New York v. United States Department of Homeland Security, Make the Road

19-3591, 19-3595
New York v. United States Department of Homeland Security, Make the Road New York v. Cuccinelli

UNITED STATES COURT OF APPEALS
FOR THE SECOND CIRCUIT

August Term, 2019

Argued: March 2, 2020 Decided: August 4, 2020

Docket Nos. 19-3591, 19-3595

STATE OF NEW YORK, CITY OF NEW YORK, STATE OF CONNECTICUT, STATE OF
VERMONT,

Plaintiffs-Appellees,

— v. —

UNITED STATES DEPARTMENT OF HOMELAND SECURITY, SECRETARY CHAD F. WOLF,
IN HIS OFFICIAL CAPACITY AS ACTING SECRETARY OF THE UNITED STATES
DEPARTMENT OF HOMELAND SECURITY, UNITED STATES CITIZENSHIP AND
IMMIGRATION SERVICES, DIRECTOR KENNETH T. CUCCINELLI II, IN HIS OFFICIAL
CAPACITY AS ACTING DIRECTOR OF UNITED STATES CITIZENSHIP AND IMMIGRATION
SERVICES, UNITED STATES OF AMERICA,

Defendants-Appellants.*

*
The Clerk of the Court is respectfully directed to amend the caption as set forth
above.

1
 MAKE THE ROAD NEW YORK, AFRICAN SERVICES COMMITTEE, ASIAN AMERICAN
FEDERATION, CATHOLIC CHARITIES COMMUNITY SERVICES, (ARCHDIOCESE OF NEW
YORK), CATHOLIC LEGAL IMMIGRATION NETWORK, INC.,

Plaintiffs-Appellees,

— v. —

KENNETH T. CUCCINELLI, IN HIS OFFICIAL CAPACITY AS ACTING DIRECTOR OF UNITED
STATES CITIZENSHIP AND IMMIGRATION SERVICES, UNITED STATES CITIZENSHIP AND
IMMIGRATION SERVICES, CHAD F. WOLF, IN HIS OFFICIAL CAPACITY AS ACTING
SECRETARY OF HOMELAND SECURITY, UNITED STATES DEPARTMENT OF HOMELAND
SECURITY,

Defendants-Appellants.

B e f o r e:

LEVAL, HALL, and LYNCH, Circuit Judges.

The Department of Homeland Security appeals from two orders of the
United States District Court for the Southern District of New York (Daniels, J.)
granting motions for preliminary injunctions in these cases. Two sets of Plaintiffs-
Appellees – one a group of state and local governments and the other a group of
non-profit organizations – filed separate suits under the Administrative
Procedure Act, both challenging the validity of a Department of Homeland
Security rule interpreting 8 U.S.C. § 1182(a)(4). This statutory provision renders
inadmissible to the United States any non-citizen deemed likely to become a
public charge. The district court concluded that Plaintiffs-Appellees

2
demonstrated a likelihood of success on the merits of their claims that the rule is
contrary to the Immigration and Nationality Act and that it is arbitrary and
capricious. After finding that the other preliminary injunction factors also
weighed in favor of granting relief, the district court entered orders in both cases
to enjoin implementation of the rule nationwide. We agree with the district court
that a preliminary injunction is warranted, but modify the scope of the
injunctions to cover only the states of New York, Connecticut, and Vermont. The
orders of the district court are thus AFFIRMED AS MODIFIED.

JUDITH N. VALE, Senior Assistant Solicitor General, State of
New York, New York, NY (Letitia James, Attorney
General, Barbara D. Underwood, Solicitor General,
Steven C. Wu, Deputy Solicitor General, Matthew
Colangelo, Chief Counsel for Federal Initiatives, Elena
Goldstein, Deputy Bureau Chief, Civil Rights, Ming-Qi
Chu, Section Chief, Labor Bureau, State of New York,
New York, NY, William Tong, Attorney General, State
of Connecticut, Hartford, CT, Thomas J. Donovan, Jr.,
Attorney General, State of Vermont, Montpelier, VT,
James E. Johnson, Corporation Counsel, City of New
York, New York, NY, on the brief), for Plaintiffs-Appellees
State of New York, City of New York, State of Connecticut,
State of Vermont.

JONATHAN H. HURWITZ, Paul, Weiss, Rifkind, Wharton &
Garrison LLP, New York, NY (Andrew J. Ehrlich, Elana
R. Beale, Robert J. O’Loughlin, Daniel S. Sinnreich, Paul,
Weiss, Rifkind, Wharton & Garrison LLP, New York,
NY, Ghita R. Schwarz, Brittany Thomas, Baher A.
Azmy, Center for Constitutional Rights, New York, NY,
Susan E. Welber, Kathleen Kelleher, Susan Cameron,
Hasan Shafiqullah, The Legal Aid Society of New York,
New York, NY, on the brief), for Plaintiffs-Appellees Make
the Road New York, African Services Committee, Asian

3
 American Federation, Catholic Charities Community
Services, (Archdiocese of New York), Catholic Legal
Immigration Network, Inc.

GERARD SINZDAK, Appellate Staff Attorney, Civil Division,
United States Department of Justice, Washington, DC
(Joseph H. Hunt, Assistant Attorney General, Civil
Division, United States Department of Justice,
Washington, DC, Geoffrey S. Berman, United States
Attorney for the Southern District of New York, New
York, NY, Daniel Tenny, Joshua Dos Santos, Appellate
Staff Attorneys, Civil Division, United States
Department of Justice, Washington, DC on the brief), for
Defendants-Appellants United States Department of
Homeland Security, Acting Secretary Chad F. Wolf, United
States Citizenship and Immigration Services, Acting Director
Kenneth T. Cuccinelli, United States of America.

WILLIAM E. HAVEMANN, Office of General Counsel, United
States House of Representatives, Washington, DC
(Douglas N. Letter, General Counsel, Todd B. Tatelman,
Principal Deputy General Counsel, Megan Barbero,
Josephine Morse, Adam A. Grogg, Office of General
Counsel, United States House of Representatives,
Washington, DC, Robert M. Loeb, Thomas M. Bondy,
Peter E. Davis, Orrick, Herrington & Sutcliffe LLP,
Washington, DC, Rene Kathawala, Orrick, Herrington &
Sutcliffe LLP, New York, NY, on the brief), for Amicus
Curiae United States House of Representatives, in
support of Plaintiffs-Appellees.

Additional amici curiae listed in Appendix A.
________________

4
GERARD E. LYNCH, Circuit Judge:

In August 2019, the Department of Homeland Security (“DHS”) issued a

final rule setting out a new agency interpretation of a longstanding provision of

our immigration law that renders inadmissible to the United States any non-

citizen who is likely to become a “public charge.” See Inadmissibility on Public

Charge Grounds, 84 Fed. Reg. 41,292 (Aug. 14, 2019) (“the Rule” or “the Final

Rule”). The Rule expands the meaning of “public charge,” with the likely result

that significantly more people will be found inadmissible on that basis. Lawsuits

challenging the lawfulness of the Rule were quickly filed around the country,

including two cases in the Southern District of New York, which we now

consider in tandem on appeal.

These two cases – one brought by New York State, New York City,

Connecticut, and Vermont, and the other brought by five non-profit

organizations that provide legal and social services to non-citizens – raise largely

identical challenges to the Rule, centering on the Rule’s validity under the

Administrative Procedure Act. After hearing combined oral argument on the

Plaintiffs’ motions for preliminary injunctions filed in both cases, the district

court (George B. Daniels, J.) concluded that the Plaintiffs had demonstrated a

5
likelihood of success on the merits of their claims and that the other preliminary

injunction factors also favored interim relief. The district court enjoined DHS

from implementing the Rule throughout the United States in the pair of orders

from which DHS now appeals.

We agree that a preliminary injunction is warranted in these cases,

but modify the scope of the injunctions to cover only the states of New York,

Connecticut, and Vermont. The orders of the district court are thus AFFIRMED

AS MODIFIED.

TABLE OF CONTENTS

BACKGROUND. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

I. 1999 Public Charge Guidance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

II. 2019 Public Charge Rule. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

A. The Proposed Rule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

B. Revised Definition and Relevant Public Benefits. . . . . . . . . . . 16

C. Adjudicative Framework. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19

III. Procedural Posture. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

DISCUSSION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

I. Threshold Arguments. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

A. Standing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

B. Zone of Interests. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

6
 II. Likelihood of Success on the Merits.. . . . . . . . . . . . . . . . . . . . . . . . . . . 38

A. Legal Framework.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

B. The Rule is Contrary to the INA... . . . . . . . . . . . . . . . . . . . . . . . 41

1. Origins of the Public Charge Ground.. . . . . . . . . . . . . . . . . 42

2. The Immigration Act of 1917. . . . . . . . . . . . . . . . . . . . . . . . . 47

3. The Immigration and Nationality Act of 1952. . . . . . . . . . 51

4. The Current Public Charge Ground. . . . . . . . . . . . . . . . . . . 54

5. The Settled Meaning of “Public Charge”. . . . . . . . . . . . . . . 56

6. The Rule’s Inconsistency with the Settled Meaning.. . . . . 67

C. The Rule is Arbitrary and Capricious.. . . . . . . . . . . . . . . . . . . . 84

1. Explanation for Changed Definition.. . . . . . . . . . . . . . . . . . 85

2. Explanation for Expanded List of Benefits. . . . . . . . . . . . . 89

III. Irreparable Harm to the Plaintiffs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97

IV. Balance of Equities and the Public Interest. . . . . . . . . . . . . . . . . . . . . . 99

V. Scope of Injunction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101

CONCLUSION. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105

Appendix A. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106

BACKGROUND

The Immigration and Nationality Act (“INA”) contains ten grounds of

inadmissibility, each listing various bases on which a non-citizen can be denied

admission to the United States. See 8 U.S.C. § 1182(a)(1)-(10). These appeals

concern the public charge ground, a constant feature of our immigration law

7
since 1882, which renders inadmissable any non-citizen who “is likely at any time

to become a public charge.” Id. § 1182(a)(4)(A). The statute itself does not define

“public charge,” and its precise meaning is the hotly contested question in this

litigation. In general terms, however, “public charge” has historically been

understood to refer to a person who is not self-sufficient and depends on the

government for support. See, e.g., 84 Fed. Reg. at 41,295.

The grounds of inadmissibility are assessed not only when a person is

physically entering the country, but at multiple points in the immigration

process. Consequently, the public charge ground of inadmissibility is applied by

three agencies that oversee different aspects of our immigration system. The

Department of State considers whether non-citizens are inadmissible as likely

public charges when adjudicating visa applications overseas. U.S. Customs and

Border Protection (“CBP”), a unit of DHS, assesses the public charge ground

when it inspects non-citizens arriving at airports or other ports of entry. And U.S.

Citizenship and Immigration Services (“USCIS”), another component of DHS,

applies the ground when adjudicating applications for adjustment of status, the

process by which a non-citizen who is already present in the United States in a

temporary immigration status can become a lawful permanent resident (“LPR”),

8
authorized to live and work in the United States indefinitely.1 See id. at 41,294 n.3.

The Department of Justice also has a role to play when it comes to public

charge adjudications, albeit on a different statutory basis. In addition to the

public charge ground of inadmissibility, the INA also contains a public charge

ground of removal.2 8 U.S.C. § 1227(a)(5). That provision authorizes the

government to remove non-citizens who have already been admitted to the

country but who became public charges within five years of their date of entry.

Id. The public charge ground of removal is primarily applied by the Executive

Office for Immigration Review, a component agency of the Department of Justice

that houses the immigration courts.

While multiple agencies are tasked with interpreting and applying the

public charge grounds of inadmissibility and removal, the Rule at issue in these

cases is an interpretation by DHS of the ground of inadmissibility. Accordingly,

the Rule governs only public charge determinations carried out by CBP and

1
LPRs are frequently referred to in popular discussion as “green card holders.”
2
“Removal” is the current legal term for the process popularly known as
“deportation.” See Karageorgious v. Ashcroft, 374 F.3d 152, 154 (2d Cir. 2004).

9
USCIS, as component agencies of DHS.3 84 Fed. Reg. at 41,294 n.3. As a practical

matter, moreover, the Rule is likely to be applied primarily by USCIS as it

adjudicates applications for adjustment of status, as the lengthy application

process provides more opportunity for a full consideration of the Rule’s

provisions than a CBP screening at a port of entry. See id. at 41,478.

I. 1999 Public Charge Guidance

For twenty years preceding the publication of the Rule at issue in these

cases, the governing agency interpretation of the public charge ground was

guidance published in 1999 (“the 1999 Guidance”) by the Immigration and

Nationality Service (“INS”), the predecessor agency of DHS.4 See Field Guidance

3
In October 2019, the State Department issued an interim final rule aligning its
interpretation of “public charge” with the Rule. See Visas: Ineligibility Based on
Public Charge Grounds, 84 Fed. Reg. 54,996 (Oct. 11, 2019). Litigation challenging
the State Department interim final rule is underway in the Southern District of
New York. See Make the Road New York v. Pompeo, No. 1:19-cv-11633 (S.D.N.Y.).
The Department of Justice has drafted a proposed rule that likewise is intended
to adopt a conforming interpretation of the public charge ground of removal,
which has been sent to the Office of Management and Budget for review, but no
actual text of such a rule has yet been published. See Inadmissibility and
Deportability on Public Charge Grounds, RIN 1125-AA84, Office of Mgmt. &
Budget, Spring 2020 Unified Agenda of Regulatory and Deregulatory Actions.
4
INS was dissolved, and many of its responsibilities were transferred to DHS, by
the Homeland Security Act of 2002. See Pub. L. No. 107-296, §§ 402(3), 471, 116
Stat. 2135, 2205, 2178.

10
on Deportability and Inadmissibility on Public Charge Grounds, 64 Fed. Reg.

28,689 (May 26, 1999). The 1999 Guidance was issued in response to two pieces of

legislation passed by Congress in 1996 that had significant impact on the public

charge ground.

The first was the Personal Responsibility and Work Opportunity

Reconciliation Act (“PRWORA”), a sweeping set of reforms to various public

benefits programs. See Pub. L. No. 104-193, 110 Stat. 2105 (1996). Among other

changes, PRWORA greatly restricted non-citizen access to public benefits in

response to concerns that non-citizens were “applying for and receiving public

benefits . . . at increasing rates.” 8 U.S.C. § 1601(3). The resulting benefits

eligibility scheme for non-citizens is complex, to say the least. It suffices for

present purposes to say that non-citizens who are present in the United States

illegally or who are admitted in a lawful non-immigrant (i.e., temporary) status

are ineligible for almost all federal benefits, see 8 U.S.C. §§ 1611(a), 1641(b), while

those who are in LPR status, which is permanent, are ineligible for means-tested

federal benefits for their first five years as an LPR, see 8 U.S.C. §§ 1613(a), 1641(b).

11
At the conclusion of this five-year waiting period, LPRs become eligible to

receive benefits for which they otherwise qualify.5

A little over a month after enacting PRWORA, Congress passed the Illegal

Immigration Reform and Immigrant Responsibility Act (“IIRIRA”). See Pub. L.

No. 104-208, div. C, 110 Stat. 3009-546 (1996). In IIRIRA, Congress revisited the

public charge ground to add five factors that adjudicators must consider when

determining whether a non-citizen is likely to become a public charge: the non-

citizen’s “[(1)] age; [(2)] health; [(3)] family status; [(4)] assets, resources, and

financial status; and [(5)] education and skills.” 8 U.S.C. § 1182(a)(4)(B)(i). IIRIRA

also required non-citizens seeking to immigrate to the United States based on

their family ties6 to obtain affidavits of support, in which a sponsor agrees to

maintain the non-citizen at an income of no less than 125% of the federal poverty

5
The majority of the public benefits to which the Rule applies are means-tested
benefits, that is, there are income and asset limits for eligibility. However, the
housing programs administered by the Department of Housing and Urban
Development are not considered means-tested benefits and there is thus no five-
year waiting period before LPRs can access these services. See Eligibility
Restrictions on Noncitizens: Inapplicability of Welfare Reform Act Restrictions on
Federal Means-Tested Public Benefits, 65 Fed. Reg. 49,994 (Aug. 16, 2000).
6
Persons seeking to immigrate to the United States are eligible for admission as
immigrants on various bases, including having certain familial relationships to
United States citizens or LPRs.

12
guidelines (“FPG”), and instructed adjudicators to consider those affidavits as a

discretionary sixth factor in their analysis. 8 U.S.C. §§ 1182(a)(4)(c), 1183a(a)(1).

After the passage of PRWORA and IIRIRA, INS observed widespread

“confusion about the relationship between the receipt of federal, state, [and] local

public benefits and the meaning of ‘public charge’ under the immigration laws.”

64 Fed. Reg. at 28,689. Concerned that this confusion was “deterr[ing] eligible

aliens and their families, including U.S. citizen children, from seeking important

health and nutrition benefits that they [we]re legally entitled to receive,” INS

issued the 1999 Guidance, thus for the first time publishing its interpretation of

“public charge” in the Federal Register.7 Id. at 28,692.

