The following post was contributed by a HarvardCRCL.org executive board member. The article offers a critique of the recent Second Circuit decision in Shaknes v. Berlin.
In Gonzaga University v. Doe,[1] the Supreme Court set a high bar for finding that a statute creates a private right of action under 42 U.S.C. § 1983.[2] But the Court has not addressed whether regulations can establish enforceable rights. Some federal courts of appeals have concluded that a regulation cannot itself create a right but “may [still] be relevant in determining the scope of [a] right conferred [in a statute]” insofar as it amounts to “an agency interpretation of [that] statute.”[3] In multiple cases, the Second Circuit has declined to answer this question, deciding the issues before it assuming that those circuits’ position is correct.[4] Recently, in Shakhnes v. Berlin,[5] the court considered whether a particular regulation was a “valid interpretation” of a statutory right.[6] Proceeding on the assumption that the statute at issue defined a right with a flexible standard, the court held that a related regulation was actionable as a “valid interpretation” under 42 U.S.C. § 1983 even though it imposed a rigid rule where none existed before.[7] Unfortunately, the decision provides no clear principle by which future courts may decide when a regulation interprets a statute in the context of 42 U.S.C. § 1983. Meanwhile, a Seventh Circuit opinion resolves an analogous issue in the context of notice-and-comment rulemaking with a principled distinction between rule interpretation and rule creation.[8] It is even consistent with the precedent upon which Shakhnes relies. The Second Circuit would have established a more manageable test for questions like the one at issue in Shakhnes had it drawn a similar distinction.
The provision of the Medicaid Act[9] at issue in Shakhnes requires states administering Medicaid programs to “grant[] an opportunity for a fair hearing before the State agency to any individual [in certain circumstances].”[10] A regulation further describing the Medicaid fair hearing process provides that “[t]he agency must take final administrative action . . . within 90 days” of the request of a fair hearing except under particular circumstances.[11]
The Shakhnes plaintiffs were a class of Medicaid applicants and recipients who had requested, or would request, fair hearings to challenge unfavorable decisions by the defendants, state and local agencies administering the program. The plaintiffs alleged that the Medicaid agencies violated their statutory fair hearing rights by failing to take action within the 90-day post-hearing window set forth in a regulation accompanying the statute.[12]
Applying Gonzaga, the district court decided that the statutory provision at issue established a right to a fair hearing and a right to agency action at some point after such a fair hearing.[13] In light of the accompanying regulation, it also held that the plaintiffs had a right to agency action within 90 days of the fair hearing.[14] The court ordered declaratory and injunctive relief.[15] On appeal, the defendants argued in part that the plaintiffs had no enforceable right to agency action within 90 days because the regulation containing the 90-day deadline was not a valid interpretation of the statutory right.[16] The Second Circuit affirmed as to the scope of the statutory right and vacated and remanded for further proceedings on other grounds.[17] Writing for the panel, Judge Straub concluded that the regulation’s 90-day deadline was enforceable as an interpretation of the statute’s fair hearing guarantee.[18]
The court purported to apply a standard “well-settled” in precedent to decide the scope of what 42 U.S.C. § 1983 rendered actionable.[19] To illuminate the standard, it first looked to the Supreme Court’s decision in Wright v. City of Roanoke Redevelopment & Housing Authority.[20] In Wright, the plaintiffs sued a housing authority under 42 U.S.C. § 1983 for billing more than thirty percent of each plaintiff’s income for the total of his or her rent and utilities.[21] A statute provided that “tenants could be charged as rent no more and no less than 30 percent of their income,” while a corresponding “regulation[] . . . expressly required that a reasonable amount for utilities be included in rent.”[22] The Court decided that the plaintiffs had an enforceable right to a calculation of “rent” that included the cost of utilities because the regulation merely defined the ambiguous term “rent.”[23]
Next, the panel turned to D.D. v. New York City Board of Education,[24] in which parents, suing on behalf of their children, alleging that delays in delivery of special education services in a New York City public school violated their children’s rights to such services.[25] The Individuals with Disabilities Education Act (IDEA) required “special education . . . services [to be] provided in conformity with [an] individualized education program [IEP].”[26] A corresponding regulation required that IEPs be “implemented as soon as possible” following their finalization.[27] The court held that the school system’s failure to implement the IEPs “as soon as possible” pursuant to the regulation’s requirement violated the plaintiffs’ right to IEP-conforming services because the regulation “define[d] the scope” of the statutory right.[28]
Finally, the court turned to another circuit’s reasoning about the enforceability of a regulation and certain related Medicaid Act provisions (all different from the provisions considered in Shakhnes).[29] In Harris v. James,[30] the plaintiffs sued under 42 U.S.C. § 1983, arguing that the state violated their right to transportation to and from Medicaid service providers by failing to provide this transportation.[31] The Harris court considered whether the guarantee of such transportation in a regulation[32] was enforceable under 42 U.S.C. § 1983, given that various statutory provisions in the Medicaid Act arguably guaranteed certain related rights.[33] The Act required, for example, that the state agency “provide medical assistance ‘with reasonable promptness.’”[34] The Shakhnes court reasoned that the regulation went “beyond explicating the specific content of the statutory provision and impose[d] distinct obligations[; accordingly it was] too far removed from Congressional intent to constitute a federal right enforceable under § 1983.”[35]
