Section 706(2) of the APA provides, in deceptively simple language, that a “reviewing court shall . . . hold unlawful and set aside agency action, findings, and conclusions . . . not in accordance with law.” In the ongoing debate over the scope of relief under the APA, much depends on the “semantic content of the phrase ‘set aside.’” In a recent working paper focused on the state-law influence on federal administrative law remedies, I offer new evidence on the original meaning of this vague language by demonstrating that the APA’s “set aside” language originates in state law. This historical perspective supports the traditional understanding that “set aside” contemplates universal vacatur.

The story of the term “set aside” in federal statutes begins with the Hepburn Act in 1906. Responding to growing political frustration with the judicial declawing of the Interstate Commerce Commission, the Hepburn Act conferred broad ratemaking power on the ICC—abrogating the Supreme Court’s 1897 decision that the ICC lacked the authority to set maximum rates. Tucked away in the Act’s venue provision, Congress implicitly recognized that courts had the power to “enjoin, set aside, annul, or suspend” any ICC order.

This language proved remarkably durable. Once “set aside” entered the legislative vocabulary, it spread quickly: first to the Urgent Deficiencies Act in 1913, from there to other federal statutes, and ultimately into § 706 of the APA.

But the federal genealogy of “set aside” tells only part of the story. By the time “set aside” was being introduced into federal law in 1906, state legislation had already been using the term for decades to describe judicial review of administrative action.

State railroad legislation offers a particularly instructive illustration. When it came to railroad regulation, states were trailblazers—establishing railroad commissions long before the ICC entered the scene. Many states were also years ahead of federal law in granting their railroad commissions expansive powers to make orders and set rates. With broad administrative power came the need for judicial oversight: Texas provided that rates set by its railroad commissioners would be “conclusive until set aside by direct action”; Kansas empowered its courts to “set aside, vacate or annul” any “regulations or orders” issued by the Kansas board of railroad commissioners; Wisconsin authorized its courts to “vacate and set aside” orders of the Wisconsin commission; Ohio copied the Wisconsin law for its own commission. Other examples abound, in a wide variety of administrative contexts.

A review of these pre-Hepburn Act state statutes reveals that “set aside” was rarely used in isolation. The term was used alongside, and—as marginal notes and headings confirm—interchangeably with, terms like “vacate” and “annul.” In this way, state statutes providing for judicial review of administrative action closely paralleled state statutes describing appellate review. In Wisconsin, for example, judges “vacate[d] and set aside . . . judgment[s]” handed down by lower courts and “vacate[d] and set aside . . . order[s]” handed down by the railroad commission. (As Emily Bremer has suggested on this blog, the appellate model of judicial review—where review of administrative action parallels review of lower court judgments—offers a useful perspective on the “set aside” power.)

Congress was well aware of this state practice. As Congress drafted the Hepburn Act, it turned for inspiration to the state statutes discussed above. The specific “set aside” provisions of both the Kansas and the Wisconsin state railroad statutes were quoted in full as models for what the federal law could look like. The “set aside” provision of the Ohio state railroad statute was floated—verbatim—as a template for the federal law. Dozens of representatives in the House and Senate referenced state laws for inspiration, guidance, and support in crafting the Hepburn Act. As one Senator explained, it was “instructive” to observe how “States have dealt with the question of court review, as applied to the acts of their own State railroad commissions exercising similar powers.” In short: state codes offer the best explanation for the origin of federal law’s “set aside” language.

These state-law origins carry several implications for the ongoing debate on the scope of relief under APA § 706. Most importantly, the earlier use of “set aside” in state statutes strengthens the case that the APA’s “set aside” provision contemplates universal vacatur. When state legislatures were using “set aside” to describe judicial review, they had the appellate court-lower court relationship in mind; it was this conception of “set aside” that Congress lifted into federal law, and which ultimately made its way into the APA. In other words, the best lens through which to understand “set aside” is in terms analogous to the judicial treatment of a lower tribunal’s judgment: the target action is “entirely destroyed” and “deprived of . . . conclusive effect” for all—not just the parties.

State practice also demonstrates that we should resist attaching too much weight to the fact that the APA uses the term “set aside,” rather than, say, “vacate.” APA § 706 only instructs courts to “set aside” the agency action; it does not instruct courts to “vacate” the action. But in early state legislation, “set aside” was often used interchangeably with other terms—like “vacate”—that sounded in universal relief. The state-law origins of the “set aside” terminology therefore undercut one textualist argument against the current understanding of “set aside” in APA § 706: that “the term ‘set aside’ never meant vacate.” The state statutes make clear that “set aside” very much did mean “vacate.” Just as courts “fold[] together vacatur, reversal, and set-asides” when describing appellate review of lower court decisions, so too did state statutes describing judicial review of administrative action. State practice teaches us that we should be skeptical of any interpretation that attaches significance to the fact that it was “set aside,” rather than “vacate,” that ultimately found its way into the APA.

The state-law origins of the “set aside” terminology are just one example of state influence on federal administrative law remedies. My paper argues that state courts should also be credited with a more transformative revolution in administrative law: the development of the appellate model of judicial review.

The common law writ of certiorari acted as the primary instrument of judicial review in state law in the eighteenth and nineteenth centuries. But the writ came with some English-law baggage: it could only be used to review “jurisdictional errors” and “errors on the face of the record.” Over the course of the nineteenth century, state courts gradually liberated certiorari from this myopic focus on “jurisdictional” error and began using the writ to review all questions of law. Through this selective departure from English precedent, state courts gradually fashioned certiorari to approximate appellate review: confined to the record, de novo on questions of law, and deferential on questions of fact.

By documenting the emergence of this earlier, certiorari-based appellate review model, the paper offers an answer to an enduring mystery at the heart of federal administrative law: where did the Supreme Court get the appellate review model from? The paper argues that, when the Court adopted the appellate review model in a series of ICC ratemaking cases in the early twentieth century, it was lifting the model from state certiorari practice. In so doing, the Court certiorari-ized the federal injunction. There are many implications for our understanding of modern judicial review, but the paper focuses on one particularly salient issue: the availability of universal relief. The debate on universal relief has thus far focused almost exclusively on federal equity practice. But, as the paper shows, the modern federal injunction has much more in common with state certiorari practice than with nineteenth-century federal equity practice. Looking to state certiorari practice—where courts routinely “set aside,” “annulled,” “vacated,” and “suspended” administrative action universally—offers new historical support for universal relief. As one court explained in a challenge to a tax assessment, “[i]t cannot be denied” that courts have the power, on a writ of certiorari, to “set aside and entirely annul the whole of the assessment” as to all those affected—not “only as to those parties who . . . br[ought] it before the court.”

The nineteenth and twentieth centuries witnessed significant cross-pollination between state and federal administrative law. There is much we can still learn from state practice.

Fred Halbhuber is a recent graduate of Yale Law School.