									6560-50-P

  SEQ CHAPTER \h \r 1 ENVIRONMENTAL PROTECTION AGENCY

40 CFR Part 52

[EPA-HQ-OAR-2009-0597; FRL-xxxx-x]

RIN 2060-AP87

Prevention of Significant Deterioration (PSD): Reconsideration of
Interpretation of Regulations that Determine Pollutants Covered by the
Federal PSD Permit Program

AGENCY:  Environmental Protection Agency.

ACTION:  Final Action on Reconsideration of Interpretation.

SUMMARY:  EPA has made a final decision to continue applying the
Agency’s existing interpretation of a regulation that determines the
scope of pollutants subject to the federal Prevention of Significant
Deterioration (PSD) program under the Clean Air Act (CAA or Act).  In a
December 18, 2008 memorandum, EPA established an interpretation
clarifying the scope of the phrase “subject to regulation” found
within the definition of the term “regulated NSR pollutant.”  After
considering comments on alternate interpretations of this term, EPA has
decided to continue to interpret it to include each pollutant subject to
either a provision in the CAA or regulation adopted by EPA under the CAA
that requires actual control of emissions of that pollutant.  Thus, this
notice explains that EPA will continue following the interpretation in
the December 18, 2008 memorandum with one exception.  EPA is refining
its interpretation to establish that the PSD permitting requirements
will not apply to a newly regulated pollutant until a regulatory
requirement to control emissions of that pollutant “takes effect.” 
In addition, this notice addresses several questions regarding the
applicability of the PSD and Title V permitting programs to greenhouse
gases (GHGs) upon the anticipated promulgation of EPA regulations
establishing limitations on emissions of GHGs from vehicles under Title
II of the CAA.  Collectively, these conclusions result in an EPA
determination that PSD and Title V permitting requirements will not
apply to GHGs until at least January 2, 2011. 

DATES:  This final action is applicable as of [INSERT DATE OF
SIGNATURE].

FOR FURTHER INFORMATION CONTACT:  Mr. David Svendsgaard, Air Quality
Policy Division (C504-03), U.S. Environmental Protection Agency,
Research Triangle Park, NC 27711, telephone number: (919) 541-2380; fax
number: (919) 541-5509, e-mail address:  HYPERLINK
"mailto:svendsgaard.dave@epa.gov" svendsgaard.dave@epa.gov .

SUPPLEMENTARY INFORMATION:

I.  	General Information

A.   Does this action apply to me?  

	Entities potentially affected by this action include sources in various
industry groups and state, local, and tribal governments.

B.   How is this document organized?

This document is organized as follows:	

I.  	General Information

II.	Background

III.	This Action

A. Overview

B. Analysis of Proposed and Alternative Interpretations for Subject to
Regulation

1. Actual Control Interpretation

2. Monitoring and Reporting Interpretation

3. State Implementation Plan (SIP) Interpretation

4. Endangerment Finding Interpretation

5. Section 209 Waiver Interpretation

C. Other Issues on Which EPA Solicited Comment

1. Prospective Codification of Interpretation

2. Section 821 of the Clean Air Act Amendments of 1990

3. Timing of When a Pollutant becomes Subject to Regulation

IV.	Application of PSD Interpretive Memo to Permitting for GHGs

A. Date by Which GHGs Will Be “Subject to Regulation”

B. Implementation Concerns 

C. Interim EPA Policy to Mitigate Concerns Regarding GHG Emissions from
Construction or Modification of Large Stationary Sources

D.  Transition for Pending Permit Applications

V.	PSD Program Implementation by EPA and States

VI.	Application of the Title V Program to Sources of GHGs

VII.	Statutory Authority

VIII. Judicial Review

II. 	Background

	On December 18, 2008, then-EPA Administrator Stephen Johnson issued a
memorandum setting forth EPA’s interpretation regarding which
pollutants were “subject to regulation” for the purposes of the
federal PSD permitting program.  See Memorandum from Stephen Johnson,
EPA Administrator, to EPA Regional Administrators, RE:  EPA’s
Interpretation of Regulations that Determine Pollutants Covered by
Federal Prevention of Significant Deterioration (PSD) Permit Program
(Dec. 18, 2008) (“PSD Interpretive Memo” or “Memo”); see also 73
FR 80300 (Dec. 31, 2008) (public notice of Dec. 18, 2008 memo).  The
Memo interprets the phrase “subject to regulation” to include
pollutants “subject to either a provision in the CAA or regulation
adopted by EPA under the CAA that requires actual control of emissions
of that pollutant,” while excluding pollutants “for which EPA
regulations only require monitoring or reporting.”  See Memo at 1. 
The Memo was necessary after issues were raised regarding the scope of
pollutants that should be addressed in PSD permitting actions following
the Supreme Court’s April 2, 2007 decision in Massachusetts v. EPA,
549 U.S. 497 (2007).  

In Massachusetts v. EPA, the Supreme Court held that GHGs, including
carbon dioxide (CO2), fit within the definition of air pollutant in the
CAA.  The case arose from EPA’s denial of a petition for rulemaking
filed by more than a dozen environmental, renewable energy, and other
organizations requesting that EPA control emissions of GHGs from new
motor vehicles under section 202(a) of the CAA.  The Court found that,
in accordance with CAA section 202(a), EPA was required to determine
whether or not emissions of GHGs from new motor vehicles cause or
contribute to air pollution which may reasonably be anticipated to
endanger public health or welfare, or whether the science is too
uncertain to make a reasoned decision.  

On November 13, 2008, the Environmental Appeals Board (EAB) issued a
decision in a challenge to a PSD permit to construct a new electric
generating unit in Bonanza, Utah.  In re Deseret Power Electric
Cooperative, PSD Appeal No. 07-03 (EAB Nov. 13, 2008) (“Deseret”). 
The permit was issued by EPA Region 8 in August 2007 and did not include
best available control technology (BACT) limits for CO2.  At the time,
the Region acknowledged Massachusetts but found that decision alone did
not require PSD permits to include limits on CO2 emissions.  In briefs
filed in the EAB case, EPA maintained the position that the Agency had a
binding, historic interpretation of the phrase “subject to
regulation” in the federal PSD regulations that required PSD permit
limits to apply only to those pollutants already subject to actual
control of emissions under other provisions of the CAA.  Response of EPA
Office of Air and Radiation and Region 8 to Briefs of Petitioner and
Supporting Amici (filed March 21, 2008).  Accordingly, EPA argued that
the regulations contained in 40 CFR Part 75, which require monitoring of
CO2 at some sources, did not make CO2 subject to PSD regulation.  The
order and opinion issued by the EAB remanded the permit after finding
that prior EPA actions were insufficient to establish a historic,
binding interpretation that “subject to regulation” for PSD purposes
included only those pollutants subject to regulations that require
actual control of emissions.  However, the EAB also rejected arguments
that the CAA compelled only one interpretation of the phrase “subject
to regulation” and found “no evidence of a Congressional intent to
compel EPA to apply BACT to pollutants that are subject only to
monitoring and reporting requirements.”  Thus, the Board remanded the
permit to the Region to “reconsider whether or not to impose a CO2
BACT limit in light of the ‘subject to regulation’ definition under
the CAA.”  The Board encouraged EPA to consider “addressing the
interpretation of the phrase ‘subject to regulation under this Act’
in the context of an action of nationwide scope, rather than through
this specific permitting proceeding.”  See Deseret at 63-64.  

EPA issued the PSD Interpretive Memo shortly after the Deseret decision
with the stated purpose to “establish[] an interpretation clarifying
the scope of the EPA regulation that determines the pollutants subject
to the federal Prevention of Significant Deterioration (PSD) program
under the Clean Air Act (CAA or Act)” by providing EPA’s
“definitive interpretation” of the definition of the term
“regulated NSR pollutants” found at 40 CFR 52.21(b)(50) and
resolving “any ambiguity in subpart ([iv]) of that paragraph, which
includes ‘any pollutant that otherwise is subject to regulation under
the Act.’”  See Memo at 1.  As the Memo explains, the statute and
regulation use similar language – the regulation defines a regulated
NSR pollutant to include “[a]ny pollutant that otherwise is subject to
regulation under the Act” and requires BACT for “each regulated NSR
pollutant,” per 40 CFR 52.21(b)(50) and (j), while the Act requires
BACT for “each pollutant subject to regulation under this [Act],”
per CAA sections 165(a)(4) and 169.  The EAB had determined that “the
meaning of the term ‘subject to regulation under this Act’ as used
in [CAA] sections 165 and 169 is not so clear and unequivocal as to
preclude the Agency from exercising discretion in interpreting the
statutory phrase” in implementing the PSD program.  See Deseret at 63.

The PSD Interpretive Memo seeks to resolve the ambiguity in
implementation of the PSD program by stating that “EPA will interpret
this definition of ‘regulated NSR pollutant’ to exclude pollutants
for which EPA regulations only require monitoring or reporting but to
include each pollutant subject to either a provision in the Clean Air
Act or regulation adopted by EPA under the Clean Air Act that requires
actual control of emissions of that pollutant.”  The Memo states that
“EPA has not previously issued a definitive interpretation of the
definition of ‘regulated NSR pollutant’ in section 52.21(b)(50) or
an interpretation of the phrase ‘subject to regulation under the
Act’ that addressed whether monitoring and reporting requirements
constitute ‘regulation’ within the meaning of this phrase.”  The
Memo, however, explains that the interpretation reflects the
“considered judgment” of then-Administrator Johnson regarding the
PSD regulatory requirements and is consistent with both historic Agency
practice and prior statements by Agency officials.  See Memo at 1-2. 

The PSD Interpretive Memo is not a substantive rule promulgated under
section 307(d) of the CAA, but rather an interpretation of the terms of
a regulation at 40 CFR 52.21(b)(50).  An interpretive document is one
that explains or clarifies, and is consistent with, existing statutes or
regulation.  See National Family Planning and Reproductive Health
Ass’n v. Sullivan, 979 F.2d 227, 236-37 (D.C. Cir. 1992).  The PSD
Interpretive Memo explains and clarifies the meaning of the definition
of “regulated NSR pollutant” in section 52.21(b)(50) of the existing
NSR regulations, and does not alter the meaning of the definition in any
way that is inconsistent with the terms of the regulation.  As a result,
EPA concluded that the PSD interpretive Memo was an interpretive rule
that could be issued without a notice and comment rulemaking process. 

However, the PSD Interpretive Memo observed that the adoption of an
interpretation of a rule without a notice and comment process does not
preclude subsequent action by the Agency to solicit public input on the
interpretation.  Indeed, given the significant public interest in the
issue addressed in the December 18, 2008 memorandum, EPA subsequently
elected to seek public input on the memorandum and alternative readings
of the regulations.  

	On December 31, 2008, EPA received a petition for reconsideration of
the position taken in the PSD Interpretive Memo from Sierra Club and 14
other environmental, renewable energy, and citizen organizations.  See
Petition for Reconsideration, In the Matter of: EPA Final Action
Published at 73 FR 80300 (Dec. 31, 2008), entitled “Clean Air Act
Prevention of Significant Deterioration (PSD) Construction Permit
Program; Interpretation of Regulations That Determine Pollutants Covered
by the Federal PSD Permit Program.”  Petitioners argued that the PSD
Interpretive Memo “was impermissible as a matter of law, because it
was issued in violation of the procedural requirements of the
Administrative Procedures [sic] Act ... and the Clean Air Act..., it
directly conflicts with prior agency actions and interpretations, and it
purports to establish an interpretation of the Act that conflicts with
the plain language of the statute.”  See Petition at 2.  Accordingly,
Petitioners requested that EPA reconsider and retract the PSD
Interpretive Memo.  Petitioners later amended their Petition for
Reconsideration to include a request to stay the effect of the Memo
pending the outcome of the reconsideration request.  Amended Petition
for Reconsideration (filed Jan. 6, 2009).

On February 17, 2009, EPA granted the petition for reconsideration on
the PSD Interpretive Memo, on the basis of the authority conferred by
section 553(e) of the Administrative Procedure Act (APA), and announced
its intent to conduct a rulemaking to allow for public comment on the
issues raised in the Memo and on any issues raised by the EAB’s
Deseret opinion, to the extent they do not overlap with the issues
raised in the Memo.  Because the Memo was not a substantive rule
promulgated under section 307(d) of the APA, the reconsideration action
was not a reconsideration under the authority of section 307(d)(7)(B) of
the CAA.  See Letter from Lisa P. Jackson, EPA Administrator, to David
Bookbinder, Chief Climate Counsel at Sierra Club (Feb. 17, 2009).  EPA
did not stay the effectiveness of the PSD Interpretive Memo pending
reconsideration, but the Agency did reiterate that the Memo “does not
bind States issuing [PSD] permits under their own State Implementation
Plans.”  Id. at 1. 

On October 7, 2009 (74 FR 51535), EPA proposed a reconsideration of the
PSD Interpretive Memo that solicited comment on five possible
interpretations of the regulatory phrase “subject to regulation” –
the “actual control” interpretation (adopted by the Memo); the
“monitoring and reporting” interpretation (advocated by
Petitioners); the inclusion of regulatory requirements for specific
pollutants in SIPs (discussed in both the Memo and the Petition for
Reconsideration); an EPA finding of endangerment (discussed in the
Memo); and the grant of a section 209 waiver interpretation (raised by
commenters in another EPA action).  EPA also addressed, and requested
public comment on, other issues raised in the PSD Interpretive Memo and
related actions that may influence this reconsideration. 

Of the five interpretations described in the proposed reconsideration
notice, EPA expressly favored the actual control interpretation, which
has remained in effect since issuing the memorandum, notwithstanding the
EPA’s grant of reconsideration.  The proposal explained that the
actual control interpretation best reflects EPA’s past policy and
practice, is in keeping with the structure and language of the statute
and regulations, and best allows for the necessary coordination of
approaches to controlling emissions of newly identified pollutants. 
While the other interpretations may represent reasoned approaches for
interpreting “subject to regulation,” no particular one is compelled
by the statute, nor did the EAB determine that any one of them was so
compelled.  Because EPA had overarching concerns over the policy and
practical application of each of the alternative interpretations, the
Agency proposed to retain the actual control interpretation. 
Nevertheless, EPA requested comment on all five of the interpretations. 


III. This Action

A. 	Overview

EPA has made a final decision to continue applying (with one limited
refinement) the Agency’s existing interpretation of 40 CFR
52.21(b)(50) that is articulated in the PSD Interpretive Memo.  For
reasons explained below, and addressed in further detail in the document
“EPA’s Response to Public Comments on Reconsideration of the PSD
Interpretive Memo”, after reviewing the comments, EPA has concluded
that the “actual control interpretation” is a permissible
interpretation of the CAA and is the most appropriate interpretation to
apply given the policy implications.  However, EPA is refining its
interpretation in one respect to establish that PSD permitting
requirements apply to a newly regulated pollutant at the time a
regulatory requirement to control emissions of that pollutant “takes
effect” (rather than upon promulgation or the legal effective date of
the regulation containing such a requirement).  In addition, this notice
addresses several outstanding questions regarding the applicability of
the PSD and Title V permitting programs to GHGs upon the anticipated
promulgation of EPA regulations establishing limitations on emissions of
GHGs from vehicles under Title II of the CAA.

