Shawnee Tribe v. Yellen

Opinion

UNITED STATES DISTRICT COURT FOR THE DISTRICT OF COLUMBIA

_________________________________________

)

THE SHAWNEE TRIBE, )

)

Plaintiff, )

)

v. ) Case No. 20-cv-1999 (APM)

)

JANET L. YELLEN, 1 in her official capacity as ) Secretary of the Treasury, et al., )

)

Defendants. )

_________________________________________ )

)

THE MICCOSUKEE TRIBE OF ) INDIANS OF FLORIDA, )

)

Plaintiff, )

)

v. ) Case No. 20-cv-2792 (APM)

)

UNITED STATES DEPARTMENT ) OF THE TREASURY et al., )

)

Defendants. )

_________________________________________ )

)

PRAIRIE BAND POTAWATOMI NATION, )

)

Plaintiff, )

)

v. ) Case No. 21-cv-0012 (APM)

)

JANET L. YELLEN, in her official capacity as ) Secretary of the Treasury, )

)

Defendant. )

_________________________________________ )

1 Here and in Prairie Band Potawatomi Nation below, pursuant to Rule 25(d) of the Federal Rules of Civil Procedure, the court substitutes the current Secretary of the Treasury as the defendant in this case.

MEMORANDUM OPINION

I. INTRODUCTION In March 2020, Congress passed the Coronavirus Aid, Relief, and Economic Security (“CARES”) Act in response to the COVID-19 public health crisis. The Act created a Coronavirus Relief Fund, which reserved $8 billion for payments to Tribal governments, leaving the precise method of allocation to the Secretary of the Treasury. The Department of Treasury ultimately selected an allocation method in May 2020 that undercounted—as Treasury itself now acknowledges—the membership of certain Tribes. Three such Tribes brought suit challenging that methodology. Although initially unsuccessful before this court, one of the Tribes prevailed in the D.C. Circuit, which ruled that Treasury’s May 2020 allocation methodology was likely arbitrary and capricious under the Administrative Procedure Act (“APA”) as to certain undercounted Tribes. Following remand, on April 30, 2021, Treasury announced a new allocation methodology as to certain undercounted Tribes, resulting in additional distributions of CARES Act funds, including to the three Plaintiff Tribes.

Despite receiving these additional funds, two Plaintiff Tribes—the Miccosukee Tribe of Indians of Florida (“Miccosukee”) and Prairie Band Potawatomi Nation (“Prairie Band Potawatomi”)—continue to challenge Treasury’s allocation methodology. They maintain that not only was Treasury’s May 2020 allocation arbitrary and capricious under the APA, but so too was the agency’s April 2021 distribution of additional funds. The third Plaintiff Tribe—the Shawnee Tribe (“Shawnee”)—is satisfied with its supplemental distribution and has abandoned its legal challenges.

3

Now before the court are the amended motion for summary judgment brought by Miccosukee and Prairie Band Potawatomi (“the Tribes” or “Plaintiffs”) and Defendants’ 2 cross- motion for summary judgment. For the reasons that follow, the court denies Plaintiffs’ motion for summary judgment and grants Defendants’ cross-motion. II. BACKGROUND The court sets forth below the long, complicated factual and procedural history of these three cases. Some detail is left out in the interest of brevity.

A. 2020 Title V Funding Allocation In March 2020, Congress passed the CARES Act in response to the developing COVID-

19 public health crisis. CARES Act, Pub. L. No. 116-136, 134 Stat. 281 (2020). Title V of the Act amends the Social Security Act and appropriates $150 billion for “payments to States, Tribal governments, and units of local government,” of which $8 billion was set aside for payments to Tribal governments. 42 U.S.C. § 801(a)(1), (2)(B). Such payments were to be allocated “based on increased expenditures of each . . . Tribal government” due to COVID-19, to be “determined in such manner as the Secretary [of the Treasury] determines appropriate,” “in consultation with the Secretary of the Interior and Indian Tribes.” Id. § 801(c)(7).

On May 5, 2020, Treasury announced its methodology for allocating the $8 billion in Title V funds set aside for Tribal governments. Pls.’ Mot. for Summ. J., ECF No. 70 [hereinafter Pls.’ First MSJ], Ex. H, ECF No. 70-8 [hereinafter 2020 Allocation Methodology], at 2. 3 Tribal

2 Defendants in this consolidated action are Deb Haaland, in her official capacity as Secretary of the Interior; the Department of the Interior; the Department of the Treasury; Janet Yellen, in her official capacity as Secretary of the Department of the Treasury; and the United States. 3 Unless otherwise indicated, all references to the docket are in the consolidated action, Shawnee Tribe v. Yellen, 20- cv-1999 (APM).

4

population would serve as a proxy for a portion of the allocation: “60[%] of the $8 billion reserved for Tribal governments” would be distributed “immediately based on population”; the remaining 40% would be based on employment and expenditures data. Id. Treasury explained that it expected “Tribal population . . . to correlate reasonably well with the amount of increased expenditures of Tribal governments related directly to the public health emergency, such as increased costs to address medical and public health needs.” Id. at 1. Treasury advised that it would use “reliable and consistently prepared data” for Tribal population—namely, “Tribal population data used by the Department of Housing and Urban Development (HUD) in connection with the Indian Housing Block Grant (IHBG) Program.” Id. at 2. The IHBG data relies on Census data tied to geographic “formula area[s]” developed by HUD. Id.; 24 C.F.R. § 1000.302. For any Tribes not included in the IHBG data, Treasury obtained population figures from HUD. 2020 Allocation Methodology at 3. Tribal governments with an IHBG population below 37 were assigned a “minimum payment of $100,000.” Id.

The IHBG population data, however, severely undercounted the enrollment of certain Tribes. See Pls.’ Am. Mot. for Summ. J, ECF No. 88 [hereinafter Pls.’ Second MSJ], Ex. B, ECF No. 88-2 [hereinafter 2021 Reallocation Methodology], at 1. Indeed, “several Tribes . . . were assigned a population of zero, or a low population that approaches zero, for their IHBG formula areas despite having substantial enrollment.” Id.

B. Litigation Arising from the May 2020 Allocation Three Tribes whose populations were substantially undercounted based on the IHBG data—all Plaintiffs in this consolidated action—brought suit to challenge Treasury’s reliance on the IHBG data. See Compl., Prairie Band Potawatomi Nation v. Mnuchin, No. 20-cv-01491

5

(APM) (D.D.C.) [hereinafter Prairie Band Potawatomi I Docket], ECF No. 1 [hereinafter Prairie Band Potawatomi’s 2020 Compl.]; Compl., ECF No. 2 [hereinafter Shawnee’s Compl.]; Miccosukee Tribe of Indians of Fla. v. U.S. Dep’t of the Treasury, No. 20-cv-02792 (APM) (D.D.C.) [hereinafter Miccosukee Docket], ECF No. 1 [hereinafter Miccosukee’s Compl.]. The relevant procedural aspects of their cases are discussed in greater detail below.

