Stewart v. Hargan ( 2019 )

UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF COLUMBIA
RONNIE MAURICE STEWART, et al.,
Plaintiffs,
v. Civil Action No. 18-152 (JEB)
ALEX M. AZAR II, et al.,
Defendants.
MEMORANDUM OPINION
This Court again takes up a challenge to the federal approval of Kentucky HEALTH, an
experimental project proposed by the Commonwealth of Kentucky intended to “comprehensively
transform” its Medicaid program. The Secretary of Health and Human Services has authority to
approve such experimental proposals — or “demonstration projects” — as long as they promote
the objectives of the Medicaid Act. Kentucky HEALTH, which the Secretary initially approved
on January 12, 2018, would condition Medicaid eligibility for a large portion of its beneficiaries
on work or community-engagement requirements and impose several additional obligations
intended to make Medicaid more like commercial insurance.
Plaintiffs, Kentucky residents currently enrolled in the Commonwealth’s Medicaid
program, believed HHS’s approval unlawful. In a ruling last summer, this Court agreed.
Finding that the “Secretary never adequately considered whether Kentucky HEALTH would in
fact help the state furnish medical assistance to its citizens” and thus promote a central objective
of the Medicaid Act, the Court concluded that this “signal omission render[ed] his determination
arbitrary and capricious.” Stewart v. Azar, 313 F. Supp. 3d 237, 243 (D.D.C. 2018). In
1
particular, it found that the Secretary had not grappled with Kentucky’s estimate that a
substantial number of people were likely to lose coverage under Kentucky HEALTH. Id. at 260.
The Court, consequently, vacated the approval and remanded to HHS for further review.
The bell now rings for round two. Following the Court’s remand and an additional
notice-and-comment period, the Secretary reapproved the program last November, this time
relying on somewhat different reasoning. Plaintiffs now challenge the reapproval, contending
principally that the Secretary has not remedied the defects that rendered his prior action
unlawful. Specifically, they maintain that he has still not adequately considered Kentucky
HEALTH’s likelihood to cause significant coverage loss. The Secretary, by contrast, believes
that this time around he has cured any critical omission. Defendants now rely primarily on a new
argument to that effect — namely that, although Kentucky HEALTH may cause nearly 100,000
people to lose coverage, that number will be dwarfed by the approximately 450,000 people who
would suffer that fate if Kentucky ends its coverage entirely of those who have joined the
Medicaid rolls via the Affordable Care Act, as it has threatened to do if this project is not
approved.
The Supreme Court, in holding that Congress could not require states to adopt that
Medicaid expansion by conditioning all their Medicaid funding on a decision to do so, explained
that the states could not be compelled to engage in a program they had not bargained for with “a
gun to the head.” Nat’l Fed. of Indep. Business v. Sebelius, 567 U.S. 519, 581 (2012).
Kentucky, it seems, has now picked up that gun by threatening to de-expand Medicaid.
Defendants urge the Court to adopt the proposition that the Secretary need not grapple with the
coverage-loss implications of a state’s proposed project as long as it is accompanied by a threat
that the state will de-expand — or, indeed, discontinue all of Medicaid. By definition, so this
2
argument goes, any number of people covered by an experimental Medicaid program would be
greater than the number if there were no Medicaid at all; as a result, any demonstration project
that leaves any individual on a state’s Medicaid rolls promotes coverage. The Court cannot
concur that the Medicaid Act leaves the Secretary so unconstrained, nor that the states are so
armed to refashion the program Congress designed in any way they choose. As a consequence,
once again finding the reapproval was both contrary to the Act and arbitrary and capricious, the
Court will vacate it and remand to HHS for further review.
I. BACKGROUND
The details of the statutory scheme and the facts of the dispute will be familiar to readers
of the Court’s prior Opinion. See Stewart I, 313 F. Supp. 3d 237. The Court nevertheless offers
a brief refresher on both before setting out the Secretary’s actions on remand.
A. Statutory Scheme
Medicaid is a cooperative federal-state scheme that aims to provide medical assistance to
certain vulnerable populations. See 42 U.S.C. § 1396-1. Specifically, Congress implemented the
program “[f]or the purpose of enabling each state, as far as practicable . . . to furnish (1) medical
assistance . . . [to] individuals[] whose income and resources are insufficient to meet the costs of
necessary medical services, and (2) rehabilitation and other services to help such families and
individuals attain or retain capability for independence and self-care.” Id. The Centers for
Medicare and Medicaid Services (CMS), a federal agency within HHS, has primary
responsibility for overseeing the Medicaid program. To receive federal funding, states must
submit their “plans for medical assistance” for the HHS Secretary’s approval. Id. Currently, all
states have chosen to participate in the program.
3
The Medicaid Act sets out certain minimum requirements to which all state plans must
conform. See 42 U.S.C. § 1396a. Those provisions ensure that individuals receive a minimum
level of coverage and stipulate that state plans “mak[e] medical assistance available” to certain
sets of low-income individuals. See 42 U.S.C. § 1396a(a)(10)(A). Originally, that group
included only pregnant women, children, and their families; some foster children; the elderly;
and people with certain disabilities. Id. In 2010, the passage of the Affordable Care Act,
colloquially known as Obamacare, gave states a choice to expand their Medicaid coverage to
include additional low-income adults under the age of 65 who would not otherwise qualify — a
group now commonly referred to as the “expansion population.” 42 U.S.C.
§ 1396a(10)(A)(i)(VIII).
The Act also allows states wishing to deviate from either the original or the additional
requirements of Medicaid to obtain a waiver from the Secretary of HHS. See 42 U.S.C. § 1315.
Section 1115 of the Social Security Act, accordingly, permits the Secretary to approve
“experimental, pilot, or demonstration project[s]” in state plans that would otherwise fall outside
the Medicaid Act’s parameters. The Secretary, however, can approve only those projects that “in
[her] judgment . . . [are] likely to assist in promoting the [Act’s] objectives.” 42 U.S.C.
§ 1315(a). If a project, in the Secretary’s judgment, passes muster, she can then waive
compliance with the terms of § 1396a “to the extent and for the period . . . necessary to enable
[the] State . . . to carry out such project.” 42 U.S.C. § 1315(a)(1).
B. Factual Background
1. Kentucky HEALTH
In 2018, CMS released a State Medical Director (SMD) letter that indicated its new
commitment to “support[ing] state efforts to test incentives that make participation in work or
4
other community engagement a requirement for continued Medicaid eligibility” and that
encouraged states to apply for § 1115 waivers for this purpose. See AR 90. The Commonwealth
of Kentucky submitted one such waiver application. As the Court has previously detailed, its
application has multiple components. See Stewart I, 313 F. Supp. 3d at 246. Relevant here is its
Kentucky HEALTH program, which “applies only to adult beneficiaries who do not qualify for
Medicaid on the basis of a disability.” Id. (internal quotation marks and citation omitted). That
is, it targets primarily — though not exclusively — the ACA expansion population. Id.
Kentucky “believed that this project would ‘transform’ the state’s Medicaid program by, among
other things, predicating Medicaid eligibility for most of the expansion population on workforce
participation or community service.” Id. (quoting AR 2, 15–16).
Just one day after releasing the SMD letter, the Secretary approved Kentucky HEALTH,
granting the Commonwealth waivers to implement the following six features: first, a
community-engagement requirement mandating that beneficiaries spend at least 80 hours per
month on qualifying activities (including employment, job-skills training, education, community
service, and participation in Substance Use Disorder (SUD) treatment) or lose Medicaid
coverage; second, elimination — except for pregnant women and former foster-care youth — of
the three-month period of retroactive eligibility for benefits; third, monthly premiums based on
income and/or length of time enrolled in Medicaid; fourth, elimination — except for former
foster-care youth, pregnant women, or the medically frail — of the Commonwealth’s obligation
to assure non-emergency medical transportation to and from providers; fifth, reporting
requirements; and sixth, lockouts allowing the Commonwealth to deny coverage for up to six
months to any beneficiary who failed to meet her premium or reporting requirements and has an
income above 100% of the federal poverty line. Id. at 246–47. The Kentucky HEALTH
5
program also includes features similar to health-insurance plans on the commercial market,
including “an incentive and savings account called My Rewards.” Id. at 247 (citations omitted).
2. Stewart I
Two weeks after the Secretary’s approval of Kentucky HEALTH, fifteen Kentuckians
headed to Court, filing a nine-count suit seeking declaratory and injunctive relief on behalf of
themselves and a “statewide proposed class . . . of all residents of Kentucky who are enrolled in
the Kentucky Medicaid program on or after January 12, 2018.” ECF No. 1 (Complaint), ¶ 33.
The Court granted Kentucky’s Motion to Intervene, see Minute Order of March 30, 2018, and
the parties subsequently filed competing Motions for Summary Judgment. See ECF Nos. 33, 50,
51. Because Kentucky HEALTH was slated to take effect on July 1, 2018, the Court operated on
an expedited schedule and issued its Opinion on June 29, 2018.
It found the Secretary’s approval, considered as a whole, arbitrary and capricious because
he “never adequately considered whether Kentucky HEALTH would in fact help the state furnish
medical assistance to its citizens, a central objective of Medicaid.” Stewart I, 313 F. Supp. 3d at
243. Specifically, he not only “failed to consider adequately the impact of Kentucky HEALTH
on Medicaid coverage,” but “he entirely failed to consider Kentucky’s estimate that 95,000
persons would leave its Medicaid rolls during the 5-year [demonstration] project.” Id. at 260
(internal quotation marks and citations omitted). The Court rejected the Secretary’s contention
that “he could properly focus on . . . three alternative criteria” — health and well-being, cost
considerations, and beneficiary self-sufficiency — in approving the demonstration project. Id. at
265–66. In light of the Medicaid Act’s “clear emphasis on promoting medical assistance,” the
Court found that “the Secretary could not reasonably focus on health and well-being instead.”
Id. at 268 (internal quotations omitted). It held similarly that while cost savings may be one
6
result of the demonstration project, they “cannot excuse the Secretary’s failure” to consider
coverage. Id. at 271. The Court reasoned similarly regarding self-sufficiency after expressing
“doubts whether such an objective is proper.” Id. at 271. It consequently “den[ied] Defendants’
Motions for Summary Judgment,” “grant[ed] Plaintiffs’ Motion for Summary Judgment . . .[,]
vacate[d] the Secretary’s approval of Kentucky HEALTH, and remand[ed] to the agency.” Id. at
274.
