Timothy Sandefur is the Vice President for Litigation, and Jon Riches is the Director of National Litigation and General Counsel at the Goldwater Institute.
Among the greatest threats to the liberty that our constitutional system was designed to protect is the growth of the administrative state. That’s the name given to the bureaucratic entities that, within a generation, have grown so large and powerful that they now threaten not only the future of individual freedom, but even our democratic system of government. And this is true at both the state and federal levels.
These agencies, staffed by officials who are purposely insulated from control by voters, write rules that have the effect of law. They also prosecute alleged violations of those rules—typically in their own “courts”—and issue decisions that interpret those rules. These decisions are often rubber-stamped by a judicial system that is supposed to ensure that the Constitution and the actual laws passed by elected legislators are faithfully followed. In other words, agencies combine legislative, executive, and a judicial authority in ways that violate the separation-of-powers principle long recognized as crucial to the security of individual rights.
Meanwhile, lawmakers routinely surrender their authority to these undemocratic agencies, allowing them in practice to govern wide swaths of our lives—everything from how much fish may be harvested from the sea, to the angle at which chairs in an office setting may lean back, to the thickness of ketchup in packets at fast-food restaurants. Legislatures pass broad, vaguely worded statutes and leave it to bureaucracies to fill in the details. Yet, as the old adage has it, the details are just where the devil is.
Fortunately, there seems to be a growing recognition of what Columbia Law Professor Philip Hamburger has called the “jarring disconnect between what is taught and celebrated in constitutional law and what is accepted in administrative law. One offers a vision of divided power exercised through and under law; the other presents the reality of consolidated power exercised outside and above the law.”1 Fixing this problem presents many challenges and will require a concentrated and persistent approach. But we do have the tools necessary to rein in the administrative state, hold its regulators accountable, and better fulfill the promises of our Constitution.
Or more accurately, constitutions. An often forgotten point is the United States has not just one constitution, but 51—and states are free to provide greater protections for individual liberty in their fundamental law than those the federal Constitution provides. While the U.S. Constitution sets forth the minimum basic rules that states must obey, states are free to add stronger security for freedom when necessary. State legislatures can also pass laws that limit or otherwise shape the powers of their administrative agencies in ways that reinforce constitutional limits while still obtaining the advantages these agencies provide. This paper sets forth some of the more promising proposals for ensuring that the administrative state operates within constitutional boundaries at the state level.
Why Have Administrative Agencies in the First Place?
At the outset, it’s worth remembering why we have administrative agencies at all. In the United States, such agencies have existed since 1824, when the first—the Bureau of Indian Affairs—was created. At that time, such entities were viewed as falling within the executive branch of the government because they were overseen by the president, who bore ultimate responsibility for enforcing the law. No matter how large an agency might be, it was regarded as essentially a team of assistants whose job was to aid the president in enforcing congressionally enacted laws.
Half a century later, things began to change. In 1887, the first economic regulatory agency—the Interstate Commerce Commission—was founded and given the job of setting the prices railroads could charge. The reasons for creating commissions to do jobs like this rather than having Congress do so directly through legislation were the same as those given today for establishing bodies such as the Environmental Protection Agency, the California Air Resources Board, or the Michigan Department of Licensing and Regulatory Affairs: Agencies are expected to be staffed by experts in a particular field, who are purportedly better suited to make rules to govern that field. They are also thought to be better equipped to respond swiftly to public need and act in a more focused way than a legislature can.
During the Progressive Era (1890-1920), the fact that agencies were not subject to control by voters came to be seen as an advantage. Progressive leaders, most notably President Woodrow Wilson, saw regulatory agencies as a way to separate government from politics—to ensure that government was scientifically planned by experts and free from the influence of voters, who were considered too ignorant and capricious to make good decisions.2 During the New Deal, this concept of regulatory agencies grew stronger as President Franklin Roosevelt oversaw an immense expansion of the federal bureaucracy, filled by specialists who would plan the nation’s economy with scientific precision.
Then in 1935, the U.S. Supreme Court decided Humphrey’s Executor v. United States,3 a case that ushered in a new way of viewing regulatory agencies. The case involved Roosevelt’s effort to fire a member of the Federal Trade Commission (FTC). Because the law creating the Commission provided that members could be terminated only for certain limited reasons, Roosevelt’s decision was challenged in court. The justices unanimously held that the President could not fire the commissioner, because the FTC was not an executive-branch entity but “quasi-judicial and quasi-legislative.”4 In the court’s view, the FTC “must be free from executive control” to do its job: “It is charged with the enforcement of no policy except the policy of the law … Its members are called upon to exercise the trained judgment of a body of experts ‘appointed by law and informed by experience.’”5 So although a president could appoint members to the FTC—and although the FTC’s job was to “enforce” the “policy of the law,” which is quintessentially an executive branch responsibility—Congress could still forbid a president from firing commissioners who he believed were not doing their jobs correctly.
This marked the birth of the independent agency—an entity whose oversight was not truly clear, and that although responsible for enforcing the law, was nevertheless immune from control by the one official the Constitution entrusts with enforcing the law: the president. The Humphrey’s Executor decision gave its blessing to an entity that lacks constitutional foundation: an agency made up of experts who exercise power without meaningful democratic oversight and contrary to the separation of powers the Constitution’s authors thought essential.6
The third stage of the growth of administrative agencies came in the 1960s-1970s, when the administrations of Lyndon Johnson and Richard Nixon sought once more to harness the expertise of the “best and the brightest” to scientifically plan solutions to social problems. The number of independent agencies expanded during this period to include everything from the Commodity Futures Trading Commission to the Federal Election Commission.
State agencies grew in a similar pattern, roughly doubling in number since the 1950s.7 It’s impossible to calculate the number of state-based regulatory agencies, in part because such agencies can quickly be created and transformed. Massachusetts law, for example, allows the Commissioner of Administration to create new agencies at will, no legislative approval necessary.8 By contrast, California law requires state agencies to be reviewed periodically for effectiveness and dissolved if no longer needed—a process known as “sunsetting”—except that when agencies are dissolved, their powers are sometimes not actually eliminated but simply transferred to the state’s Department of General Services.9
In many cases, state agencies are, as in Humphrey’s Executor, so insulated against control even by the state’s chief executive, that they largely function as autonomous governments beyond the reach of the voters or their elected representatives. In short, the growth of the administrative state—and the vast expansion of its powers—has been exponential and largely unchecked for the better part of a century at both the federal and state level.
