# District of Columbia v. U.S. Department of Agriculture

> District Court, District of Columbia · March 13, 2020

URL: https://www.frixlaw.com/law-library/cases/4516443

## Case

- **Court:** District Court, District of Columbia
- **Decided:** March 13, 2020
- **Precedential status:** Published
- **Opinion:** Opinion
- **Judges:** Chief Judge Beryl A. Howell
- **Nature of suit:** Civil
- **Cited by:** 0 later opinions in the Frix Law Library

## Citator (automated)

- No negative treatment found by the automated citator. That is not the same as a confirmation that the case is good law; read the citing cases.
- Full citator and citing cases: https://www.frixlaw.com/law-library/cases/4516443

## How later opinions describe it (automated extraction)

- holding that no “vested right” existed for purposes of retroactivity analysis where the prior “system offered no vested right to any specific terms” of a license and the agency “always retained the power to alter the term of existing licenses by rulemaking”
- stating that an agency’s “reasoning cannot be internally inconsistent” (citing Sierra Club v. EPA, 884 F.3d 1185, 1194–96 (D.C. Cir. 2018)
- explaining, based on Clapper, that a too- speculative future injury is one that depends on a “series of contingent events, none of which was alleged to have occurred by the time of the lawsuit”

## Opinion text

UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF COLUMBIA

DISTRICT OF COLUMBIA, et al.,

Plaintiffs, Civil Action No. 20-119 (BAH)

v. Chief Judge Beryl A. Howell

U.S. DEPARTMENT OF AGRICULTURE,
et al.,

Defendants.

BREAD FOR THE CITY, et al.,

Plaintiffs,

v.

U.S. DEPARTMENT OF AGRICULTURE,
et al.,

Defendants.

MEMORANDUM OPINION

In this country of plenty, the federal and state governments work together to ensure that

low-income Americans and their families do not go hungry. The largest federal food assistance

program that serves as the cornerstone of this joint federal-state effort to reduce hunger — and

hunger’s adverse effects on health, educational achievement, and housing security — is the

Supplemental Nutrition Assistance Program (SNAP), formerly known as the food stamp

program. A new federal rule poised to go into effect in a few weeks, in April 2020, would

dramatically alter the long-standing operations of the SNAP program, placing more stringent

requirements on states’ award of SNAP benefits with concomitant, virtually immediate effects

on the lives, by the federal government’s estimate, of over one million individuals currently

1
receiving SNAP benefits. Of those million, nearly 700,000 would lose their benefits. Especially

now, as a global pandemic poses widespread health risks, guaranteeing that government officials

at both the federal and state levels have flexibility to address the nutritional needs of residents

and ensure their well-being through programs like SNAP, is essential.

Nineteen states, the District of Columbia, the City of New York, and three private

plaintiffs have moved to enjoin preliminarily and to stay this new federal rule, issued by the

United States Department of Agriculture (USDA), that would limit state-implemented waivers of

the work requirements on which receipt of food assistance from SNAP may be conditioned. See

Final Rule, Supplemental Nutrition Assistance Program: Requirements for Able-Bodied Adults

Without Dependents, 84 Fed. Reg. 66782 (Dec. 5, 2019); see also State Pls.’ Mot. for Prelim.

Inj. or 5 U.S.C. § 705 Stay Pending Judicial Review (“State Pls.’ Mot.”), ECF No. 3; Pls.’ Mot.

for Prelim. Inj. (“Private Pls.’ Mot.”), Bread For the City, et al. v. U.S. Dep’t of Agric., et al., 20-

cv-127, ECF No. 4.

The low-income Americans targeted by USDA’s Final Rule depend on monthly SNAP

benefits to avoid hunger. These SNAP participants may wield little political or economic power,

but, nonetheless, USDA’s proposed changes to take away nutrition benefits from almost 700,000

people prompted “more than 100,000 comments,” the “majority” of which the agency concedes

were opposed to the proposed changes. 84 Fed. Reg. at 66783–84. Notwithstanding these

critical comments, USDA proceeded in the challenged Final Rule to adopt changes that, in some

respects, were more draconian than those initially proposed. Although the hundreds of

thousands of low-income individuals who stand to lose their benefits had little direct voice in

that rulemaking process, the process exists to protect them and ensure that the agency cannot

terminate their benefits arbitrarily. Under the Administrative Procedure Act (APA), agency

2
rules, like USDA’s, are unlawful unless the agency has considered the relevant evidence, has

weighed the consequences of its actions, and has rationally justified its choices. USDA says it

did all that here, but USDA is not the arbiter of the Final Rule’s legality. The courts are, and this

Court has determined that aspects of the Final Rule are likely unlawful because they are arbitrary

and capricious. USDA will be enjoined from implementing those aspects of the Final Rule

nationwide pending final judicial review.

The Final Rule relates to provisions of the Personal Responsibility and Work Opportunity

Reconciliation Act of 1996 (PRWORA) that conditioned the eligibility for SNAP benefits of

able-bodied adults without disabilities, or ABAWDs, on meeting work-related requirements. See

Pub. L. No. 104-193 § 824, 110 Stat. 2105, 2323 (1996) (codified at 7 U.S.C. § 2015).

Recognizing that the imposition of inflexible work requirements would undermine the SNAP

program’s effectiveness in alleviating hunger, Congress created two relevant exceptions. First,

PRWORA provided that, “[o]n the request of a State,” USDA “may waive the work

requirements” in “area[s]” that “do not have a sufficient number of jobs” for ABAWDs. 7

U.S.C. § 2015(o)(4)(A). Portions of the challenged Final Rule set to become effective on April

1, 2020, redefine waiver “area[s]” and limit the ways that states can show lack of sufficient jobs.

See 84 Fed. Reg. at 66802; see also id. (printing parts of the regulation to be codified, including

the challenged portions: (f)(2) and (f)(4)). Second, the Balanced Budget Act of 1997 (BBA)

allowed states to exempt from the work requirements up to 15% of all “covered individuals in

the State.” 7 U.S.C. § 2015(o)(6)(C)–(D); see also Balanced Budget Act of 1997, Pub. L. No.

105-33 § 1001, 111 Stat. 251, 252 (1997). Portions of the Final Rule set to become effective on

October 1, 2020, limit states’ ability to carry unused discretionary exemptions to later years. See

84 Fed. Reg. at 66802. For the reasons stated below, plaintiffs’ motions are DENIED as to the

3
discretionary exemption portions of the Final Rule. Plaintiffs’ motions are GRANTED as to the

waiver portions of the Final Rule.

I. BACKGROUND

Review of the procedural background follows discussion of the statutory framework, the

regulatory framework, and the challenged Rule.

A. Statutory Framework

Congress created SNAP in 1964 “to alleviate . . . hunger and malnutrition” by providing

“supplemental nutrition assistance” to “low-income households.” 7 U.S.C. § 2011; see also

Food Stamp Act of 1964, Pub. L. No. 88-525, 78 Stat. 703 (1964). SNAP offers non-cash

benefits that can be used to buy eligible food at approved retail stores. See 7 U.S.C. § 2011; id.

at § 2013(a). The program served an average of 42.1 million recipients per month in fiscal year

2017. See 83 Fed. Reg. 8013, 8013 (Feb. 23, 2018). Average monthly benefits are about $123,

while average monthly benefits for ABAWDs are around $160. See USDA, Characteristics of

Able-Bodied Adults Without Dependents (2018), https://www.fns.usda.gov/snap/characteristics-

able-bodied-adults-without-dependents; see also Edward Bolen Decl. (“Bolen Decl.”), Att. 2 at

228, ECF No. 3-2 (quoting the figure for ABAWDs).

USDA “is authorized to formulate and administer” SNAP, 7 U.S.C. § 2013(a), and has

delegated those responsibilities to the Food and Nutrition Service (FNS), a part of USDA, 7

C.F.R. § 271.3(a). Congress also explicitly granted USDA authority to “issue” regulations

“necessary or appropriate” to implement SNAP and prescribed that USDA “shall promulgate all

such regulations in accordance with” the APA’s notice and comment rulemaking procedures. 7

U.S.C. § 2013(c).

States, including the District, also play a significant role in administering SNAP. See 7

U.S.C. § 2012(r) (including the District in the relevant definition of “State”). By statute, state

4
agencies “have responsibility for certifying applicant households” as eligible for SNAP and for

“issuing” benefits. Id. § 2020(a)(1); see also Garnett v. Zeilinger, 313 F. Supp. 3d 147, 151–52

(D.D.C. 2018) (describing state responsibilities). The federal government funds at least half of

each state’s costs to “operat[e]” SNAP, 7 U.S.C. § 2025(a), and federal funds cover the full cost

of SNAP benefits paid, see id. § 2019.

1. Work Requirement Waivers

PRWORA, which amended the Food Stamp Act of 1977, conditioned ABAWD’s

eligibility for SNAP benefits on meeting work-related requirements. ABAWDS are specifically

defined as individuals between 18 and 49 years of age who are not “medically certified as

physically or mentally unfit for employment,” are not “a parent or other member of a household

with responsibility for a dependent child,” are not pregnant, and are not “otherwise exempt under

subsection (d)(2).” 7 U.S.C. § 2015(o)(3); see also id. § 2015(d)(2) (providing a variety of

exemptions, including for students and for those participating in drug treatment programs).

Under 7 U.S.C. § 2015(o), ABAWDs can only receive SNAP benefits for three months in a “36-

month period” unless they are working or participating in job training for “20 hours or more per

week, averaged monthly.” Id. § 2015(o)(2).1

Congress recognized that blunt application of those work requirements in areas where

few jobs are available would advance neither PRWORA’s goal of engaging ABAWDs in the

workforce nor the SNAP program’s goal of ensuring food security. See, e.g., 142 Cong. Rec.

H7905 (1996) (statement of Rep. Kasich) (stating that the program aims to encourage

employment in areas where “there are jobs available”). Congress therefore created a process for

1
Job-substitute and training programs include “a work program . . . as determined by the State agency” and
“a program under section 2029 of this title or a comparable program established by a State or political subdivision of
a State.” 7 U.S.C. § 2015(o)(2).

5
waiving the work requirements in certain areas. “On the request of a State agency,” USDA “may

waive” the work requirements “to any group of individuals in the State if” USDA “makes a

determination that the area in which the individuals reside” either “(i) has an unemployment rate

of over 10 percent; or (ii) does not have a sufficient number of jobs to provide employment for

the individuals.” Id. § 2015(o)(4)(A). “[T]he individuals” referenced are ABAWDs. See State

Pls.’ Mem. Supp. Mot. for Prelim. Inj. (“State Pls.’ Mem.”) at 21, ECF No. 3; Private Pls.’ Mem.

Supp. Mot. for Prelim. Inj. (“Private Pls.’ Mem.”) at 12, Bread for the City, et al., 20-cv-127,

ECF No. 4; Defs.’ Consolidated Opp’n to Pls.’ Mots. for Prelim. Inj. (“USDA’s Opp’n”) at 24,

ECF No. 26 (describing the statute as providing “whether an area has ‘a sufficient number of

jobs to provide employment for’ ABAWDs” (quoting 7 U.S.C. § 2015(o)(4)(A)(ii)).

As discussed in more detail below, through guidance and regulation, USDA has

implemented the statutory phrases “area in which the individuals reside” and “sufficient number

of jobs to provide employment for the individuals.” The challenged Final Rule radically rewrites

the policies that have been in effect for 25 years.

2. Discretionary Exemptions

As stated, in the BBA, Congress allowed states to exempt from a month’s work

requirements up to 15% of all “covered individuals in the State.” 7 U.S.C. § 2015(o)(6)(C)–(D).

“Covered individuals” are ABAWDs subject to the work requirements. Id. § 2015(o)(6)(A)(ii).

Thus, ABAWDs residing in waived areas do not count as “covered individuals,” and states with

statewide waivers in a given year do not earn any exemptions in that year. For ABAWDs, being

granted an exemption in a month means that the month’s SNAP benefits will not count towards

the three-month limit. Congress reduced these discretionary state exemptions to 12% of

“covered individuals” starting in fiscal year 2020. Id. § 2015(o)(6)(E).

6
By statute, USDA calculates each state’s available exemptions annually based on an

“estimate[]” of the total number of “covered individuals” in a given state. Id. § 2015(o)(6)(C)–

(E). Under 7 U.S.C. § 2015(o)(6)(G), USDA “shall increase or decrease the number of” each

state’s discretionary exemptions “to the extent that the average monthly number of exemptions in

effect in the State for the preceding fiscal year” was “lesser or greater than the average monthly

number of exemptions estimated for” that state for the “preceding fiscal year.” In other words,

USDA is required to increase a state’s allotted discretionary exemptions for a given fiscal year

“to the extent” that the state used fewer exemptions in the prior fiscal year than it had earned.

USDA is required to decrease a state’s allotted discretionary exemptions for a given fiscal year

“to the extent” that the state used more exemptions in the prior fiscal year than it had earned.

The challenged Rule changes how USDA implements the carrying over of exemptions.

B. Regulatory Framework

The full description of that challenged Rule is preceded by a description of the 2001

regulation, which is currently in effect.

1. The 2001 Regulation

Prior to 2001, in 1996, USDA had published guidance similar in substance to the 2001

regulation. Compare Food Stamp Program: Personal Responsibility Provisions of the Personal

Responsibility and Work Opportunity Reconciliation Act of 1996 (“2001 Regulation”), 66 Fed.

Reg. 4438 (codified at 7 C.F.R. § 273.24), with State Pls.’ Mot., Kathleen Konopka Decl.

(“Konopka Decl.”), Ex. C, Att.2, Guidance for States Seeking Waivers for Food Stamp Limits

(Dec. 3, 1996), ECF No. 3-12.

The 2001 regulation generally permits a state to “submit whatever data it deems

appropriate to support its” waiver requests, but requires data reliant on “standard Bureau of

Labor Statistics (BLS) data or methods” for “waiver requests based on unemployment rates or

7
labor force data.” 7 C.F.R. § 273.24(f)(2). The regulation also provides “[a] non-exhaustive list

of the kinds of data a State agency may submit” to support a “claim of lack of sufficient jobs.”