Prairie Band Potawatomi filed the first case that came before the court. It sought to enjoin Treasury from disbursing the IHBG population–based portion of Title V funds to Tribal governments because of flaws in the IHBG data that resulted in a substantial undercount (though not a zero count) of its population. See Prairie Band Potawatomi Nation v. Mnuchin, No. 20-cv- 1491 (APM), 2020 WL 3402298, at *1 (D.D.C. June 11), appeal dismissed, No. 20-5171, 2020 WL 4931697 (D.C. Cir. July 16, 2020). The court declined to grant an injunction in part on the ground that the Secretary’s allocation decision was “committed to agency discretion by law” under the APA, 5 U.S.C. § 701(a)(2), and therefore not judicially reviewable. Prairie Band Potawatomi Nation, 2020 WL 3402298, at *1–2. After initially appealing the court’s decision, Prairie Band Potawatomi abandoned its appeal and voluntarily dismissed its action. See Prairie Band Potawatomi Nation v. Mnuchin, No. 20-5171, 2020 WL 4931697, at *1 (D.C. Cir. July 16, 2020); Notice of Voluntary Dismissal, Prairie Band Potawatomi I Docket, ECF No. 30.

Next came the action filed by Shawnee. Shawnee filed its suit in the Northern District of Oklahoma, but the case was transferred to this District a month into the litigation. See Op. & Order, ECF No. 27. Much like Prairie Band Potawatomi before it, Shawnee sought to enjoin the disbursement of additional Title V funds on the ground that the IHBG population–based allocation of Title V funds was arbitrary and capricious as applied to it. In Shawnee’s case, the IHBG data

6

showed the Tribe as having a population of zero despite its 3,021 enrolled members (and despite the fact that other, non-IHBG, HUD data reflected an enrollment number of 2,113 for the Tribe). See Shawnee Tribe v. Mnuchin, 480 F. Supp. 3d 230, 232 (D.D.C. 2020), rev’d, 984 F.3d 94 (D.C. Cir. 2021); Shawnee’s Compl. ¶¶ 26–27. Not deviating from its decision in Prairie Band Potawatomi, the court denied injunctive relief in part on the ground that the Secretary’s use of IHBG data as a proxy for allocation was not judicially reviewable. See Shawnee, 480 F. Supp. 3d at 232. The court later dismissed Shawnee’s complaint on the same basis. See Shawnee Tribe v. Mnuchin, No. 20-cv-1999 (APM), 2020 WL 5440552, at *1 (D.D.C. Sept. 10, 2020), rev’d, 984 F.3d 94.

Shawnee appealed these decisions, and on January 5, 2021, the D.C. Circuit reversed them both. See Shawnee, 984 F.3d at 103. First, the D.C. Circuit held that, although Title V conferred broad authority on the Secretary, his decision to use the IHBG data as a proxy for allocating Title V funds was judicially reviewable and the statute provided a meaningful standard—“increased expenditures”—against which to judge his decision. See id. at 99–101. The court also ruled on the Tribe’s APA claim: it found, under the standard for preliminary injunctions, that Shawnee was likely to succeed on the merits because “the IHBG formula area population data is, at least with respect to some tribes, an unsuitable proxy” for “increased expenditures.” Id. at 102–03. The D.C. Circuit went on to enjoin the Secretary from distributing $12 million in remaining Title V funds, as Shawnee had requested, and it reversed the dismissal of the complaint. See id. at 103. On remand, this court formally entered the same injunction pending the outcome of the case. Order, ECF No. 55.

7

Meanwhile, Miccosukee had filed its own action in the Southern District of Florida and likewise sought preliminary relief for Treasury’s treatment of it as a zero-population Tribe based on the IHBG data. Miccosukee’s Compl; Expedited Mot. for Prelim. Inj., Miccosukee Docket, ECF No. 5 [hereinafter Miccosukee’s Sept. 2020 Mot. for Prelim. Inj.]. That case eventually made its way to this court. See Order Transferring Case, Miccosukee Docket, ECF No. 21. Thereafter, likely sensing that the D.C. Circuit would reverse this court in Shawnee, Prairie Band Potawatomi refiled suit and again sought preliminary injunctive relief. See Mot. for Prelim. Inj., Prairie Band Potawatomi Nation v. Mnuchin, No. 21-cv-00012 (APM) (D.D.C.) [hereinafter Prairie Band Potawatomi II Docket], ECF No. 4 [hereinafter Prairie Band Potawatomi’s Jan. 2021 Mot. for Prelim. Inj.].

After the remand in Shawnee, this court consolidated all three Tribes’ cases under the Shawnee docket. Minute Order, Miccosukee Docket, Jan. 14, 2021; Minute Order, Prairie Band Potawatomi II Docket, Jan. 14, 2021.

C. 2021 Title V Funding Reallocation By the time Shawnee returned to this court, limited Title V funds remained to compensate Tribes underfunded based on IHBG data. That was so for two reasons: First, in June 2020, this court had ordered Treasury to distribute all remaining Title V funds not encumbered by ongoing litigation. See Agua Caliente Band of Cahuilla Indians v. Mnuchin, No. 20-cv-01136 (APM), 2020 WL 3250701, at *1 (D.D.C. June 15, 2020). Second, the funds that remained subject to litigation were funds that Treasury had allocated to Alaska Native Corporations (“ANCs”). This court had ordered over $500 million to be held back pending litigation over whether ANCs qualified as “Tribal governments” for purposes of the CARES Act. See Confederated Tribes of

8

Chehalis Rsrv. v. Mnuchin, No. 20-cv-01002 (APM), 2020 WL 3791874, at *1 (D.D.C. July 7, 2020). At the time of remand, litigation over ANC eligibility was pending before the Supreme Court. See Yellen v. Confederated Tribes of Chehalis Rsrv., 141 S. Ct. 2434 (2021). The Supreme Court eventually held that ANCs were indeed eligible to receive Title V funds, see id. at 2452, thereby shrinking the available pot of Title V funds to distribute to undercompensated Tribes.

In response to the D.C. Circuit’s Shawnee ruling, Treasury decided to “reconsider its prior decision” to rely on IHBG data as a population-based proxy for increased expenditures and “reallocate a portion of the remaining, unpaid [Coronavirus Relief Fund] funds pursuant to a new methodology.” 2021 Reallocation Methodology at 2; see also Status Report, ECF No. 60, at 1 (“[T]he Department of the Treasury will construct a revised methodology to determine how Treasury will allocate any funds remaining . . . . In constructing this revised methodology, Treasury intends to re-consider the formula and/or data used to allocate the funds to Plaintiffs, and similarly-situated Tribes . . . .”). Treasury officially announced this “reallocation” of funds—and its new methodology—on April 30, 2021. 2021 Reallocation Methodology at 1–3. The “reallocation” assumed that ANCs would be eligible for Title V funding. See id. at 3.

In its announcement, Treasury explained that the “reallocation [would] provide additional payments when there is a substantial disparity between the Tribe’s IHBG formula area population count and its Tribal enrollment count,” because for those Tribes “formula area population is less likely to be an accurate proxy for increased expenditures.” Id. at 2. Treasury explained its reallocation methodology as follows: First, “to determine which Tribes may have received a payment that significantly undercounted their potential expenditures,” the Department would “compare the IHBG formula area population against enrollment data recently collected by the

9

Bureau of Indian Affairs (BIA).” Id. at 3. If enrollment data for a particular Tribe was not available from BIA, Treasury would instead rely on the enrollment data the Tribe had submitted in response to the Department’s April 2020 solicitation of enrollment data or, as a third option, the enrollment data “available in HUD’s IHBG data” (as distinct from the IHBG population data used in the 2020 allocation). Id. With that data in hand, Treasury would calculate the ratio between each Tribe’s IHBG population and its enrollment. Id. It would subtract each ratio from 1, yielding each Tribe’s “population-to-enrollment ratio.” Id. Treasury stressed that “funds available for reallocation are limited and therefore only the most substantial disparities can be addressed.” Id. To that end, Treasury would rank the population-to-enrollment ratios and provide additional payments only to the Tribes in the 85th percentile and above; stated differently, Treasury would allocate the remaining funds to the 15% of Tribes with the greatest disparity between IHBG population and actual enrollment. Id.

