Oklahoma v. United States Environmental Protection Agency

FILED
United States Court of Appeals
Tenth Circuit

PUBLISH July 19, 2013
Elisabeth A. Shumaker
UNITED STATES COURT OF APPEALS Clerk of Court

TENTH CIRCUIT

STATE OF OKLAHOMA;
OKLAHOMA INDUSTRIAL
ENERGY CONSUMERS, an
unincorporated association,

Petitioners,
v. No. 12-9526
UNITED STATES
ENVIRONMENTAL PROTECTION
AGENCY,

Respondent.

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

SIERRA CLUB,

Intervenor-Respondent,

and

PACIFICORP; AMERICAN
COALITION FOR CLEAN COAL
ELECTRICITY; NATIONAL PARKS
CONSERVATION ASSOCIATION,

Amici Curiae.

OKLAHOMA GAS & ELECTRIC
COMPANY,

Petitioner,
v. No. 12-9527
 UNITED STATES
ENVIRONMENTAL PROTECTION
AGENCY,

Respondent.

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

SIERRA CLUB,

Intervenor-Respondent,

and

PACIFICORP; AMERICAN
COALITION FOR CLEAN COAL
ELECTRICITY; NATIONAL PARKS
CONSERVATION ASSOCIATION,

Amici Curiae.

PETITION FOR REVIEW OF FINAL DECISION ISSUED BY THE
UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
EPA-R06-OAR-2010-0190

E. Scott Pruitt, Oklahoma Attorney General, (P. Clayton Eubanks, Assistant
Attorney General; Michael Graves and Thomas P. Schroedter of Hall Estill,
Tulsa, Oklahoma, with him on the briefs), Oklahoma City, Oklahoma, for
Petitioners State of Oklahoma and Oklahoma Industrial Energy Consumers.

Brian J. Murray of Jones Day, Chicago, Illinois, (Thomas E. Fennell of Jones
Day, Dallas, Texas; Michael L. Rice of Jones Day, Houston, Texas; Charles T.
Wehland of Jones Day, Chicago, Illinois, on the briefs), for Petitioner Oklahoma
City Gas & Electric Company.

Stephanie J. Talbert, United States Department of Justice, Environment & Natural
Resources Division, Environmental Defense Section, Washington, D.C. (Ignacia

2
S. Moreno, Assistant Attorney General; M. Lea Anderson and Barbara Nann, Of
Counsel, United States Environmental Protection Agency, with her on the brief),
for Respondent.

Andrea Issod, (Elena Saxonhouse and Sanjay Narayan with her on the brief), San
Francisco, California, for the Intervenor, Sierra Club.

Michael G. Jenkins, Assistant General Counsel, PacifiCorp Energy, Salt Lake
City, Utah, and E. Blaine Rawson, Ray Quinney & Nebeker, P.C., Salt Lake City,
Utah, filed an amicus curiae brief on behalf of PacifiCorp, Amicus Curiae.

Paul M. Seby and Marian C. Larsen of Moye White LLP, Denver, Colorado, filed
an amicus curiae brief on behalf of the American Coalition for Clean Coal
Electricity, Amicus Curiae.

Stephanie Kodish, Knoxville, Tennessee, filed an amicus curiae brief on behalf of
the National Parks Conservation Association, Amicus Curiae.

Before BRISCOE, Chief Judge, KELLY and LUCERO, Circuit Judges.

BRISCOE, Chief Judge.

In these consolidated petitions for review, petitioners 1 challenge a final rule

promulgated by the United States Environmental Protection Agency under the

Clean Air Act. The petitioners argue that the EPA impermissibly rejected

Oklahoma’s plan to limit the emissions of sulfur dioxide at Oklahoma Gas and

Electric Company power plants and replaced it with its own more stringent

1
The petitioners include the state of Oklahoma, the Oklahoma Industrial
Energy Consumers interest group, and the Oklahoma Gas and Electric Company.

3
regulations, which petitioners contend usurped the state’s authority and will

require sizable expenditures on unnecessary technology. We conclude that the

EPA has authority to review the state’s plan and that it lawfully exercised that

authority in rejecting it and promulgating its own. Exercising our jurisdiction

under 42 U.S.C. § 7607(b)(1), we deny the petitions for review.

I

A. Statutory Background

The Clean Air Act “uses a cooperative-federalism approach to regulate air

quality.” U.S. Magnesium, LLC v. EPA, 690 F.3d 1157, 1159 (10th Cir. 2012).

Under the Clean Air Act (CAA), the Environmental Protection Agency (EPA)

must create and review national ambient air quality standards for certain

pollutants. See 42 U.S.C. §§ 7408, 7409. States then have the responsibility to

adopt state implementation plans (SIPs), “which provide[] for implementation,

maintenance, and enforcement” of those primary and secondary air quality

standards. § 7410(a)(1).

States, however, exercise this authority with federal oversight. The EPA

reviews all SIPs to ensure that the plans comply with the statute. The EPA may

not approve any plan that “would interfere with any applicable requirement” of

this chapter of the United States Code. § 7410(l). The EPA has a duty to create

its own federal implementation plan (FIP) if either: 1) it “finds that a State has

failed to make a required submission or finds that the plan or plan revision

4
submitted by the State does not satisfy the minimum criteria established under

subsection (k)(1)(A) of this section”; or 2) it “disapproves a State implementation

plan submission in whole or in part.” § 7410(c)(1). The duty to promulgate a FIP

exists “unless the State corrects the deficiency, and the Administrator approves

the plan or plan revision, before the Administrator promulgates such Federal

implementation plan.” Id.

At issue in this case are the portions of the CAA that seek to protect

visibility at certain national parks and wildlife areas. The CAA requires that the

EPA promulgate regulations “to assure . . . reasonable progress toward”

preventing any future and “remedying . . . any existing, impairment of visibility

in mandatory class I Federal areas which impairment results from manmade air

pollution.” § 7491(a)(1), (a)(4). It also requires that the EPA ensure that each

state plan “contain[s] such emission limits, schedules of compliance and other

measures as may be necessary to make reasonable progress toward meeting the

national goal.” § 7491(b)(2).

Relevant in this case are the CAA’s mandates regarding sources that

contribute to visibility impairments. SIPs must include:

except as otherwise provided . . . a requirement that each
major stationary source which is in existence on August 7,
1977, but which has not been in operation for more than fifteen
years as of such date, and which, as determined by the State
(or the Administrator in the case of a [FIP]) emits any air
pollutant which may reasonably be anticipated to cause or
contribute to any impairment of visibility in any such area,

5
 shall procure, install, and operate, as expeditiously as
practicable (and maintain thereafter) the best available retrofit
technology, as determined by the State (or the Administrator in
the case of a [FIP]) for controlling emissions from such source
for the purpose of eliminating or reducing any such
impairment.

§ 7491(b)(2)(A). To simplify, a state—or the EPA, when promulgating a

FIP—must: 1) determine which of the eligible major stationary sources in their

state contributes to visibility impairment; and then 2) determine the “best

available retrofit technology” for controlling the emissions causing that

impairment at that source. Id. When determining “best available retrofit

technology” (BART):

the State (or the Administrator in determining emission
limitations which reflect such technology) shall take into
consideration [1] the costs of compliance, [2] the energy and
nonair quality environmental impacts of compliance, [3] any
existing pollution control technology in use at the source, [4]
the remaining useful life of the source, and [5] the degree of
improvement in visibility which may reasonably be anticipated
to result from the use of such technology.

§ 7491(g)(2).

The CAA requires that the EPA create guidelines for the states “on

appropriate techniques and methods for implementing this section.” § 7491(b)(1).

For “a fossil-fuel fired generating powerplant having a total generating capacity

in excess of 750 megawatts, the emission limitations required under this

paragraph shall be determined pursuant” to the regulations promulgated by the

EPA. § 7491(b). The EPA has promulgated these BART guidelines at 40 C.F.R. §

6
51.308(e).

B. Procedural Background

In 2005, the EPA issued an updated version of its Regional Haze Rule that

required states to submit SIP revisions by December 17, 2007. See Regional

Haze Program Requirements, 40 C.F.R. § 51.308(b). On January 15, 2009, the

EPA took final action in finding that Oklahoma—along with 31 other states, the

District of Columbia, and the U.S. Virgin Islands—failed to submit a SIP that

addressed any of the Regional Haze elements by this deadline. See Finding of

Failure To Submit State Implementation Plans Required by the 1999 Regional

Haze Rule, 74 Fed. Reg. 2392-01 (Jan. 15, 2009). This triggered the EPA’s duty

to promulgate a federal implementation plan within two years. See 42 U.S.C. §

7410(c)(1).

Before the EPA promulgated a FIP, however, Oklahoma submitted its SIP.

See Oklahoma Regional Haze State Implementation Plan, Joint Appendix (JA) at

55 (Feb. 17, 2010). At issue in this petition are the SIP’s BART determinations

with respect to two units at Oklahoma Gas & Electricity’s (OG&E’s) Muskogee

Generating Station and two units at its Sooner Generating Station. The Oklahoma

SIP set a sulfur dioxide (SO 2) emissions limits of 0.65 lb/mmBtu (thirty-day

average) and 0.55 lb/mmBtu (annual average) for each of these four units. See

OG&E Muskogee Generating Station BART Review, JA at 187 (Jan. 15, 2010);

OG&E Sooner Generating Station BART Review, JA at 221 (Jan. 15, 2010). The

7
BART for each of these units included OG&E’s continued use of low-sulfur coal.

The SIP considered, but rejected, an emissions limit that would require the

installation of so-called scrubbers to remove SO 2. See Muskogee BART Review,

JA at 213; Sooner BART Review, JA at 247. “The cost for [dry scrubbers] is too

high, the benefit too low and these costs, if borne, further extend the life

expectancy of coal as the primary fuel in the Sooner [and Muskogee] facilit[ies]

for at least 20 years and beyond,” according to OG&E’s BART analyses. See id.