Id. The six criteria on this “non-exhaustive list” include data showing that an area: (1) “is

designated as a Labor Surplus Area (LSA) by the” Department of Labor (DOL); (2) qualifies, as

determined by DOL, “for extended unemployment benefits;” (3) “is described in an academic

study or other publications as an area where there are lack of jobs;” (4) “has a 24-month average

unemployment rate 20 percent above the national average for the same 24-month period.” Id.

§ 273.24(f)(2)(ii). The regulation further permits states to “define areas to be covered by

waivers.” Id. § 273.24(f)(6).

The 2001 regulation implements the statutory provisions about discretionary exemptions

by explaining simply that, if a “State agency does not use all of its exemptions by the end of the

fiscal year, FNS will increase the estimated number of exemptions allocated to the State agency

for the subsequent fiscal year by the remaining balance.” Id. § 273.24(h)(2)(i). Conversely, “[i]f

the State agency exceeds its exemptions by the end of the fiscal year, FNS will reduce the

estimated number of exemptions allocated to the State agency for the subsequent fiscal year by

the corresponding number.” Id. § 273.24(h)(2)(ii).

2. The Challenged Rule

USDA issued an Advanced Notice of Proposed Rulemaking (ANPRM) on February 23,

2018 to solicit comments from the public about potential changes to the 2001 regulation’s

treatment of state waivers. 83 Fed. Reg. 8013, 8013 (Feb. 23, 2018). “The ANPRM generated

nearly 39,000 comments . . . .” Proposed Rule, 84 Fed. Reg. 980, 982 (Feb. 1, 2019).

On February 1, 2019, USDA published a Proposed Rule that, inter alia, altered the 2001

regulation’s criteria governing approval of state waivers, redefined “area” for purposes of state

waiver applications, and revised how USDA calculated carried over discretionary exemptions.
8
Id. at 983. The Proposed Rule cited as one “rationale for modifying waiver standards,” id. at 980

(capitalization altered), the President’s April 2018 “Executive Order on Reducing Poverty in

America by Promoting Opportunity and Economic Mobility,” which “directed Federal agencies

to review regulations and guidance documents to determine whether such documents are

consistent with the principles of increasing self-sufficiency, well-being, and economic mobility,”

id. at 980–981. USDA voiced “confide[nce] that” the proposed “changes would encourage more

ABAWDs to engage in work or work activities if they wish to continue to receive SNAP

benefits.” Id. at 981.

USDA “received more than 100,000 comments” on the Proposed Rule. 84 Fed. Reg. at

66783. Like the Proposed Rule, the Final Rule tightened the criteria for demonstrating lack of

“sufficient jobs,” reworked the definition of waiver “area,” and changed how discretionary

exemptions would be carried over. See id. at 66811–12. 2 The waiver-related changes are

effective April 1, 2020, while the discretionary exemption change is effective October 1, 2020.

Id. at 66782. What follows is an overview of the Final Rule and of USDA’s reasoning in

adopting it.

a. Waivers for Lack of Sufficient Jobs

The Final Rule provides that USDA will approve states’ waivers for areas satisfying one

of two “core standards:” (1) “a recent 12-month average unemployment rate over 10 percent;” or

(2) a recent 24-month “average unemployment rate 20 percent or more above the national rate”

and above 6%. Id. at 66811. The first standard corresponds to the statutory provision allowing a

waiver if an area “has an unemployment rate of over 10 percent.” 7 U.S.C. § 2015(o)(4)(A)(i).

2
The differences between the Proposed Rule and the Final Rule are not comprehensively detailed because
this decision does not resolve the plaintiffs’ procedural challenges that depend on those differences. See infra, note
12 (describing those procedural challenges).

9
The second standard is intended to correspond to the statutory provision allowing a waiver if an

area “does not have a sufficient number of jobs to provide employment for the individuals.” Id.

§ 2015(o)(4)(A)(ii); see 84 Fed. Reg. at 66791 (indicating this). In short, under the Final Rule, to

qualify an area for a waiver based on insufficient jobs, a state will generally need to show that

the area’s 24-month average unemployment rate is both 20% above the national average and

above 6%. The Final Rule thus adds, for the first time in this regulatory scheme, an

unemployment rate floor to the waiver criteria. Finally, as discussed further later, USDA will

also consider waiver requests based on “other data and evidence” in an “exceptional

circumstance.” 84 Fed. Reg. at 66811.

As already noted, the 2001 regulation’s non-exhaustive list of criteria for establishing

lack of sufficient jobs comprises six examples of data sources and types that may be used. 7

C.F.R. § 273.24(f)(2)(ii). The Final Rule eliminates five of those criteria and adds the “6 percent

unemployment rate floor” to the 2001 regulation’s criterion requiring a 24-month average

unemployment rate. 84 Fed. Reg. at 66785. “Not including a floor,” the Final Rule stated, “has

had the effect of allowing areas with low rates of unemployment to qualify for waivers.” Id. at

66784.

The Final Rule addressed USDA’s decision to eliminate the following five criteria listed

in the 2001 regulation: qualification for federal extended unemployment benefits; designation as

an LSA by DOL; a declining employment-to-population ratio; jobs in declining industries or

occupations; and description in an academic study or publication as an area lacking jobs. 7

C.F.R. § 273.24(f)(2)(ii). As to qualification for federal extended unemployment benefits, which

DOL evaluates statewide, USDA expressed “concern[]” that retaining this “criterion would allow

States to receive statewide waivers even when there is not a lack of sufficient jobs within certain

10
areas of the State.” Id. at 66789–90. The Final Rule thus tied elimination of this criterion to the

redefinition of “waiver area,” explaining that USDA “is choosing to provide a strict definition of

a waiver area that will also restrict statewide waivers.” Id. at 66790. As to LSA designation,

USDA similarly reasoned that “including LSA designation as a waiver criterion would be

inconsistent with the final rule’s definition of an area.” Id. at 66800.

The other three criteria listed in the 2001 regulation — a declining employment-to-

population ratio, jobs in declining industries or occupations, and description in an academic

study or publication as an area lacking jobs — were rejected as a group for inclusion in the Final

Rule. Id. at 66790.3 “Many commenters,” USDA observed, “opposed excluding these criteria.”

Id. Most importantly, some commenters made the “broad[] argu[ment] that” elimination of these

criteria “results in an overreliance on unemployment rates” in measuring lack of sufficient jobs.

Id. at 66791. USDA acknowledged that non-unemployment rate–based measures “can enhance

the understanding of the job market,” but, citing its “operational experience,” said that “the

Department has recognized that” these measures “are less reliable and consistent than standard

unemployment data in demonstrating a lack of sufficient jobs.” Id. USDA also referred readers

to the discussion of unemployment rates in the section of the Rule justifying the 6%

unemployment rate floor. Id.

That discussion’s summary of comments concerned with overreliance on unemployment

rates was extensive, with “the vast majority of those who commented on the unemployment rate

floor opposed setting any unemployment rate floor.” See id. at 66786–88. These commenters

gave statutory and policy reasons for their position. See id. at 66786–87. First, “[c]ommenters

3
The Final Rule continued to include these criteria as appropriate supporting evidence for waiver
applications for areas, such as Indian reservations and U.S. Territories, for which BLS produces limited data. See 84
Fed. Reg. at 66791; see also id. at 66811.

11
argued that the statutory language,” id. at 66786 — which allows a waiver if an area “does not

have a sufficient number of jobs to provide employment for the individuals,” 7 U.S.C.

§ 2015(o)(4)(A)(ii) (emphasis added) — “requires the Department to base the waiver standards

on whether there are lack of sufficient jobs for the specific ABAWD population, not the broader

population,” 84 Fed. Reg. at 66786. USDA responded to this line of argument with its view that

“limiting the number of ways that a State may demonstrate a lack of jobs in order to prevent the

misapplication of waivers in areas in which the lack of jobs is questionable” is “well within the

authority” granted at 7 U.S.C. § 2015(o)(4)(A), which USDA interpreted “to provide[] the

Secretary with broad discretion on how to define what does and does not constitute a lack of

sufficient jobs,” 84 Fed. Reg. at 66788.

Second, commenters offered “analysis” and “research,” id. at 66787, supporting their

understanding that the standard unemployment rate “does not accurately reflect labor market

prospects for ABAWDs[] and . . . does not fully account for the ability of ABAWDs to find and

keep jobs due to lack of skills, training, or other barriers,” id. at 66786. This research included

studies “indicating” that ABAWDs tended to have lower levels of education, higher levels of

disability, higher levels of homelessness, were more likely to have been incarcerated, were more

likely to experience racial discrimination, and were more likely to be underemployed than the

general population and studies “suggesting” that these characteristics correlated with

unemployment rates higher than the unemployment rate of the general population. Id. at 66787.

“Commenters also asked the Department to consider access to transportation, housing stability,

and forced moves among the ABAWD population that lead to particular problems maintaining

stable employment.” Id. Finally, commenters observed that USDA “has previously

acknowledged,” in the 1996 guidance and the 2001 rulemaking, “that time limit waivers were

12
intended by Congress to recognize the challenges that the ABAWD population faces when

finding permanent employment.” Id. at 66787.

“In response to these comments,” USDA “recognize[d] that ABAWDs may face barriers

to employment and have more limited employment prospects than the general public.” Id.

“[N]otwithstanding” this recognition, “the Department [wa]s resolute” that the Rule’s changes,

and specifically the 6% unemployment rate floor, were “necessary to ensure that the standard is

designed to accurately reflect a lack of sufficient jobs in a given area.” Id. “The Department’s

position,” USDA stated, “is based on its operational experience, during which it has recognized

that, without” such changes “areas that do not clearly lack sufficient jobs will continue to qualify

for waivers.” Id.

b. Redefinition of “area”

The 2001 regulation, which permits states to “define areas to be covered by waivers,” 7

C.F.R. § 273.24(f)(6), allows waivers for single counties or towns or for groups of jurisdictions,

see 84 Fed. Reg. at 66794. The Final Rule would eliminate states’ longstanding discretion to

tailor waiver requests. Specifically, the Final Rule “prohibit[s] states from combining

unemployment data from individual substate areas to calculate an unemployment rate for the

combined area . . ., unless the combined area is designated as a Labor Market Area (LMA)” by

DOL. Id. at 66793.4 According to the Final Rule, DOL describes an LMA as “an economically

integrated area within which individuals can reside and find employment within a reasonable

distance or can readily change jobs without changing their place of residence.” Id.

In short, the Final Rule redefines “area in which the individuals reside,” 7 U.S.C.

§ 2015(o)(4)(A), as the LMA in which the individuals reside, see 84 Fed. Reg. at 66811. “[T]his

4
The Rule also “consider[s]” any “reservation area” and “U.S. Territory” “to be an area for the purposes of
waivers.” See 84 Fed. Reg. at 66811.

13
means that the final rule will only allow for waivers covering LMAs.” Id. at 66796. “Many

commenters opposed the proposed restriction on grouping only to LMAs.” Id. at 66793. These

commenters argued that states are best positioned to “understand[] . . . regional patterns in their

labor markets, local commuting burdens, and other local nuances specific to ABAWDs,” id., and

that “years of FNS guidance” and “USDA’s longstanding position” granting states “flexibility to

group substate areas” had embraced that assessment of state expertise, id. at 66794. “[T]he

Department disagree[d],” concluding that arguments supporting state “flexibility” were

“outweigh[ed]” by “the need to address” a “problem” identified by USDA. “The Department

has learned, through its extensive operational experience,” the Final Rule stated, “that this

flexibility allows States to strategically group substate areas to maximize the geographic

coverage of waived areas rather than to demonstrate high unemployment or a lack of sufficient

jobs for ABAWDs . . . .” Id.

Commenters further objected, citing examples, that the area definition adopted was

problematic because LMAs “do not account for specific ABAWD commuting patterns and other

factors specific to ABAWDs.” Id. at 66793. USDA was “not compelled by” these comments or

their suggested alternative groupings, however, concluding that “LMAs remain the best available

and most appropriate delineation” because “there are no Federally-designated areas that

specifically assess commuting patterns and other related economic factors for ABAWDs.” Id.

The Final Rule further explained that USDA had redefined waiver area as LMA because USDA

“expects unemployed ABAWDs to proactively pursue any and all work and/or work training

opportunities within reasonable commuting distance of their homes.” Id. at 66796. “[G]enerally

restricting waivers to qualifying LMAs,” the Final Rule concluded, “will result in a broader

14
application of the time limit, encourage geographic mobility among ABAWDs, and reduce

dependence on government benefits.” Id. at 66796.

c. Carryover of Discretionary Exemptions

Under the 2001 regulation, when a state did not use all of its exemptions in a given fiscal

year, USDA increased the state’s discretionary exemptions for the following year by the unused

balance. States were permitted to carryover indefinitely any unused exemptions. “As a result,”

the Final Rule stated, “States have accumulated extremely high amounts of unused discretionary

exemptions that well exceed the number allotted to each State for the fiscal year.” Id. at 66803.

In fiscal year (FY) 2019, “States earned approximately 1.3 million exemptions, but had about 7.4

million exemptions available for use in total due to the carryover of unused exemptions from

previous fiscal years.” Id. In USDA’s view, “the indefinite carryover and accumulation of

unused exemptions is inconsistent with Congress’ decision to limit the number of exemptions

available to States in a given fiscal year, as expressed” in § 2015(o)(6)(C), (D), and (E). Id.

Under the Final Rule, each state is of course still granted discretionary exemptions

totaling 12% of the individuals in the state covered by the work requirements, as required by the

statute. 7 U.S.C. § 2015(o)(6)(E). This 12% is also supplemented by exemptions carried over

from previous years, but this carry over is much more limited than under the 2001 regulation.

Specifically, the Final Rule “allow[s] States to carry over only one year’s worth of exemptions

from previous years.” Id. at 66803. The Final Rule does “not allow States to retain their existing

accumulated exemptions past the end of FY 2020.” Id. USDA planned to implement this

change “during the next scheduled” calculation of exemptions, for FY 2021, on October 1, 2020.