After identifying the most undercompensated Tribes, Treasury would then need to decide how much to pay each of them. It made that allocation in two steps. Treasury would begin by determining “the amount [each] Tribe would have received under an enrollment-based allocation” by “calculating each Tribe’s pro rata share of the amount of the population-based allocation available for Tribes in the aggregate.” Id. In other words, Treasury determined what each Tribe’s share would have been if the 2020 allocation had been based on enrolled population for all Tribes. Next, Treasury “linearly phased [the payments] out such that the Tribe with the highest population- to-enrollment ratio [would] receive the largest percentage of the difference between the amount it would have received under the enrollment-based allocation and the IHBG formula area population- based allocation.” Id. That is, Treasury ran a calculation to ensure that the most underfunded

10

Tribes would receive a higher percentage of the available funds than Tribes that were underfunded to a lesser extent.

All three Plaintiff Tribes in this consolidated action were determined to be eligible for reallocated funds based on their population-to-enrollment ratios. On May 3, 2021, Treasury notified the three Plaintiff Tribes via email of the “supplemental amount each Plaintiff Tribe will receive under the revised methodology.” Pls.’ Second MSJ, Ex. D, ECF No. 88-4, at 1. Shawnee was to receive an additional $5,202,604; Miccosukee was to receive an additional $825,196; and Prairie Band Potawatomi was to receive an additional $864,141. Id. Or, in different terms, Shawnee received $1,712 per initially uncounted member under the reallocation methodology; Miccosukee received $1,355 per uncounted member; and Prairie Band Potawatomi received $225 per uncounted member. See Mem. Op. & Order, ECF No. 90 [hereinafter Mem. Op. Denying PI Modification], at 7; Pls.’ Reply in Supp. of Mot. for Summ. J. & Resp. to Def.’s Cross-Mot. for Summ. J., ECF No. 93 [hereinafter Pls.’ Reply], at 1.

D. Litigation After Remand At the time of the remand in Shawnee, both Prairie Band Potawatomi and Miccosukee had pending motions for preliminary injunctions in their individual cases. Prairie Band Potawatomi’s Jan. 2021 Mot. for Prelim. Inj.; Miccosukee’s Sept. 2020 Mot. for Prelim. Inj. After consolidation, the three Plaintiff Tribes moved together for an additional preliminary injunction, asking the court to order Treasury to immediately distribute Title V funds to them. Suppl. Mot. for Prelim. Inj., ECF No. 65. Plaintiffs then also filed their first motion for summary judgment. Pls.’ First MSJ.

With the summary judgment briefing schedule on its way, the court denied Plaintiffs’

request for an immediate release of funds but, based on the D.C. Circuit’s holding in Shawnee,

11

granted Miccosukee’s and Prairie Band Potawatomi’s motions to “enjoin Treasury’s distribution of [a] combined $9,647,063 in remaining Title V funds pending resolution of th[e] litigation.” Shawnee Tribe v. Yellen, No. 20-cv-1999 (APM), 2021 WL 1634597, at *2 (D.D.C. Apr. 26, 2021). 4 That sum represented the combined sum the Plaintiff Tribes estimated that they were due under a population-based allocation using enrollment data rather than IHBG population counts.

Treasury announced its reallocation and revised methodology four days later. 2021 Reallocation Methodology. The court stayed the summary judgment briefing schedule “[i]n light of Treasury’s stated intent to imminently distribute Title V funds to the Plaintiff Tribes.” Minute Order, May 4, 2021; see also Mot. to Stay Summ. J. Briefing Schedule, ECF No. 76. Unsatisfied with the amounts they had received, Miccosukee and Prairie Band Potawatomi filed amended complaints, which included claims challenging the 2021 reallocation. Miccosukee Tribe of Indians of Florida’s Am. Compl., ECF No. 79; Prairie Band Potawatomi Nation’s Am. Compl., ECF No. 82.

Then, Prairie Band Potawatomi moved to modify the April 2021 preliminary injunction by increasing the amount of undistributed Title V funds set aside for it from $7.6 million to $11,680,105. See Pl.’s Mot., ECF No. 83, at 1. The court denied that motion, concluding that Prairie Band Potawatomi had failed to explain how its challenge to the May 2020 allocation was not now moot and that the court likely lacked the authority to direct Treasury to pay the Tribe a sum certain. Mem. Op. Denying PI Modification at 5–6 (noting that “the text of the CARES Act

4 The court noted that “[i]n Shawnee Tribe, the D.C. Circuit made clear that the Miccosukee Tribe—also assessed by Treasury as having zero population—is similarly situated to the Shawnee Tribe,” such that Miccosukee was entitled to the same injunction as Shawnee. Shawnee, 2021 WL 1634597, at *4. And, in addition, “because Prairie Band Potawatomi Nation claims that Treasury’s original methodology vastly undercounted its population, its legal claim is substantially similar, and thus warrants the same injunctive relief.” Id. (citation omitted).

12

does not support the ‘extraordinary circumstances’ that might justify a ‘detailed remedial order’” (quoting Shawnee, 2021 WL 1634597, at *2)).

Days before the court ruled on Prairie Band Potawatomi’s motion to modify the preliminary injunction, Prairie Band Potawatomi and Miccosukee filed their amended motion for summary judgment, which “incorporates by reference the prior summary judgment motion.” Pls.’ Amended MSJ at 2. Defendants filed a cross-motion for summary judgment. Cross-Mot. for Summ. J., ECF No. 92; Cross-Mot. for Summ. J. & Opp’n to Pls.’ Mot. for Summ. J., ECF No. 91 [hereinafter Defs.’ Cross-Mot.]. The court now considers these motions. III. LEGAL STANDARD When courts review agency action under the APA, “summary judgment is the mechanism for deciding whether as a matter of law an agency action is supported by the administrative record and is otherwise consistent with the APA standard of review.” Louisiana v. Salazar, 170 F. Supp. 3d 75, 83 (D.D.C. 2016). The district court “sits as an appellate tribunal,” reviewing the entire case as a question of law. Am. Bioscience, Inc. v. Thompson, 269 F.3d 1077, 1083–84 (D.C. Cir. 2001) (collecting cases). Accordingly, the court need not engage in lengthy factfinding, and as a general rule, judicial review is limited to the administrative record. “It is black-letter administrative law that in an [APA] case, a reviewing court should have before it neither more nor less information than did the agency when it made its decision.” CTS Corp. v. EPA, 759 F.3d 52, 64 (D.C. Cir. 2014) (internal quotation marks omitted); see also 5 U.S.C. § 706 (“[T]he court shall review the whole record or those parts of it cited by a party . . . .”).