Agencies as Anti-Democratic
Administrative agencies are supposed to harness the power of expert knowledge and to shield decision-making from political interference. Yet that very independence and expertise can prove problematic. Independence means agencies cannot easily be restrained if they act improperly, and their expertise makes these agencies vulnerable to what scholars call “regulatory capture,” which occurs when an agency’s powers are exercised for the benefit of powerful regulated entities instead of for the general public.10
This typically results from the fact that a regulatory board charged with, for example, regulating the prices charged by railroads or the licensing of barbers will have an enormous effect on how railroads or barbers operate, which leads railroads and barbers to focus attention on influencing how these agencies work. Members of the profession get to know the regulators, testify before their hearings, and perhaps even win appointments to serve on the regulatory board themselves. Members of the general public, by contrast, spend little time or money seeking to influence the regulatory agency, because they have other daily concerns.
The result is a classic problem of concentrated benefits and diffuse costs. In the end, agencies become so heavily influenced by the businesses they’re supposed to regulate that they come to serve those businesses’ private interests—or the interest of that particular industry—even at the expense of the public good. For example, the U.S. Supreme Court in 2015 observed that the North Carolina Board of Dental Examiners, which was largely staffed by practicing dentists, was forbidding non-dentists from offering cosmetic teeth-whitening services, not to protect public safety, but to serve the financial interests of dentists who did not want competition from others using this safe, nonmedical procedure.11
In addition, as President Ronald Reagan said, “The first rule of bureaucracy is, protect the bureaucracy.”12 Agencies have extremely strong incentives to defend and expand their authority, a fact which doesn’t necessarily impugn the motives of bureaucratic staff. On the contrary, it is the job of conscientious officials to exercise their regulatory authority if they genuinely believe it will serve the public interest. That inherently translates, however, into agencies expanding their powers to the fullest extent that they believe appropriate. This results in pressure to regulate more than necessary, and to expand the agency’s own powers rather than taking steps to improve outcomes for the public (such as increasing competition and innovation).
Not only are agencies inherently undemocratic, they’re often designed to prevent legislatures, presidents, or governors from imposing significant changes. For example, the California Coastal Commission’s makeup is carefully staggered in a way that prevents a governor from replacing or appointing a majority of its members. The state’s Supreme Court has said that the Commission exercises “a variety of governmental functions, some generally characterized as ‘executive,’ some ‘quasi-legislative,’ and some ‘quasi-judicial,’”13 and it governs land use decisions over vast portions of the state. Voters, of course, have no direct control over the Commission, and the complex way in which its members are selected ensures it is largely immune from the control of their elected representatives, too. What’s more, the Commission’s powers are mostly exercised, not by its members, but by an executive director chosen by the Commission—meaning that governors, legislators, and judges have little power to rein in the Commission when it overreaches. And that means Californians themselves have no effective control over one of the most powerful government agencies in the nation.
Even in less extreme cases, agencies are staffed by career civil servants who often serve far longer than do elected legislators or governors. Some of these bureaucrats therefore come to view elected representatives as temporary nuisances who interfere with their operations, rather than as their legal superiors. Even where agencies are not “independent”—that is, even where elected officials exercise genuine oversight—such agencies often act undemocratically and without meaningful public control. Some are just too large to be effectively overseen by elected officials. They produce countless regulations, opinions, and rulings every year—far too many to be reviewed or even understood by a president or governor. Between 1950 and 2018, for example, the Code of Federal Regulations grew from about 15,000 pages to nearly 200,000. George Washington University’s Regulatory Studies Center has found that federal regulatory agencies alone have adopted between 20 and 40 “economically significant” new rules every year since 1985, each of which includes intricate legal terminology or scientific data—and it defines “economically significant” as rules having an effect on the national economy exceeding $100 million.14 This is just at the federal level.
Circumstances are much the same in the states. Although the total number of state regulations is impossible to calculate, the Mercatus Center estimates that each state has between 65,000 and 300,000 regulations that either forbid or require certain actions. That, of course, doesn’t count regulations that define important legal terms or control how agencies themselves operate—all of which can have important legal consequences for citizens.15
These regulations are produced by a process that stands outside our democratic system—and they pose a major threat to individual liberty. Regulatory agencies are charged with authority to determine how businesses operate, how people may use their property, how and when they may participate in the political process, what they may do for a living, where they may travel, how their children are educated, etc. Indeed, most of the rules governing how people live their lives in the United States are not statutes enacted by elected officials, but regulations created by agencies that operate without meaningful democratic oversight. And these entities have the power to inflict devastating financial and other penalties on individuals who violate their rules. When an agency accuses someone of violating a rule, that person may usually request a hearing—but most such hearings are overseen not by independent judges, but by the agency itself, in hearings where the normal rules of due process and evidence usually do not apply. In some cases, the agency doesn’t give the person a hearing at all.16
What’s worse, legal courts have adopted a legal theory called “deference” (including several different kinds of deference) that essentially allows agencies to determine their own authority. Deference minimizes the amount of independent oversight that agencies receive by requiring judges to defer to an agency’s factual determinations and legal conclusions, even though it is the job of judges, not agencies, to make those decisions. The several kinds of deference have recently come under criticism by judges and legal scholars, and limiting them is essential to repairing the constitutional damage administrative agencies have inflicted.
Ending Deference to the Administrative State in the States
Administrative agencies make regulations (as well as things such as “guidance letters” that are technically not regulations but still have legal force). They also investigate alleged violations of those regulations and adjudicate those alleged violations in hearings presided over by administrative law judges hired and employed by the agency itself. If the case then finds its way to an actual courtroom, the judge there must defer to the agency’s interpretation of disputed questions of law; in other words, courts are obligated to put a thumb on the scale for the agency.
