Jackpot Justice? An Analysis of Inconsistency in the Ninth Circuit

• In Event: US Law & Courts: Courts & Judging Mini-Conference: Lower & Administrative Courts

Abstract

Critiques of judicial decision making in the Ninth Circuit have been blistering, with one of the court’s judges even claiming that his court is the home of “jackpot justice.” As the largest circuit in the nation, many judges, members of congress, and scholars have argued that the Ninth Circuit's large size leads to inconsistent decision making between the panels of judges assigned to hear individual cases. As a result, myriad bills to split the circuit have been introduced in Congress. Although a Congressional commission concluded that there was no special problem of inconsistency in the Ninth Circuit, there was no quantitative analysis to back that finding. Using a novel dataset of all civil appeals filed in the Second and Ninth Circuits from 1995 to 2013, we undertake the first quantitative analysis of the issue. Our analysis proceeds in two main steps: causal identification and estimation.

First, we demonstrate that a sizable number of cases are not randomly assigned to panels in these two circuits, but a simple set of data cleaning steps based on qualitative knowledge of court procedures recovers randomization. This is important because in the absence of random assignment (or some other way to guarantee unbiased effects), our quantitative measure of decision making inconsistency across panels of judges could be a byproduct of differences in cases those panels hear, rather than actual differences in the way the panels come to their decisions.

Second, contrary to the commissions' conclusions, we find evidence that inconsistency is indeed a problem in the Ninth Circuit. Along multiple dimensions, the Ninth Circuit is substantially more inconsistent than the Second Circuit. For example, using machine learning methods, we show that the outcome of at least 30% of cases heard in the Ninth Circuit could have been different had they been assigned to a different panel, as compared to only 6% for the Second Circuit.

Authors

• Ryan Hubert, London School of Economics and Political Science

• Ryan Copus, University of Missouri - Kansas City School of Law