Id. at 66804.

15
C. Procedural Background

Six weeks after publication of the Final Rule, on January 16, 2020, 14 states, the District,

and New York City sued under the APA to challenge the Rule. See Complaint, D.C. et al. v.

U.S. Dep’t of Agric., et al., 20-cv-119, ECF No. 1. An amended complaint, filed January 29,

2020, added five new states as plaintiffs. See States’ Amended Compl. at 1–3, ECF No. 19.

Joining the District and the City of New York in this suit are New York, California, Colorado,

Connecticut, Hawaii, Illinois, Maine, Maryland, Massachusetts, Michigan, Minnesota, Nevada,

New Jersey, New Mexico, Oregon, Pennsylvania, Rhode Island, Virginia, and Vermont

(collectively, the state plaintiffs). See id. Their complaint challenges the Final Rule under the

APA as arbitrary and capricious, contrary to law, and promulgated without observing procedural

requirements. See 5 U.S.C. § 706(2)(A), (D); see also States’ Amended Compl. ¶¶ 461–71

(contrary to law); id. ¶¶ 472–77 (procedural claim); ¶¶ 478–82 (arbitrary and capricious). When

they filed their initial complaint, the original state plaintiffs moved for a preliminary injunction

or for a stay of the Final Rule pending judicial review. See 5 U.S.C. § 705; see generally State

Pls.’ Mot.

Also, on January 16, 2020, Bread for the City (“BFC”) and the individual plaintiffs filed

a separate suit making the same claims as the state plaintiffs about the waiver portions of the

Final Rule, see generally Private Pls.’ Compl., and moved for a preliminary injunction, see

Private Pls.’ Mot. The private plaintiffs do not challenge the discretionary exemption aspect of

the Final Rule. The two suits were consolidated under Federal Rule of Civil Procedure 42

without objection from the parties, on January 23, 2020. See Min. Order on Consolidation (Jan.

23, 2020).5

5
USDA, George Ervin Purdue, III, in his official capacity as Secretary of Agriculture, and the United States
are named as defendants in the state plaintiffs’ suit, see States’ Amended Compl. at 3, while USDA and the United

16
At the Court’s direction, the parties conferred and proposed a schedule for briefing on the

pending motion for preliminary injunctive relief, see Min. Order (Jan. 21, 2020); Notice of Joint

Proposed Schedule and Positions Regarding Consolidation, ECF No. 10. The briefing schedule

proposed by the parties was adopted, with their proposed date for a hearing moved up a week

earlier. Scheduling Order (Jan. 23, 2020).

USDA opposed the preliminary injunction, see Defs.’ Consolidated Opp’n to Pls.’ Mots.

for Prelim. Inj. (“USDA’s Opp’n”), ECF No. 26, and the plaintiffs filed replies in support of

their motions, see Private Pls.’ Reply in Supp. of Mot. for Prelim. Inj. (“Private Pls.’ Reply”),

ECF No. 30; Pls.’ Reply in Supp. of Mot. for Prelim. Inj. or 5 U.S.C. § 705 Stay Pending

Judicial Review (“State Pls.’ Reply”), ECF No. 31.6 A hearing on these pending motions was

held on March 5, 2020.

II. LEGAL STANDARD

A preliminary injunction “is a stopgap measure, generally limited as to time, and

intended to maintain a status quo or ‘to preserve the relative positions of the parties until a trial

on the merits can be held.’” Sherley v. Sebelius, 689 F.3d 776, 781–82 (D.C. Cir. 2012) (quoting

Univ. of Texas v. Camenisch, 451 U.S. 390, 395 (1981)). To obtain a preliminary injunction, the

plaintiffs must establish that (1) they are likely to succeed on the merits, (2) they are likely to

suffer irreparable harm in the absence of preliminary relief, (3) the balance of equities tips in

their favor, and (4) an injunction is in the public interest. See Winter v. Natural Res. Def.

States are named in the suit brought by BFC and the individual plaintiffs, see Private Pls.’ Compl. at 1. USDA, the
Secretary, and the United States are defendants in the consolidated suit and are collectively referred to as “USDA.”
6
The United States House of Representatives filed an amicus brief in support of the plaintiffs, see Br. of
U.S. House of Rep. as Amicus Curiae in Supp. of Pls. (“House Amicus Br.”), ECF No. 16, and organizations “that
study, develop and support” certain welfare reforms, Amici Curiae Br. of the Foundation for Gov’t Accountability et
al., ECF No. 37, as well as nine states — Arizona, Alabama, Arkansas, Georgia, Kentucky, Louisiana, Nebraska,
South Carolina, and Texas — submitted amici briefs in support of the USDA, see Br. of Amici States of Ariz. et al.
(“State Amici Br.”), ECF No. 42.

17
Counsel, 555 U.S. 7, 20 (2008).7 The first factor is also the “most important factor.” Aamer v.

Obama, 742 F.3d 1023, 1038 (D.C. Cir. 2014). A preliminary injunction “is an extraordinary . . .

remedy, one that should not be granted unless the movant, by a clear showing, carries the burden

of persuasion” on each of the factors. Mazurek v. Armstrong, 520 U.S. 968, 972 (1997) (per

curiam) (emphasis omitted) (quoting 11A C. Wright, A. Miller, & M. Kane, FEDERAL PRACTICE

AND PROCEDURE § 2948 (2d ed.1995)).

Section 705 of the APA authorizes “the reviewing court” to stay “the effective date of an

agency action” pending judicial review “to prevent irreparable injury.” 5 U.S.C. § 705. The

factors governing issuance of a preliminary injunction also govern issuance of a § 705 stay. See,

e.g., Texas v. EPA, 829 F.3d 405, 424, 435 (5th Cir. 2016); Humane Soc’y of United States v.

Gutierrez, 558 F.3d 896, 896 (9th Cir. 2009); Cuomo v. U.S. Nuclear Regulatory Comm’n, 772

F.3d 972, 974 (D.C. Cir. 1985).

III. DISCUSSION

The discussion begins with review of the state plaintiffs’ challenge to the discretionary

exemption aspect of the Final Rule and concludes this challenge is not likely to succeed on the

merits. The state plaintiffs’ motion for a preliminary injunction as to that aspect of the Final

Rule is thus denied.

The discussion next turns to plaintiffs’ arbitrary and capricious challenges to the waiver

aspects of the Final Rule and concludes that these claims are highly likely to succeed. All the

plaintiffs have also shown that implementation of those aspects of the Final Rule will cause them

7
The D.C. Circuit has previously followed a “sliding scale” approach to evaluating preliminary injunctions,
but that approach is likely inconsistent with Winter, see Archdiocese of Washington v. WMATA, 897 F.3d 314, 334
(D.C. Cir. 2018) (observing that Winter may be “properly read to suggest a ‘sliding scale’ approach to weighing the
four factors be abandoned), and therefore will not be employed here, see Singh v. Carter, 185 F. Supp. 3d 11, 16
(D.D.C. 2016).

18
irreparable harm. The balance of the equities and the public interest further weigh in favor of

preliminary relief for the plaintiffs. Finally, the discussion addresses the proper scope of the

relief, determining that a nationwide injunction of the waiver parts of the Final Rule is consistent

with D.C. Circuit precedent and necessary to provide complete relief.

A. State Plaintiffs Are Not Likely To Succeed In Challenging The Final
Rule’s Discretionary Exemption Change

The state plaintiffs claim that the discretionary exemption aspect of the Final Rule is

unlawful in two respects. According to these plaintiffs, first, the Final Rule’s limitation on the

carryover of unused exemptions to one year “runs afoul of the statutory scheme,” State Pls.’

Mem. at 24, and, second, this part of the Final Rule is arbitrary and capricious, by “fail[ing] to

address significant reliance interests,” id. at 37. States are understandably upset that USDA’s

Final Rule radically changes a longstanding policy to the states’ detriment. Indeed, most of the

fifty states may stand to lose accumulated exemptions on October 1, 2020, when the

discretionary exemption portion of the Final Rule becomes effective.8 That said, “[a]gencies are

free to change their existing policies,” even to change existing policies to the significant

detriment of states or other parties, “as long as [agencies] provide a reasoned explanation for the

change” and ground the change in a reasonable interpretation of the governing statute. Encino

Motorcars, LLC v. Navarro, 136 S. Ct. 2117, 2125 (2016). USDA likely satisfied that standard

here, so plaintiffs’ motion for preliminary relief will be denied as to the discretionary exemption

aspect of the Final Rule.

8
The current record is not clear on this point, but any state with areas subject to the work requirements at
some point since 1999, when carry over began, could lose carried over exemptions as a result of the Rule. The
District has had a statewide waiver since the enactment of the work requirements and therefore has not accumulated
any exemptions. See Rough Hr’g Tr. at 32:7–23. Four other jurisdictions currently have statewide waivers, but the
record does not indicate whether those statewide waivers have been in effect since 1999 as in the District. See id. at
32:1–6.

19
The state plaintiffs’ argument about the Final Rule being contrary to law is governed by

the test from Chevron U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837

(1984). At Chevron step one, a court asks “whether Congress has directly spoken to the precise

question at issue.” Id. at 842. If Congress has spoken, then the agency action comports with the

law only if the action “give[s] effect to the unambiguously expressed intent of Congress.” Id. at

842–43. If Congress has been “silent or ambiguous” on the question at issue, then, at Chevron

step two, the court asks whether the agency’s action “is based on a permissible construction of

the statute,” and the court must defer to the agency’s interpretation if reasonable. Id. at 843.

Section 2015(o)(6)(E) allows states to exempt from SNAP program work requirements

12% of ABAWDs, a reduction in the 15% exemption limit that applied before fiscal year 2020.

See 7 U.S.C. § 2015(o)(6)(C)–(D). Section 2015(o)(6)(G) instructs USDA to adjust this 12%

limit annually. Specifically, USDA is to “increase or decrease the number of individuals who

may be granted an exemption by a State agency . . . to the extent that the average monthly

number of exemptions in effect in the State for the preceding fiscal year . . . is lesser or greater

than the average monthly number of exemptions estimated for the State agency for such

preceding fiscal year.” Id. § 2015(o)(6)(G).

The state plaintiffs argue that § 2015(o)(6)(G) unambiguously requires USDA to

compare the number of exemptions used in the prior year to the number of exemptions available

in the prior year, including all exemptions carried over from prior years, and then carry over the

difference. See State Pls.’ Mem. at 24 (stating that the provision “requires the Secretary to . . .

effect[] an ongoing ‘carry-over’ of exemptions from prior years: the underuse of exemptions in

one year will increase the number of exemptions available in the next year, and that underuse

compared in the second year will roll over to the third and so forth”). Under this reading, the

20
state plaintiffs construe the phrase “average monthly number of exemptions estimated . . . for the

preceding fiscal year” to mean the number of exemptions available in the prior year, without

regard to whether those exemptions were initially granted in the prior year or, say, twenty years

ago in 1999. 7 U.S.C. § 2015(o)(6)(G). That phrase could bear the state plaintiffs’ reading, but

it could also bear USDA’s.

To USDA, the statute is at best ambiguous on the issue of whether “number of

exemptions estimated . . . for the preceding fiscal year” should include exemptions carried over

from years prior to the “preceding fiscal year.” 7 U.S.C. § 2015(o)(6)(G). If anything, in

USDA’s view, the statutory language’s “focus on the ‘preceding fiscal year,’” USDA’s Opp’n at

30 (quoting 7 U.S.C. § 2015(o)(6)(G)), means that Congress intended USDA to credit unused

exemptions granted in the preceding year but not unused exemptions granted in any number of

prior years. A statute that can bear multiple plausible meanings is ambiguous at Chevron step

one. See, e.g., Whitaker v. Thompson, 353 F.3d 947, 949–50 (D.C. Cir. 2004) (seeing “no basis

for finding any unambiguously expressed intent of Congress” where the sentences at issue

“might mean” one thing and “might mean” another (internal quotation marks omitted)).

In reply, the plaintiffs argue that, even if the statute’s text appears ambiguous, “the

legislative history unambiguously contemplates” their reading. State Pls.’ Reply at 16. Even

assuming that legislative history could make ambiguous text clear at Chevron step one, the

legislative history does not do what the state plaintiffs say it does.9 The 2018 Farm Bill reduced

the percentage of available exemptions from 15% to 12% starting in FY 2020. Agriculture

9
The state plaintiffs argue that “to the extent that congressional silence on time limits is ambiguous,
legislative history can fill the gap” as to Congress’s intent at Chevron step one, but they point to no case in which a
court has used legislative history to deem ambiguous statutory language clear. State Pls. Reply at 17. The typical
use of legislative history at Chevron step one is to “shed new light on” an ambiguity in congressional intent,
“notwithstanding statutory language that appears superficially clear.” Nat. Res. Def. Council, Inc. v. Browner, 57
F.3d 1122, 1125 (D.C. Cir. 1995) (quoting Am. Scholastic TV Programming Found. v. FCC, 46 F.3d 1173, 1180
(D.C. Cir. 1995)).

21
Improvement Act of 2018, Pub. L. No. 115-334, § 4005, 132 Stat. 4490, 4632 (2018); see also 7

U.S.C. § 2015(o)(6)(E). While the Senate version of the Farm Bill left exemptions untouched,

see H.R. 2, 115th Cong. § 4103(F) (June 28, 2018) (Engrossed Amendment in Senate), the

House version had not only reduced the percentage to 12%, albeit starting in 2026, but also

entirely eliminated exemption carry over, see H.R. 2, 115th Cong. § 4015(a)(G)(iv) (June 21,

2018) (Engrossed in House).