On March 22, 2011, the EPA proposed a rule that would partially approve

and partially disapprove Oklahoma’s SIP. Proposed Rule, 76 Fed. Reg.

16,168-01, 16,169 (Mar. 22, 2011). The SO 2 emission limitations for OG&E’s

four units were among the parts of the SIP that the EPA proposed disapproving.

The EPA said that Oklahoma failed to follow the promulgated regulations in

determining BART. Id. at 16,182. Specifically, the EPA said that Oklahoma “did

not properly ‘take into consideration the costs of compliance’ when it relied on

cost estimates that greatly overestimated the costs of dry and wet scrubbing to

conclude these controls were not cost effective.” Id. (quoting 40 C.F.R. §

51.308(e)(1)(ii)(A)).

“Given that scrubbers are typically considered to be highly cost-effective

controls for power plants such as those at issue, [the EPA] retained a consultant to

independently assess the suitability and costs of installing these controls.” Id.

The EPA found the scrubbers to be substantially more cost effective than

8
Oklahoma did. Id. at 16,183. For example, Oklahoma estimated the cost of the

scrubbers to be $7,147 per ton of SO 2 removed at one of the Sooner Generating

Station units. Id. The EPA projected scrubbers at that same unit would cost

$1,291 per ton of SO 2 removed. Id.

In addition to proposing the partial disapproval of the SIP, the EPA

proposed creating its own federal implementation plan in the same action. Id. at

16,168. The EPA proposed an SO 2 emissions limit of 0.06 lb/mmBtu (thirty-day

average). Id. at 16,193-94. Based on this limit, the EPA believed the use of dry

scrubbers would be cost effective. Id. at 16,183. After notice and comment, the

EPA published the final rule enacting these emissions limits. See Final Rule, 76

Fed. Reg. 81,728-01 (Dec. 28, 2011).

On February 24, 2012, the state of Oklahoma and the Oklahoma Industrial

Energy Consumers filed in this court a petition seeking review of the final rule

(Case No. 12-9526). OG&E filed its petition for review the same day (Case No.

12-9527). We later issued an order granting a motion to consolidate these

petitions.

The petitioners also took steps to stay the application of the rule. The same

day they filed petitions for review, the petitioners filed with the EPA a motion for

reconsideration and a request for an administrative stay. 2 The petitioners also

2
Under the CAA, the filing of a petition for reconsideration does not affect
(continued...)

9
filed a motion in this court seeking a stay pending a hearing on the merits. A

two-judge panel of this court granted the petitioners’ motion to stay the portion of

the rule requiring the reduction of SO 2 emissions at these four OG&E units.

Oklahoma v. EPA, Nos. 12-9526 and 12-9527, at 1-2 (10th Cir. June 22, 2012).

Meanwhile, appellate briefing progressed.

The petitioners raise a number of objections to the final rule, arguing that

the EPA has usurped the state’s authority in an effort to force OG&E to spend

more than one-billion dollars to install unnecessary technology in the next five

years. First, they argue that the EPA exceeded its statutory authority by

disapproving Oklahoma’s BART determination. Second, they argue that, even if

the EPA had this authority, the EPA acted arbitrarily and capriciously by

disapproving Oklahoma’s SIP. Third, they argue that the EPA acted arbitrarily

and capriciously in promulgating its FIP. Fourth, the petitioners argue that the

2
(...continued)
the finality of an EPA action for the purposes of judicial review. See 42 U.S.C. §
7607(b)(1) (“The filing of a petition for reconsideration by the Administrator of
any otherwise final rule or action shall not affect the finality of such rule or
action for purposes of judicial review nor extend the time within which a petition
for judicial review of such rule or action under this section may be filed, and shall
not postpone the effectiveness of such rule or action.”); see also Natural Res. Def.
Council v. Abraham, 355 F.3d 179, 203 n.11 (2d Cir. 2004). While the Third
Circuit has held that a pending petition for reconsideration deprived it of
jurisdiction under the CAA, see W. Penn Power Co. v. EPA, 860 F.2d 581, 587-
88 (3d Cir. 1988), it reached this result before the CAA was amended to prevent
petitions for reconsideration from affecting finality. Clean Air Act, Amendments,
Pub. L. No. 101-549, § 706, 104 Stat. 2399 (1990).

10
EPA failed to provide them adequate notice of aspects of the final rule. Finally,

the petitioners argue the EPA violated the CAA by promulgating the FIP in the

same action in which it partially disapproved of the SIP and after the two-year

deadline to promulgate a FIP had expired.

II

The petitioners argue that the EPA exceeded its statutory authority by

rejecting Oklahoma’s BART determinations and replacing them with its own.

The petitioners say that the EPA’s action tramples on the discretion that Congress

afforded states to make these decisions. The CAA’s cooperative-federalism

policy supports this view, the petitioners say. More specifically, the petitioners

point to the statute’s legislative history and its language—mandating BART “as

determined by the State.” In the petitioners’ view, this all indicates that the

statute unambiguously prescribes a limited role for the EPA as regards BART

determinations.

In interpreting the CAA, we must follow the guidance set forth in Chevron,

U.S.A., Inc. v. Natural Res. Def. Council, Inc., 467 U.S. 837 (1984). “If the

statute is clear, we apply its plain meaning and the inquiry ends.” Ariz. Pub.

Serv. Co. v. EPA, 562 F.3d 1116, 1123 (10th Cir. 2009) (quotation omitted). “If

the statute is silent or ambiguous about the question at issue . . . we defer to the

authorized agency and apply the agency’s construction so long as it is a

reasonable interpretation of the statute.” Id. (quotation omitted).

11
“[A]dministrative implementation of a particular statutory provision qualifies for

Chevron deference when it appears that Congress delegated authority to the

agency generally to make rules carrying the force of law, and that the agency

interpretation claiming deference was promulgated in the exercise of that

authority.” United States v. Mead Corp., 533 U.S. 218, 226-27 (2001).

We agree with the EPA that the statute provides the agency with the power

to review Oklahoma’s BART determination for these four units. The EPA may

not approve any plan revision “if the revision would interfere with any applicable

requirement concerning attainment and reasonable further progress . . . or any

other applicable requirement of this chapter.” 42 U.S.C. § 7410(l). And under

§ 7410(a)(2)(J) SIPs must “meet the applicable requirements of . . . part C of this

subchapter”—which includes the provisions of the CAA related to visibility. See

§§ 7491, 7492.

The visibility statute itself requires the EPA to promulgate regulations that

“require each applicable implementation plan . . . to contain such emission limits,

schedules of compliance and other measures as may be necessary to make

reasonable progress toward meeting the national goal.” § 7491(b)(2). That

includes a requirement that the state make BART determinations. And while it is

undoubtedly true that the statute gives states discretion in balancing the five

BART factors, it also mandates that the state adhere to certain requirements when

conducting a BART analysis. The state plan must include a BART determination

12
for any eligible plant that “may reasonably be anticipated to cause or contribute to

any impairment of visibility in any such area.” § 7491(b)(2). In addition,

§ 7491(b) requires that the BART determination for units at power plants like

those at issue here—having a total generating capacity of greater than 750

megawatts—“shall be determined pursuant” to the EPA regulations. See 42

U.S.C. § 7491(b) (“In the case of a fossil-fuel fired generating powerplant having

a total generating capacity in excess of 750 megawatts, the emission limitations

required under this paragraph shall be determined pursuant to guidelines,

promulgated by the Administrator under paragraph (1).”); see also EPA Br. at 7.

As required by the statute, the EPA has promulgated regulations providing

guidelines for making BART determinations. Like the statute, the regulations

require that BART determinations at large power plants follow these guidelines.

40 C.F.R. § 51.308(e)(1)(ii)(B) (“The determination of BART for fossil-fuel fired

power plants having a total generating capacity greater than 750 megawatts must

be made pursuant to the guidelines appendix Y of this part (Guidelines for BART

Determinations Under the Regional Haze Rule).”). The EPA rejected Oklahoma’s

SIP because the BART determinations failed to comply with these guidelines.

See EPA Br. at 22 (“Specifically, EPA concluded that Oklahoma failed to

reasonably consider the ‘cost of compliance’ factor by calculating costs as

required by the BART guidelines, which led to an ‘unreasoned and unjustified’

BART determination.”). Given that the statute mandates that the EPA must

13
ensure SIPs comply with the statute, we fail to see how the EPA would be without

the authority to review BART determinations for compliance with the guidelines.

The D.C. Circuit’s opinion in American Corn Growers Ass’n v. EPA, 291

F.3d 1 (D.C. Cir. 2002), does not alter this conclusion. At issue in Corn Growers

was a provision of the Regional Haze Rule that required states to make BART

decisions based in part on the geographical location of a source, as opposed to its

actual emissions. Id. at 4-5. The rule required BART-eligible sources be subject

to BART “even absent empirical evidence of that source’s individual contribution

to visibility impairment in a Class I area so long as the source is located within a

region that may contribute to visibility impairment.” Id. at 5. When making the

BART determination, the state needed to “analyze the degree of visibility

improvement that would be achieved . . . as a result of the emission reductions

achievable from all sources subject to BART located within the region that

contributes to visibility impairment.” Id. at 6 (quotation and emphasis omitted).

The D.C. Circuit held the EPA’s approach was “inconsistent” with the CAA. Id.

at 7-8.