There are several kinds of deference doctrines. One, called Chevron deference, after the Supreme Court decision that created it,17 requires courts to accept an agency’s interpretations of arguably ambiguous statutes. Another, called Auer deference,18 requires courts to defer to an agency’s interpretation of its own regulations.
These doctrines raise core concerns about separation of powers because they allow executive agencies—or “independent” agencies—to exercise legislative and judicial powers. Courts are supposed to exercise their own judgment when interpreting the laws created by legislatures, but deference short-circuits that process and bars courts from questioning the executive branch’s self-created rules. As Justice Neil Gorsuch observed while serving on the Tenth Circuit Court of Appeals, this kind of deference to agencies “permit[s] executive bureaucracies to swallow huge amounts of core judicial and legislative power and concentrate federal power in a way that seems more than a little difficult to square with the Constitution of the framers’ design.”19
In other words, deference doctrines create structural problems for the Constitution’s careful balancing of powers among the different branches. By immunizing the decisions of regulators from meaningful review by neutral judges, they also create practical accountability problems because regulators know that if their decisions are challenged in court, the judge must find a way to rule for the government in all but the most extreme cases.
While much commentary has focused on the problems judicial deference has created at the federal level, deference to administrative power is not a uniquely federal problem. Indeed, because many states have modeled their own administrative codes on the federal Administrative Procedure Act (APA),20 and because many state courts have copied the federal courts’ deference theories, these problems are just as pervasive at the state level.
This, however, provides an opportunity for states to lead the way. The Goldwater Institute has developed model legislation that amends state law to eliminate deference to agency decisions and restore the proper constitutional balance in cases where an agency is interpreting a constitutional provision, a state statute, or an agency rule. In 2018, Arizona became the first state in the country to adopt this law.21 Other states should follow Arizona’s lead.
What Is Administrative Deference and Where Does It Exist?
Deference to administrative agencies traces its roots to early interpretations of the federal APA, which was originally passed in 1946.22 That law has been amended several times since, including in 1966, to add a “scope of review” section that courts have interpreted as requiring deference. But the actual language of that statute does not even suggest, let alone require, deference to agency actions. Instead, it directs courts, when reviewing agency actions, to “decide all relevant questions of law, interpret constitutional and statutory provisions, and determine the meaning or applicability of the terms of an agency action.” It also requires courts to “hold unlawful and set aside agency action, findings, and conclusions found to be … not in accordance with law.”23
In other words, the statute indicates no congressional intent either to delegate lawmaking power to executive agencies, or to curtail the judiciary’s responsibility “to say what the law is.”24 On the contrary, it directs courts to do their job of interpreting the law, ambiguous or otherwise. Nevertheless, the U.S. Supreme Court and lower courts have gradually grafted onto the APA the idea that instead of impartially interpreting the law, courts must accede to agency actions in most circumstances.
The first and most notorious of the deference doctrines is Chevron deference, which requires courts to give an agency’s interpretations of an ambiguous statute “controlling weight,” even if the court would have given a different interpretation if it had been asked first.25 Judges apply Chevron through a two-step process. First, they determine whether the statute in question is “ambiguous,” and if so, they then determine whether the agency’s interpretation is “reasonable.”26 If, for example, Congress adopts a law that forbids the release of pollutants into a “water of the United States” but fails to define that term, the Chevron doctrine lets the agency define it in any way that is “reasonable”—which effectively enables the agency to decide the scope of its own authority. Of course, as then-Judge Gorsuch observed, nowhere in “all this does a court interpret the law.”27 That core judicial function is given instead to unelected bureaucrats.
The second deference doctrine, called Auer, extended Chevron’s faulty rationale to apply to an agency’s interpretations of its own rules. Under Auer, when an agency interprets its own regulation, courts must give “controlling weight” to that interpretation, unless it is plainly erroneous. This means that agencies not only get to write rules that have the force of law, but if there’s a question as to what those rules mean or to whom or what they apply, courts will accept the agency’s argument.29
This is what occurred in a case involving Flytenow, a small technology firm that created a digital platform to allow pilots of small aircraft to communicate their flight plans to potential passengers who might wish to ride along. Like carpooling, these flights are not done for profit, but instead involve splitting the expenses—mostly fuel and airport charges—in exchange for a ride. Pilots and passengers have shared flight costs since the earliest days of aviation. But in the past, they connected via telephone, word of mouth, or by posting flyers on airport bulletin boards. Flytenow just made the communication easier by allowing pilots to post the same information on the internet.
According to the Federal Aviation Administration (FAA), however, using the internet to communicate flight plans transformed private pilots of four- to six-seat aircraft into “common carriers,” making them subject to the same expensive and time-consuming FAA regulations that apply to large commercial air carriers like Delta and American Airlines.
Absurd as this might seem, Congress had not defined the term “common carrier” in its statutes governing general aviation. So the FAA defined it by adopting a regulation in the 1980s, and that definition has not been updated since. Then in the 2010s, the FAA interpreted that regulation in a new, broader fashion, meaning that Flytenow’s operations fell under it. The agency then issued an order effectively shutting down Flytenow’s website.
Flytenow sued, arguing that they were obviously not the same thing as a private commercial airline. That contention, however, required challenging the FAA’s interpretation of its regulation that interpreted Congress’s laws, and the court followed the Auer deference doctrine, deferring to the agency’s interpretation of its own regulation interpreting the statute.30 As a result, flight-sharing in the United States has been grounded.
It is an unfortunate reality that federal courts have not been alone in abandoning the judiciary’s duty to independently and impartially interpret the law and adjudicate disputes. Many state courts, too, have either expressly adopted Chevron or other deference doctrines or have fashioned their own similar versions.31 This has turned judicial deference into a nationwide foundation for a large and powerful administrative state at both the federal and state level. But nothing about this is permanent—and states have plentiful authority to address this problem at the state level, which might also lead to national changes.
