The administrative state

Deep State

King George III decried Thomas Jefferson in America’s Declaration of Independence “has erected a multitude of new offices, and sent hither swarms of officers to harass our people, and eat out their substance.” While Jefferson was channeling the colonists’ ire at imperial customs officials, his words resonate with anyone forced to deal with modern administrative bureaucracy.

Indeed, the problem with bureaucracy has gotten so bad that, in the U.S. at least, that it is often referred to as a fourth branch of government – the “administrative state,” or even the “deep state.” All of these descriptions mark an admission that administrative law has some form of existence outside the norms of western liberal democracy. Tackling the problem will require a return to those norms.

The administrative state is legitimate in the strict sense of the word. Everything about it has been set up by act of either Congress or Parliament. Courts have repeatedly found this process legitimate under the relevant constitutional requirements. The legislature, courts have found, has the right to delegate certain of its legislative powers to the executive, allowing the administration to write rules that have the force of law.

To American eyes, this may seem strange. The Constitution says that “all legislative powers … shall be vested in a Congress of the United States.” This would seem to preclude the delegation of such powers to the executive, which is a separate but equal branch of government. Delegation would seemingly breach this principle of separation of powers.

However, early in the republic’s existence, the U.S. Supreme Court blurred this once-bright line. In 1825, the court ruled that while Congress must deal with “important” subjects, mere details could be left up to the executive (Wayman v. Southard). A century later, in 1928, the court went further, saying that the legislature may delegate powers as long as there is an “intelligible principle” to which the executive must conform (J. W. Hampton, Jr. & Co. v. United States).

The edifice built on this slim foundation has proved mighty indeed. Huge numbers of executive agencies have sprung up – so many that even federal agencies cannot agree how many there are – the Department of Justice says 252, the U.S. Government Manual says 316, and says 443. Not only do these have rulemaking and enforcement powers, many of them have their own courts, whose procedures must be exhausted before one can even think about appealing to the U.S. federal courts.

In the beginning, these agencies generally regulated via enforcement, which generated great uncertainty for those on the receiving end of regulations. That meant that one might not know one was running afoul of administrative law before being charged with an offense.

Resulting quasi-judicial decisions set precedents for regulated parties to follow.

The recognition that this was a problem led to the passage of the Administrative Procedure Act, which set out “rules about rules.” All rules had to go through a period of notice and comment. The agency would publish a draft rule and request comment from affected parties and the public. The agency would assess the resulting comments in order to improve the draft rule.

Theoretically, this would mean significant stakeholder involvement in the creation of the rule, but it does not always work that way. Some controversial rules receive hundreds of thousands of comments, which are often simply ignored or dismissed. Some laws require a cost-benefit analysis, which sometimes takes the form of a discursive discussion of potential costs and benefits, when it is performed at all. Rules that are supposed to be submitted to Congress are sometimes never submitted at all, raising questions as to their actual validity.

In some cases, the agency performs an end run around the Administrative Procedure Act by implementing policy via guidance documents and other issuances. These notes, manuals, interpretations, and even web postings can significantly alter the way a given rule is applied, without a notice-and-comment period. In one case, the Obama administration changed a significant rule about the classification of independent contractors by means of a blog post.

All of this means that it is very hard for anyone to know whether they are affected by a regulation or not. According to my Competitive Enterprise Institute colleague Wayne Crews, 24,557 notices pertaining to regulation appeared in the Federal Register in 2016. He calculates there have been 550,489 public notices since 1994 and well over a million since the 1970s.

All this might be less of an issue if the courts imposed an effective block against excessive regulation. Yet courts have time and again found ways to defer to agency interpretation and imposition of law. The current guiding doctrines are known as “Chevron” and “Auer” deference.

Chevron deference holds that courts should defer to agency interpretation of a law passed by Congress as long as the agency does not misinterpret an unambiguous instruction and if the agency has a rational basis for its action. That means that even when another more reasonable interpretation is available, the courts should defer to the agency interpretation.

Auer deference goes further, saying that courts must defer to the agency that wrote a regulation for its interpretation, unless the interpretation is plainly erroneous.

Combined, the Chevron and Auer deference doctrines mean that what agencies say generally goes when it comes to regulation.