	EPA received 71 comments on the proposed reconsideration notice
published on October 7, 2009 (74 FR 51535).  Commenters represented a
range of interests, including state regulatory agencies, corporations
that may need to obtain PSD permits, trade associations representing
various industrial sectors, and environmental and public interest
groups.  Commenters representing states and regulated entities generally
expressed support for the actual control interpretation, while
environmental and public interest groups generally favored the
alternative interpretations.  States and regulated entities also
supported EPA’s proposed action to apply PSD requirements at the point
in time when an actual control requirement becomes effective, with many
entities specifically requesting that EPA interpret “effective” to
mean the compliance date of a rule.  Environmental stakeholders
supported retaining the position in the existing PSD Interpretive Memo
that PSD requirements apply to a pollutant upon the promulgation of the
relevant requirement for that pollutant. 

	EPA has not been persuaded that the Agency is compelled by the CAA, the
terms of EPA regulations, or prior EPA action to apply any of the four
alternatives to its preferred interpretation described in the October 7,
2009 notice – monitoring and reporting requirement, EPA-approved SIP,
endangerment finding, or CAA section 209 waiver.  EPA has likewise not
been persuaded that all of the alternative interpretations are precluded
by the CAA.  However, since Congress has not precisely spoken to this
issue, EPA has the discretion to choose among the range of permissible
interpretations of the statutory language.  Since EPA’s interpretation
of the regulations is not precluded by the statutory language, we are
electing to maintain that interpretation on policy grounds.  We have
concluded that the “actual control” interpretation is not only
consistent with decades of past practice, but provides the most
reasonable and workable approach to developing an appropriate regulatory
scheme to address newly identified pollutants of concern.  Thus, except
as to the one element that EPA proposed to modify, EPA is reaffirming
the PSD Interpretive Memo and its establishment of the actual control
interpretation as EPA’s definitive interpretation of the phrase
“subject to regulation” under the PSD provisions in the CAA and EPA
regulations.  

	EPA has been persuaded by public comments on the proposed
reconsideration to modify the portion of its interpretation regarding
the timing of when a pollutant becomes subject to regulation under the
CAA and thus covered by the requirements of the PSD permitting program. 
Specifically, EPA is modifying its interpretation of 40 CFR 52.21(b)(50)
of its regulations, and the parallel provision in 40 CFR 51.166(b)(49),
to establish that the PSD requirements will not apply to a newly
regulated pollutant until a regulatory requirement to control emissions
of that pollutant “takes effect.”  EPA has concluded that this
approach is consistent with the CAA and a reasonable reading of the
regulatory text.  

Based on these final determinations, EPA will continue to apply the
interpretation reflected in the PSD Interpretive Memo with one
refinement.  For the reasons discussed in more detail below, EPA has not
generally found cause to change the discussion or reasoning reflected in
the Memo.  As a result, EPA does not see a need to either withdraw or
re-issue the Memo.  However, this notice refines one paragraph of that
memorandum to reflect EPA’s current view that a pollutant becomes
subject to regulation at the time the first control requirements
applicable to a pollutant take effect.  Public comments raised several
questions regarding the application of the PSD program and Title V
permits to GHGs that EPA did not specifically raise in the October 7,
2009 proposed notice of reconsideration.  Some of these comments raised
significant issues that the Agency recognizes the need to address at
this time to ensure the orderly transition to the regulation of GHGs
under these permitting programs.  Thus, this notice reflects additional
interpretations and EPA statements of policy on topics not discussed in
the October 7, 2009 notice.  These interpretations and polices have been
developed after careful consideration of the public comments submitted
to EPA on this action and related matters.  In subsequent actions, EPA
may address additional topics raised in public comments on this action
that the Agency did not consider necessary to address at this time. 

	Regarding GHGs, EPA has concluded that PSD program requirements will
apply to GHGs upon the date that the anticipated tailpipe standards for
light-duty vehicles (known as the “LDV Rule”) take effect.  Based on
the proposed LDV Rule, those standards will take effect when the 2012
model year begins, which is no earlier than January 2, 2011.  While the
LDV Rule will become “effective” for the purposes of planning for
the upcoming model years as of 60 days following publication of the
rule, the emissions control requirements in the rule do not “take
effect” – i.e., requiring compliance through vehicular certification
before introducing any Model Year 2012 into commerce – until Jan. 2,
2011, or approximately 9 months after the planned promulgation of the
LDV Rule.  Furthermore, as EPA intends to explain soon in detail in the
final action on the PSD and Title V GHG Tailoring Rule (known as the
“Tailoring Rule”), in light of the significant administrative
challenges presented by the application of the PSD and Title V
requirements for GHGs (and considering the legislative intent of the PSD
and Title V statutory provisions), it is necessary to defer applying the
PSD and Title V provisions for sources that are major based only on
emissions of GHGs until a date that extends beyond January 2, 2011. 

B. 	Analysis of Proposed and Alternative Interpretations for Subject to
Regulation

1.  	Actual Control Interpretation

EPA has concluded that the “actual control” interpretation (as
articulated in the PSD Interpretive Memo) is permissible under the CAA
and is preferred on policy grounds.  Thus, EPA will continue to
interpret the definition of “regulated NSR pollutant” in 40 CFR
52.21(b)(50) to exclude pollutants for which EPA regulations only
require monitoring or reporting but to include each pollutant subject to
either a provision in the CAA or regulation promulgated by EPA under the
CAA that requires actual control of emissions of that pollutant.  As
discussed further below, EPA will also interpret section 51.166(b)(49)
of its regulations in this manner.  This interpretation is supported by
the language and structure of the regulations and is consistent with
past practice in the PSD program and prior EPA statements regarding
pollutants subject to the PSD program.  The CAA is most effectively
implemented by making PSD emissions limitations applicable to pollutants
after a considered judgment by EPA (or Congress) that particular
pollutants should be subject to control or limitation.  The actual
control interpretation promotes the orderly administration of the
permitting program by allowing the Agency to first assess whether there
is a justification for controlling emissions of a particular pollutant
under relevant criteria in the Act before applying the requirements of
the PSD permitting program to a pollutant. 

Because the term “regulation” is susceptible to more than one
meaning, there is ambiguity in the phrase “each pollutant subject to
regulation under the Act” that is used in both sections 165(a)(4) and
169(3) of the CAA.  As discussed in the Memo, the term “regulation”
can be used to describe a rule contained in a legal code, such as the
Code of Federal Regulations, or the act or process of controlling or
restricting an activity.  The primary meaning of the term
“regulation” in Black’s Law Dictionary (8th Ed.)  TA \l "Black’s
Law Dictionary (8th Ed.)" \s "Black’s" \c 3   is “the act or process
of controlling by rule or restriction.”  However, an alternative
meaning in this same dictionary defines the term as “a rule or order,
having legal force, usu. issued by an administrative agency or local
government.”  The primary meaning in Webster’s dictionary for the
term “regulation” is “the act of regulating: the state of being
regulated.”  Merriam-Webster’s Collegiate Dictionary 983 (10th Ed.
2001).  Webster’s secondary meaning is “an authoritative rule
dealing with details of procedure” or “a rule or order issued by an
executive authority or regulatory agency of a government and having the
force of law.”  Webster’s also defines the term “regulate” and
the inflected forms “regulated” and “regulating” (both of which
are used in Webster’s definition of “regulation”) as meaning “to
govern or direct according to rule” or to “to bring under the
control of law or constituted authority.”  Id. 

The PSD Interpretive Memo reasonably applies a common meaning of the
term “regulation” to support a permissible interpretation that the
phrase “pollutant subject to regulation” means a pollutant subject
to a provision in the CAA or a regulation issued by EPA under the Act
that requires actual control of emissions of that pollutant.  Public
comments have not demonstrated the dictionary meanings of the term
“regulation” described in the Memo are no longer accepted meanings
of this term.  In light of the different meanings of the term
“regulation,” EPA has not been persuaded by public comments that the
CAA plainly and unambiguously requires that EPA apply any of the other
interpretations described in the October 7, 2009 notice.  Moreover, the
Memo carefully explains how the actual control interpretation is
consistent with the overall context of the CAA in which sections
165(a)(4) and 169(3) are found.  After consideration of public comment,
EPA continues to find this discussion persuasive.  The “subject to
regulation” language appears in the BACT provisions of the Act, which
themselves require actual controls on emissions.  The BACT provisions
reference the New Source Performance Standards (NSPS) and other control
requirements under the Act, which establish a floor for the BACT
requirement.  See 42 U.S.C. §7479(3).  Other provisions in the CAA that
authorize EPA to establish emissions limitations or controls on
emissions provide criteria for the exercise of EPA’s judgment to
determine which pollutants or source categories to regulate.  Thus, it
follows that Congress expected that pollutants would only be regulated
for purposes of the PSD program after:  (1) the EPA promulgated
regulations requiring control of a particular pollutants on the basis of
considered judgment, taking into account the applicable criteria in the
CAA, or (2) EPA promulgates regulations on the basis of Congressional
mandate that EPA establish controls on emissions of a particular
pollutant, or (3) Congress itself directly imposes actual controls on
emissions of a particular pollutant.  In addition, considering other
sections in the Act that require reasoned decision-making and authorize
the collection of emissions data prior to establishing controls on
emissions, it is also consistent with the Congressional design to
require BACT limitations for pollutants after a period of data
collection and study that leads to a reasoned decision to establish
control requirements.  Public commenters did not demonstrate that it was
erroneous for EPA to interpret the PSD provisions in this manner, based
on the context of the Act. 

Furthermore, the actual control interpretation is consistent with the
terms of the regulations EPA promulgated in 2002.  EPA continues to find
the reasoning of the PSD Interpretive Memo to be persuasive.  The
structure and language of EPA’s definition of ‘regulated NSR
pollutant’ at 40 CFR 52.21(b)(50) supports the actual control
interpretation.  The first three parts of the definition describe
pollutants that are subject to regulatory requirements that mandate
control or limitation of the emissions of those pollutants, which
suggests that the use of “otherwise subject to regulation” in the
fourth prong of the definition also intended some prerequisite act or
process of control.  The definition’s use of “subject to
regulation” should be read in light of the primary meanings of
“regulation” described above, which each use or incorporate the
concept of control. 

 One commenter stated that EPA’s suggestion that its proposed
interpretation will allow for a more practical approach to determining
whether emissions of air pollutants endanger health and human welfare
amounts only to a policy preference.  The commenter argued that EPA’s
policy preference should be subordinate to statutory language and
Congressional intent.  Another commenter made similar comments and
stated that EPA cannot avail itself of additional, non-statutory de
facto extensions of time to fulfill its statutory obligations.  

Where the governing statutory authority is susceptible to more than one
interpretation, it is not impermissible for EPA to apply policy
preferences when determining which interpretation to apply, so long as
the interpretation EPA elects to follow is a permissible one.  The PSD
Interpretive Memo provides a persuasive explanation for why the
interpretation reflected in that memorandum is consistent with the terms
of the CAA and Congressional intent.  In this instance, EPA’s policy
preferences are fully consistent with that intent.  As explained above,
Congress intended for EPA to gather data before establishing controls on
emissions and to make reasoned decisions.

EPA continues to prefer the actual control interpretation because it
ensures an orderly and manageable process for incorporating new
pollutants into the PSD program after an opportunity for public
participation in the decision making process.  Several commenters who
supported EPA’s proposal to continue applying the “actual control”
interpretation identified these considerations as important reasons that
EPA should continue doing so.  EPA agrees with these comments.  As
discussed persuasively in the PSD Interpretive Memo, under this
interpretation, EPA may first assess whether there is a justification
for controlling emissions of a particular pollutant under relevant
criteria in the Act before imposing controls on a pollutant under the
PSD program.  In addition, this interpretation permits the Agency to
provide notice to the public and an opportunity to comment when a new
pollutant is proposed to be regulated under one or more programs in the
Act.  It also promotes the orderly administration of the permitting
program by providing an opportunity for EPA to develop regulations to
manage the incorporation of a new pollutant into the PSD program, for
example, by promulgating a significant emissions rate (or de minimis
level) for the pollutant when it becomes regulated.  See 40 CFR
52.21(b)(23)  TA \l "40 C.F.R. § 52.21(b)(23)" \s "40 C.F.R. §
52.21(b)(23)" \c 6  .  Furthermore, this interpretation preserves the
Agency’s ability to gather data on pollutant emissions to inform their
judgment regarding the need to establish controls on emissions without
automatically triggering such controls.  This interpretation preserves
EPA’s authority to require control of particular pollutants through
emissions limitations or other restrictions under various provisions of
the Act, which would then trigger the requirements of the PSD program
for any pollutant addressed in such an action.  

	Some commenters who opposed the actual control interpretation argued
that this deliberate approach leads to “analysis paralysis” and is
subject to political manipulation.  The commenter further noted that the
case-by-case BACT requirement does not contemplate waiting years for EPA
to conduct analyses and “develop” control options; rather, BACT must
be based on control options that are available.  Then, permitting
agencies are to make “case-by-case” determinations “taking into
account energy, environmental, and economic impacts and other costs,”
thereby ensuring that the decision is informed by the available
solutions, their efficacy and costs.

	While this analysis may sometimes take more time than the commenter
would prefer, a deliberative and orderly approach to regulation is in
the public interest and consistent with Congressional intent.  It would
be premature to impose the BACT requirement on a particular pollutant if
neither EPA nor Congress has made a considered judgment that a
particular pollutant is harmful to public health and welfare and merits
control.  

Once the Agency has made a determination that a pollutant should be
controlled using one or more of the regulatory tools provided in the CAA
and those controls take effect, EPA agrees that a BACT analysis must
then be completed based on available information.  As the commenter
points out, the BACT process is designed to determine the most effective
control strategies achievable in each instance, considering energy,
environmental, and economic impacts.  Thus, EPA agrees that the onset of
the BACT requirement should not be delayed in order for technology or
control strategies to be developed.  Furthermore, EPA agrees with the
commenter that delaying the application of BACT to enable development of
guidance on control strategies is not necessarily consistent with the
BACT requirement.  The BACT provisions clearly contemplate that the
permitting authority will develop control strategies on a case-by-case
basis.  Thus, EPA is not in this final action relying on the need to
develop guidance for BACT as a justification for choosing to continue
applying the actual control interpretation.  However, in the absence of
guidance on control strategies from EPA and other regulatory agencies,
the BACT process may be more time and resource intensive when applied to
a new pollutant.  Under a mature PSD permitting program, successive BACT
analyses establish guidelines and precedents for subsequent BACT
determinations.  However, when a new pollutant is regulated, the first
permit applicants and permitting authorities that are faced with
determining BACT for a new pollutant must invest more time and resources
in making an assessment of BACT under the statutory criteria.  Given the
potentially large number of sources that could be subject to the BACT
requirement when EPA regulates GHGs, the absence of guidance on BACT
determinations for GHGs presents a unique challenge for permit
applicants and permitting authorities.  EPA intends to address this
challenge in part by deferring, under the Tailoring Rule, the
applicability of the PSD permitting program for sources that would
become major based solely on GHG emissions.  EPA is also developing
guidance on BACT for GHGs. 

Several commenters expressed concern with EPA’s explanation that the
actual control interpretation best reflects EPA’s past practice.  One
commenter argued that the Deseret decision rejects the idea that “past
policy and practice” is a sufficient justification for EPA’s
preferred interpretation.  In addition, several commenters argued that
the memorandum was in fact not consistent with past EPA practice, based
on their interpretation of a statement made in the preamble to a rule
which promulgated PSD regulations in 1978.  

While the record continues to show that the actual control
interpretation is consistent with EPA’s historic practice, EPA agrees
that continuity with past practice alone does not justify maintaining a
position when there is good cause to change it.  In this case, however,
EPA has not found cause to change an interpretation that is consistent
with Congressional intent and supported by the policy considerations
described earlier.  Thus, EPA is not retaining the actual control
interpretation simply to maintain continuity with historic practice. 
The record reflects that EPA’s past practice was grounded in a
permissible interpretation of the law and supported by rational policy
considerations.  Commenters have not otherwise persuaded EPA to change
its historic practice in this area. 