The 1999 Guidance defined “public charge” to mean a person who is

“primarily dependent on the government for subsistence, as demonstrated by

either (i) the receipt of public cash assistance for income maintenance or (ii)

7
The 1999 Guidance was not a final rule, but was published in the Federal
Register as an interim measure to establish the agency’s “public charge”
definition while INS went through the rulemaking process. 64 Fed. Reg. at 28,689.
The Guidance was a reproduction of INS’s field guidance, making public the
internal directive of the agency to its officials tasked with applying the “public
charge” standard. INS did publish a proposed rule alongside the 1999 Guidance,
but it was never finalized. See Inadmissibility and Deportability on Public Charge
Grounds, 64 Fed. Reg. 28,676 (May 26, 1999). The 1999 Guidance remained the
operative agency interpretation until 2019.

13
institutionalization for long-term care at government expense.” Id. at 28,689

(internal quotation marks omitted). The Guidance identified four public benefits

that could be taken as evidence of primary dependence: Supplemental Security

Income (“SSI”), which “guarantees a minimum level of income” for older adults

and people who are blind or disabled; Temporary Assistance for Needy Families

(“TANF”), which provides cash assistance to families living in poverty;8 state and

local cash assistance programs, often called “General Assistance” programs; and

any program (including Medicaid) supporting people institutionalized for long-

term care. Id. at 28,692, 28,687; see 45 C.F.R. § 260.20.

INS explained that the nature of these benefits suggested that recipients

may be dependent on the government for subsistence, explicitly distinguishing

non-cash benefits that are “by their nature supplemental and do not, alone or in

combination, provide sufficient resources to support an individual or family.” 64

Fed. Reg. at 28,692.9 The Guidance instructed that the ultimate determination as

8
TANF also funds various forms of non-cash assistance, e.g., subsidized child
care. These additional forms of support were excluded from consideration under
the Guidance. See 64 Fed. Reg. at 28,692 n.17.
9
The 1999 Guidance explicitly stated that adjudicators “should not place any
weight on the receipt of non-cash public benefits (other than institutionalization)
or the receipt of cash benefits for purposes other than for income maintenance.”

14
to whether a non-citizen was primarily dependent on the government was to be

made by considering the totality of the circumstances: neither current nor past

“receipt of cash income-maintenance benefits . . . automatically ma[de] an alien

inadmissible as likely to become a public charge.” Id. at 28,690.

II. 2019 Public Charge Rule

A. The Proposed Rule

Nearly two decades after INS issued its 1999 interpretation of “public

charge,” DHS published a notice of proposed rulemaking (“the Proposed Rule”)

announcing its intention to change the agency’s interpretation of the public

charge ground. See Inadmissibility on Public Charge Grounds, 83 Fed. Reg.

51,114 (Oct. 10, 2018). Among other provisions, the Proposed Rule suggested

redefining “public charge” to mean “an alien who receives one or more public

benefit” at certain defined usage thresholds, and listed a broader set of benefits as

relevant to the public charge definition. Id. at 51,289-90.10

64 Fed. Reg. at 28,689.
10
Technically, the Proposed Rule defined public charge to mean any non-citizen
who received any “public benefit,” but then further defined “public benefit” to
mean one of the listed benefits if usage exceeded the prescribed threshold level,
as described in the next paragraph.

15
 The Proposed Rule divided its list of relevant benefits into two groups –

monetizable and non-monetizable – and set usage thresholds for each. The

monetizable benefits (e.g., SSI) were to be considered in the public charge

analysis if the cumulative value of the benefits received in one year exceeded 15%

of FPG for a household of one. Id. The non-monetizable benefits (e.g., Medicaid)

were counted if the non-citizen received the benefit “for more than 12 months in

the aggregate within a 36 month period.” Id. at 51,290. The Proposed Rule

garnered 266,077 comments during the notice and comment period, “the vast

majority of which opposed the rule.”11 84 Fed. Reg. at 41,297.

B. Revised Definition and Relevant Public Benefits

In August 2019, DHS published its Final Rule, which made a number of

changes from the Proposed Rule. Most relevant for our purposes, the Rule enacts

a different definition of “public charge,” interpreting the term as a person “who

receives one or more public benefits, as defined in [a subsequent] section, for

11
We note that these appeals have also generated significant public interest and
acknowledge with appreciation the contributions of the amici curiae appearing
before us. The twenty amicus briefs we received (nineteen of which support the
Plaintiffs, and one of which supports the government) represent the views of a
diverse collection of more than four hundred organizations, businesses, and
scholars and provided helpful nuance on many aspects of the complex questions
before us.

16
more than 12 months in the aggregate within any 36-month period (such that, for

instance, receipt of two benefits in one month counts as two months).” Id. at

41,501. While the Final Rule incorporates the same expanded list of relevant

public benefits as the Proposed Rule, it did away with the categorization of

“monetizable” versus “non-monetizable” benefits, eliminated the 15% of FPG

threshold requirement for monetizable benefits, and elevated the 12-month

threshold requirement for non-monetizable benefits into the definition of public

charge itself, thus making it the usage threshold for all of the listed benefits. Id. at

41,501-02.

With respect to the relevant benefits, the Final Rule retains those benefits

INS made relevant to the public charge determination in 1999 – SSI, TANF, and

state or local cash assistance programs – and adds a number of other benefits:

Medicaid;12 the Supplemental Nutrition Assistance Program (“SNAP”),

commonly referred to as food stamps; the Section 8 Housing Choice Voucher

12
The 1999 Guidance identified Medicaid as a relevant public benefit only when
it was used to fund long-term institutionalization, but the Rule broadens the
consideration to include Medicaid used to fund most forms of routine healthcare.
The Rule does not count Medicaid benefits only when they are used for
emergency medical conditions, for services provided under the Individuals with
Disabilities Education Act, for school-based services, and by children or pregnant
and newly postpartum women. 84 Fed. Reg. at 41,501.

17
Program; Section 8 Project-Based Rental Assistance; and public housing. Id. at

41,501. Thus, under the Final Rule, use of any quantity of one of these benefits in

a given month counts as one month towards the 12-months-within-36 months

limit beyond which one is considered a public charge. And because the definition

aggregates benefits usage, use of two benefits in a single month counts as two

months (and three benefits in a single month counts as three months, etc.), with

the result that a person could reach the 12-month threshold in six months or

fewer. The 12-month threshold is thus deceptive: an industrious, self-sufficient

person who, by reason of a temporary injury or illness, used three benefits per

month for four months would thereby be conclusively established as a public

charge.

The Rule’s public charge definition would be complex to apply even to

assess past or current benefits usage. But lest we forget the context in which the

Rule operates, we highlight that the vast majority of non-citizens will not have

been eligible to receive any of the relevant public benefits (and therefore

presumably will not have received such benefits) at the time the Rule is applied

18
and their likelihood of becoming a public charge is assessed.13 Recall that the Rule

applies primarily to non-citizens seeking to adjust status to become LPRs but that,

in general, non-citizens are not eligible to receive the relevant public benefits

until five years after they obtain LPR status. Accordingly, very few non-citizens

will have a history of public benefits usage at the time the forward-looking public

charge ground is applied. Under the revised public charge definition, the Rule

thus requires adjudicators to predict whether, five years or more into the future,

the non-citizen is likely to use one of the enumerated benefits for more than

twelve months, or to use two of the enumerated benefits for more than six

months, and so on, within a thirty-six-month period. Id. at 41,502.

C. Adjudicative Framework

To support adjudicators in making what might seem an impracticable

prediction about future benefits usage, the Rule lays out an adjudicative

framework. This framework fleshes out the five factors adjudicators are

13
We note that PRWORA allows states flexibility to determine non-citizen
eligibility for state-funded benefits programs and some states have chosen to
fund benefits for those who do not yet have LPR status. See 83 Fed. Reg. at 51,131;
see also 8 U.S.C. § 1621(d). However, as counsel for the government
acknowledged at oral argument, it would be the rare exception if a non-citizen
received benefits prior to the public charge determination. Oral Argument at
32:00-33:30.

19
statutorily required to consider – age, health, family status, finances, and

education – and explains how each should be analyzed to decide whether a non-

citizen is a likely future user of public benefits. The Rule further instructs

adjudicators how to assess the sixth, discretionary factor – the affidavit of

support – and adds a seventh factor for consideration, the immigration status

sought. As laid out below, the framework identifies the particular characteristics

adjudicators should look for with each factor and, for some, lists the forms of

evidence the non-citizen must submit.

Age. Adjudicators are to assess whether a non-citizen’s age affects his

ability to work. The Rule suggests that preference be given to non-citizens

between eighteen and sixty-one years of age. Being under eighteen or over sixty-

one is treated as making it more likely that the applicant will become a public

charge. 84 Fed. Reg. at 41,502.

Health. Adjudicators are to be on the lookout for non-citizens with medical

conditions that are “likely to require extensive medical treatment or

institutionalization or that will interfere with the alien’s ability to provide and

care for himself or herself, to attend school, or to work.” Id.

20
 Family Status. Adjudicators are to assess whether a non-citizen’s household

size makes him more likely to utilize the listed benefits. Large families are thus

more suspect. Id.

Assets, Resources, and Financial Status. Adjudicators are to consider whether

the non-citizen’s household has a gross income above 125% of FPG or has

significant assets, whether the household assets and resources would cover any

reasonably foreseeable medical costs, any outstanding financial liabilities, and

whether the non-citizen has ever in the past applied for or received any of the

enumerated public benefits. Id. at 41,502-03. Non-citizens are to evidence this

factor by submitting, inter alia, their tax returns, bank statements, credit history

and score, and proof of private health insurance. Id. at 41,503.

Education and Skills. Adjudicators are to consider whether the non-citizen

has adequate education and skills to obtain lawful employment with an income

sufficient to avoid becoming a public charge. Id. Non-citizens are directed to

evidence this factor by submitting, inter alia, their employment history, tax

returns, proof of degrees or licenses, and proof of proficiency in English or any

other languages. Id. at 41,503-04.

21
 Affidavit of Support. For those non-citizens who must obtain an affidavit of

support, the Rule directs adjudicators to consider “the likelihood that the sponsor

would actually provide the statutorily-required amount of financial support,”

which is to be evidenced by proof of the sponsor’s income and assets, the

relationship between the non-citizen and sponsor, and the number of other non-

citizens for whom the sponsor has executed affidavits of support. Id. at 41,504.

Desired Immigration Status. The Rule newly requires adjudicators to

consider the immigration status sought by the non-citizen, “as it relates to the

alien’s ability to financially support[ ] himself or herself during the duration of

the alien’s stay.” Id. After the publication of the Rule, USCIS updated its policy

manual to clarify that it would generally treat seeking LPR status as a negative

factor, given that LPRs are eligible for public benefits after the five-year waiting

period has elapsed. See USCIS, POLICY MANUAL vol. 8, pt. G, ch. 12 (2020). We

note that the vast majority of non-citizens who are subject to the Rule are being

assessed precisely because they are seeking LPR status; this provision therefore

would appear to automatically assign a negative factor to any applicant for

lawful immigration to the United States.

22
 After addressing these factors, the Rule concludes by identifying a number

of heavily weighted negative and positive circumstances that adjudicators should

consider in deciding a case. While cautioning that no single factor is dispositive,

the Rule directs adjudicators to give particular emphasis to four heavily weighted

negative factors: (1) lacking a current or recent employment history, (2) receiving

a relevant public benefit for more than twelve months in the preceding three

years, (3) lacking health insurance while having a diagnosed medical condition

likely to require extensive treatment or institutionalization, and (4) having been

found inadmissible or removable on public charge grounds in the past. 84 Fed.

Reg. at 41,504. In contrast, the Rule also identifies the following heavily weighted

positive factors: (1) having a household income that exceeds 250% of FPG, (2)

being employed with an income exceeding 250% of FPG, and (3) having private

health insurance that was not purchased using Affordable Care Act premium tax

credits. Id.

III. Procedural Posture

Shortly after DHS issued the Final Rule in August 2019, the two cases at

issue in these appeals were filed in the Southern District of New York. New York,

Vermont, Connecticut, and New York City (collectively “the States”) filed suit

23
first, followed a few days later by Make the Road New York, African Services

Committee, Asian American Federation, Catholic Charities Community Services

(Archdiocese of New York), and Catholic Legal Immigration Network, Inc.

(collectively, “the Organizations”). The two groups of plaintiffs (collectively, “the

Plaintiffs”) raise largely similar challenges to the Rule, arguing that it is invalid

under the Administrative Procedure Act (“APA”) as well as the Fifth

Amendment’s due process clause. The Organizations also challenge the Rule

under the Fifth Amendment’s guarantee of equal protection. Both the States and

the Organizations moved for a preliminary injunction, and the district court

heard combined oral argument.

On October 11, 2019, four days before the Rule was scheduled to take

effect, the district court granted both motions for preliminary injunctions in

largely identical decisions and orders. After concluding that both the States and

the Organizations had standing to challenge the Rule, the district court found

that they had demonstrated a likelihood of success on the merits of their claims

that the Rule was contrary to law as well as arbitrary and capricious, and that the

24
other preliminary injunction factors supported injunctive relief. The district court

thus enjoined DHS from enforcing the Rule nationwide.14

DHS timely appealed the district court’s grant of the preliminary

injunctions and moved for a stay pending appeal. A motions panel of this Court

denied DHS’s motion to stay. DHS then filed an application for a stay with the

Supreme Court, requesting that the preliminary injunctions be stayed through

the resolution of the merits of this appeal and the disposition of any petition for a

writ of certiorari. The Supreme Court granted the application in January 2020.

DHS v. New York, 140 S. Ct. 599 (2020). With the district court’s preliminary

injunctions thus stayed, the Rule went into effect nationwide on February 24,

2020.15

14
The district court also ordered that the effective date of the Rule be stayed
pursuant to 5 U.S.C. § 705.
15
Five similar cases challenging the Rule were brought in the District of
Maryland, the Northern District of Illinois, the Northern District of California,
and the Eastern District of Washington. See CASA de Maryland, Inc. v. Trump, No.
19-cv-2715 (D. Md.); Cook Cty. v. Wolf, No. 19-cv-6334 (N.D. Ill.); City and Cty. of
San Francisco v. USCIS, No. 19-cv-4717 (N.D. Cal.); California v. DHS, No. 19-cv-
4975 (N.D. Cal.); Washington v. DHS, No. 19-cv-5210 (E.D. Wash.). All five district
courts granted plaintiffs’ motions for preliminary injunctions. DHS appealed in
all cases and, as here, requested that the Fourth, Seventh, and Ninth Circuits stay
the district courts’ preliminary injunctions pending appeal. The Fourth and a
divided panel of the Ninth Circuit granted DHS’s motions to stay, the Ninth

25
 DISCUSSION

We review a district court’s decision to grant a preliminary injunction for

abuse of discretion, examining the legal conclusions underpinning the decision

de novo and the factual conclusions for clear error. Cty. of Nassau v. Leavitt, 524

F.3d 408, 414 (2d Cir. 2008). The scope of the injunctive relief ordered by the

district court is evaluated for abuse of discretion. Id.

These appeals fall under the preliminary injunction framework laid out in

Winter v. Natural Resources Defense Council, Inc., 555 U.S. 7 (2008). Winter instructs

that “[a] plaintiff seeking a preliminary injunction must establish that he is likely

to succeed on the merits, that he is likely to suffer irreparable harm in the absence

of preliminary relief, that the balance of equities tips in his favor, and that an

injunction is in the public interest.” Id. at 20. Where, as here, the government is a

party to the suit, the final two factors merge. Cf. Nken v. Holder, 556 U.S. 418, 435

Circuit doing so in a lengthy published opinion. See City & Cty. of San Francisco v.
USCIS, 944 F.3d 773 (9th Cir. 2019). The Seventh Circuit denied DHS’s motion to
stay, but DHS successfully sought a stay from the Supreme Court for the
preliminary injunction at issue in that case, which was limited in scope to the
state of Illinois. See Wolf v. Cook Cty., 140 S. Ct. 681 (2020). The Seventh Circuit has
since decided the merits of the case before it, holding that plaintiffs were likely to
succeed on the merits of their challenge to the Rule and affirming the preliminary
injunction entered by the Northern District of Illinois. See Cook Cty. v. Wolf, 962
F.3d 208 (7th Cir. 2020).

26
(2009); see California v. Azar, 911 F.3d 558, 575 (9th Cir. 2018). Before we turn to the

merits of these appeals, however, we address two threshold arguments raised by

DHS.

I. Threshold Arguments

DHS first argues that neither the States nor the Organizations meet the

“irreducible constitutional minimum of standing” and thus cannot be permitted

to challenge the Rule. Lujan v. Defs. of Wildlife, 504 U.S. 555, 560 (1992). DHS

further argues that the Plaintiffs may not bring suit because they do not fall

within the zone of interests protected by the public charge statute. See Lexmark

Int’l Inc. v. Static Control Components, Inc., 572 U.S. 118, 129 (2014). We disagree

with DHS on both counts.

A. Standing

At the preliminary injunction stage, “a plaintiff’s burden to demonstrate

standing will normally be no less than that required on a motion for summary

judgment. Accordingly, to establish standing for a preliminary injunction, a

plaintiff cannot rest on . . . mere allegations . . . but must set forth by affidavit or

other evidence specific facts” that establish the “three familiar elements of

standing: injury in fact, causation, and redressability.” Cacchillo v. Insmed, Inc.,

27
638 F.3d 401, 404 (2d Cir. 2011) (internal quotation marks and citation omitted).