The D.C. Circuit cited two ways in which the rule was inconsistent with the

statute. First, the EPA’s approach “distort[ed] the judgment Congress directed

the states to make for each BART-eligible source” by treating one of the five

BART factors differently than the others. Id. at 6. The rule, for instance,

prevented a state from “consider[ing] the degree to which new equipment at a

14
particular source would help cure the haze in some distant national park.” Id. at

7. “Under EPA’s take on the statute, it is therefore entirely possible that a source

may be forced to spend millions of dollars for new technology that will have no

appreciable effect on the haze in any Class I area.” Id.

Second, the D.C. Circuit said that the rule impermissibly “constrain[ed]

authority Congress conferred on the states.” Id. at 9. The court said that the

statute and the legislative history suggested that the states had broad authority to

weigh the statutory factors and make BART determinations. Id. at 8. The D.C.

Circuit noted that the Conference Report on the 1977 amendments to the CAA

specifically referenced “an agreement to reject the House bill’s provisions giving

EPA the power to determine whether a source contributes to visibility impairment

and, if so, what BART controls should be applied to that source.” Id. The

agreement instead added the language delegating this authority to the state. Id.

“The Conference Report thus confirms that Congress intended the states to decide

which sources impair visibility and what BART controls should apply to those

sources.” Id. The Haze Rule, though, “ties the states’ hands and forces them to

require BART controls at sources without any empirical evidence of the particular

source’s contribution to visibility impairment in a Class I area.” Id.

Here, though, the statute and the legislative history support our conclusion

that the EPA may reject BART determinations that do not comply with the

guidelines. True, the modification of the original House bill reflects an intent to

15
shift the power to determine BART from the EPA to the states. But, as above, it

still placed statutory limits on those state decisions. While the legislative history

may evidence an intent to prevent the EPA from directly making those BART

decisions, it does not necessarily evidence an intent to deprive the EPA of any

authority to ensure that these BART decisions comply with the statute. In the

present case, the EPA did not reject the petitioners’ BART determination because

it disagreed with the way it balanced the five factors. It rejected the BART

determination because it failed to follow the guidelines—as required by the

statute—in calculating one of those factors.

All the conference agreement referenced by the D.C. Circuit did was shift

the initial responsibility for making BART determinations from the EPA to the

state. But that does not differ from other parts of the CAA—states have the

ability to create SIPs, but they are subject to EPA review. In addition, the

Conference Report emphasized that the BART determinations for large power

plants must comply with EPA guidelines:

The agreement clarifies that the State, rather than the
Administrator, identifies the source that impairs visibility in
the Federal class I areas identified and thereby fall within the
requirements of this section.
....
In establishing emission limitations for any source
which impairs visibility, the State shall determine what
constitutes ‘best available retrofit technology’ (as defined in
this section) in establishing emission limitations on a source-
by-source basis to be included in the State implementation plan
so as to carry out the requirements of this section. The

16
 regulations and Federal guidelines required by the House
passed bill for determining this technology are eliminated for
all sources other than fossil fuel electric generating plants
with a total generating capacity in excess of 750 megawatts.

H.R. Rep. No. 95-564, at 155 (1977) (Conf. Rep.) (emphasis added).

The Senate discussion about the Conference Report also highlighted the role that

the guidelines play in BART determinations for large power plants:

[Senator] McClure. And while those existing sources are
limited to the 28 major sources contained in the Senate bill’s
definition of major emitting facilities, exempting any such
source which has the maximum potential to emit less than 250
tons per year, Federal guidelines apply only to fossil-fuel fired
generating plants in excess of 750 megawatts?

[Senator] Muskie. That is correct.

[Senator] McClure. Under the conference agreement, does the
State retain sole authority for identification of sources for the
purpose of visibility issues under this section?

[Senator] Muskie. Yes; the State, not the Administrator,
identifies a source that may impair visibility and thereby falls
within the requirement of section 128.

[Senator] McClure. And does this also hold true for
determination of “Best Available Retrofit Technology”?

[Senator] Muskie. Yes; here again it is the State which
determines what constitutes “Best Available Retrofit
Technology,” as defined in section 128. The Federal
guidelines apply only to the large powerplants we have
described.

123 Cong. Rec. S26,854 (daily ed. Aug. 4, 1977) (emphasis added). The last

sentence—omitted by petitioners in their brief—makes clear that the statute

17
requires that the BART determination here comply with the guidelines. See Pet.

Opening Br. at 15. And because the EPA monitors SIPs for compliance with the

statute, it must monitor BART determinations for compliance with the guidelines.

To be sure, the guidelines themselves might somehow conflict with the statute.

But the petitioners have not argued that any conflict exists here. 3 We therefore

hold that the EPA had the authority to review Oklahoma’s BART determination

3
In its amicus brief, the American Coalition for Clean Coal Electricity
asserts that some conflict between the guidelines and statute may exist because:

EPA can provide the States with guidelines only “on
appropriate techniques and methods,” including “(A) methods
for identifying, characterizing, determining, quantifying, and
measuring visibility impairment in Federal areas referred to in
paragraph (1), (B) modeling techniques (or other methods) for
determining the extent to which manmade air pollution may
reasonably be anticipated to cause or contribute to such
impairment, and methods for preventing and remedying such
manmade air pollution and resulting visibility impairment.”
See cross-reference from [42 U.S.C. § 7491(b)] (last
paragraph) to [§ 7941(b)(1) to § 7491(a)(3)]. Thus, EPA’s role
is to provide procedural and technical guidance to the States in
making BART determinations.

See Am. Coalition for Clean Coal’s Br. at 12 (emphasis added). However, the
statute does not limit the guidelines so restrictively. First, the guidelines must be
made as part of regulations that ensure “reasonable progress toward meeting the
national goals” specified in the statute. § 7491(a)(4), (b)(1). Second, those
guidelines must merely “tak[e] into account” recommendations from a report to
Congress on the methods and techniques referenced in the statute, 42 U.S.C. §
7491(b)(1), which includes “methods for preventing and remedying such
manmade air pollution and resulting visibility impairment.” 42 U.S.C. §
7491(a)(3)(C). Moreover, the amicus brief fails to explain why the EPA could
provide these regulations providing procedural and technical guidance, but yet
lacks the authority to ensure states complied with them.

18
with respect to these two power plants.

III

Having held that the EPA possesses the authority to review these BART

decisions, we must now determine whether the EPA lawfully exercised that

authority when it rejected Oklahoma’s SIP. Petitioners argue that the EPA took

arbitrary and capricious action in rejecting two sets of cost estimates they used in

determining BART. The EPA, on the other hand, argues that it properly rejected

these estimates—and, thus, the SIP that relied on them—for failure to comply

with its guidelines.

We follow the standards of the Administrative Procedure Act (APA) in

reviewing the EPA’s actions under the CAA. See Magnesium, 690 F.3d at 1164.

Under the APA, we must hold unlawful any agency action that is “arbitrary,

capricious, an abuse of discretion, or otherwise not in accordance with law.” 5

U.S.C. § 706(2). “Under the arbitrary or capricious standard, we must determine

whether the agency considered the relevant data and rationally explained its

decision.” Ariz. Pub. Serv. Co., 562 F.3d at 1122. “Agency action is arbitrary or

capricious if the agency has relied on factors which Congress has not intended it

to consider, entirely failed to consider an important aspect of the problem, offered

an explanation for its decision that runs counter to the evidence before the

agency, or is so implausible that it could not be ascribed to a difference in view or

the product of agency expertise.” Id. at 1123. (quotation omitted). “Even when

19
an agency explains its decision with less than ideal clarity, a reviewing court will

not upset the decision on that account if the agency’s path may reasonably be

discerned.” Alaska Dep’t of Envtl. Conservation, 540 U.S. at 497 (quotation

omitted). In addition, we note that “[w]hen an agency interprets its own

regulation, the Court, as a general rule, defers to it unless that interpretation is

plainly erroneous or inconsistent with the regulation.” Decker v. Nw. Envtl. Def.

Ctr., 133 S. Ct. 1326, 1337 (2013) (quotation omitted).

A. 2008 Cost Estimates

The petitioners argue that the EPA arbitrarily rejected a set of cost

estimates that OG&E submitted to the EPA in 2008 (2008 Cost Estimates). The

petitioners claim that the “EPA acknowledged that ‘OG&E did utilize the “EPA

Air Pollution Control Cost Manual” when constructing its [May 2008] cost

estimates.’” Pet. Opening Br. at 20 (quoting EPA Region 6 Comments on

ODEQ’s BART Engineering Analyses, JA at 1132 (Nov. 4, 2008)). The 2008

Cost Estimates were “more than ten times EPA’s stated average costs per ton for

this technology, and nearly five times as much as the upper limit of EPA’s

expected cost range.” Id. at 21. The EPA should have addressed these numbers,

petitioners say, because they prove the scrubbers were not cost effective.

This argument is without merit. The EPA never stated that the 2008 Cost

20
Estimates complied with the Control Cost Manual. 4 In context, the EPA simply

acknowledged that OG&E purported to have used the manual in constructing

these estimates. Indeed, it requested that OG&E note any deviations from the

cost manual—as required by the guidelines 5—after pointing out that OG&E’s

estimates “seem[ed] high compared to what EPA has seen in other BART

analysis.” EPA Region 6 Comments, JA at 1132; see also 40 C.F.R. pt. 51 app.