	A review of numerous federal PSD permits shows that EPA has been
applying the actual control interpretation in practice – issuing
permits that only contained emissions limitations for pollutants subject
to regulations requiring actual control of emissions under other
portions of the Act.  Furthermore, in 1998, well after promulgation of
the initial CO2 monitoring regulations in 1993, EPA’s General Counsel
concluded that CO2 would qualify as an “air pollutant” that EPA had
the authority to regulate under the CAA, but the General Counsel also
observed that “the Administrator has made no determination to date to
exercise that authority under the specific criteria provided under any
provision of the Act.”  The 1978 Federal Register notice promulgating
the initial PSD regulations stated that pollutants “subject to
regulation” in the PSD program included “any pollutant regulated in
Subchapter C of Title 40 of the Code of Federal Regulations.” 
Commenters argue this statement illustrates that EPA has in fact applied
the PSD BACT requirement to any pollutant subject to only a monitoring
requirement codified in this portion of the Code of Federal Regulations.
 However, this comment overlooked the discussion in the PSD Interpretive
Memo regarding the differing meanings of the term “regulation” and
“regulate.”  The 1978 preamble did not amplify the meaning of the
term “regulated in.”  Thus, commenters have not demonstrated that
EPA had concluded in 1978 that monitoring requirements equaled
“regulation” within the meaning of sections 165(a)(4) and 169(3) of
the CAA, nor have commenters provided any examples of permits issued by
EPA after 1978 that demonstrate EPA’s interpretation was inconsistent
with the practice described in the PSD Interpretive Memo.

Therefore, we affirm that the actual control interpretation expressed in
the PSD Interpretive Memo continues be the operative statement for the
EPA interpretation of the meaning of the regulatory phrase “subject to
regulation” within the federal PSD rules.  

2.  	Monitoring and Reporting Interpretation 

EPA is not persuaded that the monitoring and reporting interpretation is
compelled by the CAA, and the Agency remains concerned that application
of this approach would lead to odd results and make the PSD program
difficult to administer.  EPA continues to find the reasoning of the PSD
Interpretive Memo persuasive.  

The monitoring and reporting interpretation would make the substantive
requirements of the PSD program applicable to particular pollutants
based solely on monitoring and reporting requirements (contained in
regulations established under section 114 or other authority in the
Act).  This approach would lead to the perverse result of requiring
emissions limitations under the PSD program while the Agency is still
gathering the information necessary to conduct research or evaluate
whether to establish controls on the pollutant under other parts of the
Act.  Such a result would frustrate the Agency’s ability to gather
information using section 114 and other authority and make informed and
reasoned judgments about the need to establish controls or limitations
for particular pollutants.  If EPA interpreted the requirement to
establish emissions limitations based on BACT to apply solely on the
basis of a regulation that requires collecting and reporting emissions
data, the mere act of gathering information would essentially dictate
the result of the decision that the information is being gathered to
inform (whether or not to require control of a pollutant).  Many
commenters representing state permitting agencies and industry groups
agree with the policy arguments advanced by EPA and others that EPA’s
critical information gathering activities will be constrained, with
likely adverse environmental and public health consequences, if
monitoring requirements are necessarily associated with the potentially
significant implementation and compliance costs and resource constraints
of the PSD program.  Commenters expressed concern that without the
ability to gather data or investigate unregulated pollutants, for fear
of triggering automatic regulation under the CAA, EPA will not have the
flexibility to review the validity of controlling new pollutants. 

We agree that a monitoring and reporting interpretation would hamper the
Agency’s ability to conduct monitoring or reporting for investigative
purposes to inform future rulemakings involving actual emissions control
or limits.  In addition, it is not always possible to predict when a new
pollutant will emerge as a candidate for regulation.  In such cases, the
Memo’s reasoning is correct in that we would be unable to promulgate
any monitoring or reporting rule for such a pollutant without triggering
PSD under this interpretation.  

An environmental organization disagreed with our proposed notice of
reconsideration, and commented that EPA has issued monitoring and
reporting regulations for CO2 in 40 CFR 75, promulgated pursuant to
section 821 of the 1990 CAA Amendments.  The commenter felt that these
monitoring and reporting rules are “regulation” in that they are
contained in a legal code, have the force of law, and bring the subject
matter under the control of law and the EPA.  Furthermore, the commenter
says that EPA itself has characterized these monitoring and reporting
requirements as “regulations.”  In contrast, another commenter
argued that an agency’s interpretation of a statute should focus first
on the ordinary dictionary meaning of the terms used and that monitoring
emissions does not fit within any of the types of activities understood
to constitute “regulation” of those emissions in the ordinary
meaning of that term.  Each of these commenters focuses on only one of
the two potential meanings of the term “regulation” described above.


The commenter that favors the “monitoring and reporting”
interpretation appears to focus only on the dictionary meanings that
describe a rule contained in a legal code.  The commenter has not
demonstrated that it is impermissible for EPA to construe the CAA on the
basis of another common meaning of the term “regulation.”  In the
context of construing the Act, the EAB observed in the Deseret case that
a plain meaning could not be ascertained from looking solely at the word
“regulation.”  The Board reached this conclusion after considering
the dictionary definitions of the term “regulation” cited above. 
See Deseret slip op. at 28-29.  EPA continues to find the reasoning of
the EAB and the PSD Interpretive Memo to be persuasive.  The EAB found
“no evidence of Congressional intent to compel EPA to apply BACT to
pollutants that are subject only monitoring and reporting
requirements.”  See Deseret at 63.  

Comments have not convincingly shown that Congress clearly intended to
use the term “regulation” in section 165(a)(4) and 169(3) to
describe any type of rule in a legal code.  Some commenters presented
alternative theories of Congressional intent regarding the BACT
provisions, but they have not persuasively demonstrated that the
interpretation of Congressional intent based on the context of the CAA
described in the PSD Interpretive Memo is erroneous.  

For example, one commenter opposed to EPA’s proposed action commented
that the PSD Interpretive Memo ignores the Congressionally-established
purpose of PSD to protect public health and welfare from actual and
potential adverse effects.  See CAA §160(1).  Specifically, this
commenter stated that to limit application of BACT until after control
requirements are in place following an endangerment finding ignores the
broad, protective purpose of the PSD program.  The commenter said that
the emphasis on “potential adverse effect[s]” distinguishes PSD the
requirement from the National Ambient Air Quality Standards (NAAQS) and
NSPS programs, which require that EPA make an endangerment finding
before establishing generally applicable standards such as the NSPS or
motor vehicle emissions standards.  According to this commenter,
BACT’s case-by-case approach provides the dynamic flexibility
necessary to implement an emission limitation appropriate to each
particular source.  This commenter feels that the PSD program’s
ability to address potential adverse effects is hindered by the position
that an endangerment determination and actual control limits must be
first established.  

EPA does not agree that the terms of section 160 cited by the commenter
compel EPA to read sections 165(a)(4) and 169(3) to apply to a pollutant
before the Agency has established control requirements for the
pollutant.  Section 160(1) describes PSD’s purpose to “protect
public health and welfare from any actual or potential adverse effect
which in the Administrator’s judgment may reasonably be anticipated to
occur from air pollution.”  Thus, this goal contemplates an exercise
of judgment by EPA to determine that an actual or potential adverse
effect may reasonably be anticipated from air pollution.  In that sense,
this goal is consistent with NAAQS and NSPS programs, which contemplate
that regulation of a pollutant will not occur until a considered
judgment by EPA that a substance or source category merits control or
restriction.  The commenter has not persuasively established that the
“potential adverse effect” language in section 160(1) makes this
provision markedly different than the language used in sections
108(a)(1)(A) and 111(b)(1)(A).  All three sections use the phrase “may
reasonably be anticipated.” Furthermore, section 160 contains general
goals and purposes and does not contain explicit regulatory
requirements.  The controlling language in the PSD provisions is the
“subject to regulation” language in sections 165(a)(4) and 169(3). 
As discussed earlier, the “actual control” interpretation is based
on a common and accepted meaning of the term “regulation.”  To the
extent the goals and purpose in section 160 are instructive as to the
meaning of other provisions in Part C of the Act, section 160(1) is just
one of several purposes of the PSD program that Congress specified.  The
Act also instructs EPA to ensure that economic growth occurs consistent
with the preservation of existing clean air resources.  CAA §160(3). 
EPA’s interpretation is consistent with this goal because it allows
EPA to look at the larger picture by coordinating control of an air
pollutant under the PSD program with control under other CAA provisions.
 

EPA finds the logic of the PSD Interpretive Memo more persuasive.  The
Memo considers the full context of the CAA, including the health and
welfare criteria that generally must be satisfied to establish control
requirements under other parts of the Act, information gathering
provisions that contemplate data collection and study before pollutants
are controlled, and requirements for reasoned decision making.  While
some commenters presented arguments for why it might be possible or
beneficial to apply the BACT requirement before a control requirement is
established for a pollutant elsewhere under the Act, these arguments do
not demonstrate that the contextual reading of the CAA described in the
Memo is erroneous.  Thus, the comments have at most provided another
permissible reading of the Act, but they do not demonstrate that EPA
must require BACT limitations for pollutants that are not yet controlled
but only subject to data collection and study. 

	We continue to believe that the monitoring and reporting interpretation
is inconsistent with past agency practice because, as the Memo notes,
“EPA has not issued PSD permits containing emissions limitations for
pollutants that are only subject to monitoring and reporting
requirements,” including CO2 emissions.  Further, the Memo determines
that the monitoring and reporting interpretation is not required under
the 1978 preamble language, explaining that the preamble language could
be interpreted in a variety of ways and “did not specifically address
the issue of whether a monitoring or reporting requirement makes a
pollutant ‘regulated in’ [Subpart C of Title 40] of the Code of
Federal Regulations.”  See Memo at 11-12.  Commenters have not
demonstrated that the Agency specifically intended, through this
statement, to apply the PSD requirements to pollutants that were covered
by only a monitoring and reporting requirement codified in this part of
the CFR. 

	One commenter questioned EPA’s basis for rejecting the monitoring and
reporting interpretation because they believe EPA has not identified a
pollutant other than CO2 that would be affected by the monitoring and
reporting interpretation.  However, EPA’s GHG Reporting Rule covers
six GHGs, not just CO2.  Further, EPA has promulgated regulations that
require monitoring of oxygen (O2) in the stack of a boiler under certain
circumstances.  See 40 CFR 60.49Da(d).  These examples help demonstrate
why monitoring and reporting requirements alone should not be
interpreted to trigger PSD and BACT requirements.

For the reasons discussed above, we affirm the Memo’s rejection of the
monitoring and reporting interpretation for triggering PSD requirements
for a new pollutant.

3.  	State Implementation Plan (SIP) Interpretation 

In discussing the application of the actual control interpretation to
specific actions under the CAA, the PSD Interpretive Memo rejects an
interpretation of “subject to regulation” in which regulatory
requirements for a particular pollutant in the EPA-Approved State
Implementation Plan (SIP) for a single state would “require regulation
of that pollutant under the PSD program nationally.”  (Hereinafter,
referred to as the “SIP interpretation.”)  In this action, we affirm
and supplement our rationale for rejecting the SIP interpretation
provided in the PSD Interpretive Memo and our reconsideration proposal. 
Since the meaning of the term “subject to regulation” is ambiguous
and susceptible to multiple interpretations, the SIP interpretation is
not compelled by the structure and language of the Act.  Furthermore,
there would be negative policy implications if EPA adopted this
interpretation.  

The Memo reasons that application of the SIP interpretation would
convert EPA’s approval of regulations applicable only in one state
into a decision to regulate a pollutant on a nationwide scale for
purposes of the PSD program.  The Memo explains that the establishment
of SIPs is better read in light of the “cooperative federalism”
underlying the Act, whereby Congress allowed individual states to create
and apply some regulations more stringently than federal regulations
within its borders, without allowing individual states to set national
regulations that would impose those requirements on all states.  See
Ellis v. Gallatin Steel Co., 390 F.3d 461, 467 (6th Cir. 2004).  In
rejecting the SIP interpretation, the Memo also explains that we adopted
a similar position in EPA’s promulgation of the NSR regulations for
fine particulate matter (or “PM2.5”), without any public comments
opposing that position.  See Memo at 15-16. 

EPA continues to believe that the CAA and our implementing regulations
are intended to provide states flexibility to develop and implement SIPs
to meet the air quality goals of their individual state.  Each state’s
implementation plan is a reflection of the air quality concerns in that
state, allowing a state significant latitude in the treatment of
specific pollutants of concern (or their precursors) within its borders
based on air quality, economic, and other environmental concerns of that
state.  As such, pollutant emissions in one state may not present the
same problem for a state a thousand miles away.  As expressed in the PSD
Interpretive Memo, we continue to have concerns that the SIP
interpretation would improperly limit the flexibility of states to
develop and implement their own air quality plans, because the act of
one state to establish regulatory requirements for a particular
pollutant would drive national policy.  If EPA determined that a new
pollutant becomes “subject to regulation” nationally within the
meaning of section 165 based solely on the provisions of EPA-approved
SIP, then all states would be required to subject the new pollutant to
PSD permitting whether or not control of the air pollutant was relevant
for improving that state’s air quality.  Whether one state, five
states, or 45 states make the decision that their air quality concerns
are best addressed by imposing regulations on a new pollutant, we do not
think those actions should trump the cooperative federalism inherent in
the CAA.  While several states may face similar air quality issues and
may choose regulation as the preferred approach to dealing with a
particular pollutant, we are concerned that allowing the regulatory
choices of some number of states to impose PSD regulation on all other
states would do just that. 

Some commenters support the SIP interpretation, and fault the Agency’s
rejection of the interpretation by stating that neither the Act, nor the
Memo, provides a basis for a position that regulation by a single state
is not enough to constitute “regulation under the Act” on a
nationwide basis for purpose of section 165.  Petitioners and another
commenter also assert that CO2 is already “subject to regulation under
the Act” and take the position that any requirement EPA adopts and
approves in an implementation plan makes the covered pollutant
“subject to regulation under the Act” because it is approved by the
EPA “under the Act,” and because it becomes enforceable by the
state, by EPA and by citizens “under the Act” upon approval.  

 EPA disagrees with the Petitioner and with this commenter that this
reasoning necessarily means that a pollutant regulated in one SIP
approved by EPA must automatically be regulated through the PSD program
nationally.  In fact, Congress demonstrated intent, in the language and
structure of the Act, for SIP requirements to have only a local or
regional effect. 

In section 102(a) of the CAA, Congress directs EPA to encourage
cooperative activities among states, and the adoption of uniform state
and local laws for the control of air pollution “as practicable in
light of the varying conditions and needs.”  This language informs the
issue of whether SIP requirements have nationwide applicability in two
ways.  First, there would be no need for EPA to facilitate uniform
adoption of standards in different air quality control regions, if the
regulation of an air pollutant by one region would automatically cause
that pollutant to be regulated in another region.  Second, Congress
bounded its desire to promote uniformity by recognizing that addressing
local air quality concerns may preempt national uniformity of
regulation.  

