A Theory of Judicial Retirement

For over 50 years, narrative and empirical accounts of judicial retirement have selected variables on a range of unstated assumptions, with discordant results. This paper introduces a formal model in which the justices, the President and the Senate are rational agents who aim to shift the median ideology of the Court as close as possible to their own ideologies. The model shows that the probability of retirement depends on a set of personal, contextual and political variables. It provides a more rigorous theory for the effect of extant variables, reveals erroneous conclusions in the literature, and identifies variables that have not been previously appreciated, such as the ideologies of the non-retiring justices and whether the ideology of the retiring justice is moderate or extreme. This more complete explanation of strategic judicial retirements raises empirically testable predictions to differentiate among the disparate findings of the existing literature.