4
The entire EPA comment included the following:

Regarding its cost estimates, OG&E’s estimates seem
high compared to what EPA has seen in other BART
analyses. OG&E cites increased equipment costs, in
part due to the “sellers market” that resulted from the
CAIR program. Since the CAIR has been vacated,
OG&E should solicit revised bids from pollution control
equipment vendors. Region 6 is aware of similarly sized
and configured facilities that estimate much lower costs
for the installation of wet or dry FGD systems. Region 6
notes that OG&E did utilize the “EPA Air Pollution
Control Cost Manual” when constructing its cost
estimates. However, OG&E should also note any areas
in which where it has deviated from that guidance.

EPA Region 6 Comments, JA at 1132 (emphases added).
5
The guidelines state that “cost estimates should be based on the OAQPS
Control Cost Manual, where possible.” 40 C.F.R. pt. 51 app. Y(IV)(D)(4)(a).
However, it also instructs that parties “should include documentation for any
additional information regarding purchased equipment costs, equipment life,
replacement of major components, and any other element of the calculation that
differs from the Control Cost Manual.” 40 C.F.R. pt. 51 app. Y(IV)(D)(4)(a)
n.15.

21
Y(IV)(D)(4)(a) n.15. 6

And in any event, the EPA did not “ignore[]” the 2008 Cost Estimates.

Rather, the EPA explained that “[t]hese 2008 costs are not valid under the

overnight costing method” required by the manual. Response to Technical

Comments, JA at 1236. The 2008 Cost Estimates “contain[ed] . . . fundamental

methodological flaws, such as including escalation and Allowance for Funds Used

During Construction (AFUDC).” Id. “The cost of scrubbers would not be

substantially higher than those reported for other similar projects if OG&E had

used the costing method and basis, i.e., overnight costs in current dollars,

prescribed by the Control Cost Manual,” the EPA said. Id. The EPA therefore

had a reasonable basis for rejecting the 2008 Cost Estimates as not complying

6
Petitioners argue in their reply brief that the EPA did not explicitly raise
this response in its comments and therefore cannot use this reasoning to justify its
decision on appellate review. Pet. Reply Br. at 10 n.1 (quoting Motor Vehicles
Mfrs. Ass’n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 50 (1983) (“[A]n
agency’s action must be upheld, if at all, on the basis articulated by the agency
itself.”)). But an agency “need not address every comment, [although] it must
respond in a reasoned manner to those that raise significant problems.” Covad
Commc’ns Co. v. FCC, 450 F.3d 528, 550 (D.C. Cir. 2006) (quotation omitted).
“The failure to respond to comments is significant only insofar as it demonstrates
that the agency’s decision was not based on a consideration of the relevant
factors.” Id. (citations and quotation omitted). “In making the ‘keystone’ inquiry
whether the [agency] engaged in reasoned decisionmaking, the reviewing court is
to consider the larger administrative record.” Mt. Diablo Hosp. v. Shalala, 3 F.3d
1226, 1234 (9th Cir. 1993) (quotation omitted).
Here, the letter itself—which is part of the record—quite clearly explains
the EPA’s concerns about the 2008 Cost Estimates. And the petitioners did, in
fact, submit new cost estimates.

22
with the guidelines.

B. The 2009 Cost Estimates

The petitioners also argue that the EPA acted arbitrarily in rejecting the

cost estimates submitted in 2009 (2009 Cost Estimates). These cost estimates

included more site-specific data than the ones OG&E initially submitted. The

petitioners say that the EPA should defer to Oklahoma’s determination that the

“site-specific cost information submitted by OG&E in 2009 was credible,

detailed, and specific for the individual facilities, going well beyond the default

methodology recommended by the EPA guidance.” Pet. Opening Br. at 23

(quotation omitted). This argument appears to be premised, in part, on our

accepting the representation that the EPA conceded the 2008 Cost Estimates

complied with the guidelines. The petitioners argue in their reply brief that for

“EPA to acknowledge receipt of cost estimates that it agreed were in compliance

with the CCM, ask for site-specific cost estimates that go beyond the CCM, and

then reject those site-specific estimates for not conforming to the CCM

exemplifies arbitrary and capricious results-oriented decisionmaking.” Pet. Reply

Br. at 11.

We disagree with the petitioners’ characterization of the EPA’s actions.

The EPA did not reject the mere use of any site-specific costs as not complying

with the guidelines. It rejected this particular use of site-specific numbers

because it “recognized that how OG&E specified those vendor quotes and [its]

23
subsequent use of them in its cost analysis was flawed.” Response to Technical

Comments, JA at 1308. The guidelines require that states provide support for any

site-specific costs that depart from the generic numbers in the Control Cost

Manual. See 40 C.F.R. pt. 51 app. Y(IV)(D)(4)(a) n.15 (“You should include

documentation for any additional information you used for the cost calculations,

including any information supplied by vendors that affects your assumptions

regarding purchased equipment costs, equipment life, replacement of major

components, and any other element of the calculation that differs from the

Control Cost Manual.”). OG&E never delivered to the EPA these detailed vendor

estimates, preventing the EPA from conducting an adequate review to ensure

these departures from the manual—and Oklahoma’s approval of them—were

justified. See 76 Fed. Reg. at 81,745 (“[M]uch of the documentation OG&E and

others cite to support deviations from the Control Cost Manual was not provided

to us. Thus, we were unable to analyze their contents and determine whether

these deviations were appropriate.”); see, e.g., Response to Technical Comments,

JA at 1239 (“For instance, although OG&E provided two spreadsheets that listed

their cost line items, these spreadsheets, each over 600 lines in length (and

including line items such as seeding and fertilizing the grass at the plant sites),

were stripped of all cell calculations, preventing any meaningful review.”). And,

as we discuss below in evaluating the EPA’s action in promulgating its FIP, many

24
of OG&E’s costing assumptions were unjustified. 7

Moreover, the EPA’s consultant noted that, even if the departures from the

generic numbers had been properly documented, the resulting costs were not

analyzed in compliance with the manual. See Revised BART Cost-Effectiveness

Analysis, JA at 1517 (October 2010) (“[I]t is possible to follow the generic

costing method in the Cost Manual, relying on vendor quotes and other

information to estimate scrubber capital and O&M costs. [OG&E’s consultants]

used vendor estimates for equipment costs but did not follow the generic costing

method.”). The guidelines say that states should follow the manual’s

methodology so that projects can be more easily compared. The EPA said that

OG&E should have used the “overnight” costing methodology. Instead, “OG&E

and others incorrectly assume that BART cost effectiveness should be based on

the ‘all-in’ cost method, which includes all of the costs of a financial transaction,

including interest, commissions, and any other fees from a financial transaction

up to the date that the project goes into operation, as of the assumed commercial

operating dates of the scrubbers, 2014 and 2015.” 76 Fed. Reg. at 81,744.

For their part, the petitioners argue generally that the Control Cost Manual

7
We recognize that the EPA has less discretion when it takes actions to
reject a SIP than it does when it promulgates a FIP. However, we believe that the
EPA had reason to make the adjustments described in Section IV, Part B, even
under the higher standard we would apply when evaluating its actions in rejecting
a SIP. OG&E has yet to provide any justification for providing estimates that
departed from the guidelines.

25
does not require the use of the “overnight” method used by the EPA. However,

they do not point to any specific parts of the guidelines or the Control Cost

Manual that contradict the EPA’s approach. Instead, the petitioners argue that the

EPA itself conceded that its methodology excluded only inflation. Pet. Opening

Br. at 31. (“EPA claimed that the CCM required compliance with a ‘constant

dollar’ approach . . . . The constant dollar approach allows comparability in the

BART context by removing the effects of inflation from cost estimates.”).

Petitioners believe this means that it properly included the other costs, such as the

Allowance for Funds Used During Construction, in its analyses.

Unlike the petitioners, we do not read the EPA’s statement as proof that the

EPA believed the constant-dollar method requires removing only inflation. The

EPA’s consultant referenced the manual’s adherence to the “constant-dollar”

method as the reason for excluding inflation in the estimates. See Revised BART

Cost-Effectiveness Analysis, JA at 1517 (“The cost metric estimated in the

Manual is real or constant-dollar costs in that the effect of inflation has been

removed.”). But the consultant never said that the Control Cost Manual excluded

only inflation. In fact, the consultant explained throughout her report that the

Control Cost Manual also required excluding many of the other costs on which

OG&E had relied. See JA at 1519 (“Cost items such as escalation of costs, bond

cost, and AFUDC are not part of the Cost Manual methodology for estimating

costs.”); see, e.g., JA at 1520 (“[Allowance for Funds Used During Construction]

26
are not part of the constant dollar approach found in the EPA Control Cost

Manual and should not be included in the BART cost-effectiveness analysis.”).

Viewing these comments as a whole, we do not believe the EPA or its consultant

ever conceded inflation was the only cost that needed to be eliminated from

OG&E’s estimates.

The petitioners also refer to two affidavits from their experts that detail

what they believe is the proper costing methodology. Even if we permitted these

arguments to be incorporated by reference, we cannot consider these affidavits

because they are outside of the administrative record. 42 U.S.C. § 7607(d)(7)(A).

Aware of this hurdle, petitioners criticize the EPA’s procedure, arguing

that they had no opportunity to object to the EPA’s use of the so-called

“overnight” costing method because the EPA used it for the first time in the final

rule. To be sure, the EPA used the term “overnight” method for the first time in

the final rule. However, the EPA excluded the same costs in the final rule that it

did throughout the entire process—the “overnight” method was simply the

shorthand it used in the Final Rule to describe the exclusion of these costs. In

fact, the petitioners’ own comments to the EPA belie the argument they have

made to us, since they challenged the exclusion of these costs in the

administrative proceeding. See, e.g., Ex. B to OG&E’s Comments: May 2011

BART Cost Analysis Report, JA at 1156 (May 20, 2011) (“EPA’s consultant

incorrectly argues that an AFUDC is not part of the constant dollar approach

27
found in the EPA Control Cost Manual and should not be included in the BART

cost-effectiveness analyses.” (quotation omitted)). We see no reason to excuse

petitioners’ failure to raise these substantive arguments in their brief.