3. Action on Remand
Following the decision in Stewart I, the Secretary returned to the drawing board and re-
opened the public-comment period for Kentucky HEALTH. See AR 25,499. On November 20,
2018, he reapproved Kentucky HEALTH effective on April 1, 2019, for five years. See AR
6718–19. The program has essentially the same features as it did before — namely, a
community-engagement requirement; premium payments; a six-month lockout for failure to
complete the redetermination process or timely report changes to household circumstances;
elimination of retroactive eligibility for most enrollees; elimination of non-emergency medical
transport for most enrollees; heightened cost-sharing for non-emergency use of the emergency
room; and usage of the My Rewards account for various benefits. See AR 6756–60; 6762; 6764–
65; 6769; 6770–72; 6773–80. Indeed, the new approval letter acknowledges as much, naming
only four changes: first, waiving an additional statutory provision of the Act “in an abundance of
caution” to ensure the program limits retroactive eligibility as contemplated; second, “[r]evision
of the premium requirement for beneficiaries who are eligible for transitional medical
assistance”; third, “[u]pdated monitoring and evaluation [provisions]”; and fourth, a
“requirement for Kentucky to submit a demonstration implementation plan and . . . monitoring
protocol.” AR 6723. Defendants have confirmed that those four changes, plus the “add[ition]
7
[of] certain exemptions for survivors of domestic violence,” are the only “substantive changes to
the project.” See ECF No. 130 (Supplemental Filing) at 1–2.
In addition to those modest changes in the plan, the Secretary also altered his rationale for
approval. In the November 20 approval letter, he acknowledged that he “may approve a
demonstration project under section 1115 . . . if, in his judgment, the project is likely to assist in
promoting the objectives of [the Act].” AR 6719. He explained that, while the Act “makes clear
that an important objective of the Medicaid program is to furnish medical assistance and other
services to vulnerable populations,” “there is little intrinsic value in paying for services if those
services are not advancing the health and wellness of the individual receiving them, or otherwise
helping the individual attain independence.” Id. The Secretary concluded that his
“demonstration authority . . . allow[ed]” him to approve states’ experimenting “with different
ways of improving health outcomes and strengthening the financial independence of
beneficiaries,” while “at the same time, allow[ing] states to maintain the long-term fiscal
sustainability of their Medicaid programs and to provide more medical services to more
Medicaid beneficiaries.” AR 6720. He elaborated on the ways in which Kentucky HEALTH
would promote beneficiary health and financial independence and improve the sustainability of
the safety net. See AR 6723–28. Regarding the “signal omission” the Court identified as the
Secretary’s failure to “adequately consider[] whether Kentucky HEALTH would in fact help the
state furnish medical assistance to its citizens,” Stewart I, 313 F. Supp. 3d at 243, he reasoned
this time around that commenters did not understand the nature of the coverage loss, that the
program has exemptions and guardrails in place to minimize coverage loss, that Kentucky is not
required to cover the expansion population in any event, and that the 95,000 individuals who
may lose coverage are “likely dwarfed by the 454,000 newly eligible adults who stand to lose
8
coverage” if the state makes good on its threat to terminate its participation in the ACA
expansion in the absence of the demonstration project. See AR 6730–32.
The question here, of course, is whether this second effort gets the Secretary over the
line. Plaintiffs, not surprisingly, think not. They have returned to this Court seeking review of
the Secretary’s reapproval of Kentucky HEALTH. See ECF No. 91 (Plaintiffs’ MSJ) at 1.
Contending that “the Secretary has failed to remedy the shortcomings identified” in Stewart I,
Plaintiffs seek summary judgment on their APA claims and vacatur of the Kentucky HEALTH
waiver (and, separately, the SMD Letter). Id. at 1–2. Defendants — federal and state — have
cross-moved for summary judgment, maintaining that the reapproval was lawful. See ECF Nos.
108 (HHS MSJ) at 1–5; 110 (Kentucky MSJ) at 1–2. The Court heard oral argument on March
14, 2019, and because Kentucky HEALTH will take effect on April 1, has issued this Opinion on
an accelerated basis.
II. LEGAL STANDARD
The parties have cross-moved for summary judgment on the administrative record. The
summary-judgment standard set forth in Federal Rule of Civil Procedure 56(c), therefore, “does
not apply because of the limited role of a court in reviewing the administrative record.” Sierra
Club v. Mainella, 459 F. Supp. 2d 76, 89 (D.D.C. 2006); see also Bloch v. Powell, 227 F. Supp.
2d 25, 30 (D.D.C. 2002), aff’d, 348 F.3d 1060 (D.C. Cir. 2003). “[T]he function of the district
court is to determine whether or not as a matter of law the evidence in the administrative record
permitted the agency to make the decision it did.” Sierra Club, 459 F. Supp. 2d at 90 (quotation
marks and citation omitted). “Summary judgment is the proper mechanism for deciding, as a
matter of law, whether an agency action is supported by the administrative record and consistent
9
with the [Administrative Procedure Act] standard of review.” Loma Linda Univ. Med. Ctr. v.
Sebelius, 684 F. Supp. 2d 42, 52 (D.D.C. 2010) (citation omitted).
The Administrative Procedure Act “sets forth the full extent of judicial authority to
review executive agency action for procedural correctness.” FCC v. Fox Television Stations,
Inc., 556 U.S. 502, 513 (2009). It requires courts to “hold unlawful and set aside agency action,
findings, and conclusions” that are “arbitrary, capricious, an abuse of discretion, or otherwise not
in accordance with law.” 5 U.S.C. § 706(2). Agency action is arbitrary and capricious if, for
example, the agency “entirely failed to consider an important aspect of the problem, offered an
explanation for its decision that runs counter to the evidence before the agency, or is so
implausible that it could not be ascribed to a difference in view or the product of agency
expertise.” Motor Vehicle Mfrs. Ass’n of U.S., Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S.
29, 43 (1983).
In other words, an agency is required to “examine the relevant data and articulate a
satisfactory explanation for its action including a rational connection between the facts found and
the choice made.” Id. at 43 (quoting Burlington Truck Lines v. United States, 371 U.S. 156, 168
(1962)) (internal quotation marks omitted). Courts, accordingly, “do not defer to the agency’s
conclusory or unsupported suppositions,” United Techs. Corp. v. Dep’t of Def., 601 F.3d 557,
562 (D.C. Cir. 2010) (quoting McDonnell Douglas Corp. v. Dep’t of the Air Force, 375 F.3d
1182, 1187 (D.C. Cir. 2004)), and “agency ‘litigating positions’ are not entitled to deference
when they are merely [agency] counsel’s ‘post hoc rationalizations’ for agency action, advanced
for the first time in the reviewing court.” Martin v. Occupational Safety & Health Review
Comm’n, 499 U.S. 144, 156 (1991) (citation omitted). Although a reviewing court “may not
supply a reasoned basis for the agency’s action that the agency itself has not given,” a decision
10
that is not fully explained may, nevertheless, be upheld “if the agency’s path may reasonably be
discerned.” Bowman Transp., Inc. v. Arkansas–Best Freight System, Inc., 419 U.S. 281, 285–86
(1974) (citation omitted).
III. ANALYSIS
Plaintiffs’ challenge, fortunately, does not require the Court to start from square one;
indeed, this round of litigation resembles in many respects the one concluded in Stewart I.
Plaintiffs again essentially contend that the Secretary has sought to “rewrite the Medicaid Act in
a way that is contrary to the program’s purpose.” Pl. MSJ at 1. They elaborate that he has
violated the APA because he “failed to remedy the shortcomings identified” in Stewart I in again
“ignor[ing]” record evidence that “Kentucky HEALTH’s requirements and benefits cuts are
antithetical to Medicaid’s core purpose”: “furnishing medical assistance to those who are unable
to afford the costs of medically necessary care and services.” Id.
Plaintiffs advance those arguments in challenging both the reapproval as a whole, see
ECF No. 88 (Amended Complaint) at 83 (Count VIII under APA), and the individual
components of the program. Id. at 77–83 (Counts II through VII). Because, as the Court
explained previously, the Secretary must determine under § 1115 “whether a project would
promote the Act’s objectives, not whether each component, viewed in isolation, would,” it will
again limit its analysis to Count VIII. See Stewart I, 313 F. Supp. 3d at 257 (internal quotation
marks and citation omitted). As they did last time, Plaintiffs also assert several additional causes
of action, including that the Secretary acted in excess of statutory authority, that the SMD Letter
ought to be vacated as an improperly promulgated substantive rule, and that Defendants violated
the Take Care Clause. See Pl. MSJ at 34, 43; ECF No. 88 (Amended Complaint) at 83. These
claims will again remain in limbo, as the Court sidesteps resolving them.
11
Before moving to the substance of the dispute, the Court will address two jurisdictional
objections — one concerning standing and one on justiciability.
A. Jurisdiction
Having addressed these issues in depth previously, see Stewart I, 313 F. Supp. 3d at 250–
57, no more than a limited treatment is required here.
The Court takes standing first. Article III restricts the jurisdiction of federal courts to
actual “Cases” and “Controversies.” U.S. Const., art. III, § 2. Not every dispute clears that
hurdle. Specifically, a plaintiff must demonstrate that she suffers: 1) an injury-in-fact that is 2)
caused by the conduct complained of and is 3) “likely” to be “redressed by a favorable decision.”
Lujan v. Defs. of Wildlife, 504 U.S. 555, 560–61 (1992). Defendants here offer no more than a
brief objection. They assert, in a handful of paragraphs, that Plaintiffs lack standing to level
specific challenges at particular components of the Kentucky HEALTH program, including the
waiver of retroactive eligibility, the lockouts, the waiver of non-emergency medical
transportation, and deductions from the MyRewards account. See HHS MSJ at 35, 37, 38, 39.
Specifically, they argue that Plaintiffs’ injuries-in-fact are too speculative to establish standing as
to those particular components of Kentucky HEALTH. Id.
Because the Court will examine whether the reapproval as a whole — rather than its
individual components — violates the APA, it will again consider only whether Plaintiffs have
standing to bring that global challenge. See Davis v. Fed. Election Comm’n, 554 U.S. 724, 734
(2008) (“‘[A] plaintiff must demonstrate standing for each claim he seeks to press’ and ‘for each
form of relief’ that is sought.”) (quoting Daimler Chrysler Corp. v. Cuno, 547 U.S. 332, 352
(2006)). Defendants do not dispute that they do. Although they do cite a case for the proposition
that standing “is not dispensed in gross,” HHS MSJ at 35 (citing Davis, 554 U.S. at 734), it is
12
inapposite here. That case held only that a plaintiff’s standing to challenge one statutory
provision does not necessarily establish her standing to challenge another.
