Rolling Back Arbitrary Government
Most civil servants put the “civil” and “service” into Civil Service. It’s the political appointees empowered by arbitrary executive orders whom you have to watch out for.
Politico posted a piece titled “Many in government are worried about Trump’s return. At DOJ, they’re terrified.” The DOJ, of course, is the Department of Justice. Politico reports that personnel may feel compelled to leave any number of the various “divisions” of the Department—with the exception of the Antitrust Division. The Antitrust Division will be relieved of the terror, because the Division is already going after Big Tech firms, just the kind of thing the incoming administration would want to do, anyway.
I will agree with Politico that the Antitrust Division would likely be relieved of the terror but for a much more fundamental reason: It operates out of a very mature caselaw process. Political appointees may come and go, and many of those appointees may hate the fact that the Division operates within a caselaw process—they would prefer to convert the antitrust enterprise into an administrative “rulemaking” process by which they would get to make up “administrative law” and “rules” as they go along—but they can’t just make stuff up given the Division is subject to the discipline of a court-ordered process. If the Division wants to do anything, then outside parties can exercise their option to compel the Division to make its case in front of a judge. For example, the Division may move to block a merger, but merging parties can force the Division to make its case in court. The Division can’t simply do some type of internal analysis and block a merger just because it says so.
At the same time, however, being subject to the discipline of court-ordered process does not render the Division impotent. It can bully outside parties—again, parties to a proposed merger, say—with the threat of forcing them to put up with costly, protracted court-ordered process. Outside parties might be willing cut a deal.
It turns out that other divisions of the DOJ do maintain some capacity to make administrative law by way of “rulemaking” capacity. The Civil Rights Division, for example, maintains some capacity for promulgating rules that amount to interpreting and implementing “Title IX” of the Education Amendments of 1972. Title IX is the kind of stuff that the federal government has used to require universities to spend more on women’s sports. Basically, the universities have been required to match spending on men’s sports with spending on women’s sports. A given university might spend a lot on its football program. It could choose to keep the program and cut other men’s programs—programs in wrestling and men’s gymnastics were often cut—as a way of balancing spending between men’s and women’s programs. Universities could also go out of their way to find any number of women’s programs to fund. Some modest number of universities opted instead to cut their costly football programs and to maintain a broader portfolio of less costly men’s programs.
The main proposition here is that subjecting government agencies to robust processes of “judicial review” encourages those same agencies to discipline themselves. Agency insiders really do factor into their decision-making the prospect of outside parties exercising their option to require those same insiders to make their case in court. I’ve witnessed this kind of thing myself with respect to “Tunney Act” proceedings as colleagues gathered around a conference table will have tossed in to debate declarations in the spirit of “We can’t do that, because certain-outside-party will complain [to the court]!” The Tunney Act is short for the Antitrust Procedures and Penalties Act of 1974, and the Act made way for outside parties to comment on deals the Antitrust Division might make to resolve matters involving mergers or market conduct.
That said, a second proposition: There is important variation across agencies in the robustness and scope of processes supporting judicial review. As noted, there is variation across divisions of the Department of Justice with (I would suggest) the Antitrust Division operating way out there on the frontier of robust processes and the Civil Rights Division operating largely under much less robust processes. Then there are entities that had been designed specifically to operate under less robust processes. A prominent example would be the Consumer Financial Protection Bureau created in 2010 in the wake of the 2008-2009 financial crisis. Then there is the Federal Trade Commission (FTC), brought to life with the Federal Trade Commission Act of 1914, as a way of explicitly relieving the antitrust enterprise of robust judicial review. The FTC Act invested the FTC with some rulemaking capacity, and it made way to staff the FTC with a cadre of administrative law judges. It was through the judges that the FTC would subject itself to an ersatz judicial review.
I leave it to the reader to contemplate whether or not enlisting the services of internal administrative law judges really amounts to implementing robust processes of judicial review, but I will suggest that the administrative law judge phenomenon really does constitute one of many ways of violating the “separation of powers.” Staffing an administrative agency with its own judges concentrates rule-making capacity (the business of making laws) with the business of reviewing laws (judicial review).
