Agencies regulate through a combination of the legislative, executive, and judicial functions by issuing rules with the force of law, policing those rules, and adjudicating their enforcement. In 2021, for example, the Biden administration issued 3,257 regulations with the force and effect of law, whereas Congress passed 81 laws during that time. The last available year for comprehensive data about administrative adjudications is 2013, when the five busiest agencies convened 1,351,342 executive branch tribunals; that same year, there were 57,777 total cases (civil and criminal) filed in the U.S. district and appellate courts.

Of course, the administrative state’s concentration of legislative, executive, and judicial power operates in considerable tension with our constitutional structure, which was designed to diffuse government authority to better protect liberty. As James Madison warned in Federalist no. 47, the “the accumulation of all powers, legislative, executive, and judiciary, in the same hands … may justly be pronounced the very definition of tyranny.”

Under our constitutional design, legislating is supposed to be hard work. Congress’s bicameral structure ensures that a bill must sustain a majority in the both the House and the Senate before becoming law. After that, the Constitution further requires the president’s approval before a law can take effect. It takes significant time and resources before these three institutions—the House, the Senate, and the president—come to agreement on any given policy. The Founders made lawmaking difficult because they were animated by an awareness of the threat posed to liberty by government. In Federalist no. 62, James Madison warned that an “excess of lawmaking” is a “disease” to which “our governments are most liable.”

By contrast, it’s much simpler for the president to achieve a regulation that is the functional equivalent of a law passed by Congress. All he needs to do is pick up a “pen and phone” to initiate the executive branch’s regulatory power. Yet the comparative ease of regulation incubates the “disease” of “excess” lawmaking—again, the Biden administration in 2021 issued 3,257 lawlike regulations, whereas Congress passed only 81 laws. And this tally of Biden-era rules does not include thousands of “subregulatory” documents, like guidance memos and policy statements, which are supposedly nonbinding but which nonetheless must be followed to avoid regulatory prosecution.

Scholars have coined the phrase “presidential administration” to describe how modern domestic policymaking is driven by the White House through the administrative state. In addition to facilitating overbearing government, presidential administration engenders unprecedented instability in federal policymaking. Every time the presidency changes hands from one party to another, the lawmaking machinery of the administrative state pivots 180 degrees, in alignment with the values of the incumbent president. As a result, thousands of rules affecting almost every aspect of American life bounce back and forth between partisan extremes every four to eight years, and the swings are becoming greater as presidents push the envelope of their authority to make administrative policy.

To reassert popular accountability to administrative lawmaking, Congress should pass the Regulations from the Executive in Need of Scrutiny (REINS) Act, which would require lawmakers to vote “major rules” into law before they take effect. Under its provisions, agency rules that meet the criteria are automatically introduced into each house and fast-tracked toward an up-or-down vote within 70 days. The practical effect would be that regulatory agencies could no longer promulgate major rules without congressional ratification. In its present form, the REINS Act applies to rules that cost more than $100 million (“major rules”); in 2021, there were 387 such rules (out of 3,257). If lawmakers are concerned about unduly adding to their workloads, one possible solution is to increase the threshold that triggers the REINS Act. Even subjecting the 50 most consequential rules to congressional scrutiny would go a long way toward reining in the administrative state.

Another way for Congress to reclaim control over domestic policymaking is to put a time limit on delegations of regulatory authority to administrative agencies. Under so-called sunset provisions, regulatory programs expire after a given period—typically 5 to 10 years—unless Congress revisits the program, assesses its effectiveness, and reauthorizes the delegation for another duration of time. At present, Congress employs sunset provisions sparingly, and almost never for discretionary regulatory programs in environmental or labor policy. Sunset mechanisms are used with much greater frequency at the state level, where they play an important role in ensuring popular supervision of regulatory policy.

Finally, Congress must modernize the Administrative Procedure Act (APA). Known as the “constitution of the administrative state,” the APA is 76 years old and no longer meets the needs of the moment. When the APA was passed in 1946, agencies rarely issued legislative rules; instead, agencies created rules through case-by-case adjudication, akin to how the common law works. As a result, the APA gives scant attention to rulemakings, which are now the primary means by which agencies regulate. The absence of meaningful procedural safeguards has abetted the rise (and rise) of the administrative state. The bipartisan Regulatory Accountability Act is a promising vehicle for bringing the APA into the 21st century. In addition to codifying procedures for retrospective review of outdated rules, the act would establish a common-sense sliding scale of procedural requirements, depending on a rule’s cost. As rules become more expensive, the act would require increasingly formalized cost–benefit analysis and greater opportunity for public input.

Besides promulgating lawlike regulations, agencies also enforce these rules in prosecutions before tribunals located within the same agency that brought the enforcement action. This combination of prosecutorial and adjudicative authority coexists uneasily with our constitutional structure. As Madison observed (quoting Montesquieu), “Were it joined to the executive power, THE JUDGE might behave with all the violence of AN OPPRESSOR.”