Deference Doctrines and Agency Accountability
Deference requires courts to abandon the neutrality that is central to their role as impartial interpreters of the law and instead forces them to apply bias in favor of the executive branch. As Professor Hamburger observed, “When the government is a party to a case, the doctrines that require judicial deference to agency interpretations are precommitments in favor of the government’s legal position, and the effect is systematic judicial bias.”32 This structural accountability problem is accompanied by a practical one. Namely, regulators in the executive branch know they can create expansive rules, investigate borderline violations, and decide close cases in favor of themselves because if those rules are challenged, the courts will defer to their judgment.33
This results in more, and more arbitrary, rulemaking. It also results in more investigations and more findings of violations. And ultimately, it results in bigger, more intrusive government. In the absence of judicial deference, agencies and the regulators who staff them would be more constrained in their rulemaking, more measured in their interpretations, and more principled in their enforcement actions. In other words, they would be more careful because they would be more accountable.
A State-Based Solution
A change to deference doctrines does not need to originate in the judiciary.34 In fact, federal courts initially applied deference doctrines by interpreting the federal Administrative Procedure Act(APA).35 State legislatures later adopted their own state-law versions of the APA—modeled on the federal one—to govern their own agencies, and state courts then created their own state-law deference doctrines by interpreting those state administrative procedure acts.36 Since both the federal APA and these state-law copies set out the legal framework for how courts must review agency decisions, the result is that state legislatures are also free to direct courts as to how questions of deference should be addressed, by clarifying their own state administrative procedure acts. In other words, changes to state laws governing state agencies can help eliminate these deference doctrines across the board.
This is precisely what was done in Arizona. In 2019, based on legislation developed by the Goldwater Institute, Arizona became the first state in the country to eliminate the state equivalent of Chevron and Auer deference statutorily. That was accomplished by including a new sentence in the scope of review section of the state’s APA:
In a proceeding brought by or against the regulated party, the court shall decide all questions of law, including the interpretation of a constitutional or statutory provision or a rule adopted by an agency, without deference to any previous determination that may have been made on the question by the agency.37
This change addresses the state-law versions of both Chevron and Auer deference and reinforces the courts’ obligation to exercise their own independent judgment when interpreting the law. Although only a short time has passed since this new law went into effect, Arizona courts appear to be applying it faithfully—by applying non-deferential de novo review to the legal interpretations of administrative agencies.38
Of course, change at the state level will not apply to federal law, but it can serve as a model for federal reform. Congress often adopts changes to federal law that are modeled on innovations adopted by state legislatures. And Congress has already expressed interest in reforming administrative deference. The House of Representatives introduced and passed legislation in 2016 that would address and eliminate Chevron deference.39
Administrative deference prevents meaningful oversight over and accountability for the decisions of executive agencies, and it is offensive to the separation of powers. As a result of Arizona’s reform, Arizonans now have a more equal and fair opportunity to challenge regulatory actions in court. And because the new law makes the limits on agency authority clearer, it will also likely decrease litigation in the long run, as agencies will regulate in a manner less likely to push the envelope or to exploit vague language as a chance to expand their power. The consequence will be better due process in the administrative context and, because the law injects accountability into a system where it is currently absent, better regulatory decision-making.
Rational Basis Scrutiny
Rational basis scrutiny isn’t often considered a form of deference doctrine, but that’s what it is. Invented in 1934,40 rational basis is a legal test courts apply whenever a person contends that his or her constitutional rights have been violated by the government. The test asks whether the government’s actions are “rationally related to a legitimate government interest”—an extremely lenient standard that is only violated if government officials could not possibly have believed that their actions would benefit the public in any way. Judges using this test typically imagine whether in a hypothetical world, the governments’ actions might have been justified, and, if so, to rule in the government’s favor, regardless whether those imaginary facts bear any relationship to what actually happened. Some courts have even said that facts are irrelevant under this test.41 This contrasts sharply with the “strict scrutiny” applied to other kinds of constitutional rights, such as freedom of speech or freedom to travel. Under that test, the government’s actions must be “narrowly tailored to achieve a compelling government interest”—an extremely demanding standard that requires the government’s actions to be precisely designed to accomplish an exceptionally important goal.
There is no basis in either the state or federal constitutions for treating different kinds of rights differently. Indeed, there is no constitutional foundation for the idea of rational basis scrutiny to begin with. Nevertheless, ever since it was created, both state and federal courts have applied rational basis when addressing cases that involve such rights as private property or economic liberty. As a consequence, if government deprives a person of the right to express an opinion, that person can ask a court to intervene and is likely to prevail under “strict scrutiny”—whereas if the government deprives the same individual of the right to run a business, choose working hours, or build a home, he or she is unlikely to win a lawsuit because courts will apply deferential “rational basis scrutiny.” In short, rational basis is a deference doctrine that applies not just to administrative agencies, but to all government officials.42
One significant problem with rational basis it is that it presumes in favor of the government and places the burden of proof on the individual—who must disprove the constitutionality of a challenged law. Because it is impossible to prove a negative, this burden makes it virtually impossible to win such cases against the government, especially where courts consider the actual facts irrelevant. In sum, citizens challenging the constitutionality of restrictions on their economic liberty or private property rights must prove that those restrictions are utterly irrational—and even then, it is unlikely that courts will rule in their favor.
This creates the same accountability problem mentioned above. Regulators know their decisions will probably never be challenged, and if they are, the court will look for some way to rule in their favor. Thus, like other deference doctrines, rational basis encourages more regulations and vaguer, less reasonable regulatory behavior.