Of course, Defendants’ lack of objections to standing does not end the inquiry, since the
Court has an independent obligation to assure itself of its own jurisdiction. See Floyd v. District
of Columbia, 129 F.3d 152, 155 (D.C. Cir. 1997). It has no trouble here concluding, as it did
previously, that Plaintiffs have established standing to challenge the reapproval of Kentucky
HEALTH as a whole. See Stewart I, 313 F. Supp. 3d at 250–57. Specifically, it found there that
they had established an economic injury stemming from the rise in their premium payments the
program would entail. Id. at 251–52. While granting the Government’s contention that some of
the named Plaintiffs may be exempt from the premium requirement, the Court found that at least
some of them would indisputably be required to pay. Id. That is all that is necessary. See
Animal Legal Def. Fund, Inc. v. Glickman, 154 F.3d 426, 429 (D.C. Cir. 1998) (holding that in
suit brought by multiple plaintiffs, only single plaintiff must possess standing for case to
proceed). Vacating the approval, moreover, would afford Plaintiffs complete relief. See Stewart
I, 313 F. Supp. 3d at 252–53. The Court therefore finds no reason to depart now from its earlier
conclusion that Plaintiffs have established the three elements necessary for standing.
That leaves justiciability. The Court previously analyzed at length the justiciability of
Plaintiffs’ challenges to the program’s approval in response to Defendants’ arguments that the
Secretary’s § 1115 decisions are “committed to agency discretion by law.” Id. at 254–57.
Following every other court to consider this issue, this Court concluded that it could review those
decisions because “there is some ‘law to apply.’” Id. at 254–55. The Secretary, nevertheless,
appears to renew in a single sentence the Government’s position that the language of § 1115
“commits these determinations to the Secretary’s discretion as a matter of law.” HHS MSJ at 14.
13
There is no further analysis, however, much less a proffered reason that the Court should revisit
its prior thorough treatment of this issue. As a result, it has little trouble concluding once again
that the approval is judicially reviewable.
B. Merits
Having cleared the ground, the Court can now move to Plaintiffs’ main beef: the
Secretary’s reapproval of Kentucky HEALTH is, they contend, arbitrary and capricious
primarily because he did not adequately consider whether his § 1115 waiver promotes the
objectives of the Medicaid Act. The Court agrees.
The Secretary can only approve demonstration projects that are “likely to assist in
promoting the objectives” of the Medicaid Act. See 42 U.S.C. § 1315(a). He must,
consequently, first identify those objectives. Courts reviewing an agency’s statutory
interpretation employ the two-step Chevron framework. That is, they first ask whether
“Congress has directly spoken to the precise question at issue,” and, if not, whether “the
agency’s answer is based on a permissible construction of the statute.” Chevron U.S.A., Inc. v.
Nat’l Res. Def. Council, Inc., 467 U.S. 837, 842–43 (1984). The Court need not reach Plaintiffs’
contention that this case is an exceptional one in which Chevron should not apply at all. See
ECF No. 119 (Plaintiffs’ Reply) at 2–4. That is because, even according the Secretary
appropriate deference, his action cannot stand.
As the Court concluded in Stewart I, a central objective of the Act is “furnish[ing]
medical assistance” to needy populations. See 313 F. Supp. 3d at 243. Rather than adequately
addressing Kentucky HEALTH’s potential to cause loss of medical coverage, the Secretary
continues to press his contention that the program promotes his alternative proposed objectives
of beneficiary health, financial independence, and the fiscal sustainability of Medicaid. The
14
Court finds that the first two of those three goals are not objectives of the Act in their own right,
and, regardless, the Secretary’s failure once again to adequately consider the effects of Kentucky
HEALTH on coverage is alone — as it was in Stewart I — fatal to the approval. To explain
why, the Court will begin with an account of the role of coverage as one of the Act’s objectives
before setting out the Secretary’s analysis and counterarguments.
1. Coverage as Objective of the Medicaid Act
In Stewart I, the Court acknowledged that the “objectives” in Section 1115 may be
ambiguous. See 313 F. Supp. 3d at 260. To discern them, it followed other courts that have
considered this issue in beginning with 42 U.S.C. § 1396-1. Id. (citing Pharm. Research & Mfrs.
of Am. v. Concannon, 249 F.3d 66, 75 (1st Cir. 2001); Jonathan R. Bolton, The Case of the
Disappearing Statute: A Legal & Policy Critique of the Use of Section 1115 Waivers to
Restructure the Medicaid Program, 37 Colum. J.L. & Soc. Probs. 91, 132 & n.235 (2003)).
Section 1396-1 provides that Congress appropriated Medicaid funds “[f]or the purpose of
enabling each State, as far as practicable under the conditions in such State, to furnish (1)
medical assistance . . . [to] individuals[] whose income and resources are insufficient to meet the
costs of necessary medical services, and (2) rehabilitation and other services to help such
families and individuals attain or retain capability for independence or self-care.” Medical
assistance, in turn, is defined in the statute as “‘payment of part or all of the cost’ of medical
‘care and services’ for a defined set of individuals.” Adena Reg’l Med. Ctr. v. Leavitt, 527 F.3d
176, 180 (D.C. Cir. 2008) (citing 42 U.S.C. § 1396d(a)); see also Harris v. McRae, 448 U.S. 297,
301 (1980) (“The Medicaid program was created . . . for the purpose of providing federal
financial assistance to States that choose to reimburse certain costs of medical treatment for
needy persons.”); W. Va. Univ. Hosps. Inc. v. Casey, 885 F.2d 11, 20 (3d Cir. 1989) (“[T]he
15
primary purpose of [M]edicaid is to achieve the praiseworthy social objective of granting health
care coverage to those who cannot afford it.”).
Indeed, the Secretary agrees — as he did in the last round of litigation, see Stewart I, 313
F. Supp. 3d at 260 — that § 1396-1 is “include[d]” in the “purposes of Medicaid” and “makes
clear that an important objective of the Medicaid program is to furnish medical assistance and
other services to vulnerable populations.” AR 6719 (citing 42 U.S.C. § 1396-1). At oral
argument, the federal government again agreed that it is an objective of Medicaid, including for
the expansion population. See Oral Argument Transcript at 6; see also Tr. at 10 (acknowledging
that “the central objective of Medicaid under the Court’s analysis in Stewart I is the provision of
medical assistance” and that therefore “what the Court has to figure out is whether the Secretary
approved the project that is reasonably approximated toward enhancing the provision of medical
assistance”).
Particularly given the federal government’s position, the Court finds unpersuasive
Kentucky’s assertion that “[i]t stands to reason that the ‘objectives’ of Medicaid” for the
expansion population are different from those for the original Medicaid population. See
Kentucky MSJ at 5. As it explained after thorough consideration of this issue in Stewart I,
“[T]he Medicaid statute — taken as a whole — confirms that Congress intended to provide
medical assistance to the expansion population.” 313 F. Supp. 3d at 269. Observing that when
Congress passed the ACA, it placed the expansion population “on equal footing with other
‘vulnerable’ populations, requiring that states afford them ‘full benefits,’” id., the Court held that
the Secretary “must . . . evaluate the effect of Kentucky HEALTH on all Medicaid recipients,
including low-income individuals, and he must do so without prioritizing certain groups over
others.” Id. at 270.
16
The Court concludes, therefore, as it did previously, that § 1396-1 provides a central
objective of the Medicaid Act: to furnish medical assistance to the populations covered by the
Act. Under § 1115, the Secretary must therefore “adequately analyze” any demonstration
project’s implications for such assistance or coverage. Id. at 262 (citing Am. Wild Horse Pres.
Campaign v. Perdue, 837 F.3d 914, 932 (D.C. Cir. 2017)).
2. Secretary’s Consideration of Medicaid’s Objectives
Defendants contend that coverage is not the sole objective of Medicaid; indeed, in his
approval letter, the Secretary discusses the program’s promotion of what are, in his view, four
objectives of the Act. Specifically, he explains that one “important objective of the Medicaid
program is to furnish medical assistance and other services to vulnerable populations,” but
because “there is little intrinsic value in paying for [those] services if [they] are not advancing
the health and wellness of the individual receiving them, or otherwise helping the individual
attain independence,” another “objective of the Medicaid program . . . is to advance the health
and wellness needs of its beneficiaries.” AR 6719 (emphasis added). He also concludes that
“reforms that go beyond just routine medical care and . . . evidence-based interventions . . . may
increase beneficiaries’ financial independence.” Id. (emphasis added). Finally, he observes that
“demonstration projects also provide an opportunity for states to test policies that ensure the
fiscal sustainability of the Medicaid program.” Id. (emphasis added). In citing these four
purported objectives, the Secretary contends that he has adequately explained in the new
approval letter how Kentucky HEALTH would promote the objectives of the Medicaid Act,
“amply address[ing]” defects identified in Stewart I. See AR 6718–37; HHS MSJ 14–21.
The Court will review his analysis of those four objectives — furnishing medical
assistance, health, financial independence, and fiscal sustainability — in turn. As before, it finds
17
his consideration of the program’s effects on medical assistance inadequate. His examination of
the other three aims, two of which the Court finds are not stand-alone objectives of the statute in
the first instance, cannot make up for that failure. This is especially true where the Secretary
made no attempt to weigh any of those three aims against the coverage-loss consequences of the
program. Although the Court takes up fiscal sustainability last, the reader should be aware that
this is the principal new position Defendants press in this round of litigation and the one
requiring the most analysis.
a. Furnishing Medical Assistance
In Stewart I, the Court found that the Secretary had “ignored” the Act’s objective to
furnish medical assistance. See 313 F. Supp. 3d at 261. As the Court explained then, “There are
two basic elements to that problem” — namely, “whether the project would cause recipients to
lose coverage” and “whether the project would help promote coverage.” Id. at 262. Although he
has no longer entirely “ignored” this objective of the Act, his reapproval was nevertheless legally
inadequate because he “failed to ‘adequately analyze’ coverage.” Id. (quoting Am. Wild Horse,
873 F.3d at 932). To explain why, the Court separately examines his more recent consideration
of coverage loss and of coverage promotion.
i. Coverage Loss
In the original approval, the Secretary “never provided a bottom-line estimate of how
many people would lose Medicaid with Kentucky HEALTH in place,” an “oversight” that was
especially “glaring” since, “[i]n its application, Kentucky estimated that the project would cause”
a substantial number of people to leave its Medicaid rolls — the equivalent of 95,000 people
losing coverage for a year. Id. As the Court noted before, “Amici maintain that such number is
conservative and peg the real figure as between 175,000 and 297,500” people losing coverage in
18
the first year of the program. Id. (citing ECF No. 44 (Amicus Brief of Deans, Chairs, and
Scholars) at 18). Whatever the precise calculation, the number is undoubtedly substantial.