These three decisions that Shakhnes addressed were all appropriate analogies because each considered whether a regulation was a valid interpretation of — or went beyond interpreting — a statute. But the court did not adequately derive a unifying principle from these cases. Instead, it merely quoted conclusory language to support its conclusion. In discussing Wright, the Shakhnes court adverted to the Third Circuit’s observation about the case that “the regulation at issue . . . merely defined the specific right that Congress already had conferred through the statute.”[36] In reviewing D.D., the panel quoted multiple times the conclusion that the regulation “merely defines the scope” of the statutory right.[37] It never explained further what it was about the regulation’s relation to the statute that led the court to its conclusion. In distinguishing the case at bar from Harris, the court relied on the Eleventh Circuit’s statement that the regulation considered in that case could not “‘reasonably [be] understood to be part of the content of’ [the allegedly predicate] statutory right,”[38] and that the regulation went “beyond explicating the specific content of [a] statutory provision.”[39] Unfortunately, none of the language quoted from these cases meaningfully guides future courts deciding whether a regulation is a valid interpretation of a statutory right. Instead, each quotation effectively restates the issue that it purports to resolve — that is, it begs the question.
However, a principled distinction does exist between a regulation that is a valid interpretation of a statutory right and one that imposes new obligations. Although not rendered in the context of private rights of action under 42 U.S.C. § 1983, a decision from the Seventh Circuit explains just this distinction. The issue before the court in Hoctor v. Department of Agriculture[40] was whether an agency’s internal memorandum could bypass the Administrative Procedure Act’s notice-and-comment rulemaking requirements thanks to a provision in the Act exempting agency statements that merely interpret prior regulations.[41] The unchallenged regulation in Hoctor required that a “facility [housing animals] be constructed of such material and of such strength as appropriate for the animals involved [and that t]he indoor and outdoor housing facilities . . . be structurally sound and . . . maintained in good repair to protect the animals from injury and to contain the animals.”[42] The year after the agency adopted this regulation, it issued an internal memorandum requiring that “all dangerous animals, defined as including, [e.g.,] lions, tigers, and leopards, . . . be [kept] inside a perimeter fence at least eight feet high.”[43] Hoctor kept a variety of such animals inside a perimeter fence six feet high.[44] The Department of Agriculture sanctioned Hoctor for the inadequacy of his fence, although not because the fence was impermissible under the (unchallenged) regulation.[45] Rather, “[t]he only ground on which the Department defend[ed] sanctioning Hoctor . . . [wa]s that requiring an eight-foot-high perimeter fence for dangerous animals is an interpretation of the Department’s own structural-strength regulation.”[46]
The Seventh Circuit panel explained that the rule was not a valid interpretation of the regulation and that it was instead a novel rule inspired by the regulation.[47] Resolving this familiar question about the legislative/interpretive divide in administrative law, the court distinguished statements that “particularize [prior statements of law] through interpretation” from those that “impose a[n independent] duty” by reasoning that strict numerical limits generally cannot be derived from flexible standards.[48] This objection to the alleged “interpretation” is intuitive: that a fence must be eight feet high does not obviously follow from the statute’s requirements since, absent further evidence, a fence of such height cannot be expected to meet the regulation’s goals substantially better than a fence one inch shorter.[49] The particular height of eight feet has no special significance. A rule like this, not derived from the underlying statute, is “legislative” (or “substantive”).[50]
But numerical limits can play a part (albeit a bounded one) in interpretation:
[h]ad the Department of Agriculture said in the internal memorandum that it could not imagine a case in which a perimeter fence for dangerous animals that was lower than eight feet would provide secure containment, and would therefore presume, subject to rebuttal, that a lower fence was insecure, it would have been on stronger ground. For it would have been tying the rule to the animating standard, that of secure containment, rather than making it stand free of the standard, self-contained, unbending, arbitrary. To switch metaphors, the ‘flatter’ a rule is, the harder it is to conceive of it as merely spelling out what is in some sense latent in a statute or regulation, and the eight-foot rule in its present form is as flat as they come.[51]
The understanding of the legislative/interpretive divide described in Hoctor boils down to this: one cannot generally interpret an existing command by imposing an inflexible rule where no such rule already existed. If the prior command includes some vagueness, resolution of that vagueness into a strict rule is a legislative (not merely an interpretive) function.