Here, DHS argues that the States and Organizations have failed to establish

injury in fact, which requires the Plaintiffs to show they have suffered “an

invasion of a legally protected interest which is (a) concrete and particularized,

and (b) actual or imminent, not conjectural or hypothetical.” Lujan, 504 U.S. at 560

(internal quotation marks and citations omitted).

The States allege that they are injured because the Rule will cause many of

their residents to forgo use of public benefits programs, thereby decreasing

federal transfer payments to the states, reducing Medicaid revenue, increasing

overall healthcare costs, and causing general economic harm. DHS argues that

these projected harms do not suffice to show injury in fact because the facts

asserted by the States at most establish a possible, rather than imminent, future

injury, and because any economic losses will be offset by the money saved by not

providing public benefits to those who disenroll. See Clapper v. Amnesty Int’l USA,

568 U.S. 398, 409 (2013).

We are satisfied that the States have sufficiently established actual

imminent harms. DHS itself anticipates that a significant number of non-citizens

will disenroll from public benefits as a result of the Rule’s enactment, including

28
many who are not in fact subject to the Rule but who would be fearful of its

consequences nonetheless. See 84 Fed. Reg. at 41,300-01, 41,463. When an agency

action has a “predictable effect . . . on the decisions of third parties,” the

consequences of those third party decisions may suffice to establish standing,

even when the decisions are illogical or unnecessary. See Dep’t of Commerce v. New

York, 139 S. Ct. 2551, 2566 (2019). Contrary to its disparagement before us of the

likelihood of harm to the States from disenrollment, DHS acknowledged in its

own explication of the costs and benefits considered in adopting the Rule that

expected disenrollment will result in decreased federal funding to states, 84 Fed.

Reg. at 41,485, decreased revenue for healthcare providers, id. at 41,486, and an

increase in uncompensated care, id. at 41,384.

DHS’s own predictions thus align with declarations submitted by the

States documenting the Rule’s chilling effect on non-citizen use of public benefits

– which began even prior to the Rule taking effect – and its anticipated economic

impacts.16 Where the agency itself forecasts the injuries claimed by the States, we

16
For example, the Commissioner of Health of the State of New York stated that
“even before the Final Rule has gone into effect, consumers have been calling . . .
[and] inquiring about canceling their Medicaid or other health insurance
coverage because of the Final Rule.” New York (“N.Y.”) J. App. 512. The
Commissioner further notes that “[i]ndividuals without coverage will still need

29
agree with the Ninth Circuit that it is “disingenuous” for DHS to claim that the

injury is not sufficiently imminent. San Francisco, 944 F.3d at 787 (finding state

and local governments had standing to challenge the Rule).

We are also unpersuaded by DHS’s argument that the States cannot

establish injury in fact because any losses in funding will be offset by the savings

accrued as fewer people seek public assistance. “[T]he fact that an injury may be

outweighed by other benefits . . . does not negate standing.”17 Denney v. Deutsche

and receive care” but without insurance “those costs will be borne by the
healthcare delivery system.” Id. The President and CEO of New York City Health
and Hospitals Corporation provided specific examples of patients refusing care
or requesting disenrollment because of the Rule, and estimated that in the best-
case scenario, the Rule could result in a loss of $50 million in the first year for the
municipal hospital system. See N.Y. J. App. 266-69; see also N.Y. J. App. 183
(Commissioner of the New York City Department of Social Services providing
statistics evidencing a “striking and dramatic drop in non-citizen SNAP cases”
since the public charge rule began to get media coverage); N.Y. J. App. 227, 233-
34 (Commissioner-Designate of Connecticut Department of Social Services
estimating economic harms and increased healthcare costs); N.Y. J. App. 385-86
(Acting Secretary of the Agency of Human Services in Vermont predicting
increased use of state-funded services).
17
For largely the same reason, we are not persuaded by DHS’s argument that the
States’ losses will be offset by their continued receipt of Emergency Medicaid
funds, a benefit not impacted by the Rule. We further note that Emergency
Medicaid is limited to care provided after a “sudden onset of a medical condition
manifesting itself by acute symptoms of sufficient severity” if immediate medical
care is necessary to prevent serious health consequences. 42 C.F.R. §
440.255(b)(1), (c). That narrow definition is far from a blanket assurance that all or

30
Bank AG, 443 F.3d 253, 265 (2d Cir. 2006). In any event, this simplistic argument

fails to account for the fact that the States allege injuries that extend well beyond

reduced Medicaid revenue and federal funding to the States, including an overall

increase in healthcare costs that will be borne by public hospitals and general

economic harms. See, e.g., N.Y. J. App. 185 (explaining that “the SNAP program

has a direct economic multiplier effect: for every one dollar in SNAP benefits

received, there is an approximate $1.79 in increased economic activity”); N.Y. J.

App. 512-13. Again, DHS itself identified these same broader harms as likely

outcomes of the Rule. See, e.g., REGULATORY IMPACT ANALYSIS, INADMISSIBILITY ON

PUBLIC CHARGE GROUNDS, RIN 1615-AA22, at 105-06 (2019) (calculating that

reduced use of SNAP caused by the Rule will result in an estimated annual

decrease of approximately $550 million in economic activity). We are satisfied

that the States’ alleged economic harms are sufficiently concrete and imminent to

constitute injury in fact.

The Organizations allege injury on the grounds that the Rule has

necessitated significant and costly changes in their programmatic work and

caused increased demand on their social service programs. DHS contends that

even most services rendered in an emergency-room setting will be covered.

31
the Organizations have only shown harm to their “abstract social interests” and

that increased costs of representing clients after the Rule is not sufficient to confer

standing. Appellants’ Br. at 23 (quoting Havens Realty Corp. v. Coleman, 455 U.S.

363, 379 (1982)).

An organization need only show a “perceptible impairment” of its

activities in order to establish injury in fact. Ragin v. Harry Macklowe Real Estate

Co., 6 F.3d 898, 905 (2d Cir. 1993). Contrary to DHS’s assertion that the

Organizations have merely altered the subject matter of their existing outreach

work, the declarations submitted by the Organizations make clear that the Rule

has required significant diversion of resources. For example, over the course of

three months Make the Road New York conducted almost forty workshops for

community members devoted exclusively to the Rule, necessitating the hiring of

two part-time staff members. See Make the Road (“M.T.R.”) J. App. 319-20, 323.

The complexities of the Rule required Catholic Charities to change its educational

outreach from group sessions to time-intensive individual meetings and to

institute a series of evening phone banks. See M.T.R. J. App. 344, 349-51. The

African Services Committee is funding a campaign of radio-based public service

announcements to disseminate information about the Rule and has documented

32
an increased demand on its social service programs, as clients turn away from

public benefits programs.18 See M.T.R. J. App. 466-67, 470.

“[A] nonprofit organization establishes an injury-in-fact if, as here, it

establishes that it spent money to combat activity that harms its . . . core

activities.” Centro de la Comunidad Hispana de Locust Valley v. Town of Oyster Bay,

868 F.3d 104, 111 (2d Cir. 2017) (internal quotation marks omitted). The

Organizations are dedicated to providing an array of legal and social services to

non-citizens and they have expended significant resources to mitigate the Rule’s

impact on those they serve. In so doing, they have diverted resources that would

otherwise have been available for other programming, a “perceptible

opportunity cost” that suffices to confer standing. Nnebe v. Daus, 644 F.3d 147, 157

(2d Cir. 2011).

The Rule will also impede the Organizations’ abilities to carry out their

responsibilities in a variety of ways. Oyster Bay, 868 F.3d at 110 (finding standing

18
Similarly, the Asian American Federation has devoted resources to a press
conference and media-based outreach campaign and plans to reallocate staff to
implement new programmatic priorities in light of the Rule. See M.T.R. J. App.
487, 490-91. And the Catholic Legal Immigration Network has seen a three-fold
increase in the volume of inquiries related to the public charge ground and
anticipates redirecting staff currently assigned to other projects to respond to the
Rule. See M.T.R. J. App. 502-04.

33
where an organization “face[d] increased difficulty in meeting with and

organizing [day] laborers”). For example, the Asian American Federation, which

“support[s] culturally appropriate health and human services for Asian

American immigrants[,]” is preparing to establish a network of social service

providers that will not ask for immigration status information in order to provide

alternatives for non-citizens who will not access public benefits because of the

Rule. M.T.R. J. App. 485-86, 490. And while Catholic Charities was previously

able to assign adjustment of status cases to paralegals working under the

supervision of accredited representatives or attorneys, it anticipates that most of

the adjustment cases for its predominantly low-income clients will now need to

be handled by an attorney and require in-person representation at adjustment

interviews. M.T.R. J. App. 346-48.

These injuries constitute “far more than simply a setback to the

[Organizations’] abstract social interests.” Havens Realty, 455 U.S. at 379. Even

before its entry into force, the Rule has caused a “perceptible impairment” of the

Organizations’ activities and further harms are imminent. Oyster Bay, 868 F.3d at

110 (internal quotation marks omitted). As with the States, we conclude that the

injuries alleged by the Organizations suffice to confer Article III standing.

34
 B. Zone of Interests

DHS also argues that neither group of Plaintiffs falls within the zone of

interests of the public charge statute. The zone-of-interests test restricts the ability

to bring suit to those plaintiffs whose interests are “arguably within the zone of

interests to be protected or regulated by the statute that [they] say[] was

violated.” Match-E-Be-Nash-She-Wish Band of Pottawatomi Indians v. Patchak, 567

U.S. 209, 224 (2012) (internal quotation marks omitted). Because Congress

intended to make agency action presumptively reviewable under the APA, that

test is not especially demanding in the context of APA claims and may be

satisfied even if there is no “indication of congressional purpose to benefit the

would-be plaintiff.” Clarke v. Sec. Indus. Ass’n, 479 U.S. 388, 399-400 (1987). A

plaintiff is precluded from bringing suit only where its “interests are so

marginally related to or inconsistent with the purposes implicit in the statute that

it cannot reasonably be assumed that Congress intended to permit the suit.” Id. at

399. Here, DHS argues that the Plaintiffs’ interests fall outside the zone of

interests of the statute because the Plaintiffs seek to facilitate greater use of public

benefits by non-citizens, which it views as inconsistent with the purpose of the

public charge ground.

35
 This argument mischaracterizes both the purpose of the public charge

statute and the Plaintiffs’ interests. DHS assumes the merits of its own argument

when it identifies the purpose of the public charge ground as ensuring that non-

citizens do not use public benefits. As we conclude infra in Section II.B.6,

Congress enacted the public charge ground to refuse admission to non-citizens

who will likely be unable to support themselves in the United States, which is not

tantamount to ensuring that non-citizens do not access any public benefits.

Moreover, when we consider the role of the public charge ground within

the broader context of the INA, a fuller picture of the interests implicated in the

statute emerges. See Air Courier Conference of Am. v. Am. Postal Workers Union, 498

U.S. 517, 529 (1991) (explaining the Supreme Court’s reasoning in Clarke that, in

the context of the National Bank Act “the zone-of-interests test was to be applied

not merely in the light of § 36, which was the basis of the plaintiffs’ claim on the

merits, but also in the light of § 81, to which § 36 was an exception”). The public

charge statute delineates a category of persons who are to be denied adjustment

of status (or another form of admission) to which they would otherwise have a

claim. See, e.g., 8 U.S.C. § 1255 (detailing requirements for adjustment of status).

The grounds of inadmissibility are the fulcrum on which Congress balances its

36
interest in allowing admission where it advances goals of family unity and

economic competitiveness against its interest in preventing certain categories of

persons from entering the country. See 84 Fed. Reg. at 41,306. DHS suggests that

only those parties advocating increasingly harsher interpretations of the grounds

of inadmissibility could fall within the zone of interests protected by the statute.

That is too narrow a read of both the zone-of-interests test itself and the interests

protected by the public charge ground. Understood in context, its purpose is to

exclude where appropriate and to not exclude where exclusion would be

inappropriate. See Patchak, 567 U.S. at 225-26.

As with the interests protected by the statute, DHS mischaracterizes the

Plaintiffs’ interests when it claims they seek only increased non-citizen

enrollment in public benefits. The States actually seek to protect the economic

benefits that result from healthy, productive, and engaged immigrant

communities. And the Organizations’ interests stem from their assorted missions

to increase non-citizen well-being and status, which they express in their work to

provide legal and social services to non-citizens. An overbroad interpretation of

the public charge ground, tipping the balance too far in the direction of exclusion

at the expense of admission in the interest of family unity and economic vitality,

37
imperils both these interests. See Clarke, 479 U.S. at 399 n.14 (finding zone of

interests could apply to “those whose interests are directly affected by a broad or

narrow interpretation of the [statute]” (internal quotation marks omitted)).

The Plaintiffs are among “those who in practice can be expected to police

the interests that the statute protects[,]” namely, the admission of non-citizens

who will be self-sufficient and the exclusion of those who will not. Fed. Defs. of

N.Y., Inc. v. Fed. Bureau of Prisons, 954 F.3d 118, 131 (2d Cir. 2020) (internal

quotation marks omitted); see Bank of Am. Corp. v. City of Miami, 137 S. Ct. 1296,

1304 (2017). We conclude that the States and the Organizations have Article III

standing to challenge the Rule and that they fall within the zone of interests of

the public charge statute. We thus turn to the merits of these appeals.

II. Likelihood of Success on the Merits

We begin by considering whether the Plaintiffs are likely to succeed on the

merits of their claims, the first preliminary injunction factor. See Winter, 555 U.S.

at 20. The Plaintiffs challenge the Rule under the APA, which declares unlawful

any agency action that is “arbitrary, capricious, an abuse of discretion, or

otherwise not in accordance with law[.]” 5 U.S.C. § 706(2)(A). Though the

Plaintiffs contend that the Rule violates the APA for several reasons, we focus on

38
the arguments that the Rule is unlawful because it is contrary to the INA and

because it is arbitrary and capricious.

A. Legal Framework

“We evaluate challenges to an agency’s interpretation of a statute that it

administers within the two-step Chevron deference framework.” Catskill

Mountains Chapter of Trout Unlimited, Inc. v. EPA, 846 F.3d 492, 507 (2d Cir. 2017).

At the first step of Chevron, we consider “whether Congress has directly spoken

to the precise question at issue. If the intent of Congress is clear, that is the end of

the matter; for the court, as well as the agency, must give effect to the

unambiguously expressed intent of Congress.” Chevron, U.S.A., Inc. v. Nat. Res.

Def. Council, Inc., 467 U.S. 837, 842-43 (1984). Accordingly, we start our analysis

below by considering whether Congress has spoken to its intended meaning of

the statutory term “public charge” and conclude that it has done so. Because the

intent of Congress is clear, “Chevron leaves the stage” and we proceed to the

central question of whether DHS’s interpretation of “public charge” is consistent

with this intent. Epic Sys. Corp. v. Lewis, 138 S. Ct. 1612, 1630 (2018) (internal

quotation marks omitted). We conclude that the Rule is contrary to the INA and

39
that the Plaintiffs have thus demonstrated a likelihood of success on the merits of

this claim. See 5 U.S.C. § 706(2)(A).

We then move to the Plaintiffs’ second argument, that the Rule is unlawful

for the further reason that it is procedurally arbitrary and capricious. See id. We

consider this argument under the familiar rubric laid out in Motor Vehicle

Manufacturers Ass’n of the United States, Inc. v. State Farm Mutual Automobile

Insurance Co., 463 U.S. 29 (1983), which asks whether the agency has “articulate[d]

a satisfactory explanation for its action.”19 Id. at 43. We conclude that DHS failed

to provide a reasoned explanation for its changed definition and the expanded

list of relevant public benefits and that the Plaintiffs are thus also likely to

19
As we noted in Catskill Mountains, there has been “[m]uch confusion” about the
relationship between Chevron and the State Farm frameworks. 846 F.3d at 522. We
distinguished the two, however, on the grounds that “State Farm is used to
evaluate whether a rule is procedurally defective as a result of flaws in the
agency’s decisionmaking process” while Chevron “is generally used to evaluate
whether the conclusion reached as a result of that process . . . is reasonable.” Id. at
521. On appeal, the Plaintiffs primarily raise procedural challenges to the Rule.
We thus consider these arguments under the State Farm framework. See also Nat.
Res. Def. Council, Inc. v. U.S. EPA, 961 F.3d 160, 170-71 (2d Cir. 2020).

40
succeed on the merits of their claim that the Rule is arbitrary and capricious

under 5 U.S.C. § 706(2)(A).20

Accordingly, because the Plaintiffs have shown a likelihood of success on

these two arguments, they have established the first preliminary injunction factor

in their favor.

B. The Rule is Contrary to the INA.

“In a statutory construction case, the beginning point must be the language

of the statute[.]” Estate of Cowart v. Nicklos Drilling Co., 505 U.S. 469, 475 (1992). “If

the statutory text is ambiguous, we also examine canons of statutory

construction” to identify congressional intent. Catskill Mountains, 846 F.3d at 512;

see Chevron, 467 U.S. at 843 n.9. Here, the Plaintiffs do not argue that “public

charge” is unambiguous on its face, relying instead on the ratification canon to

ascertain the clear intent of Congress.

