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 (18 pp, 313K, About PDF)

[Federal Register: July 30, 2008 (Volume 73, Number 147)]
[Proposed Rules]
[Page 44503-44520]
From the Federal Register Online via GPO Access [wais.access.gpo.gov]
[DOCID:fr30jy08-39]

Regulating Greenhouse Gas Emissions Under the Clean Air Act

[[Continued from page 44502]]

[[Page 44503]]

cap-and-trade system alone. In addition, the potential would exist for
PSD to interfere with the efficient operation of the GHG cap-and-trade
program. Although PSD would neither reduce nor increase the overall
emission reductions achieved under the cap, it would force different
choices about the stringency and location of controls than if control
choices were based solely on market factors. Under this scenario, the
result would be to increase costs without achieving additional GHG
emissions reductions. For example, assume that a company undertakes a
change that triggers PSD at a location where controls are expensive to
retrofit but are required as BACT for that location. Without PSD, the
company could have increased emissions and still complied with the cap
by purchasing less expensive emissions reductions from another source,
and the same total GHG emissions reductions would have been achieved.
Notably, for GHGs, which have relatively uniform global concentrations,
the location of GHG emissions does not matter to global climate
impacts, so the policy reasons for the spatial component of PSD control
requirement would not apply to GHG controls.
    PSD program requirements also affect numerous CAA programs that
require stationary source controls that may increase emissions of
pollutants other than the pollutant targeted for control (i.e.
``collateral increases''), such as the increased NO_X
emissions that result when a thermal oxidizer is installed to control
VOC. Because there is no exemption from PSD requirements for such
pollution control projects, the collateral increase must be reviewed,
which can result in added costs and delay of those pollution control
projects. Regulation of GHGs would exacerbate these concerns because
the energy demands of many controls for criteria pollutants, HAP, and
other pollutants have the potential to result in increased
CO₂ emissions.
6. What Are Some Possible Tailoring Approaches to Address
Administrative Concerns for GHG NSR?
    The cost and potential broad applicability of PSD requirements
raises questions about whether GHG regulation through PSD would be more
effective in minimizing GHG increases if it operates as a broad program
targeting numerous smaller sources and modifications, or as a narrow
program targeting smaller numbers of large sources and modifications.
We ask for comment on how these cost/benefit considerations for
permitting small sources and modifications under PSD, as well as any
other factors, should be considered in EPA's deliberations regarding
the major source cutoffs and significance levels for GHGs as well as
EPA's available legal authority in this area.
    EPA believes that whether or not PSD is workable for GHGs may
depend on our ability to craft the program to deal with the unique
issues posed by GHG regulation.
    This section discusses several options, including:
     Reducing the potential universe of sources based on
``potential to emit'' approaches;
     Increasing the major source thresholds and significance
levels for GHGs, to permanently restrict the program to larger sources;
     Phasing in the applicability of PSD for GHGs;
     Developing streamlined approaches to implementing the BACT
requirement; and
     Issuing general permits for numerous similar sources.

The options are not necessarily exclusive. Many are complementary, and
we note that some combination of these options may be most effective.
We also ask for suggestions on additional tailoring options not
described below, and more generally on which options, if any, present
an appropriately balanced means of addressing the administrative
concerns.
    Before discussing each option in detail, we present an overarching
legal discussion that lays out possible rationales for such
flexibility. For at least one of the options identified (e.g., the
option of adopting higher major source sizes than those contained in
the Act), the principal legal constraint is the ``plain meaning'' of
the applicable PSD provisions, such as the major source levels.
Nonetheless, we have identified two legal doctrines that may provide
EPA with discretion to tailor the PSD program to GHGs: Absurd results
and administrative necessity.
    The Supreme Court has stated that the plain meaning of legislation
is not conclusive ``in the `rare cases [in which]
the literal
application of a statute will produce a result demonstrably at odds
with the intentions of the drafters' * * * [in which case]
the
intention of the drafters, rather than the strict language, controls.''
U.S. v. Ron Pair Enterprises, Inc., 489 U.S. 235, 242 (1989). To
determine whether ``the intentions of the drafters'' differs from the
result produced from ``literal application'' of the statutory
provisions in question, the courts may examine whether there is a
related statutory provision that conflicts, whether there is
legislative history of the provisions in question that exposes what the
legislature meant by those terms, and whether a literal application of
the provisions produces a result that the courts characterize variously
as absurd, futile, strange, or indeterminate. See, e.g., id., Nixon v.
Missouri Municipal League, 541 U.S. 125 (2004); United States v.
American Trucking Association, Inc. 310 U.S. 534 (1940); Rector of Holy
Trinity Church v. U.S., 143 U.S. 457 (1892).
    Further, the administrative burdens that would result for the
federal and state permitting authorities, as well as the sources, from
a literal application of the PSD provisions give rise to consideration
of whether EPA can craft relief from a strict interpretation based on
the judicial doctrine of administrative necessity. In Alabama Power,
the D.C. Circuit addressed various instances of claimed administrative
burdens resulting from the application of the PSD statutory provisions
and efforts by EPA to provide regulatory relief. Alabama Power Co. v.
Costle, 636 F.2d at 357-60 (D.C. Cir. 1980). In a section of its
opinion titled ``Exemptions Born of Administrative Necessity,'' the
Court stated,

    Certain limited grounds for the creation of exemptions are
inherent in the administrative process, and their unavailability
under a statutory scheme should not be presumed, save in the face of
the most unambiguous demonstration of congressional intent to
foreclose them.

Id. at 357. The Court identified several types of administrative
relief. One is ``[c]ategorical exemptions from the clear commands of a
regulatory statute,'' which the court stated are ``sometimes
permitted,'' but emphasized that they ``are not favored.'' Id. at 358.
A second is ``an administrative approach not explicitly provided in the
statute,'' such as ``streamlined agency approaches or procedures where
the conventional course, typically case-by-case determinations, would,
as a practical matter, prevent the agency from carrying out the mission
assigned to it by Congress.'' Id. A third is a delay of deadlines upon
`` `a showing by [the agency]
that publication of some of the
guidelines by that date is infeasible.' '' Id. at 359 (quoting NRDC v.
Train, 510 F.2d 692, 712 (D.C. Cir. 1974). The Court indicated it would
evaluate these choices based on the ``administrative need to adjust to
available resources * * * where the constraint was imposed * * * by a
shortage of funds * * *, by a shortage of time, or of the

[[Page 44504]]

technical personnel needed to administer a program.'' Id. at 358.
a. Potential-to-Emit: Reducing the Number of Sources Potentially
Covered
    Applicability of PSD is based in part on a source's ``potential to
emit'' or PTE. The PTE concept also is used for applicability of
nonattainment NSR, Title V, and the air toxics requirements of section
112. We discuss PTE in detail here, but the issues and questions we
discuss in this section apply equally to these other programs. As noted
above, PTE is defined as the maximum capacity of a source to emit any
air pollutant under its physical and operational design. In the case of
sources that are not operating for part of the year, the PTE for many
types of sources counts the emissions that would be possible if those
sources did emit year round.
    EPA believes that an important threshold question is how to
interpret ``maximum capacity * * * to emit * * * under its physical and
operational design'' for commercial and residential buildings, and
other types of source categories that might be subject to PSD and Title
V solely due to GHG emissions. For example, in the case of a furnace at
a residence, is it appropriate, in calculating the furnace's PTE, to
assume that a homeowner would set the thermostat at a level that would
require the furnace to operate continuously throughout the year? Even
on a cold winter day, a furnace typically turns on and off throughout
the day, and as the weather warms, the number of operating hours
decreases until the weather warms to the point where the furnace is not
needed at all and is shut off for an extended time.
    The EPA has in a few instances provided guidance on PTE calculation
methodologies to account for category-specific considerations. For
example, we issued technical guidance for calculating PTE from grain
elevators that took into account inherent limitations on the amount of
grain that could be handled due to the fact that grain is only
available for handling during a relatively short harvest period, and is
further limited by the amount of grain capable of being grown (as
represented by a record crop year adjusted for future increases in crop
yield) on the land that would ever reasonably be served by the
elevator.\273\ We ask for comment on whether, for smaller GHG sources
like these, there could be appropriate methodologies for defining PTE
in ways that consider these common-sense limitations on a source's
operation, but still reflect the maximum capacity to emit of a source.
---------------------------------------------------------------------------

    \273\ Calculating Potential to Emit (PTE) and Other Guidance for
Grain Handling Facilities: November 14, 1995 memorandum from John S.
Seitz, Director, U.S. EPA Office of Air Quality Planning and
Standards, to EPA Regional Offices.
---------------------------------------------------------------------------

    Sources with PTE exceeding the major source threshold can become
minor sources by taking legally and practically enforceable limits on
their PTE, by, for example, agreeing to operate only part of the year,
or only so many hours per day, or by employing control devices.\274\
Many sources are able to avoid classification as ``major'' by taking
such limits.
---------------------------------------------------------------------------

    \274\ Current regulatory language allows consideration of such
limits in calculating PTE only if they are federally enforceable,
but this definition was vacated or remanded in three separate
cases--one for PSD/NSR (Chemical Manufacturers Assn v. EPA, No. 89-
1514 (D.C. Cir. Sept. 15, 1995), one for Title V (Clean Air
Implementation Project v. EPA, No. 96-1224 (D.C. Cir. June 28,
1996), and one for section 112 (National Mining Association v. EPA,
59 F. 3d 1351 (D.C. Cir. 1995). EPA is developing a rule to respond
to these cases and in the meantime is following a transition policy
that does not require federal enforceability.
---------------------------------------------------------------------------

    The estimates provided for potential new permits for GHG sources
outlined in section VII.D.2 above are based on actual emissions. Were
they based on PTE, and if year-round operation were assumed to
represent PTE for all source categories, the estimates would likely be
an order of magnitude higher (in the absence of actions to limit the
scope of the programs). This emphasizes the significance of the
interpretation of ``potential to emit'' for buildings and other
categories not traditionally subject to PSD, as well as the importance
of streamlined mechanisms for obtaining limits on PTE.
    For traditional PSD and Title V permitting, the PTE limit is
typically a source specific limit that is crafted in a facility's minor
source permit and tailored to the source's individual circumstances. If
it were necessary to create PTE limits for very large numbers of GHG-
emitting sources nationwide, this would certainly require a more
efficient approach than creating them through individual minor source
permits. Not only would the sheer volume of permits and the process
required for each one severely strain permitting authority resources,
but some state and local agencies may lack the authority to establish
minor source permit limits for non-NAAQS pollutants. In addition, while
sources may not seek PTE limits for PSD until they have planned
modifications that could otherwise trigger PSD, sources may seek PTE
limits for Title V purposes as soon as the program is effective,
meaning that the approach would need to deal with a large number of
sources at essentially the same time.
    We ask for comment on whether we should also therefore consider
streamlined regulatory approaches for creating the legally and
practically enforceable limits sources need without requiring a huge
number of individual minor source permits. A possible mechanism could
involve adopting a regulation that sets forth operational restrictions
that limit PTE for a broad class of sources. We may wish to consider
adopting--or encouraging state permitting authorities to adopt--rules
for numerous categories where we expect there to be large numbers of
sources whose actual emissions are not major but who have major PTE
(unless addressed through interpreting maximum capacity as described
above). Such a rule could, for example, limit a source's natural gas
usage to 1700 MM BTU (17,000 therms) per year, which would keep it
below the 100 tpy cutoff for Title V.\275\ Typically, the rule would
also build in some operating margin so that the limit is not right at
the major source cutoff. The rule would have to include recordkeeping
and reporting, which would be simple here since fuel use is metered.
This approach may be a streamlined effective way to limit PTE for many
sources with fuel combustion equipment, provided they can agree to
comply with the limits in the rule, even in an abnormally long, cold
winter. We ask for comment on stakeholders' experience with limiting
PTE by rule rather than through individual permits, possible
considerations in tailoring this approach to GHG sources, and
identification of categories that might benefit from the use of rules
limiting PTE.
---------------------------------------------------------------------------

    \275\ Although the PSD cutoff may in some cases be 250 tpy,
sources will generally adopt PTE limits below 100 tpy to avoid both
PSD and Title V applicability where they have the option to do so.
For this reason, this example uses a 100 tpy cutoff, though in some
cases PTE limits are taken to stay below a 250 tpy cutoff.
---------------------------------------------------------------------------

