Case

Starting in the 1970s, lethal injection gained popular support in the United States as an alternate method of capital punishment. Advocates thought that it would be more humane compared to contemporaneous methods, such as the electric chair and gas chamber. In 1977, Oklahoma became the first U.S. state to pass a law authorizing execution via lethal injection; under its procedure, a chemical paralytic agent is combined with an "ultra-short acting barbiturate", neither of which were specifically intended to be the lethal agent.^[1] A day after Oklahoma passed its statute, Texas passed its own version of it, containing no specific protocol for execution.^[1][2] Two of the earliest people to be sentenced under these statutes were Larry Leon Chaney of Jenks, Oklahoma, who was convicted of the 1977 murder of Kendal Ashmore, and Doyle Skillern, who was convincted of the 1974 murder of Patrick Randel.^[3][4] Chaney was the second person in the state to be sentenced to death by lethal injection; his protracted legal battle in state and federal courts was met with little initial luck, including the U.S. Supreme Court thrice declining to review Chaney's case.^[5]

In January 1981,^{[lower-alpha 1]} Chaney and Skillern petitioned the Food and Drug Administration (FDA) to block their states from using lethal injection to execute them, alleging that the process was illegal and "may actually result in agonizingly slow and painful deaths". The NAACP Legal Defense and Educational Fund also signed onto the petition.^[6] Per the petitioners, their states were planning to use drugs for lethal injection that had not been approved by the FDA for that purpose, in violation of two provisions of the Federal Food, Drug, and Cosmetic Act's (FDCA). First, they contended, their states had violated the "new drug" provision by introducing a new drug into interstate commerce without FDA approval – the drugs were FDA-approved, but the petitioners claimed that it was a "new drug" because it had not been approved as "safe and effective" for the purpose of lethal injection. Second, they said that their states' use of approved drugs for unapproved purposes violated the "misbranding" provisions of the act.^[8] They requested that the FDA declare that the drugs were not "safe and effective" for lethal injection, seize prison stockpiles, recommend the prosecution of those involved in the sale of drugs for lethal injection, and investigate the administration of the drugs by amateurs.^[6]

That July, the FDA declined the petition in a letter to the inmates' lawyer on jurisdictional grounds, reasoning that states have the prerogative to use FDA-approved drugs for any purpose they see fit. Arthur H. Hayes Jr., the head of the FDA, wrote that he felt the agency "must defer to a state's choice" in the matter, and wrote that even if the agency did have jurisdiction, it would still decline the petition under what it asserted was its "inherent discretion".^[9][10] By this time, the number of petitioners had increased to eight. Five of them were from Oklahoma, including Chaney, Alton C. Franks, Carl Morgan, Charles William Davis, and Robyn Leroy Parks; three were from Texas, including Skillern, Jerry Joe Bird, and Henry Martinez Porter. The inmates appealed the FDA's refusal to the federal court system.^[9][11]

Administrative Procedure Act

Under section 706(2)(A) of the Administrative Procedure Act of 1946 (APA), courts are empowered to strike down actions of federal agencies that are "arbitrary, capricious, [or] an abuse of discretion".^[12] However, the act also provides two exemptions: section 701(a)(1) allows an action to be precluded from judicial review by statute, and section 701(a)(2) allows the same where it is within the agency's discretion by its enabling statute. These two sections of the APA created a conflict: 706 implied that federal agency actions should generally be reviewable. 701(a)(1) allowed some actions to be expressly exempted by statute. But, regarding 701(a)(2), the scope of "agency discretion" was unclear: if it needs to be expressly authorized, it would be redundant with 701(a)(1); if it extends to anywhere it is not expressly limited, it would move almost all agency action outside judicial review, rendering 706 near toothless.^[13]

Prior to Chaney, the Supreme Court had attempted on multiple occasions to resolve the discrepancy in different circumstances. In Abbott Laboratories v. Gardner (1967), the court interpreted the APA to provide for a "basic presumption of judicial review" that would need "clear and convincing evidence" of legislative intent for an exemption.^[15] The court first directly ruled on section 701(a)(2) in Citizens to Preserve Overton Park v. Volpe (1971), ruling that it only included cases where "statutes are drawn in such broad terms that in a given case there is no law to apply" to the case.^[16] This standard provoked criticism from the legal community, who commented that an agency action need not be limited by "law to apply" to violate the arbitrariness standard of section 706, despite its attempt to construe 701(a)(2) narrowly.^[17] Some lower courts followed this standard to its letter, but others skirted or flouted its terms. Courts in the District of Columbia Circuit in particular argued that the "law to apply" standard should be interpreted as depending on a pragmatic or "functional analysis" of whether a given case was judicially reviewable. When the pragmatic concerns outweighed arguments in favor of review, they would rule that there was "no law to apply". Some commentators said that the D.C. Circuit was completely disregarding the Supreme Court's decision under the guise of following it.^[18][19]