The APA 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

13

with law.” 5 U.S.C. § 706(2)(A). “This is a ‘narrow’ standard of review as courts defer to the agency’s expertise.” Ctr. for Food Safety v. Salazar, 898 F. Supp. 2d 130, 138 (D.D.C. 2012) (quoting Motor Vehicle Mfrs. Ass’n of U.S., Inc. v. State Farm Mut. Auto. Ins., 463 U.S. 29, 43 (1983)). The court’s review, however, is not toothless. The court must satisfy itself that the agency has “examine[d] the relevant data and articulate[d] a satisfactory explanation for its action including a rational connection between the facts found and the choice made.” State Farm, 463 U.S. at 43 (internal quotation marks omitted). When an agency “has failed to provide a reasoned explanation, or where the record belies the agency’s conclusion, [the court] must undo its action.” County of Los Angeles v. Shalala, 192 F.3d 1005, 1021 (D.C. Cir. 1999). “Moreover, an agency cannot fail to consider an important aspect of the problem or offer an explanation for its decision that runs counter to the evidence before it.” Dist. Hosp. Partners, L.P. v. Burwell, 786 F.3d 46, 57 (D.C. Cir. 2015) (cleaned up). “Post hoc rationalizations advanced to remedy inadequacies in the agency’s record” or in the agency’s explanation will not suffice. City of Brookings Mun. Tel. Co. v. FCC, 822 F.2d 1153, 1165 (D.C. Cir. 1987). IV. DISCUSSION Plaintiffs seek summary judgment on the grounds that (1) the May 2020 allocation was “arbitrary and capricious, not in accordance with law, and not fully remedied by the [April] 2021 allocation,” and (2) the April 2021 reallocation was arbitrary and capricious and not in accordance with law. Pls.’ Second MSJ at 2; see 5 U.S.C. § 706(2)(A). The court considers Plaintiffs’ challenge to the May 2020 allocation before turning to the April 2021 allocation.

14

A. Challenge to the May 2020 Allocation Methodology Plaintiffs argue that the awards made pursuant to the 2020 allocation methodology were, “at least with respect to Plaintiffs, arbitrary and capricious and not in accordance with law.” Pls.’ Second MSJ at 12 (internal quotation marks omitted). Specifically, they contend that IHBG data was insufficient for “estimating a Tribal government’s increased expenditures, when, for example, a tribe does not have a formula area or when the tribe provides COVID-related assistance to enrolled members listed outside of Tribal lands.” Id. (internal quotation marks omitted). Plaintiffs do not make this argument in any comprehensive fashion in their amended motion; rather, they “refer to[,] . . . incorporate by reference,” and rest on their first motion. Id.; see also Pls.’ First MSJ. For their part, Defendants argue that Plaintiffs’ claims against the original methodology are moot. They say the court “cannot provide the Plaintiffs with any meaningful relief for their original claim beyond ordering Treasury to do what it has already done,” and the court cannot “require Treasury to adopt any particular methodology.” Defs.’ Cross-Mot. at 12–13. The court agrees with Defendants.

“If events outrun the controversy [in any given litigation] such that the court can grant no meaningful relief” on a claim, there is no longer an “actual, ongoing controvers[y]” for Article III purposes, and so the claim “must be dismissed as moot.” Foretich v. United States, 351 F.3d 1198, 1210 (D.C. Cir. 2003) (internal quotation marks omitted). That is the case here.

“Under settled principles of administrative law, when a court reviewing agency action determines that an agency made an error of law, the court’s inquiry is at an end: the case must be remanded to the agency for further action consistent with the corrected legal standards.” PPG Indus., Inc. v. United States, 52 F.3d 363, 365 (D.C. Cir. 1995). As this court stated in its June

15

2021 Opinion and Order denying Prairie Band Potawatomi Nation’s motion to modify the preliminary injunction already in place in this case, “[t]he Circuit in Shawnee held that IHBG population data was an ‘unsuitable proxy’ for increased expenditures insofar as that data set assessed a zero population for certain tribes. But Treasury has since abandoned that methodology and announced a new approach that, presumably, seeks to rectify its error.” Mem. Op. Denying PI Modification at 5. Defendants therefore are correct that the court can do no more than order Treasury to do something it has already done: reevaluate its allocation methodology. Simply put, “when an agency has rescinded and replaced a challenged regulation, litigation over the legality of the original regulation becomes moot.” Akiachak Native Cmty. v. U.S. Dep’t of the Interior, 827 F.3d 100, 113 (D.C. Cir. 2016).

Plaintiffs point out that this is not precisely what has occurred here: at issue in this case are funding awards, not a regulation, and Treasury did not rescind the funds it allocated pursuant to the 2020 methodology. See Pls.’ Reply at 15. They are technically correct. But as a matter of administrative law, the larger point remains: upon finding the 2020 allocation methodology arbitrary and capricious, the court would be empowered to do no more than set Treasury’s action aside and remand to the agency for reconsideration of the allocation methodology. See 5 U.S.C. § 706(2)(A); PPG Indus., 52 F.3d at 365. The agency did precisely that of its own accord. Plaintiffs’ claims over awards pursuant to the 2020 allocation are moot.

It is true that under certain “extraordinary circumstances,” a court may enter a “detailed remedial order[]” in a case involving a challenge to an agency action. N.C. Fisheries Ass’n, Inc. v. Gutierrez, 550 F.3d 16, 20 (D.C. Cir. 2014). Plaintiffs appear to ask for such an order, see Pls.’ Reply at 15, although it is not entirely clear to the court what specific affirmative relief Plaintiffs

16

seek. Defs.’ Cross-Mot. at 13 (describing Plaintiffs’ requested relief as an order requiring “a different methodology that includes certain mandatory aspects that the Plaintiffs would prefer”); Pls.’ Reply at 13 (arguing that Treasury could have distributed per–uncounted enrollee reallocation awards only to Plaintiffs or to the 15% of Tribes most severely undercounted under the 2020 methodology). But for the reasons the court has previously explained, the requisite “extraordinary circumstances” to justify a detailed remedial order are not present here. See Shawnee, 2021 WL 1634597, at *2. Those circumstances “exist only when courts are permitted to compel agency action under section 706(1) of the APA.” Id. (internal quotation marks omitted). And the APA authorizes courts to “compel agency action,” 5 U.S.C. § 701(1), only “under narrow circumstances.” Elec. Priv. Info. Ctr. v. IRS, 910 F.3d 1232, 1244 (D.C. Cir. 2018). “[T]he court’s authority to issue a detailed remedial order applies only to discrete action that is legally required . . . about which an official had no discretion whatever.” Shawnee, 2021 WL 1634597, at *2 (second alteration in original) (emphasis and internal quotation marks omitted). Plaintiffs make no argument as to how that narrow circumstance is met here.

That is not surprising. When, as here, “the manner of [an agency’s] action is left to the agency’s discretion, a court can compel the agency to act, but has no power to specify what the action must be.” Norton v. S. Utah Wilderness All., 542 U.S. 55, 65 (2004). No one disputes that the CARES Act “‘gives the Secretary some discretion’ in carrying out Congress’s directive.” Shawnee, 2021 WL 1634597, at *2 (quoting Shawnee, 984 F.3d at 100). That “discretion

17

forecloses the detailed order plaintiffs seek.” In re Long-Distance Tel. Serv., 751 F.3d 629, 634 (D.C. Cir. 2014). 5 For these reasons, to the extent that the Tribes maintain their claims pertaining to only the 2020 allocation methodology, their claims are moot. If, as Defendants suggest, Plaintiffs are really arguing “only that they may pursue a claim ‘against the original methodology (as modified by the revised methodology),’” Defs.’ Reply in Supp. of Cross-Mot. for Summ. J., ECF No. 95 [hereinafter Defs.’ Reply], at 20 (quoting Pls.’ Reply at 15), the court agrees that such claims are not moot. However, such claims are best conceived of as 2021 reallocation–based claims, because the second allocation of funds cannot be considered in isolation from the first allocation, and so the court discusses those claims below.