    Finally, where the establishment of a rule-based PTE limit for an
entire source category is not recommended or is infeasible, the EPA
requests comment on whether general permitting approaches might be
useful. A general permit is a permit that the permitting authority
drafts one time, and then applies essentially identically (except for
some source-specific identifying information) to each source of the
appropriate type that requests coverage under the general permit.
Similar to the type of rules limiting PTE described above, a general
permit could also limit PTE by setting out the operational restrictions
(e.g., fuel combusted per

[[Page 44505]]

year) necessary to assure the GHG emissions stay below major source
thresholds, and would also spell out records the source would have to
keep to assure it met these restrictions. To be most useful, the permit
would need to address large numbers of similar sources. This approach
may also work well for many types of GHG sources as well. We request
comment on the use of a general permit approach to limiting PTE, and
whether it would offer additional benefit over the approach of
establishing operational restrictions directly by rule.
b. Options for Setting Higher GHG Major Source Cutoffs and Significance
Levels
    If the EPA ultimately determines that subjecting numerous small
sources and modifications to PSD is not an effective way to address GHG
emissions, one possible option for tailoring the program would be to
raise the major source cutoffs (e.g., raise the threshold only for GHGs
as a class, or perhaps only for certain individual GHGs) and establish
a significance level for GHGs at a level high enough to assure that the
program applies to larger sources and modifications, but excludes
smaller sources and modifications. Since the existing major source
thresholds are set forth in the CAA itself, EPA would need to find the
legal flexibility to raise these thresholds above 250 and 100 tons per
year. We present for discussion below several policy and legal options
for higher major source cutoffs and significance levels.
i. Higher GHG major source cutoffs--possible approaches and legal basis
    Regardless of how PTE is calculated, the major source size
threshold will be a critical consideration in tailoring the PSD program
for GHGs. There are a number of factors one might consider in choosing
an appropriate cutoff for GHGs and whether to establish the cutoff for
individual gases such as CO₂ or for GHGs as a class. One
conceptual approach might be to identify the number of sources and
modifications affected by various cutoffs, calculate the costs and
benefits of a PSD program for that universe of affected sources, and
select a cutoff that optimizes the benefit-cost ratio. Unfortunately,
we presently have the ability to quantify in dollar terms only a subset
of the climate impacts identified by the IPCC. Also, we have very
limited data on the number of sources expected at various major source
cutoffs, and even more limited data on the number of modifications at
various significance levels. More importantly, it is very difficult to
project the future number of permits or the incremental impact of any
additional GHG reductions that would result from the control technology
decisions therein. For these reasons, EPA cannot quantitatively
determine an optimal major source size or significance level.
    We could, however, consider other means of setting levels. One
example is an emissions scaling approach. This approach would compare
the emissions of other existing NSR pollutants for sources that are
major and would calculate the corresponding GHG emissions that the same
source would emit. This would be an appropriate approach if the goal
were to tailor PSD applicability for GHGs to cover a similar universe
of source sizes and types to the universe now regulated for other
pollutants. A second option would be to base the major source size on a
scientific determination of a level below which an individual source
would have a de minimis contribution to any particular adverse climate-
related impact on a relevant health, societal, or environmental
endpoint. Although it may be possible to generally estimate such a
level, we are not currently aware of any scientific literature that
establishes a specific numeric threshold below which GHG emissions are
de minimis, either in terms of their impact on climate, or on these
endpoints. By the same token, aside from an ability to use currently
available models to project temperature effects, the Agency does not
have the ability to project specific climatic impacts or endpoints
resulting from individual sources. Alternatively, we could potentially
choose a GHG major source size that is selected to harmonize with GHG
cutoffs from other regulatory programs. For example, the DOE's 1605(b)
program has a threshold of 10,000 metric tons of CO₂-
equivalent, California's AB32 regulation for mandatory reporting of
GHGs has a threshold of 25,000 metric tons of CO₂-
equivalent, and the Wisconsin emission inventory reporting requirements
has a CO₂ threshold of 100,000 short tons. Notably, these
examples are thresholds for reporting requirements only. PSD would
involve much more than simply reporting emissions, so under a
harmonizing approach we may need to evaluate whether it is feasible to
require not only reporting, but also the other PSD elements for the
sources that would be covered. We ask for comment on the range of
approaches EPA could take in selecting a major source cutoff if we
decide it is appropriate under existing legal authority, if available,
to develop a higher cutoff for GHGs. In addition, we request data that
may be useful for conducting necessary analysis to support such
approaches.
    A related issue to the establishment of the major source thresholds
and significance levels for GHGs is the selection of the metric against
which these levels are evaluated. Emissions of GHGs are typically
expressed in a common metric, usually the metric called CO₂-
equivalent, although the measure known as Carbon Equivalent (CE) is
also used. The use of either metric allows the impact of emissions of
different GHGs to be directly compared, as some gases have a higher
global warming potential or GWP than others. Since both units are
measured in weight--usually tons--either could be used for purposes of
PSD applicability. The use of either metric has the advantage of
linking emissions of a GHG directly to its ability to impact climate,
appropriately regulating more potent GHGs more stringently. The use of
CO₂-equivalent would solve the problem of leaving unreviewed
significant GHG emissions of some chemicals, such as
hydrofluorocarbons, but it would leave many small CO₂
sources with less climate impact still subject to PSD. However, the use
of Carbon Equivalent (CE) addresses both concerns. The attached table
demonstrates the possible effect of using CE in making PSD
applicability decisions:

------------------------------------------------------------------------
                                                 Emissions equal to 250
                                         GWP            tons CE
------------------------------------------------------------------------
Carbon dioxide (CO2).................       1  917 tons.
Methane (CH4)........................      21  44 tons.
Nitrous oxide (N2O)..................     310  3 tons.
Hydrofluorocarbon (HFC)-134a.........    1300  1410 lbs.
------------------------------------------------------------------------

    As the table shows, it would take more CO₂ emissions to
reach the major source size for CE. However. it would take
substantially less of several other GHGs. Such an approach would likely
result in fewer sources being added to the PSD program for GHGs in
total. While more sources for several GHGs would be considered major,
the major source population is, as noted above, dominated by
CO₂, and there would be fewer sources classified as major
due to CO₂ emissions. This approach arguably would regulate
significant sources of potent GHG while also reducing the burden on
relatively small sources of CO₂, focusing efforts on the
sources with the most important climate impacts. EPA seeks comments on
the potential use of the CE measure as the means to determine PSD
applicability. Specifically we ask for comment on the appropriateness
of the metric (considering that CO₂, rather than

[[Page 44506]]

carbon, is the air pollutant), data regarding its effect on PSD
applicability, and views concerning whether such an approach fits
within the language of the CAA.
    Whether, and the extent to which, EPA has flexibility to limit the
application of the PSD permitting requirements (and, by extension, the
nonattainment NSR permitting requirements if a NAAQS is set for GHGs)
to sources that emit larger amounts of CO₂ and other GHGs
than the 100/250 tpy thresholds depends on the interpretation of the
key PSD definitional term, ``major emitting facility.'' Under CAA
section 165(a), the basic PSD applicability requirement is that a
``major emitting facility'' may not construct unless it has received a
permit that covers specified requirements.\276\ As defined by CAA
section 169(1), a ``major emitting facility'' is defined to include (i)
``any * * * stationary source[]'' that emits or has the potential to
emit 100 tpy or more of any air pollutant and that falls into one of 28
specified industrial source categories; and (ii) ``any other source
with the potential to emit 250 tons per year or more of any air
pollutant.'' However, the last sentence of this definition allows
states to exempt ``new or modified facilities which are nonprofit
health or educational institutions'' from the PSD program. EPA's
regulations, promulgated in 1980 and revised several times since then,
make clear that emissions count toward the 100/250 tpy thresholds only
if they are ``regulated NSR pollutant[s]'' (e.g., 40 CFR
52.21(b)(1)(i)(a)), the specific meaning of which is discussed
elsewhere in this notice.
---------------------------------------------------------------------------

    \276\ The requirement to obtain a permit applies to a source
that commences construction after the effective date of the 1977
Clean Air Act Amendments (August 7, 1977), and that does so ``in any
area to which [the PSD provisions] appl[y].'' All parts of the
United States and its possessions are covered (see CAA sections 161,
302(d) and (q), and 110(a)(1)), but if EPA promulgates a NAAQS for
GHGs and designates certain areas as nonattainment, then those areas
would not be covered.
---------------------------------------------------------------------------

    Once GHGs are regulated, these PSD provisions, by their terms,
would apply to sweep into the PSD program new sources that emit 100 or
250 tpy of CO₂ or other GHGs. As indicated above, the courts
have held that the plain meaning of statutory provisions is generally
controlling. Even so, we solicit comment on whether these PSD threshold
requirements may present one of those rare cases in which congressional
intent differs, based on the legislative history.
    The legislative history indicates that Congress was aware of the
range of stationary sources that emitted pollution and did not envision
that PSD would cover the large numbers of smaller sources within that
inventory. As the D.C. Circuit stated in Alabama Power, the seminal
court decision regarding PSD that reviewed numerous challenges to EPA's
initial set of PSD regulations,

    Congress's intention was to identify facilities which, due to
their size, are financially able to bear the substantial regulatory
costs imposed by the PSD provisions and which, as a group, are
primarily responsible for emissions of the deleterious pollutants
that befoul our nation's air.

636 F.2d. 323, 353 (D.C. Cir. 1980) (emphasis added). In addition,
Congress also sought to protect permitting authorities from undue
administrative burdens. See S. Rep. 95-127 at 97; Alabama Power, 636
F.2d at 354.
    One important indication that Congress viewed PSD as limited in
scope may be found in information provided by EPA in 1976 and included
in the Congressional Record: A comprehensive list of industrial and
commercial source categories, which included the amounts of certain
pollutants emitted by ``typical'' sources in those categories and the
number of new plants in those categories constructed each year. 122
Cong. Rec. S 24548-50 (July 29, 1976) (statement of Sen. McClure). The
pollutants included particulate matter (PM), sulfur dioxide
(SO₂), carbon monoxide (CO), and hydrocarbons. The two
largest of these source categories consisted of--
     Small boilers, those that generate between 10 MMbtu/hr and
250 MMbtu/hr. EPA estimated that 1,446 new plants with boilers of this
size were, at that time, constructed each year, and that the amount of
PM emissions with controls from a ``typical'' such boiler were 53 tpy.
     Very small ``boilers,'' those that generate between 0.3
MMBtu/hr and 10 MMBtu/hr. EPA estimated that 11,215 new plants with
boilers of this size were, at that time, constructed each year, and
that the amount PM emissions with controls would be 2 tpy.
    The D.C. Circuit indicated, in Alabama Power, that Congress did not
believe sources with boilers of these small sizes should be covered by
PSD: ``[With respect to] the heating plant operating in a large high
school or in a small community college * * * [w]e have no reason to
believe that Congress intended to define such obviously minor sources
as `major' for the purposes of the PSD provision.'' \277\ 636 F.2d at
354. To support this proposition, the Court cited a statement in the
Congressional Record by Sen. Bartlett arguing that the PSD provisions
should not cover ``[s]chool buildings, shopping malls, and similar-
sized facilities with heating plants of 250 million BTUs.'' Id. at 354
(citing 122 Cong. Rec. S. 12775, 12812 (statement of Sen. Bartlett)).
Yet, boilers of even this small size could well emit at least 250 tpy
of CO₂ and therefore could fall into PSD permitting
requirements if the definition of ``major emitting facility'' is read
to include emitters of CO₂ of that size or more.
---------------------------------------------------------------------------

    \277\ Although Congress specifically authorized the States to
exempt ``nonprofit health or education institutions'' from the
definition of ``major emitting facility'' this statement by the D.C.
Circuit should be taken as the Court's view that Congress did not
design PSD to cover sources of the small size described.
---------------------------------------------------------------------------

    Thus, it is clear that Congress's construct of PSD--specifically,
the 100/250 tpy thresholds--was based on Congress's focus on
conventional pollutants at that time and its understanding that sources
emitting conventional pollutants above those levels should be subject
to PSD, with its attendant cost burdens, both because such sources have
the financial resources and because they have the responsibility to
reduce their large share of the convention pollution problems. Limited
administrative resources were also part of this equation. But the
equation is scrambled when CO₂ is the pollutant because many
smaller sources, with limited resources, and whose share of the GHG
emissions problem is no greater than their share of the conventional
pollution problem, get swept into PSD at those threshold levels.
Further, administrative resources become greatly stretched. Juxtaposing
the limited scope of the universe of PSD sources that Congress had in
mind against the broad terms that Congress used in defining ``major
emitting facility,'' which determines PSD applicability, raises the
question of whether a narrower interpretation of those terms may be
permissible under various judicial doctrines.
    We solicit comment on whether the case law cited above, concerning
narrowing the application of statutory provisions in light of other
indications of congressional intent or in light of administrative
necessity, support interpreting the term, ``major emitting facility''
in a manner that is narrower than the literal meaning of the phrase,
``any other source'' in the case of sources that emit amounts of
CO₂ that are more than 250 tpy but less than the levels
discussed above.