The ratification canon provides that “Congress is presumed to be aware of

an administrative or judicial interpretation of a statute and to adopt that

20
Because we find the Plaintiffs are likely to succeed on the merits of their two
primary arguments, we need not address their additional argument that the Rule
is contrary to the Rehabilitation Act or the Organizations’ argument that the Rule
violates equal protection.

41
interpretation when it re-enacts a statute without change.” Lorillard v. Pons, 434

U.S. 575, 580 (1978). The Plaintiffs argue that Congress ratified the settled judicial

and administrative interpretations of “public charge” as it repeatedly reenacted

the public charge ground over the course of more than a century – most recently

in 1996 – such that the current public charge statute unambiguously forecloses

the Rule’s new interpretation of the term. In response, DHS argues that its

interpretation is not precluded by the historical interpretations of “public charge”

and that other provisions of the INA show that the Rule is consistent with

Congress’s intended meaning of “public charge.” Proper application of the

ratification canon requires a thorough understanding of the evolution of the

public charge statute, from its inception in 1882 to its enactment in its current

form in 1996, as well as the accompanying body of administrative and judicial

decisions interpreting the term. Accordingly, we begin with a historical review.

1. Origins of the Public Charge Ground

The public charge ground has its roots in concerns that arose in the late

nineteenth century that foreign nations were addressing poverty within their

own borders by funding passage to the United States for their poorer citizens. See

13 CONG. REC. 5,109 (1882). As one of the primary immigrant-receiving states,

42
New York in particular was concerned that, upon arrival, these non-citizens

“bec[a]me at once a public charge . . . get[ting] into our poor-houses and alms-

houses.” Id. In response to the costs of supporting new arrivals and other

expenditures associated with its role overseeing the immigration process, New

York attempted to impose various taxes and bonds on arriving immigrants, as

well as the shipping companies providing their transport. See id. at 5,107. The

Supreme Court, however, repeatedly struck down these state statutes as

unconstitutional, on the grounds that the Constitution vested the power to

“regulate commerce with foreign nations,” U.S. CONST. art. I, § 8, cl. 3, in

Congress. See, e.g., Henderson v. Mayor of New York, 92 U.S. 259, 270 (1875).

New York thus turned to Congress for assistance, lobbying for the

enactment of two provisions that ultimately became law with the Immigration

Act of 1882: the public charge ground of exclusion and the immigrant fund. The

inaugural public charge statute directs immigration inspectors to board arriving

ships and refuse permission to land to any passenger who is “unable to take care

of himself or herself without becoming a public charge[.]” Immigration Act of

1882, Pub. L. No. 47-376, § 2, 22 Stat. 214, 214. While denying entry to those who

could not care for themselves, the Act simultaneously established an “immigrant

43
fund,” which was to be used, inter alia, “for the care of immigrants arriving in the

United States, [and] for the relief of such as are in distress.” Id. § 1. By these

provisions, the Act established a scheme that distinguished between those

arriving immigrants who were “unable to take care of [themselves]” and those

who were merely “in distress.” Id. §§ 1, 2. The former were to be excluded; the

latter provided with financial support. Representatives from New York spoke in

favor of this two-part design, applauding the effort to exclude those who would

depend on public assistance while offering words of praise for the immigrants

who may arrive in need of some aid but ultimately go on to “learn our language,

adapt themselves readily to our institutions, and become a valuable component

part of the body-politic.”13 CONG. REC. 5,108 (statement of Rep. Van Voorhis).

Early interpretations of the term “public charge” from this era come

principally from state courts and treat the term as somewhat interchangeable

with “pauper,” distinguishable from those who were simply poor by the

permanence of the condition. For example, the Massachusetts Supreme Court

explained that a bond could be required for arriving immigrants who had “been

paupers in a foreign land; that is, for those who have been a public charge in

another country; and not merely destitute persons, who, on their arrival here,

44
have no visible means of support[.]” City of Boston v. Capen, 61 Mass. 116, 121

(Mass. 1851). The court affirmed that the bond was necessary only from “those

who, by reason of some permanent disability, are unable to maintain themselves”

and who “might become a heavy and long continued charge to the city, town, or

state, in this country[.]” Id. at 122; see also State v. The S.S. Constitution, 42 Cal. 578,

582 (Cal. 1872); City of Alton v. Cty. of Madison, 21 Ill. 115, 116 (Ill. 1859).

Congress amended the immigration laws in 1891, making slight revisions

to the public charge ground of exclusion while adding for the first time a public

charge ground of deportation. See Immigration Act of 1891, Pub. L. No. 51-551,

§§ 1, 11, 26 Stat. 1084, 1084, 1086. Under the terms of the 1891 Act, “[a]ll idiots,

insane persons, [and] paupers or persons likely to become a public charge” were

to be excluded from admission, id. § 1, while a non-citizen who became “a public

charge within one year after his arrival in the United States” could be deported,

id. § 11.

In 1907, Congress again made modest revisions to the public charge

ground, amending the law to exclude “paupers; persons likely to become a

public charge; [and] professional beggars[.]” Immigration Act of 1907, Pub. L.

No. 59-96, § 2, 34 Stat. 898, 899. A few years after the 1907 Act, the Supreme Court

45
weighed in on the meaning of “public charge” in its first and (as of yet) only

interpretation of the term. In Gegiow v. Uhl, 239 U.S. 3 (1915), the Court

considered the case of two Russian immigrants who had been found likely to

become public charges because they arrived with little money; were bound for

Portland, Oregon, where work was scarce; and had no one legally obligated to

support them. Id. at 8. In its analysis, the Court emphasized that “public charge”

was listed alongside “paupers” and “professional beggars” in the statute,

reasoning that the term should “be read as generically similar to the others

mentioned before and after.” Id. at 10. Accordingly, the Court concluded that a

“public charge,” like the other categories of persons mentioned, must be defined

by some kind of “permanent personal objections.” Id. Because the Russian

immigrants had been deemed likely public charges based on the Portland labor

market, rather than on any intrinsic and problematic characteristics of their own,

the Court reversed the determination.

Citing Gegiow, we declared ourselves “convinced” in a subsequent decision

that “Congress meant [public charge] to exclude persons who were likely to

become occupants of almshouses for want of means with which to support

themselves in the future.” Howe v. United States ex rel. Savitsky, 247 F. 292, 294 (2d

46
Cir. 1917). The Ninth Circuit adopted our interpretation in Ng Fung Ho v. White,

266 F. 765, 769 (9th Cir. 1920), rev’d in part on other grounds, 259 U.S. 276, 285

(1922). Other circuits adopted a somewhat broader interpretation of the term as

encompassing “not only those persons who through misfortune cannot be self-

supporting, but also those who will not undertake honest pursuits, and who are

likely to become periodically the inmates of prisons[,]” Lam Fung Yen v. Frick, 233

F. 393, 396 (6th Cir. 1916) (internal quotation marks omitted); see United States ex

rel. Medich v. Burmaster, 24 F.2d 57, 59 (8th Cir. 1928). But those interpretations as

well emphasized the habitual and persistent nature of the dependency that

would render one a public charge.

2. The Immigration Act of 1917

In the wake of Gegiow, Congress sought to “overcome” the line of cases

that “limit[] the meaning of [public charge] because of its position between other

descriptions conceived to be of the same general and generical nature.” S. COMM.

ON IMMIGRATION , 64TH CONG ., REP. ON H.R. 10384, at 5 (1916). Thus, in the

Immigration Act of 1917, Congress relocated the public charge ground within the

list of excludable persons so that it no longer appeared between paupers and

47
professional beggars, but rather between contract laborers and people who had

been deported previously. See Pub. L. No. 64-301, § 3, 39 Stat. 874, 876.

Notwithstanding Congress’s efforts, “[s]everal courts promptly questioned

the efficacy of the [1917] amendment and affirmed the interpretation that a

‘person who is likely to become a public charge’ is one who for some cause is

about to be supported at public expense[.]” Matter of Harutunian, 14 I. & N. Dec.

583, 587 (B.I.A. 1974). The Ninth Circuit was the first to hold that “this change of

location of the words does not change the meaning that should be given them,

and that it is still to be held that a person ‘likely to become a public charge’ is one

who, by reason of poverty, insanity, or disease or disability, will probably

become a charge on the public.” Ex parte Hosaye Sakaguchi, 277 F. 913, 916 (9th Cir.

1922). A few years later, the Fifth Circuit affirmed that the public charge ground

still “intended to refer to . . . a condition of dependence on the public for

support.” Coykendall v. Skrmetta, 22 F.2d 120, 121 (5th Cir. 1927). And in United

States ex rel. Iorio v. Day, 34 F.2d 920 (2d Cir. 1929), we agreed that the change did

not require overruling the interpretation we had previously adopted in Howe,

noting that “[t]he language itself, ‘public charge,’ suggests . . . dependency.” Id. at

922.

48
 As the courts of appeals applied the public charge ground in this era, the

inquiry usually turned on whether the non-citizen could earn a living, frequently

out of a concern that a health condition might prevent the person from working.21

Conversely, courts routinely found a non-citizen’s ability and willingness to

work sufficient to defeat a public charge finding.22 Administrative interpretations

21
See, e.g., Tod v. Waldman, 266 U.S. 113, 120 (1924) (remanding based on “the
absence from the record of any finding by the department on appeal as to the
issue [of] whether the lameness of Zenia, one of the children, affected her ability
to earn a living or made her likely to become a public charge”); United States ex
rel. Minuto v. Reimer, 83 F.2d 166, 168 (2d Cir. 1936) (affirming public charge
determination where non-citizen “was a woman seventy years old with an
increasing chance of becoming dependent, disabled, and sick [and] [n]o one was
under any obligation to support her”); Tullman v. Tod, 294 F. 87, 88 (2d Cir. 1923)
(affirming public charge determination where the non-citizen “was found to be
affected with deaf mutism, which, as was certified, might affect his ability to earn
a living”); Wallis v. United States ex rel. Mannara, 273 F. 509, 511 (2d Cir. 1921) (“A
person likely to become a public charge is one whom it may be necessary to
support at public expense by reason of poverty, insanity and poverty, disease
and poverty, idiocy and poverty. We think that the finding by the administrative
authorities, showing a physical defect of a nature that may affect the ability of the
relator and appellee to earn a living, is sufficient ground for exclusion [as a likely
public charge]” (internal citation omitted)); see also “Italia” Societa Anonima Di
Navigazione v. Durning, 115 F.2d 711, 713 (2d Cir. 1940).
22
See, e.g., Ex parte Sturgess, 13 F.2d 624, 625 (6th Cir. 1926) (reversing public
charge determination where non-citizen was “39 years of age, in good health, a
skilled carpenter, and had in his possession about $75 in money”); Nocchi v.
Johnson, 6 F.2d 1, 1 (1st Cir. 1925) (reversing public charge determination where
there was “no clear showing that the boy is so feeble-minded that he is not able to
earn his own living” and his parents were wealthy); United States ex rel. Mantler v.

49
issued in the early days of the Board of Immigration Appeals (“BIA”) also

focused on non-citizens’ abilities to work and sustain themselves.23

In the context of the public charge ground of deportation, this era also saw

growing consensus among the courts that non-citizens who had been

institutionalized were deportable as public charges.24 The BIA weighed in on the

matter in one of its first published decisions to address either of the public charge

Comm’r of Immigration, 3 F.2d 234, 236 (2d Cir. 1924) (reversing public charge
determination where non-citizen “is 23 years of age, has been in the country now
for 4 years, is in good physical condition, and by industry and frugality has saved
a substantial portion of her earnings”); Sakaguchi, 277 F. at 916 (reversing public
charge determination where there was no evidence “of mental or physical
disability or any fact tending to show that the burden of supporting the appellant
is likely to be cast upon the public” and the non-citizen was “an able-bodied
woman of the age of 25 years, with a fair education . . . [and] a disposition to
work and support herself”); see also Thack v. Zurbrick, 51 F.2d 634, 635 (6th Cir.
1931); United States ex rel. De Sousa v. Day, 22 F.2d 472, 473-74 (2d Cir. 1927);
Lisotta v. United States, 3 F.2d 108, 111 (5th Cir. 1924).
23
See, e.g., Matter of C-, 3 I. & N. Dec. 96, 97 (B.I.A. 1947) (“In this case there is no
likelihood that the beneficiary will become a public charge. . . . [H]e is in good
health and is able and willing to go to work.”); Matter of V-, 2 I. & N. Dec. 78, 81
(B.I.A. 1944) (reversing public charge determination where the respondent was
employed and “has always been self-supporting” other than during a period of
hospitalization).
24
See, e.g., Canciamilla v. Haff, 64 F.2d 875, 876 (9th Cir. 1933); Fernandez v. Nagle,
58 F.2d 950, 950 (9th Cir. 1932); United States ex rel. Casimano v. Comm’r of
Immigration, 15 F.2d 555, 556 (2d Cir. 1926); United States ex rel. La Reddola v. Tod,
299 F. 592, 593 (2d Cir. 1924).

50
grounds. In Matter of B-, 3 I. & N. Dec. 323 (B.I.A. 1948), the BIA considered the

case of a non-citizen who was institutionalized in a psychiatric hospital run by

the state of Illinois. The BIA held that a non-citizen who had become a public

charge could not be deported on that basis unless the state had a law imposing a

charge for the services rendered, a demand for repayment had been made, and

the non-citizen had failed to reimburse the state. Id. at 326. These procedural

safeguards persist to this day in the public charge ground of deportation, which

considers benefits received, but are not applied in the predictive public charge

ground of inadmissibility. See Harutunian, 14 I. & N. Dec. at 589.

3. The Immigration and Nationality Act of 1952

Shortly after the Matter of B- decision, the Senate initiated “a full and

complete investigation of [the] entire immigration system[,]” the results of which

were released in a Senate Judiciary Committee Report published in 1950. SENATE

JUDICIARY COMM ., THE IMMIGRATION AND NATURALIZATION SYSTEMS OF THE

UNITED STATES, S. REP. NO. 81-1515, at 1 (1950). The investigation and report

resulted in a proposed omnibus bill to overhaul the “patchwork” of the then-

existing immigration and naturalization systems. Id. at 4. Thus was enacted the

51
Immigration and Nationality Act of 1952, the foundation of our current

immigration system.

The Judiciary Committee report devotes several pages to a review of the

public charge ground. The report notes that “courts have given varied definitions

of the phrase ‘likely to become a public charge,’” but summarizes the caselaw as

focusing on four characteristics that indicate non-citizens are likely to become

public charges: (1) impending or current imprisonment in a federal prison; (2)

limited finances; (3) a weakened physical condition “as it relate[s] to his ability

and capacity for employment[;]” and (4) traveling to the United States on a ticket

paid for by someone else. Id. at 347-48. The Judiciary Committee recommended

that the public charge ground be re-enacted in the forthcoming INA and further

recommended that the term not be defined in the statute since “the elements

constituting likelihood of becoming a public charge are varied.” Id. at 349.

Congress took both recommendations, listing “[a]liens who, in the opinion

of the consular officer at the time of application for a visa, or in the opinion of the

Attorney General at the time of application for admission, are likely at any time

to become public charges” as one of the INA’s grounds of inadmissibility. See

Pub. L. No. 82-414, § 212(a)(15), 66 Stat. 163, 183 (1952). The INA also retained the

52
corresponding ground of deportation for any non-citizen who “in the opinion of

the Attorney General, has within five years after entry become a public charge

from causes not affirmatively shown to have arisen after entry.” Id. § 241(a)(8).

Administrative interpretations of “public charge” after the enactment of

the INA largely align with the pre-1952 interpretations. In 1964, the Attorney

General noted that the term had been the subject of “extensive judicial

interpretation” and that the “general tenor” of the caselaw understood “public

charge” to require “[s]ome specific circumstance, such as mental or physical

disability, advanced age, or other fact reasonably tending to show that the

burden of supporting the alien is likely to be cast on the public[.]” Matter of

Martinez-Lopez, 10 I. & N. Dec. 409, 421 (A.G. 1964).

The BIA subsequently affirmed that “while economic factors should be

taken into account, the alien’s physical and mental condition, as it affects ability

to earn a living, is of major significance.” Harutunian, 14 I. & N. Dec. at 588. The

BIA concluded that “[e]verything in the statutes, the legislative comments and

the decisions points to one conclusion[:]” that Congress intended to exclude as a

likely public charge a non-citizen who was not self-supporting. Id. at 589; see also

Matter of Vindman, 16 I. & N. Dec. 131, 132 (B.I.A. 1977).

53
 As the BIA applied this interpretation in subsequent decisions, it focused

on the non-citizen’s capacity for work, reversing decisions that put too much

weight on temporary setbacks and affirming those where a non-citizen had no

prospects for employment by virtue of age or disability.25 And in Matter of Perez,

15 I. & N. Dec. 136, 137 (B.I.A. 1974), the BIA explicitly held that “[t]he fact that

an alien has been on welfare does not, by itself, establish that he or she is likely to

become a public charge.”