Indeed, section 116 of the CAA grants States the right to adopt more
stringent standards than the uniform, minimum requirements set forth by
EPA.  See 42 U.S.C. 7416.  The legislative history of the 1977 CAA
Amendments shows that Congress understood that States may adopt
different and more stringent standards then the federal minimum
requirements.  See, e.g., 122 Cong. Rec. S12456 (daily ed. July 26,
1976) (statement of Sen. Randolph) (“[T]he States are given latitude
in devising their own approaches to air pollution control within the
framework of broad goals. ... The State of West Virginia has established
more stringent requirements than those which, through the Environmental
Protection Agency, are considered as adequate...”); 122 Cong. Rec.
S12458 (daily ed. July 26, 1976) (statement of Sen. Scott) (“The
states have the right, however, to require higher standards, and they
should have under the police powers.”)  Congress could not have
intended states to have latitude to implement their own approaches to
air pollution control, and simultaneously, require that air pollutants
regulated by one State automatically apply in all other States.

Importantly, the legislative history also shows that Congress intended
to limit the EPA’s ability to disapprove a State’s decision to adopt
more stringent requirements in setting forth the criteria for approving
state submissions under section 110.  This intent is supported by the
following passage:

State implementation plans usually contain a unified set of requirements
and frequently do not make distinctions between the controls needed to
achieve one kind of ambient standard or another.  To try to separate
such emission limitations and make judgments as to which are necessary
to achieving the national ambient air quality standards assumes a
greater technical capability in relating emissions to ambient air
quality than actually exists.

A federal effort to inject a judgment of this kind would be an
unreasonable intrusion into protected State authority.  EPA’s role is
to determine whether or not a State’s limitations are adequate and
that State implementation plans are consistent with the statute.  Even
if a State adopts limits which may be stricter than EPA would require,
EPA cannot second guess the State judgment and must enforce the approved
State emission limit.

123 Cong. Rec. S9167 (daily ed. June 8, 1977) (statement of Sen.
Muskie). 

	This Congressional intent is reflected within the statutory language. 
Under section 110(k)(3), the EPA Administrator “shall approve” a
state’s submittal if it meets the requirements of the Act, and under
section 110(l) “shall not” approve a plan revision “if the
revision would interfere with any other applicable requirement of this
Act.”  Courts have similarly interpreted this language to limit
EPA’s discretion to approve or disapprove SIP requirements.  See,
e.g., State of Connecticut v. EPA, 656 F.2d 902, 906 (2d. Cir. 1981)
(“As is illustrated by Congress’s use of the word ‘shall,’
approval of an SIP revision by the EPA Administrator is mandatory if the
revision has been the subject of a proper hearing and the plan as a
whole continues to adhere to the requirements of section 110(a)(2)”)
(referencing Union Electric Co. v. EPA, 427 U.S. 246, 257 (1976); and
Mission Indus., Inc. v. EPA, 547 F.2d 123 (1st Cir. 1976)).  These
provisions of the statute do not establish any authority or criteria for
EPA to judge the approvability of a state’s submission based on the
implications such approval would have nationally.  The absence of such
authority or criteria in the applicable standard argues against
nationwide applicability of SIP requirements and the SIP interpretation.

Moreover, under section 307(b) of the CAA, Congress assigns review of
specific regulations promulgated by EPA and “any other nationally
applicable regulations promulgated or final action taken, by the
Administrator under this Act” only to the U.S. Court of Appeals for
the District of Columbia Circuit (“D.C. Circuit”).  In contrast,
“the Administrator’s action in approving and promulgating any
implementation plan under Section 110 ... or any other final action of
the Administrator under this Act ... which is local or regionally
applicable may be filed only in the United States Court of Appeals for
the appropriate circuit.”  42 U.S.C. §7607(b) (emphasis added). 
Thus, Congress set forth its intended applicability of these regulations
in assigning judicial venue and clearly articulated that requirements in
a SIP are generally “local or regionally applicable.”  

Even if the Act could be read to support EPA review of the national
implications of state SIP submissions, such an approach would be
undesirable for policy reasons.  As we highlighted in our
reconsideration notice, one practical effect of allowing state-specific
concerns to create national regulation is that EPA’s review of SIPs
would likely be much more time-consuming, because we would have to
consider each nuance of the SIP as a potential statement of national
policy.  Thus, we would have heightened oversight of air quality actions
in all states – even those regarding local and state issues that are
best decided by local agencies.  Our approval of SIPs would be delayed,
which would in turn, delay state’s progress toward improving air
quality.  And, EPA would be required to defend challenges to the
approval of a SIP with national implications in the D.C. Circuit Court
of Appeals rather than the local Circuit Court of Appeals.  The
potential increased burden of reviewing and approving SIPs to analyze
the national implications of each SIP, and the associated delay in
improving air quality, creates a compelling policy argument against
adoption of the SIP interpretation.

Petitioners also fault EPA’s reliance on Connecticut v. EPA, 656 F.2d
902 (2d Cir. 1981) and assert that this case has nothing to do with the
issue of whether a pollutant is “subject to regulation under the
Act.”  In the PSD Interpretive Memo, EPA cited Connecticut to support
the notion that while a state is free to adopt air quality standards
more stringent than required by the NAAQS or other federal law
provisions, Congress precludes those stricter requirements from applying
to other states.  The Agency agrees with commenter that the
circumstances involved in that case are not directly analogous, but,
nevertheless, the case supports the inference that EPA has drawn from
it.  The Court concluded that “[n]othing in the Act, however,
indicates that a state must respect its neighbor's air quality standards
(or design its SIP to avoid interference therewith) if those standards
are more stringent than the requirements of federal law.”  If a state
is not required to respect the more stringent requirements of a
neighboring State in developing its own implementation plan, then by
inference, the state would also not be compelled to follow the more
stringent standards. 

In sum, after reconsidering the legal and policy issues, we decline to
adopt the SIP interpretation.

4. 	Endangerment Finding Interpretation 

The PSD Interpretive Memo states that the fourth part of the regulated
NSR pollutant definition (“[a]ny pollutant that otherwise is subject
to regulation”) should not be interpreted “to apply at the time of
an endangerment finding.”  See Memo at 14 (hereinafter, referred to as
the “endangerment finding interpretation.”).  After considering
public comments, EPA is affirming the position expressed in the PSD
Interpretative Memo that an endangerment finding alone does not make the
requirements of the PSD program applicable to a pollutant.  EPA
maintains its view that the terms of EPA’s regulations and the
relevant provisions of the CAA do not compel EPA to conclude that an air
pollutant becomes “subject to regulation” when EPA finds that it
endangers public health or welfare without contemporaneously
promulgating control requirements for that pollutant. 

As explained in EPA’s Endangerment and Cause or Contribute Findings
for GHGs under section 202(a) of the CAA, there are actually two
separate findings involved in what is often referred to as an
endangerment finding.  74 FR 66496 (Dec. 15, 2009).  The first finding
addresses whether air pollution may reasonably be anticipated to
endanger public health or welfare.  The second finding involves an
assessment of whether emissions of an air pollutant from the relevant
source category cause or contribute to this air pollution.  In this
notice, we use the phrase “endangerment finding” to refer to EPA
findings on both of these questions.  The EPA interpretation described
here applies to both findings regardless of whether they occur together
or separately.  

As explained in the proposed reconsideration, an interpretation of
“subject to regulation” that does not include endangerment findings
is consistent with the first three parts of the definition of
“regulated NSR pollutant” in section 52.21(b)(50) of EPA’s
regulations.  Unlike the first three parts of the definition, an
endangerment finding does not itself contain any restrictions (e.g.,
regarding the level of air pollution or emissions or use).  Moreover,
two parts of the definition involve actions that can occur only after an
endangerment finding of some sort has taken place.  In other words,
other parts of the definition already bypass an endangerment finding and
apply the PSD trigger to a later step in the regulatory process.  

Specifically, under the first part of that definition, PSD regulation is
triggered by promulgation of a NAAQS under CAA section 109.  However, in
order to promulgate NAAQS standards under section 109, EPA must first
list, and issue air quality criteria for a pollutant under section 108,
which in turn can only happen after EPA makes an endangerment finding
and a version of a cause or contribute finding, in addition to meeting
other requirements.  CAA sections 108(a)(1) and 109(a)(2).  Thus, if EPA
were to conclude that an endangerment finding, cause or contribute
finding, or both would make a pollutant “subject to regulation”
within the meaning of the PSD provisions, this would read all meaning
out of the first part of the “regulated NSR pollutant” definition
because a pollutant would become subject to PSD permitting requirements
well before the promulgation of the NAAQS under section 109.  See 40 CFR
52.21(b)(50)(i).

Similarly, the second part of the definition of “regulated NSR
pollutant” includes any pollutant that is subject to a standard
promulgated under section 111 of the CAA.  Section 111 requires EPA to
list a source category, if in his or her judgment, “it causes, or
contributes significantly to, air pollution which may reasonably be
anticipated to endanger public health or welfare.”  See CAA section
111(b)(1)(A).  After EPA lists a source category, it promulgates NSPS
for that source category.  For a source category not already listed, if
EPA were to list it on the basis of its emissions of a pollutant that
was not previously regulated, and such a listing made that pollutant
“subject to regulation” within the meaning of the PSD provisions,
this chain of events would result in triggering PSD permitting
requirements for that pollutant well in advance of the point
contemplated by the second prong of the regulated NSR pollutant
definition.  See 40 CFR 52.21(b)(50)(ii).

Furthermore, as discussed in the Memo, waiting to apply PSD requirements
at least until the actual promulgation of control requirements that
follow an endangerment finding is sensible.  The Memo explains that when
promulgating the final regulations establishing the control requirements
for a pollutant, EPA often makes decisions that are also relevant to
decisions that must be made in implementing the PSD program for that
pollutant.  See Memo at 14.  For example, EPA often does not make a
final decision regarding how to identify the specific pollutant subject
to an NSPS standard until the NSPS is issued, which occurs after both
the endangerment finding and the source category listing.  

Public comments echoed these concerns.  One commenter said that
subjecting the pollutant to PSD requirements, including imposition of
BACT emission limits, before the Agency has taken regulatory action to
establish emission controls would turn the CAA process on its head. 
Another commenter indicated that triggering PSD review upon completion
of an endangerment finding, but potentially before the specific control
requirement that flows directly from the endangerment finding, clearly
undermines the orderly process created by Congress for regulation of new
air pollutants.  A third commenter added that establishing controls
without having a standard to be achieved leads to uncertainty in the
permitting program. 

In further support of EPA’s interpretation that an endangerment
finding does not make an air pollutant “subject to regulation” is
the fact that an endangerment finding is not a codified regulation; it
does not contain any regulatory text.  The PSD Interpretive Memo
explains, and numerous commenters agree, that an endangerment finding
should not be construed as “regulating” the air pollutant(s) at
issue because there is no actual regulatory language applicable to the
air pollutant at this time in the Code of Federal Regulations.  Rather,
the finding is a prerequisite to issuing regulatory language that
imposes control requirements.  This is true even if the endangerment
finding is a “rule” for purposes of administrative processes; that
does not alter the fact that there is no regulation or regulatory text
attached to the endangerment finding itself.  Since an endangerment
finding does not establish “regulation” within the common meaning of
the term applied by EPA, we do not believe the CAA compels EPA to apply
PSD requirements to a pollutant on the basis of an endangerment finding
alone.  

EPA’s interpretation is also consistent with the Supreme Court’s
decision in Massachusetts.  In its decision, the Court acknowledged that
EPA “has significant latitude as to the manner, timing, content and
coordination” of the regulations that would result from a positive
endangerment finding under section 202(a).  See 549 U.S. at 532.  Just
as EPA has discretion regarding the timing of the section 202(a) control
regulations that would flow from an endangerment finding under that
section, it also has some discretion regarding the timing of the
triggering of PSD controls that the statute requires based on those
section 202(a) regulations.  EPA has reasonably determined that PSD
controls should not precede any other control requirements.  Some
commenters cited Massachusetts in support of EPA’s position. 

 For the foregoing reasons, we affirm that the prerequisite act of
making an endangerment finding, a cause or contribute finding, or both,
does not make a pollutant “subject to regulation” for the purposes
of the PSD program.  This interpretation applies to both steps of the
endangerment finding – the finding that air pollution may reasonably
be anticipated to endanger public health or welfare, and the finding
that emissions of an air pollutant from a particular source category
causes or contributes to this air pollution – regardless of whether
the two findings occur together or separately.  As explained above, EPA
believes that there are strong legal and policy reasons for rejecting
the endangerment finding interpretation. 

5.	Section 209 Waiver Interpretation 

EPA is affirming its position that an Agency decision to grant a waiver
to a state under section 209 of the CAA does not make the PSD program
applicable to pollutants that may be regulated under state authority
following a grant of such a waiver.  For the reasons discussed below,
the granting of a waiver does not make the pollutants that are regulated
by a state after obtaining a section 209 waiver into pollutants
regulated under the CAA.  Furthermore, EPA is also affirming the
position that PSD requirements are not applicable to a pollutant in all
states when a handful of states besides the one obtaining the waiver
adopt identical standards under section 177 of the CAA that are then
approved into state SIPs by EPA. 

As explained in the proposal, neither the PSD Interpretive Memo nor the
Petition for Reconsideration raise the issue of whether a decision to
grant a waiver under the section 209 of the CAA triggers PSD
requirements for a pollutant regulated by a state after obtaining a
waiver.  EPA received comments in response to the notice of an
application by California for a CAA section 209 waiver to the state of
California to adopt and enforce GHG emission standards for new motor
vehicles that suggested that arguments might be made that the grant of
the waiver made GHGs subject to regulation across the country for the
purposes of PSD.  See 74 FR 32744, 32783 (July 8, 2009).  Those
commenters requested that EPA state clearly that granting the California
Waiver did not render GHGs subject to regulation under the CAA, while
others commented that the question of when and how GHGs should be
addressed in the PSD program or otherwise regulated under the Act should
instead be addressed in separate proceedings.  At that time, EPA stated
that these interpretation issues were not a part of the waiver decision
and would be more appropriately addressed in another forum.  

	In our reconsideration of the PSD Interpretive Memo, EPA proposed to
affirm the following position that we previously explained to Congress:
“a decision to grant a waiver under section 209 of the Act removes the
preemption of state law otherwise imposed by the Act.  Such a decision
is fundamentally different from the decisions to establish requirements
under the CAA that the Agency and the [EAB] have considered in
interpreting the provisions governing the applicability of the PSD
program.”  Letter from Lisa P. Jackson, EPA Administrator, to Senator
James M. Inhofe (March 17, 2009).  Specifically, EPA proposed to find
that neither the CAA nor the Agency’s PSD regulations make the PSD
program applicable to pollutants that may be regulated by states after
EPA has granted a waiver of preemption under section 209 of the CAA. 
Accordingly, EPA said that the Agency’s decision to grant a section
209 waiver to the state of California to establish its own GHG emission
standards for new motor vehicles does not trigger PSD requirements for
GHGs.  

Several commenters disagreed with EPA’s proposed position on the
section 209 waiver provisions, and assert that EPA’s granting of the
waiver results in “actual control.”  According to these commenters,
even under EPA’s interpretation of “subject to regulation,” CO2 is
now subject to BACT.  One of these commenters argues that EPA’s
granting of a waiver is an EPA regulatory action that “controls” CO2
by allowing California and 10 other states to “regulate” CO2 under
the Act.  Another one of these commenters states that 10 states used
section 177 of the CAA to adopt the California Standards into their
SIPs, thus making these provisions enforceable by both EPA and citizens
under the CAA.  See, e.g., 42 U.S.C. §7413; 42 U.S.C. §7604(a)(1),
(f)(3).  EPA has not been persuaded to change its proposed position
based on these comments.  

EPA does not disagree that the regulations promulgated by the state
pursuant to the waiver will require control of emissions and thus
constitute “regulation” of GHGs under the meaning applied by EPA. 
However, the principal issue here is whether this regulation occurs
under the authority of the Clean Air Act (i.e., “under the Act.”).  