Some judges have expressed concern about the excessive deference that rational basis embodies. For example, in 2012, two D.C. Circuit Court judges wrote that the “practical effect of rational basis review … is the absence of any check on the group interests that all too often control the democratic process. It allows the legislature free rein to subjugate the common good and individual liberty to the electoral calculus of politicians, the whim of majorities, or the self-interest of factions.”43 And the Texas Supreme Court refused to use the rational basis approach in a 2015 decision called Patel in which one of the justices called it “a misnomer, wrapped in an anomaly, inside a contradiction … less objective reason than subjective rationalization.”44
Patel is significant because one bizarre fact about rational basis is that it was created by federal courts as a way to apply the federal Constitution, but was later adopted by state courts as a way to apply their state constitutions. This makes little sense. State courts are not bound by federal legal doctrines when they interpret their own state constitutions. In fact, most state constitutions were written long before the invention of rational basis, meaning that the authors and ratifiers of those state constitutions cannot have expected courts to copy-and-paste the federal rational basis standard. In fact, as one Alabama Supreme Court justice observed in a 2007 decision, the Alabama Constitution was written in 1901, more than 30 years before federal courts invented rational basis deference. Its authors and the people who ratified their work expected courts to apply the much more skeptical legal test that courts used in 1901. There’s no justification for state judges to employ the lenient rational basis theory when enforcing the state constitution.45
Nevertheless, most state judges continue to echo federal legal theories such as rational basis deference without considering such questions at all. For example, in a sweeping decision that overruled 51 separate state-level precedents, the Washington Supreme Court declared in 2019 that the state constitution’s “due process of law” clause would from now on be interpreted identically with the federal Constitution’s “due process of law” clause—meaning it would follow federal decisions from the 1930s, even though the Washington Constitution was written in the 1890s.
The Patel case represents a far better approach. There, the Texas Supreme Court refused to adopt the extreme deference of the federal rational basis test, noting that the Texas Constitution’s wording does not mirror the federal Constitution, and there’s no reason for state judges to simply follow federal courts in lockstep. It concluded that Texas courts should use a test more protective of individual rights—and decide for themselves whether restrictions on those rights are constitutional.
The Right to Earn a Living Act
But even where state judges don’t take the lead in resisting the excessive deference of the rational basis test, state legislatures can act. Arizona recently adopted two important pieces of legislation that restrict the power of agencies to impose regulations that burden economic freedom in the context of occupational licensing.
An occupational license is a government permission slip to work in one’s chosen field. Occupational licenses have been required for some professions, such as doctors and lawyers, for decades. Other restrictions—such as licensing laws for florists, interior designers, and landscape architects—are more recent. The historical justification for requiring government preapproval before engaging in an occupation is that regulators can protect the public from harm or fraud by requiring that the practitioner prove that he or she meets certain standards before engaging in dangerous or risky professions. But over the years, such limits have been increasingly imposed on professions that pose no serious risk of harm, or for reasons that bear no connection to public health or safety.
In the 1950s, only 5% of jobs required an occupational license. Today, roughly one in four require government permission.46 While fewer than 30 occupations are licensed in all states (mostly in the health professions), over half of all state-licensed occupations are only licensed in one state—a strong indication that these occupations are not a real danger to public safety. These include graphic designers, audio engineers, braille instructors, and travel agents.47 States require occupational licenses for such innocuous professions as chimney sweepers (in Vermont),48 bed salespeople (in West Virginia),49 and florists (in Louisiana).50 Other examples include locksmiths, alarm installers, and furniture upholsterers.
Of course, when it comes to the right to earn an honest living in these and other occupations, our system should presume in favor of freedom and require regulators to at least provide a good reason when they undermine a person’s freedom to get a job. The Right to Earn a Living Act, developed by the Goldwater Institute and enacted in Arizona in 2017,51 corrects this accountability problem and restores the right to earn a living to its proper status as a protected right.
The Act puts the burden of proof back where it belongs: on the regulators who restrict economic freedom, instead of on job-seekers. Whenever bureaucrats restrict people’s right to use their skills to provide for themselves and their families, the Act requires government to show a genuine public health and safety need for that restriction. If the government cannot prove the regulation is necessary to serve the public, then it should not impose it.
During legislative debate over the Act, opponents argued that it would result in a flood of litigation. Advocates countered that it would actually result in less regulation as well as more sensible regulation, and thereby protect individual rights without an increase in lawsuits. Since this law was enacted, the latter prediction has proven correct.
One illustrative example is the case of Annette Stanley, a behavioral health counselor who received her license in Kansas after completing more than 2,000 hours of supervised work experience. That experience was overseen by a counselor who had been practicing in the field for 20-plus years. After Stanley was licensed, her husband got a job in Arizona, and the family relocated there. But when Stanley asked Arizona regulators to issue her a license, the state’s Board of Behavioral Health said no, citing a rule that required license applicants to demonstrate that they had a certain number of hours of supervised work experience—but did not allow applicants to count any hours that were obtained from a firm in which the applicant had an “ownership interest.” Since Stanley had completed her thousands of hours of required experience in Kansas in an office that she co-owned with her supervisor, none of her supervised experience hours counted.
Stanley challenged the Board’s denial of her application on the grounds that its no-ownership rule violated the requirements of the Right to Earn a Living Act. It could not be plausibly argued that such the rule actually protected public safety. Rather than face litigation—where a court would have applied its own independent analysis instead of rational basis deference—the Board decided to change the rule and issued Stanley a license.52 In other words, because there was now accountability in a system that previously lacked it, the regulators engaged in more sensible rulemaking.
Arizona took another step in licensing reform in 2019 by adopting the Universal Recognition Act—a law that automatically recognizes occupational licenses issued by other states, with certain narrow exceptions. This Act imposes an important limit on occupational licensing by recognizing this simple fact: If someone has been safely and productively practicing an occupation in another state, that person does not suddenly become unsafe or unqualified by crossing state lines.
Under this law, if an individual has been practicing in good standing with a license in another state for at least a year and relocates to Arizona, Arizona will recognize that license upon relocating.53 This reduces one of the largest barriers to occupational mobility and prevents regulatory boards from imposing arbitrary criteria on individuals who are already trained and experienced in their trades. It also holds regulators accountable by requiring them to justify imposing restrictions on in-state licensees that do not exist in other states.
The Right to Earn a Living Act and the Universal Recognition Act also help address one of the most persistent problems with regulatory agencies: their tendency to serve the interests of regulated industries instead of the public by creating barriers to competition. In many cases, regulators have imposed restrictions on entry into a trade or profession that do not protect public health or safety, but only make it harder to compete against those who already have licenses. For example, barbering and cosmetology licensing boards have sought to prohibit people from braiding hair or even blow-drying hair if they lack government-issued barbering licenses—even though these individuals are not cutting hair, using chemicals, or otherwise threatening the public health or safety. (Fortunately in 2019, Arizona also passed legislation specifically allowing blow-drying without a license.54) Such restrictions do not protect the public but do protect the private financial interest of existing companies who prefer not to compete economically.
