Industry Cries Foul to EPA's Attempt to Regulate GHG Emissions Using the Clean Air Act
| Author | William J. Walsh - Mark A. Erman - Jane C. Luxton |
| Position | The authors assisted several clients (including trade associations) in preparing their comments on the Environmental Protection Agency Prevention of Significant Deterioration Tailoring Rule discussed in this article |
| Pages | 39-42 |
SUSTAINABLE DEVELOPMENT LAW & POLICY
39
inDuStRy cRieS foul to epa’S attempt to
Regulate ghg emiSSionS uSing the clean
aiR act
by William J. Walsh, Mark A. Erman, & Jane C. Luxton*
*Pepper Hami lton LLP. The authors assiste d several clie nts (including trade
associations) in preparing th eir comments on the Environ mental Prote ction
Agency Prevention of Significant Deterioration Tailoring Rule discussed in this
article.
INTRODUCTION
The U.S. House of Representatives passed a comprehen-
sive, albeit fla wed, climate change bill, th e Waxman/
Markey bill, in June 2009, 1 and the Senate Environ-
ment Committee voted to bring a similar, but measurably more
demanding, bill, the Kerry/Boxer bill, to the floor of the Senate.2
The Ho use and Se nate bills cover the s ame greenhouse gases
(“GHGs”) and facilities, require an eighty three percent reduction
in emissions between 2005 and 2050, and create a GHG emission
allowance trading program, which lowers the cost of compliance,
generates funds to provide incentives for the use of carbon cap-
ture and sequestration, and encourages use of more energy-effi-
cient buildings, among other things.3 The Senate bill: (a) requires
covered sources to reduce their GHG emissions twenty percent
below 2005 levels by 2020, as opposed to the House bill’s seven-
teen percent reduction; (b) codifies the Environmental Protection
Agency (“EPA”) Clean Air Act (“CAA”) GHG rule (ensuring the
worst of both worlds (cap-and-trade and command and control
CAA regulation)); (c) imposes a lower offset limit, which will
increase the price of allowances and the cost of the program,
according to the EPA; (d) reduces the total amount of free allow-
ances, primarily to reduce the national deficit, and (e) provides
a $28 price cap on GHG emission allowances, lower than the
House bill’s cap.4 After this strong beginning, both bills stalled,
however, and prospects for passage remain uncertain.
As the year wore on, the climate change spotlight moved dra-
matically from the legislative arena and complementary interna-
tional efforts5 to the development of EPA’s CAA regulations that
will impose GHG-related requirements on industry. In particular,
EPA’s proposed Prevention of Significant Deterioration (“PSD”)
tailoring rule (“PSD Tailoring Rule”) will require the installation
of “best available cont rol technologies” (“BACT”) on new or
modified “major” sources that exceed certain GHG thresholds.6
Even if—as some believe—the Obama Administration’s motiva-
tion in proposing to use the CAA to reduce GHG emissions is to
provide leverage for a legislative solution, now that EPA has pro-
posed the PSD Tailoring Rule, industry has had no choice but to
comment on it. This article provides an overview of these industry
comments regarding the merits of the CAA PSD Tailoring Rule.7
BACKGROUND AND SUMMARY OF THE PROPOSED
PSD TAILORING RULE
In 2007, the Supreme Court in Massachusetts v. EPA held
that carbon dioxide (“CO2”), the most common GHG, was a “pol-
lutant” under the CAA, and, although the Court did not compel
regulation of GHGs, it did require an evaluation of whether GHG
emissions from all sources were causing an endangerment to pub-
lic health and the environment, whether automobile emissions
were contributing to that endangerment, and whether regulation
of mobile sources was required.8 The Court also directed EPA to
“ground its reasons for action or inaction in the statute.”9
The CAA requires PSD permits in attainment areas (areas
that comply with air quality standards) when a new or modified
major source causes a significant net emissions increase, but this
only applies for “each pollutant subject to regulation.”10 Once
GHGs are “subject to regulation” under the CAA, the regulatory
authority must assess if a technology that meets the definition of
BACT exists for GHGs and, if so, must mandate installation of
such BACT as part of the PSD permitting process.11
EPA’s pre-2009 interpretation was that only a pollutant that
is presently subject to a statutory requirement or regulatory provi-
sion that requires actual control of a pollutant is “subject to regula-
tion” under the new source review (“NSR”) program described
above. Under this interpretation, CO2 is not “subject to regula-
tion” because EPA has not established a National Ambient Air
Quality Standard (“NAAQS”) or New Source Performance Stan-
dard (“NSPS”) for CO2, classified CO2 as a Title VI substance, or
otherwise regulated CO2 under any other provision of the Act.12
In response to the remand in Massachusetts v. EPA, EPA dis-