While the Secretary has now nominally acknowledged that estimate, none of his responses
evinces the kind of “reasoned decisionmaking” that arbitrary-and-capricious review requires.
See Michigan v. EPA, 135 S. Ct. 2699, 2706 (2015).
The most significant point that the Secretary now makes about coverage is one he
couches as a fiscal-sustainability consideration — namely, that because Kentucky is facing
budget woes and has as a result threatened to terminate the entire Medicaid expansion if this
demonstration project is not approved, any coverage loss from the project should be viewed
against the Commonwealth’s unbridled prerogative to scrap the entire population. See AR 6726,
6731; HHS MSJ at 3–4, 22. The Court will, accordingly, address this point in its discussion of
fiscal sustainability. See Section B.2.d, infra. For now, it will note only that the argument is
inconsistent with and relies on an unreasonable reading of the Secretary’s § 1115 authority. It
cannot, as a result, satisfy his obligation to analyze coverage loss.
Defendants next make two arguments questioning the extent of his obligation to consider
coverage losses. The Secretary first contends that § 1115 contemplates that demonstrations may
result in an impact on eligibility, meaning coverage loss does not necessarily render a project
unlawful. See AR 6726, 6729–30. That is certainly true: the Act expressly provides for a
“demonstration project . . . that would result in an impact on eligibility.” 42 U.S.C. § 1315(d).
That acknowledgment does not, however, sanction a demonstration that would result in
significant coverage loss, nor does it relieve the Secretary of his obligation to consider the
magnitude of coverage loss here. Indeed, the limitation remains that demonstration projects can
only be approved it they “promote the objectives” of the Act. Here, the coverage loss is
19
sufficiently significant — even at the low end of the estimated range — that it cannot be waved
off by the rejoinder that some amount of coverage loss is legally permissible.
Second, the Secretary protests that he has no obligation to provide an exact estimate of
the number of people who will lose coverage, particularly because demonstration projects are
experiments intended to gather data and “predictive calculations are a murky science in the best
of circumstances.” See AR 6730; HHS MSJ at 22–23 (citing Cablevision Sys. Corp. v. FCC,
597 F.3d 1306, 1314 (D.C. Cir. 2010)). As an initial matter, the experimental nature of the
project cannot relieve the Secretary of the obligation to do the analysis that § 1115 itself
demands — viz., whether a demonstration project promotes the objectives of the Act.
Of course, the exact number of people who will lose coverage under Kentucky HEALTH
is admittedly subject to some uncertainty. As the D.C. Circuit acknowledged when a petitioner
challenged the potential imprecision of an agency’s numbers, even “in the best of
circumstances,” the agency “has no access to infallible data.” Cablevision, 597 F.3d at 1314.
Nevertheless, Kentucky itself provided an estimate equivalent to 95,000, while amici suggest
“Medicaid losses will be much higher and faster than the Commonwealth predicted.” Am. Br. of
Deans, Chairs, and Scholars at 18; see Stewart I, 313 F. Supp. 3d at 262; see also, e.g., AR
13175, 14665. The central point is that no one questions that the loss will be substantial; as a
result, it requires attention in any approval of the program. See Am. Wild Horse, 873 F.3d at
923 (explaining that agency must “examine all relevant factors and record evidence”); see also
Humane Soc’y of United States v. Zinke, 865 F.3d 585, 606–07 (D.C. Cir. 2017) (“failure to
address . . . salient factor” in statute renders agency’s approval arbitrary and capricious).
Regardless of the precise loss number, the Secretary next maintains that his consideration
was adequate because he has now “expressly stated that he considered Kentucky’s estimated
20
coverage effects and Plaintiffs’ interpretation of those figures, and he has explained why
approval of the project is nonetheless warranted.” HHS MSJ at 23–24 (citing AR 6730–31). He
offers two broad sets of arguments. The first quibbles with the nature of the estimate. The
second concerns safeguards in Kentucky HEALTH meant to assist beneficiaries in avoiding
coverage loss.
The Secretary first explains that the 95,000 number is misleading. It represents the
number of member months Kentucky projected will be reduced by Kentucky HEALTH, divided
by twelve to reach a figure representing one year’s worth of coverage for a given individual. He
argues that, rather than commenting on the number of member months, “Plaintiffs incorrectly
assume, with no foundation for doing so, that every member month of coverage lost under the
demonstration is part of a full year of coverage for a person who never regains coverage.” HHS
MSJ at 24. In addition, the Secretary elaborates, the people who do leave Medicaid permanently
may have lost their eligibility for a variety of reasons, including transitioning to commercial
coverage. See AR 6730–31; see also Kentucky MSJ at 19–20.
This argument is unpersuasive and is, ultimately, legerdemain intended to undercut the
significance of the number. 95,000 is one way to represent the annualized number of member
months of lost coverage under Kentucky HEALTH, as commenters plainly understood. See,
e.g., AR 16707–08. There are, as the Secretary acknowledges, multiple ways to slice the number
of lost months: 95,000 people may lose coverage for a year; a larger number may lose coverage
for shorter periods of time; or fewer people may be deprived of coverage for lengthier periods.
Regardless of how the number of lost member months is distributed among Medicaid
beneficiaries, it indisputably reflects that a substantial number of people will lose coverage. As
such, the Secretary cannot avoid addressing that number. This is especially so where
21
commenters detailed the widespread predicted nature of coverage loss and its devastating effects,
see AR 13175, 15482, 19489, including the destructive effects of coverage gaps. See AR 12918,
12967, 15486, 19388–89, 19985–86; see also ECF No. 99 (Amicus Brief of American Academy
of Pediatrics) at 10–15, 19–20. In other words, understanding of the loss estimate was baked
into their analysis of the magnitude of coverage loss, and the Secretary is not relieved of his
obligation to consider the significance of the number — whether it represents primarily
permanent losses of coverage or a high incidence of gaps.
He makes no effort, moreover, to cite evidence or otherwise provide a reasoned basis for
the assertion that some number of people will transition to commercial coverage and, if so, how
many he might expect. Once again, “[w]hile the agency spoke generally of ‘creating incentives
for individuals to obtain and maintain coverage through private, employer-sponsored insurance,’
it cited no research or evidence that this would happen, nor did it make concrete estimates of
how many beneficiaries might make that transition.” Stewart I, 313 F. Supp. 3d at 264. In
addition, he made no effort to explain how — given that Kentucky HEALTH contains
community-engagement rather than work requirements — beneficiaries could reasonably expect
to get commercial insurance from “education, job skills training, job search activities, and
community service.” AR 6730; see also AR 12823–24, 12858, 12973–75, 14044–45, 16716–18
(explaining beneficiaries unlikely to get coverage on labor market).
Second, the Secretary emphasizes that the coverage-loss “projections were made prior to
the inclusion of changes made to the demonstration at approval, including additional beneficiary
guardrails expected to help beneficiaries maintain enrollment.” AR 6731; see also HHS MSJ at
24. He elaborates that there are a number of exemptions from program requirements — for the
medically frail, pregnant women, former foster youth, and domestic-violence survivors — and
22
also several guardrails — including a good-cause exemption to certain penalties, an opportunity
for re-enrollment after coming back into compliance with program requirements, screening
beneficiaries for other eligibility possibilities before the lockout, full appeal rights prior to
eligibility loss, and maintaining a system for “reasonable modification[]” of the requirements for
persons with disabilities, “among other assurances.” AR 6729; see AR 6725–26; 6727–28;
6734–35; HHS MSJ at 25–27; Kentucky MSJ at 20–21.
Yet, again, this “response . . . is no answer at all.” Stewart I, 313 F. Supp. 3d at 263. The
Commonwealth’s original application provided all of these exemptions, except that for domestic-
violence survivors, which is now an exemption rather than a good-cause exception. It also
included opportunities for beneficiaries to come back into compliance before the end of a penalty
period. “Even with those reforms baked in, Kentucky estimated that 95,000 people would lose
coverage,” and the “commenters, too, expressed their concerns about coverage losses with those
features in mind.” Id. (citing AR 3694–95, 3937). Before Stewart I, the Commonwealth also
added the good-cause exemption. In fact, between then and now, there have been very few
substantive changes to Kentucky HEALTH. See Notice of Supplemental Filing at 1–2. Yet the
Secretary has “never revised Kentucky’s estimate on coverage loss with” the exemptions,
guardrails, or any other set of reforms in mind. See Stewart I, 313 F. Supp. 3d at 264. “Rather,
he granted the waivers with no idea of how many people might lose Medicaid coverage and thus
‘failed to consider an important aspect of the problem.’” Id. (quoting State Farm, 463 U.S. at
43).
At oral argument, counsel for Kentucky suggested that the state had added additional
guardrails — including extending an exemption to those exempt from work requirements
associated with other entitlement programs and a physician-attestation option to support the
23
medical-frailty exemption — that predated the Secretary’s original approval but postdated its
95,000 estimate. These additions, the Commonwealth contends, would have reduced its
estimate. See Tr. at 23–25. Because the reapproval letter evinces consideration of neither of
those things, the Court cannot consider them either, see SEC v. Chenery Corp., 318 U.S. 80, 87
(1943), nor can they be imagined to mitigate the coverage-loss estimate.
ii. Coverage Promotion
The Secretary devotes little space, conversely, to describing how Kentucky HEALTH
would promote coverage. He does elaborate that the “My Rewards Account incentives for
healthy behaviors are intended to increase uptake of preventive services,” and the “waiver of
retroactive eligibility” is designed to “encourage preventive care.” AR 6724. He also explains
that the program will allow Kentucky to “evaluate whether the My Rewards and Deductible
accounts, as well as redetermination and reporting requirements, will strengthen beneficiary
engagement in their personal health and provide an incentive structure to support
responsible consumer decision­making about maintaining health and accessing care and
services,” particularly given that a “prior evaluation of one demonstration project with
beneficiary engagement components has shown some promise that these strategies can have
a positive impact on beneficiary behavior.” Id.