Additionally, we do not have jurisdiction to consider any procedural error

that might have occurred as a result of the EPA allegedly using the “overnight”

method for the first time in the Final Rule. Under the CAA, “[o]nly an objection

to a rule or procedure which was raised with reasonable specificity during the

period for public comment . . . may be raised during judicial review.” 42 U.S.C.

§ 7607(d)(7)(B). The fact that the petitioners could not have raised their

objection about the use of the “overnight” method until after the final rule was

published does not excuse them from the requirement they first raise the issue

with the EPA. “Rather, the CAA requires a petitioner to first raise its objection to

the agency th[r]ough a petition for reconsideration.” Appalachian Power Co. v.

EPA, 249 F.3d 1032, 1065 (D.C. Cir. 2001). Petitioners here, though, filed their

petition for reconsideration the same day they filed this petition for review. We

therefore lack jurisdiction to rule on this procedural objection. See 42 U.S.C. §

7607(d)(7)(B) (“If the Administrator refuses to convene [a reconsideration

proceeding], such person may seek review of such refusal in the United States

court of appeals for the appropriate circuit . . . .”) (emphasis added).

IV

Petitioners also challenge decisions made by the EPA in promulgating the

28
FIP. First, the petitioners argue that the EPA itself failed to follow the guidelines

because it did not base its cost-effectiveness analysis on the historical emissions

baseline. Second, petitioners argue that the EPA based its analysis on incorrect

technological assumptions about the size of the scrubber that needs to be built.

Third, the petitioners criticize many of the adjustments the EPA made to its cost

estimates. Finally, the petitioners assert that “[s]crubbers at the OG&E units

would not have a significant impact on visibility.” Id. at 36. We review these

challenges under the same arbitrary and capricious standard we used to evaluate

the EPA’s rejection of the SIP. However, we do so while recognizing this

requires a slightly different perspective: evaluating the EPA’s own choices under

the guidelines, as opposed to evaluating its choice to reject the Oklahoma SIP

under the guidelines.

A. Baseline Emissions/Technical Feasibility

Petitioners argue the EPA acted arbitrarily when it promulgated its FIP

because it ignored the units’ past rates of SO 2 emissions in conducting its

analysis. Under the guidelines, “[a]verage cost effectiveness means the total

annualized costs of control divided by annual emissions reductions (the difference

between baseline annual emissions and the estimate of emissions after controls).”

40 C.F.R. pt. 51 app. Y(IV)(D)(4)(c). “The baseline emissions rate should

represent a realistic depiction of anticipated annual emissions for the source.” Id.

at Y(IV)(D)(4)(d). “In general, for the existing sources subject to BART, you

29
will estimate the anticipated annual emissions based upon actual emissions from a

baseline period.” Id. “When you project that future operating parameters (e.g.,

limited hours of operation or capacity utilization, type of fuel, raw materials or

product mix or type) will differ from past practice, and if this projection has a

deciding effect in the BART determination, then you must make these parameters

or assumptions into enforceable limitations.” Id.

OG&E has been voluntarily using low-sulfur coal at these power plants. It

therefore estimated that scrubbers would remove around 14,000 tons of SO 2 per

year at each of the power plants. Comments of OG&E on Proposed EPA Rule, JA

at 1106. The EPA, however, assumed that OG&E would begin to use high-sulfur

coal if it installed the proposed scrubbers. The EPA, therefore, estimated that

scrubbers would remove 43,428 tons of SO 2 per year at one of the power plants,

and 46,458 tons of SO 2 per year at the other. Id. By assuming OG&E would

remove larger amounts of SO 2, while using cheaper, high-sulfur coal, the resulting

conclusion was that the scrubbers appeared more cost effective.

This links with petitioners’ second objection: that the EPA impermissibly

based its analysis on the construction of smaller, less expensive—but allegedly

technically infeasible—scrubbers. Petitioners claim that EPA wrongly assumed

that OG&E had the option of building a smaller scrubber based on the sulfur

content of the coal it would burn. Scrubber size does not depend on the sulfur

content of coal, the petitioners say. Rather, scrubber size must “reflect the

30
maximum potential heat input from the facility, and that number is essentially the

same whether a facility burns high or low sulfur coal.” Pet. Opening Br. at 28. A

smaller scrubber would be impossible to build, or would significantly diminish

the units’ electricity production, petitioners say.

The EPA says it made its calculations in response to the flawed

assumptions made by the petitioners. In the EPA’s view, the petitioners’ analysis

assumed that they would be building a much more powerful—and, more

important, costly—scrubber system than was actually needed. The petitioners

still assumed, though, that they would use low-sulfur coal in the future, making

the scrubbers seem less cost effective than they actually would be, the EPA said.

The EPA’s consultant, Dr. Phyllis Fox, analyzed two options to account for

these alleged flaws in OG&E’s analysis. In Option 1, Fox evaluated the cost

effectiveness of the high powered scrubbers using an ahistorical baseline.

Revised BART Cost-Effectiveness Analysis, JA at 1513. That is, the analysis

assumed that OG&E—while currently using low-sulfur coal—would begin using

cheaper, higher sulfur coal if it actually built these larger scrubbers. Id. at 1280.

This increased the scrubbers’ cost effectiveness because it anticipated the removal

of greater amounts of SO 2. Id. at 1513-14.

In Option 2, the EPA evaluated the cost effectiveness of a lower powered

scrubber using the historical baseline emissions. Id. at 1514. In order to design

this less powerful scrubber, the EPA used a model that OG&E’s consultants had

31
created for the EPA in a different action. 8 JA at 1283. This option increased the

cost effectiveness of the scrubbers—assuming the continued use of low-sulfur

coal—because it assumed the scrubbers would be less expensive.

The evaluation of the petitioners’ argument essentially hinges on the

technical feasibility of a smaller unit. We agree with the petitioners that the

guidelines typically require the use of the historical emissions baseline. And we

agree that if the EPA were to, without justification, increase the historical

emissions baseline, it would encourage the use of high-sulfur coal simply for the

purpose of removing greater amounts of SO 2.

But what is framed as an argument over the baseline is really an argument

over the size of the proposed scrubbers. The cost effectiveness of scrubbers

would routinely be understated if proposals included costing for scrubbers much

larger than needed. “While it may be prudent to overdesign for many reasons, the

cost of overdesign should not be attributed to BART, especially when the

emission reductions do not consider the overdesign,” the EPA said. Response to

Technical Comments. Id. at 1283. If OG&E wanted larger scrubbers, then the

EPA needed some way to isolate the BART-related costs from the non-BART-

related costs. The EPA took this into account by adjusting the baseline emissions

8
The EPA used this model in the final rule. The EPA’s consultant had
used a less precise model for basic engineering in the proposal, which the EPA
conceded was an “oversimplification.” Response to Technical Comments, JA at
1283.

32
under the assumption that, if OG&E built the scrubbers it proposed, it would

presumably shift to using cheaper forms of coal.

Left to evaluate the arguments of the parties’ experts, we must give

deference to the EPA. See San Juan Citizens Alliance v. Stiles, 654 F.3d 1038,

1045 (10th Cir. 2011) (“The deference we give agency action is especially strong

where the challenged decisions involve technical or scientific matters within the

agency’s area of expertise.” (quotation omitted)). 9 While the petitioners criticize

some of the engineering assumptions made by the EPA, they do not explain why

the EPA was not justified in relying on OG&E’s own consultant’s model, or why

the EPA’s detailed responses in its technical support document were insufficient

in addressing its concerns. For example, the petitioners state that “the smaller

scrubber envisioned by EPA [option 2] might work for some days of operation at

the OG&E Units, but it would preclude OG&E from producing electricity at

higher levels when needed.” Pet. Opening Br. at 29 n.18. But the EPA assumed

in its analysis that the plants would operate at a 100% capacity factor. Technical

Support Document, JA at 1348.

9
The petitioners argue that we should not afford the EPA deference
because the EPA’s consultant did not speak to OG&E directly and did not visit
the site. Pet. Br. at 19 n.9. They cite no authority for this proposition. Further,
the EPA said in its final rule that it “met with OG&E and its consultant
concerning the development of our proposal and had extensive communications
clarifying particular technical points,” which it conveyed to its consultant to
incorporate in her report. 76 Fed. Reg. at 81,728.

33
 To be sure, our dissenting colleague raises a number of valid concerns

about the EPA’s actions, and we acknowledge that this is a close case. But,

ultimately, we cannot adopt OG&E’s analysis given that the EPA was aware of,

and provided explanations contradicting, petitioners’ comments. 10 For instance,

the petitioners contend that the EPA relied too heavily on the sulfur content of the

fuel in downsizing the scrubbers. “A scrubber must be sized to reflect the

maximum potential heat input from the facility, and that number is essentially the

same whether a facility burns high or low sulfur coal,” the petitioners say. Pet.

Opening Br. at 28. The EPA “agree[d] that the sulfur content of the fuel, taken

by itself, will not significantly affect the size (or cost) of the gas path portions of

the FGD system.” Response to Technical Comments, JA at 1283. However, the

EPA noted that the “design and sizing of a scrubber is generally divided into two

major systems: (1) flue gas path and (2) reactant handling system.” Id. at 1284.

