Copyright FryeLaw PLLC 2005-2008.

    New Source Review Reform Regulations Subject of D.C. Circuit Argument
    Russ Frye
    February 5, 2005
On January 25, 2005, the U.S. Court of Appeals for the District of Columbia held oral
argument in the consolidated challenges to EPA’s December 31, 2002 rulemaking to
improve the Prevention of Significant Deterioration/New Source Review Program, 67
Fed. Reg. 80,920 (the “NSR Reform Rule”).  This case involves claims by numerous
states and environmental groups that the NSR Reform regulation relaxes the NSR
regulations too much and imposes those relaxations on unwilling states, as well as
challenges by industry groups arguing that the regulation improperly treats as
modifications activities but do not change the source’s capacity to emit.  Russ Frye
represented steel companies who intervened in support of EPA.
The case was heard by a panel consisting of Judge Rodgers, Judge Tatel, and Senior
Judge Williams.  The argument was extraordinary, lasting almost four hours, with
numerous parties arguing each of half a dozen issues.  The panel was very well-
prepared and questioned all of the parties extensively.  
Judges Rodgers and Tatel are considered among the more liberal of the judges on the
D.C. Circuit, while Judge Williams is considered one of the more conservative, and
from the questioning it seemed as if Judge Tatel was more inclined to the
state/environmental petitioners’ positions and Judge Williams was leaning more
towards the EPA/industry position on those issues.  Judge Rodgers, who is known for
asking few questions during oral arguments, was harder to read.
Without recounting here all the four hours worth of colloquy, I have attempted to
summarize some of the key points made on the various issues and give my perception
of how the panel was leaning.  It is, however, usually very difficult to predict how the
panel is likely to rule based on the questions asked at oral arguments.  The Judges
gave no overt indication of how they might rule on any of the issues, nor when.  
There also a few issues (such as the procedure for calculating emissions increases for
electric utilities) which the parties agreed not to present during the oral argument
and to rely on the written argument in the briefs, so of course there was no
discussion by the Court of those issues.

Industry Argument On Capacity To Emit
Industry groups were both petitioners challenging EPA’s action, and
intervenors on behalf of EPA, although some industry groups chose not file
petitions for review and participated only as intervenors.  The industry
petitioners argued that NSR should not apply to an activity that does not
increase the capacity of the source to emit, i.e., its maximum hourly
emission rate.  Their argument relied primarily on the fact that Congress
adopted, for NSR purposes, the statutory definition of “modification” in the
New Source Performance Standard section of the CAA, and EPA at that time
had a history of applying NSPS only to changes that increased the source’s
maximum emission rate.  The Court seemed skeptical, noting that the D.C.
Circuit already approved some EPA departures from the NSPS regulatory
definition of “modification” when implementing NSR, in the Alabama Power
decision.  Unfortunately, the Court was also skeptical of industry’s assertion
that NSPS applicability is based on an increase in emissions capacity rather
than in actual emissions.  The judges focused on the facts that the NSPS
definition of “modification” describes a change “which results in an
increase in the emission rate” and provides for the determination of
whether an increase in emission rate has occurred by comparing before and
after stack tests.  Industry counsel failed to clear up this misunderstanding
during the argument (although it is dealt with more clearly in the briefs).  
My guess is that the Court will reject the industry argument unanimously.  
Most of us thought it was a long shot to begin with, though.  It also may
have had a beneficial effect in that the discussion of the
state/environmental petitioners’ objections, described below, seemed
largely to accept EPA’s move towards using future actual emissions rather
than future potential emissions when calculating emissions increases.

