Today's New York Times reports on a new study by three economists purporting to show that the Supreme Court's decisions are increasingly tilted toward the interests of the rich, and suggesting that this may play a role in rising economic inequality. Here's how the NYT story begins:

Supreme Court justices take two oaths. The first, required of all federal officials, is a promise to support the Constitution. The second, a judicial oath, is more specific. It requires them, among other things, to "do equal right to the poor and to the rich."

A new study being released on Monday from economists at Yale and Columbia contends that the Supreme Court has in recent decades fallen short of that vow.

The study, called "Ruling for the Rich," concludes that the wealthy have the wind at their backs before the justices and that a good way to guess the outcome of a case is to follow the money.

The study, "Ruling for the Rich: The Supreme Court Over Time," by Andrea Prat, Fiona Scott Morton, and Jacob Spitz, was posted as an NBER working paper, and has not yet been peer-reviewed or (to my knowledge) accepted anywhere for publication. But that did not stop it from getting a the full-article treatment in the NYT.

Having looked at the study, I don't think it does much of anything to reliably substantiate its claims. It is both conceptually muddled and poorly executed. At most it shows that (surprise!) Republican Supreme Court nominees have become more conservative over time (a point I made to the Times), but that is hardly a revelation.

The biggest problem with the study's design is its adoption of overly simplistic conceptions of what makes a decision "pro-rich" or "pro-poor." Here is what the authors say they are doing:

We measure whether a justice's vote moves money from poor to rich, and not on the words they use to justify that decision. Our approach provides a practical alternative to predicting judicial behavior, and one that is based on an empirical methodology. Our framing is particularly helpful to the prediction exercise when the case is not about a hot-button cultural issue like abortion, but instead on a topic like tax or regulation. Making the rich richer may not be an ideology that is easily justifiable to ordinary citizens, but does a better job at explaining decisions than theories of statutory or constitutional interpretation, e.g. originalism. . . .

We categorize the parties in these cases as "rich" or "poor" according to their likelihood of being wealthy. A justice's vote is pro-rich if its outcome would directly shift resources to the party that is more likely to be wealthy. . . .

Set aside whether it is meaningful to look at cases in this way, to make their assessment the authors adopt a set of arbitrary (and value-laden) assumptions about what makes a given decision "pro-rich" or "pro-poor." Among other things, they assume that all decisions in favor of a private party challenging a government regulation are "pro rich." As they explain:

Any type of government constraint on a business prevents it from doing what it otherwise would, which is most typically maximizing its profits. Thus economic regulation is generally costly to business, but can nonetheless be enacted into law in a democracy because of its benefits to citizens. For example, a company must bear the cost of abating pollution, but after it does so, citizens are no longer harmed by dirty air and water. Governments create many rules that constrain corporations' actions, including financial regulation, labor regulation, safety regulation, environmental regulation, etc., which are designed for the broader benefit of society. We categorize votes on the Court that support businesses over a government rule that the business is challenging as pro-rich.

This analysis conflates the social desirability of government intervention, perhaps to serve some public interest, with the potential distributional consequences of such a policy, as if they are one and the same. And do the authors really mean to embrace the proposition that government policy is never manipulated to serve the interests of the wealthy? Should regulations constricting housing supply should be categorized as pro-poor and anti-rich?

Focusing on environmental regulation (which the authors choose to highlight), it has been long understood, environmental regulation in particular is often a reflection of elite policy preferences and such regulation often has regressive economic effects. Protecting environmental resources and values often means adopting policies that increase the costs of goods and services. Good or bad, it is not clear why either side should be characterized as "pro-rich" or "pro-poor." And have these authors never heard of rent-seeking? Are they really unaware that regulation (including environmental regulation) often advances the interests of privileged interests or incumbent firms at the expense of consumers?

When it comes to cases pitting firms against firms, the authors again embed their ideological priors into their study design. They write:

When one corporation is in a dispute with another, we determine if consumers (the "poor" in our taxonomy) belong on one side. A win for the plaintiff in a private antitrust case, for example, should redistribute monopoly profits to consumers, and therefore a vote for the plaintiff is pro-poor.

Here they have simply adopted the highly contests assumption that whether an antitrust plaintiff prevails tells us anything about whether the decision benefits the rich or the poor. Indeed, the entire premise of contemporary antitrust jurisprudence is that outcomes should be driven by the pursuit of "consumer welfare" such that if an antitrust defendant prevails, this is because there are reasons to presume that this outcome--and not an outcome for the plaintiff--is better for consumers. My point here is not that current antitrust doctrine is right or wrong, only that the authors' case characterization assumes what existing doctrine contests: that a victory for a plaintiff is a ruling against the rich. Similar concerns could be raised about their coding of class-action cases, which some would argue benefit wealthy plaintiffs' lawyers at the expense of actual consumers.

All this is bad enough, but it is when the authors turn to their "counterfactual analysis" to see what might have happened had President Trump not altered the composition of the Supreme Court that things really go off the rails.

In this part of the paper, the authors look at how their conclusions might have turned out differently had Democratic Presidents filled one or more of the Supreme Court seats ultimately filled by President Trump. In their first counterfactual scenario, for instance, they consider what would have happened had Justice Ginsburg retired in 2016, allowing President Obama to appoint her replacement and preventing the ultimate appointment of Justice Barrett. According to the authors, this would have changed the outcome in four of cases from "pro-rich" to "pro-poor." (See Table 5 on page 29.)

The first such example they give is Washington State Department of Licensing v. Cougar Den, Inc. Set aside how the authors concluded that this was a "pro-rich" decision, how is it that the result would have flipped in their counterfactual? Justice Ginsburg joined the judgement of the Court and all of the other Democratic appointees supported the outcome. Perhaps more importantly, this case was decided in March 2019, so how could replacing RBG with a Kagan clone in 2016 (and preventing ACB's appointment in 2020) have affected this outcome at all?

The second example in this table is perhaps even worse: June Medical Services LLC v. Russo. Here, again, we have a case in which RBG and the Democratic appointees were already in the majority, so replacing RBG with a Democratic appointee would have no conceivable impact. And here again we have a case that could not be affected at all by preventing the appointment of ACB because it was decided before ACB was on the bench (June 2020).

June Medical also seems like an odd case to highlight if one is focused on whether Supreme Court decisions are "pro-rich." I guess this case is categorized by the authors as "pro-rich" because the Court ruled against the government on behalf of a business entity, but did the authors really mean to characterize this case in such terms? June Medical was an abortion case. I doubt any justice viewed this case in economic terms, and I am highly suspect that many of those celebrating this study on BlueSky would agree that a decision striking down a state abortion regulation is "pro-rich." (After all, this would mean Dobbs was "pro-poor.")

I think it is important to note that I did not cherry-pick these examples out of the authors' underlying data set; I do not have access to their unpublished data or coding. Rather, these examples are literally the first highlighted by the authors themselves in Table 5. So if the coding and analysis is problematic here, I shudder to think what might be found if other researchers are allowed to check under the hood and scrutinize the data underlying the authors' results.

Longtime readers know that I have long been skeptical of studies purporting to claim that the Supreme Court is "pro-business." I think that such labels often obscure more than they illuminate. I have also argued that insofar as one thinks "pro-business" analyses are useful, there are reasons to think the Trump appointees may make the Court less pro-business than it had been before. But this sort of conclusion requires more careful consideration of the actual cases the Court is considering and what is actually at issue. Sloppy characterizations that embed contestable ideological premises is not any way to develop a better understanding of this Court or its effect on American life.