cussed its options in an Advance Notice of Proposed Rulemaking
(“ANPR”) in June 2008,13 and the new Administration proposed
on September 28, 2009, to regulate GHG emissions from light-
duty vehicles (based on EPA’s proposed endangerment finding).14
On December 7, 2009, EPA found that GHG emissions from all
sources endanger public health and welfare and that mobile source
emissions contributed to that endangerment.15
On October 27, 2009, EPA proposed its PSD Tailoring Rule
to address industrial stationary sources of GHG emissions.16 EPA
felt that such a rule was necessary because, once the light-duty
vehicle rule is final, GHGs will be “subject to regulation,” and,
therefore, the GHGs from stationary sources will also immedi-
ately be “subject to regulation” under the PSD program.17
For criteria pollutants (i.e., nitrogen oxides, sulfur oxides,
particulates, lead, ozone, and carbon monoxide), the CAA PSD
and Title V programs define “major” sources as those that emit
more than 100 tons per year for applicability and 250 tons per year
WINTER 2010 40
for PSD significance. If these thresholds are applied to GHGs,
hundreds of thousands, if not millions, of companies (including
many small businesses) will be, in EPA’s words, “burdened by
the costs of individualized PSD control technology requirements
and permit applications . . . . State permitting authorities would
be paralyzed.” 18 To avoid this, EPA invoked the judicial doc-
trines of avoiding absurd results and administrative necessity19
in a two-phase approach. First, EPA proposed establishing appli-
cability thresholds of 25,000 tons per year of CO2 equivalents
(“CO2e”) and a PSD significance level of between 10,000 and
25,000 tons per year of CO2e. Then, EPA proposed that it would
issue a rule within six years that will either confirm the first-phase
permitting levels or establish revised levels or other streamlining
techniques.20
COMMENTS ON THE PROPOSED PSD
TAILORING RULE
The Proposed PSD Tail oring Rule has the potential to
adversely affect millions of plants from an extremely diverse
range of industries and of widely differing sizes. All industry
comments concluded that the rule, if issued as written, will signifi-
cantly impact industrial operations in the United States. More than
5,800 comments (many from individual companies, trade associa-
tions, and industry coalitions representing thousands of compa-
nies) were filed on the PSD Tailoring Rule.21 These comments
express an interesting diversity of views, as well as some clear
and consistent messages.
congReSS DiD not intenD to Regulate ghg
emiSSionS uSing the caa
Virtually every industry comment stated the obvious and irre-
futable fact that Congress simply did not have GHG emissions in
mind when it originally drafted the CAA in 1970 or subsequently
amended it in 1977 to include the PSD program.22 The nature of
GHG emissions (i.e., a global, very long-term impact on climate)
and their control are fundamentally different from the criteria pol-
lutant emissions intended to be addressed by the original CAA
(i.e., protection of local or regional ambient air quality). Thus,
the square peg of GHG emissions does not fit the round hole of
the CAA. This is precisely the reason why Congress has devoted
so much time to considering climate change legislation and why
the presidential candidates from both parties in the last election
favored legislation during the campaign.
Regulation of ghg emiSSionS puRSuant to the
caa iS NOT ReQuiReD by the SupReme couRt
Most industry comments argued persuasively that regulation
of GHG emissions pursuant to the CAA is not required by Mas-
sachusetts v. EPA (see discussion above). Some comments, but by
no means all, argued that climate change regulation was so impor-
tant that it should be addressed by Congress, but such comments
naturally provided little detail concerning what such legislation
might include. In essence, some argue that GHG is a political
issue of global impact that should be decided by Congress. Con-
gress, however, could decide to take no action.
inDuStRy Split conceRning whetheR the abSuRD
ReSultS anD aDminiStRative neceSSity DoctRineS
applieD
Interes tingly, th e industry comments split concern ing
whether, on one hand, the “absurd results” and “administrative
necessity” legal doctrines applied to GHG emissions at all. Thus,
some comments concluded that, if EPA was required to regulate
stationary sources, EPA was compelled to regulate every source
emitting more than 250 tons per year, arguably an absurd result
to be avoided. This legal argument also provides an incentive for
Congress to intervene by amending the CAA to bar or at least
delay use of the CAA to regulate GHG emissions, and proposed
legislation along those lines has already been introduced. The
question remains whether there are enough votes in the House and
Senate to pass legislation barring use of the CAA, no less override
an anticipated Presidential veto.