B. Challenge to the 2021 Reallocation Methodology There is no dispute that the 2021 reallocation methodology—and the 2020 methodology as modified by the 2021 reallocation methodology—poses a live controversy. See Defs.’ Cross-Mot. at 13. Plaintiffs argue that “the awards allocated as part of the 2021 Distribution were arbitrary and capricious, and not in accordance with law.” Pls.’ Second MSJ at 2. Specifically, they contend that the 2021 “allocation methodology . . . [(1)] has no coherent connection to the statute, [(2)] is

5 Plaintiffs face an additional potential hurdle: little, if any, Title V funds are left over to compensate Plaintiffs for an amount greater than the enjoined sums. It is not clear to the court how many Tribes were awarded additional funds through the 2021 reallocation, how much of the remaining Title V funds Treasury reserved for the 2021 reallocation, and how much of that reserved amount it has distributed to this point. See Pls.’ Reply at 8 n.4 (“It bears mentioning that Treasury has not disclosed the supplemental payments that were issued to other Tribes apart from the awards to the three named Plaintiffs.”). It is also unclear whether any funds allocated to ANCs remain undistributed. “It is a well-settled matter of constitutional law that when an appropriation has lapsed or has been fully obligated, federal courts cannot order the expenditure of funds that were covered by that appropriation.” City of Houston v. HUD, 24 F.3d 1421, 1424 (D.C. Cir. 1994). “[O]nce the relevant funds have been obligated, a court cannot reach them in order to award relief.” Id. at 1426. Thus, if what Plaintiffs seek is an award greater than the amounts enjoined, there may be no money left to grant such relief.

18

undermined by the actual enrollment data Treasury has on hand, and [(3)] . . . treats similarly situated parties differently.” Id. at 12.

The court first addresses the parties’ dispute as to the degree of deference the court owes Treasury’s 2021 methodology, before turning to Plaintiffs’ specific APA arguments.

1. Degree of Deference Owed Plaintiffs’ primary position is that “Treasury is not entitled to any deference” because “its action is at odds with the plain language of the statute that it is implementing.” Pls.’ Reply at 10. The court understands this to refer to Plaintiffs’ argument that the 2021 reallocation methodology was not “based on” increased expenditures at all; the court discusses this argument in greater detail below. See Pls.’ Second MSJ at 3 (“Treasury paid out the awards based on the percentage by which a tribe’s enrollment was undercounted in the 2020 Distribution, a figure that has no cognizable connection to actual increased expenditures or to Treasury’s previously announced proxy for increased expenditures: Tribal population, or enrollment.”). Defendants, on the other hand, argue that the court should grant the agency “an extreme degree of deference” under the circumstances. Defs.’ Cross-Mot. at 14 (internal quotation marks omitted). In support of their argument, Defendants point to cases involving “complex judgments,” “technical matters,” and “data analysis,” where courts explained that “great deference” is appropriate. Defs.’ Cross-Mot. at 14 (internal quotation marks omitted) (first citing Alaska Airlines, Inc. v. TSA, 588 F.3d 1116, 1120 (D.C. Cir. 2009); and then citing Appalachian Power Co. v. EPA, 249 F.3d 1032, 1051–52 (D.C. Cir. 2001)). Plaintiffs counter that, even if some deference is due, “the allocation[s] [here] . . . [are] not the sort of agency action that [are] eligible for an extreme degree of deference,” because the cases cited by Defendants involved far more complex judgments and

19

technical expertise than what the CARES Act required of the Secretary—and “courts cannot afford agencies ‘extreme deference’ every time they do some math.” Pls.’ Reply at 3, 10 (internal quotation marks omitted).

But whether the court owes Treasury an “extreme” level of deference or simply the normal amount, the outcome is the same. It is manifest that the court owes some deference to the agency in the present circumstances. That deference is embodied in both the APA and the CARES Act. First, the arbitrary-and-capricious standard set forth in 5 U.S.C. § 706(2)(A) requires “that agency action be reasonable and reasonably explained,” but no more. FCC v. Prometheus Radio Project, 141 S. Ct. 1150, 1158 (2021). Indeed, “[a] court may not substitute its own policy judgment for that of the agency,” and so judicial review under the APA is necessarily “deferential.” Id. Second, the procedure the CARES Act established for developing methodologies to allocate Title V funds necessarily entails deference to the Secretary. The Act provides that “the amount paid . . . to a Tribal government shall be the amount the Secretary shall determine, in consultation with the Secretary of the Interior and Indian Tribes, that is based on increased expenditures of each such Tribal Government”—“determined in such manner as the Secretary determines appropriate to ensure that all amounts available . . . are distributed to Tribal governments.” 42 U.S.C. § 801(c)(7) (emphases added). Congress gave the Secretary latitude to “exercise his discretion,” subject, of course, to certain “restrictions.” Shawnee, 984 F.3d at 100. And when Congress delegates discretion to an agency, courts must in turn grant the agency deference for the actions it takes pursuant to that discretion. See Chevron, USA, Inc. v. NRDC, 467 U.S. 837, 844–45 (1984). It is therefore plain that deference to the Secretary is warranted; whether that deference is “extreme” or not does not, in this case, change the court’s analysis.

20

The court also notes that “the parameters of the ‘arbitrary and capricious’ standard of review will vary with the context of the case.” WWHT, Inc. v. FCC, 656 F.2d 807, 817 (D.C. Cir. 1981). The parties agree that arbitrary-and-capricious review is “contextually tailored,” Maggard v. O’Connell, 671 F.2d 568, 571 (D.C. Cir. 1982), but they disagree about how to characterize the relevant context. See Defs.’ Reply at 7; Pls.’ Reply at 11–12. The “relevant context” for defining the contours of the court’s arbitrary-and-capricious review in this case includes the following: (1) the D.C. Circuit’s Shawnee decision and this court’s subsequent orders; (2) the status of Title V funding allocated for ANCs and the Supreme Court’s Chehalis decision; (3) the process undertaken by Treasury to develop its methodologies, including consultation with Tribal leaders; and (4) critically, the limited remaining Title V funds available for distribution to underfunded Tribes upon remand. The court considers each of these factors, as appropriate, in evaluating whether the agency acted arbitrarily and capriciously.

2. “Based On” Expenditures The CARES Act expressly provides that “the amount paid . . . to a Tribal government shall be . . . based on increased expenditures of each such Tribal government.” 42 U.S.C. § 801(c)(7) (emphasis added). Plaintiffs argue that the 2021 reallocation methodology “has no coherent connection to increased expenditures due to COVID.” Pls.’ Second MSJ at 12. More specifically, they contend that the new approach and the awards it generated “are disconnected from both the statutory objective (increased expenditures) and the agency’s own original premise that a pro-rata allocation per tribal member is the proper mechanism for meeting that objective.” Id. at 13. Instead, the Tribes assert, “Treasury allocated the awards for the 2021 Distribution not according to Tribal population, enrollment, or expenditures, but based on the percentage by which a tribe’s

21

enrollment was undercounted in the 2020 distribution.” Id. And “[t]he percentage by which a tribe was undercounted does not correspond to its actual COVID expenditures.” Id. (emphasis omitted). Therefore, in Plaintiffs’ view, Treasury did not “operate within the confines of the statute it is implementing” as it is “bound” to do. Pls.’ Second MSJ at 14.