Baseline For Calculating Increases
Next, state and environmental petitioners argued that EPA should not have
allowed the calculation of whether a change results in a significant net
increase in emissions to be based on emissions from the average of any two
years during the past 10 years, but rather should have kept the “baseline” at
the average of the immediately preceding two years.
The Court seemed troubled primarily with two aspects of this argument.  
First, even under the previous NSR regulations, the baseline was the
average emissions for the preceding two years or another period that the
state determined to be more representative.  Thus, most of the problems
the state and environmental petitioners are ascribing to the new rule could
have existed under the old rule, too.  (There was some suggestion, though,
that the new rule would allow baseline emissions to the calculating using
two years within the past 10 that are no longer  representative of now-
normal emissions from the source.)  
Secondly, the judges got the states’ lawyer to concede that there was
nothing magical about the last two years, and EPA could, for example,
previously have specified the average of the preceding three years, instead
of the average of the preceding two years.  Given that, the judges seemed
to have trouble seeing how they could draw a bright line, e.g., going back
five years is acceptable, but 10 years is not.  
My guess (and that of other observers with whom I talked) is that the Court
will uphold the rule with respect to baseline emissions.  (Petitioner
Newmont Mining argued that 10 years was not long enough, because the
business cycle for gold mining is longer than 10 years.  At least one of the
judges expressed skepticism about that, and in any event the same concerns
about holding EPA to some bright-line test would apply to Newmont’s
argument, as well.)  

Enforcement Mechanism
The states argued that the NSR Reform Rule is arbitrary and capricious
because determinations of whether there is a modification are to be made
by the source, but the source is not required by the regulations to keep any
records or provide any reports to support a determination that no
modification will occur and no NSR permit is required, unless there is a
“reasonable possibility” that a change will result in a net emissions increase
above the NSR thresholds.  While there were some questions from the
Court about whether this was any different than the old rule (the states
argued that under the old rule most sources would need to get a permit
limitation to cap their potential to emit), all the judges wondered how it
could really be a “simplification” of NSR implementation to say that states
can go to a facility and inspect it to see what has changed and review the
facility’s capital expenditure records, rather than just asking to see the
record of the facility’s non-applicability determinations for any changes.  
The Court also seemed troubled by the vagueness of the requirement that
sources maintain records of their non-applicability determinations in cases
where there is a “reasonable possibility” that the future actual emissions
may in fact produce a significant net increase in emissions.
EPA did not do a particularly good job of explaining how the Title V
permitting program provides several mechanisms for oversight of sources’
non-applicability determinations.  I believe there is a good chance the Court
will agree with the state and environmental petitioners that it was arbitrary
and capricious for EPA to recognize the importance of recordkeeping to
support non-applicability determinations, especially when they are based on
projected future actual emissions, and yet not include any recordkeeping or
reporting requirements in the final rule.  

Plantwide Applicability Limits
The environmental petitioners argued that the plantwide applicability limit
option contained in the NSR Reform Rule has similar problems to the ten-
year look back provisions on baseline emissions, only worse:  a PAL could be
based on an emission rate the source has not achieved for years, and it
would allow the source to increase its emissions up to that level over the
next five years.  The environmental petitioners argued that, under the D.C.
Circuit’s Alabama Power decision, emission reductions could only be used to
“net out” emissions increases if they are “substantially contemporaneous”,
which a PAL would not require.  
The Court seemed not to be particularly excited about this argument,
especially since it largely followed the same approach as the baseline
emissions argument.  A couple of the judges pointed out that EPA had cited
two pilot program studies showing that PALs resulted in reductions in
emissions, although the environmental petitioners’ lawyer said that those
pilot programs contained more restrictive PAL requirements.  Judge
Williams asked whether there was any reason why a company would prefer
to use the “Clean Unit” provision, rather than a PAL.  He asked the question
again during the Clean Unit portion of the argument, described below, and
the government finally explained that the PAL is more burdensome to
develop and administer, as it applies to the entire facility.  I could not
figure how why Judge Williams was so interested in this question, unless
perhaps he was thinking he could uphold the PAL provisions and strike down
the Clean Unit provisions without limiting the rule’s flexibility very much.  
Overall, my sense was that the environmental groups failed to carry their
burden of demonstrating why the PAL provision is unlawful.  

