A competing risks model of supreme court vacancies, 1789-1992

We hypothesize that Supreme Court justices will consider the likely ideological disposition of their successor in their decision to retire or remain on the Court. Furthermore, because a justice's decision to remain on the Court places him or her "at risk" of dying in office, it is necessary to consider a model of both voluntary and involuntary vacancies. Our study examines three broad classes of factors influential to Supreme Court vacancies: personal considerations, institutional context, and political influences. We assess the factors that affect the probability of a vacancy on the U.S. Supreme Court due to mortality and retirement at the individual level from 1789 to 1992, using a competing risk duration model and incorporating time-varying covariates. We find significant differences in the hazards of vacancy due to these two causes, and a number of factors are shown to influence the probability of a vacancy, including a general propensity to retire near the beginning of presidents' second terms. However, we find little evidence of the influence of political factors in either retirement-or death-related vacancies, suggesting that justices who retire do not generally do so for expressly political reasons and those who die in office rarely do so as a result of "holding out" for a like-minded replacement.