The Shakhnes defendants failed to make a clear Hoctor-style argument to the court. At times their brief seems to suggest that as between “rigid” and “flexible” language in putatively interpretive regulations, only the latter is genuinely “tied . . . to the underlying statute itself.” [52] But the brief does not elaborate on this point. At oral argument, too, counsel for the defendants failed to explain this distinction.
Despite counsel’s failure to raise this precise argument, the Second Circuit might have drawn the Hoctor distinction on its own. Had it done so, it would still have remained faithful to Wright, D.D., and Harris. Applying the reasoning in Hoctor leaves the holdings of all three of those cases undisturbed. In Wright, the regulation interpreted a statutory requirement that a family “‘shall pay as rent’ a specified percentage of its income” to encompass payment not just for “the use of the dwelling accommodation” but also for “reasonable amounts of utilities.”[53] It did not veer from the underlying statutory guarantee that a certain amount will be paid for rent. It did not impose a strict rule where the statute was vague. It simply interpreted the meaning of the ambiguous term “rent.” In D.D., the Second Circuit enforced a flexible interpretation (“as soon as possible”) of a vague standard (“special education . . . services [shall be] provided in conformity with the [IEP]”).[54] This, too, is permitted under Hoctor. The “as soon as possible” standard imposes no new absolute requirements. It infers from the statutory guarantee of services that those services will come quickly. It leaves the issue whether a given IEP has been implemented soon enough to triers of fact in individual lawsuits. It does not make a legislative determination that after, say, two weeks, a school has failed to provide “special education . . . services in conformity with the [IEP].”[55] In sum, the regulations at issue in Wright and D.D. avoid imposing new requirements. The regulation in Wright interprets a specific (but ambiguous) statutory guarantee by imposing a specific regulatory guarantee. The regulation in D.D. interprets an open-ended statutory guarantee with an open-ended regulation.
Unlike the regulations in Wright and D.D., the regulation in Harris falls on the legislative side of the legislative/interpretive divide under the reasoning in Hoctor. In Harris, the plaintiffs pointed to several statutory provisions that made broad, standard-based guarantees — for example, that state agencies would provide services “promptly.”[56] They argued that a regulation describing a right to transportation to medical service providers interpreted these statutory provisions.[57] The Eleventh Circuit disagreed, finding the nexus between the statutory and regulatory guarantees to be “too tenuous.”[58] The Harris court respected the legislative/interpretive divide under the Hoctor approach because it held that a regulation that imposed a strict rule was not sufficiently related to statutory provisions that imposed no such requirements.
Although none of these cases cites Hoctor, they all comply with the Seventh Circuit’s distinction between rule interpretation and rule creation, which, admittedly, Hoctor draws in the context of administrative law. None calls it “interpretation” when a regulation imposes a rigid rule although the underlying statute imposes none itself.