The EPA pointed out that OG&E’s own consultant had said in a previous case that

“gross unit size in MW . . . and sulfur content of the fuel are the major variables”

needed “in order to predict future retrofit costs.” Id. at 1285. Ultimately, the

10
The dissent argues that “[t]he EPA deserves no deference [on technical
determinations], however, where it does not support a conclusion contradicting
Oklahoma’s first, reasonable, detailed technical analysis.” Dissent at 3.
However, the dissent does not disagree with our conclusion that the EPA had
sufficient reasons for rejecting cost estimates—rife with errors—submitted by
OG&E. It is not clear how the dissent reached the conclusion that, despite these
errors, other portions of Oklahoma’s analysis were “reasonable.”

34
EPA concluded that “the use of the lower sulfur coal alone would reduce the

capital cost of the scrubber by about $7 million or 3%.” Id. at 1284. Given that

we must defer to the EPA’s technical judgments, we cannot say the EPA acted

arbitrarily on the basis of the record before us and the petitioners’ arguments in

their brief.

B. 2009 Cost Estimates

The petitioners also argue that the EPA improperly adjusted the 2009 Cost

Estimates in promulgating the FIP. Petitioners make three arguments. First, the

petitioners argue that the EPA made unreasonable adjustments to the site-specific

numbers put forth by petitioners. Second, the petitioners argue that the EPA

improperly relied on the “overnight” cost method instead of the constant-dollar

approach. Finally, the petitioners argue the EPA selectively manipulated many of

the other input variables in order to force OG&E to install scrubbers.

1) Double Counting and Discounts

The petitioners claim that in performing the analysis that formed the basis

of the FIP, the EPA consultant arbitrarily discounted some of the costs that the

petitioners had used in their analysis. Reviewing the record, we do not believe

these adjustments were arbitrary or capricious. The EPA explained the basis for

each of the adjustments it made. For example, the EPA discounted vendor quotes

to reflect the likely efficiencies from building multiple identical units. Revised

BART Cost-Effectiveness Analysis, JA at 1530-31. It applied a 5% discount after

35
a “search of the literature revealed a range of 4% to 10% savings from optimized

equipment.” Response to Technical Comments, JA at 1229.

Further, the EPA provided an extensive response to the petitioners’

comments on this point. The petitioners argued during the notice-and-comment

period that the EPA should not have assumed that the quotes did not already

contain a multiple-unit discount because the vendors knew they were constructing

multiple units. In response, the EPA explained why it believed these quotes did

not factor in efficiency discounts. Id. at 1228. OG&E had specifically asked for

single-unit estimates; the prices OG&E used in its report were “exactly double the

per-unit prices in the vendor quotes.” Id. at 1229. There was “no evidence that

any of these vendor quotes considered multiple unit discounts,” even though

“[b]ased on common industry practice, [the EPA] expected to see vendor

discounts for multiple units in OG&E’s site specific cost estimates.” Id. at 1228.

In addition, one of the vendors specifically “noted that the cost of the equipment

and the design could be optimized to provide more economical operation.” Id. at

1229.

Likewise, the EPA explained many of the other adjustments it made to the

costs submitted by OG&E. For instance, OG&E’s proposal estimated contingency

costs at 14% of the total project capital cost plus escalation, without providing

any details on how this number was calculated. Revised BART Cost-

Effectiveness Analysis, JA at 1520. The consultant flagged this number as high,

36
noting that the Control Cost Manual only permits the consideration of a limited

number of contingency costs. “A contingency factor in a cost effectiveness

analysis ‘should be reserved (and applied to) only those items that could incur a

reasonable but unanticipated increase but are not directly related to the

demolition, fabrication, and installation of the system,’” the consultant said. Id.

at 1521 (quoting Control Cost Manual, Chapter 2, Cost Estimation: Concepts and

Methodology, Sec. 2.5.4, p. 2-30). In the absence of any justification for

OG&E’s estimate, the consultant adjusted contingency costs to the standard 3% of

purchased equipment costs. Id.

Similarly, the consultant adequately explained the decision to apply a

discount to the “owner’s costs” estimated by OG&E. OG&E estimated “owner’s

cost” at 5% of total capital expended, much higher than the 2% its consultant had

used in other BART analyses. Id. at 1527. The EPA’s consultant acknowledged

that owners incur some costs related to construction. Id. But the consultant

explained that many of the costs that OG&E included in its estimate of “owner’s

costs”—such as site oversight—had already been included under the indirect

capital costs of engineering/procurement and construction. Id. In light of this,

the consultant removed owner’s costs from the analysis. Id. (“Owner’s costs are

not separately included in BART cost effectiveness analyses and have been more

than double counted here.”).

We do not believe these types of adjustments were arbitrary or capricious.

37
Moreover, even after the EPA made these adjustments, at least some of these site-

specific costs were higher than the generic numbers the EPA could otherwise

have used. See, e.g., id. at 1528 (“This change reduces the engineering and

procurement costs by $12,733,100 at Sooner and by $12,944,277 at Muskogee.

These values are still high compared to estimates based on the Cost Manual

method of 10% purchased equipment costs.” (footnotes omitted)). In fact, the

EPA consultant concluded that the EPA’s cost estimates were likely higher than

actual costs would be. Id. at 1510 (“Actual costs could be even lower as I was

unable to correct all of the overestimates that I identified due to lack of support

and underlying calculations.”).

2) Constant-Dollar Basis

Petitioners also argue that the EPA departed from the Control Cost Manual

in removing costs aside from inflation from its estimates. As above, we see no

merit to this argument. The petitioners have not made a persuasive case in their

briefs that either the EPA’s methodology conflicts with the manual 11 or that the

EPA conceded it should only have excluded inflation.

3) Selective Manipulation

Finally, the petitioners claim the EPA selectively manipulated its data to

11
We are not necessarily endorsing the EPA’s approach to costing
methodology. We note only that the petitioners chose on appeal to abandon the
arguments that they made in the administrative hearing that the manual did not
support the EPA’s methodology.

38
meet its desired result. The petitioners argue the EPA manipulated the data in

three ways. First, petitioners argue that the EPA arbitrarily accepted some

site-specific numbers and not others. Second, petitioners again reiterate

complaints about the EPA’s adjustments to its estimates. Third, petitioners argue

that the EPA assumed too long of a useful life for the scrubbers.

a) Site-specific costs

First, the petitioners argue that the EPA chose only to adopt site-specific

cost estimates that were higher than the generic estimates used in the manual. We

cannot agree. As the EPA explained, it used these site-specific numbers when

there was, in fact, accurate documentation—supporting a departure from the

manual. Response to Technical Comments, JA at 1273 (“We used the Control

Cost Methodology (overnight costs, no inflation, no AFUDC, no income taxes,

etc.) and site-specific values when they were valuable and correct.”). Petitioners

want us to view it as suspicious that the EPA accepted the site-specific estimates

where they were lower than the generic numbers in the guidelines. But one could

find it equally suspicious that the petitioners did not provide documentation for

its estimates that were much greater than the generic cost numbers in the manual.

Given that many of the site-specific numbers provided to the EPA were not

properly documented in accordance with the guidelines, we do not find it

arbitrary or capricious that the EPA rejected them.

39
 b) Double counting

The petitioners also argue that the EPA relied on flawed assumptions when

it reduced some of the cost estimates for double counting or overestimation.

Again, though, the consultant and the EPA thoroughly documented why they

made these adjustments. We see no basis for reversing the EPA’s decision on this

ground.

c) Useful life of scrubbers

Finally, the petitioners argue that the EPA assumed too long of a useful life

for the scrubbers. The longer the useful life of the scrubbers, the more cost

effective they will appear to be. The EPA assumed a useful life of thirty years in

its analysis. The petitioners argue that the EPA departed from the Control Cost

Manual’s standard useful life of twenty years for large pollution control systems.

But the EPA adequately explained why it chose to assume a useful life of

thirty years. As the petitioners acknowledge, the Control Cost Manual does not

reference any specific useful life for scrubbers. See Pet. Opening Br. at 34.

Therefore, the EPA considered, among other factors: 1) the fact that scrubbers

installed between 1975 and 1985 are still in use; 2) the standards from cost

estimates handbooks and published papers, and 3) the fact that the EPA has

assumed a 30-year lifetime for scrubbers since “at least 1981.” See JA at 1263.

40
The petitioners do not explain why this explanation was inadequate. 12

Instead, the petitioners argue that the EPA’s decision was unjustified

because the EPA assumed that the petitioners would use high-sulfur coal in

analyzing Option 1, but then, in its response to the comments, said that the

scrubbers would operate in the mild, low-sulfur environment in assuming a useful

life of thirty years. This argument is premised on an inaccurate characterization

of the EPA’s comments. In context, the EPA said that even scrubbers in

high-sulfur environments have useful lives of at least thirty years:

12
The petitioners characterize the EPA as departing from the manual’s
standard useful life of twenty years based solely on an unsupported determination
“that significant advances have been made in the material of construction and
baghouse design since this publication.” Pet. Opening Br. at 34. But the EPA
was merely explaining that its thirty-year useful life for dry scrubbers was
reasonable even in the context of the other control systems actually mentioned in
the manual:

OG&E also asserts that the Control Cost Manual assumes a 20
year useful life for other large air pollution systems, citing 20
years for a fabric filter baghouse and 20 years for SCR. The
cite of 20 years for the baghouse actually says: “For fabric
filters, the system lifetime varies from 5 to 40 years, with 20
years being typical,” citing to a 1980 report. Significant
advances have been made in the material of construction and
baghouse design since this publication. Further, the 1981
EPA/TVA report cited above, “Technical Review of Dry FGD
Sytems and Economic Evaluation of Spray Dryer FGD
System” assumes a 30 year economic and tax life for an
SDA/FF, which includes the fabric filter. Regardless, our 30
year estimate for a scrubber, which includes a baghouse, is
well within the reported range.