In our reconsideration notice, we explained that a waiver granted under
CAA section 209(b)(1) simply removes the prohibition found in section
209(a) that forbids states from adopting or enforcing their own
standards relating to control of emissions from new motor vehicles or
new motor vehicle engines.  Thus, the grant of the waiver does not lead
to regulation “under the Act” because it simply allows California to
exercise the same authority to adopt and enforce state emissions
standards for new motor vehicles that California could have exercised
without the initial prohibition in section 209(a).  Several other
commenters agreed with our position and reasoning.  They explained that
a waiver constitutes a withdrawal of federal preemption that allows a
state to develop its own state standards to regulate vehicle emissions;
the waiver does not transform these state standards into federal
standards.  Other supporting commenters also assert that there is
nothing in the legislative history that supports a conclusion that
Congress intended section 209 waivers to result in application of PSD
requirements. The opposing comments have not convincingly articulated a
mechanism through which EPA’s action granting the waiver in fact
requires control of emissions (as opposed to the states action under
state law).  If EPA granted the waiver alone and the state ultimately
decided not to implement its regulation, there would be no control
requirement in effect under the CAA.  

As we explained in our reconsideration notice, we also find it
instructive that enforcement of any emission standard by the state after
we grant a section 209 waiver would occur pursuant to state enforcement
authority, not federal authority.  EPA would continue to enforce the
federal emission standards we promulgate under section 202.  EPA does
not enforce the state standard.  EPA only conducts testing to determine
compliance with the federal standard promulgated by EPA and any
enforcement would be for violation of EPA standards, not the state
standards.  As one commenter noted, CAA section 209(b)(3) provides that
where a state has adopted standards that have been granted a waiver
“compliance with such state standards shall be treated as compliance
with applicable Federal standards for purposes of this subchapter,”
but does not say that such state standards actually become the federal
standards.  Accordingly, we find the absence of legislative history
supporting the contrary position, and the language in section 209(b)(3)
instructive as Congress clearly recognized the co-existence of the
federal and state standards.  This shows Congress did not intend that
state regulations replace, or transform state standards into federal
regulations “under the Act.”  We agree with supporting commenters’
conclusions summarized here, and are not persuaded to change our
proposed position.  

EPA has also concluded that the adoption of identical standards by
several states under section 177 does not make a pollutant covered by
those standards “subject to regulation under the Act” in all states.
 Like section 209, section 177 only grants states authority to regulate
under state authority by removing federal preemption.  Adoption of
California standards by other states does not change the fact that those
standards are still state standards enforced under state law and federal
law is approved in a SIP.  However, EPA agrees that when a state adopts
alternate vehicle standards into its SIP pursuant to section 177, and
EPA approves the SIP, these standards become enforceable by EPA and
citizens under the CAA.  Nonetheless, we do not agree that this compels
an interpretation that any pollutant included in an individual state SIP
requirement becomes “subject to regulation” in all states under the
CAA.  As discussed earlier, EPA rejects the theory that a regulation of
a pollutant in one or more states in an EPA-approved implementation plan
necessarily makes that pollutant subject to regulation in all states. 
Such an approach is inconsistent with the fundamental principle of
cooperative federalism embodied in the CAA.  

In summary, we conclude that neither the act of granting a section 209
waiver of preemption for state emission standards nor the EPA-approval
of standards adopted into a SIP pursuant to section 177 makes a
pollutant “subject to regulation under the Act” in all states for
the purposes of the PSD program.

C.   Other Issues on Which EPA Solicited Comment 

1. 	Prospective Codification of Interpretation

Through the proposed reconsideration notice, EPA  SEQ CHAPTER \h \r 1 
requested comment on whether the Agency should codify its final
interpretation of the “subject to regulation” in the statute and
regulation by amending the federal PSD rules at 40 CFR 52.21.  We
received a number of comments both in support of and opposing
codification.

EPA does not believe it is necessary to codify its interpretation in the
regulatory text.  The Agency feels it is important to promptly
communicate and apply our final decisions regarding the applicability of
the PSD program in light of recent and upcoming actions related to GHGs.
 More specifically, EPA recently finalized the “Mandatory Reporting of
Greenhouse Gases” rule (known as the “Reporting Rule”), which
added monitoring requirements for additional GHGs not covered in the
Part 75 regulations.  Further, EPA is poised to finalize by the end of
March 2010 the LDV Rule that will establish controls on GHGs that take
effect in Model Year 2012, which starts as early as January 2, 2011. 
Thus, these actions make it important that EPA immediately apply its
final interpretation of the PSD regulations on this issue (as refined in
this action). Furthermore, even if EPA modified the text of the federal
rules, many states may continue to proceed under an interpretation of
their rules.  EPA thus believes overall implementation of PSD permitting
programs is facilitated by this notice that describes how existing
requirements in federal regulations at 40 CFR 52.21 are interpreted by
EPA and how similar state provisions may be interpreted by states.

Likewise, EPA does not believe it is necessary to re-issue the PSD
Interpretive Memorandum.  The Agency has not identified any legal
requirement for the Agency to re-issue an interpretive rule after a
process of reconsideration.  No comparable procedure is required after
the reconsideration of substantive rule.  In the latter situation, a
notice of final action is sufficient to conclude the reconsideration
process and an Agency may simply decline to revise an existing
regulation that remains in effect.  EPA has therefore concluded that
this notice of final action is sufficient to conclude the
reconsideration process initiated on February 17, 2009 and that there is
no need to re-issue the entire memorandum in order for EPA to continue
applying the interpretation reflected therein, as refined in this
notice. 

2. 	Section 821 of the Clean Air Act Amendments of 1990

In the October 7, 2009 notice, EPA also solicited comment on the
question of whether section 821 of the Clean Air Act Amendments of 1990
is part of the Clean Air Act.  EPA indicated that the Agency was
inclined against continuing to argue that section 821 was not a part of
the CAA, as the Office of Air and Radiation and Region 8 had done in
briefs submitted to the EAB in the Deseret matter.  This question bears
on the determination of whether the CO2 monitoring requirements in
EPA’s Part 75 regulations are requirements “under the Act.”  In
the proposed reconsideration notice, EPA explained that it would be
necessary to resolve whether or not the CO2 monitoring and reporting
regulations in Part 75 were promulgated “under the Act” if EPA
adopted the monitoring and reporting interpretation.  EPA received
public comments on both sides of this issue, with one environmental
organization pressing EPA to drop the position that section 821 is not a
part of the CAA and several industry parties requesting that EPA affirm
it.

EPA has not yet made a final decision on this question, and it is not
necessary for the Agency to do so at this time.  Since EPA is not
adopting the monitoring and reporting interpretation, the status of
section 821 is not material to the question of whether and when CO2 is
“subject to regulation under the Act.”  Because there are currently
no controls on CO2 emissions, the pollutant is not “subject to
regulation.”  Given that the provisions in Part 75 do not
“regulate” emissions of CO2, it is unnecessary determine whether
such provisions are “under the Act” or not to determine PSD
applicability.  Furthermore, the promulgation of EPA’s Reporting Rule
makes this issue even less material.  In that rule, which became
effective in December 2009 and required monitoring to begin in January
of this year, EPA established monitoring and reporting requirements for
CO2 and other GHGs under sections 114 and 208 of the CAA.  Thus, there
can be no dispute that monitoring and reporting of CO2 (as well as other
GHGs) is now occurring under the CAA, regardless of the status of
section 821 of the 1990 amendments.  At this point, the section 821
issue would only become relevant if a court were to find that the
monitoring and reporting interpretation is compelled by the CAA and a
party subsequently seeks to retroactively enforce such a finding against
sources that had not obtained a PSD permit with any limit on CO2
emissions.  If this situation were to arise, EPA will address the
section 821 issue as necessary.  

3.  	Timing of When a Pollutant becomes Subject to Regulation

The October 7, 2009 notice also solicited comment on whether the
interpretation of “subject to regulation” should also more clearly
identify the specific date on which PSD regulatory requirements would
apply.  In the PSD Interpretive Memo, EPA states that the language in
the definition of “regulated NSR pollutant” should be interpreted to
mean that the fourth part of the definition should “apply to a
pollutant upon promulgation of a regulation that requires actual control
of emissions.”  See Memo at 14.  After evaluating the underlying
statutory requirement in the CAA and the language in all parts of the
regulatory definition more closely, EPA proposed to modify its
interpretation of the fourth part of the definition with respect to the
timing of PSD applicability.  The Agency proposed to interpret the term
“subject to regulation” in the statute and regulation to mean that
PSD requirements apply when the regulations addressing a particular
pollutant become final and effective.

Based on public comments and other considerations raised in the
proposal, EPA has determined that it is necessary to refine the portion
of the PSD Interpretive Memo that addresses the precise point in time
when a pollutant becomes subject to regulation for purposes of the PSD
program.  As a result, while the Memo is otherwise unchanged by the
reconsideration proceeding, this final notice will adjust the first
paragraph of section II.F of the Memo (bottom of page 14) to reflect
EPA’s conclusion that it is more appropriate and consistent with the
reasoning of the Memo to construe EPA regulations and the CAA to make a
pollutant subject to PSD program requirements when the first controls on
a pollutant take effect.  This refines the approach proposed in the
October 7, 2009 notice.  

	Like the PSD Interpretive Memorandum itself, the refinement to EPA’s
interpretation described in this final notice is an interpretation of
the regulation at 40 CFR 52.21 and the CAA provisions that provide the
statutory foundation for EPA’s regulations.  The refinement reflected
in this notice explains, clarifies, and is consistent with existing
statutes and the text of regulatory provisions at 40 CFR
52.21(b)(50)(ii) through (iv).  Some commenters argued that courts have
limited an Agency’s ability to fundamentally change a long-standing,
definitive, and authoritative interpretation of a regulation without
engaging in a notice and comment rulemaking.  See, e.g., Alaska
Professional Hunters Association v. FAA, 177 F.3d 1030, 1033-34 (D.C.
Cir. 1999); Paralyzed Veterans of America v. D.C. Arena L.P., 117 F.3d
579, 586 (D.C. Cir. 1997).  Since EPA’s interpretation of the PSD
program regulations is unchanged in most respects by this action, it is
not clear that the particular refinement to that interpretation that EPA
is making in this action would invoke the doctrine described in these
cases.  Even if this refinement is viewed as a fundamental change, EPA
has completed the revision reflected in this action after a notice and
comment process.  Furthermore, since EPA initiated a process of
reconsidering and soliciting comment on the PSD Interpretive Memo within
three months of its issuance, the memorandum had not yet become
particularly well-established or long-standing.  See MetWest Inc. v.
Secretary of Labor, 560 F.3d 506, 511 n.4 (D.C. Cir. 2009).  Thus, the
doctrines reflected in these cases do not preclude the action EPA has
taken here to refine its interpretation of the regulations. 

The regulatory language of 40 CFR 52.21(b)(50)(iv) does not specify the
exact time at which the PSD requirements should apply to pollutants in
the fourth category of the definition of “regulated NSR pollutant.” 
In the PSD Interpretive Memo, EPA states that EPA interprets the
language in this definition to mean that the fourth part of the
definition should “apply to a pollutant upon promulgation of a
regulation that requires actual control of emissions.”  See Memo at
14.  However, after continuing to consider the underlying statutory
requirement in the CAA and the language in all parts of the regulatory
definition more closely, EPA proposed in the October 7, 2009 notice to
modify its interpretation of the fourth part of the definition with
respect to the timing of PSD applicability.  In the notice of
reconsideration, EPA observed that the term “subject to regulation”
in the statute and regulation is most naturally interpreted to mean that
PSD requirements apply when the regulations addressing a particular
pollutant become final and effective.  In addition, EPA expressed a
desire to harmonize the application of the PSD requirements with the
limitation in the Congressional Review Act (CRA) that a major rule
cannot take effect until 60 days after it is published in the Federal
Register.  

In this final notice on reconsideration, based on information provided
in public comments, EPA is refining its interpretation of the time the
PSD requirements will apply to a newly-regulated pollutant.  Under the
PSD program, EPA will henceforth interpret the date that a pollutant
becomes subject to regulation under the Act to be the point in time when
a control or restriction that functions to limit pollutant emissions
takes effect or becomes operative to control or restrict the regulated
activity.  As discussed further below, this date may vary depending on
the nature of the first regulatory requirement that applies to control
or restrict emissions of a pollutant.

Several public comments observed that a date a control requirement
becomes “final and effective” and the date it actually “takes
effect” may differ.  Some commenters supported these points with
reference to federal court decisions that suggest the date that the
terms of a regulation become effective can take more than one form.  In
one case involving the Congressional Review Act, the United States Court
of Appeals for the Federal Circuit observed that the date a regulation
may “take effect” in accordance with the CRA is distinct from the
“effective date” of the regulation.  See Liesegang v. Sec’y of
Veterans Affairs, 312 F.3d 1368, 1374-75 (Fed. Cir. 2002), amended on
reh’g in part on other grounds, 65 Fed. Appx. 717 (Fed. Cir. 2003). 
In this opinion, the court observed that “[t]he ordinary meaning of
‘take effect’ is ‘[t]o be in force; go into operation’”  Id.
at 1375 (quoting Black’s Law Dictionary at 1466 (7th ed. 1999).  Based
on this, the court reasoned that the CRA does not “change the date on
which the regulation becomes effective” but rather “only affects the
date when the rule becomes operative.”  Id.  In another case, the
Second Circuit Court of Appeals described a distinction between the date
a rule may “take effect” under the CRA, the “effective date” for
application of the rule to regulated manufacturers, and the “effective
date” for purposes of modifying the Code of Federal Regulations. See
Natural Resources Defense Council v. Abraham, 355 F.3d 179, 202 (2d Cir.
2004).

	The Office of Federal Register (OFR) uses the term “effective date”
to describe the date that amendments in a rulemaking document affect the
current Code of Federal Regulations.  See Federal Register Document
Drafting Handbook, at p. 2-10 (Oct. 10, 1998).  However, OFR draws a
contrast between such a date and the compliance or applicability date of
a rule, which is described as “the date that the affected person must
start following the rule.”  Id. at 2-11.  Thus, the “effective
date” of a regulation is commonly used to describe the date by which a
provision in the Code of Federal Regulations is enacted as law, but it
is not necessarily the same as the time when provision enacted in the
Code of Federal Regulations is operative on the regulated activity or
entity.  The latter may be described as the “compliance,”
“applicability,” or “takes effect” date.  

The terms of the CAA also recognize a similar distinction in some
instances.  Section 112(i)(3)(A) of the CAA provides that “after the
effective date of any emissions standard, limitation or regulation
promulgated under this section and applicable to a source, no person may
operate such source in violation of such standard, limitation, or
regulation except, in the case of an existing source, the Administrator
shall establish a compliance date or dates for each category or
subcategory of existing sources, which shall provide for compliance as
expeditiously as practicable, but in no event later than 3 years after
the effective date of such standard.”  Another example in section 202
of the Act is discussed in more detail below. 

Another formulation may be found in Section 553(c) of the APA (5 U.S.C.
§553(c)), which provides, with some exceptions, that “[t]he required
publication or service of a substantive rule shall be made not less than
30 days before its effective date.”  The APA does not define the term
“effective date” or make precisely clear whether it is referring to
the date a regulation has the force of law or the date by which a
regulatory requirement applies to a regulated entity or activity.  The
APA also separately recognizes the concept of finality of Agency action
for purposes of judicial review.  See 5 U.S.C. §704. 