In practice, the agencies’ home-field advantage is sometimes conspicuous. For example, the Securities and Exchange Commission (SEC) acts both as the prosecutor and the judge when the agency pursues financial penalties through enforcement of its regulations for publicly traded companies and investment activities. According to an analysis conducted by the Wall Street Journal, the SEC had a 90 percent win rate in contested cases it brought before its administrative law judges from 2010 through 2015, while it prevailed in only 69 percent of federal court trials over the same period. During this period, regulated parties filed an official complaint regarding an alleged lack of impartiality by one SEC in-house judge, whose record ruling in favor of the agency had been 51–0.

Another example is the Federal Trade Commission’s (FTC’s) enforcement of its antitrust rules. According to former FTC commissioner Joshua Wright, “Whatever the congressionally intended promise of expert agency administrative adjudication [is] in theory, in practice, the application has been problematic and raises significant concerns that the deck is stacked against firms and in the agency’s favor.” Indeed, the FTC has not lost a single case on its home court for a quarter century. As the Ninth Circuit quipped, “Even the 1972 Miami Dolphins would envy that type of record.” In a scholarly paper, ex-commissioner Wright and Angela Diveley collected data showing that appeals courts reverse FTC decisions at four times the rate of federal district court judges in antitrust cases.

To be sure, many adjudicative regimes in the executive branch do not raise these sorts of problems. For example, no one complains about a pro-agency bias in administrative adjudications that delineate federal rights under disability programs, such as Social Security or veterans benefits, because these proceedings are nonadversarial, meaning that there is no “prosecution” and that the judge operates under a presumption in favor of the beneficiary. In fact, these two relatively innocuous regimes account for almost 75 percent of agency adjudications. The rest involve adversarial proceedings, where the government prosecutes alleged violations and seeks sanctions. It is this latter class of cases, involving regulatory enforcement, that incur the threat of a tilted playing field.

Congress should take certain of these adversarial regimes out of the executive branch altogether. One example is the enforcement of “market manipulation” rules for the securities, commodities, and energy markets by the SEC, Commodity Futures Trading Commission, and Federal Energy Regulatory Commission, respectively. In practice, prosecution for “market manipulation” centers on alleged dishonesty by the regulated party, which is functionally no different than a common-law fraud claim. And the government typically seeks exorbitant civil penalties, including fines totaling scores of millions of dollars and permanent trading bans. In this context—that is, government prosecutions of multimillion-dollar fraud claims—controversies should be heard before Article III courts, not agency tribunals. In our constitutional system, fraud on this scale is for juries to decide. For these market manipulation cases, there is too much at stake to allow agencies to play prosecutor and judge. Congress should move this and any similar adjudicative regime out of the executive branch and into the judicial branch, where regulated parties enjoy impartial justice, as guaranteed by the Constitution. The Consumer Financial Protection Bureau, too, routinely seeks huge civil penalties in its in-house courts for fraudlike regulatory violations, including “lying” and “deceptive practices.” These cases belong before the courts of law.

For the remaining adversarial adjudications, Congress should consider structural changes to ensure a level playing field. Specifically, lawmakers should make greater use of the “separate function” model of agency adjudication. Under this framework, judging responsibilities are vested in principal officers other than the ones who perform the prosecutorial function. For example, under the Occupational Safety and Health Act, Congress delegated prosecuting duties to the Labor Department and judging responsibilities to the Occupational Safety and Health Review Commission. The federal regulatory regime for mine safety is similarly divided between the Labor Department and the Federal Mine Safety and Health Review Commission. The judging function can be located within the agency or in a separate commission, if whoever makes the final decision is different from whoever makes the final decisions on prosecutions. Congress should consider switching to this “separate function” wherever government prosecutes adversarial proceedings.

Adler, Jonathan H., and Christopher J. Walker. “Delegation and Time.” Iowa Law Review 105, no. 5 (2020): 1931–93.

Crews, Clyde Wayne. Ten Thousand Commandments 2021. Washington: Competitive Enterprise Institute, June 30, 2021.

Nielson, Aaron L. “Three Wrong Turns in Agency Adjudication.” George Mason Law Review 28, no. 2 (2021): 657–81.

Walker, Christopher J. “Modernizing the Administrative Procedure Act.” Administrative Law Review 69, no. 3 (2019): 629–70.

———. “A Reform Agenda for Administrative Adjudication.” Regulation 44, no. 1 (2021): 30–35.

Yeatman, William. “Bombshell Admission Undermines Integrity of SEC’s In‐​House Courts.” Cato at Liberty (blog), June 20, 2022.

———. “A Fuller Picture of the Trump Administration’s Regulatory Agenda.” Regulatory Review, May 25, 2020.