FILED
NOT FOR PUBLICATION
APR 13 2022
UNITED STATES COURT OF APPEALS MOLLY C. DWYER, CLERK
U.S. COURT OF APPEALS
FOR THE NINTH CIRCUIT
MEDICAL ADVOCATES FOR No. 20-72780
HEALTHY AIR; NATIONAL PARKS
CONSERVATION ASSOCIATION; EPA Nos.
ASSOCIATION OF IRRITATED EPA-R09-OAR-2019-0318
RESIDENTS; SIERRA CLUB, FRL-10011-44-Region 9
Petitioners,
MEMORANDUM*
v.
U.S. ENVIRONMENTAL PROTECTION
AGENCY; ANDREW WHEELER,
Administrator, U.S. Environmental
Protection Agency; JOHN BUSTERUD,
Regional Administrator, Region IX, U.S.
Environmental Protection Agency,
Respondents,
______________________________
SAN JOAQUIN VALLEY UNIFIED AIR
POLLUTION CONTROL DISTRICT,
Intervenor.
On Petition for Review of an Order of the
Environmental Protection Agency
*
This disposition is not appropriate for publication and is not precedent
except as provided by Ninth Circuit Rule 36-3.
Argued and Submitted February 11, 2022
San Francisco, California
Before: WARDLAW, IKUTA, and BADE, Circuit Judges.
Petitioners seek review of a final rule issued by the Environmental
Protection Agency (the EPA) under the Clean Air Act (the CAA), 85 Fed. Reg.
44,192-01 (July 22, 2020), approving an attainment plan (the “2018 Plan”)
submitted by California and adopted by both the San Joaquin Valley Unified Air
Pollution Control District (the “District”) and the California Air Resources Board
(CARB) to meet the 2006 24-hour PM2.5 national ambient air quality standards
(NAAQS) in the San Joaquin Valley.1 The final rule also approved California’s
request to extend the attainment date for meeting these standards from December
31, 2019 to December 31, 2024.
Petitioners argue that the EPA’s approval was arbitrary and capricious
because: (1) the 2018 Plan’s aggregate commitments did not satisfy applicable
requirements, and (2) the EPA’s analysis regarding implementation of the best
available control measures (“BACM”) and most stringent measures (“MSM”) was
1
PM2.5 means particles measuring less than 2.5 microns in diameter. PM2.5
may form due to the emissions of other chemicals such as nitrogen oxides (NOx).
See 85 Fed. Reg. 17,382 (Mar. 27, 2020).
2
arbitrary and capricious, such that the attainment deadline extension should not
have been granted.
We first consider petitioners’ challenges to the 2018 Plan’s aggregate
commitments, involving the use of incentive-based control measures to achieve
specified reductions of emissions according to a schedule. To the extent
petitioners argue that the EPA’s approval of aggregate commitments in the 2018
Plan is illegal because such commitments allow states to impermissibly defer plan
development, we reject this argument. We have previously concluded that the
EPA may reasonably rely on a control strategy that includes aggregate
commitments, and that such an approach does not violate the CAA. See Comm. for
a Better Arvin v. EPA, 786 F.3d 1169, 1179 (9th Cir. 2015); see also 42 U.S.C.
§ 7410(a)(2)(A) (providing that a state implementation plan may include
“schedules and timetables for compliance”). We also reject petitioners’ argument
that the aggregate commitments are invalid on the grounds that they are
unenforceable. Upon the EPA’s approval of the aggregate commitments as part of
the 2018 Plan, they become federally enforceable requirements of an applicable
implementation plan, which can be enforced by the public and the EPA. See 42
U.S.C. §§ 7602(q), 7604(a)(1), 7413(a)(1); see also Arvin, 786 F.3d at 1180.
Petitioners’ reliance on 42 U.S.C. §§ 7511a(e)(5) and 7410(k)(4) to support their
3
argument that aggregate commitments are precluded by the CAA is misplaced.