In the October 7, 2009 notice, EPA did not clearly distinguish between
the various forms of the date when a regulatory requirement may become
effective.  One commenter observed that the EPA analysis in the proposed
reconsideration appeared to blur the distinction between the
“effective date” set by EPA and the date that Congress allows a
regulation to become effective under the CRA.  EPA in fact discussed all
of these concepts in its notice, with part of the discussion focused on
the date a regulation becomes “final” and “effective” and a part
on when a regulation may “take effect” under the CRA.  EPA viewed
these forms of the date when a regulation becomes “effective” to be
essentially the same, but the case law suggests that administrative
agencies do not necessarily need to harmonize the date that regulatory
requirements take effect with the “effective date” of a regulation,
meaning the date a regulation has the force of law and amends the Code
of Federal Regulations.  Since these are distinct concepts, the
effective date of a regulation for purposes of amending the CFR may
precede the date when a regulatory requirement “takes effect” or
when a regulated entity must comply with a regulatory requirement.  A
regulation may “take effect” subsequent to its stated “effective
date” where it has been published in final form but does not require
immediate implementation by the agency or compliance by regulated
entities.  

The key issue raised by EPA in the October 7, 2009 notice was
determining which date should be determined by EPA to be the date when a
pollutant becomes “subject to regulation” and, thus, the date when
the requirements of the PSD permitting program apply to that pollutant. 
In recognition of the distinction between the “effective date” of
the regulation for purposes of amending the CFR and the point at which a
regulatory restriction may “take effect,” EPA has considered whether
it is permissible to construe sections 165(a)(4) and 169(3) of the CAA
to mean that a pollutant becomes “subject to regulation” at the
point that a regulatory restriction or control “takes effect.”  In
the October notice, EPA observed that the use of “subject to” in the
Act suggests that PSD requirements are intended to be triggered when
those standards become effective for the pollutant.  We also said that
no party is required to comply with a regulation until it has become
final and effective.  Prior to that date, an activity covered by a rule
is not in the ordinary sense “subject to” any regulation. 
Regardless of whether one interprets regulation to mean monitoring or
actual control of emissions, prior to the effective date of a rule there
is no regulatory requirement to monitor or control emissions.

The same reasoning applies to the date that a regulation “takes
effect,” as that term is used in the judicial decisions described
above.  Regulated entities are not required to comply with a regulatory
requirement until it takes effect.  Prior to the date a regulatory
requirement takes effect, the activity covered by a rule is not in the
ordinary sense subject to any regulation. 

As discussed in the PSD Interpretive Memo, as used in the context of the
PSD provisions in EPA regulations and the CAA, EPA interprets the term
“regulation” in the context of sections 165(a)(4) and 169 of the CAA
to mean the act or process of controlling or restricting an activity. 
This interpretation applies a common meaning of the term regulation
reflected in dictionaries.  

 	Thus, EPA agrees with commenters that the term “subject to
regulation” used in both the CAA and EPA’s regulations may be
construed to mean the point at which a requirement to control a
pollutant takes effect.  The CAA does not necessarily preclude
construing a pollutant to become subject to regulation upon the
promulgation date or the date that a regulation becomes final and
effective for purposes of amending the CFR or judicial review.  However,
EPA has been persuaded by public comments that the phrase “subject to
regulation” may also be interpreted to mean the date by which a
control requirement takes effect.

Indeed, EPA has concluded that the latter interpretation is more
consistent with the actual control interpretation reflected in the PSD
Interpretive Memo.  As one commenter observed, a regulation would have
to have become actually effective, in the sense that actual legal
obligations created by the regulation have become currently applicable
for regulated entities and are no longer merely prospective obligations,
before that regulation could make a pollutant subject to actual control.
 Another commenter noted that a regulated entity has no immediate
compliance obligations and cannot be held in violation of the regulation
until a legal obligation becomes applicable to them on the “takes
effect” date.  Thus, based on this reasoning, EPA has decided that it
will construe the point at which a pollutant becomes “subject to
regulation” within the meaning of section 52.21(b)(50)(iv) of EPA’s
regulations to be when a control or restriction is operative on the
activity regulated.  EPA agrees with commenters that there is generally
no legally enforceable obligation to control a pollutant when a
regulation is promulgated or, in some instances, even when a regulation
becomes effective for some purposes.

Thus, EPA currently interprets the time that a pollutant becomes a
“regulated NSR pollutant” under section 52.21(b)(50)(iv) to be the
time when a control or restriction on emissions of the pollutant takes
effect or becomes operative on the regulated activity.  Given EPA’s
conclusion that this is a permissible interpretation of the “subject
to regulation” language in sections 165(a)(4) and 169(3) of the CAA,
we will also interpret other parts of section 52.21(b)(50) to make a
pollutant a regulated NSR pollutant on the date that a control
requirement takes effect, provided such an interpretation is not
inconsistent with the existing language of the regulations.  

EPA does not agree with several commenters who suggested that EPA
determine that a pollutant does not become subject to regulation until
the time that an individual source engages in the regulated activity. 
EPA does not believe such a reading is consistent with the “subject to
regulation” language in the CAA.  Even if no source is actually
engaged in the activity, once a standard or control requirement has
taken effect, no source may engage in the regulated activity without
complying with the standard.  At this point, the regulated activity and
the emissions from that activity are controlled or restricted, thus
being subject to regulation within the common meaning of the term
regulation used in EPA’s regulations and section 165(a)(4) and 169(3)
of the CAA.

Likewise, EPA does not agree with commenters who argued that a pollutant
does not become subject to regulation until the date when a source must
certify compliance with regulatory requirements or submit a compliance
report.  In some instances, a compliance report or certification of
compliance may not be required until well after the point that a
regulation operates to control or restrict the regulated activity. 
Thus, EPA does not feel that it would be appropriate as a general rule
to establish the date when a source certifies compliance or submits its
compliance report as the date that a pollutant becomes subject to
regulation.

Since the fourth part of the definition of “regulated NSR pollutant”
functions as a catch-all provision, it may cover a variety of different
types of control requirements established by EPA under the CAA.  These
different types of regulations may contain a variety of different
mechanisms for controlling emissions and have varying amounts of lead
time before controls take effect under the particular regulatory
framework.  Thus, whenever the Agency adopts controls on a new pollutant
under a portion of the CAA covered by the fourth part of the definition,
EPA anticipates that it will be helpful to states and regulated sources
for EPA to identify the date when a new pollutant becomes subject to
regulation.  In section IV.A of this notice, EPA provides such an
analysis for the forthcoming LDV Rule that is anticipated to establish
the first controls on GHGs.  

EPA has also concluded that it is appropriate to extend the reasoning of
this interpretation across all parts of the definition of the term
“regulated NSR pollutant.”  The reasoning described above is equally
applicable to the regulation of additional pollutants under the specific
sections of the Act delineated in the first three parts of the
definition of “regulated NSR pollutant.”  While the date a control
requirement may take effect could vary across sections 109, section 111,
and Title VI, we do not see any distinction in the applicability of the
legal reasoning above to these provisions of the CAA.  There should be
less variability among rules promulgated under the same statutory
section, so EPA does not expect that it will be necessary for EPA to
identify the date that a new pollutant becomes subject to regulation
each time EPA regulates a new pollutant in a NAAQS or NSPS.  EPA can
more readily identify the specific dates when controls under such rules
take effect. 

By way of example, the NSPS under section 111 of the Act preclude
operation of a new source in violation of such a standard after the
effective date of the standard.  See 42 U.S.C. 7411(e).  Thus, the
control requirements in an NSPS take effect on the effective date of the
rule.  Once such a standard takes effect and operates to preclude
operations in violation of the standards, then EPA interprets the
statute and EPA’s PSD regulations to also require that the BACT
requirement apply to a pollutant that is subject to NSPS.  Consistent
with our October 7, 2009 proposal, EPA has determined that the existing
language in section 52.21(b)(50)(ii) of its regulations may be construed
to apply to a new pollutant upon the effective date of an NSPS.  This
part of the definition covers “[a]ny pollutant that is subject to any
standard promulgated under section 111 of the Act.”  See 40 CFR
52.21(b)(50)(ii).  While the word “promulgated” appears in this part
of the definition, this term modifies the term “standard” and does
not directly address the timing of PSD requirements.  Under the language
in this part of the definition, the PSD requirements apply when a
pollutant becomes “subject to” the underlying standard, which is
“promulgated under” section 111 of the Act.  Thus, this language can
be interpreted to make an NSPS pollutant a regulated NSR pollutant upon
the effective date of an NSPS.  EPA did not receive any public comments
that opposed reading this portion of the definition to invoke PSD
requirements upon the effective date of an NSPS.  This can logically be
extended to be consistent with the general view described above that the
time a pollutant becomes subject to regulation is the time when a
control requirement “takes effect.”  As discussed above, the
effective date of an NSPS is also that date when the controls in an NSPS
“take effect.” 

Likewise, under section 169(a)(3) of the Act, a source applying for a
PSD permit must demonstrate that it will not cause or contribute to a
violation of the NAAQS in order to obtain the permit.  Once a NAAQS is
effective with respect to a pollutant, the standard operates through
section 169(a)(3) of the Act and section 52.21(k) of EPA’s regulations
to preclude construction of a new source that would cause or contribute
to a violation of such standard. 

Using the effective date of a NAAQS to determine when a pollutant
covered by a NAAQS becomes a regulated NSR pollutant is more consistent
with EPA’s general approach for determining when a new NAAQS applies
to pending permit applications.  EPA generally interprets a revised
NAAQS that establishes either a lower level for the standard or a new
averaging time for a pollutant already regulated to apply upon the
effective date of the revised NAAQS.  Thus, unless EPA promulgates a
grandfathering provision that allows pending applications to apply
standards in effect when the application is complete, a final permit
decision issued after the effective date of a NAAQS must consider such a
NAAQS.  As described above, the effective date of the NAAQS is also the
date a NAAQS takes effect through the PSD permitting program to regulate
construction of a new or modified source.

Since a NAAQS covering a new pollutant would operate through the PSD
permitting program to control emissions of that pollutant from the
construction or modification of a major source upon the effective date
of the NAAQS, a NAAQS covering a new pollutant takes effect on the
effective date of the regulation promulgating the NAAQS.  EPA does not
agree with one commenter’s suggestion that such a NAAQS would not take
effect until the time a state first promulgates limitations for the
pollutant in a SIP.  Under section 165(a)(3) of the Act and the federal
PSD permitting regulations at 52.21(k), to obtain a PSD permit, a major
source must demonstrate that the proposed construction will not cause or
contribute to a violation of a NAAQS.  Due to these requirements, the
PSD program operates to incorporate the NAAQS as a governing standard
for permitting construction of large sources.  Thus, under the federal
PSD program regulations at least, a new pollutant covered by a NAAQS
becomes subject to regulation at a much earlier date.  These PSD
provisions require emissions limitations for the NAAQS pollutant before
construction at a major source may commence and thereby function to
protect the NAAQS from new source construction and modifications of
existing major sources in the SIP development period before a completion
of the planning process necessary to determine whether additional
standards for a new NAAQS pollutant need to be developed.  The timing
when the NAAQS operates in this manner under SIP-approved programs is
potentially more nuanced and depends on whether state laws are
sufficiently open-ended to call for application of a new NAAQS as a
governing standard for PSD permits upon the effective date.  EPA
believes that state laws that use the same language as in EPA’s PSD
program regulations at 52.21(k) and 51.166(k) are sufficiently
open-ended and allow such a NAAQS to “take effect” through the PSD
program upon the effective date of the NAAQS.  Notwithstanding this
complexity in SIP-approved programs, the applicability of the federal
PSD program regulations to a new NAAQS pollutant upon the effective date
of the NAAQS is sufficient to determine that a new pollutant is subject
to regulation on this date. 

In the October 7, 2009 notice, EPA observed that one portion of its
existing regulations was not necessarily consistent with this reading of
the CAA.  For the first class of pollutants described in the definition
of “regulated NSR pollutant,” the PSD requirements apply once a
“standard has been promulgated” for a pollutant or its precursors. 
See 40 CFR 52.21(b)(50)(i).  The use of “has been” in the regulation
indicates that a pollutant becomes a "regulated NSR pollutant," and
hence PSD requirements for the pollutant are triggered, on the date a
NAAQS is promulgated. Thus, EPA observed in the October 7, 2009 notice
that it may not be possible for EPA to read the regulatory language in
this provision to make PSD applicable to a NAAQS pollutant upon the
effective date of the NAAQS.  EPA did not propose to modify the language
in 40 CFR 52.21(b)(50)(i) in the October 2009 notice because EPA had not
yet reached a final decision to interpret the CAA to mean that a
pollutant is subject to regulation on the date a regulatory requirement
becomes effective.  Since EPA was not proposing to establish a NAAQS for
any additional pollutants, the timing of PSD applicability for a newly
identified NAAQS pollutant did not appear to be of concern at the time. 
No public comments on the October 2009 notice addressed this issue. 
Since EPA is now adopting a variation of the proposed interpretation
with respect to the timing of PSD applicability, we believe it will be
appropriate to propose a revision of the regulatory language in section
52.21(b)(50)(i) at such time as EPA may consider promulgation of a NAAQS
for an additional pollutant.  Until that time, however, EPA will
continue to apply the terms of section 52.21(b)(50)(i) of the
regulation.  This is permissible because, even though EPA believes the
better reading of the Act is to apply PSD upon the date that a control
requirement “takes effect,” the Agency has not determined in this
action that the Clean Air Act precludes applying PSD requirements upon
the promulgation of a control requirement. 

IV.	Application of PSD Interpretive Memo to PSD Permitting for GHGs

A. 	Date by Which GHGs Will Be “Subject to Regulation”

Although the PSD Interpretive Memo and this reconsideration reflect a
broad consideration of the most appropriate legal interpretation and
policy for all pollutants regulated under the CAA, the need to clarify
this issue as a general matter has been driven by concerns over the
effects of GHG emissions on global climate and the contention made by
some parties in permit proceedings that EPA began regulating CO2 as
early as the promulgation of monitoring and reporting requirements in
EPA’s Part 75 rules to implement section 821 of the CAA Amendments of
1990.  The vast majority of public comments on the October 7, 2009
notice focused on the regulation of GHGs under the PSD program.  As a
result, EPA recognizes that it is critically important at this time for
the Agency to make clear when the requirements of the PSD permitting
program for stationary sources will apply to GHGs.  For the reasons
discussed below, GHGs will initially become “subject to regulation”
under the CAA on January 2, 2011, assuming that EPA issues final GHG
emissions standards under section 202(a) applicable to model year 2012
new motor vehicles as proposed.  As a result, with that assumption, the
PSD permitting program would apply to GHGs on that date.  However, the
Tailoring Rule, noted above, proposed various options for phasing in PSD
requirements for sources emitting GHGs in various amounts above 100 or
250 tons per year.  Since EPA has not yet completed that rulemaking,
today’s action concludes only that, under the approach envisioned for
the vehicle standards, GHGs would not be considered “subject to
regulation” (and no source would be subject to PSD permitting
requirements for GHGs) earlier than January 2, 2011.  The final
Tailoring Rule will address the applicability of PSD requirements for
GHG-emitting sources that are not presently subject to PSD permitting. 

EPA’s determination that PSD will begin to apply to GHGs on January 2,
2011 is based on the following considerations: (1) the overall
interpretation reflected in the PSD Interpretive Memo; (2) EPA’s
conclusion in this notice that a pollutant becomes subject to regulation
when controls “take effect,” and (3) the assumption that the agency
will establish emissions standards for model year 2012 vehicles when it
completes the proposed LDV Rule. 