[[Page 44507]]

ii. Modifications: Options and Legal Basis for Higher GHG Significance
Levels
    Regarding the selection of a significance level for GHG emissions,
we could follow a de minimis approach, as we have done in setting the
existing PSD significance levels. We could base the significance level
on the level below which an individual modification has a de minimis
contribution to climate change. A scaling approach similar to that
discussed above for the major source threshold is also an option for
setting the significance level. We could set the significance level to
a level of GHG emissions that corresponds to the same activity level as
the significance levels for other pollutants, so as to roughly maintain
the same permitting burden for GHGs as for ``traditional'' pollutants.
We ask for comment on the merits of these approaches and invite
suggestions on other approaches. We are also interested in specific
information that would help us analyze how the selection of various
significance levels would affect the number and types of modifications
affected.
    The legal rationale for establishing a significance level is found
in the D.C. Circuit's Alabama Power decision, 636 F.2d at 405, where
the Court authorized EPA to establish ``a de minimis standard
rationally designed to alleviate severe administrative burdens.'' The
Court elaborated:

    A rational approach would consider the administrative burden
with respect to each statutory context: what level of emission is de
minimis for modification, what level de minimis for application of
BACT. Concerning the application of BACT, a rational approach would
consider whether the de minimis threshold should vary depending on
the specific pollutant and the danger posed by increases in its
emission. The Agency should look at the degree of administrative
burden posed by enforcement at various de minimis threshold levels.*
* * It may * * * be relevant * * * that Congress made a judgment in
the Act that new facilities emitting less than 100 or 250 tons per
year are not sizeable enough to warrant PSD review.

    Id. (emphasis added). We believe that this approach entails broad
discretion in fashioning a de minimis level, consistent with the
overarching principle of obviating administrative burdens that are not
commensurate with the contribution of the amount of emissions to the
pollution problem. We consider the Court's emphasized statement to
leave the door open to setting significance levels at the same level as
the applicability threshold levels. We solicit comment on appropriate
GHG significance levels, and on the relationship of significance levels
to the GHG applicability thresholds discussed above.
c. Phase-In of PSD Permitting Requirements
    Absent higher major source cutoffs and significance levels, it
would be necessary to formulate a strategy for dealing with the tenfold
increase in required permits that EPA projects permitting authorities
will experience if GHGs become regulated for PSD purposes. Even with
advance notice, an increase of this magnitude over a very short time
could overwhelm permitting authorities. They would likely need to fund
and hire new permit writers, and staff would need to develop expertise
necessary to identify sources, review permits, assess control
technology options for a new group of pollutants (and for a mix of
familiar and unfamiliar source categories), and carry out the various
procedural requirements necessary to issue permits. Sources would also
face transition issues. Many new source owners and operators would need
to become familiar with the PSD regulations, control technology
options, and procedural requirements for many different types of
equipment. If the transition were not effectively managed, an
overwhelmed permit system would not be able to keep up with the demand
for new pre-construction permits, and construction could be delayed on
a large number of projects under this scenario.
    The size of the increase in workload that must be accommodated and
the potentially serious consequences of an overly abrupt transition
demonstrate that a phase-in approach may have merit. Under one concept
of a phase-in approach, EPA could phase-in PSD applicability beginning
with the largest sources of GHGs and gradually include smaller sources.
This could be accomplished by initially adopting a relatively high
major source size and significance level, and then periodically
lowering the level until the full coverage level is reached. We ask for
comment on what an appropriate transition time would be, what the
appropriate starting, middle, and end points would be in terms of
coverage, and what requirements, if any, should be put into place for
sources prior to their being phased in. For example, if the ultimate
goal is to reach a 250 tpy major source cutoff, what would be the
appropriate starting cutoff (e.g., 10,000 tpy) and how should it be
determined? Would the phase-in need to be complete by a certain date,
and if so how long should the phase-in take? Alternatively, could the
phase-in of the smaller sources proceed by setting up periodic EPA
evaluations of the administrative necessity for deferring applicability
for such sources, and applying PSD only after we determine that it is
feasible to do so? We also ask for comment on what activities occurring
over this time we should consider in structuring a phase-in.
    As noted elsewhere, in its broad review of the initial PSD program
promulgated under the 1977 Clean Air Act Amendments, the D.C. Circuit
set out a range of mechanisms through which an agency can, at least
under ``limited'' circumstances, provide relief on grounds of
``administrative necessity'' from even clear statutory mandates, as
long as those mandates do not unambiguously foreclose such relief.
Alabama Power, 636 F.2d at 357. The Court noted that an agency could
establish the need for such relief based on ``a shortage of funds[,] *
* * time, or * * * technical personnel.'' Id. at 358.
    As described above, the large number of sources that would become
subject to the PSD requirements at the 100/250 tpy levels would strain
the administrative resources of the State permitting authorities and
perhaps also of the EPA regional offices that issue PSD permits. Each
of the constraints noted by the Court in Alabama Power--funds, time,
and technical personnel--would arise.
    Elsewhere in this notice, we solicit comment on whether
``administrative necessity'' authorizes EPA to exempt categories of
smaller GHG emitters. Here, we solicit comment on phasing-in the
applicability of the permit program over a multi-year period, with
successively smaller sources becoming subject. This method could allow
an orderly ramp-up in funding and in essential human capital. Under
such an approach, we also seek comment on whether it would be necessary
to set a firm schedule for phase-in, or whether it is sufficient for
the agency to select a future date to assess the level of program
coverage and the associated administrative burden, and determine at
that time whether it is appropriate to add them to the program, and if
not, to set an additional future date to revisit the issue. We request
information that would help us determine the appropriate timeframe for
such assessments, including the current and anticipated state resources
for processing PSD permits, including numbers of permitting personnel,
and the time period and person-hours needed to issue a typical permit.
d. Streamlining Determinations of Required Controls
    As previously noted, one of the most significant aspects of the PSD
program

[[Page 44508]]

for GHGs is the BACT requirement. While permitting authorities are
accustomed to making BACT determinations on a case-by-case basis for
major sources and modifications under the current PSD program, BACT for
GHGs (particularly CO2) presents significant additional permitting
challenges. The primary challenge is the dramatic increase in the
number of sources and modifications that under the 100/250-ton
thresholds would be subject to BACT review and the new source
categories that would be brought into the PSD program, which could
exceed the capacity of the permitting system and have negative effects
described above in section VII.D.4. An additional challenge stems from
the fact that for some GHG-emitting activities, primarily CO2 from
combustion sources, permitting authorities will need to look at
alternative approaches to determining BACT such as setting efficiency
targets, if add-on controls are not viewed as adequately demonstrated.
While there is much information available on efficiency for some of the
various kinds of equipment used by these newly applicable sources,
permit engineers will need to understand this information for a very
wide range of source categories.
    This section seeks comment on approaches for streamlining the BACT
process for many new smaller sources that could be brought into the PSD
program based on their GHG emissions. Under PSD, BACT is a case-by-case
decision that reflects the state-of-the-art demonstrated control
technology at the time of the permit action. Thus, BACT changes over
time and requires continual updating. Determining BACT is also a
decision that affords permitting authorities flexibility to consider a
range of case-specific factors such as cost, energy, and environmental
impacts. However, full case-by-case consideration of those factors
requires significant data and analysis in order for permitting
authorities to arrive at a permitting decision that is appropriate for
each individual source or modification
    EPA is interested in whether there would be ways to move from a PSD
permit system in which BACT limits are set on an individual case-by-
case basis to a system in which BACT determinations could be made for
common types of equipment and sources, and those determinations could
be applied to individual permits with little to no additional tailoring
or analysis. EPA has previously introduced this concept, known as
``presumptive BACT,'' as an aid to streamlining permitting for
desulfurization projects at refineries as well as in other
instances,\278\ and some state permitting authorities have adopted
similar approaches in their air permitting programs.\279\ Based on our
understanding of the types of sources that will become subject to PSD
if GHGs are regulated with a major source size of 250 tpy of emissions,
we believe the presumptive BACT process could offer significant
streamlining benefits. These benefits arise because many of these
smaller sources will likely have very similar emissions producing
equipment, and there will be little variation across sources with
respect to the cost, energy, and environmental considerations in the
BACT decision.
---------------------------------------------------------------------------

    \278\ See January 19, 2001 memo from John S. Seitz, Director,
Office of Air Quality Planning and Standards to the Regional Air
Division Directors entitled, ``BACT and LAER for Emissions of
Nitrogen Oxides and Volatile Organic Compounds at Tier 2/Gasoline
Sulfur Refinery Projects.''
    \279\ For example, Wyoming has a minor source permitting program
that includes a BACT analysis, and they use a presumptive BACT
process for issuing minor source permits to a particular source
category--oil and gas production facilities. See Permitting Guidance
for Oil and Gas Production Facilities, Wyoming Dept. of
Environmental Quality, Air Quality Division (August 2007 revision).
---------------------------------------------------------------------------

    While the CAA states that PSD permits shall be issued with BACT
determinations made for each pollutant on a ``case-by-case basis,'' the
court in Alabama Power recognized that exceptions may be appropriate
where ``case-by-case determinations, would, as a practical matter,
prevent the agency from carrying out the mission assigned to it by
Congress.'' 636 F.2d at 358 (emphasis added). The court recognized that
such streamlining measures may be needed when time or personnel
constraints or other practical considerations ``would make it
impossible for the agency to carry out its mandate.'' See id. at 359.
Given the more-than-tenfold increase in new sources that would likely
be brought into the PSD program once GHGs are regulated and the other
challenges described above, maintaining a traditional PSD permitting
program with individual case-by-case BACT determinations may be
impractical, warranting streamlined regulatory approaches as allowed
under the Act. A presumptive BACT permitting program would allow EPA,
state and local permitting authorities to carry out the PSD program in
a timely and efficient manner necessary to promote (rather than hinder)
control of GHG emissions from the many new, small source categories
that would be required to have PSD permits based on their GHG
emissions, while still preserving opportunities for public
participation.
    In considering a change from case-by-case BACT determinations to a
presumptive BACT process for some specific source categories within the
PSD program, EPA is considering how such presumptive BACT limits should
be established and used, and what provisions in the CAA would set
requirements or limits on their establishment and use. In particular,
EPA recognizes the statutory requirement to set BACT limits on a case-
by-case basis after taking into account site-specific energy, economic,
and environmental impacts (otherwise known as collateral impacts). One
option would be to allow permitting authorities to adjust any BACT
limit that was based on presumptive BACT, as necessary, upon
identifying significant collateral impacts applicable to a specific
source. EPA also recognizes the requirement to subject proposed PSD
permits, and the BACT limits contained within them, to public notice
and comment before such permits become final. A presumptive BACT
program could be designed to establish presumptive emissions limits for
a particular category of sources through guidance that would be issued
only after public notice and comment procedures. Another approach could
be to allow presumptive BACT limits in each permit to become final only
if public comments fail to establish that significant case-specific
energy, economic, and/or environmental impacts require adjustment of
the presumed limit for that particular source.
    In addition, while case-by-case BACT determinations allow for the
continual evolution of BACT requirements over time (as controls applied
in prior permits are considered in each subsequent case-by-case BACT
determination), EPA recognizes that application of presumptive BACT to
a category of sources over many permitting decisions may somewhat
diminish PSD's incentives for improved technology. EPA is interested in
options that would help maintain advances in control technologies, such
as a requirement to update and/or strengthen the presumptive BACT at
set intervals (such as after 3 years). EPA seeks comment on all aspects
of the use of presumptive BACT limits within the PSD program, including
EPA's authority to do so, whether there is need for and value to such
an approach, and suggestions for how such limits could be established,
updated, and used consistent with the requirements of the CAA.