Clean Units
The Clean Unit provision allows a unit that has undergone a BACT
determination, or the equivalent, within the last five years to increase
emissions up to its maximum emission rate without undergoing NSR.  The
state and environmental petitioners attacked this provision primarily with
respect to its effect in non-attainment areas.  Clean Units would be allowed
to increase emissions in non-attainment areas without ever having to install
technology to meet the Lowest Achievable Emission Rate, and without ever
having had to obtain emissions offsets for those increased emissions.  
Both Judge Tatel and Judge Williams questioned how, in this case, EPA
could look at potential (maximum capacity) emissions, when the remainder
of the NSR Reform Rule looks at increases in actual emissions.  Judge Tatel
suggested that the plain language of the statute, which refers to an increase
in pollutants “emitted by such source,” necessarily suggests that an increase
in actual emissions, rather than potential emissions, will trigger NSR.  Judge
Williams asked for some citation to legislative history that suggests EPA can
look at increases in potential, rather than actual, emissions.  The Justice
Department lawyer did not have good answers to these questions, and again
there was confusion about whether a modification under the NSPS
regulations is based on actual emissions or potential emission rate.  My guess
is that the Court will, at minimum, remand the Clean Unit provision to EPA
for better explanation.  

Pollution Control Projects
The environmental petitioners attacked the exemption in the NSR Reform
Rule for pollution control projects (PCPs) outside the electric utility industry
(which was addressed in the earlier “WEPCo Rule”).  They pointed out that
EPA acknowledged that PCPs are physical changes, but nevertheless decided
to exempt, in most cases, increased emissions resulting from PCPs.  Judge
Tatel suggested that it was not unreasonable to think that Congress intended
to control increased emissions from PCPs, if the PCP resulted in an increase
in emissions above significance levels, and Judge Williams pointed out that
EPA had not shown that the PCP exemption would assure that the increased
emissions being exempted were much less important than the emissions
reductions produced by the PCP.  In short, it seemed to me that the Court
might remand the PCP provisions of the rule for further explanation.

Limits On State Authority
The state petitioners attacked EPA requirements that even states with
approved NSR programs adopt all of what they termed as “relaxations”
contained in the NSR Reform Rule.  They asserted that this violates
provisions of the Clean Air Act that reserve to the states the right to be
more stringent than the federal rules require.  Judge Williams recited EPA’s
argument that the NSR Reform provisions will result in more efficiency and
fewer cases of equipment being run without needed repairs, such that the
new rules will result in the same or less emissions than under the old rules.  
He  suggested EPA’s assertion that a state program therefore would be “less
stringent” than the federal rules if it failed to adopt the NSR Reform Rule
provisions was not clearly unreasonable.  
A least two of the judges questioned the states rigorously on why this claim
was ripe – while EPA has said that states should adopt all of the NSR Reform
Rule provisions, it also has said that it will look at the totality of a state’s
NSR program to determine whether the state program is more or less
stringent than the federal rules.  The states are presuming that EPA would
reject, for example, a state program that failed to adopt the ten-year look-
back for baseline emissions, but EPA has not established any such bright
line.  I think the Court will probably adopt a wait-and-see attitude here,
declining to rule on the states’ objection until presented with an appeal of
EPA’s disapproval of a particular state NSR program.  

As noted at the outset, it is notoriously risky to try to predict the outcome
of a case based on questioning during oral argument.  My guesses about how
the Court may come out on the various issues need to be viewed with that
in mind.  It does seem likely, though, that some aspects of the NSR Reform
Rule will be upheld, and some parts will at least be remanded, if not
vacated.  It also seems likely that the Court will not issue an opinion until
late spring, at the earliest, given the number and complexity of the issues
that will have to be addressed in its written opinion.  
There were a number of comments from counsel and the Court referring to
the fact that another aspect of the NSR regulations, EPA’s October 27, 2003
rule exempting certain equipment replacement projects as “routine
maintenance, repair, and replacement,” will be argued later.  Oral argument
has not yet been scheduled in that case, pending EPA’s action on the
state/environmental group petitions for reconsideration.  When the opinion
is issued in this case, it may provide some insights into how the Court may
come out when the equipment replacement rule is before it.