4. The Current Public Charge Ground

It was against this backdrop of judicial and administrative interpretations

that Congress enacted PRWORA and IIRIRA in 1996, creating the public charge

ground as it exists today. While leaving the principal statutory language intact –

rendering inadmissible any non-citizen who is “likely at any time to become a

public charge” – IIRIRA amended the ground to require consideration of the non-

25
See, e.g., Matter of A-, 19 I. & N. Dec. 867, 870 (B.I.A. 1988) (“There may be
circumstances beyond the control of the alien which temporarily prevent an alien
from joining the work force. . . . [T]he director placed undue weight on [the
family’s financial circumstances], thereby overshadowing the more important
factors; namely, that the applicant has now joined the work force, that she is
young, and that she has no physical or mental defects which might affect her
earning capacity.”); Vindman, 16 I. & N. Dec. at 132 (affirming public charge
finding where respondents were older adults and had no employment
prospects); cf. Matter of Kowalski, 10 I. & N. Dec. 159, 160 (B.I.A. 1963).

54
citizen’s age, health, family status, financial status, and education. See IIRIRA

§ 531(a). IIRIRA also required certain non-citizens to obtain affidavits of support,

id. § 551(a), building on PRWORA’s requirement that such affidavits of support

be legally enforceable against the sponsor, see PRWORA § 423.

Congress considered, and nearly enacted, a more sweeping set of changes

to the public charge ground with IIRIRA. The conference report of the bill

included a statutory definition of public charge, which would have defined the

term to cover “any alien who receives [means-tested public benefits] for an

aggregate period of at least 12 months[.]” CONFERENCE REPORT, H.R. REP. 104-828,

at 138 (1996). While the House passed the conference report containing this

language, it was ultimately dropped under threat of presidential veto. See 142

CONG. REC. S11,882 (daily ed. Sept. 30, 1996) (statement of Sen. Kyl); cf. Statement

on Senate Action on the “Immigration Control and Financial Responsibility Act

of 1996,” 32 WEEKLY COMP. PRES. DOC. 783 (May 2, 1996) (President Clinton

critiquing prior version of the bill for “go[ing] too far in denying legal

immigrants access to vital safety net programs which could jeopardize public

health and safety”).

55
 We end our historical review back where we started this opinion, with

INS’s release of the 1999 Guidance to counteract public confusion after IIRIRA

and PRWORA. We have already explored in some detail INS’s 1999

interpretation, which defines “public charge” as one who is “primarily

dependent on the Government for subsistence, as demonstrated by either the

receipt of public cash assistance for income maintenance or institutionalization

for long-term care at Government expense.” 64 Fed. Reg. at 28,677. We simply

note here that INS concluded that its interpretation was warranted by “the plain

meaning of the word ‘charge,’ the historical context of public dependency when

the public charge immigration provisions were first enacted more than a century

ago, . . . the expertise of the benefit-granting agencies that deal with subsistence

issues[, and the] factual situations presented in the public charge case law.” Id.

5. The Settled Meaning of “Public Charge”

With this understanding of the history of the public charge ground, we

turn to the applicability of the ratification canon. We first examine whether

Congress changed the statutory language as it amended the ground over the

years, so as to render the canon inapposite. See Holder v. Martinez Gutierrez, 566

U.S. 583, 593 (2012). We then determine whether the caselaw interpretations of

56
the term produced a sufficiently consistent and settled meaning of the term, such

that we may presume Congress ratified that understanding when it created the

current public charge statute in 1996. See Commodity Futures Trading Comm’n v.

Schor, 478 U.S. 833, 846 (1986).

We quickly dispose of the first question. There can be no dispute that, since

its origins in the Immigration Act of 1882, Congress reenacted the public charge

ground without pertinent change in 1891, 1907, 1917, 1952, and 1996. We note

that Congress made minor alterations to the ground over the course of its history.

For example, the 1882 public charge ground excluded anyone who was “unable

to take care of himself or herself without becoming a public charge” at their time

of arrival in the United States while subsequent acts established the forward-

looking likelihood standard. Compare Immigration Act of 1882 § 2 with

Immigration Act of 1891 § 1. And with IIRIRA, Congress added the list of

mandatory factors to consider when applying the ground. See IIRIRA § 531(a).

But Congress has unwaveringly described the fundamental characteristic at issue

as being a “public charge” since 1882. We easily conclude that Congress

“adopt[ed] the language used in [its] earlier act[s]” in its most recent reenactment

of the public charge ground in 1996. See Hecht v. Malley, 265 U.S. 144, 153 (1924).

57
 With respect to the second question, our review of the historical

administrative and judicial interpretations of the ground over the years leaves us

convinced that there was a settled meaning of “public charge” well before

Congress enacted IIRIRA. The absolute bulk of the caselaw, from the Supreme

Court, the circuit courts, and the BIA interprets “public charge” to mean a person

who is unable to support herself, either through work, savings, or family ties. See,

e.g., Day, 34 F.2d at 922; Harutunian, 14 I. & N. Dec. at 588-89. Indeed, we think

this interpretation was established early enough that it was ratified by Congress

in the INA of 1952. But the subsequent and consistent administrative

interpretations of the term from the 1960s and 1970s remove any doubt that it

was adopted by Congress in IIRIRA. See United Airlines, Inc. v. Brien, 588 F.3d 158,

173 (2d Cir. 2009) (noting that “Congress’s repeated amendment of the relevant

provisions of the statute without expressing any disapproval” of the BIA’s

interpretation is “persuasive evidence that the [Agency’s] interpretation is the

one intended by Congress” (internal quotation marks omitted)).

We find particularly significant the Attorney General’s decision from 1962,

which summarizes the “extensive judicial interpretation” of the term as requiring

a particular circumstance, like disability or age, that shows that “the burden of

58
supporting the alien is likely to be cast on the public[.]” Martinez-Lopez, 10 I. & N.

Dec. at 421. Accordingly, the Attorney General held that “[a] healthy person in

the prime of life cannot ordinarily be considered likely to become a public

charge[.]” Id. The BIA came to a similar conclusion after its own review of the

public charge caselaw and legislative history, holding that “any alien who is

incapable of earning a livelihood, who does not have sufficient funds in the

United States for his support, and [who] has no person in the United States

willing and able to assure that he will not need public support is excludable as

likely to become a public charge[.]” Harutunian, 14 I. & N. Dec. at 589-90.

Subsequent administrative decisions affirmed that the public charge

determination must be made based on the totality of the circumstances and

rejected the notion that receipt of public benefits categorically renders one a

public charge. See Perez, 15 I. & N. Dec. at 137; Vindman, 16 I. & N. Dec. at 132; A-,

19 I. & N. Dec. at 870.

The scope and consistency of these administrative decisions warrants the

application of the ratification canon. These published decisions are nationally

binding, issued under the agency’s mandate to “provide clear and uniform

guidance to [other components of the government] and the general public on the

59
proper interpretation and administration of the [INA].” 8 C.F.R. § 1003.1(d)(1).

The broad principles articulated in the decisions are grounded in comprehensive

reviews of public charge history and offer a consistent understanding of the term

as they carry that history forward. While we may not derive a settled rule from

isolated or contradictory decisions, see Jama v. Immigration and Customs Enf’t, 543

U.S. 335, 350-52 (2005), that is not the case here. For more than twenty years prior

to IIRIRA, the agency interpreted “public charge” to mean a person not capable

of supporting himself. In the face of this consistent agency interpretation – which

itself aligns with the earlier judicial interpretations – we conclude that when

Congress reenacted the public charge ground in 1996 it ratified this settled

construction of the term. See FDA v. Brown & Williamson Tobacco Corp., 529 U.S.

157 (2000) (concluding that Congress ratified agency interpretation that had been

its “unwavering position since its inception” and that was consistent “with the

position that its predecessor agency had first taken”).

Our conclusion finds further support in the legislative history of IIRIRA.

“Although we are generally reluctant to employ legislative history at step one of

Chevron” it may be helpful “when the interpretive clues speak almost

unanimously, making Congress’s intent clear beyond reasonable doubt.” Catskill

60
Mountains, 846 F.3d at 515 (internal quotation marks and alterations omitted). We

thus look to the legislative history only to confirm what we have already

concluded.

As noted above, Congress very nearly included a statutory definition of

“public charge” in IIRIRA that would have redefined the term to mean receipt of

any form of means-tested public benefits for more than twelve months. See

CONFERENCE REPORT, H.R. REP. NO. 104-828, at 138. That proposed definition was

intended to overcome the BIA’s Matter of B- decision, which interpreted the

public charge ground of deportation, but it was deleted from the final enactment

under threat of presidential veto.26 See 142 CONG. REC. S4,408-09 (daily ed. April

30, 1996) (statement of Sen. Simpson); 142 CONG. REC. S11,882 (statement of Sen.

26
The definition was proposed as part of the public charge ground of
deportation. CONFERENCE REPORT, H.R. REP. NO. 104-828, at 138. We nevertheless
think it reasonable to look to this language as we interpret the public charge
ground of inadmissibility on the principle that a term appearing in multiple
places within a statute is “generally read the same way each time it appears.”
Ratzlaf v. United States, 510 U.S. 135, 143 (1994). And while the definition was
proposed to overcome Matter of B-’s procedural safeguards for public charge
deportation, the definition is confined to identifying the relevant benefits and
time period of use that made one a “public charge” and could have easily been
transposed to the inadmissibility context. Cf. Harutunian, 14 I. & N. Dec. at 589.

61
Kyl). In effect, an effort was made to change the prior administrative and judicial

consensus as to the meaning of public charge, but that effort failed.

While we agree with DHS that failed legislative proposals are, as a general

matter, unreliable sources of legislative history because bills may fail for any

number of reasons, here we know exactly why the definition was removed from

IIRIRA and find it directly relevant to our analysis. See Solid Waste Agency of N.

Cook Cty. v. U.S. Army Corps of Eng’rs, 531 U.S. 159, 169-70 (2001). We read this

legislative history as further evidence that Congress was aware of prevailing

administrative interpretations of “public charge” when it enacted IIRIRA. See

Lindahl v. Office of Pers. Mgmt., 470 U.S. 768, 782-83 (1985) (applying ratification

canon where legislative history demonstrated Congress was aware of

interpretation). Congress’s abandonment of its efforts to change the meaning of

the term further suggests that it ratified the existing interpretation of “public

charge” in 1996. See Bob Jones Univ. v. United States, 461 U.S. 574, 600-01 (1983).

DHS urges us to conclude that Congress did not ratify any interpretation of

“public charge” because the term has never had a fixed definition. To support its

contention, DHS points to the 1950 Senate Judiciary Committee report, which

observed that “the elements constituting likelihood of becoming a public charge

62
are varied[.]” S. REP. NO. 81-1515, at 349. Consequently, the report recommends

that the term not be defined in the statute and that the determination of whether

a given non-citizen is likely to become a public charge should “rest[] within the

discretion of the [agency].” Id.; see id. at 347 (noting the term has been given

“varied definitions” by the courts).

Rather than suggesting that the core meaning of “public charge” is unclear,

the language on which DHS relies refers to the fact there are a variety of personal

circumstances that may be relied on to show the likelihood that a would-be

immigrant would fall within that category. As described above, the report distills

from the caselaw four circumstances that indicate a non-citizen is likely to

become a public charge. Id. at 348. It thereby recognizes that there are many paths

to dependency, and that administrative flexibility in determining whether a non-

citizen is likely to be dependent is desirable. But the fact that many and varied

circumstances may show that one is likely to become a public charge does not

mean that the underlying term is undefined or lacks a core meaning. And while

DHS makes much of the fact that it retains discretion to decide whether the

ground applies in a given case, the allowance of discretion in individual cases

63
does not mean that the term itself is standardless or without a core, established

meaning.

We recognize that our conclusion that Congress ratified the settled

meaning of “public charge” in 1996 conflicts with decisions from the only two

circuits to have addressed this argument to date. See City and Cty. of San Francisco

v. USCIS, 944 F.3d 773, 798 (9th Cir. 2019); Cook Cty. v. Wolf, 962 F.3d 208, 226 (7th

Cir. 2020). In the context of granting DHS’s motion to stay the injunctions against

the enforcement of the Rule entered by district courts in California and

Washington, the Ninth Circuit decided that “public charge” had been subjected

to “varying historical interpretations” by 1996, such that the ratification canon

did not apply. San Francisco, 944 F.3d at 797. The Ninth Circuit reasoned that

there was no consistent interpretation because

[i]nitially, the likelihood of being housed in a government
or charitable institution was most important. Then, the
focus shifted in 1948 to whether public benefits received
by an immigrant could be monetized, and the immigrant
refused to pay for them. In 1974, it shifted again to
whether the immigrant was employable and self-
sufficient. That was subsequently narrowed in 1987 to

64
 whether the immigrant had received public cash
assistance, which excluded in-kind benefits. 27

Id. at 796. We think the Ninth Circuit goes astray in pinning the definition of

“public charge” on the form of public care provided to the dependent non-citizen.

That the face of our welfare system has changed over time does not mean that the

fundamental inquiry of the public charge ground – whether the non-citizen is

likely to depend on that system – has also changed. The settled meaning of

“public charge,” as the plain meaning of the term already suggests, is

dependency: being a persistent “charge” on the public purse. And as we explain

further below, the mere receipt of benefits from the government does not

constitute such dependency.

We are similarly unpersuaded by the Seventh Circuit’s “admittedly

incomplete” historical review and its conclusion that plaintiffs in that case had

failed to establish that Congress ratified the settled meaning of the term. See Cook

Cty., 962 F.3d at 226. The Seventh Circuit focuses almost exclusively on the state

of the law prior to 1927 and enactments post-dating Congress’s 1996 amendment

27
We explain below our further disagreement with these characterizations of the
1948 Matter of B- decision and the 1987 public charge provision established in the
Immigration Reform and Control Act of 1986, which only applied to those non-
citizens participating in an ad hoc legalization program.

65
to the public charge ground, the point at which any existing interpretation would

have been ratified. Id. at 222-26. Critically, this limited analysis omits the

administrative interpretations of the 1960s and 1970s that established uniform

and nationally binding interpretations of the public charge ground, a key

component of our determination that Congress ratified the prevailing

interpretation of the term in 1996. See id. at 225.

In light of the judicial, administrative, and legislative treatments of the

public charge ground from 1882 to 1996, we hold that Congress ratified the

settled meaning of “public charge” when it enacted IIRIRA. Congress intended

the public charge ground of inadmissibility to apply to those non-citizens who

were likely to be unable to support themselves in the future and to rely on the

government for subsistence. “[D]eference is not due unless a court, employing

traditional tools of statutory construction, is left with an unresolved ambiguity.”

Epic Sys., 138 S. Ct. at 1630 (internal quotation marks omitted). Here, because the

ratification canon reveals the intent of Congress “on the precise question at issue,

that intention is the law and must be given effect.” Chevron, 467 U.S. at 843 n.9.

We thus owe no deference to the Rule and consider only whether it comports

with congressional intent.

66
 6. The Rule’s Inconsistency with the Settled Meaning

“No matter how it is framed, the question a court faces when confronted

with an agency’s interpretation of a statute it administers is always, simply,

whether the agency has stayed within the bounds of its statutory authority.” City

of Arlington v. FCC, 569 U.S. 290, 297 (2013) (emphasis omitted). Having marked

out the interpretive boundaries of “public charge,” we now consider whether the

Rule’s interpretation falls within the ambit of congressional intent. DHS

repeatedly claims that the Rule aligns with the intent of Congress because it

excludes those non-citizens who lack “self-sufficiency and . . . need to rely on the

government for support.” 84 Fed. Reg. at 41,317; see, e.g., id. at 41,295, 41,306,

41,318, 41,320, 41,348. As we have just concluded, Congress did indeed ratify a

consistent and long-standing judicial and administrative understanding of

“public charge” as focused on non-citizens’ abilities to support themselves. But

DHS’s generalized assurance that it shares Congress’s interest in self-sufficiency

is belied by the Rule’s actual definition of “public charge,” which reveals that

DHS and Congress have dramatically different notions of the term.

The prevailing administrative and judicial interpretation of “public

charge” ratified by Congress understood the term to mean a non-citizen who

67
cannot support himself, in the sense that he “is incapable of earning a livelihood,

. . . does not have sufficient funds in the United States for his support, and has no

person in the United States willing and able to assure that he will not need public

support[.]” Harutunian, 14 I. & N. Dec. at 589. Moreover, under that

interpretation the “determination of whether an alien is likely to become a public

charge . . . is a prediction based upon the totality of the alien’s circumstances . . . .

The fact that an alien has been on welfare does not, by itself, establish that he or

she is likely to become a public charge.” Perez, 15 I. & N. Dec. at 137. In contrast,

the Rule categorically renders non-citizens public charges – i.e., not self-sufficient

– if they are likely to access any quantity of the enumerated benefits for a limited

number of months. See 84 Fed. Reg. at 41,349. We think it plain on the face of

these different interpretations that the Rule falls outside the statutory bounds

marked out by Congress. Our conclusion is bolstered by the fact that many of the

benefits newly considered by the Rule have relatively generous eligibility criteria

and are designed to provide supplemental assistance to those living well above

the poverty level, as we discuss in greater detail below.28 See infra Section II.C.2.