The CAA’s inclusion of § 7511a(e)(5), which allows the EPA to approve
provisions of an implementation plan in certain ozone nonattainment areas “which
anticipate development of new control techniques or improvement of existing
control technologies,” does not prevent the EPA from approving a plan such as the
one here, which neither relies on anticipated developments in technology nor
involves an ozone nonattainment area. Nor does § 7410(k)(4), which allows the
EPA to “approve a plan revision based on a commitment of the State to adopt
specific enforceable measures by a date certain,” prevent the EPA from approving
a state’s aggregate commitments as part of a larger plan. See 85 Fed. Reg. at
44,204.
Petitioners next challenge the EPA’s application of its traditional three-
factor test for evaluating the permissibility of aggregate commitments. Under this
established test, the EPA considers: (1) whether the commitment addresses a
limited portion of the statutorily required program; (2) whether the state is capable
of fulfilling its commitment; and (3) whether the commitment is for a reasonable
and appropriate period of time. See 75 Fed. Reg. 74,518-01, 74,535–36; see also
BCCA Appeal Grp. v. EPA, 355 F.3d 817, 840 (5th Cir. 2003). We conclude that
4
the EPA reasonably determined that the first and third factor were met, but erred in
concluding that the second factor was met.
First, the EPA reasonably determined that the commitment addressed a
limited portion of the statutorily-required program in light of “the facts and
circumstances of the nonattainment area at issue.” 85 Fed. Reg. at 44,198.
Petitioners argue that the EPA erred in allowing California to make aggregate
commitments to address 17 to 31 percent of the emissions reductions, because
historically the EPA had approved aggregate commitments that addressed
approximately 10 percent of the emissions to be reduced. We reject this argument
because petitioners fail to identify any statutory or regulatory authority precluding
the EPA from allowing aggregate commitments to address a portion of total
emission reductions larger than 10 percent, and because the petitioners fail to
otherwise establish that the EPA’s approval of higher percentages in this case was
so unreasonable as to be arbitrary and capricious. See Bahr v. EPA, 836 F.3d 1218,
1228 (9th Cir. 2016) (“[The arbitrary and capricious] standard is ‘highly
deferential, presuming the [EPA’s] action to be valid and affirming the agency
action if a reasonable basis exists for its decision.’” (citation omitted)).
Second, the EPA reasonably determined that the aggregate commitments
were for a reasonable and appropriate period of time, therefore satisfying the third
5
factor of the three-factor test. Although the record shows that the District intends
to implement certain measures after January 1, 2024, the date by which California
must achieve its commitments, the EPA points out that the 2018 Plan does not rely
on these measures to demonstrate attainment. Petitioners do not affirmatively
dispute this point. Thus, the EPA’s approval of the time period for the aggregate
commitments is not arbitrary or capricious.
Petitioners also argue that the aggregate commitments did not satisfy the
second factor of the three-factor test—whether the state is capable of fulfilling its
commitments—because California is unable to adequately fund the Plan’s
incentive-based control measures. We agree with this argument. Based on
CARB’s own calculations, there is a potential $2.6 billion shortfall in funding for
the incentive-based measures, which could result in emission reduction shortfalls
of approximately 7% of the total NOx reductions and 8% of the total PM2.5
reductions necessary for attainment. See 85 Fed. Reg. at 44,201. The EPA fails to
provide evidence or reasoned explanation for its conclusion that California will be
able to fulfill its commitment despite this shortfall. Instead, the EPA speculates
that circumstances may arise in the future that will allow the state to do so.
