Merry Christmas to the readers of Eli Lee blog. Thank you all for putting up with my verbosity and for providing feedback over these past few months. More to come in 2025!

Is Admin Law Epiphenomenal?

One of the best words that gets thrown around a lot at Yale Law School is “epiphenomenal.” If you ever find yourself in a Yale Law School classroom, and you haven’t been paying attention or don’t know what’s going on, just raise your hand and ask “excuse me professor, but isn’t this epiphenomenal?” Your question will be hailed as very intelligent and insightful.

All jokes aside, the question of whether, to what degree, and how legal doctrine is epiphenomenal is actually quite important. If legal decisionmaking is driven by forces other than the nature of doctrine, it would be essential for a practicing lawyer to have an understanding of these forces.

I want to offer a brief contribution to this discussion by looking at how markets reacted (or didn’t react) to what many legal scholars regard as a major change in administrative law doctrine. What’s interesting is that commentators on both sides of the debate–both liberal defenders and conservative critics of the admin state–agreed that the Supreme (EXTREME) Court’s decision in Loper Bright worked a major change in the balance of power between the administrative state, Congress, and the judiciary.

Ian Milhiser at Vox writes:

The Supreme Court handed down what is likely to be one of its most consequential modern-day decisions on Friday. Loper Bright Enterprises v. Raimondo fully consolidates the Court’s dominance over federal agencies within the executive branch of government. It is a radical reordering of the US separation of powers, giving the one unelected branch of government all of its own power, plus much of the power that Congress has vested in the executive branch.

On the other side, we have the conservative advocacy group ADF’s amicus brief:

Chevron deference threatens more than economic vibrancy and separation of powers principles. It also threatens fundamental rights. The absence of accountability for federal agency officials—combined with immense pressure on the executive branch to placate its political base—has made the federal administrative state increasingly susceptible to agendas that abuse the fundamental freedoms and values of the American people…When left to their own devices—or to the political calculations of the White House—agencies stretch and strain their authority to burden the everyday lives of American citizens in ways Congress never imagined, much less prescribed.

The wonderful thing about Loper Bright is that these kinds of dire predictions are empirically testable.

Some Background on the case

I actually don’t really know what Loper Bright says because I’ve never read it. You can think of this post as an elaborate justification for my refusal to care about these kinds of minutiae. It’s generally perceived as having overruled a legal doctrine called Chevron deference that required courts to defer to agencies’ interpretations of statutes as long as the statute was ambigious and the interpretation is reasonable. Loper Bright–as far as I can tell–got rid of this system and created a new doctrine that requires courts to impose their own interpretations of statutes when reviewing administrative agency actions.

At the very least, we can say that what Loper Bright does is alter the formal process by which courts justify their decisions about whether and how to supervise the administrative state. If you’re tempted by the thought that such formal processes are mere epiphenomena, i.e. that they are the results of other forces and have little or no causal weight of their own, then you wouldn’t think that changing the formal processes by which courts arrive at their decisions would matter very much. That’s precisely the hypothesis I intend to test.

My method

In securities litigation, a method called an event study is used to test whether a development created a material change in the price of a security. For example, in an insider trading case, the prosecution would use an event study to demonstrate that upon the release of a piece of information improperly disclosed or traded on by an insider, the company’s stock price changed.

The principle behind an event study on a single security is to isolate the changes in that securities’ price unique to that security. The movement of stock prices of comparable companies are used to isolate changes that were unique to the specific security being studied as distinguished from macro trends like changes in interest rates or industry-wide developments.

My aim is to replicate the logic of event studies to see whether the announcement of the Loper Bright created a significant change in the areas of the U.S. economy most subject to regulation. If Loper Bright really redistributes power away from the administrative state either to Congress (as its defenders on the right would say) or the judiciary (as its left-wing critics would have it), then we should expect highly-regulated industries to receive more favorable regulatory treatment and generate more value for their investors. This is because both of the institutions I mentioned (Congress and the judiciary) are more conservative than the administrative state, so if they gain more power to supervise regulators in the administrative state, then we should expect regulations to end up more lenient and therefore allow more profit in highly-regulated industries.

I don’t want to pretend that my method is the most statistically sophisticated one out there. I’m sure that there’s other and better ways to study how a court decision affected markets’ perceptions of certain industries fortunes. But understanding the precise way details of how markets reacted to Loper Bright is not my objective here. Instead, I just want to show that it did not create any big changes in (investors’ perceptions of) the fortunes of America’s highly-regulated industries. That seems like enough to provide evidence in favor of the view that administrative law doctrine is epiphenomenal and has somewhere between limited to zero causal power of its own.

A final elephant in the room that has to be addressed is the efficient market hypothesis (EMH). My argument requires some form of the EMH to make the deductive leap from the perspective implicit in market movements to the actual truth of how the decision affected the economy. Some legal scholars who spend too much time baking their brains on caselaw may want to say that if markets didn’t react to Loper Bright by adjusting the valuations of highly-regulated inustries, that’s just because the markets are wrong and don’t understand how significant the decision was.