The Supreme Court previously ruled on the reviewability of agency inaction in Dunlop v. Bachowski (1975), affirming the Third Circuit Court of Appeals' decision that the Secretary of Labor's decision not to bring a lawsuit when petitioned was reviewable by the courts. The court reached this decision on the grounds that it was the secretary's "heavy burden" to prove that Congress had intended to make their decision under the relevant statute unreviewable, and came to the conclusion that Congress had no such intentions. It also dismissed by footnote the secretary's argument that their decision whether or not to enforce was protected by prosecutorial discretion.^[20] Justice William Rehnquist dissented, arguing that the secretary's action was an exercise of agency discretion under section 701(a)(2).^[21]

Lower courts

The inmates appealed the FDA's refusal to the United States District Court for the District of Columbia in Chaney v. Schweiker.^{[lower-alpha 2]} On August 30, 1982, the court issued summary judgment against them, holding that the nonenforcement actions are "essentially unreviewable by the courts".^[22][8] The D.C. Circuit Court of Appeals, however, reversed the district court. Judge J. Skelly Wright, writing the majority opinion, invoked the broad assumption of reviewability set out in Gardner and Dunlop, and narrowly construed the "no law to apply" test set out in Overton Park to primarily rely on pragmatic considerations such as "whether judicial supervision is necessary to safeguard plaintiffs' interests, whether judicial review will unnecessarily impede the agency in effectively carrying out its congressionally assigned role, and whether the issues are appropriate for judicial review."^[24] Wright also cited an FDA policy statement in which the agency committed itself to acting against unapproved uses of approved drugs as additional "law to apply".^[25] Having asserted that the agency's nonenforcement decision was reviewable, Wright ruled the agency's decision to be "arbitrary and capricious".^[26]

Judge Antonin Scalia dissented, arguing that the majority cited Dunlop, Overton Park, and Gardner erroneously and that they did not apply to the case at hand. While he agreed that there is a strong presumption of reviewability towards agency action in general, he argued that agency enforcement decisions carried a strong presumption of nonreviewability, and that the three precedential cases were not intended to address section 701(a)(2) and this case.^[27] He also noted that the FDA policy statement was attached to a proposed rule that was never adopted.^[28] In a subsequent opinion, Scalia also criticized the court's regular reliance on pragmatic considerations.^[29]

Supreme Court

The Supreme Court reversed the D.C. Circuit's decision, ruling unanimously against Chaney; Justice Thurgood Marshall concurred in judgement only, and did not join the majority opinion.^[30][31] The majority, written by Justice Rehnquist, held that agency nonenforcement decisions are presumptively unreviewable by the courts under 701(a)(2). The court's decision repudiated the appeals court's use of pragmatic considerations to grant review – it upheld Overton Park's "no law to apply" test, reaffirming the doctrine that statutory limits on discretion were necessary to allow review. In declaring a presumption of unreviewability, however, the court primarily relied on pragmatic factors to render its decision, arguing the following:^[32]

1. Agency inaction lends itself to a vagueness that makes it difficult to draw up standards of review for a given inaction case.^[33]
2. Agency inaction, as opposed to agency action, does not inherently violate anyone's rights directly, and therefore does not trigger immediate cause for review.^[34]
3. An agency nonenforcement decision involves an expert weighing of various factors, which should generally be deferred to by the courts.^[35]
4. An agency nonenforcement decision shares some similarities with a prosecutor's decision of whether or not to bring charges, which is covered by the doctrine of prosecutorial discretion.^[36]

The court rejected the petitioners' argument that the FDA acted outside of its statutory discretion; applying Overton Park's "law to apply" test, the court found that the FDCA contained no such constraints because the misbranding and new drug provisions did not limit agency discretion. The court dismissed the FDA policy statement on the same grounds Scalia did: the proposed rule that statement was attached to was never adopted by the agency, and the court additionally found the language to be ambiguous on the matter. The court also refused to interpret a clause of the FDCA exempting the Secretary from exercising authority for minor violations as implying a requirement for action against major violations, arguing that it could only apply where the agency had already established that a major violation took place, which does not happen if the agency declines to investigate.^[25][37]

The court distinguished Dunlop as precedent, holding that in that case, there was reasonable "law to apply" in that the relevant act could have required the secretary to take the requested action under certain circumstances.^[25][37] Speaking to the presumption of reviewability affirmed in Dunlop that was relied on by the D.C. Circuit Court of Appeals, Rehnquist remarked that "our textual references to the 'strong presumption' of reviewability in ... [Bachowski] were addressed only to the § (a)(1) exception; we were content to rely on the [Third Circuit] Court of Appeals' opinion to hold that the § (a)(2) exception did not apply."^[38] The court emphasized that the presumption of unreviewability is not absolute, and can be overcome if the petitioner can find "law to apply" in the enabling statue that would require action be taken. It also left open the possibility that there could be judicial review under several other circumstances, including where an agency refuses to make a rule, resolves to fully ignore its statutory obligations, or makes a nonenforcement decision that is based solely on jurisdictional grounds or violates a plaintiff's constitutional rights.^[39]