28
DHS assumes that receipt of SNAP, Medicaid, or housing assistance shows that
a non-citizen is per se unable to meet basic needs. See, e.g., 84 Fed. Reg. at 41,349.
Because DHS primarily invokes this assumption as a justification for its changed

68
 To be sure, we do not find the intent of Congress evidenced by the

ratification canon to be so precise as to support only one interpretation. On the

contrary, the principles at issue are broad enough that they may support a variety

of agency interpretations. But while an agency may “give authoritative meaning

to the statute within the bounds of th[e] uncertainty” implicit in congressional

intent, “the presence of some uncertainty” does not prevent us from “discern[ing]

the outer limits of [a statutory] term[.]” Cuomo v. Clearing House Ass’n, LLC, 557

U.S. 519, 525 (2009). When the meaning of a statutory term is unclear, federal

agencies specialized in the area receive deference from courts in assigning

meaning to the uncertain language. But the deference is not unlimited. If

Congress passed a statute leaving it unclear whether a term of the statute means

A, B, or C, an appropriate federal agency will receive deference in

concluding that the proper meaning is any one of A, B, or C. But it does not

follow that, because the statutory term could mean either A, B, or C, the

agency will receive deference in interpreting it to mean X or Y or Z, because

interpretation, we address (and reject) it in our analysis of the Plaintiffs’ arbitrary
and capricious challenge. We note it here because the fact that the Rule
incorporates public benefits with broader programmatic aims than basic
subsistence evidences the Rule’s inconsistency with congressional intent.

69
such an interpretation would be inconsistent with the meaning of the

statute.

Whatever gray area may exist at the margins, we need only decide today

whether Congress “has unambiguously foreclosed the [specific] statutory

interpretation” at issue. Catawba Cty. v. EPA, 571 F.3d 20, 35 (D.C. Cir. 2009). And

we conclude that Congress’s intended meaning of “public charge”

unambiguously forecloses the Rule’s expansive interpretation. We are not

persuaded by DHS’s efforts to argue otherwise.

DHS first attempts to argue that its definition of “public charge” is

consistent with the historical caselaw interpretations of the term. DHS points to

two district court cases from the 1920s to claim that Congress ratified a definition

of “public charge” that encompasses minimal and temporary public benefits

usage. The first, Guimond v. Howes, 9 F.2d 412 (D. Me. 1925), held that members of

an immigrant family were likely to become public charges when the husband

was a bootlegger who had been incarcerated for two periods of sixty and ninety

days, respectively. Id. at 413. Because the family had been “supported” by the

town while he was imprisoned, and because the husband’s occupation made it

likely he would return to jail in the future, the district court found the family

70
likely public charges. Id. at 413-14. The second case, Ex parte Turner, 10 F.2d 816,

817 (S.D. Cal. 1926), also found that members of a family were likely public

charges because the husband was “predisposed to physical infirmity” and would

“likely be incapacitated from performing any work or earning support for

himself and [his] family” when his ailments flared up in the future. Because his

wife and children had received charitable aid during his previous two-month

hospitalization, the court anticipated they would do so again when he became

sick in the future and found the family to be likely public charges. Id.

In both cases, the district court did look at previous, short-term receipt of

public benefits in making the public charge determination. But neither case

suggests that this receipt alone rendered the immigrant a public charge. Rather,

the district courts found it significant that the families were likely to repeatedly

become dependent on the public in the future, as the breadwinners of the families

were unlikely to stop bootlegging or to overcome physical infirmity. Guimond, 9

F.2d at 414; Turner, 10 F.2d at 817. These cases thus do not suggest that courts

have historically considered the temporary receipt of benefits as sufficient to

enter a public charge finding. To the contrary, both Guimond and Turner rest on

71
the finding that the benefits usage was not merely temporary but was likely to

regularly reoccur.29

The only other case on which DHS relies is Matter of B-, 3 I. & N. Dec. at

323. DHS argues that Matter of B- shows that a non-citizen is deportable as a

public charge if she fails to reimburse the government for the benefits used, even

if the non-citizen was not primarily dependent on the benefits. DHS reads far too

much into the case. In Matter of B-, the non-citizen was institutionalized, the

paradigmatic example of a public charge. Id. at 324. The only issue in the case

was whether she could escape deportation by reimbursing the state for the

services received. Id. at 326. The BIA held that a non-citizen is not deportable as a

public charge unless the state has asked for reimbursement and the non-citizen or

her relatives have failed to pay. Id. at 325. Rather than broadening the definition

of the public charge ground of inadmissibility – or saying anything that casts

doubt on other cases suggesting that a “public charge” must be persistently and

primarily dependent on the government – Matter of B- held that even an

29
In any event, even if these cases could be read to support DHS’s proposition,
they would not outweigh the prior or subsequent caselaw – particularly the
agency decisions from the 1960s and 1970s setting out nationally binding public
charge standards – endorsing a different interpretation of “public charge.”

72
immigrant who had been institutionalized at public expense because she was

unable to care for herself and was likely to require permanent hospitalization, still

was not categorically a public charge if the state had not sought payment and

been unable to collect.

Matter of B- thus offers a procedural escape hatch to those who need

government services but have money to pay. While the decision alters the

mechanics of deportation as a public charge, it hardly presents a new

interpretation of the term “public charge” itself.30 It seems particularly odd to cite

this somewhat unusual case, with its generous treatment of a non-citizen who

might well seem to fall within the established meaning of “public charge,” in

support of a sweeping expansion of that category.

DHS next argues that a series of policy statements enacted as part of

PRWORA show that the Rule’s interpretation is consistent with congressional

30
Moreover, in Matter of Harutunian, the BIA held that the Matter of B- test is
limited to the public charge ground of deportation and should not be read into
the public charge ground of inadmissibility. 14 I. & N. Dec. at 589-90. This
distinction between the public charge grounds of inadmissibility and deportation
was affirmed by INS in its 1999 Guidance, where it explained that while “the
definition of public charge is the same for both admission/adjustment and
deportation, the standards applied to public charge adjudications in each context
are significantly different.” 64 Fed. Reg. at 28,689.

73
intent regarding the meaning of “public charge.” See 8 U.S.C. § 1601 (describing

the “national policy with respect to welfare and immigration”). In the policy

statements, which lay out the rationale for enacting restrictions on non-citizen

eligibility for public benefits, Congress emphasized that “[s]elf-sufficiency has

been a basic principle of United States immigration law since this country’s

earliest immigration statutes” and affirmed that

[i]t continues to be the immigration policy of the United
States that (A) aliens within the Nation’s borders not
depend on public resources to meet their needs, but rather
rely on their own capabilities and the resources of their
families, their sponsors, and private organizations, and (B)
the availability of public benefits not constitute an
incentive for immigration to the United States.

Id. § 1601(1), (2). The policy statements further explain that PRWORA creates

“new rules for eligibility and sponsorship agreements in order to assure that

aliens be self-reliant in accordance with national immigration policy.” Id.

§ 1601(5). The policy statements conclude by noting that any state adopting the

federal benefits eligibility scheme laid out in PRWORA “shall be considered to

have chosen the least restrictive means available for achieving the compelling

governmental interest of assuring that aliens be self-reliant in accordance with

national immigration policy.” Id. § 1601(7).

74
 DHS reads these policy statements to mean that “Congress expressly

equated a lack of ‘self-sufficiency’ with the receipt of ‘public benefits’” and that

Congress intended “public charge” to mean “individuals who rely on taxpayer-

funded benefits to meet their basic needs.” Appellants’ Br. at 30-31. We are

thoroughly unpersuaded by this argument. PRWORA implemented Congress’s

goal of self-sufficiency by restricting non-citizen eligibility for benefits, including

the establishment of the five-year waiting period for LPRs. PRWORA did not

eliminate non-citizen eligibility for benefits nor does it suggest that such drastic

action is necessary. Still less does it indicate any congressional intention that non-

citizens who receive the benefits for which Congress did not render them

ineligible risk being considered “public charges.” On the contrary, the policy

statements specifically proclaim that the new eligibility restrictions sufficiently

“achiev[ed] the compelling governmental interest of assuring that aliens be self-

reliant in accordance with national immigration policy.” Id. § 1601(7) (emphasis

added). Clearly, Congress decided that the benefits it preserved for non-citizens

after PRWORA did not interfere with its interest in assuring non-citizen self-

sufficiency. Rather than supporting DHS’s expanded interpretation, both the

policy statements, taken as a whole, and the actual implementation of those

75
policy goals in the substantive provisions of PRWORA, are in considerable

tension with the Rule’s new interpretation of “public charge,” which penalizes

non-citizens for the possibility that they will access the very benefits PRWORA

preserved for them.31

DHS attempts to salvage this argument by pointing out that the 1999

Guidance made various cash benefits relevant to the public charge analysis,

notwithstanding that PRWORA also preserved non-citizen eligibility for those

benefits. DHS argues that this shows that Congress did not intend to preclude the

agency from considering the receipt of PRWORA-approved benefits in the public

charge determination. DHS is correct that the 1999 Guidance makes “receipt of

public cash assistance for income maintenance” one of two ways “primary

dependence” on the government could be shown. 64 Fed. Reg. at 28,689. But the

Guidance was also clear that receipt of such benefits alone was insufficient to

establish dependency, and that any such receipt needed to be weighed in the

31
We note that in its decision on DHS’s motion to stay the preliminary
injunctions, the Ninth Circuit accepted DHS’s argument on this point. San
Francisco, 944 F.3d at 799. The Ninth Circuit based its analysis on the first two
policy statements but did not consider the impact of the subsequent statements in
which Congress explained that the PRWORA eligibility scheme satisfied its
notions of self-sufficiency. Id.

76
context of the non-citizen’s overall circumstances. The Guidance explicitly noted

that “an alien receiving a small amount of cash for income maintenance purposes

could be determined not likely to become a public charge due to other positive

factors under the totality of the circumstances test.” 64 Fed. Reg. at 28,690. While

the 1999 Guidance permissibly looked at receipt of cash benefits as one factor

indicating dependence on the government, the Rule elevates receipt of any

quantity of a broad list of benefits to be the very definition of “public charge.” See

84 Fed. Reg. at 41,295.

The question under consideration is whether the Rule’s understanding of

the term “public charge” goes beyond the bounds of the settled meaning of the

term. The Plaintiffs do not argue, and we do not hold, that the receipt of various

kinds of public benefits is irrelevant to the determination of whether a non-

citizen is likely to become a public charge. But defining public charge to mean the

receipt, even for a limited period, of any of a wide range of public benefits –

particularly, as we discuss below, ones that are designed to supplement an

individual’s or family’s efforts to support themselves, rather than to deal with

their likely permanent inability to do so – is inconsistent with the traditional

77
understanding of what it means to be a “public charge,” which was well-

established by 1996.

Finally, DHS points to three other statutory provisions to support its

argument that the Rule is consistent with the intent of Congress. First, DHS

points to 8 U.S.C. § 1182(s), which exempts from the public charge analysis “any

benefits” received by a non-citizen who qualified for such benefits as a survivor

of domestic violence. See 8 U.S.C. § 1641(c). DHS argues that because § 1182(s)

excuses any benefits received, Congress understood that past receipt of even non-

cash benefits would otherwise generally be relevant to a public charge

determination. But 8 U.S.C. § 1182(s) was added in 2000, shortly after INS issued

its 1999 Guidance in which it clarified that benefits like TANF and SSI would be

relevant for the public charge determination. See Victims of Trafficking and

Violence Protection Act of 2000, Pub. L. No. 106-386, § 1505(f), 114 Stat. 1464,

1526. Without § 1182(s), survivors of domestic violence could thus have had their

receipt of cash benefits used against them. By far the most natural reading of

§ 1182(s) is that Congress was preventing domestic violence victims from being

penalized under the then-existing framework. In any event, the statute concerns

what is relevant to the determination, and gives no indication that Congress

78
somehow understood that the receipt of the benefits covered by the 1999

Guidance, let alone a broader set of benefits, could categorically render non-

citizens who were not domestic violence survivors “public charges.” Once again,

what is impermissible in DHS’s interpretation is not that it renders receipt of

supplemental non-cash benefits relevant to a non-citizen’s classification as a

public charge, but rather that it makes the receipt of such benefits determinative.

Second, DHS argues that the provisions requiring affidavits of support for

family-based immigrants and allowing the government to seek reimbursement

from the sponsor for any means-tested public benefit used by the non-citizen

support its interpretation. DHS contends that these provisions show that

Congress considered any non-citizen who might receive an unreimbursed public

benefit in the future a likely public charge. See 8 U.S.C. §§ 1182(a)(4)(c)(ii),

1183a(b)(1)(A). We are not convinced that the affidavit reimbursement

mechanism shows congressional intent to broaden the meaning of “public

charge.” For one thing, not all immigrants have to provide affidavits of support;

the requirement is limited to family-based immigrants and we see no reason it

79
should be taken to alter the underlying terms that apply to all non-citizens.32 We

also note that the statute includes a corollary, allowing the non-citizen herself to

take the sponsor to court if the sponsor fails to support the non-citizen as

promised. See id. § 1183a(a)(1)(B). The reimbursement provision thus serves

primarily as a mechanism to get sponsors to take their commitments seriously by

making them legally enforceable, a longstanding point of concern. See S. REP. NO.

81-1515, at 347 (Senate Judiciary Committee Report from 1950 critiquing the

affidavit of support as an unenforceable document that “at most, appears to be

merely a moral obligation upon the affiant”).

Third and last, DHS looks to the 1986 Immigration Reform and Control Act

(“IRCA”) to support its claim that the Rule is consistent with congressional

intent. IRCA established an ad hoc legalization program for undocumented

immigrants. See 8 U.S.C. § 1255a. To qualify for legalization, applicants needed to

prove that most of the grounds of inadmissibility did not apply to them,

including the public charge ground. See id. § 1255a(a)(4), (d)(2). However, IRCA

established a “special rule” with respect to the public charge inquiry, under

32
The affidavit of support requirement is also applied to a small subset of
employment-based immigrants, where the non-citizen’s prospective employer is
a relative or an entity owned in large part by a relative. 8 U.S.C. § 1182(a)(4)(D).

80
which a non-citizen would not be deemed a likely public charge “if the alien

demonstrates a history of employment in the United States evidencing self-

support without receipt of public cash assistance.” Id. § 1255a(d)(2)(B)(iii). DHS

argues that because the IRCA special rule specifically incorporates only cash

assistance, the generic ground in § 1182(a)(4) must necessarily have a broader

reach.

The implementing regulations of IRCA, however, suggest that, rather than

refining the benefits relevant to the public charge inquiry, the special rule

allowed a non-citizen who may otherwise be deemed a likely public charge

because of his limited financial resources an additional manner of showing self-

sufficiency. The relevant regulations provide that a non-citizen “who has a

consistent employment history which shows the ability to support himself or

herself even though his or her income may be below the poverty level, may be

admissible.” 8 C.F.R. § 245a.2(k)(4). Accordingly, even if an applicant was

“determined likely to become a public charge[,]” adjudicators were to find a non-

citizen inadmissible on this ground only if he was “unable to overcome this

determination after application of the special rule” and consideration of his

employment history. Id. § 245a.2(d)(4). The BIA applied the IRCA special rule in

81
Matter of A-, reversing a public charge finding that put too much weight on the

applicant’s “financial circumstances” in the face of “the more important factors;

namely, that the applicant has now joined the work force, . . . is young, and . . .

has no physical or mental defects which might affect her earning capacity.” 19 I.

& N. Dec. at 870. In other words, the BIA read the IRCA special rule as fully

consistent with the long-standing view that the ultimate issue in defining a

“public charge” is the non-citizen’s anticipated ability, over a protracted period,

to be able to work to support himself or herself. IRCA and its implementing

regulations thus show that Congress continued to emphasize capacity for work

as a core element of the public charge ground.

All three of these statutory arguments share a common flaw. DHS attempts

to justify a sweeping redefinition of “public charge” by pointing to tangential

details within the extensive patchwork that makes up American immigration law

– none of which express any intention by Congress to revise or depart from the

settled meaning of the term “public charge.” DHS’s argument that these statutory

provisions are “consistent” with its interpretation is of no relevance. The question

is whether, by passing these statutes, Congress undertook to change the

long-established meaning of public charge. While the statutes to which DHS

82
points may be “consistent” with the meaning DHS has assigned to public charge,

they are no less consistent with the long established meaning of public charge

that DHS seeks to overturn. These enactments do nothing to demonstrate that

Congress changed the meaning of public charge. The arguments thus “run[ ]

afoul of the usual rule that Congress does not alter the fundamental details of a

regulatory scheme in vague terms or ancillary provisions – it does not, one might

say, hide elephants in mouseholes.” Epic Sys., 138 S. Ct. at 1626-27 (internal

quotation marks omitted).

We conclude that the Plaintiffs have demonstrated a likelihood of success

on the merits of their argument that the Rule is contrary to the INA. In reenacting

the public charge ground in 1996, Congress endorsed the settled administrative

and judicial interpretation of that ground as requiring a holistic examination of a

non-citizen’s self-sufficiency focused on ability to work and eschewing any idea

that simply receiving welfare benefits made one a public charge. The Rule makes

receipt of a broad range of public benefits on even a short-term basis the very

definition of “public charge.” That exceedingly broad definition is not in

accordance with the law. See 5 U.S.C. § 706(2)(A).