These two pieces of legislation help refocus the attention of regulators on their true mission, which is protecting the public interest.
Still, licensing and permit requirements can be a significant burden on individual rights and on economic productivity simply due to their vagueness or lack of time constraints. That issue has been the focus of another reform proposal offered in Arizona and other states in recent months: Permit Freedom.
The Permit Freedom Act offers three commonsense—and constitutionally obligatory—proposals to reform the way permits and licenses are issued.
First, when government requires a person to obtain a license or a permit for anything, the criteria for obtaining that permit must be clear and unambiguous—as opposed to vague language such as “good cause.” Second, the applicant should be given a specific date on which the permit will either be granted or officially denied, as opposed to letting the agency indefinitely delay action on the application.55 Finally, if an applicant is denied a permit and wishes to challenge that in court, the applicant should be given an unbiased hearing before an independent judge instead of having the case heard by the same bureaucrats who denied the permit to begin with.
These three requirements are technically already required by the federal Constitution. In the 1950s, the U.S. Supreme Court issued a series of decisions involving permits to show films, hold parades, or engage in other expressive activities. The Court said that while it is acceptable for government to require permits, it must also give applicants these three “procedural safeguards.”56 And these safeguards apply not only to permit requirements that relate to freedom of speech, but to permit requirements that affect any of the “freedoms which the Constitution guarantees.”57
Nevertheless, these requirements are routinely ignored by state and local officials who phrase permit requirements in ambiguous terms, provide no specific deadlines, and then force rejected applicants to appeal a decision to an administrative agency hearing rather than a court.
Consider the case of Lee Sepanek, who was threatened by the city of Phoenix with a series of vague demands for decorating his home with holiday lights. For over 30 years, Lee and his wife created one of the most impressive and anticipated holiday displays in Arizona and opened their beautiful display up to friends, neighbors, and visitors to enjoy. But in 2018, city representatives threatened to shut down the Sepaneks’ display and ordered them to stop offering free hot chocolate and cookies to visitors. Among other things, the city claimed that the Sepaneks would need a “mobile food vending” permit—even though they were not selling the cocoa or cookies and were offering them to people from a table in their yard, so were not “mobile.” Securing such a permit would have required the family to obtain space at a restaurant to serve as a vending facility and to obtain health department approval. Yet the city also refused to say whether obtaining this permit would allow the Sepaneks to continue displaying their lights and serving cocoa. Instead, officials ordered the couple to provide parking for visitors—without ever specifying what would satisfy the city’s demands.
Lee’s case is not unusual: Permit applicants often are not given objective guidelines that clarify what will and won’t qualify them for permits, but instead are provided vague, incomprehensible criteria such as “good cause,” which maximize the power of bureaucrats to decide when to issue a permit. And absent a specific deadline on a permit application, the government can delay a decision indefinitely, knowing that the applicant cannot sue as long as the government doesn’t issue a final decision, a rule judges call “ripeness.” And if the government does issue a final decision, and the individual sues, he or she is usually required to participate in an administrative hearing first—a hearing overseen by the agency itself—instead of going to a judicial court. This deprives people of the right to have their cases decided by an unbiased decision-maker. In fact, because administrative hearings are often exempt from the evidentiary and procedural rules that apply in courts, people have little real chance of prevailing.
Fortunately, after being threatened with litigation, the city of Phoenix backed down in Lee’s case. But similar acts of arbitrary government action occur daily. The Permit Freedom Act remedies these problems, not by repealing any existing permit or licensing requirements, but by ensuring that the “procedural safeguards” that should already apply are in fact respected as a matter of state law.
Since at least the 1970s, sunset requirements have been viewed as a helpful way to periodically review the effectiveness of agency regulations and, when they prove obsolete, to eliminate them. Sunsetting takes different forms: Some sunset provisions provide that a law will expire automatically unless it is renewed; others require government agencies to reevaluate their effectiveness at certain intervals; still others simply transfer power from one agency to another after a certain lapse of time. The effectiveness of sunset requirements is debatable, however. One U.S. News study found that most states rarely eliminate agencies or laws through sunsetting.58 This may, however, undervalue the degree to which sunsetting works as a threat to prevent agencies from engaging in abuse. And in some instances, sunsetting requirements have proven quite effective.
For example, two states recently allowed their codes of administrative regulations to sunset. In Idaho, the legislature must reauthorize the administrative code each year.59 In 2019, the legislature did not do so, and the code expired on July 1, 2019.60 The governor then worked with the legislature to review the code and reauthorize those rules that were effective, but not those that were not. Similarly, in 2016, Rhode Island required state regulators to refile existing rules as it moved to an electronic system for administrative rules.61 If agencies failed to do so, the rule would no longer be enforceable. Rhode Island’s sunset review process led to the elimination of several outdated or ineffective rules, some 31% of the state’s administrative code.62
Arizona recently introduced legislation called the Red Tape Review Act, developed by the Goldwater Institute, which combines the Rhode Island and Idaho measures. The Act would require the state legislature to annually review and reauthorize the administrative code and would provide for a one-time sunset of the entire code so that it can be reviewed. During this process, rules that are outdated, costly, or ineffective need not be reauthorized, while those that work well can be renewed and strengthened if necessary. Although state law already requires agencies to periodically review their own rules, it is imperative that the legislature ultimately take charge—as state constitutions require—by reviewing and reauthorizing the state’s regulations. That alone can avoid self-serving by agencies and respect both democratic principles and the separation of powers by ensuring that the elected, lawmaking branch of government has the final say on the restrictions that affect the life of every Arizonan.
Many regulatory restrictions were put on the books decades ago and have not been seriously examined since. In Arizona, for example, there are over 64,000 restrictions that include such antiquated and odd rules as dictating which prizes can be given away at the Arizona State Fair (water guns are prohibited!63) and barring individuals who hold liquor licenses from storing liquor at any location other than that identified on the license (hopefully the restaurant doesn’t flood!).64
Most Americans clean their homes periodically, getting rid of clutter and keeping the good stuff. When thousands of restrictions govern our daily lives, we should expect our government to do the same. This is particularly true when most of these laws were not put in place by lawmakers in the first place but were created by unelected and mostly unaccountable administrative agencies.