Specifically, the EPA asserts (1) that its own analysis indicates the funding
shortfall may be closer to $2 billion as opposed to $2.6 billion, see id., and the
6
actual funding shortfall may be even smaller, (2) that certain emission reduction
measures might be less expensive to implement than the state currently predicts,
(3) that certain yet-to-be-quantified sources of emission reductions in the 2018
Plan might make up for shortfalls in emission reductions caused by lack of
funding, and finally, (4) that California and the District may identify other
measures in the future to fulfill the commitments. Because these speculative
assertions are unsupported by evidence, they fail to ensure that California and the
District have a plausible strategy for achieving this portion of the attainment
strategy, and therefore do not collectively satisfy the second factor of the EPA’s
three-factor test. See also 42 U.S.C. § 7410(a)(2)(E) (requiring, inter alia, that
state implementation plans provide necessary assurances of adequate funding for
carrying out the plan). Therefore, we conclude that this portion of the EPA’s
analysis is arbitrary and capricious because there is no “rational connection
between the facts found” by the EPA regarding the funding shortfall and the EPA’s
conclusion that the District will nevertheless meet the attainment goals. Barnes v.
U.S. Dep’t of Transp., 655 F.3d 1124, 1132 (9th Cir. 2011).
Having addressed petitioners’ challenges relating to the EPA’s approval of
the aggregate commitments, we now turn to petitioners’ argument that the state
failed to meet BACM and MSM requirements, and that the EPA’s approval of the
7
request for a five-year extension was therefore arbitrary and capricious on that
ground. We conclude that the EPA undertook a rigorous analysis of compliance
with BACM and MSM requirements in granting the five-year extension. See 42
U.S.C. § 7513(e). We reject petitioners’ arguments to the contrary.
We first reject petitioners’ argument that the EPA cannot evaluate the
stringency of control measures in the aggregate, but must evaluate each control
measure individually to ensure each is the most stringent available. That is, we
defer to the EPA’s reasonable conclusion that it can apply a holistic MSM analysis
when the rule as a whole is as or more stringent than any analogous rule in other
jurisdictions. Petitioners have not pointed to any provision precluding such an
approach or requiring the EPA to conduct an individualized MSM analysis for each
control measure. For similar reasons, we also reject petitioners’ argument that the
EPA erred in using different levels of generality to evaluate the stringency of
control measures, because the petitioners cite no legal authority preventing the
EPA from doing so. The EPA’s thorough analysis provided a sufficient
explanation to support its conclusion.
We reject petitioners’ argument that Rule 4901 of the 2018 Plan is not
sufficiently stringent because the rule involves implementing relevant measures in
some areas of the San Joaquin Valley and not others. The EPA reasonably
8
determined that this type of targeted approach is acceptable provided the rule as a
whole is as or more stringent than any analogous rule in other jurisdictions. As the
EPA determined, and as petitioners do not affirmatively dispute, Rule 4901 results
“in greater emission reductions than the provisions of any other analogous rule.”
Therefore, petitioners fail to demonstrate the EPA’s conclusion was arbitrary and
capricious.
Petitioners’ argument that the EPA failed to adequately consider increased
building electrification requirements also fails, because the EPA considered such
an approach and reasonably accepted the state’s determination that it was not
feasible at this time.
Finally, we reject petitioners’ argument that the EPA was unreasonable in
approving the District’s Rule 4550 “menu” approach to controlling dust emissions
from agricultural sources—i.e., allowing the regulated community to choose
among various control measures—because such an approach is not sufficiently
stringent. We have previously upheld the approval of such an approach as a
“common and accepted practice for the control of dust . . . [especially] because of
the variable nature of farming.” See Vigil v. Leavitt, 381 F.3d 826, 837 (9th Cir.
2004) (citation omitted).
9
Accordingly, we vacate the portion of the final rule that determined that the
relevant aggregate commitments satisfied the second factor of the three-factor test
and remand to the EPA for further consideration of the second factor, and for
further proceedings consistent with this decision. See also 42 U.S.C.
§ 7410(a)(2)(E). In all other respects, the petition is denied. See Vigil, 381 F.3d at
847.
PETITION GRANTED IN PART AND DENIED IN PART.2
2
Each party will bear its own costs.
10