State plaintiffs rest their argument on the following sentence in the Conference Report

explaining the Farm Bill’s ultimate compromise: “States will maintain the ability to exempt up to

12% of their SNAP population subject to ABAWD work requirements, down from 15%, and

continue to accrue exemptions and retain any carryover exemptions from previous years,

consistent with current law.” H.R. Rep. No. 115-1072, at 616 (2018) (Conf. Rep.) (quoted in

State Pls.’ Mem. at 24 and State Pls.’ Reply at 17). Crediting plaintiffs’ reading of this sentence

as a congressional attempt to clarify that exemptions carry over “from previous years,” in the

plural, does not, however, dictate the result they want of rendering clear otherwise ambiguous

statutory language on this point. See NLRB v. Health Care & Ret. Corp. of Am., 511 U.S. 571,

582 (1994) (“[I]t is the function of the courts and not the Legislature, much less a Committee of

one House of the Legislature, to say what an enacted statute means.” (alteration in original)

(quoting Pierce v. Underwood, 487 U.S. 552, 566 (1988)); Consumer Prod. Safety Comm’n v.

GTE Sylvania, Inc., 447 U.S. 102, 118 n.13 (1980) (dismissing similar argument by recognizing

that “a mere statement in a conference report” of subsequent legislation “as to what the

Committee believes an earlier statute meant is obviously” not authoritative of the original

statute’s meaning); Eagle Pharm. Inc. v. Azar, No. 18-5207, slip op. at 30 (D.C. Cir. Mar. 13,

2020) (stating that legislative history from subsequent legislation is “particularly unhelpful” and

22
“less persuasive” than legislative history from the original legislation); Verizon v. FCC, 740 F.3d

623, 639 (D.C. Cir. 2014) (stating the general rule of statutory interpretation that “subsequent

failed legislation tell[s] us little if anything about the original meaning of” a statutory provision).

Moreover, the Conference Report’s statement was made in the specific context of explaining the

Conference Committee’s decision to reject the House’s total elimination of carryover exemptions

and does not purport to bar agency action in the future. Put another way, this legislative history

does not supplant statutory language nor amount to a congressional command requiring USDA

to retain forever into future rulemakings the 2001 regulation’s method of carrying over

exemptions indefinitely. Such a command would need to be clearly stated in the statute. See

Health Care & Ret. Corp. of Am., 511 U.S. at 582 (reaffirming that statements in committee

reports “do not have ‘the force of law, for the Constitution is quite explicit about the procedure

that Congress must follow in legislating’” (quoting Am. Hosp. Ass’n. v. NLRB, 499 U.S. 606, 616

(1991)); Puerto Rico Dep't of Consumer Affairs v. Isla Petroleum Corp., 485 U.S. 495, 501

(1988) (“[U]nenacted approvals, beliefs, and desires are not laws.”).

The state plaintiffs’ statutory argument thus fails because the statute bears multiple

readings and USDA has settled on one permissible reading of the statute. See Chevron, 467 U.S.

at 843 (instructing courts to reach Chevron step two if the statute is ambiguous and to defer to

the agency if the agency’s reading is a “permissible construction”).10

10
The state plaintiffs also argue, in a single sentence, that the Final Rule is contrary to law because it violates
the APA’s prohibition on retroactive legislative rules. See State Pls.’ Mem. at 25; see also State Pls.’ Reply at 17
n.7 (preserving this argument). State plaintiffs are also unlikely to succeed on this theory, at least as currently
presented. Distinctions are drawn, as a legal mater, “between a rule that imposes new sanctions on past conduct,
which is retroactive and invalid [under the APA] unless specifically authorized, and one that merely ‘upsets
expectations,’ which is secondarily retroactive and invalid only if arbitrary and capricious.” Nat’l Petrochemical &
Refiners Ass’n v. EPA, 630 F.3d 145, 159 (D.C. Cir. 2010) (citing Nat’l Cable & Telecomms. Ass’n v. FCC, 567
F.3d 659, 670–71 (D.C. Cir. 2009)). The Final Rule merely upsets expectations by “impair[ing] the future value of
past” years’ carried over exemptions and is thus valid unless arbitrary and capricious. Id. at 161 (quoting Nat’l
Cable, 567 F.3d at 670). The Final Rule does not alter the past legal consequences of past decisions about carrying
over exemptions. See Mobile Relay Assocs. v. FCC, 457 F.3d 1, 11 (D.C. Cir. 2006) (“[A]n agency order that ‘alters

23
State plaintiffs next argue that the Final Rule is arbitrary and capricious for several

reasons. First, state plaintiffs believe the Final Rule is “unreasoned.” State Pls.’ Mem. at 37.

The Final Rule explained USDA’s view that “the indefinite carryover . . . of unused exemptions

is inconsistent with Congress’ decision to limit the number of exemptions available to States in a

given fiscal year, as expressed by sections 6(o)(6)(C), (D), and (E).” 84 Fed. Reg. at 66802

(citing 7 U.S.C. § 2015(o)(6)(C)–(E)). The Final Rule further illustrated that the number of

“unused discretionary exemptions” accumulated by states “well exceed[s] the number allotted to

each state for the fiscal year.” Id. “In FY 2019, States earned approximately 1.3 million

exemptions, but had about 7.4 million exemptions available for use in total due to the carryover

of unused exemptions from previous fiscal years.” Id. As the agency noted in the Proposed

Rule, a 2016 USDA Office of the Inspector General (OIG) report “expressed” “[c]oncerns”

about whether such vast accumulation of exemptions was what the statute intended. 84 Fed.

Reg. at 988; see also USDA’s Opp’n, Casey McConnell Decl. (“McConnell Decl.”), Ex. 1, FNS

Controls Over SNAP Benefits for Able-Bodied Adults Without Dependents (“OIG Report”) at

10–11 (Sept. 2016), ECF No. 26-1. These were “entirely rational” reasons to revise how

discretionary exemptions were carried over. FCC v. Fox Television Stations, Inc., 556 U.S. 502,

517 (2009); see also Rust v. Sullivan, 500 U.S. 173, 187 (1991) (deeming “justified” with

“reasoned analysis” a change in policy that the agency viewed as “more in keeping with the

the future effect, not the past legal consequences’ of an action, Sinclair Broad. Group v. FCC, 284 F.3d 148, 166
(D.C. Cir. 2002), or that ‘upsets expectations based on prior law,’ DIRECTV, Inc. v. FCC, 110 F.3d 816, 826 (D.C.
Cir. 1997) (quotation omitted), is not retroactive” in the way the APA forbids.). State plaintiffs argue that “[b]y
cancelling previously accumulated exemptions aside from those accrued in the preceding fiscal year, the Rule”
impermissibly “extinguishes prior state rights” without specific authorization. State Pls.’ Mem. at 24. The analysis
of the statute above, however, demonstrates that “the legal status quo ante” did not create inalienable rights to
maintain unused exemptions beyond the next fiscal year. Bergerco Canada, a Div. of Conagra, Ltd. v. U.S.
Treasury Dep’t, 129 F.3d 189, 193 (D.C. Cir. 1997); see also Celetronix Telemetry, Inc. v. FCC, 272 F.3d 585, 589
(D.C. Cir. 2001) (holding that no “vested right” existed for purposes of retroactivity analysis where the prior
“system offered no vested right to any specific terms” of a license and the agency “always retained the power to
alter the term of existing licenses by rulemaking”).

24
original intent of the statute” and that relied on OIG reports concluding that prior policy did not

properly implement the statute).

Second, the state plaintiffs argue that the Final Rule “summarily discounts the numerous

comments that relay the importance of the exemptions in enabling states to nimbly respond to

swift economic downturns or changes in employment conditions.” State Pls.’ Mem. at 37. To

the contrary, USDA “agree[d]” with comments that “demonstrate[d] the importance of

discretionary exemptions.” 84 Fed. Reg. at 66803. The agency simply concluded that those

comments did not demonstrate that states needed to retain unused exemptions “indefinitely” to

enable nimble response to changing conditions. Id. That conclusion appears rational in light of

the evidence before the agency. States’ history of accumulating rather than using exemptions —

such that 82.5% of exemptions in effect in 2019 were carried over from past years — is

consistent with the agency’s determination that exemptions could be scaled back without

threatening states’ ability to “deal with potential unforeseen sharp economic declines or other

quickly changing circumstances.” Id.

Finally, the state plaintiffs argue that the Final Rule “fails to confront the serious reliance

interests of states that have accumulated exemptions for many years.” State Pls.’ Mem. at 37.

Those reliance interests are undoubtedly serious, but “reliance does not overwhelm good reasons

for a policy change,” Encino Motorcars, 136 S. Ct. at 2128 (Ginsburg, J., concurring), and

USDA appears to have given good reasons here. An explanation of why one “policy ‘is more

consistent with statutory language’ than alternative policies” is a good reason for a policy

change, id. (majority opinion) (quoting Long Island Care at Home, Ltd. v. Coke, 551 U.S. 158,

175 (2007)), as is remediating a concern identified in an OIG report, see Rust, 500 U.S. at 187

(upholding a change in policy that relied in part on OIG reports).

25
Furthermore, the Final Rule explicitly confronted and considered the states’ reliance

interests. As already noted, USDA acknowledged that discretionary exemptions were important

to states but reasonably concluded that more limited carryover would still allow states the

flexibility to respond to suddenly changing circumstances. 84 Fed. Reg. at 66803. In addition,

recognizing that states with accumulated exemptions may have created policies and plans on the

assumption that those exemptions would be available into the future, USDA chose to implement

the change in October 2020, rather than April 2020, to give states time to “make significant

changes” to their current systems of exemption use “to ensure they do not overuse exemptions”

once the Final Rule goes into effect. Id. at 66804.

In sum, the state plaintiffs have not shown a likelihood of success on the merits of their

challenges to the discretionary exemption aspects of the Final Rule.

Further, without fully analyzing the matter, the Court notes that the plaintiffs may face an

uphill battle in showing irreparable harm. For one, this aspect of the Final Rule does not become

effective until October 1, 2020, and harm that is eight months away may not be sufficiently

immediate. For another, on this record, states may struggle to establish that ending indefinite

carryover of discretionary exemptions will have a certain impact. This is not to diminish the

importance of discretionary exemptions, especially in responding to sudden events like

manufacturing plant closures, or even pandemics. That said, some states and areas, including the

District and New York City, have earned no or very few exemptions because they have been

under waivers for quite some time, and thus are not impacted at all by the elimination of

indefinite carryover. Further, states will continue to earn exemptions yearly at the rate of 12% of

covered individuals, and the states’ declarations are vague in their assertions about whether that

continued accrual is insufficient to continue to exempt the individuals whom the states have been

26
exempting. See, e.g., Catherine Buhrig Decl. (“Buhrig Decl.”) ¶ 17, ECF No. 3-4 (noting that

Pennsylvania currently has approximately 160,000 exemptions, will lose 140,000 with the result

that “the State may not be able to exempt everyone that it has exempted in the past,” but without

stating the number actually needed (emphasis added)); Alexis Carmen Fernández (“Fernández

Decl.”) ¶¶ 19–22, ECF No. 3-5 (stating simply that California has 850,000 carried over

exemptions and asserting that these are “instrumental in providing food benefits in the six

counties currently implementing the time limit,” but without stating the number actually needed).

The state plaintiffs’ motion for preliminary relief as to those aspects of the Final Rule is

denied. See Munaf v. Geren, 553 U.S. 674, 705 (2008) (reversing entry of a preliminary

injunction where the movant failed to show likelihood of success on the merits); see also, e.g.,

Guedes v. Bureau of Alcohol, Tobacco, Firearms & Explosives, 920 F.3d 1, 10 (D.C. Cir. 2019)

(“[B]ecause the plaintiffs have shown no likelihood of success on the merits, we choose not to

‘proceed to review the other three preliminary injunction factors.’” (quoting Ark. Dairy Coop.

Ass’n v. USDA, 573 F.3d 815, 832 (D.C. Cir. 2009))); Davis v. Pension Benefit Guar. Corp., 571

F.3d 1288, 1296 (D.C. Cir. 2009) (Kavanaugh, J., concurring) (“Munaf means that a strong

showing of irreparable harm, for example, cannot make up for a failure to demonstrate a

likelihood of success on the merits.”); cf. Citizens for Responsibility & Ethics in Wash. v. FEC,

904 F.3d 1014, 1019 (D.C. Cir. 2018) (per curiam) (stating that a movant’s failure to show

likelihood of success on the merits is “an arguably fatal flaw for a stay [pending appeal]

application.”).

B. Plaintiffs Are Likely To Succeed On The Merits Of Their Arbitrary
And Capricious Claim Challenging The Final Rule’s Waiver Changes

The APA requires reviewing courts to “hold unlawful and set aside agency action” that is

“arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law.” 5 U.S.C.

27
§ 706(2)(A). “An agency action that ‘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’ is arbitrary and capricious.” Gresham v. Azar, 950 F.3d 93, 99 (D.C. Cir.

2020) (quoting Motor Vehicle Mfrs. Ass’n of U.S., Inc. v. State Farm Mut. Auto. Ins. Co., 463

U.S. 29, 43 (1983)). “That is not a high bar,” the Supreme Court has observed, “but it is an

unwavering one.” Judulang v. Holder, 565 U.S. 42, 45 (2011).

The plaintiffs are likely to succeed on their claim that the waiver changes in the Final

Rule are arbitrary and capricious. Given this, the plaintiffs’ other APA claims — that the Rule is

contrary to law and was promulgated without observing procedural requirements — need not be

addressed at this stage of the case.11 The Rule’s severe reduction of the allowable criteria for

demonstrating lack of sufficient jobs is addressed first before turning to the Rule’s redefinition of

“area.”