As the Court noted before, the invocation of the incentive created by the waiver of
retroactive eligibility is a “‘conclusory’ reference” to coverage promotion that “cannot suffice,
‘especially when viewed in light of’ an obvious counterargument.” Stewart I, 313 F. Supp. 3d at
265 (quoting Getty v. Fed. Sav. and Loans Ins. Corp., 805 F.2d 1050, 1057 (D.C. Cir. 1986)). In
fact, “restricting retroactive eligibility will, by definition, reduce coverage.” Id. Whether or not
the program generally will lead to an uptick in preventive care, the Secretary makes no effort to
24
quantify that uptick or to weigh it against coverage losses for those whom Kentucky HEALTH
may deprive of all access to care, preventive and otherwise. Likewise, even if beneficiaries
become more engaged with their care, the Secretary must balance that with the possibility that
there will be widespread lack of access to care. In light of the failure to weigh any coverage
promotion in the face of the likelihood of substantial coverage loss, the Secretary did not
“adequately analyze the . . . consequences” of the reapproval. See Am. Wild Horse, 873 F.3d at
932.
b. Health
Moving off of coverage, the reapproval relies in part on the Secretary’s conclusion that
Kentucky HEALTH will promote the health and wellness of its beneficiaries. Indeed, the
Government contends that “Kentucky HEALTH is independently justified because the Secretary
found that it is likely to improve the health of the Medicaid recipients receiving coverage under
the demonstration.” HHS MSJ at 4. To the extent Plaintiffs believe that it “cannot be a
freestanding objective of Medicaid to improve the health of the people that program covers,” the
“Secretary emphatically disagrees.” Id. In the approval letter, he explains this is so because
“there is little intrinsic value in paying for services if those services are not advancing the health
and wellness of the individual receiving them, or otherwise helping the individual attain
independence.” AR 6719.
The Court rejected this argument in Stewart I. Treating health — rather than the
furnishing of medical services — as the Act’s ultimate goal is nothing “more than a sleight of
hand.” Stewart I, 313 F. Supp. 3d at 266. While the Court assumes that the Secretary is entitled
to Chevron deference in isolating health as one of the “objectives” contemplated in § 1115, it
cannot uphold his interpretation even under that standard. It fails at step two because it falls
25
outside “the bounds of reasonableness.” Abbott Labs v. Young, 920 F.2d 984, 988 (D.C. Cir.
1990). “‘The reasonableness of an agency’s construction depends,’ in part, ‘on the
construction’s fit with the statutory language, as well as its conformity to statutory purposes.’”
Goldstein v. SEC, 451 F.3d 873, 881 (D.C. Cir. 2006) (quoting Abbott Labs, 920 F.2d at 988).
On those scores, the Court cannot sustain the Secretary’s generalization of health from the Act’s
objective of furnishing medical care.
The Secretary’s primary contention is that health must be an independent objective
because there is little value in paying for healthcare if it is not advancing that goal. As the Court
explained in its prior Opinion, health was not necessarily the ultimate aim Congress pursued
when it decided to “provide health insurance to needy populations.” Stewart I, 313 F. Supp. 3d
at 267. “It . . . had an interest in making healthcare more affordable for such people.” Id. That
focus is evident in the structure of the Act, which does not require states “to provide direct
medical services to its citizens” but rather “provides federal funding so that the state can
‘pay[] . . . [for] part or all of the cost of medical care and services for a defined set of
individuals.’” Id. (quoting Adena Reg’l Med. Ctr. v. Leavitt, 527 F.3d 176, 180 (D.C. Cir.
2008)). Indeed, the history of the ACA’s passage indicates that Congress was moved by the
problem of medical bankruptcies and the concern that one of now two equally healthy people
could have seen their life irrevocably altered by the cost of care. Id. Congress thus designed a
scheme to address not health generally but the provision of care to needy populations. The
Secretary is not free instead to extrapolate the objectives of the statute to a higher level of
generality and pursue that aim in the way he prefers. See Waterkeeper Alliance v. EPA, 853
F.3d 527, 535 (D.C. Cir. 2017) (“[A]gencies are . . . bound not only by the ultimate purposes
26
Congress has selected, but by the means it has deemed appropriate, and prescribed, for the
pursuit of those purposes.”) (internal quotation marks omitted).
No more persuasive is Kentucky’s argument that the ACA altered the objectives of the
Act to include health as a stand-alone aim. To begin, for the reasons discussed in Section B.1,
supra, the Court is dubious that the ACA altered the objectives of Medicaid at all. None of the
Commonwealth’s specific arguments here alters that conclusion.
It first contends that because the ACA amended the definition of “medical assistance” to
include care itself as well as the payment for that care, the Act’s objectives must now include
health as well as furnishing medical assistance. See Kentucky MSJ at 7. It next maintains that
because the Act requires that medical care be furnished with “reasonable promptness,” the statute
must be concerned with health because receiving care promptly “is undeniably linked to health
and wellness.” Id. at 8. Neither of those statutory features suggests that promoting health as a
stand-alone objective has replaced furnishing medical assistance as the statute’s primary aim.
Medical care is not health; health may be the result of medical care. Faring no better is the
state’s citation-less assertion that “it makes sense to treat expanded Medicaid as a transition
program, at least in part, given the inevitable fluctuations in the economy.” Id. at 18. While the
Commonwealth may believe it makes sense to treat the expansion population differently from
the traditional Medicaid population, there is no evidence that Congress intended to so treat them.
Indeed, the expansion is part of a comprehensive coverage regime. See NFIB, 567 U.S. at 581.
The Court will not explain again why holding health to be a freestanding objective of the
Act would “have bizarre results.” Stewart I, 313 F. Supp. 3d at 267. Were that the case, nothing
would prevent the Secretary from conditioning coverage on a special diet or certain exercise
regime. Id. at 267–68. While there would likely be a basis to conclude that “[e]ither of those
27
conditions could promote ‘health’ or ‘well-being’ . . .[,] both are far afield of the basic purpose
of Medicaid: ‘reimburs[ing] certain costs of medical treatment for needy persons.’” Id. at 268
(quoting Harris, 448 U.S. at 301).
Because the Court finds that health is not a freestanding objective of the statute, it need
go no further, since, if that is so, the Secretary’s consideration of it cannot support his § 1115
analysis. Even if health were such an objective, approving Kentucky HEALTH on this basis
would still be arbitrary and capricious. The Secretary, most significantly, did not weigh health
gains against coverage losses in justifying the approval. Because the provision of Medicaid
coverage is indisputably a central objective of the Act, the Secretary’s consideration of the
separate objective of health does not excuse him from addressing an “important aspect of the
problem.” State Farm, 463 U.S. at 43. This is especially true where the Secretary himself
acknowledged that there is a conflict between his reasoning for why the program promotes health
and the possibility that it will cause widespread coverage deprivation: “To create an effective
incentive for beneficiaries to take measures that promote health and independence, it may be
necessary for states to attach penalties to failure to take those measures,” which “may mean that
beneficiaries who fail to comply will lose Medicaid coverage, at least temporarily.” AR 1631.
Even solely in the realm of health, moreover, the agency fell short. That is because it did
not consider the health benefits of the project relative to its harms to the health of those who
might lose their coverage. Commenters made clear that those health effects were significant.
See AR 12918, 12967, 13175, 15482, 15486, 19388–89, 19985–86. Even if health were an
appropriate consideration under § 1115, therefore, these oversights demonstrate why it would be
necessary to weigh purported health benefits against the coverage losses and resulting health
consequences. This the Secretary did not do, rendering his decision arbitrary and capricious.
28
c. Financial Independence
The Secretary also posits that the project will “test[] measures designed to help adults
transition from Medicaid to greater financial independence and other forms of health coverage,”
including by preparing them for the commercial health market. See HHS MSJ at 2; AR 6724–
25. As the Court found before, financial self-sufficiency is not an independent objective of the
Act and, as such, cannot undergird the Secretary’s finding under § 1115 that the project promotes
the Act’s goals. This is so even where the Court accords Chevron deference to his interpretation
of financial independence as an “objective” contemplated in § 1115. For the reasons that follow,
it is an unreasonable reading of the relevant provision because it is incompatible with the
surrounding statutory language and aims. See Goldstein, 451 F.3d at 881.
The Secretary does not specify the statutory basis from which he derives financial
independence as a purpose. Rather, he explains that “there is little intrinsic value in paying for
services if those services are not” improving beneficiaries’ health “or otherwise helping . . .
individual[s] attain independence.” AR 6719. As before, the Secretary is not free to generalize
or otherwise extrapolate the ultimate value of the program Congress designed. Rather, he must
employ the means Congress prescribed to tackle the problem it identified. See Waterkeeper
Alliance, 853 F.3d at 535.
In his previous approval, the Secretary did specify a statutory basis to believe self-
sufficiency is an objective of the Act — namely, § 1396-1’s appropriation to states to
“furnish . . . rehabilitation and other services to help such families and individuals attain or retain
capability for independence or self-care.” From that, he relied on the mention of “independence
or self-care” to conclude that “greater independence” and “reduc[ing] reliance on public
assistance” are stand-alone objectives of the Act. See Stewart I, 313 F.3d at 271 (citing AR 4,
29
5). The Court found that contention unpersuasive because “[t]he text . . . quite clearly limits its
objectives to helping States furnish rehabilitation and other services that might promote self-care
and independence,” so that it “does not follow that limiting access to medical assistance would
further that same end.” Id. The Secretary, in any case, does not seem to renew that argument
here.
The Commonwealth does offer additional reasons as to why the Court should find that
financial independence is a stand-alone objective of the Act. Again, because the Secretary did
not rely on them in his approval, the Court need not consider them. See Chenery, 318 U.S. at 87.
To add a set of suspenders to its belt, it nevertheless does so briefly here. In Kentucky’s view, it
is an “extreme position to believe that Medicaid is unconcerned with whether able-bodied
persons stay on government assistance.” Kentucky MSJ at 15. The Commonwealth
acknowledges that the Court previously found that financial independence was not a stand-alone
objective of the Act, but reasons that it may make sense as an objective for the expansion
population, if not the traditional one. This is so, it believes, because “Section 1396-1 makes
clear that it does not list the universe of ‘purposes’ for appropriations for expanded Medicaid.”
Id.