Response to Technical Comments, JA at 1264 (footnotes omitted).

41
 The subject application, the use of a scrubber to remove SO 2
from low sulfur coal, is a mild environment for a scrubber,
compared to high sulfur applications, which have already
demonstrated 30 year lifetimes. The corrosion potential and
bag plugging issues in a low sulfur application are much lower
than in a comparable high sulfur application.

Response to Technical Comments, JA at 1264 (emphasis added).

Finally, the petitioners argue that the EPA has assumed a shorter useful life

for scrubbers attached to other projects. However, the petitioners did not raise

this particular argument regarding the useful life of the scrubbers during the

administrative review period. Under the CAA, we may review “[o]nly an

objection to a rule or procedure which was raised with reasonable specificity

during the period for public comment.” 42 U.S.C. § 7607(d)(7)(B); see, e.g., Am.

Farm Bureau Fed’n v. EPA, 559 F.3d 512, 538 (D.C. Cir. 2009) (per curiam) (“In

the comments submitted in response to the EPA’s proposed revocation of the

annual standard, the environmental petitioners argued only that the record

evidence demonstrated adverse effects from long-term coarse PM exposure; they

did not raise their current argument that an annual standard is necessary to

prevent adverse effects from short-term exposure.”). The EPA raised this

jurisdictional bar in its brief, EPA Response Br. at 43 n.9; the petitioners offered

no response in their reply. Therefore, even if this argument had merit, we decline

to consider it in this petition for review.

42
 C. Scrubbers Would Not Have a Significant Impact on Visibility

Finally, the petitioners assert that the emissions limits proposed in the FIP

would not have a significant impact on visibility in the region. The petitioners

argue: 1) that the EPA should have used the dollar-per-deciview method in

evaluating the benefits to visibility in installing the scrubbers; and 2) that the

EPA impermissibly aggregated the visibility improvements the scrubbers would

create across facilities. Neither of these claims has merit.

Oklahoma first suggests EPA should not have rejected the visibility

analysis it conducted in the SIP, which used the dollar-per-deciview method.

This argument is misguided. The EPA rejected the SIP because of the flawed cost

estimates. When promulgating its own implementation plan, it did not need to

use the same metric as Oklahoma. The guidelines merely permit the BART-

determining authority to use dollar per deciview as an optional method of

evaluating cost effectiveness. See 40 C.F.R. pt. 51 app. Y(IV)(E)(1). 13

13
We note, however, that in both its final rule and in its brief the EPA
asserts that the guidelines require the use of the dollar-per-ton metric in
evaluating cost effectiveness. The guidelines themselves are a bit unclear. In the
section on cost effectiveness, the guidelines mention only the dollar-per-ton
metric. 40 C.F.R. pt. 51 app. Y(IV)(D)(4)(c). However, the guidelines later state
that in evaluating alternatives, “we recommend you develop a chart (or charts)
displaying for each of the alternatives” that includes, among other factors, the
cost of compliance defined as “compliance—total annualized costs ($), cost
effectiveness ($/ton), and incremental cost effectiveness ($/ton), and/or any other
cost-effectiveness measures (such as $/deciview).” Id. app. Y(IV)(E)(1)
(emphasis added).
(continued...)

43
 And in the final rule, the EPA explained why it did not use the dollar-per-

deciview metric used by Oklahoma. “Generally speaking, while the metric can be

useful if thoughtfully applied, we view the use of the $/deciview metric as

suggesting a level of precision in the calculation of visibility impacts that is not

justified in many cases.” 76 Fed. Reg. at 81,747. The EPA has never mandated

the use of this metric, and has not developed “thresholds of acceptable costs per

deciview improvement.” Id. While the federal land managers have developed

threshholds, these threshholds were apparently developed without input from the

EPA and without notice-and-comment review. EPA Br. at 54 n.13. In light of

this, we do not find it arbitrary or capricious that the EPA chose not to use the

dollar-per-deciview metric in evaluating BART options in creating the FIP. We

therefore also conclude that any argument by the petitioners that the dollar-per-

deciview measurement proves the scrubbers are not cost effective lacks merit.

See Pet. Reply Br. at 16.

The petitioners next argue that the EPA impermissibly aggregated the

visibility improvements across facilities to make the scrubbers look more

effective. The petitioners say that this is evidenced by the EPA’s reference to the

13
(...continued)
We do not rule here on whether the EPA would be justified in rejecting a
SIP because it relied on the dollar-per-deciview metric. We hold only that it was
reasonable for the EPA to use the dollar-per-ton metric despite Oklahoma’s
earlier analysis.

44
total improvement in visibility (2.89 deciviews) that would result from placing

scrubbers on all four of these units. Viewed separately, the improvements from

the addition of scrubbers at each unit would not be worth the cost of the

scrubbers, the petitioners say.

While we agree that the EPA referenced the aggregated visibility

improvement across the Class I areas, we do not agree that it used that number to

guide its analysis. Instead, it evaluated the improvements facility-by-facility.

The Response to the Technical Comments includes data on the visibility

improvement at a number of different areas, broken down by facility. See JA at

1495-98. Further, the EPA said its modeling “indicates that visibility

improvements anticipated from the installation of dry scrubbers at each facility

will result in reducing modeled impacts . . . from each facility at all nearby Class

I areas to levels below 0.5 dv, with improvements greater than 1.0 dv at some

Class I areas.” See 76 Fed. Reg. at 81,739 (emphasis added).

The petitioners also argue that the EPA erred because it considered the

visibility improvement facility-by-facility instead of unit-by-unit. Even if this

argument had merit, we have no jurisdiction to consider it. The petitioners never

raised this objection during the rulemaking process. See 42 U.S.C. §

7607(d)(7)(B) (“Only an objection to a rule or procedure which was raised with

reasonable specificity during the period for public comment (including any public

hearing) may be raised during judicial review.”). In fact, the petitioners actually

45
commented that the EPA needed to make the visibility determinations on a

facility-by-facility basis. See JA at 1108 (“Whether reviewing a SIP or proposing

its own FIP, EPA, like individual states, is required to consider the visibility

improvement associated with scrubbers on a facility-by-facility basis.”).

Likewise, the petitioners argue that the EPA did not provide sufficient

notice of the approach it used in its final rule. The petitioners say the EPA used a

new metric—days of visibility improvement—that it had never used before. Even

if this argument has merit, we cannot consider it on appeal. Again, we may only

address issues that were raised during the rulemaking process. As discussed

above, it does not matter that the petitioners could not have raised their objection

before the promulgation of the final rule. “Rather, the CAA requires a petitioner

to first raise its objection to the agency through a petition for reconsideration.”

Appalachian Power Co., 249 F.3d at 1066. We therefore decline to consider the

petitioners’ argument here. See 42 U.S.C. § 7607(d)(7)(B) (“If the Administrator

refuses to convene [a reconsideration proceeding], such person may seek review

of such refusal in the United States court of appeals for the appropriate

circuit . . . .” (emphasis added)).

V

In addition to these arguments concerning the EPA’s substantive analysis,

the petitioners raise a number of challenges to the procedures the EPA used in

promulgating the rule. First, they argue that the EPA may not promulgate a FIP

46
in the same action in which the agency disapproves a SIP. Second, the petitioners

argue that the EPA lost the authority to promulgate a FIP because the agency

failed to act within two years after its duty to promulgate a FIP was first

triggered.

The CAA creates a high bar for any petitioner challenging an EPA action

on procedural grounds. The petitioner must prove: 1) that the failure to observe

the procedure was “arbitrary and capricious”; 2) that the objection was “raised

with reasonable specificity during the period for public comment”; and 3) that the

errors were “so serious and related to matters of such central relevance to the rule

that there is a substantial likelihood that the rule would have been significantly

changed if such errors had not been made.” 42 U.S.C. § 7607(d)(9)(D), (d)(7),

(d)(8). “The essential message of so rigorous a standard is that Congress was

concerned that EPA’s rulemaking not be casually overturned for procedural

reasons, and we of course must respect that judgment.” Sierra Club v. Costle, 657

F.2d 298, 391 (D.C. Cir. 1981).

The petitioners first argue that the EPA violated required procedures by

promulgating its FIP in the same action in which it disapproved the SIP. The

petitioners argue the statute requires the EPA first take action on the SIP because

it says, according to petitioners, that the “EPA shall propose a FIP ‘unless the

State corrects the deficiency,’ thereby reflecting Congress’s intention for States to

have the power to design their own SIP and have an opportunity to correct a SIP

47
before a FIP is issued.” Pet. Opening Br. at 40 (quoting § 7410(c)). It also

makes a policy argument that permitting the EPA to disapprove SIPs in the same

action in which it promulgates a FIP will “blur [the] distinction” between the

EPA’s role in reviewing SIPs and promulgating FIPs. Id. at 41.

We do not agree that the EPA’s actions violated the procedural

requirements of the CAA. The petitioners’ parsing of the statute relies on a

truncated quotation. Under 42 U.S.C. § 7410(c)(1), the EPA must create a FIP

after either the state has failed to make the required SIP submission or the EPA

has disapproved part of the state’s SIP. This duty continues to exist “unless the

State corrects the deficiency, and the Administrator approves the plan or plan

revision, before the Administrator promulgates such Federal implementation

plan.” § 7410(c)(1) (emphasis added). Once the EPA issued findings that

Oklahoma failed to submit the required SIP under the Regional Haze Rule, the

EPA had an obligation to promulgate a FIP. The statute itself makes clear that

the mere filing of a SIP by Oklahoma does not relieve the EPA of its duty. And

the petitioners do not point to any language that requires the EPA to delay its

promulgation of a FIP until it rules on a proposed SIP. As the EPA points out,

such a rule would essentially nullify any time limits the EPA placed on states.