11
To be clear, the plaintiffs’ contrary to law arguments are addressed to the extent those arguments overlap
with their arbitrary and capricious claim. See infra Part III.B.1. The private plaintiffs’ contrary to law argument that
USDA exceeded its rulemaking authority by promulgating a rule that “substantially displace[d] the waiver
adjudication process established by Congress,” Private Pls.’ Mem. at 21, is not addressed. Further, the state
plaintiffs’ procedural claim stemming from USDA’s reliance on the phrase “operational experience,” see State Pls.’
Mem. at 17, is addressed to the extent it overlaps with their arbitrary and capricious claim, see infra Part III.B.2.
The decision to postpone consideration of any arguments at this stage in no way reflects any assessment that those
arguments are less likely to succeed. Pointing to case law holding that the APA’s notice and comment requirements
are not satisfied where the agency does the opposite of what it proposed, see State Pls.’ Mem. at 14 (citing, for
example, Allina Health Servs. v. Sebelius, 746 F.3d 1102, 1110–11 (D.C. Cir. 2014)), the plaintiffs argue that USDA
did not provide adequate notice of its decision to eliminate state waivers based on the state’s qualification for
extended unemployment benefits, id. The Proposed Rule stated the opposite: that USDA “would continue to
approve a state’s waiver request that is based upon the requesting state’s qualification for extended unemployment
benefits.” 84 Fed. Reg. at 985. The state plaintiffs further argue that the Proposed Rule did not give notice that
LMAs “would be the sole geographical unit for a waiver,” and that this failure “deprived Plaintiffs and the public of
meaningful comment” on a consequential aspect of the Final Rule. State Pls.’ Mem. at 14, 16 (emphasis added).
USDA counters that the Final Rule was a “logical outgrowth” of the notice and by attempting to distinguish the case
law plaintiffs rely on. See USDA’s Opp’n at 48–52. USDA suggests that these procedural challenges may be best
evaluated on a full administrative record, which is currently being compiled. See USDA’s Opp’n at 4 (arguing that
plaintiffs’ procedural “claims founder in light of the administrative record”); see also McConnell Decl. ¶ 34
(“USDA-FNS is in the process of compiling the administrative record for this case.”). The Court accepts that
suggestion given that the preliminary injunction motion may readily be resolved on other grounds.

28
1. The Rule’s Adoption of New Criteria Governing Approval of Waivers
is Likely Arbitrary and Capricious

The plaintiffs are likely to show that the Rule’s adoption of new criteria for evaluation of

waivers is arbitrary and capricious in two respects. First, the agency’s decision to eliminate

measures of insufficient jobs other than 24-month average unemployment rate runs counter to

evidence before the agency, and USDA’s decision to depart from that evidence was inadequately

explained. Second, USDA’s singular reliance on unemployment rates in implementing 7 U.S.C.

§2015(o)(4)(A) was arbitrary and capricious in light of the statutory language, structure, and

other indicia of congressional intent.

a. Decision to Eliminate Measures of Insufficient Jobs Other than 24-
Month Average Unemployment Rate

Voluminous evidence supported the commenters’ view that general unemployment rates

alone cannot measure whether an area has “a sufficient number of jobs to provide employment

for the individuals.” 7 U.S.C. § 2015(o)(4)(A)(ii) (emphasis added). As already reviewed, the

Rule summarized “analysis” and “research” submitted by the commenters “indicating” that

ABAWDs faced greater barriers to employment than the general population because ABAWD

tended to have lower educational levels, higher levels of disability and homelessness, worse

access to transportation, and were more likely to have prior criminal convictions and to

experience racial discrimination. See, e.g., 84 Fed. Reg. at 66787 (reciting data from critical

commenters “based on the 2017 USDA Household Characteristics data, that non-disabled

individuals aged 18 through 49 in households without children in SNAP report lower than

average educational attainment” (citation omitted)); id. (noting submitted “research indicating

that, on average, unemployment rates for people with low-education attainment are much higher

than what BLS unemployment rates for the general public indicate”); id. (stating that

“[c]ommenters provided research indicating that lower unemployment rates are less indicative of

29
strong labor markets in recent years than in the past, and particularly for those with lower levels

of education”); id. (observing that “research indicat[es] that employment rates for workers with

low levels of education still have not recovered from the recession and pointed to evidence that

workers with less education may be hit harder by recessions”); id. (considering USDA-

“commissioned” analysis “that those subject to SNAP work requirements face substantial

barriers to employment.”).

USDA waved away these commenters’ concerns and their supporting evidence in a few

sentences of defective argument. Although USDA conceded “that ABAWDs may face barriers

to employment and have more limited employment prospects than the general public,” 84 Fed.

Reg. at 66787, it insisted, “based on its operational experience,” that “areas that do not clearly

lack sufficient jobs will continue to qualify for waivers” without the Rule’s limitations. Id.

Putting the lesson slightly differently, the Rule also stated that, “[t]hrough its operational

experience, the Department has recognized that” measures not based on unemployment rates

“are less reliable and consistent that standard unemployment data in demonstrating a lack of

sufficient jobs.” Id. at 66791. USDA’s sole supporting example was that, if evaluation of

sufficient jobs were not pegged to unemployment rates, and specifically to unemployment rates

over 6%, then “a state could request and qualify for a waiver in areas with an unemployment rate

as low as of 4.7 percent.” Id. at 66787.

USDA’s logic is arbitrary because it is circular. The agency erases the concern that

general unemployment rates are an inappropriate measure of lack of sufficient jobs for

ABAWDs by redefining the thing to be measured in terms of the agency’s preferred measure,

namely, general unemployment rates. USDA’s example suffers from the same defect of

30
circularity. USDA’s semantic trick does not address the evidence that the Final Rule simply is

not measuring what the statute requires: the lack of sufficient jobs for ABAWDs.

USDA gave no other reason for its tunnel-vision embrace of unemployment rates.

Indeed, USDA has embraced non-unemployment rate–based measures since 1996, including in

the 2001 regulation. A new policy of viewing these measures as unreliable or inconsistent for

purposes of assessing whether an area has “a sufficient number of jobs to provide employment

for the individuals,” 7 U.S.C. § 2015(o)(4)(A)(ii), cannot be justified by “merely recit[ing] the

terms” reliable and consistent. State Farm, 463 U.S. at 52.12 Instead, when faced with

considerable evidence that its preferred measure was inappropriate or incomplete in the precise

context defined by the statute, the agency needed to provide a meaningful response to that

evidence and “articulate a satisfactory explanation for its action including a ‘rational connection

between the facts found and the choice made.’” State Farm, 463 U.S. at 43 (quoting Burlington

Truck Lines, Inc. v. United States, 371 U.S. 156, 168 (1962)); see also, e.g., Gresham, 950 F.3d

at 103 (deeming an agency’s decision arbitrary and capricious when it “dismiss[ed] important

“concerns” “in a handful of conclusory sentences”); Am. Wild Horse Pres. Campaign v. Purdue,

873 F.3d 914, 932 (D.C. Cir. 2017) (vacating an agency’s action that “brushed aside critical

facts” and failed to “adequately explain” or “adequately analyze” its policy choice); Tesoro

Alaska Petroleum Co. v. FERC, 234 F.3d 1286, 1294 (D.C. Cir. 2000) (“The Commission's

failure to respond meaningfully to the evidence renders its decisions arbitrary and capricious.”).

12
In addition, a look at the full Final Rule indicates that the agency embraced DOL data inconsistently.
USDA invoked the supposedly superior reliability of DOL data in eliminating three measures of job availability in
the 2001 regulation — a declining employment-to-population ratio, jobs in declining industries or occupations, and
description in an academic study or publication as an area lacking jobs, 84 Fed. Reg. at 66790 — but failed to
acknowledge or address this preference for DOL data in rejecting a measure that relied on DOL data — LSA
designation by DOL, see id. at 66799. Such inconsistency is a signal of arbitrary decisionmaking. ANR Storage Co.
v. FERC, 904 F.3d 1020, 1024 (D.C. Cir. 2018) (stating that an agency’s “reasoning cannot be internally
inconsistent” (citing Sierra Club v. EPA, 884 F.3d 1185, 1194–96 (D.C. Cir. 2018)).

31
USDA now argues that the eliminated metrics were “rarely used,” including by the state

plaintiffs. USDA’s Opp’n at 40 (citing McConnell Decl., Exs. 13–28). To USDA, this shows

that “Plaintiffs’ true complaint is not with the use of unemployment rates,” but with the height of

the unemployment rate floor. Id. USDA’s view of the plaintiffs’ “true complaint” is beside the

point: that many past or current waivers are based on unemployment rates does not change the

record before the agency indicating problems with its approach or the inadequacy of the agency’s

response to those problems.

b. Singular Reliance on Unemployment Rates

In § 2015, Congress provided two criteria for waivers: (i) that an area has “an

unemployment rate of over 10 percent; or” that the area (ii) “does not have a sufficient number

of jobs to provide employment for the individuals.” 7 U.S.C. § 2015(o)(4)(A). State plaintiffs

argue that the Rule unlawfully implements § 2015(o)(4)(A) because neither of the Final Rule’s

two core standards “answers the question that Congress asked — whether an area actually has

insufficient jobs to provide employment for ABAWDs.” State Pls.’ Mem. at 22.13 Plaintiffs are

13
State plaintiffs frame their arguments about the Rule’s singular reliance on unemployment rates under the
test from Chevron. In their view, the Rule falls at Chevron’s step one because the statutory language
unambiguously “forecloses the Rule’s near-total reliance on the general unemployment rate to meet the test.” State
Pls.’ Mem. at 21. USDA counters that “the phrase in question . . . is the ‘antithesis of a Chevron step one statutory
directive.’” USDA’s Opp’n at 23 (quoting Anna Jacques Hosp. v. Burwell, 797 F.3d 1155, 1164 (D.C. Cir. 2015)).
The phrase “sufficient number of jobs to provide employment for the individuals” is not obviously unambiguous at
Chevron step one, especially when read in light of the context, which signals congressional intent to delegate aspects
of waiver determinations to the agency: “the Secretary may waive” the work requirements if “the Secretary makes a
determination that the area in which the individuals reside . . . does not have a sufficient number of jobs to provide
employment for the individuals.” 7 U.S.C. § 2015(o)(4)(A). Thus, a Chevron analysis would likely proceed to step
two.
In any event, there is little distance between Chevron’s step two and arbitrary and capricious review under
the APA. See Judulang v. Holder, 565 U.S. 42, 52 n.7 (2011) (stating that the “analysis would be the same” under
Chevron step two and APA arbitrary and capricious review); see also Humane Soc’y of United States v. Zinke, 865
F.3d 585, 605 (D.C. Cir. 2017) (“[A]nalysis of the reasonableness of agency action ‘under Chevron Step Two and
arbitrary and capricious review is often the same.’” (quoting Pharm. Res. & Mfrs. of Am. v. FTC, 790 F.3d 198, 209
(D.C. Cir. 2015)). “[T]he more apt analytical framework in this case is standard ‘arbitrary [or] capricious’ review
under the APA.” Judulang, 565 U.S. at 52 n.7 (alteration in original) (quoting 5 U.S.C. § 706(2)(A)). Although the
Rule implements the statutory language of 7 U.S.C. § 2015(o)(4)(A), the Rule does not dwell on “an interpretation
of any statutory language.” Judulang, 565 U.S. at 52 n.7; see also Nat’l Ass’n of Regulatory Util. Comm’rs v. ICC,
41 F.3d 721, 727 (D.C. Cir. 1994) (“[A]lthough Chevron’s second step sounds closely akin to plain vanilla arbitrary-

32
likely to succeed on this claim that the Final Rule is an unreasonable construction of the statute,

Guedes, 920 F.3d at 32 (articulating the Chevron step two standard), is arbitrary or capricious in

substance, see Judulang, 565 U.S. at 52 n.7 (discussing Chevron step two), ignores “relevant

factors,” and amounts to “a clear error of judgment,” State Farm, 463 U.S. at 42–43.

The Final Rule adopts unreasonably narrow criteria for implementing the statutory

requirements for evaluating waivers. Start with the statutory structure: prong one of

§ 2015(o)(4)(A) addresses with precision one circumstance, an unemployment rate over 10%,

that would warrant a waiver, while prong two states with greater generality that waivers may be

granted for lack of “sufficient jobs.” USDA is right that prong two, in not prescribing a single

permissible measure, necessarily leaves USDA with some discretion to define “sufficient jobs.”

See USDA’s Opp’n at 23–24. Nevertheless, the statute’s two-pronged structure also plainly

limits USDA’s discretion: if prong one addresses the circumstances under which unemployment

rates would warrant a waiver, prong two must address a distinctive measure. By making both

prongs of § 2015(o)(4)(A) dependent on unemployment rate, USDA has arbitrarily written this

distinction out of the Rule. See New York v. EPA, 443 F.3d 880, 887 (D.C. Cir. 2006) (setting

aside an agency action, albeit at Chevron step one, based on an interpretation “contrary to ‘[the]

cardinal principle of statutory construction’” that words in the statute should not be made

“insignificant” or “superfluous” (quoting TRW Inc. v. Andrews, 534 U.S. 19, 31 (2001)); Shays v.

FEC, 414 F.3d 76, 105 (D.C. Cir. 2005) (vacating agency rules at Chevron step one because the

and-capricious style review, interpreting a statute is quite a different enterprise than policy-making” (internal
quotation marks and ellipsis omitted)). Rather, the Rule primarily engages in policymaking. See Mozilla Corp. v.
FCC, 940 F.3d 1, 49 (D.C. Cir. 2019) (stating that “the Venn Diagram of the two inquiries is not a circle” and
explaining that, even where an agency adopts “a reasonable interpretation of the statute for purposes of Chevron,”
the policy rationales underlying the agency action may still be arbitrary and capricious). As a result, this decision
adopts the arbitrary and capricious framework but imports the parties’ Chevron arguments where applicable. Cf.
USDA’s Opp’n at 31 n.12 (urging that its arguments that the Rule is not arbitrary and capricious should be applied
in a Chevron step two analysis).

33
agency’s too “narrow interpretation of” a statutory term was “implausible,” and acknowledging

that the district court reached the same result at Chevron step two).

Section 2015(o)(4)(A)(ii)’s use of the phrase “employment for the individuals” cements

this conclusion. That phrase is a guardrail on USDA’s discretion. It instructs USDA to

determine whether an area has “a sufficient number of jobs” with some attention given to

ABAWD-related measures.14 As already discussed, USDA has not adequately explained in this

Rule that unemployment rates, taken alone, are such a metric.