[[Page 44509]]

    The central component of a presumptive BACT approach would be the
recurring technical determination, subject to notice and comment, of
the presumptive BACT levels for various categories. Because of the
limited data we currently have about the number and types of sources
that would become subject to the BACT requirement for GHGs, we cannot
at this time predict how many or which source categories might benefit
from such an approach if we opt to pursue it. We seek comment on the
basis we could use in setting the presumptive BACT level. Considerable
work will be needed to determine what options exist for controlling GHG
emissions from these categories of smaller sources and the various
emitting equipment they use. Even if a determination is made that add-
on controls for CO₂ from combustion sources are adequately
demonstrated, it is unlikely that the application of these controls
would be cost-effective at these small sources in the relatively near
future. Thus the focus of presumptive BACT for CO₂ would
likely be on energy efficiency standards for the installed equipment.
    While PSD permitting staff generally would not possess specialized
knowledge in the area of energy efficiency for categories of small
sources, there is experience within EPA and other agencies that could
help inform the establishment of presumptive BACT. Both EPA and DOE,
for example, have extensive experience in deploying cost effective
technologies and practices to reduce greenhouse gases from a wide range
of emissions sources in support of the President's GHG intensity goal.
For example the Energy Star program promotes efficient technologies
through a labeling program that establishes performance-based
specifications for determining the most efficient products in a
particular category, which then qualify for the Energy Star label. To
develop these specifications, EPA and DOE use a systematic process that
relies on rigorous market, engineering, and pollution savings analyses
as well as input from stakeholders. While Energy Star specifications
generally cover electrical appliances or fuel combusting appliances
that would be smaller than those triggering the BACT requirement, the
types of analyses conducted for Energy Star could inform the
presumptive BACT process. In addition, DOE's Energy Efficiency and
Renewable Energy program sets standards for several types of equipment,
some of which may be affected by the BACT requirement if GHGs are
regulated, including furnaces, boilers, and water heaters. The DOE
standards are similar to the concept of presumptive BACT in that they
take cost into consideration and are updated over time.\280\ They also
take into account effects on competitiveness among equipment
manufacturers, which could be a significant concern if left unaddressed
in determining presumptive BACT. We ask for comment on whether these or
other similar programs could serve as a basis for the setting of
presumptive BACT where applicable.
---------------------------------------------------------------------------

    \280\ See, e.g., 42 U.S.C. 6295(o).
---------------------------------------------------------------------------

    Regarding LAER, we note that, as previously discussed, if a NAAQS
were established for GHG at levels lower than current concentrations,
the relevant technology requirement would be LAER, not BACT. We ask for
comment on whether the presumptive BACT approach would have utility for
LAER and whether the particular statutory language of the LAER
requirement would allow a presumptive approach under the same legal
principles laid out for BACT.
    Finally, while presumptive BACT or LAER may have the potential to
help address the problem of numerous small but similar types of
sources, it is likely of less value in making BACT or LAER
determinations at the types of large sources that have generally been
subject to PSD for traditional pollutants. This is because there is
generally less similarity among these traditional sources. Nonetheless,
as noted above, there may be numerous modifications that will be newly
subject to PSD for GHGs at such sources, and there may also be issues
unique to establishing control technology requirements for GHGs that do
not presently exist for such sources. We ask for comment on whether
there are issues at traditional PSD major sources that arise for GHGs
and that would not be addressed by a presumptive BACT approach. If so,
we ask for comment on additional options for tailoring the BACT
requirement to address these issues.
e. General Permits for Streamlined Permitting of Numerous Similar
Sources
    An approach closely linked with the presumptive BACT concept is the
concept of a general permit for PSD. A general permit is a permit that
the permitting authority drafts one time, and then applies essentially
identically (except for some source specific identifying information)
to each source of the appropriate type that requests coverage under the
general permit. Congress expressly codified the concept of general
permits when it enacted the Title V program (discussed below) and
states have been using general permits and similar process for years in
their own permit programs, particularly for minor source NSR \281\ and
operating permits. Due to the case-by-case nature of PSD permitting for
``traditional'' major sources and the differences among individual PSD
sources, there has not been much interest or activity in general
permitting for PSD. However, if one or more GHGs (particularly
CO₂) become regulated pollutants, this approach merits
strong consideration due to the large number of sources that EPA
expects will become newly subject to PSD for their GHG emissions and
the similar characteristics of many of these sources.
---------------------------------------------------------------------------

    \281\ The minor NSR is a NAAQS-based program for review of minor
sources that is distinct from the PSD program. It is not discussed
here.
---------------------------------------------------------------------------

    Although there is no provision in the CAA that expressly authorizes
the use of general permits in the PSD program, the D.C. Circuit, in the
Alabama Power case described above, recognized that ``[c]onsiderations
of administrative necessity may be a basis for finding implied
authority for an administrative approach not explicitly provided in the
statute'' and expressly identified general permits as an alternative to
the exemptions that were at issue in that case. See 636 F.2d at 360.
Further, courts have recognized EPA's authority to use general permits
under section 402 of the Clean Water Act without an express provision
authorizing such general permits. Environmental Defense Center v. EPA,
344 F.3d 832, 853 (9th Cir. 2003) (``General permitting has long been
recognized as a lawful means of authorizing discharges.'') (citing
NRDC. v. Costle., 568 F.2d 1369, 1381 (D.C. Cir. 1977)); NRDC v.
Train., 396 F. Supp. 1393, 1402 (D.D.C. 1975) (EPA has ``substantial
discretion to use administrative devices, such as area permits, to make
EPA's burden manageable.'').
    In considering the use of general permits within the PSD program,
EPA is considering how such general permits would be established and
used, and what provisions in the CAA might limit their establishment
and use. One consideration in establishing PSD general permits is the
requirement in CAA section 165(a)(2) that permits be issued after ``a
public hearing has been held with opportunity for interested persons
including representatives of the Administrator to appear and submit
written or oral presentations.'' One possible approach for fulfilling
the public participation requirement is the approach followed for Title
V general

[[Page 44510]]

permits in 40 CFR 70.6(d), which provide that permitting authorities
may establish general permits after following notice and comment
procedures required under 40 CFR 70.7(h) and then grant a source's
request to operate under a general permit without repeating the public
participation procedures. Other considerations for establishing general
permits under the PSD program include determining BACT on a case-by-
case basis (as discussed in the previous section), and the other
requirements referred to earlier in this section concerning the
evaluation of impacts on AQRVs in Class I areas and the analysis of air
quality and other potential impacts under CAA section 165(e).
    EPA seeks comment on the use of general permits within the PSD
program, including both EPA's authority to do so and suggestions for
how general permits would be established and used consistent with the
requirements of the CAA and identification of source categories that
could benefit from such an approach. We also ask for comment on whether
a general permit program approach could also work for nonattainment NSR
in the event the EPA promulgates a NAAQS for GHGs and designates areas
as nonattainment.
f. Coordinating Timing of PSD Streamlining With GHG Regulation Under
the Act
    Regardless of how EPA might tailor the NSR program for GHGs, the
timing of these approaches must be coordinated with other GHG actions
under the CAA. As described above, the applicability of PSD is tied to
whether a pollutant is subject to a control program under the Act. EPA
strongly believes that we should be prepared the first time we regulate
one or more GHGs under any part of the CAA to explain our approach to
permitting, including full consideration of the ideas presented above
for responding to the PSD implementation challenges. Coordination of
the timing of tailoring strategies for PSD or nonattainment NSR to
match with the effective date of the first GHG regulation is necessary
to minimize confusion on the part of sources, permitting authorities,
and the public, to provide for as effective a transition as possible,
and to ensure that the strategies intended to avoid problems can be in
place in time to prevent those problems. We seek comment on timing
issues in general, and particularly on the coordination of the timing
of permitting requirements with the timing of GHG regulation under
other parts of the Act.

F. Title V Operating Permits Program

1. What Are the Clean Air Act Requirements Describing the Operating
Permits Program?
    The Title V operating permits program was enacted in 1990 to
improve sources' compliance with the requirements of the CAA.\282\ In
summary, it provides for facility operating permits that consolidate
all Act requirements into a single document, provides for review of
these documents by EPA, States, and the public, and requires permit
holders to track, report, and certify annually to their compliance
status with respect to their permit requirements. Through these
measures, it is more likely that compliance status will be known, any
noncompliance will be discovered and corrected, and emissions
reductions will result. Title V generally does not add new substantive
requirements for pollution control, but it does require that each
permit contain all a facility's ``applicable requirements'' under the
Act, and that certain procedural requirements be followed, especially
with respect to compliance with these requirements. ``Applicable
requirements'' for Title V purposes generally include all stationary
source requirements, but mobile source requirements are excluded.
---------------------------------------------------------------------------

    \282\ The operating permits program requirements are contained
in title V of the CAA, and are codified in EPA regulations at 40 CFR
parts 70 and 71.
---------------------------------------------------------------------------

    Presently there are generally not any applicable requirements for
control of GHGs that would be included in Title V permits, but
regulation of GHGs under any of the approaches described above,
including PSD, could give rise to applicable requirements that would be
included. Even if a particular source emitting 100 tpy of a GHG is not
subject to GHG regulations that are ``applicable requirements,'' under
a literal reading of Title V, the Title V permit for that source must
include any other applicable requirements for other pollutants. For
example, while a 100 tpy CO₂ source would usually have
relatively small criteria pollutant emissions that would not by
themselves have subjected the source to title V, once subjected to
title V for CO₂ emissions, the source would then need to
include any SIP rules (e.g., generally applicable opacity limitations
that exist in several SIPs) that apply to the source.
    When a source becomes subject to Title V, it must apply for a
permit within one year of the date it became subject.\283\ The
application must include identifying information, description of
emissions and other information necessary to determine applicability of
CAA requirements, identification and certification of the source's
compliance status with these requirements (including a schedule to come
into compliance for any requirements for which the source is currently
out of compliance), a statement of the methods for determining
compliance, and other information. The permitting authority then uses
this information to issue the source a permit to operate, as
appropriate. A Title V source may not operate without a permit, except
that if it has submitted a complete application, it can operate under
an ``application shield'' while awaiting issuance of its permit.
---------------------------------------------------------------------------

    \283\ The deadline may be earlier if the permitting authority
(usually an approved state or local air pollution control agency,
but in some cases the EPA) sets an earlier date.
---------------------------------------------------------------------------

    Title V permits must contain the following main elements: (1)
Emissions standards to assure compliance with all applicable
requirements; (2) a duration of no more than 5 years, after which the
permit must be renewed; (3) monitoring, recordkeeping, and reporting
requirements necessary to assure compliance, including a semiannual
report of all required monitoring and a prompt report of each deviation
from a permit term; (4) provisions for payment of permit fees as
established by the permitting authority such that total fees collected
are adequate to cover the costs of running the program; and (5) a
requirement for an annual compliance certification by a responsible
official at the source. An additional specific monitoring requirement,
compliance assurance monitoring (CAM), also applies to some emissions
units operating at major sources with Title V permits.\284\ The CAM
rule requires source owners to design and conduct monitoring of the
operation of add-on control devices used to control emissions from
moderately large emissions units. Source owners use the monitoring data
to evaluate, verify, and certify the compliance status for applicable
emissions limits.\285\ The CAM rule is implemented in conjunction with
the schedule of the operating permits program.
---------------------------------------------------------------------------

    \284\ Specifically, CAM applies to units with add-on control
devices whose pre-control emissions exceed the applicable major
source threshold for the regulated pollutant.
    \285\ CAM requirements are codified in 40 CFR part 64.
---------------------------------------------------------------------------

    While these are the main elements relevant to a discussion of GHGs,
there are numerous other permit content requirements and optional
elements, as set forth in the Title V implementing regulations at 40
CFR 70.6. One of these