On the other hand, some industry comments argued not only
that these doctrines applied but that they dictated that EPA must
delay application of the CAA until a regulatory scheme crafted to
address the unique challenges presented by GHG emissions was
developed.
inDuStRy oppoSeD acting befoRe a moRe ReaSoneD
Scheme coulD be DeviSeD
Many of the comments argued that EPA should delay any
regulation—or at least its effective date—for three to six years.
This delay will prevent or minimize ad hoc industry-by-industry
and plant-by plant determinations of whether BACT exists and
will otherwise avoid inadvertently establishing a regulatory pro-
gram without assessing whether it will accomplish the desired
ends, will be cost-effective, or may otherwise result in unintended
adverse consequences.
Such an ad hoc approach to regulating GHG emissions
through permit challenges and enforcement actions presents sev-
eral problems. For coal-fired electric-generating plants, convert-
ing to oil and gas means using more expensive and less reliable
alternative fuels. Forcing the relocation of a coal-fired plant to
another location fails to reduce GHG emissions and may actually
increase them, because of the inefficiency involved in transmitting
power over distance. There has not been a successful large-scale
demonstration of the technical, economic, and environmental per-
formance of geological carbon sequestration, which is considered
to be one of the most promising GHG emission reduction tech-
nologies.23 Immediate application of the PSD applicability thresh-
old and triggers will result in unacceptable delays in permitting
and, therefore, in the construction of new industrial plants and
major modifications of existing plants, a cost not advocated by
Congress.24 Such delays will have a direct and significant adverse
economic impact (including a disincentive to convert to “green”
technologies, which would also need permits).
This concern about delay is more than theoretical. Environ-
mental groups have filed administrative or legal challenges in
more than 166 existing coal-fired electric plant permit proceed-
ings, with 113 claimed “victories” (which includes remands,
delays, and other non-final determinations).25 In fact, the Sierra
41 SUSTAINABLE DEVELOPMENT LAW & POLICY
Club settled one lawsuit in exchange for the utility “voluntarily”
agreeing to add a legally enforce able permit provision that
requires capture and sequestration of fifty eight percent of the CO2
generated by the plant.26
Also, as some comments noted, there is precedent in EPA’s
implementation of the CAA for delaying implementation of
aspects of the PSD program in order to avoid administrativ e
impracticability. For example, the 1980 PSD regulations con-
tained a number of transition provisions that delayed applicability
to certain classes of sources. EPA, in effect, has deferred applica-
tion of PSD provisions based on PM2.5 emissions, despite adop-
tion of National Ambient Air Quality Standards for PM2.5 in 1997,
relying on PM10 (larger-sized particulate matter) instead because
of p roblems measuring and modeling PM2.5 emissions.27 As a
practical matter, delaying any regulatory decision would provide
Congress a reasonable amount of time to act.
one SiZe DoeS not fit all emitteRS
Some industries argued that EPA should not use a one-
size-fits-all approach but rather should tailor the trigger to each
industry (i.e., apply an industry-specific applicability and GHG
emission trigger). A plant-by-plant BACT determination is cost-
ineffective and, in any case, either will inevitably result in a deter-
mination that there is no BACT, as discussed below. However, the
mere existence of such a process creates uncertainty in planning,
obtaining capital, and reacting nimbly to new business opportuni-
ties (such as expanding the production of renewable energy and
more energy-efficient products).
Similarly, some industries argued that the global nature of
endangerment required EPA to take into account on an industry-
by-industry basis, not the percentage of U.S. emissions covered,
but the percentage that each facility within each industry repre-
sents compared to worldwide GHG emissions from all sources in
all countries.