83
 C. The Rule is Arbitrary and Capricious.

We next consider whether the Plaintiffs are likely to succeed on the merits

of their argument that the Rule is arbitrary and capricious. See id. “The scope of

review under the ‘arbitrary and capricious’ standard is narrow and a court is not

to substitute its judgment for that of the agency.” State Farm, 463 U.S. at 43. But

“[t]his is not to suggest that judicial review of agency action is merely

perfunctory. To the contrary, within the prescribed narrow sphere, judicial

inquiry must be searching and careful.” Islander E. Pipeline Co., LLC v. McCarthy,

525 F.3d 141, 151 (2d Cir. 2008) (internal quotation marks omitted).

“When an administrative agency sets policy, it must provide a reasoned

explanation for its action. This is not a high bar, but it is an unwavering one.”

Judulang v. Holder, 565 U.S. 42, 45 (2011). The Plaintiffs argue that the Rule is

arbitrary and capricious because DHS has not provided a reasoned explanation

for its changed definition of “public charge” or the Rule’s expanded list of

relevant benefits. DHS contends that it has adequately explained its action,

stating that it adopted its new public charge definition to “improve upon” the

1999 Guidance by “aligning public charge policy with the self-sufficiency

principles set forth in [PRWORA].” 83 Fed. Reg. at 51,123; see also 84 Fed. Reg. at

84
41,319-20. DHS further explains that it expanded the list of relevant benefits

because the 1999 Guidance relied on an “artificial distinction between cash and

non-cash benefits” that is not warranted under DHS’s new definition. 83 Fed.

Reg. at 51,123; see 64 Fed. Reg. at 28,689. For the reasons laid out below, we agree

with the district court that the Plaintiffs are likely to succeed on the merits of

their claim that the Rule is arbitrary and capricious because neither rationale is a

“satisfactory explanation” for DHS’s actions.33 State Farm, 463 U.S. at 43.

1. Explanation for Changed Definition

DHS justifies its revised definition of “public charge” – one who uses a

relevant public benefit for more than twelve months in the aggregate – as a

“superior interpretation of the statute to the 1999 Interim Field Guidance”

because it “furthers congressional intent behind both the public charge

inadmissibility statute and PRWORA in ensuring that aliens . . . be self-sufficient

and not reliant on public resources.” 84 Fed. Reg. at 41,319. “In fact, DHS believes

it would be contrary to congressional intent to promulgate regulations that . . .

33
Because we find the Plaintiffs likely to succeed on this basis, we do not address
the Plaintiffs’ additional contentions that we could find the Rule arbitrary and
capricious based on its aggregation principle, selection of factors indicative of
future benefits use, or cost-benefit analysis.

85
ignore the[] receipt” of the benefits listed in the Rule “as this would be contrary

to Congress’s intent in ensuring that aliens within the United States are self-

sufficient.” Id. at 41,318 (citing the PRWORA policy statements at 8 U.S.C.

§ 1601(2)(A)); see, e.g., id. at 41,295, 41,305, 41,308. In short, DHS justifies its

changed interpretation as necessary to implement Congress’s view that “the

receipt of any public benefits, including noncash benefits, [is] indicative of a lack

of self-sufficiency.” Appellants’ Br. at 43.

This explanation fails for the same reasons as DHS’s related argument that

the PRWORA policy statements show that the Rule is consistent with Congress’s

intended meaning of “public charge.” See supra Section II.B.6. As we discussed

above, the PRWORA policy statements do show a congressional interest in

ensuring non-citizen self-sufficiency. See 8 U.S.C. § 1601(1), (2). But the statements

also show that, contrary to DHS’s belief, Congress’s vision of self-sufficiency does

not anticipate abstention from all benefits use. See Cook Cty., 962 F.3d at 232

(rejecting DHS’s “absolutist sense of self-sufficiency that no person in a modern

society could satisfy”). Rather, Congress realized its notion of self-sufficiency

with a new benefits eligibility scheme that greatly reduced – but did not

eliminate – non-citizen eligibility for public benefits. See 8 U.S.C. § 1601(7)

86
(describing the PRWORA eligibility scheme as “achieving the compelling

governmental interest of assuring that aliens be self-reliant in accordance with

national immigration policy”). “The Supreme Court and [other] court[s] have

consistently reminded agencies that they are bound, not only by the ultimate

purposes Congress has selected, but by the means it has deemed appropriate,

and prescribed, for the pursuit of those purposes.” Gresham v. Azar, 950 F.3d 93,

101 (D.C. Cir. 2020) (internal quotation marks omitted).

Had Congress thought that any benefits use was incompatible with self-

sufficiency, it could have said so, either by making non-citizens ineligible for all

such benefits or by making those who did receive them inadmissible. But it did

not. We are thus left with an agency justification that is unmoored from the

nuanced views of Congress. See Yale-New Haven Hosp. v. Leavitt, 470 F.3d 71, 85-86

(2d Cir. 2006) (finding agency failed to provide reasoned explanation as to “how

adoption of a per se coverage standard comports with congressional purposes in

enacting the Medicare Act,” which prioritized individualized care

determinations). As the Supreme Court has explained,

no legislation pursues its purposes at all costs. Deciding
what competing values will or will not be sacrificed to the
achievement of a particular objective is the very essence of

87
 legislative choice – and it frustrates rather than effectuates
legislative intent simplistically to assume that whatever
furthers the statute’s primary objective must be the law.

Rodriguez v. United States, 480 U.S. 522, 525-26 (1987) (emphasis omitted).

DHS’s misconception of the PRWORA policy statements and Congress’s

intended notion of self-sufficiency is its principal justification for its revised

definition; it identifies no other “deficienc[y]” in the 1999 Guidance, apart from

its limited list of relevant benefits, discussed below. See 84 Fed. Reg. at 41,319; see

also id. at 41,349 (describing the Guidance’s interpretation as “suboptimal when

considered in relation to the goals of the INA and PRWORA”).

To be sure, we do not suggest that DHS must, as a general matter, show

that the Guidance was deficient or that the Rule is necessarily a better

interpretation than the prior policy reflected in the Guidance to avoid being

found arbitrary and capricious. See FCC v. Fox Television Stations, Inc., 556 U.S.

502, 515 (2009) (clarifying that agencies are not required to show “that the reasons

for the new policy are better than the reasons for the old one”). Nor do we suggest

that, when an agency offers a statutory interpretation as part of its reason for

adopting a policy, and a reviewing court later rejects the agency’s statutory

interpretation, that the policy is per se arbitrary and capricious. But where, as

88
here, DHS anchors its decision to change its interpretation in the perceived

shortcomings of the prior interpretation, and then fails to identify any actual

defect, it has not provided a “reasoned explanation” for its actions – particularly

when it bases its changed position on its reading of a statute, and it is the new

Rule, rather than the old Guidance, that strays from congressional intent. Encino

Motorcars, LLC v. Navarro, 136 S. Ct. 2117, 2125 (2016).

2. Explanation for Expanded List of Benefits

This brings us to DHS’s rationale for expanding the list of benefits relevant

to the public charge determination. DHS explains that it included a broader

group of benefits in the Rule because the distinction made in the 1999 Guidance

between cash and non-cash benefits was no longer appropriate in light of the

more restrictive notions of self-sufficiency DHS enacted with the changed

definition. See 84 Fed. Reg. at 41,356; see also id. at 41,349, 41,351, 41,375; 83 Fed.

Reg. at 51,123. Though this explanation is in some ways subsidiary to DHS’s

explanation for the changed definition, DHS argues this as an additional

justification and we thus address its additional shortcomings. See Appellants’ Br.

at 43.

89
 In the 1999 Guidance, INS explained that “[a]fter extensive consultation

with benefit-granting agencies” it “determined that the best evidence of whether

an alien is primarily dependent on the government for subsistence is . . . the

receipt of public cash assistance for income maintenance.” 64 Fed. Reg. at 28,692;

see 83 Fed. Reg. 51,133. The Guidance consequently excluded non-cash benefits

(e.g., SNAP, housing assistance, and Medicaid) from consideration because those

benefits were “increasingly being made available to families with incomes far

above the poverty level, reflecting broad public policy decisions about improving

general public health and nutrition.” 64 Fed. Reg. at 28,692. In other words,

“participation in [those] programs [was] not evidence of poverty or dependence”

because they are “by their nature supplemental and do not, alone or in

combination, provide sufficient resources to support an individual or family.” Id.;

see id. at 28,678.

In justifying its decision to include these non-cash benefits in the Rule,

DHS explains that they are relevant to its revamped public charge definition

because they “bear directly on self-sufficiency.” 84 Fed. Reg. at 41,366. DHS

reasons that because “[f]ood, shelter, and necessary medical treatment are basic

necessities of life[, a] person who needs the public’s assistance to provide for

90
these basic necessities is not self-sufficient.” 83 Fed. Reg. at 51,159. Thus, the Rule

includes these benefits as relevant to the public charge determination to ensure

that all benefits bearing on self-sufficiency are considered. Id.; see 84 Fed. Reg. at

41,356.

The fundamental flaw of this justification is that while DHS repeatedly

contends that the non-citizens using these programs would be unable to provide

for their basic necessities without governmental support, it does not provide any

factual basis for this belief. See, e.g., 83 Fed. Reg. at 51,159; 84 Fed. Reg. at 41,354,

41,366, 41,375, 41,381, 41,389. While the 1999 Guidance was developed in

consultation with the benefits-granting agencies, DHS does not claim that their

expertise again informed its decision that people who use non-cash benefits

would be otherwise unable to meet their basic needs.34 Of course, DHS is free to

change its interpretation and we do not suggest it is under any obligation to

consult with its sister agencies in so doing. But what DHS may not do is rest its

changed interpretation on unsupported speculation, particularly when its

categorical assumptions run counter to the realities of the non-cash benefits at

34
In response to a comment directly asking whether any such consultation took
place, DHS invoked the deliberative process privilege. 84 Fed. Reg. at 41,460.

91
issue. The goals and eligibility criteria of these benefits programs belie DHS’s

assumption and show that these programs are designed to provide supplemental

support, rather than subsistence, to a broad swath of the population – as INS

recognized in 1999.

Take, for example, SNAP – the Supplemental Nutrition Assistance Program

– which was born of a desire to “raise levels of nutrition among low-income

households.” Food Stamp Act of 1964, Pub. L. No. 88-525, § 2, 78 Stat. 703, 703.

SNAP benefits are intended for all those whose “financial resources . . . are

determined to be a substantial limiting factor in permitting them to obtain a more

nutritious diet.” Food and Agriculture Act of 1977, Pub. L. No. 95-113, § 1301, 91

Stat. 913, 962 (emphasis added); see 7 C.F.R. § 273.9(a). Because SNAP is not

intended only for those who might otherwise face starvation, the program is

open to households with incomes exceeding the federal poverty guideline, 7

C.F.R. § 273.9(a)(1), and its supplemental nature is underscored by the fact that

the average SNAP recipient receives only $127 a month in benefits, see House of

Representatives Amicus Br. at 19 (citing 2018 statistics). Large numbers of SNAP

92
recipients, far from being incapable of productive employment, work for some of

America’s largest corporations.35

The housing benefits included in the Rule have a similar aim, intended to

ensure “a decent home and a suitable living environment for all persons, but

principally those of low and moderate income.” Housing and Community

Development Act of 1974, Pub. L. No. 93-383, § 101(c)(3), 88 Stat. 633, 634

(emphasis added). Indeed, while the majority of those using housing programs

are low-income families, benefits remain available to those earning up to 80% of

the area median income – $85,350 for a family of four in New York City in 2019.

M.T.R. J. App. 164; see 42 U.S.C. § 1437a(b)(2)(A). It makes little sense to treat the

mere receipt of housing benefits as proof of inability to survive by one’s own

efforts when the program is intended for, among others, people who can and do

earn moderate incomes. In contrast, TANF – one of the three benefits listed in the

35
See Public Justice Center Amicus Br. at 12 (citing Senate report concluding that
SNAP beneficiaries are “‘far more’ likely to be employed than to rely on cash
assistance” (quoting S. Rep. No. 11-220, at 8 (2007)); see also Dennis Green, Data
From States Shows Thousands of Amazon Employees Are on Food Stamps, BUSINESS
INSIDER (Aug. 25, 2018) (discussing SNAP usage by Amazon, Walmart, and
McDonald’s employees).

93
1999 Guidance – is generally only available to families with incomes well below

the federal poverty guideline.36

While the Rule declares non-citizens dependent for using Medicaid instead

of private health insurance, it cannot be ignored that in this country, access to

private healthcare depends for many people on whether an employer offers

coverage. See National Housing Law Project Amicus Br. at 22 (noting that

roughly 40% of employed Medicaid beneficiaries work for small businesses,

many of which are not legally required to provide health insurance). Considering

that access to insurance is often determined by factors beyond an individual’s

control, we are dubious of DHS’s unsupported claim that using public health

insurance shows a lack of self-sufficiency.37 To the contrary, studies show that

36
The TANF earnings thresholds for new applicants vary by state and range from
approximately 16% of FPG in Alabama to 91% of FPG in Nevada. See CONG.
RESEARCH SERV., TANF: ELIGIBILITY AND BENEFIT AMOUNTS IN STATE TANF CASH
ASSISTANCE PROGRAMS at 3 (2014). In the majority of states, however, TANF was
only available to those earning less than 50% of FPG, which means an annual
income of less than $13,100 for a family of four in 2020. Id.; see Annual Update of
the HHS Poverty Guidelines, 85 Fed. Reg. 3,060, 30,060 (Jan. 17, 2020).
37
DHS also suggests that Medicaid is included because “the total Federal
expenditure for the Medicaid program overall is by far larger than any other
program for low-income people,” 84 Fed. Reg. at 41,379, which DHS takes as
evidence that it is “a more significant form of public support” for individuals
than other benefits, Appellants’ Br. at 43; see 83 Fed. Reg. at 51,160. We are not

94
more than 60% of Medicaid beneficiaries who are not children, older adults, or

people with disabilities are employed. See Public Justice Center Amicus Br. at 20

(citing RACHEL GARFIELD ET AL., KAISER FAMILY FOUND., UNDERSTANDING THE

INTERSECTION OF MEDICAID AND WORK: WHAT DOES THE DATA SAY? 2 (2019)). To

be sure, it is easier for individuals to purchase private coverage in the wake of the

Affordable Care Act (“ACA”), but the Rule implies that even using ACA tax

credits to purchase health insurance evidences an inability to meet one’s needs

without government support. 84 Fed. Reg. at 41,299.

Of course, SNAP and housing benefits may very well be all that stands

between some non-citizens and hunger or homelessness. Some families may

actually fail to meet these basic needs without government support. But these

programs sweep more broadly than just families on the margin, encompassing

those who would no doubt keep their families fed and housed without

persuaded that the difference in dollars expended is an appropriate indicator of a
non-citizen’s level of self-sufficiency; rather, it seems plain to us that the
difference is due to the high cost of providing healthcare in the United States. Cf.
Public Citizen, Inc. v. Mineta, 340 F.3d 39, 58 (2d Cir. 2003) (“The notion that
‘cheapest is best’ is contrary to State Farm.”). The size of the government
expenditure on Medicaid may be relevant to a policy debate about the costs and
benefits of the program, but it has little bearing on whether Medicaid recipients
should be considered “public charges.”

95
government support but are able to do so in a healthier and safer way because

they receive supplemental assistance. See Cook Cty., 962 F.3d at 232 (noting that

the benefits covered by the Rule “are largely supplemental” and that “[m]any

recipients could get by without them” (emphasis omitted)). Accepting help that is

offered to elevate one to a higher standard of living, help that was created by

Congress for that precise purpose, does not mean a person is not self-sufficient –

particularly when such programs are available not just to persons living in abject

poverty but to a broad swath of low- and moderate-income Americans, including

those who are productively employed. DHS goes too far in assuming that all

those who participate in non-cash benefits programs would be otherwise unable

to meet their needs and that they can thus be categorically considered “public

charges.” Its unsupported and conclusory claim that receipt of such benefits

indicates an inability to support oneself does not satisfy DHS’s obligation to

explain its actions. See Gen. Chem. Corp. v. United States, 817 F.2d 844, 855 (D.C.

Cir. 1987) (rejecting agency’s “conclusory” explanation and noting that “[s]uch

intuitional forms of decisionmaking . . . fall somewhere on the distant side of

arbitrary” (internal quotation marks omitted)); see also State Farm, 463 U.S. at 51.

96
 “Agencies are free to change their existing policies as long as they provide

a reasoned explanation for the change. When an agency changes its existing

position, it need not always provide a more detailed justification than what

would suffice for a new policy created on a blank slate. But the agency must at

least . . . show that there are good reasons for the new policy.” Encino Motorcars,

136 S. Ct. at 2125-26 (internal quotation marks and citations omitted). DHS has

failed to do so here. Accordingly, the Plaintiffs have shown they are likely to

succeed on the merits of their claim that DHS’s failure to provide a reasoned

explanation renders the Rule arbitrary and capricious.

III. Irreparable Harm to the Plaintiffs

The second preliminary injunction factor under Winter requires the

Plaintiffs to show they are likely to suffer irreparable harm in the absence of

injunctive relief. 555 U.S. at 20. “Irreparable harm is injury that is neither remote

nor speculative, but actual and imminent and that cannot be remedied by an

award of monetary damages.” New York ex rel. Schneiderman v. Actavis PLC, 787

F.3d 638, 660 (2d Cir. 2015) (internal quotation marks omitted). We have already

discussed the Plaintiffs’ claimed injuries in evaluating their standing to challenge

the Rule and both the States and Organizations point to largely similar harms to

97
establish this injunctive factor. See League of Women Voters of the United States v.