As proposed, the LDV Rule consists of two kinds of standards — fleet
average standards determined by the emissions performance of a
manufacturer’s fleet of various models, and separate vehicle standards
that apply for the useful life of a vehicle to the various models that
make up the manufacturer’s fleet.  CAA section 203(a)(1) prohibits
manufacturers from introducing a new motor vehicle into commerce unless
the vehicle is covered by an EPA-issued certificate of conformity for
the appropriate model year.  Section 206(a)(1) of the CAA describes the
requirements for EPA issuance of a certificate of conformity, based on a
demonstration of compliance with the emission standards established by
EPA under section 202 of the Act.  A certification demonstration
requires emission testing, and must be done for each model year.  

The certificate covers both fleet average and vehicle standards, and the
manufacturer has to demonstrate compliance with both of these standards
for purposes of receiving a certificate of conformity.  The
demonstration for the fleet average is based on a projection of sales
for the model year, and the demonstration for the vehicle standard is
based on emissions testing and other information. 

Both the fleet average and vehicle standards in the LDV Rule will
require that automakers control or limit GHG emissions from the
tailpipes of these vehicles.  As such, they clearly constitute
“regulation” of GHGs under the interpretation in the PSD
Interpretive Memo.  This view is consistent with the position originally
expressed by EPA in 1978 that a pollutant regulated in a Title II
regulation is a pollutant subject to regulation.  See 42 FR at 57481. 
However, the regulation of GHGs will not actually take effect upon
promulgation of the LDV Rule or on the effective date of the LDV Rule
when the provisions of the rule are incorporated into the Code of
Federal Regulations. 

Under the LDV Rule, the standards for GHG emissions are not operative
until the 2012 model year, which may begin as early as January 2, 2011. 
In accordance with the requirements of Title II of the CAA and
associated regulations, vehicle manufacturers may not introduce a model
year 2012 vehicle into commerce without a model year 2012 certificate of
conformity.  CAA section 203(a)(1).  A model year 2012 certificate only
applies to vehicles produced during that model year, and the model year
production period may begin no earlier than January 2, 2011.  See
Section 202(b)(3)(A) of the CAA and implementing regulations at 40 CFR
85.2302 through 85.2305.  Thus, a vehicle manufacturer may not introduce
a model year 2012 vehicle into commerce prior to January 2, 2011.  

There will be no controls or limitations on GHG emissions from model
year 2011 vehicles.  The obligation on an automaker for a model year
2012 vehicle would be to have a certificate of conformity showing
compliance with the emissions standards for GHGs when the vehicle is
introduced into commerce, which can occur on or after January 2, 2011. 
Therefore, the controls on GHG emissions in the Light Duty Rule will not
take effect until the first date when a 2012 model year vehicle may be
introduced into commerce.  In other words, the compliance obligation
under the LDV Rule does not occur until a manufacturer may introduce
into commerce vehicles that are required to comply with GHG standards,
which will begin with MY 2012 and will not occur before January 2, 2011.
 Since section 203(a)(1) of the CAA prohibits manufacturers from
introducing a new motor vehicle into commerce unless the vehicle is
covered by an EPA-issued certificate of conformity for the appropriate
model year, as of January 2, 2011, manufacturers will be precluded from
introducing into commerce any model year 2012 vehicle that has not been
certified to meet the applicable standards for GHGs.  

This interpretation of when the GHG controls in the LDV Rule take
effect, and therefore, make GHGs subject to regulation under the Act for
PSD purposes, is consistent with the statutory language in section
202(a)(2) of the CAA.  This section provides that “any regulation
prescribed under paragraph (1) of this subsection (and any revision
thereof) shall take effect after such period as the Administrator finds
necessary to permit the development and application of the requisite
technology, giving appropriate consideration to the cost of compliance
within such period.”  See 42 U.S.C. 7521(a)(2) (emphasis added).  The
final LDV Rule will apply to model years 2012 through 2016.  The time
leading up to the introduction of model year 2012 is the time that EPA
“finds necessary to permit the development and application of the
requisite technology, giving appropriate consideration to the cost of
compliance within such period.”  Model year 2012 is therefore when the
GHG standards in the rule “take effect.”  

	EPA does not agree with several commenters who have suggested that the
GHG standards in the proposed LDV Rule would not take effect until
October 1, 2011.  The latter date appears to be based on how the
National Highway Traffic Safety Administration (NHTSA) determines the
beginning of the 2012 model year under the Energy Policy and
Conservation Act (EPCA).  Under EPCA, a more stringent CAFE standard
must be prescribed by NHTSA at least 18 months before the beginning of
the model year.  For purposes of this EPCA provision, NHTSA has
historically construed the beginning of the model year to be October 1
of the preceding calendar year.  See 49 U.S.C. §32902(g)(2); 74 FR
49454, 49644 n.447 (Sep. 28, 2009).  Although EPA has endeavored to
harmonize its section 202(a) standards with the NHTSA CAFE standards,
EPA’s standards are promulgated under distinct legal authority in the
CAA.  Thus, the section 202(a) standards promulgated in the LDV Rule are
not subject to EPCA or NHTSA’s interpretation of when a model year
begins for purposes of EPCA.  Under EPA’s planned LDV regulations,
model year 2012 vehicles may be introduced into commerce as early as
January 2, 2011.  Although as a practical matter, some U.S. automakers
may not begin introducing model year 2012 vehicles into commerce until
later in 2011, they may nevertheless do so as early as January 2, 2011
under EPA’s regulations.  Consistent with the discussion above, EPA
construes the phrase “subject to regulation” in section 165(a)(4)
and 169(3) of the Act to mean that the BACT requirement applies when
controls on a pollutant first apply to a regulated activity, and not the
point at which an entity first engages in the regulated activity.  In
this instance, the regulated activity is the introduction of model year
2012 vehicles into commerce.  As of January 2, 2011, a manufacturer may
not engage in this activity without complying with the applicable GHG
standards.  

	Likewise, EPA does not agree with commenters who argued that EPA should
not consider the GHG controls in the LDV Rule to take effect until
automakers have to demonstrate compliance with the fleet average
standards at the end of the model year, based on actual vehicle model
production.  As discussed above, the LDV Rule includes both fleet
average standards and vehicle standards that apply to individual
vehicles throughout their useful lives.  As discussed above, both of
these standards for GHG emissions are operative on model year 2012
vehicles introduced into commerce on or after January 2, 2011.  Thus,
controls on GHG emissions from automobiles will take effect prior to the
date that a manufacturer must demonstrate compliance with the fleet
average standards.  The fact that the manufacturer demonstrates final
compliance with the fleet average at a later date, based on production
for the entire year, does not change the fact that their conduct was
controlled by both the fleet average and the vehicle standards, and
subject to regulation, prior to that date.   

B.	Implementation Concerns

A substantial number of commenters requested that EPA defer application
of the PSD program requirements to GHGs based on various practical
implementation considerations, and several of these comments argued that
the CAA affords EPA the discretion to set an implementation date based
on such concerns.  EPA agrees that application of PSD program
requirements to GHGs presents several significant implementation
challenges for EPA, states and other entities that issue permits, and
the sources that must obtain permits.  Indeed, many of the public
comments have illustrated the magnitude of the challenge beyond what is
described in the notices on reconsideration of the PSD Interpretive Memo
and the proposed Tailoring Rule.  

In recognition of the substantial challenges associated with
incorporating GHGs into the PSD program, EPA’s preference would be to
establish a specific date when the PSD permitting requirements initially
apply to GHGs based solely on these practical implementation
considerations.  However, EPA has not been persuaded that it has the
authority to proceed in this manner.  While EPA may have discretion as
to the manner and time for regulating GHG emissions under the CAA, once
EPA has determined to regulate a pollutant in some form under the Act
and such regulation is operative on the regulated activity, the terms of
the Act make clear that the PSD program is automatically applicable. 

Nonetheless, given the substantial magnitude of the PSD implementation
challenges presented by the regulation of GHGs, EPA proposed in the
Tailoring Rule to at least temporarily limit the scope of GHG sources
covered by the PSD program to ensure that permitting authorities can
effectively implement it.  EPA based the proposal primarily on two legal
doctrines:  the “absurd results” doctrine, which we proposed to
apply on the basis that Congress did not envision that the PSD program
would apply to the many small sources that emit GHGs; and the
“administrative necessity” doctrine, which we proposed to apply
because of the extremely large administrative burdens that permitting
authorities would confront in permitting the GHG sources.  In comment on
that action, as well as in comments on the PSD Interpretive Memo
reconsideration notice, EPA received numerous suggestions that it is
necessary to limit the scope of sources covered at the time GHGs become
subject to regulation.  Commenters further stated that it is necessary
to select a “trigger date” for GHG permitting that takes into
account the time needed for permitting authorities to adopt any
scope-limiting measures (including the need to amend state law), to
secure the necessary additional financial and other resources, and to
hire and train the staff needed to respond to the increase in permitting
workload.  These comments make clear that more time will be needed
beyond January 2, 2011 before permitting of many GHG stationary sources
can begin.  Thus, EPA will be taking additional action in the near
future in the context of the Tailoring Rule to address GHG-specific
circumstances that will exist beyond January 2, 2011.

C.  	Interim EPA Policy to Mitigate Concerns Regarding GHG Emissions
from Construction or Modification of Large Stationary Sources

While EPA has concluded that GHGs will not become subject to regulation
(and hence the PSD BACT requirement will not apply to them) earlier than
January 2, 2011, permitting authorities that issue permits before
January 2, 2011 are already in a position to, and should, use the
discretion currently available under the BACT provisions of the PSD
program to promote technology choices for control of criteria pollutants
that will also facilitate the reduction of GHG emissions.  More
specifically, the CAA BACT definition requires permitting authorities
selecting BACT to consider the reductions available through application
of not only control methods, systems, and techniques, but also through
production processes, and requires them to take into account energy,
environmental, and economic impacts.  Thus, the statute expresses the
need for a comprehensive review of available pollution control methods
when evaluating BACT that clearly requires consideration of energy
efficiency.  The consideration of energy efficiency is important because
it contributes to reduction of pollutants to which the PSD requirements
currently apply and have historically been applied.  We further note
that, although BACT does not now apply to GHG, BACT for other pollutants
can, through application of more efficient production processes,
indirectly result in lower GHG emissions.  

Neither the statute nor EPA regulations specify precisely how to address
energy efficiency in BACT determinations, nor have we fully articulated
how to take climate considerations into account under the “energy,
environmental, and economic impacts” considerations of BACT.  Further,
while our BACT guidance for currently regulated pollutants has addressed
some facets of these issues, we believe that, given the potential
importance of the indirect GHG benefits, it will be useful for EPA to
summarize this guidance and further clarify it as necessary in order to
further illustrate where PSD permitting authorities should be using
existing BACT authority for pollutants that are presently regulated in
ways that can indirectly address concerns about GHG emissions from large
stationary sources.  We are developing such guidance and plan to issue
it in the near future.

D.	Transition for Pending Permit Applications

Some commenters requested that EPA address the question of how the
application of PSD requirements to GHGs will affect applications for PSD
permits that are pending on the date GHGs initially become “subject to
regulation.”  These commenters generally asked that EPA establish an
exclusion for any PSD permit application that was submitted in complete
form before the date on which PSD begins to apply to GHGs.

In light of EPA’s conclusion that pollutants become subject to
regulation for PSD purposes when control requirements on that pollutant
take effect and that such requirements will not take effect for GHGs
until January 2, 2011 if EPA finalizes the proposed LDV rule as
anticipated, EPA does not see any grounds to establish a transition
period for permit applications that are pending before GHGs become
subject to regulation.  As a general matter, permitting and licensing
decisions of regulatory agencies must reflect the law in effect at the
time the agency makes a final determination on a pending application. 
See Ziffrin v. United States, 318 U.S. 73, 78 (1943); State of Alabama
v. EPA, 557 F.2d 1101, 1110 (5th Cir. 1977); In re: Dominion Energy
Brayton Point, LLC, 12 E.A.D. 490, 614-616 (EAB 2006); In re Phelps
Dodge Corp., 10 E.A.D. 460, 478 n.10 (EAB 2002).  Thus, in the absence
of an explicit transition or grandfathering provision in the applicable
regulations (and assuming EPA finalizes the LDV rule as planned), each
PSD permit issued on or after January 2, 2011 would need to contain
provisions that satisfy the PSD requirements that will apply to GHGs as
of that date.

Under certain circumstances, EPA has previously allowed proposed new
major sources and major modifications that have submitted a complete PSD
permit application before a new requirement becomes applicable under PSD
regulations, but have not yet received a final and effective PSD permit,
to continue relying on information already in the application rather
than immediately having to amend applications to demonstrate compliance
with the new PSD requirements.  In such a way, these proposed sources
and modifications were “grandfathered” or exempted from the new PSD
requirements that would otherwise have applied to them.

For example, EPA adopted a grandfathering provision when it changed the
indicator for the particulate matter NAAQS from total suspended
particulate matter (TSP) to particulate matter less than 10 microns
(PM10).  The federal PSD regulations at 40 CFR 52.21(i)(1)(x) provide
that the owners or operators of proposed sources or modifications that
submitted a complete permit application before July 31, 1987, but did
not yet receive the PSD permit, are not required to meet the
requirements for PM10, but could instead satisfy the requirements for
TSP that were previously in effect.

  	In addition, EPA has allowed some grandfathering for permit
applications submitted before the effective date of an amendment to the
PSD regulations establishing new maximum allowable increases in
pollutant concentrations (also known as PSD “increments”).  The
federal PSD regulations at 40 CFR 52.21(i)(10) provide that proposed
sources or modifications that submitted a complete permit application
before the effective date of the increment in the applicable
implementation plan are not required to meet the increment requirements
for PM10, but could instead satisfy the increment requirements for TSP
that were previously in effect.  Also, 40 CFR 52.21(i)(9) provides that
sources or modifications that submitted a complete permit application
before the provisions embodying the maximum allowable increase for
nitrogen oxides (NOx) took effect, but did not yet receive a final and
effective PSD permit, are not required to demonstrate compliance with
the new increment requirements to be eligible to receive the permit.

	Under the particular circumstances presented by the forthcoming
application of PSD requirements to GHGs, EPA does not see a
justification for adopting an explicit grandfathering provision of the
nature described above.  Permit applications submitted prior to the
publication of this notice should in most cases be issued prior to
January 2, 2011 and, thus, effectively have a transition period of nine
months to complete processing before PSD requirements become applicable.
 Additional time for completion of action on applications submitted
prior to the onset of PSD requirements for GHGs therefore does not
appear warranted to ensure a smooth transition and avoid delays for
pending applications.  To the extent any pending permit review cannot
otherwise be completed within the next nine months based on the
requirements for pollutants other than GHGs, it should be feasible for
permitting authorities to begin incorporating GHG considerations into
permit reviews in parallel with the completion of work on other
pollutants without adding any additional delay to permit processing.  