Defendants respond that the Tribes’ argument “isolates one variable—the IHBG-to-

enrollment ratio—in a broader formula,” and so misapprehends the ways in which the 2021 reallocation methodology is “based on” increased expenditures. Defs.’ Cross-Mot. at 18. Defendants make two arguments as to why their reallocation methodology is “based on” increased expenditures. First, they maintain that the IHBG-to-enrollment ratios “estimate the degree to which IHBG data may have proven inadequate” in approximating increased expenditures for certain Tribes, and so the ratios “help ensure that qualifying tribes would receive,” considering the 2020 and 2021 payments in combination, “payments ‘based on’ their expenditures.” Defs.’ Reply at 10. Second, they assert that, “even assuming that the IHBG-to-enrollment ratios do not help ensure that payments are issued ‘based on’ expenditures, the Revised Methodology did not calculate the Tribes’ supplemental payments by referring only to their IHBG-to-enrollment ratios.” Id. Rather, the reallocation methodology “also relied on the Tribes’ enrollment data—an undeniable proxy for their expenditures—and so the Tribes’ supplemental payments were ‘based on’ expenditures, even if they were not based solely on expenditures.” Id.

The court agrees that the 2021 reallocation methodology is “based on” increased expenditures. The CARES Act does not require Title V awards to be precisely correlated to increased expenditures—which would be nearly impossible to accomplish—and it does not require them to be distributed pro rata based on population. The Act requires only that the allocation of

22

funds be “based on increased expenditures.” 42 U.S.C. § 801(c)(7). The D.C. Circuit has explained that, when a statute directs an agency to make a decision “based on” a particular factor or condition, that decision need “not necessarily . . . rest solely on” that factor. Sierra Club v. EPA, 356 F.3d 296, 306 (D.C. Cir.), amended, No. 03-1084, 2004 WL 877850 (D.C. Cir. Apr. 16, 2004). In other words, the term “based on” does not restrict the agency’s consideration to the statutory factor alone, but the agency may not “wholly abandon[]” it. Id.; see also Anna Jaques Hosp. v. Sebelius, 583 F.3d 1, 5–6 (D.C. Cir. 2009) (construing the phrase “on the basis of” in a statute, noting its similarity to the phrase “based on” because “both phrases use a variant of the word ‘base,’” and concluding that “‘[b]asis does not mean ‘entire’ or ‘only’”); Catawba County v. EPA, 571 F.3d 20, 37 (D.C. Cir. 2009) (“[W]e have repeatedly held that the words ‘based on’ are . . . ambiguous: they neither compel the agency to rest its decisions solely on the specified factor nor indicate the extent to which the agency may rely on additional factors. . . . Instead, they simply constrain the agency from abandoning or supplanting the specified factor altogether.” (alterations and internal quotation marks omitted)). Thus, the CARES Act does not require Treasury to consider only increased expenditures. It instead requires Treasury to consider increased expenditures as a “starting point” or “primary basis.” Int’l Union v. Mine Safety & Health Admin., 626 F.3d 84, 92 (D.C. Cir. 2010) (citing Sierra Club, 356 F.3d at 306).

Here, when Treasury first developed its 2020 allocation methodology, it selected what Plaintiffs agree to be a reasonable proxy for increased expenditures: population. 2020 Allocation Methodology at 2. It did so knowing that “any methodology, by necessity, must rely on a rough estimate of any given tribe’s expenditures.” Defs.’ Reply at 9. That decision was made against a backdrop in which Congress required the Secretary to distribute funds with exceptional speed—

23

within 30 days. See 42 U.S.C. § 801(b)(1) (“[N]ot later than 30 days after March 27, 2020, the Secretary shall pay each . . . Tribal government . . . the amount determined . . . .”); CARES Act, Pub. L. No. 116-136, 134 Stat. 281 (2020) (passed on March 27, 2020). Treasury made a misstep in using IHBG data as the source of population counts for certain Tribes, including the two Plaintiffs here, but its 2021 methodology sought to cure the shortfall created by that judgment. The second time around, it again relied on population data: using both the IHBG population counts and actual Tribal enrollment, Treasury sought to estimate the degree to which the IHBG population–based count had underestimated increased expenditures for undercounted Tribes and compensate those Tribes accordingly. That approach was not an arbitrary and capricious approximation of the statutorily required benchmark of “increased expenditures.”

The Tribes appear to argue that Treasury could choose no reallocation methodology other than a pro-rata methodology that simply replaced IHBG population with actual Tribal enrollment. See Pls.’ Reply at 3. But the CARES Act does not require Treasury to adopt any particular methodology, let alone the pro-rata approach preferred by Plaintiffs. Importantly, by the time Treasury was developing its 2021 methodology, the landscape with respect to available CARES Act funding had changed dramatically. Treasury had already distributed most of the $8 billion that Congress appropriated to Tribal governments, and what remained was a pool of available funds meant for ANCs, pending the outcome of litigation before the Supreme Court. See 2021 Reallocation Methodology at 3 (“The question of ANCs’ eligibility for [Coronavirus Relief Fund] funding is currently before the Supreme Court.”). At that point, Treasury “[could] not pay each Tribe what it would have received if Treasury had used enrollment data from the start. There simply [were] not enough remaining funds to do so.” Defs.’ Cross-Mot. at 18. In view of that

24

impossibility, Treasury decided to pay the Tribes with the worst IHBG-to-enrollment ratios the most, tracking the D.C. Circuit’s reasoning in Shawnee and the Plaintiffs’ complaints. Id. That decision was consistent with the agency’s discretion to select a model that, in its estimation, was faithful to the intent of Congress. See Appalachian Power Co., 249 F.3d at 1052 (“[Courts] must defer to the agency’s decision on how to balance the cost and complexity of a more elaborate model against the oversimplification of a simpler model.”); Kennecott Utah Copper Corp. v. U.S. Dep’t of the Interior, 88 F.3d 1191, 1225 (D.C. Cir. 1996). The court therefore concludes that the 2021 methodology for allocating Title V funds was “based on” increased expenditures.

3. The Phase-Out

Plaintiffs next argue that “Treasury’s decision to allocate awards” pursuant to a “phase-out method” that was “clearly contradicted by the actual enrollment data that it had available was arbitrary and capricious.” Pls.’ Second MSJ at 15. The court understands Plaintiffs to be referring to the fact that, once Treasury identified the 15% of Tribes with the highest population-to- enrollment ratios, it awarded a percentage of a Tribe’s “enrollment allocation”—what the Tribe “would have received [if] Treasury had used enrollment data, rather than IHBG data, from the start”—on a descending scale. Defs.’ Reply at 7–8. “The qualifying tribes with the lowest ratios received a significant portion of their enrollment-allocation, and conversely, the qualifying tribes ranked toward the bottom of the ‘top 15%’ received less significant portions of their enrollment allocations.” Id. at 8. Plaintiffs argue that Treasury did not have the authority to “pick new ‘winners’ from the top 15% of ‘losers’ from the 2020 Distribution,” instead of “allocat[ing] the new awards based on an enrollment-based allocation,” especially since Tribal enrollment data was available to it and Tribal leaders had previously communicated to Treasury “the wisdom and value

25

of using enrollment data.” Pls.’ Second MSJ at 14. Notably, Plaintiffs do not challenge Treasury’s decision to “pick . . . the top 15% of ‘losers’ from the 2020 Distribution,” which benefited the Plaintiff Tribes by reducing the number of Tribal governments eligible for what little funding remained. Rather, their arguments attach to the descending percentage of awards of those Tribes selected for further compensation. See id.; see also Pls.’ Reply at 13.