In USDA’s view, the Final Rule reasonably implements § 2015(o)(4)(A) because “[e]ven

if ABAWDs may face more challenges to finding employment than the average person, the

unemployment rate is not ‘a matter irrelevant to’ that inquiry.” USDA’s Opp’n at 24 (quoting

Judulang, 565 U.S. at 53). The statute undoubtedly makes unemployment rate a factor relevant

to evaluating waivers, but that point sidesteps the flaw in the Final Rule’s core standards. The

real issue is whether the Final Rule, in making general unemployment rates the only factor for

evaluating waivers, permissibly interpret § 2015(o)(4)(A). As just analyzed, based on the

statutory structure and text and the Final Rule’s justification, the answer is no.

Falling back, USDA says that, even if the statute does require consideration of more than

unemployment rates, the Final Rule does so. See USDA’s Opp’n at 25. A provision states that

USDA will grant a waiver based on other “data or evidence” if the state can “demonstrate that

[an] exceptional circumstance has caused a lack of sufficient jobs” in an area. 84 Fed. Reg. at

66811. Although the provision parrots the statutory phrase “sufficient jobs,” the commentary

14
USDA relies on faulty logic to protest this reading, as “untenable” because it would “bar[]” USDA “from
approving a waiver under clause (ii) unless it has job availability information specific to the ABAWD population.”
USDA’s Opp’n at 24 (emphasis omitted). Not so: this reading would bar USDA from relying solely on measures so
general that the ABAWD unemployment rate is essentially not counted, without requiring USDA to rely solely on
ABAWD specific measures.

34
makes clear the provision is designed to address “extreme, dynamic circumstances,” such as “the

rapid disintegration of an economically and regionally important industry” or the “prolonged

impact of a natural disaster.” Id. at 66792. The Final Rule’s examples illustrate the kind of other

“data or evidence” USDA has in mind: “a state might provide unemployment data or other

evidence that is similar to the core standards except in that it covers a shorter duration because

the area’s economy suffered a rapid decline . . . that is not yet demonstrated by a full 12-month

or 24-month data period.” Id. This is the same data, over a “shorter duration,” required by the

“core standards.” The “exceptional circumstances” part of the Final Rule does not meaningfully

broaden those criteria for evaluating waivers, and thus cannot save USDA here.

Finally, recent legislative history is consistent with the conclusion that Congress intended

the phrase lack of “sufficient jobs” for ABAWDs to capture a broader set of circumstances than

those encompassed by the Rule. The House version of the 2018 Farm Bill proposed amending

§ 2015(o)(4)(A) to limit waivers to areas (1) with an unemployment rate of over 10%; (2)

designated as an LSA by DOL; (3) with a 24-month average unemployment rate at least 20%

higher than the national average and above 7%; or (4) in states that are in “an extended benefit

period” under federal law or “in which temporary or emergency unemployment compensation is

being provided under any Federal law.” H.R. 2, 115th Cong. § 4015(a)(1)(F) (June 21, 2018)

(Engrossed in House). In other words, the House bill would have eliminated two of the current

data sources available to states while retaining the criteria of LSA designation and eligibility for

federal extended unemployment benefits. For the criterion measuring a higher unemployment

rate than the national average, the House bill imposed a floor of an unemployment rate of 7%.

The Senate version of the Farm Bill contained no restrictions on waivers. The Conference

Committee rejected the House bill’s restrictions on waivers. In doing so, the Conference

35
Committee “acknowledge[d] that waivers from the ABAWD time limit are necessary in times of

recession and in areas with labor surpluses or higher rates of unemployment.” H. Rep. No. 115-

1072 (2018) (Conf. Rept.) at 616. Congress’s recent rejection of waiver criteria less stringent

than those ultimately adopted by USDA is probative of congressional intent to maintain

flexibility in the waiver system. Tellingly, even the House version of the bill did not rely solely

on unemployment rates, as USDA does in its Final Rule.

USDA dismisses this legislative history, citing cases like those discussed above in

rejecting the state plaintiffs’ legislative history argument about the discretionary exemptions.

See USDA’s Opp’n at 31–32. There, the Conference Committee Report, which adopted one of

two plausible readings of the statute, could not make the ambiguous statutory text clear or render

USDA’s plausible reading unreasonable. See supra Part III.A. Here, by contrast, the

Conference Committee Report and the history of the 2018 Farm Bill merely confirm the

conclusion that USDA’s reading is unreasonable and arbitrary and capricious in substance, a

conclusion that comport with the statute’s text and structure. Council for Urological Interests v.

Burwell, 790 F.3d 212, 222–24 (D.C. Cir. 2015) (rejecting challenger’s argument based on a

Conference Report that agency’s interpretation failed at Chevron step one but then remanding to

the agency at Chevron step two because the agency’s interpretation was unreasonable in light of

the Conference Report).

2. The Final Rule’s Redefinition of Waiver Area is Likely Arbitrary and
Capricious

Plaintiffs are likely to show that USDA’s decision to redefine area was arbitrary and

capricious in two respects. First, USDA’s state abuse justification for this aspect of the Final

Rule was wholly unsupported. Second, USDA responded inadequately to the considerable

evidence suggesting that the new definition of area was inappropriate.

36
a. State Abuse Rationale

USDA adopted a “strict definition of waiver ‘area’” because of its view, developed

through “operational experience,” that “[s]tates are grouping areas in such a way to maximize

waived areas rather than demonstrate high unemployment or a lack of sufficient jobs for

ABAWDs.” 84 Fed. Reg. at 66794. Yet, USDA “has provided no evidence” that such state

manipulation is “a real problem.” Nat’l Fuel Gas Supply Corp. v. FERC, 468 F.3d 831, 841

(D.C. Cir. 2006). “Professing that an” agency action “ameliorates a real . . . problem but then

citing no evidence demonstrating that there is in fact a[] . . . problem is not reasoned

decisionmaking.” Id. at 843 (citing State Farm, 463 U.S. at 42–43); see also id. at 841 (finding

an agency order that posited a solution to a problem but “provided no evidence of a real

problem” nor “a single example of abuse” to be arbitrary and capricious).

In USDA’s view, it was “enough under the APA” that the Final Rule “cited” “operational

experience” and “explained the precise nature of [the agency’s] concerns” through “examples.”

USDA’s Opp’n at 36. This correctly articulates but misapplies the standard here. USDA relies

on National Tour Brokers Association v. ICC, 671 F.2d 528 (D.C. Cir. 1982), which explained,

“[w]hile the practice is not without problems, an agency may rely on its ‘experience’ to provide

the necessary factual support” for a rule. Id. at 533. “A problem for the agency in relying on its

experience as factual support for its decision,” however, “is the necessity of adequately recording

and explaining that experience on the record.” Id. Part of recording and explaining experience,

and part of judicial review of an agency’s reliance on experience, is assuring that any agency

“findings” based on experience are “themselves . . . factually supported in the administrative

record.” Id. USDA flatly fails that test. Factual support does not necessarily need to come in

the form of empirical evidence or other concrete examples of abuse, as USDA recognizes.

USDA’s Opp’n at 35 (citing Stilwell v. Office of Thrift Supervision, 569 F.3d 514, 519 (D.C. Cir.
37
2009)). At the same time, factual support does need to be based in fact and “real” rather than

manufactured. Nat’l Fuel Gas Supply Corp., 468 F.3d at 841; see also Nat’l Tour Brokers, 671

F.2d at 533 (looking to the agency’s factual support to decide whether the agency’s decision

“was the rational result of reasoned analysis”).15

USDA attempted to support its view that states are manipulating waiver groupings with

three assertions. None holds up on examination. First, the Final Rule surmised that state waiver

manipulation is “one of the primary reasons why about half of the ABAWDs participating in

SNAP live in waived areas, despite current low unemployment levels across the majority of the

country.” 84 Fed. Reg. at 66794. As the state plaintiffs observe, this statistic is “unsurprising.”

State Pls. Mem. at 28. Of course, a substantial proportion of ABAWDs receiving benefits live in

waived areas: ABAWDs outside of waived areas who cannot meet the work requirements have

their benefits terminated rapidly, after just three months. This statistic could thus reflect the

difficulties ABAWDs face in finding work to maintain their benefits outside of waived areas.

Despite the existence of at least one readily apparent alternative account for this statistic, USDA

articulated zero reasons for its assumption that “about half” is a suspiciously high share of

ABAWDs to be living in waived areas. An unsupported assumption cannot sustain a key

justification for a rule. See Nat’l Tour Brokers, 671 F.2d at 534 (requiring lessons from agency

experience to have factual support).

15
USDA also relies on Pharm. Research & Mfrs. of Am. V. FTC (PhRMA), 44 F. Supp. 3d 95 (D.D.C. 2014)
(Howell, C.J.), for the notion that “[a]gency reliance is at its strongest when the agency invokes its experience not in
defense of an ‘objectively or scientifically determinable’ decision, but rather to explain the very need for the very
rule in question.” USDA’s Opp’n at 36 (quoting PhRMA, 44 F. Supp. 3d at 129). That distinction is a red herring,
as PhRMA applies the same standard as Nat’l Tour Brokers. In PhRMA, as here, the agency invoked its experience
to explain the need for the rule. See 44 F. Supp. 3d at 128–29; see also Pharm. Research & Mfrs., 790 F.3d at 208
(explaining the rule). In PhRMA, unlike here, the agency provided specific and substantial factual support for its
invocation of experience and “clearly and reasonably explained why it adopted the Rule.” Pharm. Research &
Mfrs., 790 F.3d at 208 (observing that, in explaining its experience, the agency “[s]pecifically” described 66 relevant
filings supporting the assertion about experience); see also PhRMA, 44 F. Supp. 3d at 129 n.15 (distinguishing cases
in which agencies did not “provide reasoning” and did not reasonably reject “the contrary evidence”).

38
Second, USDA saw as “manipulative,” 84 Fed. Reg. at 66793, state waiver applications

that “grouped nearly all contiguous counties in the State together while omitting a few counties

with relatively low unemployment,” id. at 66794. USDA’s assertion that these waivers reflect

state abuse is not a finding supported by facts. See Nat’l Fuel Gas Supply Corp., 468 F.3d at

843. As support for this perceived problem, the Final Rule states only that such waivers allow

states to “maximiz[e] waived areas,” 84 Fed. Reg. at 66794, but maximization is not self-

evidently manipulative.

In a belated attempt to better connect the dots between the facts and the assertion of state

abuse, USDA now offers the slightly more specific argument that such waiver requests reveal

that states “conspicuously left out counties that would have jeopardized the request,” USDA’s

Opp’n at 35 (citing 84 Fed. Reg. at 986), while keeping in areas with low unemployment to

obtain waivers for the latter areas “simply by grouping them with” areas of high unemployment,

id. at 37.16 Grouping contiguous counties with relatively high unemployment while omitting

contiguous counties with relatively low unemployment is entirely consistent with good faith

efforts by states to target waiver requests to areas that lack sufficient jobs and to apply the work

requirements in areas with enough jobs. The Final Rule does nothing to rebut this common

16
USDA cites the Proposed Rule for this jeopardy point, see USDA’s Opp’n at 35 (citing Rule 84 Fed. Reg.
at 986), but the Proposed Rule simply states: “Some States have grouped nearly all contiguous counties in the State
together while omitting a few counties with relatively low unemployment in order to maximize the waived areas in
the State.” 84 Fed. Reg. at 986. That statement does nothing to provide concrete, factual examples that would fill in
the gaps of the Final Rule.
In opposing the plaintiffs’ motions, USDA also points to a 2016 USDA Office of the Inspector General
(OIG) report stating that “officials in three States told us they specifically requested ABAWD time limit waivers in
as many parts of the State as possible to minimize the areas where they needed to track the ABAWD time limits.”
OIG Report at 5; see also USDA’s Opp’n at 35. The OIG Report is a post hoc rationale not mentioned in the Final
Rule. See, e.g., SEC v. Chenery Corp., 318 U.S. 80, 94–95 (1943) (agency actions cannot be upheld based on after-
the-fact reasoning). Nor is this admission indicative of abuse. The OIG finding came in the context of the OIG’s
discussion of the difficulty and complexity of implementing PRWORA’s ABAWD time-tracking provisions. See
OIG Report at 4–8 (findings and recommendations related to the difficulty of implementing the ABAWD
provisions). The OIG’s report is consistent with the idea that the state grouping behavior USDA deemed
manipulative is a feature of the statutory design rather than a product of state abuse.

39
sense assessment of state behavior. A single comment referenced in the Final Rule speculates

that states “will seek to maximize any discretion in order to receive and use as much Federal

money as possible.” 84 Fed. Reg. at 66973. Speculation cannot substantiate the USDA’s

assumption of bad faith by states here. See, e.g., Nat. Res. Def. Council, Inc. v. EPA, 859 F.2d

156, 210 (D.C. Cir. 1988) (holding that agency speculation is “not adequate grounds upon which

to sustain an agency’s action”).

Third, USDA saw as “manipulative,” id. at 66793, state waiver applications that

“grouped certain towns together that share the same economic region while omitting others with

relatively low unemployment from the group,” id. at 66794. The Final Rule argues that this

allowed states to obtain waivers for jurisdictions “without consideration for job availability” in

nearby jurisdictions. Id. at 66796. For example, the Final Rule said, “if a State has the choice to

obtain waivers . . . for individual counties, . . . the State could waive the county without

consideration for the job availability in its surrounding LMA.” Id.17 Comments suggested

motivations other than state manipulation for such a waiver request. For example,

“[c]ommenters argued that States have the best understanding of the regional patterns in their

labor markets, local commuting burdens, and other local nuances specific to ABAWDs.” Id. at

17
Although the Final Rule provided only this hypothetical example, USDA now points to two real examples
that do not help its case. USDA says (1) that a resident of the neighborhood of Takoma in the District is subject to a
waiver while a resident of Takoma Park, Maryland — “immediately across the border (and perhaps even the street)
— is not,” USDA’s Opp’n at 38 (citing McConnell Decl., Ex. 14 (Maryland Waiver) and then citing McConnell
Decl., Ex. 15 (D.C. Waiver); and (2) that a resident of Manhattanville in New York City is subject to a waiver even
though residents of a community district blocks away might not be, id. (citing McConnell Decl., Exs. 9 & 16 (New
York Waivers)). The first example provides no support for state waiver abuse because the District and Maryland, as
different states, apply separately for waivers, based on analysis of data particular to their own jurisdictions. The
second example is consistent with state expertise, not state abuse, as the waiver applications provided by USDA
show that New York requested waivers for all of “Upper Manhattan,” including Manhattanville, but not elsewhere
in New York City, such as the community district just beneath Manhattanville, which covers the neighborhood
around Columbia University and the Upper West Side. See McConnell Decl., Ex. 9 at 5, Map 2; see also Steven
Banks Decl. (“Banks Decl.”) ¶ 14, ECF No. 3-1 (explaining that Upper Manhattan is “by many measures” “distinct
and separate from the rest of Manhattan”).