Furthermore, the circumstances surrounding the onset of requirements for
GHGs are distinguishable from prior situations where EPA has allowed
grandfathering of applications that were deemed complete prior to the
applicability new PSD permitting requirements.  First, this action and
the PSD Interpretive Memo do not involve a revision of the PSD
permitting regulations but rather involves clarifications of how EPA
interprets the existing regulatory text.  This action articulates what
has, in most respects, been EPA’s longstanding practice.  It has been
EPA’s consistent position since 1978 that regulation of a pollutant
under Title II triggers PSD requirements for such a pollutant.  See 42
FR 57481.  Thus, permitting authorities and permit applicants could
reasonably anticipate that completion of the LDV Rule would trigger PSD
and prepare for this action.  Many commenters interpreted EPA’s
October 7, 2009 notice as proposing to trigger PSD requirements within
60 days of the promulgation of the LDV Rule rather than the January 2,
2011 date that EPA has determined to be the date the controls in that
rule take effect.  Second, there are presently no regulatory
requirements in effect for GHGs.  On the other hand, at the time EPA
moved from using TSP to using PM10 as the indicator for the particulate
matter NAAQS, grandfathered sources were still required to satisfy PSD
requirements for particulate matter based on the TSP indicator. 
Likewise, when EPA later updated the PSD increment for particulate
matter to use the PM10 indicator, the grandfathered sources were still
required to demonstrate that they would not cause or contribute to a
violation of the particulate matter increment based on TSP.  In the case
of the adoption of the NO2 increment, grandfathered sources were still
required to demonstrate that they would not cause or contribute to a
violation of the NO2 NAAQS.  In contrast, for GHGs, there are no
measures currently in effect that serve to limit emission of GHGs from
stationary sources.

For these reasons, EPA does not intend to promulgate a transition or
grandfathering provision that exempts pending permit applications from
the onset of GHG requirements in the PSD program.  As discussed above,
in the absence of such a provision, PSD permits that are issued on or
after January 2, 2011 (in accordance with limitations promulgated in the
upcoming Tailoring Rule) will be required to contain provisions that
fulfill the applicable program requirements for GHGs.

V. 	PSD Program Implementation by EPA and States

	Consistent with the PSD Interpretive Memo, the refined interpretation
reflected in this notice (that a pollutant subject to actual control
becomes subject to regulation at the time such controls take effect) is
an interpretation of the language in 40 CFR 52.21(b)(50) of EPA’s
regulations.  EPA will apply the PSD Interpretive Memo, with the
refinement described above, when implementing the federal permitting
program under 40 CFR 52.21.  Furthermore, we will expect that states
that implement the federal PSD permit program under delegation from an
EPA Regional Office will do the same.  

In addition, EPA will apply the interpretation reflected in this notice
and the PSD Interpretive Memo in its oversight of existing state
programs and review and approval of new program submissions.  Many
states implement the PSD program pursuant to state laws that have been
approved by EPA as part of the SIP, pursuant to a determination by EPA
that such laws meet the PSD program criteria set forth in 40 CFR 51.166.
 The EPA regulation setting forth PSD program requirements for SIPs also
includes the same definition of the term “regulated NSR pollutant”
as the federal program regulation.  See 40 CFR 51.166(b)(49).  Because
this regulation uses the same language as contained in 40 CFR 52.21 and
the same considerations apply to implementation of the PSD program under
state laws, EPA will interpret section 51.166(b)(49) in the same manner
as section 52.21(b)(50).  However, in doing so, EPA will be mindful that
permitting authorities in SIP approved states have some independent
discretion to interpret state laws, provided those interpretations are
consistent with minimum requirements under the federal law.   

To the extent approved SIPs contain the same language as used in 40 CFR
52.21(b)(50) or 40 CFR 51.166(b)(49), SIP-approved state permitting
authorities may interpret that language in state regulations in the same
manner reflected in the PSD Interpretive Memo and this notice.  However,
EPA will not seek to preclude actions to address GHGs in PSD permitting
actions prior to January 2, 2011 where a state permitting authority
feels it has the necessary legal foundation and resources to do so.  

EPA has not called on any states to make a SIP submission that addresses
the interpretive issues addressed in this notice and the PSD
Interpretive Memo.  As long as states are applying their approved
program regulations consistent with the minimum program elements
established in 40 CFR 51.166, EPA does not believe it will be necessary
to issue a SIP call for all states to address this issue.  However,
permitting authorities in SIP-approved states do not have the discretion
to apply state laws in a manner that does not meet the minimum federal
standards in 40 CFR 51.166, as interpreted and applied by EPA.  Thus, if
a state is not applying the PSD requirements to GHGs for the required
sources after January 2, 2011, or lacks the legal authority to do so,
EPA will exercise its oversight authority as appropriate to call for
revisions to SIPs and to otherwise ensure sources do not commence
construction without permits that satisfy the minimum requirements of
the federal PSD program.  

To enable EPA to assess the consistency of a state’s action with any
PSD program requirements for GHGs, states should ensure that the record
for each PSD permitting decision addresses whether the state has elected
to follow EPA’s interpretation or believes it is appropriate to apply
a different interpretation of state laws that is nonetheless consistent
with the requirements of EPA’s PSD program regulations.  In light of
additional actions to be taken by EPA in the Tailoring Rule, states that
issue permits in the near term may want to preserve the discretion to
modify their approach after other EPA actions are finalized.  In light
of this contingency, one option states may consider is to establish that
the state will not interpret its laws to require PSD permits for sources
that are not required to obtain PSD permits under EPA regulations.   

VI. 	Application of the Title V Program to Sources of GHGs

Although the PSD Interpretive Memorandum and the October 7, 2009
reconsideration notice addressed only PSD permitting issues, EPA
received several comments on the reconsideration notice that also
addressed the application of Title V permitting requirements to GHGs. 
Most of these comments urged EPA to apply the same approach for
determining major source applicability for Title V permitting that EPA
applies to PSD. EPA has in fact been following the PSD approach in many
respects.  As with the PSD program, currently GHGs are not considered to
be subject to regulation and have not been considered to trigger
applicability under Title V.  EPA discussed this in the preamble to the
proposed Tailoring Rule as described below.  See 74 FR at 55300 n.8.  

	Title V requires, among other things, that any “major source” –
defined, as relevant here, under CAA sections 302(j) and 501(2)(b), as
“any stationary facility or source of air pollutants which directly
emits, or has the potential to emit, one hundred tons per year or more
of any air pollutant...” – apply for a Title V permit.  EPA
interprets this requirement to apply to sources of pollutants “subject
to regulation” under the Act.  EPA previously articulated its
interpretation that this Title V permitting requirement applies to
“pollutants subject to regulation” in a 1993 memorandum from EPA’s
air program.  Memorandum from Lydia N. Wegman, Deputy Director, Office
of Air Quality Planning and Standards, U.S. EPA, “Definition of
Regulated Air Pollutant for Purposes of Title V” (Apr. 26, 1993)
(“Wegman Memo”).  EPA continues to maintain this interpretation. 
The interpretation in this memorandum was based on: (1) EPA’s reading
of the definitional chain for “major source” under Title V,
including the definition of “air pollutant” under section 302(g) and
the definition of “major source” under 302(j); (2) the view that
Congress did not intend to require a variety of sources to obtain Title
V permits if they are not otherwise regulated under the Act (see also
CAA section 504(a), providing that Title V permits are to include and
assure compliance with applicable requirements under the Act); and (3)
consistency with the approach under the PSD program.  While the specific
narrow interpretation in the Wegman Memo of the definition of “air
pollutant” in CAA section 302(g) is in question in light of
Massachusetts (finding this definition to be “sweeping”), EPA
believes the core rationale for its interpretation of the applicability
of Title V remains sound.  EPA continues to maintain its interpretation,
consistent with CAA sections 302(j), 501, 502 and 504(a), that the
provisions governing Title V applicability for “a major stationary
source” can only be triggered by emissions of pollutants subject to
regulation.  This interpretation is based primarily on the purpose of
Title V to collect all regulatory requirements applicable to a source
and to assure compliance with such requirements, see, e.g., CAA section
504(a), and on the desire to promote consistency with the approach under
the PSD program.

In applying this interpretation under Title V, the Wegman Memo also
explains that EPA does not consider CO2 to be a pollutant subject to
regulation based on the monitoring and reporting requirements of section
821 of the Clean Air Act Amendments of 1990.  As articulated in numerous
orders issued by EPA in response to petitions to object to Title V
permits, EPA views the Title V operating permits program as a vehicle
for ensuring that air quality control requirements are appropriately
applied to facility emission units and that compliance with these
requirements is assured.  See, e.g., In the Matter of Fort James Camas
Mill, Petition No. X-1999-1 at 3-4 (Dec. 22, 2000); In the Matter of
Cash Creek Generation, LLC, Petition Nos. IV-2008-1 & IV-2008-2 at 2
(Dec. 15, 2009).  The Wegman Memo points out that section 821 involves
reporting and study of emissions, but is not related to actual control
of emissions.  Since the reporting requirements of section 821 have no
connection to existing air quality control requirements, it is
appropriate not to treat them as making CO2 “subject to regulation”
for purposes of Title V.  Cf. Section 504(b) (providing EPA authority to
specify requirements for “monitoring and analysis of pollutants
regulated under this Act.”).

EPA has not previously explicitly considered the question of when a
pollutant becomes “subject to regulation” under this established
interpretation of the Title V requirements.  EPA received comments in
this reconsideration proceeding specifically on the question of when a
pollutant becomes subject to regulation for purposes of Title V.  In
light of these comments, and the decision to adopt a “takes effect”
approach for PSD, EPA believes it is appropriate to address this issue
for Title V with respect to GHG.  

We are mindful of the different purposes for the PSD and Title V
programs under the statute.  While PSD results in substantive control
requirements as necessary to meet air quality goals, Title V is focused
on identifying, collecting, and assuring compliance with other Act
requirements (including PSD), and generally does not itself result in
new control requirements.  Nevertheless, as reflected in the Wegman
Memo, the two programs have historically followed the same approach for
determining when a pollutant is “subject to regulation.”  We believe
that a “takes effect” approach to the triggering of new pollutants
is desirable and appropriate for Title V, for many of the reasons
described above for PSD.  We are therefore generally inclined to follow
the approach adopted today for PSD, and conclude that GHGs are
“subject to regulation,” for purposes of determining whether a
source of GHGs is a “major source” for Title V, no earlier than the
date on which a control requirement for GHGs “takes effect.”  EPA
currently anticipates that the LDV rule will be the first control
requirement for GHGs to take effect.  Under this approach, as with PSD,
if the LDV rule takes effect as of January 2, 2011, a source that is not
currently subject to Title V for its GHG emissions could become so no
earlier than January 2, 2011.  

Finally, we note that, as with PSD, we expect that, beyond January 2,
2011, there will remain significant administrative and programmatic
considerations associated with permitting of GHGs under Title V.  In
light of this, as discussed above with regard to PSD permitting, EPA
will be further addressing in the final Tailoring Rule (to be
promulgated in the near future) the manner in which sources can become
subject to Title V as a result of their GHG emissions. 

VII.	Statutory Authority

	The statutory authority for this action is provided by section 553 of
the Administrative Procedure Act (5 U.S.C. 553) and the Clean Air Act
(CAA), as amended (42 U.S.C. 7401 et seq.).  Relevant portions of the
CAA include, but are not necessarily limited to, sections 101, 165, 169,
301, 302, 307, 501, 502, and 504 (42 U.S.C. 7401, 7475, 7479, 7601,
7602, 7607, 7661, 7661a, and 7661d).

Page 115 of 115 - Prevention of Significant Deterioration (PSD): 
Reconsideration of Interpretation of Regulations that Determine
Pollutants Covered by the Federal PSD Permit Program

VIII. Judicial Review

This action is a nationally applicable final action under section 307(b)
of the Act.  As a result, any legal challenges to this action must be
brought to the United States Court of Appeals for the District of
Columbia Circuit by [INSERT DATE 60 DAYS FROM DATE OF PUBLICATION]. 

_______________________________

Dated:

_______________________________

Lisa P. Jackson,

Administrator.

 On December 15, 2009, EPA published the final endangerment and cause or
contribute findings for GHGs under section 202(a) of the CAA.  See 74 FR
66495.  

 The PSD Interpretive Memo also reflects EPA’s interpretation of
sections 165(a)(4) and 169(3) of the CAA, which use language similar to
the EPA regulations that are based on these provisions of the statute. 
The Memo discusses the Agency’s interpretation of the CAA and
concludes that the Agency’s interpretation of its regulations is not
precluded by the terms of the CAA.  

 On January 15, 2009, a number of environmental organizations that filed
this Petition for Reconsideration also filed a petition challenging the
PSD Interpretive Memo in U.S. Court of Appeals for the District of
Columbia Circuit. Sierra Club v. E.P.A., No. 09-1018 (D.C. Cir., filed
Jan. 15, 2009).  Thereafter, various parties moved to intervene in that
action or filed similar petitions challenging the Memo.  The
consolidated D.C. Circuit cases have been held in abeyance pending this
reconsideration process.  Id., Order (filed March 9, 2009).

  Because the grant of reconsideration directed the Agency to conduct
this reconsideration using a notice and comment process, the proposal
did not address the procedural challenge presented in the Petition for
Reconsideration. 

   On September 28, 2009, EPA proposed a rule establishing emissions
standards for new motor vehicles, starting with Model Year 2012, that
would reduce GHGs and improve fuel economy from motor vehicles.  This
proposal was a joint proposal by EPA and the U.S. Department of
Transportation (DOT), with DOT proposing to adopt corporate average fuel
economy (CAFE) standards for model years 2012 and after.  See 74 FR
49453.  

 In some cases, a commenter on the proposed reconsideration of the PSD
Interpretive Memo addressed an issue or topic that is under
consideration in the forthcoming PSD and Title V GHG Tailoring Rule. 
Accordingly, we refer the reader to that rulemaking for EPA responses to
those comments.

  The proposed “Tailoring Rule” can be found at 74 FR 55291 (Oct.
27, 2009).

 The CAA requires BACT for “each pollutant subject to regulation under
this Act.”  CAA §§165(a)(4), 169(3).  The United States Code refers
to “each pollutant regulated under this chapter,” which is a
reference to Chapter 85 of Title 42 of the Code, where the CAA is
codified.  See 42 U.S.C. §§7475(a)(4),  TA \s "42 U.S.C. §
7475(a)(4)"   7479(3)  TA \s "42 U.S.C. § 7479(3)"  .  For simplicity,
this notice generally uses “the Act” and the Clean Air Act section
numbers rather than the U.S. Code citation.  

  See 67 FR 80186-80289.

 Memorandum from Jonathan Z. Cannon, General Counsel to Carol M.
Browner, Administrator, entitled EPA’s Authority to Regulate
Pollutants Emitted by Electric Power Generation Sources (April 10,
1998).

 Notably, the legislative record refers to “State” emission limit,
and makes no note of this State emission limitation having broader
applicability.  

   See 74 FR 56259 (Oct. 30, 2009).

 To EPA’s knowledge, no court has required a rulemaking procedure when
the Agency seeks to issue or change its interpretation of a statute. 
Nevertheless, EPA has completed this notice and comment proceeding
before deciding to adopt the revised interpretation of the CAA described
in this notice. 

  The increments for emissions of the various oxides of nitrogen are
expressed as concentrations of nitrogen dioxide (NO2). 

 We note that the preamble to the Proposed GHG Tailoring Rule implicitly
assumed that a pollutant will become “subject to regulation” for PSD
and Title V at the same time (and, in one case, suggests that time will
be on promulgation of the LDV Rule).  The latter statement was based on
the interpretation in current PSD Interpretive Memorandum, but failed to
note that we had proposed to change that interpretation in the October 7
Reconsideration notice (signed the same day as the Proposed GHG
Tailoring Rule).  See 74 FR at 55300 and 55340-41.  

  Wegman Memo at 5.

 EPA notes that this date is also when we expect the first GHG control
program addressing stationary sources will be in place under the Act
(i.e., the PSD program). 

Draft – Do Not Cite or Quote  		March 19, 2010

For Internal EPA Deliberations Only

  PAGE   \* MERGEFORMAT  73 