All I have to say to that is that without being committed to any particular version, strong or weak, of the EMH, if you really think that the markets analyzed the decision incorrectly, then you’re a genius investor, and you can make billions of dollars by trading on your superior insight. You could also win a Nobel Prize for finding such clear evidence of the EMH being wrong in an area where it should be relatively strong (U.S. markets are highly liquid and secure). Anyone who is attracted by the argument I mentioned in the previous paragraph should seriously consider how likely those propositions are to be true.

Results

In an article on the use of event studies in insider trading litigation, a group of UPenn Carey Law professors summarize the method as follows:

A typical event study has five basic steps:

  1. Identify one or more appropriate event dates
  2. Calculate the security’s return on each event date
  3. Determine the security’s expected return for each event date
  4. Subtract the actual return from the expected return to compute the excess return for each event date
  5. Evaluate whether the resulting excess return is statistically significant at a chosen level of statistical significance

Let’s start with the dates. When should markets have priced in the effects of Loper Bright? The most obvious candidate would be the date on which the decision was released1. But it’s also possible that, given the conservative majority on the court, investors assumed that the Chevron would be overturned as soon as the case was granted cert.

I’m pretty skeptical of this view because even if investors priced in a high probability of Chevron being overturned when Loper Bright was granted cert, that probability would have been <1, so we should still observe price changes on the day when the decision was announced and raised the probability from <1 to 1. But just to be safe, we’ll test both the decision announcement date and the date on which cert was granted.

In fact, the above argument applies to any claim that the effects of Loper Bright were already priced in when some previous event happened (Kavanaugh confirmation, Trump election victory, Chevron already being overruled by the nth Circuit’s opinion in the Ooga v. Booga case, etc.) However strongly these events may have allowed investors to predict a result like the one in Loper Bright, they did not create complete certainty that the result would occur. We should therefore still see some price effect associated with the transition from high likelihood to certainty. Anyone who doubts the significance of the price effects when a highly probable event actually occurs should recall that there is an entire class of hedge funds pursuing merger arbitrage strategies whose profitability depends precisely on these kinds of high-probability-event-driven price effects.

Now we have to look at some security’s return. In what securities would Loper Bright’s effects show up? Given the extreme reactions of legal commentatators that we saw at the beginning of this post, it really should affect the entire American economy. But the decision would have a particularly large effect on the most highly-regulated industries because those are the industries most exposed to the caprices of the administrative state.

I hope it’s not too controversial to say that a diversified and representative group of the most highly regulated industries in the United States would be the following: healthcare, insurance, pharma, energy, and telecomms. Since we are looking at the fates of these industries in comparison to the entire U.S. economy, rather than examining a single security as event studies typically do, we will instead look at indices for the entire industry.

How then do we determine the expected return for these industry indices? That term refers to the return we should expect based on securities that share exposure to the same factors except for the news of which the event study is intended to test the causal signficance.

That’s a difficult concept when dealing with indices rather than single securities. In the latter scenario, we would typically use peer firms (similar business and market cap). But what constitutes a peer industry? We could look at industries of similar size, but what we are really trying to measure is the way that highly-regulated industries changed in ways that differ from the rest of the economy. If Loper Bright was such a significant regulatory development, then more regulated industries should see changes that the rest of the economy didn’t.

I think that the best way to calculate the expected return of the industry group we’re testing is therefore to use a measure of the entire American economy: the S&P Total Market Index. So, we’re going to look at if the indices for highly-regulated industries differed from the perfomance of the economy as a whole.

Let’s take a look at the daily returns surrounding both dates:2

Loper Bright Graph 1

Loper Bright Graph 2

I used a simple CAPM equation to estimate expected returns for these indices based on their correlation with the market index3

\[AAR_{i,t} = R_{i,t}-(α_i+β_iR_{m,t})\]

Decision date:

Screen Shot 2024-12-24 at 16 58 02

Cert date:

Screen Shot 2024-12-24 at 17 02 55

note: CAAR = cumulative above-average returns

Discussion

I would summarize these results as supporting three propositions:

  1. The markets don’t think Loper Bright affected the fortunes of highly-regulated industries in any decisive way.
  2. Either that is the correct assessment or you are a genius hedge fund investor (EMH).
  3. If Loper Bright did not affect any highly-regulated industries’ fortunes, then it’s not very plausible that excessively permissive legal doctrines are responsible for the administrative state’s power.

It may be worth it to say a little more about what that last point means.

The story that’s most popular on the American institutional right is that the Constitution intended for Congress to make federal laws such that much administrative rulemaking is therefore an invasion of Congress’s power by overeager federal bureacrats. Many of the people who believe in this story also believe that the reason the administrative state has gained this improper power is that courts have not been sufficiently aggressive in policing the boundary between Congressional and administrative power. If only courts would stop the administrative state from operating in zones that the Constitution empowers Congress to govern, these conservatives think, then the administrative state’s power could be limited.