Newby, 838 F.3d 1, 9 (D.C. Cir. 2016) (looking to same injuries to establish

standing and irreparable harm).

The States contend that the implementation of the Rule will result in

reduced Medicaid revenue and federal funding and a greater number of

uninsured patients seeking care, putting public hospitals that are already

insufficiently funded at risk of closure. See, e.g., N.Y. J. App. 512-13. Additionally,

as the administrators of the benefits programs at issue, the States allege that they

will be required to undertake costly revisions to their eligibility systems to ensure

that non-citizens are not automatically made eligible for or enrolled in benefits

they may no longer wish to receive after the Rule’s implementation. See, e.g., N.Y.

J. App. 236, 381-82. The Organizations point to the economic harms of expending

funds to mitigate the impact of the Rule on the communities they serve. See, e.g.,

M.T.R. J. App. 466-67. As noted, DHS predicted that the Rule would have

economic harms, and the Rule has already had a chilling effect on non-citizen use

of public benefits. See supra Section I.A. These injuries claimed by the States and

the Organizations are actual and imminent. Moreover, because money damages

are prohibited in APA actions, they are irreparable. See 5 U.S.C. § 702; Ward v.

98
Brown, 22 F.3d 516, 520 (2d Cir. 1994). We thus conclude that the Plaintiffs have

established the second factor of the preliminary injunction standard.38

IV. Balance of Equities and the Public Interest

The final inquiry in our preliminary injunction analysis requires us to

consider whether the balance of equities tips in favor of granting the injunction

and whether that injunction is in the public interest, the third and fourth Winter

factors. Winter, 555 U.S. at 20; Azar, 911 F.3d at 575 (considering the final two

factors together where the government is a party). DHS argues that it would be

harmed by a preliminary injunction because an injunction would force the

agency to retain its prior policy, which grants status to some non-citizens that

DHS believes should be denied under a proper interpretation of the public

charge ground. Because there is no apparent means by which DHS could revisit

adjustment determinations made while the Rule is enjoined, this harm is

irreparable.

38
We note that our precedents suggest that the Plaintiffs may be able to show that
a preliminary injunction is warranted on the strength of these first two factors
alone. See Trump v. Deutsche Bank AG, 943 F.3d 627, 636, 640-41 (2d Cir. 2019),
rev’d on other grounds, – U.S. – 2020 WL 3848061 (July 9, 2020). Notwithstanding
this possibility, we consider the balance of equities and the public interest, as
discussed in Winter.

99
 While DHS has a valid interest in applying its preferred immigration

policy, we think the balance of equities clearly tips in favor of the Plaintiffs. For

one, DHS’s claimed harm is, to some extent, inevitable in the preliminary

injunction context. Any time the government is subject to a preliminary

injunction, it necessarily suffers the injury of being prevented from enacting its

preferred policy. Without additional considerations at play – for example,

national security implications, Winter, 555 U.S. at 26, or the need to correct a

previous policy that had been deemed unlawful – we do not think DHS’s

inability to implement a standard that is as strict as it would like outweighs the

wide-ranging economic harms that await the States and Organizations upon the

implementation of the Rule.

The public interest also favors a preliminary injunction. DHS itself

acknowledges that the Rule will likely result in “[w]orse health outcomes,

including increased prevalence of obesity and malnutrition, . . . [i]ncreased

prevalence of communicable diseases, . . . [i]ncreased rates of poverty and

housing instability[,] and [r]educed productivity and educational attainment.” 83

Fed. Reg. at 51,270. To say the least, the public interest does not favor the

immediate implementation of the Rule.

100
 Thus, the Plaintiffs have met their burden of showing that a preliminary

injunction is warranted in these cases. Accordingly, we affirm the district court

orders granting such relief in these cases.

V. Scope of Injunction

While we hold that the district court properly granted the Plaintiffs’

preliminary injunction motions, there remains one final issue for our

consideration: whether the district court abused its discretion by entering a

nationwide injunction, rather than a geographically limited measure. DHS argues

that a national injunction is insufficiently tailored to the Plaintiffs’ particular

injuries and allows the decision of a single district court to override contrary

decisions of other courts, an outcome not warranted by the need for uniform

application of immigration law. The Plaintiffs respond that the scope of relief is

determined by the extent of the violation and that the APA authorizes the broad

relief issued here.

The issuance of nationwide injunctions has been the subject of increasing

scrutiny in recent years, a topic that has already touched these cases on their brief

foray to the Supreme Court. See New York, 140 S. Ct. at 599-601 (Gorsuch, J.,

concurring in the grant of stay); see also Trump v. Hawaii, 138 S. Ct. 2392, 2424-29

101
(2018) (Thomas, J., concurring). The difficult questions implicated in this debate

are evidenced by the fact that both DHS and the Plaintiffs marshal persuasive

points to support their arguments. As the Plaintiffs point out, courts have long

held that when an agency action is found unlawful under the APA, “the ordinary

result is that the rules are vacated – not that their application to the individual

petitioners is proscribed.” Nat’l Mining Ass’n v. U.S. Army Corps of Eng’rs, 145

F.3d 1399, 1409 (D.C. Cir. 1998) (internal quotation marks omitted). This aligns

with the general principle that “the scope of injunctive relief is dictated by the

extent of the violation established, not by the geographical extent of the plaintiff

class.” Califano v. Yamasaki, 442 U.S. 682, 702 (1979). Moreover, courts have

recognized that nationwide injunctions may be particularly appropriate in the

immigration context, given the interest in a uniform immigration policy. See

Texas v. United States, 809 F.3d 134, 187-88 (5th Cir. 2015); see also Hawaii v. Trump,

878 F.3d 662, 701 (9th Cir. 2017), rev’d on other grounds, 138 S. Ct. 2392 (2018).

On the other hand, we share DHS’s concern that a district judge issuing a

nationwide injunction may in effect override contrary decisions from co-equal

and appellate courts, imposing its view of the law within the geographic

jurisdiction of courts that have reached contrary conclusions. That result may

102
well be more unseemly than the application of inconsistent interpretations of

immigration law across the circuits – a situation that is hardly unusual, and may

well persist without injustice or intolerable disruption. See, e.g., Orellana-Monson

v. Holder, 685 F.3d 511, 520 (5th Cir. 2012) (discussing circuit variance in a

substantive asylum standard).

We have no doubt that the law, as it stands today, permits district courts to

enter nationwide injunctions, and agree that such injunctions may be an

appropriate remedy in certain circumstances – for example, where only a single

case challenges the action or where multiple courts have spoken unanimously on

the issue. The issuance of unqualified nationwide injunctions is a less desirable

practice where, as here, numerous challenges to the same agency action are being

litigated simultaneously in district and circuit courts across the country. It is not

clear to us that, where contrary views could be or have been taken by courts of

parallel or superior authority, entitled to determine the law within their own

geographical jurisdictions, the court that imposes the most sweeping injunction

should control the nationwide legal landscape.

When confronted with such a volatile litigation landscape, we encourage

district courts to consider crafting preliminary injunctions that anticipate the

103
possibility of conflict with other courts and provide for such a contingency. Such

approaches could take the form of limiting language providing that the

injunction would not supersede contrary rulings of other courts, an invitation to

the parties to return and request modification as the situation changes, or the

limitation of the injunction to the situation of particular plaintiffs or to similarly

situated persons within the geographic jurisdiction of the court.

We need not decide whether the able district judge in these cases abused

his discretion in entering nationwide injunctions. Instead, we exercise our own

discretion, in light of the divergent decisions that have emerged in our sister

circuits since the district court entered its orders, to modify the injunction,

limiting it to the states of New York, Connecticut, and Vermont. Cf. Smith v.

Woosley, 399 F.3d 428, 436 (2d Cir. 2005). As modified, the injunction covers the

State plaintiffs and the vast majority of the Organizations’ operations. We see no

need for a broader injunction at this point, particularly in light of the somewhat

unusual posture of this case, namely that the preliminary injunction has already

been stayed by the Supreme Court, not only through our disposition of the case,

but also through the disposition of DHS’s petition for a writ of certiorari, should

DHS seek review of this decision. See New York, 140 S. Ct. at 599.

104
 CONCLUSION

For the reasons stated above, we agree with the district court that a

preliminary injunction is warranted in these cases but modify the scope of the

injunctions to cover only the states of New York, Connecticut, and Vermont. The

orders of the district court are therefore AFFIRMED AS MODIFIED.

105
 Appendix A

Amici Curiae in New York v. DHS Elizabeth B. Wydra, Brianne J.
and Make the Road v. Cuccinelli Gorod, Dayna J. Zolle, Constitutional
Accountability Center, Washington,
Maureen P. Alger, Priyamvada DC, for Amici Curiae Immigration
Arora, Cooley LLP, Palo Alto, CA, for History Scholars, in support of
Amici Curiae American Academy of Plaintiffs-Appellees.
Pediatrics, American Medical
Association, American College of Johanna Dennehy, Steptoe & Johnson
Physicians, American College of LLP, Washington, DC, for Amici
Obstetricians and Gynecologists, Curiae Immigration Law Professors,
New York State American Academy in support of Plaintiffs-Appellees.
of Pediatrics, American Academy of
Pediatrics – Vermont Chapter, and Richard L. Revesz, Jack Lienke, Max
Medical Society of the State of New Sarinsky, Institute for Policy Integrity
York, in support of Plaintiffs-Appellees. at New York University School of
Law, New York, NY, for Amicus
Emily Tomoko Kuwahara, Crowell & Curiae Institute for Policy Integrity at
Moring LLP, Los Angeles, CA, New York University School of Law,
Austin Sutta, Crowell & Moring LLP, in support of Plaintiffs-Appellees.
San Francisco, CA, for Amici Curiae
Asian Americans Advancing Justice| Russell L. Hirschhorn, Proskauer
AAJC, Asian American Legal Rose LLP, New York, NY, for Amici
Defense and Education Fund, Curiae Justice in Aging, American
National Women’s Law Center, and Society on Aging, Caring Across
40 Other Organizations, in support of Generations, Jewish Family Service
Plaintiffs-Appellees. of Los Angeles, Jewish Federations of
North America, National Asian
Hillary Schneller, Center for Pacific Center on Aging, National
Reproductive Rights, New York, NY, Council on Aging, National Hispanic
for Amicus Curiae Center for Council on Aging, MAZON, PHI,
Reproductive Rights, in support of and Center for Medicare Advocacy,
Plaintiffs-Appellees. in support of Plaintiffs-Appellees.

106
Nilda Isidro, Goodwin Procter LLP, McDermott Will & Emery LLP,
New York, NY, for Amici Curiae Washington, DC, for Amici Curiae 105
Members of Congressional Black Businesses and Organizations, in
Caucus, Congressional Hispanic support of Plaintiffs-Appellees.
Caucus, and Congressional Asian
Pacific American Caucus, in support
of Plaintiffs-Appellees.

Stuart Rossman, National Consumer
Law Center, Boston, MA, for Amici
Curiae National Consumer Law
Center, Legal Aid Justice Center,
Public Citizen, Inc., Consumer
Action, Equal Justice Society, Impact
Fund, Secure Justice, Media Alliance,
Americans for Financial Reform
Education Fund, and New Economy
Project, in support of Plaintiffs-
Appellees.

Paul J. Lawrence, Alanna E. Peterson,
Pacifica Law Group LLP, Seattle,
WA, for Amici Curiae Nonprofit Anti-
Domestic Violence and Sexual
Assault Organizations, in support of
Plaintiffs-Appellees.

Debra Gardner, Public Justice Center,
Baltimore, MD, for Amicus Curiae
Public Justice Center, in support of
Plaintiffs-Appellees.

Paul W. Hughes, Michael B.
Kimberly, Matthew A. Waring,

107
Amici Curiae in New York v. DHS Jacqueline Chávez, Foley Hoag LLP,
Boston, MA, Justin Lowe, Wendy
Sarah M. Ray, Kyle A. Virgien, Diana Parmet, Health Law Advocates, Inc.,
A. Aguilar, Charles F. Sprague, Boston, MA, for Amici Curiae Health
Latham & Watkins LLP, San Law Advocates, Inc. and Other
Francisco, CA, Tyce R. Walters, Organizations Interested in Public
Latham & Watkins LLP, Washington, Health, in support of Plaintiffs-
DC, for Amici Curiae American Civil Appellees.
Liberties Union, Center for Public
Representation, American R. Adam Lauridsen, Chessie Thacher,
Association of People with Victor H. Yu, Nicholas R. Green,
Disabilities, Association of University Keker, Van Nest & Peters LLP, San
Centers on Disabilities, Autistic Self Francisco, CA, for Amici Curiae
Advocacy Network, Civil Rights National Housing Law Project, Food
Education and Enforcement Center, Research & Action Center, Center for
Coelho Center for Disability Law, Law & Social Policy, National
Policy, and Innovation, Disability Education Association, Service
Rights Advocates, Disability Rights Employees International Union,
Education and Defense Fund, California League of United Latin
Disability Rights New York, Judge American Citizens, California Food
David L. Bazelon Center for Mental Policy Advocates, Center for the
Health Law, Little Lobbyists, Mental Study of Social Policy, Children’s
Health America, National HealthWatch, Comunidades
Association of Councils on Unidas/Communities United, First
Developmental Disabilities, National Focus on Children, Los Angeles
Council on Independent Living, Regional Food Bank, Mississippi
National Disability Rights Network, Center for Justice, National WIC
National Federation of the Blind, Association, National Low Income
New York Civil Liberties Union, The Housing Coalition, Prevention
Arc of the United States, and United Institute, Sant La Haitian
Spinal Association, in support of Neighborhood Center, South
Plaintiffs-Appellees. Carolina Appleseed Legal Justice
Center, Virginia Poverty Law Center,
Lisa C. Wood, Kristyn DeFilipp, in support of Plaintiffs-Appellees.
Andrew London, Emily Nash, E.

108
Matthew S. Freedus, Edward T. City of Houston, TX, Howard Phillip
Waters, Phillip A. Escoriaza, Schneiderman, King County, WA,
Amanda N. Pervine, Feldesman Michael P. May, City of Madison, WI,
Tucker Leifer Fidell LLP, Brian E. Washington, Marin County,
Washington, DC, for Amici Curiae CA, Erik Nilsson, City of
Public Health, Health Policy, Minneapolis, MN, Leslie J. Girard,
Medicine, and Nursing Deans, William M. Litt, Anne K. Brereton,
Chairs, and Scholars, American Marina S. Pantchenko, Monterey
Public Health Association, American County, CA, Kathryn E. Doi, Rachael
Academy of Nursing, and Public E. Blucher, Natalie M. Smith, Hanson
Health Solutions, in support of Bridgett LLP, Sacramento, CA,
Plaintiffs-Appellees. Susana Alcala Wood, City of
Sacramento, CA, John C. Beiers,
Danielle L. Goldstein, Michael David A. Silberman, Ilana Parmer
Dundas, Office of the Los Angeles Mandelbaum, San Mateo County,
City Attorney, Barbara J. Parker, Erin CA, Peter S. Holmes, City of Seattle,
Bernstein, Office of the Oakland City WA, Francis X. Wright, Jr., City of
Attorney, Margaret L. Carter, Daniel Somerville, MA, Michael Rankin,
R. Suvor, O’Melveny & Myers LLP, City of Tucson, AZ, Michael Jenkins,
Los Angeles, CA, Vince Ryan, Robert City of West Hollywood, CA, for
Hazeltine-Shedd, Harris County, TX, Amici Curiae 26 Cities and Counties,
Donna R. Ziegler, Alameda County, in support of Plaintiffs-Appellees.
CA, Esteban A. Aguilar, Jr., City of
Albuquerque, NM, Anne L. Morgan,
City of Austin, TX, Andre M. Davis,
City of Baltimore, MD, Mark A.
Flessner, Benna Ruth Solomon, City
of Chicago, IL, Christopher J. Caso,
City of Dallas, TX, Kristin M.
Bronson, City and County of Denver,
CO, Lawrence Garcia, Eli Savit, City
of Detroit, MI, Rodney Pol, Jr., City of
Gary, IN, Crystal Barnes, City of
Holyoke, MA, Ronald C. Lewis,
Judith L. Ramsey, Collyn Peddie,

109
Amici Curiae in Make the Road v.
Cuccinelli

Sadik Huseny, Brittany N. Lovejoy,
Joseph C. Hansen, Tess L. Curet,
Alexandra B. Plutshack, Latham &
Watkins LLP, San Francisco, CA, for
Amici Curiae Fiscal Policy Institute,
Presidents’ Alliance on Higher
Education and Immigration,
National Center for Law and
Economic Justice, American
Federation of State, County and
Municipal Employees, California
Immigrant Policy Center, Child Care
Law Center, Colorado Fiscal
Institute, Community Action Marin,
Kids Forward, Michigan Immigrant
Rights Center, Oasis Legal Services,
Economic Progress Institute, United
African Organization, and Virginia
Interfaith Center for Public Policy, in
support of Plaintiffs-Appellees.

Lawrence J. Joseph, Law Office of
Lawrence J. Joseph, Washington, DC,
for Amicus Curiae Immigration
Reform Law Institute, in support of
Defendants-Appellants.

110