The legislative/interpretive distinction described in Hoctor is readily applicable to the facts of Shakhnes. The underlying statute guarantees “an opportunity for a fair hearing.”[59] This statute imposes no strict requirements as to the timing of post-hearing agency action. Thus, under Hoctor, an enforceable timing-oriented regulation that is a valid interpretation of this statute would itself have to refrain from imposing any strict timing requirement. Such a regulation could provide a qualified deadline for administrative action: “in general, within 90 days,” or “except under extenuating circumstances, within 90 days.”[60] Alternatively, it could interpret the “opportunity for a fair hearing” to imply a general right to subsequent agency action “within a reasonable time.”[61] But the actual regulation’s inflexible timing requirement — “within 90 days”[62] — fails the Hoctor test: it imposes a strict rule where none existed before. Accordingly, had it followed the distinction just described, the Shakhnes court would have held that the regulation at issue imposes an independent duty where the statute is silent and is thus not enforceable under 42 U.S.C. § 1983.
Unfortunately, the Shakhnes court noted no such distinction between rule interpretation and rule creation. Instead, the Second Circuit rested its decision primarily on unhelpful language from prior decisions.[63] It ultimately held that “the 90-day requirement in the regulation here at issue merely defines the time frame with respect to Plaintiffs’ right to an opportunity for Medicaid fair hearings.”[64] But the opinion did not distinguish meaningfully between regulations that “merely define” the content of a statutory right and those that impose additional obligations.
Of course, this decision appears to be a particular boon for civil rights plaintiffs, whether suing individually or in class actions. Shakhnes affirms that no plaintiff need prove that a particular delay deprived her of a right when suing under this regulation; it is hereafter settled that after 90 days, a violation of a right has occurred. Even defendants will benefit from the Second Circuit’s settling of the issue by confirming that a bright-line rule determines the scope of the right: the decision forecloses thorny disputes about whether an agency was “reasonabl[y] prompt[]” in a given case and the resulting possibility that experts and discovery will be needed to answer the question. This should reduce litigation expenses for all parties in such cases.
But while the Hoctor rule draws a clean line forbidding imposition of strict rules where none existed prior, the decision provides little guidance for future courts inquiring whether a regulation is a valid interpretation of a statute. The reasoning leaves judges unconstrained when deciding whether a given regulation does or does not “interpret” the content of an underlying statutory right in future cases. This is a blow to the rule of law.
The decision also imposes other costs. While it may increase efficiency in a small set of cases, it also invites regulators to tangle with the creation, modification, or elimination of rights independently of congressional control. This is a bargain of dubious worth. Under this scheme, agencies may change the scope of individual rights markedly from administration to administration, and Congress may be tempted to relinquish more control of policymaking in the future.
The Shakhnes court would have been better off leaving the plaintiffs free to vindicate their statutory rights and holding, following the reasoning in Hoctor, that the regulation’s 90-day deadline imposed an independent obligation and was thus unenforceable under 42 U.S.C. § 1983.
[1] 536 U.S. 273 (2002).
[2] 42 U.S.C. § 1983 (2012).
[3] Save Our Valley v. Sound Transit, 335 F.3d 932, 939 (9th Cir. 2003) (emphasis added). See also Johnson v. City of Detroit, 446 F.3d 614, 629 (6th Cir. 2006).
[4] See, e.g., D.D. v. N.Y.C. Bd. of Educ., 465 F.3d 503, 513 (2d Cir. 2006).
[5] 689 F.3d 244 (2d Cir. 2012).
[6] Id. at 253 (quoting Harris v. James, 127 F.3d 993, 1011 (11th Cir. 1997)) (internal quotation marks omitted).
[7] Id. at 254.
[8] See Hoctor v. Dep’t of Agric., 82 F.3d 165 (7th Cir. 1996).
[9] 42 U.S.C. §§ 1396–1396w-5 (2012).
[10] 42 U.S.C. § 1396a(a)(3) (2012).
[11] 42 C.F.R. 431.244 (f)(1) (2002).
[12] See Shakhnes, 689 F.3d at 247.
[13] Id. at 249.
[14] Id.
[15] Id.
[16] Id. at 251. The admittedly imprecise term “valid interpretation” here is drawn from Shakhnes (which itself derives the language from Harris). See supra note 6. A “valid” interpretation of a right is one that supposedly goes no further than the underlying right and is consequently enforceable under 42 U.S.C. § 1983; an “invalid” interpretation is one that imposes distinct rights or obligations (and is thus unenforceable). This useful shorthand is present throughout this Comment, as is an analog in the notice-and-comment context. See infra text accompanying notes 40–51.