For the reasons already explained, the Court does not find that the objectives for the
expansion are different. Renewing its argument in this specific context, the Commonwealth
urges that “‘[i]ndependence’ for [the traditional] four populations and what they need to attain it
is quite different from ‘independence’ for the expanded Medicaid population and what they need
to attain it.” Id. at 16. Kentucky concludes that “independence and self-care” should take on a
different meaning for the expansion population because they are able-bodied adults who do not
require “rehabilitation and other services” to attain self-sufficiency. See Kentucky MSJ at 16–
30
17. The Court finds this position unconvincing because even able-bodied adults may require
rehabilitation or other services to “retain” that capacity, even if they need not “attain” it. See 42
U.S.C. § 1396-1. (And, of course, it is worth noting here that 20% of the beneficiaries to which
Kentucky HEALTH would apply are members of the traditional Medicaid population. See
Stewart I, 313 F. Supp. 3d at 268.)
Kentucky contends finally that its “interpretation of ‘independence’ is bolstered by 42
U.S.C. § 1396u-1(b)(3)(A), which permits termination of Medicaid benefits to those individuals
who have had Temporary Assistance for Needy Families benefits terminated ‘because of
refusing to work,’” and the Second Circuit’s decision in Aguayo v. Richardson, 473 F.2d 1090
(2d Cir. 1973), in which it believes that the court approved a similar waiver application for a
different entitlement program on the basis that it would promote beneficiaries’ self-sufficiency.
See Kentucky MSJ at 18–19. Neither the statutory provision nor the Second Circuit’s decision,
however, sheds light on the objectives of Medicaid. Section 1396u-1(b) is a specific statutory
provision allowing states to coordinate eligibility for people who are covered by both Medicaid
and TANF. TANF has job preparation as one of its objectives and includes work requirements.
See 42 U.S.C. § 601; 42 U.S.C. § 607. That Congress allows for states to coordinate their
administration of these two different programs does not transform the purposes of Medicaid.
Aguayo is no more instructive. In that case, the Second Circuit upheld a waiver allowing
for work requirements in the Aid to Families with Dependent Children statute. The
Commonwealth asserts that the purpose language in the AFDC statute is “remarkably similar” to
that in the Medicaid statute. See Kentucky MSJ at 19. Although the phrase “maximum self-
support and personal independence” does appear in the AFDC statute, see Aguayo, 473 F.2d at
1103–04, the AFDC and Medicaid statutes are very different. The Court cannot credit the
31
Commonwealth’s “argument for uniform usage” and “ignore[] the cardinal rule that statutory
language must be read in context since a phrase gathers meaning from the words around it.”
Gen. Dynamics Land Sys., Inc. v. Cline, 540 U.S. 581, 596 (2004). That is so because the
AFDC statute also contained purposes such as keeping children in their own homes, in addition
to achieving “maximum self-support.” See Aguayo, 473 F.3d at 1104. The Medicaid Act lacks
those additional objectives. The AFDC statute, moreover, already included some work
requirements when the court upheld it in Aguayo, including some from the inception of the
program “quite similar to those” at issue in the waiver. Id.; see also H.R. Rep. No. 74-615 at 3
(1935); Public Welfare Amendments of 1962, Pub. L. No. 87-543, § 105, 76 Stat. 172, 186. The
Medicaid Act does not. The Court is not suggesting a waiver application approving work
requirements in some form could never be lawful because the Medicaid Act does not contain
work requirements. Rather, the point is only that the existence of work requirements in the
AFDC statute is probative of their compatibility with that Act’s objectives and the possibility
that additional similar requirements would promote those aims.
Even if financial independence were an objective of the Act and thus a proper
consideration for § 1115 approval, the Secretary’s invocation of it cannot support the approval
here for a separate reason. That is because, absent any attempt to estimate the number of people
who will gain employment and move onto commercial coverage or otherwise attain financial
independence — or any analysis of the mechanism by which they are likely to do so — his
assertion that some significant number of people would migrate is insufficient. As the Court
pointed out before, “[I]t is not obvious that the community-engagement requirement alone would
help a person shift to private insurance,” particularly given that “individuals can meet it, for
example, by volunteering in the community” — an activity that “may have long term benefits”
32
but would not come with health coverage. See Stewart I, 313 F. Supp. 3d at 264; see also AR
12823–24, 12858, 12973–75, 14044–45, 16716–18 (explaining beneficiaries unlikely to get
coverage on labor market). Even if some number of beneficiaries were to gain independence, the
Secretary does not weigh the benefits of their self-sufficiency against the consequences of
coverage loss, which would harm and undermine the financial self-sufficiency of others. See AR
12916–17, 13547, 16723–24, 17464–65, 19985–87, 26311. These deficiencies render his
determination arbitrary and capricious.
d. Fiscal Sustainability
Long diverted into myriad other byways, the Court now arrives at the broad avenue that
constitutes Defendants’ key position. Fiscal sustainability is, in fact, the primary rationale on
which the Secretary relied in approving this demonstration. In his view, “Demonstration projects
that seek to improve beneficiary health and financial independence” improve well-being and “at
the same time, allow states to maintain the long-term fiscal sustainability of their Medicaid
programs.” AR 6720. The Secretary explained that “Kentucky expects that the reforms included
in the demonstration will enable the Commonwealth to continue to offer Medicaid to the ACA
expansion population,” since Kentucky “has repeatedly stated that if it is unable to move forward
with its Kentucky HEALTH demonstration project, it will discontinue coverage for the ACA
expansion population.” AR 6726. “[E]ven assuming” that the program would result in the
estimated eligibility losses, he posits that the number of people who lose coverage under
Kentucky HEALTH “is likely dwarfed by the 454,000 newly eligible adults who stand to lose
coverage if Kentucky elects to terminate the non-mandatory ACA expansion.” AR 6732. And
because “the demonstration provides coverage to individuals that the state is not required to
cover[,] [a]ny potential loss of coverage that may result from a demonstration is properly
33
considered in the context of a state’s substantial discretion to eliminate non-mandatory benefits
or to eliminate coverage” altogether for the expansion population. See AR 6731.
In this explanation, the Secretary does not make entirely clear whether he interprets fiscal
sustainability to be an independent objective of the Act, or whether making the program more
fiscally sustainable is essentially a point about coverage promotion — that is, whether saving
money by covering fewer people is ultimately coverage promoting because any number of
people Kentucky still covers under the demonstration would be greater than the number of
people covered if it terminated the ACA expansion. Based on federal Defendants’
representations during oral argument, it seems that the Government primarily presses the latter
iteration. See Tr. at 8, 53. The Court, nevertheless, will address each in turn, finding that either
way the argument is sliced, it cannot support the Secretary’s reapproval here.
i. Fiscal Sustainability as Independent Objective
The first issue is whether the Secretary could reasonably conclude that the approval of
Kentucky HEALTH was justified because it advanced, as an independent objective of the Act,
the fiscal sustainability of the safety net. The Court preliminarily considers under the Chevron
framework whether the Secretary permissibly interpreted the Act to contain fiscal sustainability
as a stand-alone objective; finding in the affirmative, it next addresses whether the agency’s
conclusion that the program would advance that objective is arbitrary and capricious.
The Commonwealth is the party that most clearly contends that fiscal sustainability is an
independent objective of the Medicaid Act. It maintains first that, as a textual matter, specific
language in § 1396-1 supports this position. See Kentucky MSJ at 10. Although the Secretary
did not clearly rely on this rationale in the reapproval letter — and does not do so in his briefing
here — the letter does at least understand “Section 1115 demonstration projects . . . [to] provide
34
an opportunity for states to test policies that ensure the fiscal sustainability of the Medicaid
program, better ‘enabling each [s]tate, as far as practicable under the conditions in such [s]tate,’
to furnish medical assistance.” AR 6719 (citing 42 U.S.C. § 1396-1). Federal Defendants, at
oral argument, offered that portion of the letter in support of this contention. See Tr. at 50–51.
As discussed previously, the Court finds at Chevron step one that the word “objectives”
as it appears in § 1115 is ambiguous. It therefore proceeds to Chevron’s second step and asks
whether the Secretary’s interpretation is reasonable. The statutory text on which Defendants rely
provides that the Act aims to “enabl[e] each State, as far as practicable under the conditions in
such state, to furnish” first, medical assistance and second, rehabilitation and other services. See
42 U.S.C. § 1396-1. In context, practicability is at least a qualifier of the extent to which states
must furnish medical assistance. Given that the Act stipulates mandatory floors for benefits and
coverage populations, see 42 U.S.C. §§ 1396a, 1396a(a)(10)(A), the Court believes the qualifier
is as easily understood as establishing a ceiling as it is lowering the floor. That said, it cannot
find his interpretation unreasonable under Chevron. Defendants may, as a result, take into
account fiscal sustainability in determining under § 1115 whether a demonstration project
promotes the objectives of the Act.
Identifying an objective of the Act is just the first step, however. The Secretary must also
give an adequate explanation for why Kentucky HEALTH advances that objective and why, if it
is adverse to other Medicaid objectives, he could reasonably conclude that, on balance, it
promotes the objectives of the Act as required by § 1115. On these fronts, he fell short. First, he
made no finding, supported by substantial evidence, that Kentucky HEALTH would improve the
sustainability of Kentucky’s Medicaid program — either by accruing savings to the state or by
any other mechanism. Second, he unreasonably prioritized program savings without weighing
35
those against the consequences of lost coverage, rendering his determination arbitrary and
capricious.
Beginning with the first, the Secretary made no finding that Kentucky HEALTH would
save the Commonwealth any amount of money or otherwise make the program more sustainable
in some way. In the last round of litigation, the Court isolated some confusion about savings
attributable to the program: Defendants “repeatedly highlight[ed] that the program could save
$2.2 billion,” but the “Commonwealth’s own records show that . . . the state’s actual savings
would be $331 million.” Stewart I, 313 F. Supp. 3d at 271. While not a trivial number, the
back-and-forth highlights the importance of analyzing the evidence to come to some reasonable
conclusion. The Court is not suggesting here that the Secretary must quantify some exact
amount of savings, but he must make some finding that supports his conclusion that the project
would actually advance Kentucky Medicaid’s fiscal sustainability. This is especially so where
— although the Court does not pass on the persuasiveness of the evidence — the record contains
some reason to believe that full administration of the expansion will save the Commonwealth
money, while reducing coverage of the expansion population would cost the state. See AR 4970,
4974–75 (study determining Kentucky will experience significant “positive fiscal impact” from
expansion through 2021).