Now, a third proposition specific to the antitrust enterprise: Converting antitrust to an enterprise organized around centralized, administrative, rule-making processes from an enterprise organized around decentralized, court-ordered, caselaw processes amounts to the soft Sovietization of antitrust. It amounts to elevating bureaucrats to the role of industry-specific ministers. Imagine a Minister of Automobile Production or a Minister of Silicon Chip Manufacturing.
Or, imagine a Minister of Franchising. I am going to suggest that “Minister of Franchising” is not too fanciful. Specifically, the Biden FTC has been on something of a jihad against all forms of contracting involving small businesses. Among other things, the FTC has endeavored to impose a panoply of new rules governing franchising agreements. Why is a puzzle, but the Democratic party elites seem hostile to any form of business organization that does not conform to large corporate entities. Franchising constitutes one avenue through which Mom-and-Pop shops can get into business. (It also turns out that many franchisees are themselves very big, sophisticated entities that operate any number of sites and span many franchisor-franchisee relationships.)
Both franchisees and franchisors have puzzled about the motivations for these things and have come forward with prodigious streams of comments relating to FTC rule-making processes. This business of comments is explicitly laid out in the Administrative Procedures Act of 1946, and the Act was put in place in order to contain the excesses of administrative rule-making experienced during the years of Franklin Roosevelt’s administration. That said, it is not obvious how much discipline formal comment procedures really place on administrative agencies. The agencies can take the comments, respond to them and yet proceed to implement their new rules.
I myself was asked to assemble comments on the FTC’s proposed rules, and one complaint I’ve heard is that the FTC might demand from franchisees evidence (data) that would substantiate franchisors’ and franchisees’ claims that there really is little or no demand for new rules. A difficulty is that franchising is a very mature business, so there is almost nothing in the way of good natural experiments that could reveal how changes in rules governing franchising could generate much-improved results. (Conceivably, one could look at how different states in the country themselves treat franchising agreements, or one could examine changes in franchising contracts and behaviors before and after important changes in rules, but “mature business” means that there is very little scope for this kind of thing.)
I would not be the first person since 1914 to suggest that the FTC is just a menace-making agency by way of its rule-making capacity, but I will suggest that the Biden administration has made subtle efforts to convert the Antitrust Division into a quasi-rule-making entity. Specifically, the new administration came in and imposed some important changes to the Division’s Horizontal Merger Guidelines. Guidelines were first rolled out in 1968 as a way of making some commitment to outside parties to not conduct antitrust investigations in an arbitrary way. Guidelines effectively amount to a kind of law situated at the bottom of the totem pole, but that quasi-law would go some modest way (as intended) toward constraining the government. Guidelines have been updated over the decades, usually with technical tweaks, but until 2010, the rules have not really violated the idea that an important aspect of the court-ordered process is that the burden of proving a case is on the government. Since at least 2010, however, the language of “presumption” has entered the Guidelines. Certain noisy indications of “market power” like market shares might be presumed to indicate market power. It would be up to outside parties to present evidence of a lack of market power. That is, some of “the burden of proof” would be shifted from the government to private parties.
The new administration has made some effort to shift more of the burden of proof from the government to outside parties. In this way, the Guidelines concept is turned upside-down in that guidelines now constitute less of a constraint on government and more of a license for government to arbitrarily launch and protract investigations of private parties.
The current 2023 iteration of the Guidelines makes heavy use of the language of “presumption,” and it introduces much language about “dominant firms.” “Dominance” makes way for all types of arbitrary interpretations of the law, for, at any given time, some number of firms may appear “dominant,” but so what? There was a time when IBM appeared dominant. Microsoft was (and may still be) dominant in its way, but even it has been left out of the FAANG’s (Facebook, Apple, Amazon, Netflix, Google). A very long time ago, Sears-Roebuck was the dominant Amazon. And now we have firms like Nvidia on the AI frontier. Perhaps Nvidia should replace Netflix in the FAANG cluster.
Dominant firms come and go, but investing the antitrust enterprise with the language of dominance really just sets up the antitrust agencies as Ministries of Dominant Firms. What could go wrong? Politically-motivated abuse is one possibility.