[[Page 44511]]

optional elements is of particular interest when considering the
implications of GHG permitting: The provisions for general permits,
which, as discussed in more detail below, can allow for more
streamlined permitting of numerous similar sources.
    In addition to the permit content requirements, there are
procedural requirements that the permitting authority must follow in
issuing Title V permits, including (1) determining and notifying the
applicant that its application is complete; (2) public notice and a 30-
day public comment period on the draft permit, as well as the
opportunity for a public hearing; (3) notice to EPA and affected
states, and (4) preparing and providing to anyone who requests it a
statement of the legal and factual basis of the draft permit. The
permitting authority must take final action on permit applications
within 18 months of receipt. EPA also has 45 days from receipt of a
proposed permit to object to its issuance, and citizens have 60 days to
petition EPA to object. Permits may also need to be revised or reopened
if new requirements come into effect or if the source makes changes
that conflict with, or necessitate changes to, the current permit.
Permit revisions and reopenings follow procedural requirements which
vary depending on the nature of the necessary changes to the permit.
2. What Sources Would Be Affected If GHGs Were Regulated Under Title V?
    Title V requires permitting for several types of sources subject to
CAA requirements including all sources that are required to have PSD
permits. However, it also applies to all sources that emit or have the
potential to emit 100 tpy of an air pollutant.\286\ As discussed above
for the PSD program, the addition of GHG sources to the program would
trigger permitting requirements for numerous sources that are not
currently subject to Title V because their emissions of other
pollutants are too small. The Title V cutoff would bring in even more
sources than PSD because the 100 tpy (rather than 250 tpy) cutoff
applies to all source categories, not just the ones specified in the
Act's PSD provisions.
---------------------------------------------------------------------------

    \286\ Other sources required to obtain Title V permits are
``affected sources'' under the acid rain program, and sources
subject to NSPS or MACT standards (though non-major sources under
these programs can be exempted by rule). It does not apply to mobile
sources.
---------------------------------------------------------------------------

    Using available data, which we acknowledge are limited, and
engineering judgment in a manner similar to what was done for PSD, EPA
estimates that more than 550,000 additional sources would require Title
V permits, as compared to the current universe of about 15,000-16,000
Title V sources. If actually implemented, this would be more than a
tenfold increase, and many of the newly subject sources would be in
categories not traditionally regulated by Title V, such as large
residential and commercial buildings. However, as described below, EPA
believes that, if appropriate, there may be grounds to exclude most of
these sources from Title V coverage, either temporarily or permanently,
under legal theories similar to those for PSD.
    The CAM requirement also applies to major sources that require
Title V permits, meaning that a number of smaller sources are
potentially newly subject to CAM as well. Under the current CAM
requirements, applicability is limited to the monitoring of add-on
control devices (e.g., scrubbers, ESPs). Presently there are few known
add-on control devices for CO₂, and for many smaller
sources, it is unlikely that there will be cost effective add-on
controls for CO₂ for many years. Thus, we generally expect
source owners to comply with any applicable GHG limits through the use
of improved energy efficiency and other process operational changes
rather than the use of add-on emissions reduction devices. As a result,
even with the large number of sources that will exceed the
applicability cutoffs, the CAM rule will have very limited application
for sources subject to GHG rules. We ask for comment on this assessment
of CAM applicability, and whether there may be CAM impacts that we have
not described here.
    As an additional note, if GHGs were regulated under section 112
authority, Title V could apply at an even smaller threshold. This
consideration adds to the list of difficulties with using section 112
to regulate GHGs that were identified in section VII.C. Although HAPs
are excluded from the definition of ``regulated NSR pollutant,'' Title
V explicitly includes major sources as defined in section 112 on the
list of sources required to obtain an operating permit. While minor
sources of HAP can be excluded by rule, major sources of HAP cannot.
For HAPs, the major source cutoffs are (as noted previously) 25 tons
for any combination of HAPs, and 10 tons for any single HAP. Thus, if
GHGs were regulated as HAPs, a 10 ton CO₂ source would
require an operating permit under Title V. Under this approach, the
number of new Title V sources would easily number in the millions
absent a means to limit PTE. In addition the major source definition
under section 112 does not exclude fugitive emissions, as it does under
PSD for unlisted categories. Thus, if GHGs were designated as HAPs, an
uncertain number of additional new kinds of sources (e.g., agriculture,
mining), would become newly subject to Title V due to fugitive
emissions of GHGs. We ask for comment on whether there are factors EPA
should consider in its description of the universe of potentially
affected sources.
3. What Are the Key Milestones and Implementation Timeline if Title V
Were Applicable for GHGs?
    Under an interpretation of the Act parallel to that for PSD, Title
V would become applicable for GHGs as soon as GHGs become subject to
any actual control requirement. This timing is perhaps even more
important for Title V than for PSD because of the potential for an
extremely large number of new sources (unless EPA administratively
reduced coverage) combined with the fact that Title V applications
would all be due at the same time (unless a phase-in approach were
adopted). This is because Title V requires permit applications within
one year of a source becoming subject to the program, in contrast to
the PSD program, where permitting authorities would receive
applications over time as sources construct or modify.
    Permitting authorities generally must act on Title V applications
within 18 months. However, Congress addressed the burden imposed by the
initial influx of (what turned out to be less than 20,000) initial
Title V permits when it enacted Title V in 1990 by providing for a 3-
year phased permit issuance timeline. Although the initial phase-in
period is over, we discuss below the possibility of interpreting Title
V provisions to authorize a phase-in period for GHG sources becoming
newly subject to Title V as well. We ask for comment on whether there
are factors EPA should consider in its description of these timelines.
4. What Are Possible Cost and Emission Impacts of Title V for GHGs?
    Title V generally does not impose additional applicable
requirements on a source. However, sources, permitting authorities,
EPA, and the public (to the extent that they participate in the
permitting process) all may incur administrative burden due to numerous
activities associated with applying for, reviewing, commenting on, and
complying with Title V permits. There are significant challenges that
would arise if GHG sources become subject to Title V. The sheer volume
of new permits would heavily strain the

[[Page 44512]]

resources of state and local Title V programs. These programs may have
to tailor their fee requirements or other program elements to address
the strain caused by the influx of numerous smaller sources, even if
the permits for each individual source are relatively straightforward.
Many new types of sources would need to understand and comply with a
new and unfamiliar program. Even under streamlined approaches like
general permits (discussed below), there would be administrative burden
imposed as sources would have to determine whether they are covered
and, if so, would need to submit annual reports and certifications. EPA
would see additional burden as well, both because we are the permitting
authority in some areas and because we would probably see an increase
in the number of Title V petitions. Because Title V does not create new
applicable requirements, the new costs of Title V would be mainly
attributable to administrative burden. Nonetheless, this overall
administrative burden is likely to be unreasonable unless EPA reduces
the number of covered sources as discussed below.
    Title V of the CAA also contains a self-funding mechanism requiring
that permitting authorities collect permit fees adequate to support the
costs of running a Title V program. Title V fees must be used solely to
run the permit program. For GHGs, the possibility of a huge influx of
smaller sources raises questions about how permitting authorities
should adjust their fee schedules to ensure that they have adequate
resources to permit these sources without causing undue financial
hardship to the sources. The most common approach, a cost per ton fee
that is equal for all pollutants, would likely result in excessive
costs to GHG-emitting sources because of the large mass emissions of
GHGs compared to other pollutants. This is particularly true for the
universe of small sources brought into Title V solely for their GHG
emissions, because those permits are expected to be relatively simple
and may even be addressed through general permits (which would not
require as many resources or as high a fee). Although it may be
permissible for permitting authorities to adopt lower fees specifically
for GHGs, they would have to assess the new resources needed for
permitting these sources and determine some basis for an appropriate
fee and a workable mechanism for collecting it.
    As noted above, the benefits of Title V stem primarily from the way
its various provisions contribute to improved compliance with CAA
requirements. However, for the particular sources that would be added
to the program solely due to their GHG emissions, it is unclear whether
there would be much benefit from these provisions given the small size
of most of these new sources, the uniform design and operation of many
of their emissions points, the anticipated lack of add-on control
devices, and the relatively small number of applicable requirements
that would be included in the permit. We ask for comment on the
expected overall costs and benefits of running a Title V program for
small GHG sources and for larger GHG sources (e.g., those emitting more
than 10,000 tons per year).
5. What Possible Implications Would Use of This Authority for GHGs Have
for Other CAA Programs?
    Because Title V is designed to work in concert with other CAA
requirements and is self-funding, we have not identified any impacts it
would have on other programs.
6. What Are Possible Tailoring Approaches To Address Administrative
Concerns for Title V for GHGs?
    As we did in section VII.D regarding NSR, we present here for
comment some possible tailoring options to address concerns about
implementing Title V for GHGs. As was previously noted for NSR, we must
consider how the Act's language may constrain these options.
Nonetheless, we see at least two possible legal theories for reducing
administrative concerns through limiting the scope of coverage of Title
V that would otherwise result from regulating GHGs. First, case law
indicates that in rare cases, the courts will interpret or apply
statutory provisions in a manner other than what is indicated by their
plain meaning. Courts will do so when Congress's intent differs from
the plain meaning, as indicated by other statutory provisions,
legislative history, or the absurd, futile, strange, or indeterminate
results produced by literal application. Second, the administrative
burden of literal application of the Title V provisions may also
provide a basis for EPA, based on the judicial doctrine of
administrative necessity, to craft relief in the form of narrowed
source coverage, exemptions, streamlined approaches or procedures, or a
delay of deadlines. Some specific options are discussed in the
remainder of this section, and we invite comment on these and other
suggested approaches.
a. Potential for Higher Major Source Cutoffs
    As discussed above in section VII.A.5, Title V applies to several
types of sources under the Act, including, among others, all PSD
sources, as well as 100 tpy sources that are not subject to PSD. In
section VII.D, we described the reasons why a higher major source
cutoff for PSD might make sense to improve the effectiveness of the
program by focusing resources away from numerous small sources for
which the environmental benefits gained from permitting may not justify
the associated administrative burdens. We believe such an approach
might be even more important for Title V because many small sources
that could become subject to the program solely because of their GHG
emissions may have few or no applicable requirements. Unless GHG
emissions from these small sources are regulated elsewhere under the
Act, the only GHG-related applicable requirements for these sources
would come from PSD permitting. Thus, if EPA adopts a higher major
source size for PSD, it would arguably be incongruous to require 100
tpy GHG sources to obtain permits under Title V. In that case, adopting
a higher applicability threshold for GHGs under Title V in parallel
with, and at the same level as for PSD, would make even more sense.
Similarly, if EPA were to regulate GHGs for certain source categories
under CAA section 111 or 112, and were to include size cutoffs in those
regulations, then it could make sense for the size-cutoffs for Title V
purposes to reflect the cutoffs for those source categories under those
regulations. Indeed, it could make sense to apply Title V only to those
sources of GHGs that are themselves subject to regulation for GHG
emissions.
    We have found several indications of congressional intent that
could serve as a basis for interpreting the Title V applicability
provisions to implement the above-described size-cutoffs or other
limitations, instead of interpreting them literally. First, other
provisions in Title V and the legislative history indicate that the
purpose of Title V is to promote compliance and facilitate enforcement
by gathering into one document the requirements that apply to a
particular source. See section 504(a) (each Title V permit must contain
terms ``necessary to assure compliance with applicable requirements''
of the CAA), H.R. Rep. No. 101-490, at 351 (1990) (``It should be
emphasized that the operating permit to be issued under this title is
intended by the Administration to be the single document or source of
all of the requirements under the Act applicable

[[Page 44513]]

to the source.''). Limiting the applicability of Title V to sources
that emit GHGs in the same quantity as sources that would be subject to
GHG limits under PSD (or other CAA requirements) for GHGs--and
excluding sources that emit GHGs in lower quantities and therefore are
not subject to CAA requirements for GHGs--would be consistent with that
purpose. Second, the legislative history of Title V indicates that
Congress expected the provisions to apply to a much smaller set of
sources than would become subject at 100 tpy GHG levels. See S. Rep.
101-228, at 353 (``[T]he additional workload in managing the air
pollution permit system is estimated to be roughly comparable to the
burden that States and EPA have successfully managed under the Clean
Water Act[,]'' under which ``some 70,000 sources receive permits,
including more than 16,000 major sources'').
    We ask for comment on whether we should consider higher GHG
applicability cutoffs for Title V, what the appropriate cutoffs might
be, and whether there are additional policy reasons and legal
justifications for doing so or concerns about such an approach.
b. Potential for Phase-In of Title V Requirements
    Due to the severe administrative burden that would result if
hundreds of thousands of sources were all to become subject to Title V
at the same time, as could be the case if EPA regulates GHGs elsewhere
under the Act, and because many of the sources could become subject
before the development of any stationary source controls for GHGs, it
may make sense to defer Title V applicability for GHG sources that are
subject to Title V solely due to GHG emissions. One deferral approach
would be to defer Title V for such sources until such time as they
become subject to applicable requirements for GHGs. Alternatively, it
may make sense to phase in Title V applicability with the largest
sources applying soonest, similar to what was discussed above for PSD
permitting.
    Legal support for some type of deferral may be found in the case
law, described above, that identifies deferral as one of the tools in
the ``administrative necessity'' toolbox. In the case of Title V,
deferral may find further legal support by reference to provisions of
Title V itself: Congress addressed the burden imposed by the initial
influx of tens of thousands of Title V permits when it originally
enacted Title V in 1990 by providing for a 3-year phased permit
issuance timeline.\287\ A similar phased approach may have even greater
merit here due to the even greater number of permits. We ask for
comment on the legal and policy arguments for or against a phase-in
approach, and request suggestions for workable permit application and
issuance timelines for Title V permits for small GHG sources.
---------------------------------------------------------------------------