The point is not to evaluate how Kentucky ought to spend its money. See Tr. at 29
(maintaining that Kentucky’s “budgetary priorities” are “[its] prerogative”). The
Commonwealth, moreover, may well be free to pull out of the expansion entirely (or, indeed, all
of Medicaid) if it chooses not to spend its money that way. The central point is that — without a
finding about the savings that Kentucky HEALTH could be expected to yield — the Secretary
could not make a reasoned decision that it would promote fiscal sustainability. If he is to rely on
36
that rationale, he must so find. Otherwise, as here, he has not marshaled substantial evidence for
that position and, indeed, has ignored contrary evidence in the record. See Fred Meyer Stores,
Inc. v. NLRB, 865 F.3d 630, 638 (D.C. Cir. 2017) (finding agency acted arbitrarily and
capriciously when it “evidence[d] a complete failure to reasonably reflect upon the information
contained in the record and grapple with contrary evidence”); Am. Wild Horse, 873 F.3d at 923
(agency must “examine all relevant factors and record evidence”).
Second, the Secretary’s reliance on fiscal sustainability was arbitrary and capricious
because he did not compare the benefit of savings to the consequences for coverage. The Ninth
Circuit has twice invalidated similar approvals, holding that the agency acted in an arbitrary and
capricious manner in approving a § 1115 waiver that — as here — derived cost savings from
shrinking a state’s Medicaid rolls without adequately considering the program’s coverage
consequences. See Newton-Nations v. Betlach, 660 F.3d 370, 381 (9th Cir. 2011) (holding
Secretary’s approval of § 1115 waiver arbitrary and capricious where its purpose was “to save
money”); Beno v. Shalala, 30 F.3d 1057, 1069 (9th Cir. 1994) (rejecting benefits cut conceived
as work incentive and finding that the “statute was not enacted to enable states to save money or
to evade federal requirements”).
The Government attempts to distinguish these cases by arguing that they held that § 1115
authority did not extend to the approval of simple benefits cuts enacted to save money.
Kentucky HEALTH, Defendants argue, is not a simple “benefit cut.” Tr. at 7, 10–11. That
distinction, however, is not convincing. As a technical matter, Kentucky HEALTH does not, on
its face, simply cut benefits. But the Secretary’s analysis of the program was defective precisely
because he did not adequately consider the significant number of people for whom the program
would entail a loss of all benefits. Their loss of coverage appears, from the record in this case, to
37
be how the Commonwealth would save money. That is precisely what the Ninth Circuit said
states cannot do with a § 1115 waiver.
Defendants rejoin with several cases that they claim stand for the proposition that the
manner in which they considered fiscal sustainability here was reasonable. See HHS MSJ at 15–
18 (citing Pharm. Research & Mfrs. of Am. v. Walsh, 538 U.S. 644, 663, 666–67 (2003)
(plurality); N.Y. State Dept. of Soc. Servs. v. Dublino, 413 U.S. 405 (1973); Pharm. Research &
Mfrs. of Am. v. Thompson, 362 F.3d 817, 824–25 (D.C. Cir. 2004); Aguayo, 473 F.2d at 1103–
04); Kentucky MSJ at 11 (citing similar plus Crane v. Matthews, 417 F. Supp. 532, 540 (N.D.
Ga. 1976), and Cal. Welfare Rights Org. v. Richardson, 348 F. Supp. 491, 496 (N.D. Cal. 1972)).
These cases are of mixed relevance. For example, as the Court discussed previously, Aguayo’s
conclusion that the imposition of work requirements promotes the objectives of the AFDC
statute — which differ from Medicaid’s — does not bear squarely on the Medicaid context,
especially where the reasoning hinged in no small part on the long history of work requirements
in the AFDC program. See 473 F.2d at 1103–04. Dublino holds that federal work requirements
do not preempt the state’s imposition of work requirements on AFDC recipients but does not
analyze the propriety of the state program or its approval. See 413 U.S. 405. Crane and
Richardson affirm that the Secretary’s authority under § 1115 is broad, both in the context of
affirming copayment requirements. See Crane, 417 F. Supp. at 540; Richardson, 348 F. Supp. at
496. The Court does not gainsay that he has “considerable discretion” under § 1115. See
Stewart I, 313 F. Supp. 3d at 256. The question here, however, is whether he has lawfully
exercised it.
The Government relies particularly on Thompson, in which, in their view, the D.C.
Circuit held it permissible to “impose a burden on Medicaid recipients to keep other people off
38
of Medicaid.” Tr. at 11–12. This case, they believe, is parallel because it seeks to lift people out
of Medicaid to make the program smaller and therefore more fiscally viable. Id. at 12.
Thompson, in turn, relies significantly on Walsh. See 362 F.3d at 821. Neither, however, aids
Defendants here. In fact, their analysis demonstrates why this approval was legally defective.
Walsh is a fractured opinion upholding the vacatur of a preliminary injunction preventing
the implementation of a Medicaid-covered outpatient drug program that the state of Maine
sought to implement. See 538 U.S. 644. In the part of the opinion on which the Government
relies, three Justices affirmed that the program served “two Medicaid-related interests” in
benefiting the “medically needy” and in “enabling some borderline aged and infirm persons
better access to prescription drugs earlier,” thereby “reduc[ing]” Medicaid expenses. Id. at 663
(plurality opinion). “A third rather obvious Medicaid purpose [would]” also “be fostered”
because “[a]voiding unnecessary costs in the administration of a State’s Medicaid program
obviously serves the interests of both the Federal Government and the States that pay the cost of
providing prescription drugs to Medicaid patients.” Id. at 663–64 (plurality opinion).
“The analyses in Walsh enlighten[ed]” that of the D.C. Circuit in Thompson, which
concerned a similar program. See 362 F.3d at 821. There, the court considered a challenge to
the Secretary’s approval of “a low-cost state prescription drug coverage program . . . for
beneficiaries of Medicaid and two non-Medicaid state health programs.” Id. at 819. The D.C.
Circuit held that the approval did not violate “the general statutory mandate that Medicaid
services be provided in a manner consistent with the best interests of recipients.” Id. (citing 42
U.S.C. § 1396a(a)(19)). The court upheld the Secretary’s determination that “the best interests
requirement . . . allow[s] a state to establish a Medicaid prior authorization program in order to
secure rebates on drugs for non-Medicaid populations” if a state demonstrates that the program
39
will further “the goals and objectives of the Medicaid program.” Id. at 824–25. In that case, the
prior authorization program furthered Medicaid’s objectives because it allowed the state to make
prescription drugs accessible to borderline Medicaid populations who were, in turn, less likely to
become Medicaid eligible. The rebate program thereby preserved Medicaid resources. Id. at
825. In Defendants’ view, Thompson demonstrates that it is permissible to “impose a burden on
Medicaid recipients to keep other people off of Medicaid.” Tr. at 11–12.
Those cases do not establish that the Secretary acted reasonably here. If anything, they
illuminate how the project in this case — and the reasoning given to support it — departs from
previous ones. Most importantly, the programs in those cases involved only incidental burdens
on Medicaid recipients. Specifically, the drug-rebate programs at issue in Walsh and Thompson
made certain drugs, but not others, more difficult to obtain and in so doing provided reduced-cost
medication to all individuals in the state. Neither program entirely stripped coverage or a
mandatory benefit from Medicaid recipients. Language in those opinions addressed this very
concern. As Walsh noted, “[P]roviding benefits to needy persons and . . . curtailing the State’s
Medicaid costs . . . would not provide a sufficient basis for upholding the program if it severely
curtailed Medicaid recipients’ access to prescription drugs.” 538 U.S. at 664–65 (emphasis
added). Thompson reasoned similarly, pointing out “the absence of any demonstrable significant
impediment to Medicaid services from [the challenged] prior authorization requirement.” 362
F.3d at 826 (citing Walsh, 538 U.S. at 664, 688).
Those disclaimers make eminently clear that a project that enhances financial
sustainability may not advance the objectives of Medicaid if it significantly impedes or curtails
Medicaid services or coverage. Important to both the D.C. Circuit and the Supreme Court was
the fact that neither program threatened the entirety of beneficiaries’ Medicaid coverage — or
40
even an aspect of their coverage, like that for prescription drugs — in the name of cost savings.
Rather, both of those cases explicitly sanctioned an incidental burden on Medicaid recipients.
They do not suggest that Medicaid recipients can be significantly burdened — that is, for
example, their eligibility significantly restricted or benefits significantly cut — in the name of
saving money. That there are limits on the extent to which fiscal sustainability can justify cuts
like those outlined in these cases makes sense. Most cuts to Medicaid services would reduce the
cost of Medicaid and thus advance the sustainability of the program to some extent. But it would
be nonsensical to conclude that any cut therefore always promotes the Act’s objectives.
Perhaps the most important takeaway from these cases is what the Court has been saying
all along: the Secretary must engage in considered analysis of the fiscal-sustainability concern,
both alone and relative to the issue of coverage loss. See Thompson, 362 F.3d at 826. The D.C.
Circuit considered in some detail the Secretary’s explanation for the mechanism by which the
project under review would save money — including the nature of the incidental burden it
imposed — and the amount of money that would be saved, ultimately finding that it passed
muster. Id. The agency had the same responsibility here. In failing to analyze the nature of
expected savings, id. at 825–26, whether the burden on Medicaid recipients was minimal, id. at
826, and how the savings should be balanced against the burdens, the Secretary acted arbitrarily
and capriciously.
ii. Fiscal Sustainability as Coverage Promoting
Defendants’ alternative fiscal-sustainability position is that Kentucky HEALTH is
coverage promoting because, absent its approval, the Commonwealth will — given fiscal strain
— simply de-expand Medicaid. This coverage-promotion argument, in fact, does not depend on
fiscal sustainability at all. Rather, all that matters is that a state, like Kentucky, has threatened to
41
de-expand Medicaid if its proposed demonstration is not approved. The underlying reason for
the threat — whether budgetary priorities, fiscal crisis, or other policy goals — is of no moment.
This route is legally permissible, Defendants assert at multiple points, because the expansion
population is optional and therefore need not receive any coverage. As the Secretary indicated in
the reapproval letter, Kentucky HEALTH should be viewed “in the context of a state’s
substantial discretion” to eliminate coverage for the expansion population. See AR 6731; see
also Tr. at 18, 52. A demonstration that shrinks coverage may thus be coverage promoting for
the purposes of § 1115 as long as the state threatens that if the demonstration is not approved, it
will discontinue coverage entirely. See HHS MSJ at 18–20; AR 6729.