Many industries noted that EPA simply had not performed
even the bare minimum level of evaluation needed to promul-
gate a regulation of this magnitude and import. Various com-
ments demanded that EPA gather sufficient information to tailor
its rules to the circumstances of each industry before issuing a
rule. In evaluating the significance of the GHG emissions from an
individual industry, the EPA should take into account the larger
quantities of GHGs emitted compared to other CAA-regulated
pollutants, the level of significance compared to total GHG emis-
sions, the effectiveness on a global scale of such regulation (e.g.,
the carbon leakage issue) for a particular industry, and the other
issues discussed in the various comments.
higheR thReSholDS ShoulD apply
Many industries28 argued for higher thresholds than 25,000
tons per year because the PSD program was intended to regu-
late only the “major” emitters, such as electric generating plants,
which are financially able to bear the regulatory costs of PSD and
are collectively responsible for most of the nation’s air pollution.
One industry, in effect, recommended changes that result in a
threshold of 777,000 tons per year.29 PSD was not designed to
cover the small- and medium-sized emitters that form a substantial
portion of the nation’s core manufacturing base, but the proposed
rule would do so.30
EPA estimated that if the major source threshold is set at
25,000 tons per year, 13,661 facilities would exceed this thresh-
old, which would cover sixty-eight percent of national station-
ary source emissions.31 At 100,000 tons per year, 4,850 facilities
would be covered, corresponding to sixty-four percent of national
GHG emissions.32 Thus, increasing the threshold from 25,000 to
100,000 tons per year would reduce the number of “major emit-
ters” by almost two-thirds but would only decrease the GHG
emissions subject to regulation by four percent. This marginal
incremental benefit is not consistent with the intent of the PSD
program. One solution presented by an ethanol industry tra de
group is to subject plants to PSD for GHGs only if the plant is
already covered by BACT requirements for other regulated pol-
lutants such as nitrous oxides or sulfur oxides.33
The Small Business Administration ’s Office of Advocacy
also took issue with the 25,000 tons per year threshold by argu-
ing that EPA improperly certified that the Tailoring Rule would
not harm a substantial number of small businesses, thus evading
the Regulatory Flexibility Act’s requirement that a special Small
Business Regulatory Enforcement Act (“SBREFA”) panel be
convened.34 Under EPA’s Regulatory Flexibility Act Guidance,
rules cause a significant economic impact when the compliance
cost for a small business is one to three percent of operating rev-
enues. If less than 1,000 small entities are significantly affected,
the rule is presumed to be ineligible for a SBREFA panel.35 The
Small Business Administration asserted that, had EPA thoroughly
analyzed the potential reach of the GHG permitting requirements
on small entities, it would have learned that the Tailoring Rule
would adversely affect much more than 1,000 small businesses;
therefore, EPA would have to convene a SBREFA panel prior to
promulgating its rule.36
pRoceSS emiSSionS ShoulD be excluDeD
Those industries that utilize intense heat to process raw mate-
rials naturally containing carbonate (e.g., the cement industry, the
limestone mineral processing industry, and the glass manufac-
turing industry) will release CO2, and there simply is no BACT
for these process emissions. Typically, there are no substitutes
for these raw materials and nothing as a practical measure can
be implemented to reduce these emissions. Moreover, some of
these industries meet new tough energy efficiency requirements or
make products that will reduce GHG emissions when utilized in
other energy-saving applications downstream. Nothing in EPA’s
administrative record to the PSD Tailoring Rule demonstrates
that GHG emissions from process emissions can be significantly
reduced with any existing technology. Put simply, there is nothing
meaningful that can be required at this time. Attempting to regu-
late these industries will be a useless act.
the tailoRing Rule ShoulD not apply to plantS
that might ReSult in caRbon leakage
Several industries and industry coalitions noted that so called
carbon leakage is almost certain to increase the net global GHG
emission if the PSD Tailoring Rule prompts regulated entities
42WINTER 2010
to move operations abroad. Many manufacturing industries are
energy-intensive and trade-sensitive, according to EPA,37 indus-
try groups’ testimony to Congress,38 the General Accountability
Office,39 and the comments provided in this rulemaking.
The costs (direct transactional costs, delay costs, and the reg-
ulatory uncertainty’s effect on ability to raise capital) will increase
at U.S. plants in regulated industries. Additional costs will be
imposed if costly BACT is required by states (with little reduction
in GHG emissions). Since no comparable costs will be imposed
on such energy-intensive industries in developing countries, their
U.S. counterparts will suffer a competitive disadvantage. EPA’s
and virtually every other analysis has found that such competitive
disadvantage moves production from the United States to other
countries with less stringent GHG controls.40 Thus, carbon “leak-
age” occurs and, in reality, the total global emissions increase,
not decrease, thereby increasing the endangerment, not reducing
it. The law should not (and does not) require such a truly absurd
result.
theRe aRe no bactS
None of the traditional air pollution con trols are designed
to control CO2 since it has not yet been regulated. Industry com-
ments could n ot identify any BACTs fo r any industry. Even
carbon capture and sequestration ( “CCS”) has not been imple-
mented in the United States at a large coal-fired electric generat-
ing plant. In fact, the Department of Energy is offering billions
of dollars in research to establish wh ether such technology can
be implemented. The smaller the GHG emission source, the less
likely that such a technology wi ll be conside red BACT unde r
EPA’s “top-down” analysis, which eliminates technologies that
may have a high removal efficiency, but low cost-effectiveness.