    \287\ CAA section 503(c).
---------------------------------------------------------------------------

c. General Permits
    The use of general permits is an additional option for addressing
the potentially large numbers of GHG sources that could become subject
to Title V. While general permits would not completely eliminate the
resource burden, and may not work for every type of source, they
clearly offer an option for meeting the Title V requirements in a more
efficient way. Congress expressly provided for general permits for
Title V and many states have experience issuing them. They appear to be
a good fit for the numerous similar small sources we are primarily
concerned about. Nonetheless, we still expect that the sheer volume of
sources and number of different types of sources affected will present
challenges. Further, any Title V general permit must comply with all
requirements applicable to permits under Title V, and no source covered
by a general permit may be relieved from the obligation to file a
permit application under section 503 of the Act. We seek comment on
whether source characteristics and applicable requirements are similar
enough for a general permit approach to be helpful, for what categories
it would provide the greatest benefit, and the degree to which it would
or would not ease the expected difficulties with implementing a GHG
Title V program.
d. Fees
    Title V contains a self-funding mechanism requiring that permitting
authorities collect permit fees adequate to support the costs of
running a Title V program. Title V fees must be used solely to run the
permit program. For GHGs, the possibility of a huge influx of new
sources raises questions about how permitting authorities should adjust
their fee schedules to ensure that they have adequate resources to
permit these sources. Title V provides significant flexibility to
permitting authorities in setting their fee schedules so long as they
can demonstrate that fees are adequate to cover all reasonable costs
required to develop and administer the Title V program
requirements.\288\ The additional resource burden imposed by GHG
sources will depend heavily on what approaches EPA and states
ultimately adopt for tailoring the program for these sources, but EPA
does expect that some additional resources will be necessary under
virtually any scenario.
---------------------------------------------------------------------------

    \288\ See CAA section 502(b)(3), which also lists specific
activities whose costs must be covered.
---------------------------------------------------------------------------

    Most states charge Title V fees on a dollar/ton basis, and actual
amounts vary from state to state. For 2008, EPA charges $43.40 per ton,
but only for regulated pollutants for the fee calculation (which
generally includes all regulated pollutants but excludes carbon
monoxide and some other pollutants). Because of the large mass
emissions of GHGs and especially of CO₂ compared to other
pollutants, if EPA and states charge fees for GHG emissions based on
cost/ton numbers for criteria pollutants or HAPs, we expect that the
fee revenues would be grossly excessive for what is needed to process
permits for GHG sources. This is particularly true for the universe of
small sources brought into Title V solely for their GHG emissions
because those permits are expected to be relatively simple and may be
addressed through general permits. Therefore we believe that it is
appropriate for permitting authorities to consider other available
options for covering their GHG source permitting costs, including:
substantially lower cost per ton fees for GHGs, fixed fees (e.g., one
time or annual processing fee that is the same for all applicants below
a certain size), and/or charging no fees for smaller GHG sources. We
ask for comment on these and other suggestions for permitting
authorities to use on structuring their fee provisions. We also request
comment on the expected resource burden resulting from new GHG
permitting, and how EPA should determine the adequacy of fees. EPA
rules contain an optional method for permitting authorities to use in
calculating a presumptively adequate fee. These regulations do not
include GHGs as a regulated pollutant for this calculation but could in
the future if GHGs were regulated under certain parts of the Act. For
permitting authorities that still use this presumptive calculation, we
ask for comment on whether, for the reasons described above, EPA should
specifically exclude GHGs from this calculation or address it in a
different manner. Finally, because EPA itself is the permitting
authority for some sources, we are also interested in comments on
whether and how EPA should change its fee structure in its part 71
permitting regulations to meet

[[Page 44514]]

its own increased resource needs from GHG permitting.\289\
---------------------------------------------------------------------------

    \289\ Technically these increased resources would need to be
provided to EPA through increased appropriation, as the EPA fee
revenues would go to the general treasury.
---------------------------------------------------------------------------

e. Coordinating Timing With Other Actions
    Like PSD, the timing of any approach to streamline Title V must be
coordinated with other GHG actions under the CAA. We believe that any
EPA determination about the applicability of the Title V program to
GHGs should be accompanied by an explanation of how EPA plans to
address--and how we recommend that State and local permitting
authorities address--the numerous implementation challenges such a
determination would pose. This timing is perhaps even more important
for Title V than for PSD because of the potential for an extremely
large number of new sources and the fact that Title V applications
would (unless a phase-in approach is adopted) all be due at the same
time, whereas PSD applications would come in over time as sources
construct or modify. We seek comment on timing issues in general, and
particularly on the coordination of the timing of Title V applicability
with the timing of GHG regulation under other parts of the Act.
    We specifically request comment on the timing of the applicability
of Title V permit requirements in relation to the applicability of GHG
control requirements. Consider the scenario where EPA issues a rule
regulating GHGs from mobile sources, and then issues a series of rules
regulating GHGs from categories of stationary sources. One possible
interpretation of the Act and EPA's regulations is that the mobile
source rule would trigger the applicability of Title V, at which point
the hundreds of thousands of 100-ton and above sources would become
subject toTtitle V and would have one year to apply for Title V
permits. Generally, however, these permits would initially contain no
applicable requirements for control of GHGs (mobile source requirements
are not included in Title V permits), and would likely contain no
applicable requirements for other pollutants, or only some generally
applicable SIP rules that apply to sources which had previously not
needed Title V permits. We have discussed the challenges of issuing
even these minimal permits in such large numbers. However, as EPA
proceeded to issue stationary source rules, each permit with three or
more years remaining on its term would, under current rules, have to be
reopened within 18 months of promulgation of each new rule to
incorporate any applicable requirements from the new rule that would
apply to the permitee. For permits with less than 3 years remaining,
the applicable requirements would be incorporated at permit renewal.
This scenario would result in duplicative effort as permitting
authorities issued hundreds of thousands of minimal Title V permits
with no GHG requirements, followed by a period of numerous reopenings
for some GHG source categories, while the requirements for other GHG
source categories would remain off-permit until renewal, at which point
they would need to be included in the renewal permit. We ask for
comment on how best to tailor the options above to minimize duplicative
effort and maximize administrative efficiency in light of these timing
concerns, and on whether additional options may be needed.

G. Alternative Designs for Market-Oriented Regulatory Mechanisms for
Stationary Sources

    EPA believes that market-oriented regulatory approaches merit
consideration under section 111 or other CAA authorities for regulating
stationary source emissions, along with other forms of regulation.
Economic efficiency advantages of market-oriented approaches that have
the effect of establishing a price for emissions were discussed in
section III. This section discusses four types of market-oriented
approaches:
     A cap-and-trade program, which caps total emissions from
covered sources, providing certainty regarding their future emission
levels, but not their costs.
     A rate-based emission credit program (also called a
tradable performance standard), which imposes an average mass-based
emission rate across covered sources but does not cap total emissions,
so emissions could rise with increased production.
     An emissions fee, which sets a price for emissions but
doesn't limit total emissions from covered sources.
     A hybrid approach, which could combine some attributes of
a rate-based emissions trading system and some attributes of a tax. A
variety of hybrid approaches are possible; the best-known is the
combination of a cap-and-trade system with a ``price ceiling.'' With a
price ceiling, if the price of allowances exceeds a certain level, the
government makes allowances available to the market at the ceiling
price.
    For a local pollutant, a regulatory approach that provides
certainty concerning future emissions can provide a predictable level
of protection, within modeling uncertainties. In the GHG context,
certainty concerning the amount of emission reduction to be achieved by
a U.S. program can make possible an estimated change in predicted
warming, but does not provide certainty that the U.S. will achieve a
desired level of climate protection. This is because GHGs are global
pollutants and the level of climate protection provided depends on the
actions of other countries as well as the U.S.
    There is a robust debate about the respective merits of policies
that provide price certainty, but not emissions certainty, and policies
that provide emissions certainty, but not price certainty. A variety of
cost-containment mechanisms have been proposed for GHG cap-and-trade
systems; these mechanisms offer different tradeoffs between emissions
certainty and price certainty.
    EPA requests comment on the extent to which CAA legal authorities
would accommodate each of these regulatory approaches. In the section
111 context, we note that these market-oriented approaches could be
used in lieu of, or in addition to, other options including emission
rate standards, technology-based standards, or work practices. With
respect to section 111, EPA recognizes that these market-oriented
approaches may differ in significant ways from the manner in which we
have historically designed emission standards and required compliance
with those standards. For this reason, we request comment on the extent
to which each of these approaches could meet the statutory definition
of a ``standard of performance'' and on what additional criteria or
conditions could be considered to ensure that they do so. We also seek
comment on how these options compare based on the policy design
considerations listed in section III.F.1, including effectiveness of
risk reduction, certainty and transparency of results, economic
efficiency, incentives for technology development, and enforceability.
1. Emissions Cap-and-Trade
    A cap-and-trade system limits GHG emissions by placing a cap on
aggregate emissions from covered sources. Authorizations to emit, known
as emissions allowances, are distributed to companies or other entities
consistent with the level of the cap. Each allowance gives the holder
an authorization to emit a fixed amount of

[[Page 44515]]

emissions (e.g., one ton) during a given compliance period. At the
close of the compliance period, sources must surrender allowances equal
to their emissions during that period. Such a system does not impose
limits on emissions from individual sources; rather, it caps emissions
across a group of sources (e.g., an industry sector) and allows
entities to buy and sell those allowances with few restrictions. Key
features of a well-designed cap-and-trade program include accurate
tracking and reporting of all emissions, compliance flexibility, and
certainty (provided by the cap) in achieving emission reductions. While
the cap provides certainty in future emissions, cap-and-trade does not
provide certainty of the price, which is determined by the market
(price uncertainty diminishes as certainty regarding control costs
increases).
    EPA has previously authorized emissions trading under section 111.
For instance, EPA promulgated standards of performance for new and
existing electric utility steam generating units on May 18, 2005 (70 FR
28606), establishing a mercury emissions cap-and-trade program for
coal-fired electric generating units that states could use to meet
their section 111 obligations to control mercury for coal-fired
electric generating units. While the court subsequently vacated this
action, the ruling did not address the legality of trading under
section 111.
    If EPA designed a cap-and-trade program that could cover certain
source categories covered by section 111, such a program could be
modeled after similar trading programs the Agency has developed under
sections 110 and 111 of the Act, such as the NO_X Budget
Trading Program, the Clean Air Interstate Rule NO_X and
SO₂ Trading Programs, and the Clean Air Mercury Rule Trading
Program. Under this model, EPA would establish appropriate state GHG
emissions budgets covering emissions of GHG for each covered source
category. EPA would establish consistent rules related to subjects such
as monitoring, applicability and timing of allocations that states
would be required to meet. EPA would develop and administer a GHG
allowance tracking system, similar to tracking systems the Agency
administers for SO₂, and NO_X. EPA would determine
provisions for monitoring, reporting, and enforcement. If states
promulgated rules consistent with the requirements set forth by EPA,
sources in their State could participate in the trading program.
Alternatively, states could develop alternative regulatory mechanisms
to meet the emissions budgets.
    A key component of an emissions cap-and-trade program is the
ability to accurately monitor emissions.\290\ For many, but possibly
not all, large stationary sources, there are methods to monitor
CO₂ that may provide enough accuracy for a cap-and-trade
program. Most large utility boilers are already required to monitor and
report CO₂ emissions under the Acid Rain Program. Utility
and industrial boilers are well suited to cap-and-trade; many
participate in SO₂ and NO_X trading under the Acid
Rain and NO_X SIP Call programs. At refineries, some emission
sources could be well suited to cap-and-trade, while for others,
accurate monitoring methods or other ways to track and verify emissions
may not be available. More analysis is needed to determine availability
of monitoring methods for all refinery emission sources. The cement
industry is another that may be well suited to emissions cap-and-trade,
since monitoring is available and a number of facilities currently
participate in NO_X trading under the NO_X SIP
Call. Cap-and-trade may not be an appropriate mechanism for the
landfills, except for potential use of landfill gas projects for
offsets. The quantity of landfill methane captured and combusted (i.e.,
the emission reduction) can be measured directly; however, total
emissions are difficult to measure.
---------------------------------------------------------------------------

