Want Your Opinions Reversed? Hire a Yale Clerk (and Don't Require the Bar)
Royce de Rohan Barondes
University of Missouri-Columbia School of Law
August 30, 2008
U of Missouri-Columbia School of Law Legal Studies Research Paper No. 2008-8
CLEA 2008 Meetings Paper
3rd Annual Conference on Empirical Legal Studies Papers
This paper examines whether clerks to federal district court judges influence judicial opinions by reviewing whether there is a relationship between characteristics of clerks hypothesized to be important and the likelihood of a subsequent Shepard's "warning" signal (roughly equivalent to a reversal). The clerk characteristics addressed are whether clerks are expected to take the bar and the law schools they attended. A finding that there is such a relationship provides evidence consistent with the hypothesis that clerks matter.
Using a sample of 36,107 opinions written by 220 federal district court judges, this paper finds: There is a negative relationship between the likelihood of a reversal and the fraction of a judge's clerks from U.S. News & World Report top 6 law schools where the clerks are expected to take the bar. Where clerks are not expected to take the bar, the higher fraction of the judge's clerks from Yale Law School is associated with a higher likelihood of a reversal. In this regard, Yale Law School seems in a class of its own among the top 18 law schools. Where judges do not expect clerks to take the bar, one can reject the hypothesis of equality of the parameter estimates for the fraction of clerks from Yale Law School and either the fraction of clerks from Harvard Law School or Stanford Law School, with p-values of 0.01 or smaller. These results are consistent with the hypothesis that clerks matter, and that Yale Law School is true to a description by its former Dean as "the most theoretically and academically oriented law school in America."
This paper reports results of conditional (fixed effects) logistic regressions, which estimate the likelihood of reversal conditional on the number of reversals each judge has in the sample. The results thus account for the possibility that judges inherently more likely to be reversed hire clerks with particular attributes.
Number of Pages in PDF File: 79
Keywords: judicial clerks, reversal rates, law schools
JEL Classification: K00, K40, K41, I21
Date posted: April 16, 2008 ; Last revised: December 10, 2012