40
66793. Once again, USDA’s failure to engage with any of these readily apparent alternative

explanations renders its assumption of state abuse inadequate.

USDA and the state amici now point to waivers that will be in the administrative record,

which has not yet been filed on the docket or produced to plaintiffs, allegedly “exhibiting” this

“pattern of gerrymandering,” USDA’s Opp’n at 35, but these examples, which USDA’s brief

makes no attempt to analyze, do not advance USDA’s defense, see McConnell Decl., Exs. 4–12,

ECF No. 26-1 (providing example waiver requests). The state amici highlight a waiver

application from California, which covered 52 counties that, together, had an average

unemployment rate of 5.5%, “exactly” 20% above the national average unemployment rate at the

time. State Amici Br. at 6 (discussing McConnell Decl., Ex. 4, ECF No. 26-1). The amici see

three problems with this application: first, that the unemployment rate for the whole region is

precisely at the threshold needed to get a waiver; second, that the waiver “region spans over 800

miles;” and, third, that some of the counties included in the waiver area had unemployment rates

below 5.5%. See id. at 6–7.

At the outset, in light of USDA’s insensitivity to the concerns of states, like the District,

that LMAs are too large to be appropriate waiver areas, the irony cannot be missed that large

waiver areas are cited as cause for suspicion. See Rough Hr’g Tr. at 80:25 (the Court noting this

in question to USDA). Further, responding to state amici’s argument at the hearing, the state

plaintiffs argued that amici’s analysis was overly simplistic. States consider more than just

unemployment rate in designing waiver area, the state plaintiffs pointed out, including what

areas are “appropriately connected to each other for purposes of considering the inquiry of jobs

for ABAWDs, whether [areas] are connected by affordable transportation or not, whether they

share similar employment opportunities or not, and even whether the areas for which waivers are

41
not requested . . . had sufficient jobs for SNAP recipients living there” as well as “for the

additional SNAP recipients in the waived areas if those waivers were not in place.” Id. at 41:17–

25. After all, USDA’s current regulation, which permits States “to define areas to be covered by

waivers” and instructs them to provide “data and analyses that correspond to the defined area” or

“as closely to the area as possible,” 7 C.F.R. §273.23(f)(6), gives states this responsibility

because States are more familiar with the reality of employment challenges confronting

AWAWBs in particular areas.

More broadly, the state plaintiffs argued at the hearing that attempts to maximize waived

areas are not in themselves manipulative. See id. at 49:24–50:14. To the contrary, plaintiff

states viewed maximization within the waiver rules laid down by USDA as responsible state

policy, preferring that funds appropriated for SNAP go directly to feed the needy rather than to

bloat state agencies that enforce work requirements in unwaived areas. Id. at 50:10–14; cf.

House Amicus Br. at 8 (stating that, during debate about the House version of the 2018 Farm Bill

curtailing waivers, “[c]oncerns were raised that, while millions of people under the House bill

would be losing their benefits, much of this money would be used to pay for a new SNAP

bureaucracy to enforce work requirements” (internal quotation marks and alteration omitted)).

Finally, the state plaintiffs argued that, if USDA had a different view of maximization, the

agency could have “use[d] a scalpel to look at . . . particular waiver requests carefully,” seeking

“more information” about or “deny[ing]” applications that USDA viewed as not “supported by

appropriate evidence.” Rough Hr’g Tr. at 42:13–16. Instead, in state plaintiffs’ view, USDA

“thr[ew] out the baby with the bath water” in radically redefining waiver areas. Id. at 42:17.

If USDA did indeed think that states were manipulating the waiver process, then asking

states why they were grouping as they were, requesting that states resubmit waiver applications

42
that USDA viewed as inappropriate, or outright denying such waiver applications, would have

been a sensible course. USDA, it appears, did none of that.18 That history is another sign that

USDA’s new policy eliminating state discretion in designing waiver areas is a solution in search

of a problem.

b. Problems with LMAs

The plaintiffs also say that problems with USDA’s new definition of “area in which the

individuals reside” rendered it arbitrary and capricious. 7 U.S.C. § 2015(o)(4)(A). On the

current record, the plaintiffs are likely to succeed on this argument because considerable

evidence before USDA posited that LMA was an inappropriate definition of area, and USDA’s

treatment of this evidence was “wanting.” Gresham, 950 F.3d at 102. That evidence showed

“LMAs are based on commuting patterns of the general workforce” and thus cannot capture

available employment for “low-income, low-skilled ABAWDs who lack affordable

transportation options.” 84 Fed. Reg. at 66793.

The mismatch between available employment for ABAWDs and available employment

across an LMA is starkest in large urban areas. Take the District, an example before the agency.

See id. at 66796 (alluding to the District). The LMA that includes the District comprises the

District, sixteen counties in Virginia, five counties in Maryland, and one county in West

Virginia. See Bolen Decl. ¶ 29. As a result, District ABAWDs would be subject to the work

requirement if the unemployment rate is low in suburbs and exurbs of the District as far out as

18
The Final Rule pointed out that USDA “attempted” in 2016 “to clarify its intention that areas be
economically tied through policy guidance.” 84 Fed. Reg. at 66794 & n.8 (citing Guide to Supporting Requests to
Waive the Time Limit for Able-Bodied Adults without Dependents). That guidance states: “In order to be
combined, the areas must be contiguous or considered parts of the same economic region. For example, two or more
contiguous counties could be grouped together in order to consider their aggregate average unemployment rate.”
McConnell Decl., Ex. 3, Guide to Supporting Requests to Waive the Time Limit for Able-Bodied Adults without
Dependents at 10, ECF No. 26-1. Since USDA did not then begin denying or questioning waiver areas that it
viewed as inconsistent with this guidance, unsurprisingly, this guidance did not “prevent[] States from strategically
using grouping to maximize waived areas.” 84 Fed. Reg. at 66794.

43
West Virginia, although these out-of-state counties are inaccessible by public transportation from

the District and although unemployment rates in parts of the District are as high as 11.6%.

Zeilinger Decl. ¶ 11. Another example is New York City, which is part of an LMA including 25

counties in three states — New York, New Jersey, and Pennsylvania — that extends more than

100 miles from the City. Banks Decl. ¶ 22–24. New York City’s current waiver encompasses

the boroughs of the Bronx, Brooklyn, and Staten Island, along with four community districts in

Manhattan, and three community districts in Queens. Id. ¶¶ 13–15. About 64,500 ABAWDs

who live in these areas, see id. ¶ 16, would be subject to the work requirement if the

unemployment rate is sufficiently low across the tristate LMA, regardless of the barriers to

employment within New York City. Indeed, the plaintiffs anticipate that both the District and

New York City would lose existing waivers under the Final Rule, given the relatively low

unemployment rates in the LMAs as a whole. See Bolen Decl. ¶ 29; Banks Decl. ¶ 24; State

Pls.’ Mot., Laura Zeilinger Decl. (“Zeilinger Decl.”) ¶¶ 8, 11–12, ECF No. 3-19.

The Final Rule alludes to the fact that commenters raised examples like the District and

New York City, see 84 Fed. Reg. at 66794 (acknowledging that “[c]ommenters gave examples in

which some LMAs are too big to properly define commuting patterns for ABAWDs”), but

USDA brushed off those examples, stating that the agency was “not compelled by the

commenters’ suggestions . . . that LMAs do not account for specific ABAWD commuting

patterns and other factors specific to ABAWDs.” Id. at 66793. Rejecting alternatives to LMAs

proposed by commenters, USDA concluded that “LMAs remain the best available and most

appropriate delineation to address the issue of grouping, as there are no Federally-designated

areas that specifically assess commuting patterns and other related economic factors for

ABAWDs.” Id. Faced with evidence that, under the LMA definition, ABAWDs, especially

44
those in large urban areas, would be subject to the work requirement despite lacking access to

sufficient jobs, USDA did no more than state that that this evidence was rejected. “Nodding to

concerns raised by commenters only to dismiss them in a conclusory manner is not a hallmark of

reasoned decisionmaking.” Gresham, 950 F.3d at 103 (citing Am. Wild Horse Pres. Campaign,

873 F.3d at 932 and then citing Getty v. Fed. Savs. & Loan Ins. Corp., 805 F.2d 1050, 1055

(D.C. Cir. 1986)).

Moreover, as justified, USDA’s preference for LMAs is arbitrary. Elsewhere in the Final

Rule, USDA explicitly “recognize[d] that ABAWDs may face barriers to employment” not faced

by the general population. 84 Fed. Reg. at 66787. The LMA section of the Final Rule implicitly

recognizes the same reality and acknowledges that USDA has discretion to adopt an approach

that accounts for that reality. USDA “note[d] that if in the future a more robust delineation” —

meaning, a statistical area measuring “commuting patterns and other related economic factors for

ABAWDs” — “becomes available from a Federal source,” USDA would consider adopting that

definition of area. Id. at 66793. USDA thus chose to exercise its discretion in a way that

admittedly ignored critical factors about ABAWDs in service of two goals — ending grouping

and relying only on federal government data —unsupported by the record before the agency and

by the agency’s explanation. As already discussed, no evidence supported USDA’s assertions

about state abuse, and nothing required the agency to rely on already available federal data

sources. Those reasons do not stick here either.

* * *

USDA’s conclusory reasoning and summary dismissal of extensive evidence submitted

by commenters is particularly troubling in light of the reliance interests involved. Encino

Motorcars, LLC, 136 S. Ct. at 2126 (requiring “reasoned explanation” given the “significant

45
reliance interests involved”).19 The Final Rule, USDA calculated, would “newly subject an

estimated 1,087,000 ABAWDs to the time limit,” and “the vast majority, approximately

688,000” people, would lose their benefits for failure to find work. 84 Fed. Reg. at 66809.

These almost 700,000 individuals have depended on SNAP benefits to avoid hunger. The states

where these individuals live have depended on SNAP and the concomitant federal funding to

ensure that the basic needs of their residents are met. Especially when a change in policy risks

disrupting interests as fundamental as access to nutrition, the law demands that the agency take

care to explain and justify the change.

C. Absent Preliminary Relief, Plaintiffs will Suffer Immediate,
Irreparable Harm

The plaintiffs have shown that they “would suffer irreparable injury if the injunction were

not granted.” Chaplaincy of Full Gospel Churches v. England, 454 F.3d 290, 297 (D.C. Cir.

2006). The analysis starts with the state plaintiffs, then turns to BFC and the individual

plaintiffs.

1. State Plaintiffs

The state plaintiffs assert that they will suffer irreparable harm in the form of significant

administrative burdens and costs, including staffing and training costs, notification costs, and

costs from expanding employment and training programs. See State Pls.’ Mem. at 38–43. The

Rule would trigger these burdens by restricting states’ flexibility in using waivers. The Center

for Budget and Policy Priorities’ declaration predicts that 65% of currently waived counties and

19
Encino Motorcars is one of several cases addressing the obligations to explain that attach to agency actions
departing from “longstanding policies [that] may have ‘engendered serious reliance interests.’” 136 S. Ct. at 2126
(quoting Fox Television., 556 U.S. at 515). Some read these cases as imposing requirements on agencies beyond
those imposed where reliance interests are not found, see State Pls.’ Mem. 35–38 (arguing that a heightened
standard applies in this case), but a heightened standard has not been applied here. Instead, simply put, USDA failed
to “articulate a satisfactory explanation” for its choice under the typical arbitrary and capricious standard. State
Farm, 463 U.S. at 43.

46
92% of currently waived towns will lose their waivers. Bolen Decl. ¶ 5. State plaintiffs predict

that areas of each of their states will lose existing waivers as a result of the Final Rule.20 The

District, for example, anticipates losing the statewide waiver held since PRWORA became

effective in 1997. See Zeilinger Decl. ¶ 10. New York state “estimates that as a result of the

Final Rule, only one county in the State consisting of seven ABAWDs, will qualify for” a

waiver. Jeffrey Gaskell Decl. (“Gaskell Decl.”), ¶ 18 ECF No. 3-8. Currently, 113,445 New

York ABAWDs live in waived areas. Id. Across the plaintiff states, and indeed the country, loss

of waivers will mean applying the work requirements to hundreds of thousands of additional

ABAWDs, even where the lack of sufficient jobs for this group may persist and the states,

though scrambling now, currently lack the infrastructure necessary to provide alternative

mechanisms to meet the work requirements, such as employment and training programs, or

process status exemptions.