The thrust of Plaintiffs’ argument is that “Treasury’s ‘phase-out’ method led to absurd results.” Pls.’ Second MSJ at 15. For example, Shawnee (which was assigned an IHBG population count of zero) received an award of $5,200,000 pursuant to the 2021 reallocation and the minimum award of $100,000 pursuant to the 2020 allocation—around $5.3 million in total for its 3,021 members. Id. By contrast, Prairie Band Potawatomi, which had been assigned an IHBG population count of 747, received around $3.3 million in total funding for 4,561 enrolled members across the two distributions. Id. All told, Shawnee received around $2 million more with around 1,500 fewer enrolled members: “40% more funds to Shawnee [than Prairie Band Potawatomi] for an enrolled population that is 33% lower.” Id. These “absurd results,” Plaintiffs argue, contradicted the “actual enrollment data that [Treasury] had available,” rendering Treasury’s decision to use the phase-out method arbitrary and capricious. Id.

Defendants respond that Treasury could not pay each Tribe a “fixed, meaningful amount per ‘uncounted enrollee’” while at the same time reserving adequate funds for ANCs, and that the phase-out method was a reasonable alternative given the circumstances. Defs.’ Cross-Mot. at 16. Because of the limited pool of funds available for the reallocation, “Treasury had to use its judgment to decide not only which Tribes would receive supplemental payments, but also the amount of each payment.” Id. Treasury decided that only the Tribes with the “most substantial

26

disparities” would be eligible for reallocation payments. 2021 Reallocation Methodology at 3. From there, Defendants submit, the same reasoning that justified setting only the top 15% of Tribes (by population-to-enrollment ratio) as eligible for payment justified phasing out the percentage of the difference between population- and enrollment-based pro-rata awards. Defs.’ Reply at 15.

As noted, Plaintiffs agree that “Treasury was well within its rights to limit the supplemental payments to the 15% of Tribes that had been severely undercounted by the IHBG population data,” because “an agency is generally afforded discretion in line drawing.” Pls.’ Reply at 13. But they nonetheless argue that it was unreasonable for Treasury to distribute funds within the 15% of Tribes using a phase-out method and not a set per-enrollee amount. Pls.’ Second MSJ at 14–15. This contention exposes the fundamental flaw in Plaintiffs’ reasoning. As Defendants point out, “[i]f Treasury cannot treat tribes within the top 15% differently based on IHBG-to-enrollment ratios, then it presumably cannot treat any tribes—top 15% or not—differently based on these ratios.” Defs. Reply at 15. Plaintiffs do not offer a principled distinction between using the ratios to set the 15% cut-off and using the ratios to implement a phase-out method between Tribes within that group, and indeed their arguments against the phase-out method prove too much. Plaintiffs accept that given the “relatively limited funding available,” Treasury had to focus on ameliorating the starkest disparities among Tribes based on the 2020 allocation methodology and draw a line at the 85th percentile of enrollment-to-population ratios. Pls.’ Reply at 13. Yet that is the same rationale Treasury drew on to land on the phase-out method: Treasury could not afford to make all undercounted Tribes whole at the per-member rate of the 2020 methodology, swapping enrollment in for IHBG population. So it used the phase-out method to, again, focus on ameliorating the starkest disparities. Not surprisingly, under that approach, a zero-count Tribe like Shawnee came

27

out better than undercounted Tribes like Prairie Band Potawatomi. It cannot be that the 15% cut- off was not arbitrary and capricious but the phase-out violated the APA. 6 In sum, Treasury used the population-to-enrollment ratio to “identify the . . . tribes that were likely most harmed by the IHBG data, and to what degree.” Defs.’ Reply at 15. Plaintiffs would have this court hold that Treasury was within its discretion to use the ratios to figure out the first part of the new methodology—which Tribes were likely most harmed by IHBG data (the top 15%)—but not the second part—“to what degree” the Tribes were harmed (the degree of disparity, which the phase-out method targets by ameliorating the starkest disparities the most). Plaintiffs have simply offered no real account for how both of those things can be true at the same time. The court agrees with Defendants that “[i]f [the] ratios . . . justify drawing distinctions among tribes to determine which ones should receive supplemental payments,” then “those same ratios should justify drawing distinctions among the qualifying tribes to determine how much each should receive.” Id. It may be true that Treasury “needlessly overcomplicated its methodology . . . by phasing out payments,” Pls.’ Reply at 3, but it was within its discretion to do so, and that judgment was not arbitrary and capricious.

4. Different Treatment of Similarly Situated Tribes To some extent, Plaintiffs’ third argument overlaps with their second (regarding the phase-

out method), and indeed the parties treat the two interchangeably at times. As a matter of

6 Plaintiffs argue that Treasury could have granted supplemental awards to only the top 15% on a pro-rata basis, distributing the same total amount of Title V funds in the reallocation phase. See Pls.’ Reply at 13–14. But the question is not whether Treasury chose the best methodology; it is whether Treasury chose a methodology that was not arbitrary and capricious. Treasury has explained that, in addition to seeking to prioritize Tribes facing the greatest funding disparities, it chose the phase-out method in part to avoid a “sharp cliff” in which a Tribe “just below the 15% cut-off would receive no supplemental payment, whereas the Tribe ranked just one spot higher . . . would receive the same ‘per enrollee’ rate as the tribe at the top of the list.” Defs.’ Reply at 15.

28

administrative law, agencies “must treat similar cases in a similar manner unless [they] can provide a legitimate reason for failing to do so.” Kreis v. Sec’y of the Air Force, 406 F.3d 684, 687 (D.C. Cir. 2005); see also Westar Energy, Inc. v. FERC, 473 F.3d 1239, 1241 (D.C. Cir. 2007) (describing the “fundamental norm of administrative procedure [that] requires an agency to treat like cases alike”). Plaintiffs maintain that Treasury treated similarly situated Tribes differently, and that there was “no legitimate basis” for doing so. Pls.’ Second MSJ at 15–16. They identify two dimensions of differential treatment: First, they point to the way that Treasury distinguished them from other Tribes that received more significant Title V funds per initially uncounted member after the reallocation. Pls.’ Second MSJ at 16. Second, they point to the “disparities” between their own Title V awards and those of the Tribes that “received no May 2021 Distribution because they were not severely undercounted.” Id. This different treatment, they say, was arbitrary and capricious. Pls.’ Second MSJ at 15.

Treasury counters that it has adequately explained any different treatment of similarly situated parties. See Petroleum Commc’ns, Inc. v. FCC, 22 F.3d 1164, 1172 (D.C. Cir. 1994) (noting that agencies may “treat[] similarly situated parties differently” when they “provide adequate explanation” for doing so). “Treasury could not both pay each Tribe a fixed, meaningful amount per uncounted enrollee, and also reserve a sufficient amount for ANCs,” and so it “focus[ed] limited funds on the Tribes whose IHBG figures were especially low in comparison to their enrollment figures, and who initially received the lowest population-based payments.” Defs.’ Cross-Mot. at 16–17 (internal quotation marks omitted). Plaintiffs respond that the administrative record does not support this explanation and that it is demonstrably false. The court disagrees.