The genius of the U.S. Constitution is the careful separation and balance of powers. As James Madison observed, “The accumulation of all powers legislative, executive and judiciary in the same hands, whether of one, a few or many, and whether hereditary, self-appointed, or elective, may justly be pronounced the very definition of tyranny.”65 Yet that is precisely what the modern administrative state does.
Governors, presidents, and legislators have often spoken of the need to reduce the size and scope of administrative power and have taken steps to repeal specific rules or eliminate certain regulations. Yet the power of the administrative state continues to grow because its agents are unaccountable. The package of proposals outlined here attempts to address this accountability problem. Madison knew, as lawmakers should, that a government of separate powers under which its actors are made to answer for their decisions is the only type of government capable of preserving liberty.
1Philip Hamburger, Is Administrative Law Unlawful? (Chicago: University of Chicago Press, 2014), 509.
2John Marini and Ken Masugi, Unmasking the Administrative State: The Crisis of American Politics in the Twenty-First Century (New York: Encounter Books, 2019).
3295 U.S. 602 (1935).
4Id. at 629.
5Id. at 624, 628.
6Geoffrey P. Miller, “Independent Agencies,” Supreme Court Review 1986: 44.
7Richard G. Niemi & Joshua J. Dyck, Guide to State Politics and Policy (Washington, D.C.: CQ Press, 2013), 248.
8Mass. Gen. Laws ch. 7, § 4A(a).
9Ca. Gov. Code § 9148.50 et seq. requires periodic review of state agencies, but unlike true “sunsetting,” the elimination of an agency is not accompanied by an elimination of its powers.
10See Adam Thierer, “Regulatory Capture: What the Experts Have Found,” Technology Liberation Front, December 19, 2010, https://techliberation.com/2010/12/19/regulatory-capture-what-the-experts-have-found/.
11See, e.g., North Carolina State Bd. of Dental Examiners v. FTC, 574 U.S. 494 (2015).
122 Public Papers of the Presidents: Ronald Reagan 1988-1989 (Washington, D.C.: Government Printing Office, 1991), 1610.
13Marine Forests Soc’y v. California Coastal Com., 36 Cal. 4th 1, 25 (2005).
14Reg Stats, GW Columbian College of Arts & Sciences, https://regulatorystudies.columbian.gwu.edu/reg-stats.
15Patrick McLaughlin, “Comparing Regulations Across U.S. States,” Mercatus Center, August 7, 2018, https:// www.mercatus.org/publications/regulation/comparing-regulations-across-us-states
16See, e.g., Sackett v. EPA, 566 U.S. 120 (2012).
17Chevron, U.S.A., Inc. v. Nat. Res. Def. Council, Inc., 467 U.S. 837, 843 (1984).
18Auer v. Robbins, 519 U.S. 452, 461 (1997).
19Gutierrez-Brizuela v. Lynch, 834 F.3d 1142, 1149 (10th Cir. 2016).
205 U.S.C. § 706.
21Pub. L. 79-404, 60 Stat. 237.
225 U.S.C. § 706.
23Marbury v. Madison, 5 U.S. 137, 177 (1803).
24Chevron, 467 U.S. at 844.
25Encino Motorcars, LLC v. Navarro, 136 S. Ct. 2117, 2125 (2016).
26Gutierrez-Brizuela, 834 F.3d at 1152.
27Auer, 519 U.S. at 461.
28In the 2019 term, the Supreme Court heard Kisor v. Wilkie, a case that, among other things, asked the Justices to overrule Auer deference. There, the Court arguably modified how Auer deference works—but declined to overrule it. 139 S. Ct. 2400, 2423 (2019).
29Flytenow, Inc. v. F.A.A., 808 F.3d 882, 890 (D.C. Cir. 2015).
30See, e.g., QCC, Inc. v. Hall, 757 So. 2d 1115, 1117 (Ala. 2000); Bell Atl. Mobile, Inc. v. Dep’t of Pub. Util. Control, 754 A.2d 128 (Conn. 2000); In re Water Use Permit Applications, 94 Hawai’i 97, 144-145 (2000); Canty v. Idaho State Tax Com’n, 59 P.3d 983, 988-989 (2002); People ex rel. Birkett v. City of Chicago, 779 N.E.2d 875, 881 (2002); Reifschneider v. State, 17 P.3d 907, 913 (2001); Davis v. State Bd. of Certified Public Accountants of Louisiana, 131 So.3d 391, 399 (2013); Maryland Aviation Admin. V. Noland, 386 Md. 556, 572 (2005); Project Extra Mile v. Nebraska Liquor Control Com’n, 283 Neb. 379, 395 (2012); In re Town of Sebrook, 44 A.3d 518, 524-525 (2012); TAC Associates v. New Jersey Dept. of Environmental Protection, 202 N.J. 533, 541-542 (2010); Lorillard Tobacco Co. v. Roth, 786 N.E.2d 7, 10 (2003); North Carolina Acupuncture Licensing Board v. North Carolina Board of Physical Therapy Examiners, 821 S.E.2d 376, 379-380 (2018); Industrial Contractors Inc., v. Workforce Safety & Insurance, 772 N.W.2d 582, 585 (2009); In re Protest of Betts Telecom Oklahoma, Inc., 178 P.3d 197, 199 (2008); Seeton v. Pennsylvania Game Com’n, 594 Pa. 563, 578 (2007); In re Williston Inn Group, 949 A.2d 1073, 1077 (2008).
31Hamburger, The Administrative State, 43.
32William Yeatman, “An Empirical Defense of Auer Step Zero,” Georgetown Law Journal, 106, no. 2, (2018): 515, 545 (finding that government agencies won 78% of cases from 1993-2006, when Auer deference was most permissive compared to 71% of cases after 2006, when the Supreme Court began to limit that deference doctrine).