As a threshold matter, USDA claims that “the costs of compliance with a regulatory

scheme do not constitute irreparable injury.” USDA’s Opp’n at 58 (quoting Nat’l Med. Care,

Inc. v. Shalala, No. 95-0860 (WBB), 1995 WL 465650, at *3 (D.D.C. June 6, 1995)).21

20
Banks Decl. ¶ 24 (predicting that New York City will lose its waiver); Tikki Brown Decl. (“Brown Decl.”)
¶ 5, ECF No. 3-3 (anticipating that only 3 of Minnesota’s 26 currently waived counties will continue to qualify for a
waiver); Buhrig Decl. ¶ 8 (stating that only two of Pennsylvania’s 60 currently waived counties will qualify under
the Final Rule); Fernández Decl. ¶ 28 (stating that 34 counties in California will likely lose their waivers); Steve
Fisher Decl. (“Fisher Decl.”) ¶ 7, ECF No. 3-6 (stating that three Nevada counties will lose their waivers); Holly
Freishtat Decl. (“Freishtat Decl.”) ¶ 10, ECF No. 3-7 (stating that the Final Rule “effectively disqualifies the
Baltimore Region from qualifying for an area waiver”); Deirdre Gifford Decl. (“Gifford Decl.”) ¶ 14, ECF No. 3-9
(calculating that five Connecticut cities will lose their waivers); Daniel Haun Decl. (“Haun Decl.”) ¶ 13, ECF No. 3-
11 (stating that “only 6 counties” in Oregon “may be eligible for a waiver,” while 31 counties currently have
waivers); Brittany Mangini Decl. (“Mangini Decl.”) ¶ 7, ECF No. 3-14 (stating that Massachusetts currently has 83
waived towns and cities and none would be eligible for waivers because none “are located in Labor Market Areas
that meet the six percent floor”); Elisa Neira Decl. (“Neira Decl.”) ¶¶ 8–9, ECF No. 3-16 (stating that New Jersey
currently has a waiver in fifteen counties and predicting that thousands will lose their benefits in the state); S. Duke
Storen Decl. (“Storen Decl.”) ¶ 10, ECF No. 3-17 (stating that Virginia has a waiver for 55 localities, but that at
most 5 of those would qualify for waivers under the Final Rule); Dawn Sweeny Decl. (“Sweeney Decl.”) ¶¶ 8,11
(stating that at most 11 counties in Michigan would qualify for waivers under the Final Rule, compared to 77
counties that would have qualified under the 2001 regulation).
21
USDA cites Nat’l Med. Care, Inc. v. Shalala, No. 95-0860 (WBB), 1995 WL 465650, at *3 (D.D.C. June
6, 1995), which relies on an outdated, out-of-circuit case, A.O. Smith Corp v. FTC, 530 F.2d 515, 527 (3rd Cir.

47
Although as a “general rule[,] . . . economic harm does not constitute irreparable injury,” Davis,

571 F.3d at 1295, economic loss caused by federal agency action is an exception: typical

economic harm is not irreparable because it is generally recoverable as monetary damages, see

id. (citing Taylor v. Resolution Trust Corp., 56 F.3d 1497, 1507 (D.C. Cir. 1995)), whereas

economic injury caused by federal agency action is unrecoverable because the APA’s waiver of

sovereign immunity does not extend to damages claims, see 5 U.S.C. § 702 (allowing relief

“other than money damages”); see California v. Azar, 911 F.3d 558, 581 (9th Cir. 2018) (stating

that economic harm caused by federal agency action “is irreparable . . . because the states will

not be able to recover monetary damages”); Chamber of Commerce v. Edmondson, 594 F.3d

742, 770–71 (10th Cir. 2010) (“Imposition of monetary damages that cannot later be recovered

for reasons such as sovereign immunity constitutes irreparable injury.”); Iowa Utilities Bd. v.

FCC, 109 F.3d 418, 426 (8th Cir. 1996) (“The threat of unrecoverable economic loss, however,

does qualify as irreparable harm.”); Pennsylvania v. Trump, 930 F.3d at 574 (“Because the States

cannot collect money damages under the APA, the States will suffer irreparable harm if the

Rules are enforced.” (citations and footnote omitted)).22

USDA presses that, “even if the court considers these costs,” they cannot constitute

irreparable harm because the training harms are “self-inflicted” and the other harms are “neither

1976). See Am. Hosp. v. Harris, 625 F. 2d 1328, 1331 (7th Cir. 1980) (relying on A.O. Smith as well). Most
problematic for USDA is the fact that the Third Circuit recently rejected the principal USDA advances, holding in
no uncertain terms that “financial consequences” resulting from federal regulation are irreparable harm.
Pennsylvania v. Trump, 930 F.3d 543, 574 (3d Cir. 2019), cert. granted sub nom. Little Sisters of Poor v.
Pennsylvania, No. 19-431, 2020 WL 254158 (U.S. Jan. 17, 2020), and cert. granted sub nom. Trump v.
Pennsylvania, No. 19-454, 2020 WL 254168 (U.S. Jan. 17, 2020).
22
The D.C. Circuit has not squarely addressed the irreparability of economic losses caused by federal agency
action but has repeatedly affirmed that economic injury is “irreparable where no ‘adequate compensatory or other
corrective relief will be available at a later date, in the ordinary course of litigation.’” Mexichem Specialty Resins,
Inc. v. EPA, 787 F.3d 544, 555 (D.C. Cir. 2015) (quoting Wis. Gas Co., 758 F. 2d at 674)).

48
‘certain’ nor ‘great.’” USDA’s Opp’n at 58 (quoting Wis. Gas Co., 758 F.2d at 674). These

characterizations are inaccurate.

To begin, the state plaintiffs claim that implementing the Rule will require allocating

resources to train new or existing staff to implement the ABAWD time limits in previously

waived areas. See Brown Decl. ¶ 10 (asserting that staff time and training will cost Minnesota at

least $500,000); Fernández Decl. ¶¶ 49–50 (averring that the Rule will require California “to

provide extensive training to our forty counties on a very tight timeline” and describing the

extensive training previously provided to the six counties in California “currently implementing

the time limit”); Gifford Decl. ¶ 21 (stating that “all eligibility staff [in Connecticut] will need to

be retrained in ABAWD case administration, given that the new rule greatly expands the number

of statewide ABAWD cases”); Haun Decl. ¶¶ 18–19 (declaring that “[t]en (10) months of

onboarding activities will be necessary in a matter of three months . . . to ensure those areas [of

Oregon] are ready to serve ABAWDs”); Storen Decl. ¶ 23 (explaining that Virginia will need to

“train staff from 55 local departments of social services who now need to implement this

policy”); Zeilinger Decl. ¶ 13 (anticipating hiring new staff because the District does not

currently administer the work requirements).

USDA argues that this harm is self-inflicted because states have always been required to

“track whether ABAWDs subject to the time limit are or are not meeting the work

requirements,” even in waived areas. USDA’s Opp’n at 59 (quoting McConnell Decl., Ex. 30,

FNS, Memorandum to All Regional Directors of SNAP (May 25, 2018) (“2018 FNS Memo”) at

3). True, USDA had advised states, at least since 2018, to “measure the 3-year period and track

ABAWDs on a continuous basis.” 2018 FNS Memo at 3. In USDA’s own words, however, that

was “so that [the state] will be ready to transition off-of [sic] the waiver when it expires and

49
reintroduce the time limit,” 2018 FNS Memo at 3. States with no reason to anticipate the demise

of a waiver under the 2001 regulation would have no reason to implement the time limits in

waived areas. Indeed, to have done so, as the states put it, “would have been a misuse of

government resources.” State Pls.’ Reply at 27. In other words, states were not previously

subject to a blanket requirement to track in waived areas, so the complained-of training costs

would “directly result from” the regulatory changes announced in the Rule, Wis. Gas Co., 758

F.2d at 674, not from state “neglect[],” USDA’s Opp’n at 58.

In any event, the case cited by USDA for the idea that self-inflicted harm cannot be

irreparable is not talking about the kind of harm USDA calls self-inflicted. See USDA’s Opp’n

at 58–59 (relying on Barton v. DC, 131 F. Supp. 2d 236, 247 (D.D.C. 2001)). Rather, self-

inflicted generally means curable by the moving party without an injunction. So, in Barton, a

business owner could not make the requisite showing of harm threatening the business’s

existence because the business owner had ways to avoid destruction of the business short of an

injunction. Barton, 131 F. Supp. 2d at 247.23 This conception of self-inflicted irreparable harm

is echoed in standing doctrine, where plaintiffs “cannot manufacture standing merely by

inflicting harm on themselves.” Clapper v. Amnesty Int’l USA, 568 U.S. 398, 416 (2013); see

also California, 911 F.3d at 573–74 (analyzing the argument that states’ asserted harms

stemming from a federal administrative action were “self-inflicted” by the states as a standing

issue); Texas v. United States, 809 F.3d 134, 159 (5th Cir. 2015) (same). Here, USDA has not

23
Similarly, in San Francisco Real Estate Investors v. Real Estate Investment Trust, 692 F.2d 814 (1st Cir.
1982), relied on by Barton, the injury asserted by plaintiff’s motion for preliminary injunction was market
uncertainty stemming from the plaintiff’s filing of the lawsuit. Id. at 818; see also Fish v. Kobach, 840 F.3d 710,
753 (10th Cir. 2016) (“[T]ypically, a finding of self-inflicted harm results from either misconduct or something akin
to entering a freely negotiated contractual arrangement.”).

50
alleged that states administered their SNAP programs to manufacture injuries, nor has the USDA

contended that the states could now avoid the training costs absent equitable relief.24

Next, the state plaintiffs assert that implementing the time limits in previously waived

areas requires states to give notice to ABAWDs about the time limits, the work requirements, the

available exemptions, and about any loss of benefits. Developing these notices, which must

comport with complex legal requirements, see, e.g., 7 C.F.R. § 273.13 (Notice of Adverse

Action), is burdensome, especially for states like the District that do not currently administer the

time limits, see Zeilinger Decl. ¶ 13 (“The District would also need to develop and disseminate

notices to impacted and potentially impacted households advising of their ABAWD status, rights

and responsibilities with respect to the new work requirements, reporting obligations, and

advanced notice of any adverse actions taken on their SNAP cases as a result of the Final Rule’s

implementation.”). Even states that have notices drawn up for use in areas currently subject to

the time limits, aver that “significant amendments” to those sample notices “will be required to

address issues specific to the expedited timeframe for implementation and the anticipated

elimination of the State’s carryover of discretionary exemptions.” Fernández Decl. ¶ 52

(California).

Once the notices are drafted, states will need to assess which of the thousands of

ABAWDs in the state need notice and will need to “execute the mailings.” Gifford Decl. ¶ 20

(Connecticut); see also Brown Decl. ¶ 10 (“Eligibility workers [in Minnesota] will have limited

time to reassess affected cases and send proper 10-day notices to ABAWDs whose SNAP

benefits must stop after implementation of the Rule.”); Mangini Decl. ¶ 10 (describing the “case

24
To the extent USDA argues that the states’ proof does not establish that the training costs are “certain,” see
USDA’s Opp’n at 59–61 (quoting Wis. Gas Co., 758 F.2d at 674), the declarations establish with the requisite
certainty that implementation of the Rule will cause them to incur the staffing and training costs already described.

51
review process to identify SNAP recipients who may become newly subject to ABAWD work

requirements under the Rule” in Massachusetts); Gaskell Decl. ¶ 19 (noting that districts in New

York state are “review[ing] their current caseload to identify individuals who will be subject to

the ABAWD requirements” through a “labor intensive, manual process” that “requires running

caseload reports and manually reviewing each case record and sending manual notices”). In

some states, completing this notice process will entail “reprogram[ming]” computer systems that

track SNAP recipients. Neira Decl. ¶ 10 (New Jersey); see also Fernández Decl. ¶ 52 (declaring

that California’s “three eligibility Information Technology (IT) systems, particularly noticing

and client reporting mechanisms, will have to be newly implemented or updated and prioritized

over existing high priority projects.”); Gifford Decl. ¶ 20 (stating that Connecticut will need to

“make system coding changes to execute the mailings”).

States estimate that the total monetary costs of noticing ABAWDs will be in the hundreds

of thousands of dollars. See id. (predicting that Connecticut’s “[a]dministrative costs will

include a minimum of $207,500 just for the costs of direct communications to all ABAWDs,

plus costs to develop notices and make system coding changes to execute the mailings);

Zeilinger ¶ 13 (averring that the District will “incur hundreds of thousands of dollars in

additional costs each year that would detract from other elements of the District’s SNAP and

potentially other programs”); Buhrig Decl. ¶ 13 (estimating that Pennsylvania’s total cost, for

both notice and administration, will be “$2,250,000, plus a one-time cost of $56,050”).

At the hearing, the state plaintiffs also emphasized that they were preparing to “add staff”

to “supply denial hearings” to beneficiaries who challenge the termination of their benefits.

Rough Hr’g Tr. at 33:14–15; see also id. at 56:24–57:1 (explaining that “people that lose benefits

are entitled to administrative hearings under the Due Process Clause”). For example,

52
Connecticut, like other states, “anticipate[s] a significant rise in the number of hearings

requested,” and plans to hire “additional hearing officers,” who, in that state, have “an average

annual salary of approximately $131,500.” Gifford Decl. ¶ 20; see also Buhrig Decl. ¶ 18 (“An

increase in appeals for case closures will add additional administrative costs.”); Fernández Decl.

¶ 72 (stating that the Final Rule will “create a substantial burden on [California’s State Hearings

Division] and will lead to delays in hearings, which violate the rules of SNAP and the many

other programs for which SHD provides administrative review”); cf. Sweeney Decl. ¶ 17

(expressing “concern[] about the substantial resources that would be needed to defend against

legal challenges from ABAWDs who lose access to SNAP benefits”).

USDA argues that these allegations of harm are too “conclusory” to be certain, USDA’s

Opp’n at 59, but USDA’s position is belied by the Final Rule itself. There, USDA

acknowledged and quantified this very burden. USDA estimated that noticing the 688,000

ABAWDs expected to lose their benefits would take “45,867 hours” and that “[t]he total start up

burden for State agencies will result in 84,463 burden hours.” 84 Fed. Reg. at 66809. So, USDA

has actually “done much of the legwork in establishing,” at least in broad strokes, the certainty,

as well as the magnitude of this harm. Massachusetts v. U.S. Dep’t of Health & Human Servs.,

923 F.3d 209, 224–25 (1st Cir. 2019) (relying on Regulatory Impact Analysis done

[Text truncated at 120,000 characters. The full text is on the page linked above.]

---

Source: Frix Law Library, https://www.frixlaw.com/law-library/cases/4516443. Public record. Not legal advice.