Finally, the EPA CAA regulations do not include GHG emis-
sion allowances. As a result, unlike the House and Senate bills,
free GHG emission allowances cannot be provided to utilities as
incentives to offset the enormous cost of CCS.
the Rule ShoulD pRoviDe incentiveS to inDuStRieS
that pRoDuce pRoDuctS that ReDuce ghg
emiSSionS oR uSe Renewable eneRgy
Some comments urged EPA to provide an incentive to indus-
tries that initiate modifications and produce products to support
other GHG emission reduction programs like manufacturers of
components or assemble rs of re newable energy sources (e .g.,
solar cells, wind powe r, and biomass energy), materials that
meet energy efficiency standards for buildings, and other energy
efficiency standards. Thus, EPA should consider the net GHG
emission impact of the entity’s project and the purpose for which
it was conducted.
CONCLUSION
In summary, addressing climate change is a scientific, eco-
nomic, and political challenge that raises equity issues within
nations and regions, and betw een develo ped and developing
nations. The inherent complexity is reflected in the fac t that
it took more than 1,400 pages to address all of these cli mate
change issues in the House bill.
EPA’s “reg ulatory fix,” altho ugh elegantly simp le, is also
fundamentally unworkable. The CAA is a technol ogy-forcing
statute that EPA is attempting to use in a situation where there is
little likelihood that GHG reduction technologies will be devel-
oped in th e foreseeable future. The rigid command and control
approach is in stark contrast to the market-based cap and trade
approach of le gislative measures, which is anticipated to lower
the cost of compliance.
Most of industry (including some companies and industries
that support comprehensive federal climate change legislation)
oppose utilizing the CAA to regulate GHG emissions. The tone
and even anger expressed in many of these comments is extraor-
dinary for comments in a rulemaking, which may be due to the
frustration faced by industry. These comments demonstrate that
the proposed PSD GHG Tailoring Rule i s not only broken, but
seems unfixable, at least in the short- to medium- term.
Legal challenges to the rule are already in the works. Sena-
tor Murkowski has proposed a bill that vetoes the endangerment
finding, thereby preven ting t he EPA from regulati ng GHGs
using the CAA. Senator Rockefeller has offered a more moder-
ate bill that will simply delay the effective date of the tailoring
rule requirements for two years. In reaction to the industry com-
ments and Congressional interes t, EPA Administrator Jackson
announced that EPA in tends to use a thresh old substa ntially
higher th an the 25,000-to n limit that EPA originally proposed
and perhaps as high as 75,000 tons. The future of this regulation
is uncertain.
Endnotes: Industry Cries Foul to EPA’s Attempt to Regulate GHG
Emissions Using the Clean Air Act
1 H.R. 2454, 111th Cong. (2009).
2 S. 1733, 111th Cong. (2009).
3 See H.R. 2454; S. 1733.
4 See S. 1733.
5 See Conference of the Parties Fifteenth Session, Copenhagen, Den., Dec.
7-18, 2009, Copenhagen Accord U.N. Doc. FCCC/CP/2009/L.7 (Dec. 18,
2009), available at http://unfccc.int/files/meetings/cop_15/application/pdf/
cop15_cph_auv.pdf (encouraging nations to commit to GHG emission goals
by “tak[ing] note of the Copenhagen Accord”). See generally United Nations
Framework Convention on Climate Change, 1771 U.N.T.S. 107, S. Treaty Doc
No. 102-38, U.N. Doc. A.AC.237/18(Part II)/Add.1, 31 I.L.M. 849 (1992),
available at http://unfccc.int/2860.php (providing other United Nations admin-
istered environmental agreements and texts).
Endnotes: Industry Cries Foul to EPA’s Attempt to Regulate
GHG Emissions Using the Clean Air Act continued on page 61