29

First, Plaintiffs argue that the administrative record does not support Treasury’s contention that it endeavored to compensate based on degree of harm resulting from use of the IHBG data. Rather, Plaintiffs contend, Treasury “repeatedly [drew] similarities between Tribes that were assigned zeros and those other Tribes . . . that were significantly undercounted, never once suggesting that one was more harmed than the other.” Pls.’ Reply at 7. In the 2021 reallocation methodology announcement, for example, Treasury identified two scenarios in which IHBG population data “may prove insufficient” as a proxy for increased expenditures: (1) where a Tribe has a “formula area population of zero,” and (2) where it has a non-zero IHBG population that is nonetheless not “sufficiently accurate.” 2021 Reallocation Methodology at 2. Plaintiffs observe that such language does not explicitly state that the degree of harm to Tribes within these two categories is any different, therefore justifying only “treating the severely undercounted Tribes the same relative to one another.” Pls.’ Reply at 7. Treasury’s attempts to draw distinctions in the degree of harm, they insist, are mere post-hoc rationalizations that cannot be credited. Id. (citing Nat’l Black Media Coal. v. FCC, 775 F.2d 342, 354 (D.C. Cir. 1985)).

Plaintiffs fail to read the administrative record as a whole. The administrative record recognizes, of course, that many Tribes facing a significant disparity between IHBG population and enrollment were undercounted by the 2020 allocation methodology. Yet, at the same time, it contains numerous references to the particular harm faced by Tribes initially counted as having a population of zero. For example, the allocation methodology announcement states that it “is particularly true” that IHBG counts are an insufficient proxy for increased expenditures of Tribes with a zero-population count. 2021 Reallocation Methodology at 2 (emphasis added). The same document explains that “[w]here there is an especially large disparity between formula area

30

population and enrollment figures”—as is more likely the case for zero-population Tribes—“the difference suggests that the Tribal government has a need for [additional] funding,” and that the reallocation would focus on these Tribes. Id. (emphasis added). Treasury also explained that the “population-to-enrollment ratio” would be used to determine which Tribes were previously “significantly undercounted” and had the “greatest need.” Id. at 3. This language implies the comparison or ranking of Tribes’ respective disparities and makes clear that Treasury’s explanation is not a mere post-hoc rationalization.

Nor does the administrative record belie Treasury’s purported reliance on Shawnee.

Plaintiffs charge that Treasury cannot rely on Shawnee to support its explanation because the administrative record shows that Treasury was trying to “rectify a broader problem than the one specifically addressed in Shawnee”—specifically, the harm to Tribes with non-zero counts. Pls.’ Reply at 7–9. But that argument ignores that post-Shawnee, Treasury was confronted with suits from both zero-count Tribes (Shawnee and Miccosukee) and an undercounted Tribe (Prairie Band Potawatomi), and it faced uncertainty as to whether other similar actions would follow. Thus, there is no inconsistency between Treasury wanting to address the harm to all Tribes that were undercompensated due to use of the IHBG data (both zero-count and undercounted Tribes), and targeting its remaining resources to redress the starkest harms to the greatest extent (consistent with the D.C. Circuit’s focus on zero-count Tribes). Nothing in the D.C. Circuit’s decision in Shawnee is incompatible with such an approach.

Plaintiffs next argue that “Treasury’s argument that Tribes that were assigned IHBG populations of zero were likely underpaid the most is demonstrably wrong based on [the enrollment data] that Treasury had when it made its decision.” Pls.’ Reply at 8 (alteration omitted)

31

(internal citation and quotation marks omitted). In other words, because Treasury knew exactly how much each Tribe was undercounted based on its own enrollment data, no guesswork was required in identifying the Tribes most adversely impacted by the initial allocation. Treasury therefore could easily have allocated funds on a pro-rata basis based on undercounted enrolled members. This argument is flawed. The CARES Act itself requires no particular formula so long as the formula selected is “based on” increased expenditures, see 42 U.S.C. § 801(c)(7), and, as this court has already concluded, there was no requirement that Treasury distribute Title V funds the second time around on a per–uncounted enrollee basis. In 2020, Treasury chose a pro-rata distribution based on population as a proxy for expenditures, but it was not required to apply that same formula in 2021 when circumstances had changed dramatically. The benchmark for the 2021 reallocation methodology was not its fidelity to the 2020 methodology but its relationship to increased expenditures, and the court has already concluded that the 2021 reallocation methodology was “based on” increased expenditures resulting from the COVID-19 pandemic.

Plaintiffs make much of the fact that the 2021 reallocation methodology overfunds some Tribes relative to others, and they suggest that a per-enrollee reallocation methodology would have avoided that result. But, as Defendants assert, “there is no indication that each tribe’s relative expenditures correlated perfectly with its enrollment count,” and any methodology would inevitably overfund some Tribes and underfund others relative to their actual increased expenditures. Defs.’ Reply at 8–9; see id. at 9 (identifying variables that could “impact[] any given tribe’s relative COVID-related expenditures, beyond simply its raw enrollment figure”). No methodology could avoid that reality, particularly not within the short timeframe prescribed by Congress for distributing the funds. Id. at 8–9. In 2021, Treasury targeted the harm Tribes faced

32

by being undercounted and underpaid in the first round of Title V funds; it was free to meet that harm the way it did.

Plaintiffs cite in their briefs this court’s observation in an earlier opinion that the “yawning disparity in per-uncounted-member funding [among the three Plaintiff Tribes] is puzzling to say the least.” Pls.’ Reply at 1; Mem. Op. Denying PI Modification at 7. When the court made that observation, however, Treasury had not offered any “defense or explanation for the seemingly inequitable results.” Pls.’ Reply at 1. It has now become clear, however, that these “inequit[ies]” are byproducts of the complex procedural history of this case and not intentional differential treatment by Treasury. It remains true—and surprising in some sense—that Prairie Band Potawatomi received, taking both awards together, around $3.3 million for its 4,561 members, while Shawnee received $2 million more for around 1,500 fewer members. Pls.’ Second MSJ at 15. But Prairie Band Potawatomi, unlike Shawnee, received some funds over the minimum amount in the initial allocation. See id. Shawnee was among the worst off after the IHBG-based allocation; it is not unreasonable that it would receive the most significant catch-up payment during the reallocation phase. This may seem inequitable at first blush, but such disparity is a function of a disbursement methodology that, when viewed in light of all relevant circumstances, was entirely reasonable.

* * *

In an ideal world, Treasury might have distributed Title V funds to all Tribal governments pro rata based on the Tribes’ enrollment data, so that no Tribe could claim underpayment. But that is not the world in which we find ourselves. Treasury was faced with the difficult challenge of how to best mitigate the harm caused to some Tribes by its use of flawed IHBG data to distribute

33

funds, as well as the passage of time and the dissipation of Title V funds. This court cannot say that Treasury’s attempts in 2021 to identify the Tribes most harmed by the original methodology and to grant them the greatest degree of relief is arbitrary and capricious, even if some Tribes ended up worse off than if Treasury had simply used better data in 2020. V. CONCLUSION For the foregoing reasons, Plaintiffs’ motions for summary judgment, ECF No. 70; ECF No. 88, are denied. Defendants’ cross-motion, ECF No. 92, is granted in full.

A final, appealable order accompanies this Memorandum Opinion.

Dated: January 28, 2022 Amit P. Mehta United States District Court Judge