The problem I have with this view is that there are lots of foreign countries that have already have all of the formal legal arrangements that conservatives want courts to affirm already baked into their constitutions, and these countries have administrative states that are just as powerful. Take the United Kingdom as an example. There is absolutely no ambiguity in the British constitution that Parliament has supreme power over all things in the country. If that’s right, then why do we see powerful administrative agencies in the UK making rules and major decisions in the exact same manner as in the United States?

I suspect the answer is something like this: any advanced economy generates its wealth through capitalism, and capitalism moves fast. Legislative bodies, on the other hand, are slow. They require lots of time for debate, analysis, and political maneuvering before they can arrive at a decision. Furthermore, the public character of their proceedings increases the costs of making certain kinds of decisions by subjecting each stage of their reasoning to the exacting scrutiny of the nosy democratic public. Because of this basic asymmetry between the rates of change in advanced capitalism and the slow pace of legislative bodies, a different kind of institution, one more able to arrive at quick and decisive decisions, is needed to effectively oversee advanced economies.

Carl Schmitt, writing almost a hundred years ago, makes this point better than I could ever hope to:

Great political and economic decisions on which the fate of mankind rests no longer result today (if they ever did) from balancing opinions in public debate and counterdebate. Such decisions are no longer the outcome of parliamentary debate…As things stand today, it is of course practically impossible [my emphasis] not to work with commitees, and increasingly smaller committees; in this way the parliamentary plenum gradually drifts away from its purpose (that is, from its public), and as a result becomes a mere façade.

Small and exclusive commmittees of parties or of party coalitions make their decisions behind closed doors, and what representatives of big capitalist interest groups agree to in the smallest committees is more important for the fate of millions of people, perhaps, than any political decisions. The idea of modern parliamentarism, the demand for checks, and the belief in openness and publicity were born in the struggle against the secret politics of absolute princes. The popular sense of freedom and justice was outraged by arcane practices that decided the fate of nations in secret resolutions. But how harmless and idyllic are the objects of cabinet politics in the seventeenth and eighteenth centuries compared with the fate that is at stake today and which is the subject of all manner of secrets [my note: the futile struggle to discpline technocapital].

In the face of this reality, the belief in a discussing public must suffer a terrible disillusionment…the smallest number still believe that just laws and the right politics can be achieved through newspaper articles, speeches at demonstrations, and parliamentary debates.

We simply can’t expect legislative bodies to competently oversee advanced economies because the latter change and move too quickly. So no matter what legal doctrines your constitution or courts espouse, it is inevitable that an administrative state will rise to power in order to handle the contingencies of capitalism. Courts can police administrative power at the margins and in individual controversial cases, but they can never transfer power back to legislative bodies. That’s because those bodies are innately unable to discharge the responsibilities of advanced economic policymaking. This is a proposition supported by the governance history of every advanced economy on earth.

If you don’t like the way the administrative state in the United States currently operates, and I certainly am not its biggest fan, then you either have to make do without advanced capitalism or propose another mechanism that can oversee advanced capitalism with equal or greater competency. Congress is not a serious candidate for such a mechanism, so in the end all we are talking about is alternative forms of the administrative state. I hope conservatives who have valid concerns about the administrative state will take this lesson seriously and stop putting energy into legal changes that in no way influence the historical forces that are actually responsible for the administrative state’s power.

All of the above are quite bold theoretical speculations that may not be fully supported by my results. But even if I’m wrong about all that, I think the results are at least interesting in that they show that the consensus among intelligent and well-informed investors was that Loper Bright didn’t change anything about how businesses are regulated in the United States in a way that would make them decisively better or worse off. Given the extremely enthusiastic and apocalyptic reception of the decision on both sides of the admin state debate, both sides could stand to reevaluate how they understand the decision. Whatever else it did, it does not appear to have affected the American economy in any significant way.

  1. A problem with this date is that it was the day after the famous Trump-Biden presidential debate, which in hindsight could be taken as a news item increasing the probability of a Trump victory. That’s why testing the cert date in addition is helpful. A longer event window should also help show us lasting effects of an admin law change as opposed to momentary boost from Trump debate buzz. I would also like to note that because the Trump admin is generally perceived as deregulatory, this issue is more likely to result in a Type 1 error (falsely rejecting a null hypothesis) than a Type 2. Since my argument relies on declining to reject the null hypothesis, this is not as dangerous of a problem. ↩

  2. To make replicating my results easier for any interested readers, I’m sharing the dataset I used. Readers, especially skeptical ones, should feel encouraged to use the the the built-in Google Finance data to test different event dates, indices, estimation windows, or more sophisticated calculations for expected returns than my simple CAPM model. I would be happy to discuss and/or update this post with any interesting discoveries. ↩

  3. I used an estimation window of -120 days from event date to -11 days. I assumed a normal distribution for all indices. My event observation period was 10 trading days, which is slightly longer than the 5 day period event studies typically study. I made this choice in order to be extra sure that the market had time to digest the news of Loper Bright. For an overview of my code and additional information on how to run event studies in Stata, see this paper. ↩

<
Previous Post
A Discussion of Wasteful Plaintiff-side Litigation and Proposal for Reform
>
Next Post
Capitalism sucks when you haven’t just blown up the entire world’s capital stock