry-wide       Backstop      industry-wide       Backstop
                                                 CO2 levels                        CO2 levels
----------------------------------------------------------------------------------------------------------------
2010 (base).................................       (323)/27.5  ...............       (323)/27.5  ...............
2011........................................         309/28.7         323/27.5         301/29.5         317/28.0
2012........................................         298/29.8         319/27.8         291/30.5         314/28.3
2013........................................         285/31.1         296/30.0         276/32.1         287/30.9
2014........................................         275/32.3         287/30.9         268/33.2         281/31.6
2015........................................         264/33.6         277/32.0         260/34.1         273/32.5
2016........................................         254/34.9         266/33.4         247/35.9         258/34.4
2017........................................         244/36.3         257/34.5         244/36.4         257/34.5
2018........................................         235/37.7         245/36.2         239/37.2         249/35.7
----------------------------------------------------------------------------------------------------------------

    A second illustrative example of a universal backstop approach
could be modeled on the ``minimum standard'' in the Energy Independence
and Security Act (EISA) of 2007. EISA establishes a fuel economy
backstop for the domestic car class that is equal to 92% of the average
fuel economy level projected for all cars. EPA believes this 92% value
was derived by dividing the current car CAFE standard of 27.5 mpg by
the average industry-wide car fuel economy performance over the past
several years. The car CAFE standard, in effect, has served as a
backstop for those manufacturers that have chosen not to pay CAFE
penalties. Applying this model to a carbon dioxide emissions backstop
would involve dividing the average projected industry-wide carbon
dioxide emissions levels by 0.92, or multiplying by a factor of 1.087,
an increase of 8.7%, to generate a universal backstop level that would
apply to all manufacturers. Under this approach, the backstop levels
for the 4% per year and model-optimized standards in Table VI-3 would
be greater than the backstop levels discussed earlier in every case,
ranging from 3 to 23 g/mi higher. This alternative approach yields
backstop levels 20 to 31 g/mi higher than the projected, average
industry-wide standards.
    For the backstop approaches discussed above, all automakers would
have the same uniform backstop for domestic and import cars, and a
higher uniform backstop for trucks. These universal approaches would
make the backstop more of a constraint on those manufacturers that sold
vehicles with higher average footprint levels and less of a constraint
on those automakers that sold vehicles with lower average footprint
levels.
    An alternative backstop approach could be to establish unique
maximum numerical carbon dioxide emissions values that would apply to
different automakers (e.g., X g/mi for Automaker A, and Y g/mi for
Automaker B) and that would become the effective fleet average standard
for an individual automaker when that automaker would otherwise be
allowed to meet a higher fleetwide average carbon dioxide emissions
value based exclusively on the footprint curve. The rationale for this
type of approach would be that since manufacturers start at different
average footprint levels, manufacturer-specific backstop values could
provide greater insurance against carbon dioxide emissions backsliding
for all manufacturers, rather than just those manufacturers that sold
vehicles with higher average footprint levels. One illustrative example
of this type of approach would be to base the annual backstop for each
manufacturer on its 2010 carbon dioxide emissions baseline, reducing it
by the same percentage each year. A similar approach would base the
annual backstop for the highest-footprint manufacturer on its 2010
carbon dioxide emissions baseline reduced by a percentage each year,
the annual backstop for the lowest-footprint automaker on its 2010
carbon dioxide emissions baseline reduced by a lesser percentage per
year, and the annual backstop values for other manufacturers on annual
percentage reductions between the higher and lower percentages. This
latter approach would yield backstop values that would be somewhat more
binding on manufacturers that sold vehicles with higher average
footprint values, yet still binding to some degree on all automakers.
This approach would also limit the degree to which manufacturers that
sold vehicles with lower average footprint values could increase
average footprint values over time.
    A combination of the universal and manufacturer-specific approaches
could be to begin with manufacturer-specific backstop values, and to
transition to uniform backstop values over a 5 or 10 year period.
    Another alternative backstop approach would not set a maximum
numerical carbon dioxide emissions value for individual manufacturers,
but would establish mathematical functions that would automatically
increase the stringency of and/or ``flatten'' the footprint curves for
future years when actual industry-wide carbon dioxide emissions
performance in the future is found to fall short of EPA's projections
at the time of promulgation. For example, at the time of promulgation,
EPA could assume a certain average industry-wide carbon dioxide g/mi
emissions level for 2011-2012. If, in 2013, EPA found that the average
industry-wide emissions level in 2011-2012 was higher than projected in
the final rule (and therefore the carbon dioxide emissions reductions
were lower than projected because of higher than projected average
footprint levels), then the backstop provisions would be triggered and
the footprint curves for future years (say, 2016 and later) would be
automatically changed to be more stringent and/or flatter in shape.
This approach would reframe the backstop issue in terms of industry-
wide emissions performance, rather than in terms of individual
automaker emissions performance.
    In lieu of a backstop, another approach would be to flatten (i.e.,
reduce the slope of) the carbon dioxide emissions-footprint curve such
that there would a major disincentive for automakers to increase
vehicle footprint. EPA invites comments on the pros and cons of this
approach relative to a backstop.


[[Continued on page 44453]]
Notices For
2009
2008
2007
2006
2005
2004
2003
2002
2001
2000
1999
1998
1997
1996
1995
1994
Regulating Greenhouse Gas Emissions Under the Clean Air Act

Local Navigation