33Although in recent years, some states have begun scaling this back. The Wisconsin and Mississippi Supreme Courts issued decisions in 2018 that repudiated their versions of Chevron deference. Tetra Tech EC, Inc. v. Wisconsin Dep’t of Revenue, 382 Wis. 2d 496, 564 (2018); King v. Mississippi Military Dep’t, 245 So. 3d 404, 407-08 (Miss. 2018). Deferring to the legal interpretations of agencies, declared the Wisconsin justices, “is unsound in principle” because it “does not respect the separation of powers, gives insufficient consideration to the parties’ due process interest in a neutral and independent judiciary, and ‘risks perpetuating erroneous declarations of the law.’” Id. Their Mississippi colleagues agreed: “Deferential review of executive-branch statutory interpretations,” they said, conflicts with the state constitution’s “strict constitutional separation of powers. … [I]n deciding no longer to give deference to agency interpretations, we step fully into the role the [Mississippi Constitution] provides for the courts and the courts alone, to interpret statutes.” Id.
345 U.S.C. § 706.
35Colo. Rev. Stat. § 24-4-106; Conn. Gen. Stat. § 4-183; D.C. Code § 2-510; Ga. Code § 50-13-19; Haw. Rev. Stat. § 91-14; Idaho Code § 67-5279; Ind. Code § 4-21.5-5-14; Iowa Code § 17A.19; Kan. Stat. § 77-621; Ky. Rev. Stat. § 13B.150; La. Rev. Stat. § 49:964; Me. Rev. Stat. tit. 5, § 11007; Md. Code, State Gov’t § 10-222; Mass. Gen. Laws ch. 30A, § 14; Mich. Comp. Laws § 24.306; Mo. Rev. Stat. § 536.140; Mont. Code § 2-4-704; Neb. Rev. Stat. § 84-917; Nev. Rev. Stat. § 233B.135; N.C. Gen. Stat. §§ 150B-43, 150B-51; N.D. Cent. Code §§ 28-32-46, 28-32-47; R.I. Gen. Laws § 42-35-15; S.C. Code § 1-23-380; Tenn. Code § 4-5-322; Tex. Gov’t Code §§ 2001.172, 2001.174; Utah Code § 63G-4-403; Wash. Rev. Code § 34.05.570; W. Va. Code § 29A-5-4.
36A.R.S. § 12-910(E).
37See, e.g., JH2K I LLC v. Arizona Dep’t of Health Servs., 246 Ariz. 307, 310 (Ct. App. 2019).
38The Separation of Powers Act of 2016, 114th Cong., 2d Session, available at https://www.grassley.senate. gov/sites/default/files/judiciary/upload/Regs%2C%2003-18-16%2C%20Separation%20of%20Powers%20 Restoration%20Act%20Bill%20Text.pdf.
39Nebbia v. New York, 291 U.S. 502 (1934).
40FCC v. Beach Communications, 508 U.S. 307, 318 (1993).
41Timothy Sandefur, “Disputing the Dogma of Deference,” Texas Review of Law & Politics 18, no. 1 (Fall 2013): 121.
42Hettinga v. United States, 677 F.3d 471, 482–83 (D.C. Cir. 2012) (Brown & Sentelle, JJ., concurring).
43Patel v. Texas Dep’t of Licensing & Regulation, 469 S.W.3d 69, 98 (Tex. 2015) (Willett, J., concurring).
44State v. Lupo, 984 So. 2d 395, 408 (Ala. 2007) (Parker, J., concurring).
45Dick Carpenter and Lisa Knepper, License to Work: A National Study of Burdens from Occupational Licensing, 2nd ed. (Institute for Justice, 2017), 6, https://ij.org/wp-content/themes/ijorg/images/ltw2/License_ to_Work_2nd_Edition.pdf.
46Mark Flatten, Protection Racket: Occupational Licensing Laws and the Right to Earn a Living, Goldwater Institute, December 6, 2017, https://www.goldwaterinstitute.org/article/protection-racket-occupational-licensing-laws- and/.
47Vt. Stat. Ann. tit. 20, § 2731, et seq.
48W. Va. Code Ann. § 47-1A-7.
49Id. at § 3:3808.
50Ariz. Rev. Stat. Ann. § 41-1093, et seq.
51Angela Gonzales, “Goldwater Institute Helps Behavioral Health Counselor Continue Practicing in Arizona,” Phoenix Business Journal, November 7, 2017, https://www.bizjournals.com/phoenix/news/2017/11/07/ goldwater-institute-helps-behavioral-health.html.
52H.B. 2569 (2019), codified at A.R.S. § 32-4302.
53S.B. 1401, codified at A.R.S. § 32-506.
54Several states already have laws whereby an application for a permit is deemed approved if a licensing agency fails to act on it within a specified amount of time. See, e.g., HI St. § 91-13.5; Minn. Stat. § 15.99. Such laws are usually confined to certain narrowly limited classes of cases and are sometimes weakened by certain procedural requirements. But they provide a useful tool for protecting individuals against agency abuses while also creating a valuable incentive for agencies to act promptly rather than employ expensive and oppressive delaying tactics. See Nick Dranias, “The Local Liberty Charter: Restoring Grassroots Liberty to Restrain Cities Gone Wild,” Phoenix Law Review 3, no 113 (2010): 157
55Freedman v. Maryland, 380 U.S. 51, 58 (1965).
56Staub v. City of Baxley, 355 U.S. 313, 322 (1958).
57Brian Baugus and Feler Bose, “Sunsetting Government Waste,” U.S. News & World Report, September 1, 2015, https://www.usnews.com/opinion/economic-intelligence/2015/09/01/sunset-clauses-leads-to-political- power-plays-but-some-good-government.
58Idaho Stat. § 67-5292.
59James Freeman, “The Great Idaho Do-Over,” Wall Street Journal, May 16, 2019.
60Rhode Island Senate, RI S3015, 2016, Regular Session.
61James Broughel, “Release and Replace the Code of Federal Regulations,” Investor’s Business Daily, December 11, 2018.
64The Federalist, no. 47 (J. Cooke, ed., Middletown, CT: Wesleyan University Press, 1961), 324.