This contention is both inconsistent with the Medicaid Act and arbitrary and capricious.
See Agape Church, Inc. v. FCC, 738 F.3d 397, 410 (D.C. Cir. 2013) (“The analysis of disputed
agency action under Chevron Step Two and arbitrary and capricious review is often ‘the same,
because under Chevron Step Two, [the court asks] whether an agency interpretation is arbitrary
or capricious in substance.’”). As an initial matter, Defendants are incorrect that a state has
additional discretion to diminish or condition eligibility for the expansion — as opposed to the
traditional — population. Relying in part on NFIB, they repeatedly assert that Kentucky has the
prerogative to de-expand. See HHS MSJ at 4, 6, 18, 19–20; Kentucky MSJ at 13; see also Tr. at
13, 18, 54. That may well be correct. But the privilege Kentucky seeks to exercise here is not to
de-expand, but rather to implement the ACA expansion as an à la carte exercise, picking and
choosing which of Congress’s mandates it wishes to implement. NFIB did not sanction that.
Rather, the Court was very clear: “Nothing in [NFIB] precludes Congress from offering funds
under the Affordable Care Act to expand the availability of health care, and requiring that States
accepting such funds comply with the conditions on their use.” 567 U.S. at 585. The case held
42
only that Congress was “not free . . . to penalize States that choose not to participate in [the
expansion] by taking away their existing Medicaid funding.” Id. That is, the Court held that, as
with traditional Medicaid, Congress may impose requirements on the states for the use of
expansion funds. Nothing in that analysis allows for “additional discretion” in how the states
comply with Medicaid requirements for the expansion population as compared to the traditional
one.
While Defendants thus err in positing that their treatment of the expansion population is
undergirded with any greater discretion than their administration of any other part of the
Medicaid program, their arguments about flexibility vis-à-vis the expansion population are
ultimately a red herring. That is so because the entire Medicaid program is optional for states.
The Court does not see why — if Defendants are correct that threats to terminate the expansion
program can supply the baseline for the Secretary’s § 1115 review — that argument would not
be equally good as applied to traditional Medicaid. Their argument must thus posit that any
§ 1115 program that maintains any coverage for any set of individuals promotes the objectives of
the Medicaid Act as long as the state threatens to terminate all of Medicaid in the absence of
waiver approval.
Taken to its logical conclusion, the Secretary’s position thus makes little sense. Under
his reasoning, states may threaten that they wish to de-expand, or indeed do away with all of
Medicaid — for fiscal reasons or no reason at all — if the Secretary does not approve whatever
waiver of whatever Medicaid requirements they wish to obtain. The Secretary could then always
approve those waivers, no matter how few people remain on Medicaid thereafter because any
waiver would be coverage promoting compared to a world in which the state offers no coverage
at all. Remarkably, when asked for a limiting principle to this proposition during oral argument,
43
Defendants did not give one. See Tr. at 9–11, 13–14. Could a state decide it did not wish to
cover pregnant women? The blind? All but 100 people currently on its Medicaid rolls? The
Secretary offers no reason that his position would not allow for any of those results.
Not only does Defendants’ position entail radical results, but it is also inconsistent with
the text of § 1115. The statute requires the Secretary to evaluate whether the project will
promote the objectives of the Act. See 42 U.S.C. § 1315; see also Tr. at 35, 38–39. Against
what baseline is he supposed to evaluate the project? The structure of the waiver provision
assumes the implementation of the Act. It confirms that the relevant baseline is whether the
waiver will still promote the objectives of the Act as compared to compliance with the statute’s
requirements, not as compared with a hypothetical future universe where there is no Act. This is
so because the overarching provision authorizing these waivers stipulates that, if the Secretary
makes a judgment that a demonstration promotes the objectives of the Act, he may then waive
compliance with certain of its provisions “to the extent and for the period necessary” to carry out
the project. See 42 U.S.C. § 1315(a), (a)(1). That is, the provision contemplates a limited
waiver. It would make little sense to have such waiver authority and limitations where the
relevant consideration was not full compliance with the Act’s requirements but instead no
engagement whatsoever in the program.
The Court, furthermore, need not exclude the possibility that fiscal considerations are
ever permissible in any context to reject the staggering breadth of the argument that Defendants
present here. To summarize, their central contention is that, where a state threatens to
discontinue Medicaid coverage entirely, any waiver approval would promote coverage. The
argument does not depend on dealing with the expansion population; it is equally applicable to
traditional Medicaid. It does not depend on a state’s being in a fiscally precarious position
44
because it does not take into account the reason the state wants to discontinue participating in the
Medicaid program. It is not subject to any kind of limiting principle. The Secretary’s
interpretation constitutes “an impermissible construction of the statute . . . because [it] is utterly
unreasonable in” its “breadth” — “nothing in this record . . . indicate[s] that Congress
empowered the agency to effect” such a sweeping authority. See Aid Ass’n for Lutherans v.
U.S. Postal Serv., 321 F.3d 1166, 1178 (D.C. Cir. 2003). Its interpretation is therefore “arbitrary
[and] capricious in substance.” See Agape Church, 738 F.3d at 410. That provision of the Act
does not turn the comprehensive Medicaid program that Congress designed into a buffet for
states. Defendants’ remarkable interpretation of Section 1115 thus cannot stand.
In finding the Secretary’s position unreasonable, the Court does not suggest that the
agency may never consider the fiscal sustainability of the Medicaid program. He very well
might properly assess whether a more efficient way of administering a state’s Medicaid program
would save resources or whether, as in Thompson, a state might save money by continuing to
deliver mandatory care to mandatory populations while restricting precisely which kinds of tests
or medications are available, for example. Those considerations are not incompatible with the
prime objective of the Act being the furnishing of medical assistance. But that is not the exercise
the Secretary engaged in here.
3. Relief
Where a court concludes that an agency’s action is unlawful, “the practice . . . is
ordinarily to vacate the rule.” Ill. Pub. Telecomms. Ass’n v. FCC, 123 F.3d 693, 693 (D.C. Cir.
1997). Defendants, however, protest that any relief should be limited to Plaintiffs here, rather
than all Kentuckians who would lose coverage. See HHS MSJ at 41–42. In support, they
contend that “[t]here is no equitable reason to disrupt the statewide implementation of Kentucky
45
HEALTH and thus jeopardize the expansion coverage for hundreds of thousands of individuals
who are not before this Court.” Id. The cases Defendants cite, however, merely concern
equitable discretion in fashioning remedies or the need, for standing purposes, for a plaintiff’s
remedy to be tailored to her injury. While the Court does retain discretion in granting relief
under the APA, Defendants offer no reason that, in this case, it should depart from the ordinary
practice of vacating the agency action found unlawful under the APA.
Even weighing the harms, as Defendants suggest, the balance does not net out in their
favor. Indeed, while vacatur is warranted based on procedural deficiencies in the approval, as
opposed to the Court’s analysis of the merits of Kentucky HEALTH, it is worth noting here that
many of Plaintiffs’ objections to the program hinge on their concern that the demonstration
project would cause a great deal of harm. In other words, the Court’s holding turns on the
Secretary’s inadequate analysis of the likelihood of coverage loss and its consequences. In light
of that, it is likely that vacatur would inflict less harm than allowing the project to take effect.
As the Court explains in its Opinion today in Gresham v. Azar, No. 18-1900, Slip Opinion at 27–
28, 32 (Mar. 27, 2019), the loss of Medicaid coverage is a substantial burden on Plaintiffs and
others like them.
There is another exception to the ordinary course of vacatur under the APA, which
Defendants do not press but the Court nevertheless considers. “[A]lthough vacatur is the normal
remedy, [courts] sometimes decline to vacate,” Allina Health Servs. v. Sebelius, 746 F.3d 1102,
1110 (D.C. Cir. 2014), depending on the “seriousness of the order’s deficiencies (and thus the
extent of doubt whether the agency chose correctly) and the disruptive consequences of an
interim change.” Allied-Signal, Inc. v. U.S. Nuclear Reg. Comm’n, 988 F.2d 146, 150–51 (D.C.
46
Cir. 1993) (citation omitted). As before, “[n]either factor favors the Government.” Stewart I,
313 F. Supp. 3d at 273.
Failure to consider an important aspect of the problem is a “major shortcoming[]”
generally warranting vacatur. Human Soc’y, 865 F.3d at 614–15; see also SecurityPoint
Holdings, Inc. v. TSA, 867 F.3d 180, 185 (D.C. Cir. 2017); Wedgewood Village Pharmacy v.
DEA, 509 F.3d 541, 552–53 (D.C. Cir. 2007). Stewart I offered clear guidance that Section
1115 mandated that coverage considerations be a central part of the analysis. Rather than follow
that direction, the Secretary doubled down on his consideration of other aims of the Medicaid
Act. Given a second failure to adequately consider one of Medicaid’s central objectives, the
Court has some question about HHS’s ability to cure the defects in the approval. Vacatur would
not, moreover, be especially disruptive. Unlike in Arkansas, Kentucky HEALTH has yet to take
effect. Far from there being “no apparent way to restore the status quo ante,” Sugar Cane
Growers Co-op of Fla. v. Veneman, 289 F.3d 89, 97 (D.C. Cir. 2002), there has yet been no
departure. The Court therefore still “believes that preserving the status quo . . . is appropriate.”
Stewart I, 313 F. Supp. 3d at 273.
C. Other Issues
Plaintiffs raise a number of other issues, including that the Secretary acted in excess of
statutory authority; that the SMD letter is an improperly promulgated substantive rule that should
have been subject to notice and comment; and that they have a claim for relief under the Take
Care Clause. As before, where Plaintiffs alleged a nearly identical set of additional claims, the
Court “need not tackle Plaintiffs’ alternative bases for vacating some or all of the components”
of Kentucky HEALTH or the SMD letter “[b]ecause [it] invalidates [the entire] approval” of
Kentucky HEALTH. Id. at 272. Vacating the reapproval will give Plaintiffs all the relief they
47
seek. “While those [other] questions may resurface on remand, they will not trouble the Court
now.” Id.
IV. CONCLUSION
For these reasons, the Court will grant Plaintiffs’ Motion for Summary Judgment on
Count VIII. It will also deny Defendants’ Cross-Motions, vacate the Secretary’s approval of
Kentucky HEALTH, and remand to the agency. A separate Order so stating will issue this day.
/s/ James E. Boasberg
JAMES E. BOASBERG
United States District Judge
Date: March 27, 2019
48