States could forestall the promulgation of a FIP by submitting one inadequate SIP

after another.

In any case, even if we agreed that the EPA should not have promulgated

48
the FIP in the same action as it rejected the SIP, it is not clear the petitioners

would meet the high bar for overturning an EPA action on procedural grounds. It

may be poor policy to try to distinguish between the SIP and FIP in a single

action. But the petitioners make no attempt to show the procedural error was “so

serious and related to matters of such central relevance to the rule that there is a

substantial likelihood that the rule would have been significantly changed if such

errors had not been made.” § 7607(d)(9)(D), (d)(8). Indeed, after the EPA raised

this heightened standard of review of procedural actions in its brief, the

petitioners were silent as to any issue regarding this procedural deficiency in their

reply brief.

The petitioners also assert the EPA violated the statute because

§ 7410(c)(1) says that the “Administrator shall promulgate a Federal

implementation plan at any time within 2 years after” the EPA’s duty to

promulgate a FIP is triggered. The petitioners argue the EPA lost authority to

promulgate its FIP because more than two years had passed since the EPA made

its initial finding that Oklahoma failed to submit a SIP. Although the statute

undoubtedly requires that the EPA promulgate a FIP within two years, it does not

stand to reason that it loses its ability to do so after this two-year period expires.

Rather, the appropriate remedy when the EPA violates the statute is an order

compelling agency action.

The Supreme Court rejected an argument similar to the petitioners’

49
argument in Brock v. Pierce County, 476 U.S. 253 (1986). At issue in Brock was

a law requiring that the Secretary of Labor “issue a final determination as to the

misuse of [Comprehensive Employment and Training Act] funds by a grant

recipient within 120 days after receiving a complaint alleging such misuse.”

Brock, Id. at 254-55 (quotation omitted). The petitioners in that case argued that

the Secretary lost his power to recover those funds if he did not make a final

determination within 120 days. Id. at 255.

In the absence of a more clear statutory directive, the Court refused to

accept this argument. Rather, when “there are less drastic remedies available for

failure to meet a statutory deadline”—such as a motion to compel agency

action—“courts should not assume that Congress intended the agency to lose its

power to act.” Id. at 260. The Court “would be most reluctant to conclude that

every failure of an agency to observe a procedural requirement voids subsequent

agency action, especially when important public rights are at stake.” Id.

The petitioners do not explain why the principles of Brock would not also

control this case. The provision here is “clearly intended to spur [the EPA] to

action, not to limit the scope of [its] authority.” Id. at 265. In the absence of any

other indication from Congress, the appropriate remedy is simply a suit to compel

agency action, not to eliminate the EPA’s authority to file a FIP. See Mont.

Sulphur & Chem. Co. v. EPA, 666 F.3d 1174, 1190-91 (9th Cir. 2012).

50
 VI

In conclusion, we hold that the EPA had the authority to review

Oklahoma’s BART determinations. Moreover, it exercised that authority

properly. Accordingly, we DENY the petition for review of the EPA’s final rule.

The stay pending hearing by the merits panel is hereby lifted.

51
Nos. 12-9526 and 12-9527, State of Oklahoma, et al. v. United States
Environmental Protection Agency

KELLY, Circuit Judge, concurring in part and dissenting in part.

Although I agree with much of the court’s analysis, I dissent with respect to

whether certain EPA actions were arbitrary and capricious. See Ct. Op. Pt. IV(A)

(analyzing the EPA’s calculations of baseline emissions and its determination

regarding the technical feasibility of the smaller scrubbers on which it based its

cost/benefit analysis). Therefore, I would grant the petition for review.

As an initial matter, the court states that “[w]e review these challenges [to

the EPA’s FIP] under the same arbitrary and capricious standard we used to

evaluate the EPA’s rejection of the SIP.” Ct. Op. Pt. IV. The court notes,

however, that “we do so while recognizing this requires a slightly different

perspective: evaluating the EPA’s own choices under the guidelines, as opposed

to evaluating its choice to reject the Oklahoma SIP under the guidelines.” Id.; see

also id. at Pt. III(B), n.7 (“We recognize that the EPA has less discretion when it

takes actions to reject a SIP than it does when it promulgates a FIP.”). That may

usually be the case, but here the EPA rejected Oklahoma’s SIP and promulgated

its own FIP in the same rulemaking action. Many of the same reasons for

rejecting the SIP were used to justify the FIP. Therefore, to the extent it makes a

difference, I am not convinced we owe any more deference to the EPA in

evaluating these challenges than we would if this action were solely a rejection of
a state SIP.

The EPA rejected Oklahoma’s cost estimates for scrubbers and provided

two options of its own. These options arbitrarily and capriciously (1) assumed

OG&E would burn coal they are not burning and have no plans to burn and (2)

used scrubbers that do not fit and are not technically feasible.

For purposes of the cost benefit analysis, the Control Cost Manual (CCM)

requires that anticipated annual emissions be calculated based upon past actual

emissions. 40 C.F.R. pt. 51, App. Y(IV)(D)(4)(d). The EPA ignored the

historical emissions baseline in Option 1. OG&E has been using low-sulfur coal

and its continued use would have resulted in scrubbers removing about 14,000

tons of SO 2 per year at each affected power plant. JA 1106. The EPA, however,

assumed that OG&E would begin using high-sulfur coal which would result in the

removal of between roughly 43,000 to 46,000 tons of SO 2 per year at each

affected power plant. Id. at 1106, 1513–14. Petitioners commented that any

assumption by the EPA of a change in coal was improper and unsupportable. See

id. at 1088, 1096–99 (OG&E EPA comments); see also id. at 300–02, 313 (OG&E

state comments). Knowing these calculations violated the manual, the EPA

developed Option 2.

Option 2 was based on OG&E’s continued use of low-sulfur coal—the

correct emissions baseline as required by the CCM. However, the EPA changed

the size of the scrubbers to smaller, less expensive ones. Id. at 1514–16.

-2-
Petitioners argued extensively that these smaller scrubbers were technically

infeasible. See, e.g., id. at 1099–1102. In particular, the comments point out the

size of a scrubber is not dependent on the type of coal used but on gas flow and

the maximum potential heat input. Id. at 1101–02, 1144 (OG&E EPA comments);

see also id. at 384–430 (OG&E state comments).

EPA admitted in response that the type of coal alone does not affect

scrubber size and its estimate for a smaller scrubber was a result of

oversimplification. See id. at 1283–84. Therefore, the only question is whether

the EPA provided support for the technical feasibility of the smaller scrubbers it

ultimately required. The EPA, however, relies almost exclusively on a cost model

by OG&E’s consultant Sargent & Lundy in a different action and its statements

that the amount of SO 2 removed can in fact affect scrubber size at least somewhat.

See id. at 1283, 1348. This report does not describe why it matters or how it

would affect the size of the scrubbers in this case.

The EPA rejected Oklahoma’s evidentiary support with no clear evidence

of its own to support its contrary conclusion. Usually the court grants deference

to the EPA’s technical determinations. See San Juan Citizens Alliance v. Stiles,

654 F.3d 1038, 1045 (10th Cir. 2011). The EPA deserves no such deference,

however, where it does not support a conclusion contradicting Oklahoma’s first,

reasonable, detailed technical conclusion. See JA 384–430; Lockheed Martin

Corp. v. Admin. Review Bd., U.S. Dep’t of Labor, No. 11-9524, 2013 WL

-3-
2398691, at *3 (10th Cir. June 4, 2013). Therefore I would conclude that the

failure of the agency to provide any evidence that a significantly smaller scrubber

was sufficient to meet OG&E’s needs is arbitrary and capricious.

Finally, it is worth noting that the EPA’s regional haze program is distinct

in the amount of power given to the states. See, e.g., Train v. Nat. Res. Def.

Council, Inc., 421 U.S. 60, 79 (1975); 70 Fed. Reg. 39,104, 39,137 (July 6, 2005)

(“[H]ow states make BART determinations or how they determine which sources

are subject to BART” are among the issues “where the Act and legislative history

indicate that Congress evinced a special concern with insuring that States would

be the decision makers.”). There are a number of reasons for this approach, not

the least of which is that its goals and standards are purely aesthetic rather than

directly related to health and safety. The EPA’s rule here requires OG&E to

make a $1.2 billion dollar investment over the next five years that will, even

under EPA’s estimate, result in no appreciable change in visibility. Moreover,

there is no evidence this investment will have any effect whatsoever on air

quality. It surely will, however, result in adverse changes to what Oklahoma

ratepayers will pay for electricity.

Although the EPA has at least some authority to review BART

determinations within a state’s SIP, it has no authority to condition approval of a

SIP based simply on a preference for a particular control measure. Texas v. EPA,

690 F.3d 670, 684 (5th Cir. 2012); see EME Homer City Generation, L.P. v. EPA,

-4-
696 F.3d 7, 29 (D.C. Cir. 2012) (reviewing a different rule and concluding that

the CAA “prohibits EPA from using the SIP process to force States to adopt

specific control measures”). Oklahoma considered the cost and resulting benefit

of such a large investment in scrubbers, and its conclusion was not unreasonable.

Assuming the EPA has authority to review Oklahoma’s SIP in the manner it

did, its rejection of the SIP and promulgation of its own FIP was arbitrary and

capricious. Accordingly, I respectfully dissent on this issue.

-5-