    \290\ While monitoring is important for determining compli,ance
in all regulatory emission reduction approaches, in a cap-and-trade
system monitoring is also important for functioning of the allowance
market.
---------------------------------------------------------------------------

    We request comments generally on the use of cap-and-trade programs
for GHGs under section 111 and other CAA authorities, including design
elements such as opportunities for sources to opt into such programs,
inter-sector trading and offsets, allowance auctions, cost containment
mechanisms, and conditions or safeguards to ensure that emission
reduction goals are met and that local air quality is protected.
Particular issues to consider include whether it be allowable under
section 111 to develop a cap-and-trade program that covered multiple
source categories or would each source category have to be covered
under a source-category-specific cap-and-trade program. Another issue
is whether it would be legally permissible to allow offsets (i.e.,
obtaining emission reductions from sources outside of the capped
sector) to meet the requirements of section 111.
2. Rate-Based Emissions Credit Program
    A rate-based emissions credit program--also called a tradable
credit standard or intensity target program--is an emissions trading
mechanism. Unlike cap-and-trade, however, a rate-based credit program
does not impose a cap on aggregate emissions from covered sources.
Rather, a rate-based emissions credit program establishes a regulatory
standard based on emissions intensity (e.g., emissions per unit of
input, emissions per unit of product produced, emissions per revenue/
value-added generated). To the extent that a covered source has an
emission rate below the regulatory intensity standard, the source
generates credits that it can sell to sources with emission rates
higher than the regulatory intensity standard. The price of credits
would be determined by the market.\291\ The regulatory intensity
standard might be set below the recent average intensity for a given
industry.\292\ Once in place, the standard would determine the average
emissions intensity (or rate) of the regulated industry.
---------------------------------------------------------------------------

    \291\ Credits are generated by a source with emissions below the
regulatory intensity (or rate). Credits are measured in a fixed unit
of emissions, e.g., a ton. A source that emits at an intensity
higher than the regulatory intensity must surrender credits--
purchased from a source with emissions below the regulatory
intensity or other entity holding credits--equivalent to the
difference between their actual emissions and the allowable
emissions.
    \292\ The average intensity could be set using any of a number
of metrics and baselines. For example, the metric might be tons of
CO₂ emitted per ton of cement produced. The baseline year
for calculating average intensity might be the same as the
compliance year, i.e., after the close of the compliance year, the
average tons CO₂ emitted per ton of cement produced would
be calculated across the industry and a source that produced with
emissions above the average would need to buy credits while a source
that produced with emissions below the average could sell credits.
Alternatively, the average intensity could be based on a year prior
to the initial compliance year.
---------------------------------------------------------------------------

    Like a cap-and-trade approach, a rate-based trading approach can
reduce the cost of reducing emissions from a group of sources, relative
to the cost of requiring every source to reach the same emission rate.
A drawback of the rate-based approach is that it provides an incentive
to increase whatever is used in the denominator of the rate (e.g., the
output of a good or the amount of a particular input). Therefore, rate-
based policies can encourage increased production because production
can be rewarded with additional credits. This in turn has the potential
to encourage increased emissions and thus to raise the overall cost of
achieving a given level of emissions.
    Many of the considerations described above for cap-and-trade
program design

[[Page 44516]]

would also apply to design of a rate-based credit program. Measuring
outputs to determine the regulatory intensity may present some
difficulty. In particular, determining the intensity for facilities
that generate multiple products would be challenging. Sectors that use
multiple inputs (e.g., different fuels) might require use of a common
metric (e.g., Btu combusted) to support a rate-based approach based on
inputs.
    Rate-based trading programs are most easily applied in a specific
sector where facilities have similar emissions characteristics. For
utility and industrial boilers, a rate-based credit standard could be
established for GHG emissions. For refineries, rate-based credit
standards could be established for individual processes or equipment
but would be difficult to set at the facility level. A GHG emissions
rate-based tradable credit standard could be developed for the Portland
cement industry. This mechanism may not be appropriate for landfills
(see discussion of monitoring above).
    We request comments on the use of emission rate trading programs
under section 111 or other CAA authorities. Similar to cap-and-trade
programs, we are seeking comment on whether sector-specific programs or
inter-sector programs might be more appropriate. We also request
comment on issues related to defining emission rates for facilities
producing multiple types of products.
3. Emissions Fee
    A GHG fee would limit GHG emissions by placing a price on those
emissions. The price is fixed up front (unlike cap-and-trade where the
price depends on the market), and a source covered by the tax would pay
to the government the fixed price for every ton of GHG that it emits. A
GHG fee permits the aggregate amount of emissions to adjust in response
to the tax, in contrast to a cap-and-trade system where the quantity of
emissions is fixed. Some key features of a GHG fee include accurate
tracking and reporting of all emissions from covered sources,
compliance flexibility, and certainty in the price of emissions (but
not certainty in future emissions because there is no cap). As noted in
the cap-and-trade subsection above, the emissions of CO2 from most
large utility boilers are already accurately monitored; this attribute
would facilitate application of an emissions tax (as well as
facilitating application of a cap-and-trade system).
    Depending on the specific authority granted by Congress with
respect to the disposition of revenue, the revenue generated by the fee
(as with potential auction revenues under a cap-and-trade approach)
could theoretically be used for any number of public purposes. Note
that depending on how the money was spent, the use of the revenues
would have the potential either to reduce or to increase market
distortions that reduce economic welfare.
    The issue of whether the CAA authorizes emissions fees is discussed
above in section III.F.2.
4. Hybrid Market Based Approach
    A hybrid, market-oriented approach that could be used to regulate
GHG borrows from pollution control options that are based on setting
emissions rates, emissions credit trading, and emissions fees. This
approach starts with a rate-based emissions credit program in which an
average emission rate (e.g., tons of GHGs emitted per unit of output or
input) would be established for a given industry. As with a typical
rate-based policy, a source in the given industry would need to buy
credits to the extent it produces with emissions over the average
intensity, and could sell credits to the extent it produces with
emissions below the average. An element of an emissions fee approach
would then be added to this policy in which the government would also
buy and sell credits. The government could set a price for credits
based on selected policy criteria, and offer credits to sources at that
predetermined price. Sources could then buy credits from the government
as well as other regulated sources. Therefore, the government-set price
would act as a price ceiling (or ``safety-valve''), and the potential
for price fluctuations in emissions credits would be diminished
(because the government's predetermined price would act as a ceiling
price). As long as relatively cost-effective GHG emissions reductions
could occur within a covered sector over time, the average emissions
intensity may decline and total reductions in emissions would occur in
a relatively cost-effective manner without significant government
handling of emissions fee revenues. In addition to being a seller, the
government could also act as a buyer (so the government sales of
credits would not result in an excess supply). A similar approach
without the government's role in selling credits at a ceiling price and
with a fixed schedule of allowable average annual rate of allowable
emissions was actually successfully used in the phase down of lead in
gasoline in the 1980s by EPA.
    Some have suggested that the government could set a price for GHG
credits or allowances based on its assessment of those benefits to be
gained from the GHG emissions reduction per unit of output or input. In
theory, under this approach the marginal compliance costs would never
exceed the marginal benefits of reducing emissions. Note, however, that
there are serious issues to be resolved regarding whether and how a
defensible single estimate of marginal GHG reduction benefits can be
developed for this purpose (see section III.G). First, whether the
scope of benefits counted is global or domestic could significantly
affect the marginal benefits estimate. Second, for benefits categories
that can be quantified and monetized, there are many uncertainties that
result in a range of legitimate estimates, making it difficult to
pinpoint an appropriate number. Third, there is a bias toward
underestimating benefits of GHG reductions because many impacts
categories identified by the IPCC are not quantified and
monetized.\293\ As a result, the price might be set too low to achieve
the amount of emissions reductions that would be warranted considering
all benefits and policy goals.
---------------------------------------------------------------------------

    \293\ There also are policy considerations that would be
neglected by an approach attempting to find a point at which
marginal costs equal marginal benefits. Examples include
irreversibility of changes in climate with adverse impacts affecting
future generations who cannot take part in today's decision-making,
and unequal geographic distribution of adverse climate change
impacts.
---------------------------------------------------------------------------

    By including this discussion, EPA is not taking a position on
whether it has legal authority to pursue a hybrid market-oriented
approach. (See section III.F.2 above.) However, the agency seeks
comment on the general matter of how the pricing of credits within an
emissions intensity approach might be designed and established, what
legal authority would be necessary for this action, and what impact
different price-setting approaches would have on aggregate emissions
reductions, costs and benefits.

VIII. Stratospheric Ozone Protection Authorities, Background, and
Potential Regulation

A. Ozone Depleting Substances and Title VI of the Clean Air Act

    Title VI of the CAA provides authority to protect stratospheric
ozone, a layer high in the atmosphere that protects the Earth from
harmful UVB radiation. Added to the CAA in 1990, Title VI establishes a
number of regulatory programs to phase out and otherwise control
substances that deplete stratospheric ozone. These ozone-depleting
substances (ODS) are used in many consumer and industrial applications,
such as refrigeration,

[[Page 44517]]

building and vehicle air conditioning, solvent cleaning, civil
aviation, foam blowing, and fire extinguishing, and even in small but
important uses such as metered dose inhalers.
    Many ODS and some of the substances developed to replace them
(e.g., HFCs) are also potent GHGs. As described below, Title VI
programs have already achieved significant reductions in emissions of
ODS and thus in emissions of GHGs. However, the ODS being phased out
are not among the six major GHGs addressed by this notice. Because
these ODS are already being addressed by international and national
requirements for protecting stratospheric ozone, they are not covered
by UNFCCC requirements, the President's May 2007 directive or many
other efforts to address climate change. Similarly, the discussion in
this notice of a potential endangerment finding for GHGs does not
include in its analysis the ODS being phased out.
    In this section of the notice, we briefly describe Title VI
regulatory programs as they relate to ODS because of the GHG emission
reductions they achieve. We also consider the Title VI program for
regulating ODS substitutes, since some substitutes are also GHGs. Since
our focus in this notice is on potential use of the CAA to control the
six major GHGs, we also examine the general authority in section 615 as
it might be used to control those GHGs. However, as further explained
below, section 615 would be available for that purpose only to the
extent that EPA finds that emissions of the major GHGs are known or
reasonably anticipated to cause or contribute to harmful effects on
stratospheric ozone or otherwise affect the stratosphere in a way that
may reasonably be anticipated to endanger public health or welfare.
Unlike other CAA provisions examined in this notice, section 615 would
not be triggered by a finding that one or more GHGs cause or contribute
to air pollution that may reasonably be anticipated to endanger public
health or welfare. The potential applicability of section 615 to the
major GHGs depends on whether specified findings related to the
stratosphere or ozone in the stratosphere could be made. In this way,
Title VI is significantly different from other CAA titles that provide
more general regulatory authority to address air pollutants that meet
an endangerment test.
1. Title VI Regulatory Programs
    Existing Title VI programs are largely focused on reducing and
otherwise controlling ODS to protect stratospheric ozone. The
cornerstone Title VI program is a graduated phaseout of ODS that
implements similar requirements in the Montreal Protocol on Substances
that Deplete the Ozone Layer, an international treaty to which the U.S.
is a party. The Title VI phaseout program relies on a system of
marketable allowances to control overall U.S. consumption (defined as
production + imports-exports) consistent with the Protocol's
requirements. EPA tracks production, export, and import of ODS, as well
as transactions in ODS allowances reflecting the flexibility inherent
in the program's market-oriented approach. This ensures compliance with
U.S. consumption caps established under the Protocol. The program also
allows exemptions from the phaseout to ensure that supplies of ODS
critical to certain sectors, like the agricultural fumigant methyl
bromide, are available until alternatives adequately penetrate the
marketplace.
    Other Title VI provisions supplement the phaseout program in a
variety of ways that enhance ozone layer protection. Under these
provisions, EPA has established a national ODS recycling and emission
reduction program, bans on nonessential ODS uses, a program for
labeling ODS-containing products, and the Significant New Alternatives
Policy (SNAP). Under the SNAP program, EPA reviews and approves
substitutes for ODS to help spur the development and uptake of safer
alternatives. Finally, Title VI authorizes EPA to accelerate the
schedule for phasing out ODS as warranted by scientific information,
the availability of substitutes, or the evolution of the treaty's
requirements pursuant to international negotiations among Parties to
the Montreal Protocol.
    Title VI has achieved large reductions in ODS consumption and
emissions, and consequently has reduced GHG emissions and slowed
climate change. According to a recent study, by 2010 ozone layer
protection will have done more to mitigate climate change than the
initial reduction target under the Kyoto Protocol, amounting to avoided
emissions of 11 billion metric tons of CO₂ equivalent per
year, or a delay in climate impacts by about 10 years.\294\
---------------------------------------------------------------------------