[17] Id. at 263.
[18] See id. at 251–55.
[19] Id. at 251.
[20] 479 U.S. 418 (1987). See Shakhnes, 689 F.3d at 251–52.
[21] 689 F.3d at 251–52 (citing 479 U.S. at 419–20).
[22] See id. at 252 (quoting 479 U.S. at 430) (internal quotation marks omitted).
[23] See id. (citing 479 U.S. at 431 n. 11).
[24] D.D. v. N.Y.C. Bd. of Educ., 465 F.3d 503 (2d Cir. 2006).
[25] See 689 F.3d at 252–53.
[26] Id. at 252 (quoting 465 F.3d at 512) (internal quotation marks omitted).
[27] Id. (emphasis in original) (quoting 465 F.3d at 512–13) (internal quotation marks omitted).
[28] Id. at 253 (quoting 465 F.3d at 513) (internal quotation marks omitted).
[29] See id. at 253–54.
[30] 127 F.3d 993 (11th Cir. 1997).
[31] See Shakhnes, 689 F.3d at 253 (citing 127 F.3d at 996).
[32] 42 C.F.R. § 431.53 (1991) (requiring that a state’s “Medicaid agency . . . ensure necessary transportation for recipients to and from providers”).
[33] See 689 F.3d at 253.
[34] Id. (quoting 127 F.3d at 1005).
[35] 127 F.3d at 1009 (internal quotation marks omitted).
[36] Shakhnes, 689 F.3d at 252 (emphasis added) (quoting S. Camden Citizens in Action v N.J. Dep’t of Envtl. Prot., 274 F.3d 771, 783 (3d Cir. 2001)) (internal quotation marks omitted).
[37] See id. at 253, 254 (quoting D.D. v. N.Y.C. Bd. of Educ., 465 F.3d 503, 513 (2d Cir. 2006)) (internal quotation marks omitted).
[38] Id. at 255 (emphasis and first alteration in original) (quoting Harris, 127 F.3d at 1012).
[39] Id. at 253 (quoting Harris, 127 F.3d at 1009).
[40] 82 F.3d 165 (7th Cir. 1996). Of course, other cases have treated this distinction in the same (rulemaking) context. See, e.g., Chamber of Commerce v. Dep’t of Labor, 174 F.3d 206 (D.C. Cir. 1999).
[41] See 82 F.3d at 167; see also 5 U.S.C. § 553(b)(A) (2012).
[42] 9 C.F.R. § 3.125(a) (1995).
[43] 82 F.3d at 168 (internal quotation marks omitted).
[44] See id. at 169.
[45] Id.
[46] Id.
[47] See id. at 172.
[48] Id. at 169–70 (arguing that “[a] rule that turns on a number is likely to be arbitrary in th[e] sense” that it is merely a “choice among methods of implementation”).
[49] The general rule is subject to rebuttal by evidence that a particular rule is “uniquely appropriate” to meet the underlying standard. Id. If, for example, evidence showed that lions, tigers, and leopards can typically escape from fences seven feet, eleven inches high because of their jumping abilities, but not from fences one inch taller, the regulation would be on much stronger ground as an “interpretation.” See id.
[50] Id.
[51] Id. at 171 (emphases added).
[52] Brief of Appellant at 29–30, Shakhnes v. Berlin, 689 F.3d 244 (2d Cir. 2012) (No. 11-2003), 2011 WL 3882067, at *29–30.
[53] Wright v. City of Roanoke Redev. & Hous. Auth., 479 U.S. 418, 420 & n.3 (1987) (citation omitted).
[54] D.D. v. N.Y.C. Bd. of Educ., 465 F.3d 503, 512–13 (2d Cir. 2006).
[55] Id. at 512.
[56] See Harris v. James, 127 F.3d 993, 1018 (11th Cir. 1997) (citing 42 U.S.C. § 1396a(a)(8)).
[57] See id. at 996.
[58] Id. at 1010.
[59] 42 U.S.C. § 1396(a)(3) (2012).
[60] Cf. Hoctor, 82 F.3d at 171.
[61] Cf. D.D., 689 F.3d 252.
[62] 42 C.F.R. 431.244 (f)(1)(ii) (2002).
[63] See supra text accompanying notes 36–39.
[64] 689 F.3d 244, 255.