Do Politics Matter?

Happy post-Superbowl Monday (a day that I think should be a federal holiday, but that is a post for another time).

Over the weekend I received an email about a recent paper written by Harvard Law Professor Alma Cohen. Prof. Cohen’s paper is an empirical study that shows the political affiliation of federal appellate judges can help predict the outcome in some cases. From the abstract:

This paper contributes to the long-standing debate on the extent to which the political affiliation of federal circuit court judges, as proxied by the party of their nominating president, predicts case outcomes. I compile and analyze a novel dataset containing more than 650,000 circuit court cases from 1985 to 2020. I find a pervasive association between political affiliation and outcomes in the vast universe of circuit court decisions that is not limited to the ideologically salient cases or published cases on which previous research has focused. The political composition of circuit court panels, I show, has a broad impact on practically all fields of US federal law.

In particular, I find an association between political affiliation and outcomes in each of the six categories of cases between litigants that judges could perceive as having unequal power. These categories include cases in which the government litigates against criminal defendants, prisoners, and immigrants, and cases in which individuals litigate against institutions. In each of the six case categories that I identify, the more Democratic judges a panel has, the higher the odds of the panel siding with the seemingly weaker party.

Furthermore, political affiliation is associated with outcomes in the large set of civil cases between parties that could be perceived to be of equal power. In these cases, panels with more Democratic judges are less likely than panels with fewer Democratic judges to defer to lower-court decisions.

Altogether, my analysis shows that political composition can help to predict outcomes in case categories that together represent over 90% of all circuit court cases. I conclude by discussing the implications of my findings for understanding the evolving body of decisions issued by circuit courts and for assessing the legal rules and arrangements that govern such courts.

A two interesting points from the paper. First, Prof. Cohen uses an exceptionally large sample of cases (670,000)–significantly larger than samples used in previous studies. She also includes unpublished cases.  Second, Prof. Cohen recognizes that using the politics of the appointing president might be an imperfect proxy for the politics of the judge. I agree. To the extent that future research wants to look at this issue, I would recommend also looking at which party controls the US Senate and the political affiliation of the home state senators (less relevant for circuit court noms, but still relevant).

I look forward to seeing more papers from Prof. Cohen as she explores this topic.