    \294\ Velders, G.J. et al., The Importance of the Montreal
Protocol in Protecting Climate, Proceedings of the National Academy
of Sciences, March 2007.
---------------------------------------------------------------------------

    Because some ODS substitutes are GHGs, some have asked whether the
net effect of the Protocol on climate has been beneficial. Recent
research has demonstrated that the climate impact of ODS (e.g.,
chlorofluorocarbons (CFCs), hydrochlorofluorocarbons (HCFCs)), compared
to CO₂ emissions from fossil fuel combustion, fell from
about 33 percent in 1990 to about 10 percent in 2000. The following
graph shows how the shift over time toward ODS alternatives under Title
VI has created a marked downward trend for GHG consumption in sectors
that use ODS and their substitutes, even while these uses have grown
with the U.S. economy and population. As can be seen below, consumption
of the ODS (CFCs, HCFCs, etc.) in 2004, although significantly lower
than peak ODS emissions in 1990, were actually greater than consumption
of HFCs, which are substitutes for CFCs and HCFCs.
    In view of the GHG emission reduction benefits of existing Title VI
programs, EPA seeks public comment on how elements of the existing
Title VI program could be used to provide further climate protection
while assuring a successful completion of the ODS phaseout, including a
smooth transition to alternatives.

BILLING CODE 6560-50-P

[[Page 44518]]

[GRAPHIC] [TIFF OMITTED] TP30JY08.033

2. Further Action Under the Montreal Protocol
    The Montreal Protocol has been and will continue to be an
important, if limited, step in addressing climate change. At the 19th
Meeting of the Parties in September 2007, the Parties agreed to more
aggressively phase out a class of ODS, the hydrochlorofluorocarbons
(HCFCs). The agreement to adjust the phase-out schedule for HCFCs is
expected to reduce emissions of HCFCs to the atmosphere by 47 percent,
compared to the prior commitments under the treaty over the 30-year
period of 2010 to 2040. For the developing countries, the agreement
means there will be about a 58 percent reduction in HCFC emissions over
the same period.
    The climate benefits of the faster phase-out of HCFCs will depend
to some extent on technology choices in the transition from HCFCs. The
estimated climate benefit of the new, stronger HCFC phase-out may be
approximately 9,000 million metric tons of CO₂e. A byproduct
of the manufacture of HCFC-22 is hydrofluorocarbon-23 (HFC-23), a gas
that does not damage ozone in the stratosphere but has a very high GWP.
Because this gas is produced in higher quantities in lower efficiency
production, to the extent that HCFC-22 production in the developing
world remains uncontrolled, additional HFC-23 would be created. Thus,
the agreement to sharply limit future developing world production of
ODS represents an important opportunity for climate protection, as well
as ozone layer recovery, as the President recognized in his April 16,
2008 speech on climate change.

B. Title VI Authorities Potentially Applicable to the Major GHGs

    As mentioned previously, the framework created by Title VI could be
effective in achieving GHG reductions by reducing and controlling ODS
and ODS substitutes through existing mechanisms for tracking
production, evaluating new safer alternatives, and addressing the needs
of the major contributing subsector, refrigeration and air
conditioning, through technician training, emission reduction and
recycling. In this section we review Title VI provisions that could
potentially apply to efforts to reduce the major GHGs that are not also
ODS or ODS substitutes.
    Title VI mostly includes provisions specific to individual ODS and
programs. The provisions generally apply to ``class I'' or ``class II''
ODS. Title VI requires EPA to list specified substances as class I and
class II ODS, and authorizes EPA to add other substances to either
category if the Agency makes certain findings regarding the substance's
effect on stratospheric ozone (see sections 602(a) and (b)). One
important difference between class I and class II ODS is that class I
substances include the most potent ODS; section 602(a) requires EPA to
list as class I substances all substances with an ozone depletion
potential of more than 0.2.\295\
---------------------------------------------------------------------------

    \295\ The ozone depletion potential (ODP) of a chemical measures
its ability to reduce stratospheric ozone compared to a common ODS
known as CFC-11. While this and another common ODS have ODPs of 1.0,
the ODPs of class I and class II ODSs known to be in use range from
0.02 to 10.
---------------------------------------------------------------------------

    Title VI also requires EPA to publish the global warming potential
(GWP) of each listed ODS. Section 602(e) further provides that the
requirement to publish

[[Page 44519]]

GWP for a listed substance ``shall not be construed to be the basis of
any additional regulation under'' the CAA.
    Since the major GHGs being addressed in this notice have no ozone
depletion potential, it appears that the Title VI provisions that
authorize regulation of listed ODS are of limited potential use for
regulating those GHGs. EPA requests comment on the potential
applicability of ODS-specific Title VI authorities, and the
significance of the section 602(e) language quoted above for regulation
of GHGs under Title VI.
1. Section 615
    In addition to the specific provisions that authorize regulation of
listed ODS and in some cases ODS substitutes, Title VI also includes
general authority in section 615 to protect the stratosphere,
especially stratospheric ozone. Section 615 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, and
shall submit notice of the proposal and promulgation of such
regulation to the Congress.

While Title VI was added to the CAA in 1990, a provision largely
identical to section 615 was added to the Act in 1977, soon after
concerns about the effects of some substances on the stratosphere were
initially raised. In 1988, EPA promulgated regulations implementing the
first round of requirements of the Montreal Protocol through a system
of tradable allowances under section 157(b) of the CAA as amended in
1977. Section 157(b) was subsequently modified by the 1990 Amendments
and became section 615.
    Since 1990, EPA has rarely relied on the authority in section 615
to support rulemaking activity, since the activities that the Agency
regulates to protect stratospheric ozone have generally been addressed
under the more specific Title VI authorities. However, in 1993 EPA did
rely on section 615 to promulgate trade restrictions in order to
conform EPA regulations to Montreal Protocol provisions on trade with
countries that were not Parties to the Protocol. (March 18, 1993, 58 FR
15014, 15039 and December 10, 1993, 58 FR 65018, 65044). These trade
restrictions prevented shipments of ODS from the U.S. to countries with
no regulatory infrastructure to control their use. Promulgating these
restrictions reduced the release of ODS into the atmosphere, thereby
reducing harmful effects on public health and welfare. The restrictions
also resulted in eliminating the U.S. as a potential market for ODS
produced in non-Parties, thereby discouraging shifts of production to
non-Parties and limiting the potential for undermining the phaseout.
    Section 615 authority remains available when other CAA authorities
are not sufficient to address effects on the stratosphere, especially
ozone in the stratosphere. For example, in the late 1990s, EPA, the
National Aeronautics and Space Administration (NASA), and the Federal
Aviation Administration (FAA) considered options for addressing
potential ozone depletion resulting from supersonic commercial
aircraft. EPA and NASA analyzed the impacts from a theoretical fleet of
supersonic commercial aircraft, known as High Speed Civil Transport
(HSCT), and in an October 1998 Memorandum of Agreement between the two
agencies (signed by Spence M. Armstrong, Associate Administrator for
Aeronautics and Space Transportation Technology (NASA) and Robert
Perciasepe, Assistant Administrator for Air and Radiation (EPA)) noted
the potential to rely on section 615 in conjunction with other
regulatory authorities.\296\
    While section 615 sets forth the authority and responsibility of
the Administrator to address effects on the stratosphere in order to
protect public health and welfare, EPA recognizes that this authority
was intended to augment other authorities and responsibilities
established by Title VI. EPA does not believe this authority is a basis
for prohibiting practices, processes, or activities that Congress
specifically exempted elsewhere. For example, EPA does not intend to
promulgate regulations eliminating the exceptions from the ODS phaseout
for essential uses as established by section 604.
    For section 615 authority to be used, a two-part endangerment test
unique to that section must be met. First, the Administrator must find,
in his judgment, that ``a substance, practice, process or activity may
reasonably be anticipated to affect the stratosphere, especially ozone
in the stratosphere.'' Second, he must determine that ``such effect may
reasonably be anticipated to endanger health or welfare.'' To determine
the potential applicability of section 615 to major GHGs, EPA thus
would have to consider whether available scientific information
supports making the requisite findings.
    The effect on the stratosphere of GHG emissions and of climate
change generally is a topic of ongoing scientific study.\297\ Recent
science suggests that feedback mechanisms exist that allow temperatures
in the stratosphere and troposphere to be mutually reinforcing or
mutually antagonistic depending on a number of factors, including the
latitude at which the ozone loss occurs. Further research is underway
to better understand these interactions. While it is beyond the scope
of this notice to assess and analyze the available scientific
information on the effect of GHGs on the stratosphere, EPA requests
comment on how evolving science might be relevant to the Agency's
potential use of section 615. More specifically, EPA requests comment
on how scientific research might help resolve areas of ambiguity in the
relationship between GHGs, effects on the stratosphere, and climate
change, and how this might help the Administrator make appropriate
judgments in applying the two-part test of section 615.
---------------------------------------------------------------------------

    \297\ See, e.g., World Meteorological Organization, Global Ozone
Research and Monitoring Project--Report No. 50, Scientific
Assessment of Ozone Depletion: 2006, Ch. 5, Climate-Ozone
Connections.
---------------------------------------------------------------------------

    If the requisite endangerment finding is made, the regulatory
authority provided by section 615 is broad. While most Title VI
authorities are applicable to class I or class II substances or their
substitutes, section 615 authorizes regulation of ``any substance,
practice, process, or activity'' which EPA finds meets the two-part
endangerment test. As noted elsewhere in this notice, depending on the
nature of any finding made, section 615 authority may be broad enough
to establish a cap-and-trade program for the substance, practice,
process or activity covered by the finding, if appropriate. Title VI
provisions provide other examples of possible regulatory approaches,
such as maximizing recapture and recycling and requiring product
labeling. EPA requests comment on possible regulatory approaches under
section 615 and how those approaches would be affected by the
particular endangerment finding that is a prerequisite to the use of
section 615 authority.
2. Section 612
    Section 612 is also relevant to today's notice to the extent a GHG
may be used as a substitute for an ODS. CAA section 612 provides for
the review of alternatives to ODS and the approval of substitutes that
do not present a risk more significant than other alternatives that are
available. Under that authority, the SNAP program has worked
collaboratively for many years with industries, user groups, and other

[[Page 44520]]

stakeholders to create a menu of alternatives that can be substituted
for the ODS as they are phased out of production in the U.S.
    In recent years, industry partners in the motor vehicle air
conditioning (MVAC) sector have urged EPA to identify and approve
appropriate new substitutes to allow for the implementation of a world-
wide platform that will satisfy the needs of the U.S. market while also
meeting new requirements in the European Union, which call for a
transition over approximately six years beginning with the 2011 model
year into non-ODS alternatives with Global Warming Potentials (GWPs) of
less than 150.
    To address these concerns, EPA proposed in September 2006 a SNAP
rulemaking that provided for the use of CO2 and HFC-152a in MVACs (71
FR 55140 docket no. EPA-HQ-OAR-2004-0488). In a separate action (INSERT
FR CITE), EPA has made final the portion of the rulemaking related to
HFC-152a. This substitute meets the EU requirements, while also
providing a new avenue for automakers to replace ODS. We believe we
should issue guidance on the use of CO2 as an MVAC alternative in the
context of the broader considerations of regulating GHGs set forth in
this notice. We have included in the docket cited above a summary of
our proposal regarding CO2 as an alternative from MVACs. This summary
reflects our latest thinking on the safe use of CO2 in those systems.

List of Subjects in 40 CFR Chapter I

    Environmental protection, Air pollution control.

    Dated: July 11, 2008.
Stephen L. Johnson,
Administrator.
[FR Doc. E8-16432 Filed 7-29-08; 8:45 am]
BILLING CODE 6560-50-P
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