These doctrines are based on the notion that agencies generally have more expertise regarding the details of regulation and regulated industries than courts. It would take up a lot of court time to fully understand the complexities of a regulation and come to a ruling, so it is easier for courts to defer to the experts. Such an assumption of expertise is an example of what economist Harold Demsetz called the “Nirvana fallacy,” which assumes a perfect alternative, in this case to the unregulated state.

Predictably, Chevron and Auer deference tilt the playing field in favor of more activist government. All an agency has to do is come up with a rational basis for its regulation and write it in a manner consistent with that rationale – and the courts will not stand in the way.

It makes judicial challenge to a burdensome regulation extremely difficult. This is especially a problem in cases where agencies have discretion to promote new rules consistent with a general mission. The lawmaking process becomes a matter of executive whim rather than deliberation involving the people’s elected representatives, constrained by judicial review.

Yet, executive agencies still face a modicum of democratic accountability. The president, as head of the executive branch, may constrain hyperactive deputies via executive order. But Congress has set up agencies that are independent of the executive branch. These agencies, which are even more removed from the democratic process, generally have wide discretion to make rules or undertake enforcement actions. A commission structure provides a democratic fig leaf. Under this arrangement, commissioners are nominated by the president and confirmed by Congress. The ruling party generally has three commissioners and the opposition party has two, allowing for a form of internal deliberation that mimics Congressional decision-making.

Recently, however, there has been a move towards ignoring even the commission structure. Agencies like the Federal Housing Finance Agency, the Office of Financial Research, and the controversial Bureau of Consumer Financial Protection (BCFP) have been set up to be led by one individual director. Isolated from presidential control and effective oversight by the Congress or courts, these agencies have almost complete discretion to do what the director likes within the powers granted to them by Congress. In the case of the BCFP, those powers are extensive.

With such concentrated power and so little oversight, other considerations come into play. The theory of “public choice,” developed by the Virginia school of economics, suggests that bureaucrats will work to expand their power to their own advantage. The suggestion of a former head of the BCFP that it should “push the envelope” in exercising its powers is a case in point.

Similarly, during the Obama administration, a team of Department of Justice lawyers came up with a way to stop businesses they did not like from operating. “Operation Choke Point,” as they named it, used increased supervision of banks to dissuade them from providing banking services to a range of industries, including porn stars, firearms dealers and payday lenders. The rational basis was that these industries posed a “reputational risk” to the banks.

Yet, Congress had passed no law restricting these industries. Indeed, porn stars and firearms dealers are supposed to be protected by the First and Second Amendments, respectively. Nevertheless, a team of bureaucrats took it upon themselves to try to regulate them out of existence.

Another public choice-based critique of the administrative state is that the people who go to work at interventionist agencies generally agree with the interventionist mission. This implies that they are more likely to seek to expand their power than not, even in a burdensome manner. This problem is compounded by labor and employment law, which generally make it extremely difficult to fire a civil servant working under a union-negotiated collective bargaining agreement. This further insulates bureaucrats from any repercussions for following an agenda aimed at expanding their powers.

The modern administrative state has become a significant burden on the American economy. Wayne Crews estimates that the cost of regulation is roughly $2 trillion annually, around 10 percent of gross domestic product, and imposes a substantial burden on every U.S. household.

Remedies to this situation need to begin with the delegation principle. Congress should stop delegating as much as it has and reclaim some of the powers it has delegated away. One way to do this would be to establish a bipartisan “Regulatory Reduction Commission” that would be tasked with reviewing existing regulations and recommending a package of regulations that are too costly, burdensome, or outdated to roll back every year. New regulations and agencies should be established with sunset clauses whereby the rules can be repealed and agencies shuttered when they have run their course (although the persistence of the Export-Import Bank after its authority expired suggests this is more easily said than done).

Courts must also play their part. There is some appetite on the current Supreme Court to revise the deference doctrines, but if it declines to do so, Congress can instruct them. The rational basis test needs to be addressed. If a less burdensome solution presents itself, there should be some way for an affected citizen to petition the government to adopt it.

The administrative state has been around for about a century. Its evolution into the current overarching problem should not have been a surprise. When British ministers were beginning to accrue similar powers to themselves in 1929, the then Lord Chief Justice, Lord Hewart, wrote a book warning of the erosion of self-government it represented. He titled it, “The New Despotism.”