The Obama Two-step
Administrative rule-making enables a lot of mischief, but it doesn’t have to be that way. What I am finding in some recent research, however, is that mischief (if any) really does come from the top. Outside observers may speculate about the “Deep State,” about a cadre of unseen, unaccountable bureaucrats who make mischief. But, what I have seen from being on the inside and from (more recently) systematically examining rule-making activity at the Department of Energy is that the bureaucrats who staff the Civil Service may or may not be very ideological. In fact, many of them may very much appreciate the integrity of their processes and may chafe against demands for them to subordinate those processes to arbitrary, flavor-of-the-month interpretations of the law. Rather, it is the cadre of senior political appointees that you have to watch out for. These are the people, for example, who endeavor to turn the Guidelines on their head, converting them from (admittedly soft) commitments on the part of the government to not be abusive to licenses for the government to affirmatively be abusive. These are the people who Presidents may empower with executive orders to implement abusive rules.
An example of abuse would be the “Dear Colleague letters” sent out from the Department of Education during the Obama years. The 2011 letter to universities encouraged them to shift the burden of proof in sexual harassment cases from accusers to the accused. Basically, you’re guilty until you prove yourself (through arbitrary processes afforded by university Star Chambers) innocent. Good luck with that. “Believe all women!”
Then there was the “Dear Colleague letter” of 2014. The letter to public schools implied that schools might be subject to federal investigations and might lose access to federal funds were they not to do something about racial disparities involving the disciplining of students. Black students routinely had been subject to school suspensions and other modes of discipline far more than students of other races. Schools subsequently made efforts to make the frequencies of disciplinary actions conform more closely to the proportions or students by race in the student populations. The result was much less discipline in schools. And we wonder why, despite spending about $70 billion a year on the Department of Education, student performance has continued to decline. Hmm…
President Obama would provide cover for the abuse the administrative agencies would impose on the country by appealing to his one and only rhetorical gimmick: He would say something open to positive interpretation while, literally in the next breath, declaring something contrary to what he just had said. Hence his justification in 2014 for his aggressive and arbitrary abuse of “executive orders”: “Where congress isn't acting, I'll act on my own. I have got a pen and I got a phone.” So, Congress isn’t acting to do good stuff, but the Messiah will sweep in and dictate the good result. Even better: “We’re a nation of laws, but we’re also respecting the fact that we’re a nation of immigrants.” That was Obama in 2015 setting himself up to further undermine immigration law.
This is all part of Obama’s actual theory of government: The purpose of government is to “punish our enemies” and “reward our friends.”
Government-by-Executive-Order versus Sausage-making
The incoming Trump administration has made an explicit point of implementing a host of executive orders “on Day One.” Many of these orders will restore executive orders the previous Trump administration had implemented—and which the Biden administration had undone with its own host of executive orders on day one of the Biden administration. Is this really the best way to govern?
I am not sure about how to contain the abuse of executive orders. And they can be abused. “Japanese internment” was enabled by Franklin Roosevelt’s Executive Order 9066 in 1942. Obviously it struck the panicked military community and much of a panicked public as a good idea at the time. But, some observers would surely argue that there can be some place for executive orders. Sometimes stuff just needs to get done. Do we really have to always wait for the creaky sausage-making mechanisms of proper legislation to roll out its sausages? But it is no accident that executive orders occupy a place very low on the totem pole—somewhere above guidelines but far, far below legislation that will yet have had to pass through both houses of Congress and secure the President’s signature.
I am thinking that the election of 2024 may prove as consequential as the elections of 1980 and 1932. The 1980 election matched the ascendance of Margaret Thatcher in Britain and the Solidarity movement in Poland. These developments seem to have aligned for a reason: The global order was collapsing, although that collapse didn’t really become entirely obvious until, say, 1989. The 1932 election also marked the collapse of a precarious global order and enabled Franklin Roosevelt, supported by massive majorities in both houses of Congress, to implement much of a New Deal. Starting now, will we get something of a rollback of a global Great Reset and something of a Restoration of some sanity?