Retirement and Death in Office of U.S. Supreme Court Justices
Demography, Vol. 47, No. 2, pp. 269-298, May 2010
30 Pages Posted: 28 Jun 2012 Last revised: 29 Jun 2012
Date Written: May 31, 2010
Abstract
In this study, we construct demographic models of retirement and death in office of U.S. Supreme Court justices, a group that has gained demographic notice, evaded demographic analysis, and is said to diverge from expected retirement patterns. Models build on prior multi-state labor force status studies and data on justices permit an unusually clear distinction between voluntary and 'induced' retirement. Using data on every justice from 1789 through 2006, with robust, cluster-corrected, discrete time, censored, event history methods, we estimate retirement effects of pension eligibility, age, health, and tenure on the timing of justices’ retirements and deaths in office, resolve decades of debate over the Politicized Departure Hypothesis that justices tend to alter the timing of their retirements for the political benefit or detriment of the incumbent President, reconsider the nature of rationality in retirement decisions, and consider the relevance of organizational conditions as well as personal circumstances to retirement decisions.
Computing robust standard errors with adjustments for clustering by justice, we find that the odds that a justice will retire (or resign or take senior status) in the first two years of the term of a president of the same political party as the president who first appointed him to the Court are about 2.6 times the odds of retiring under a president of the opposing party in the last two years of his presidential term. As hypothesized, roughly the opposite pattern is observed for dying: The odds of death in office odds are about three times higher when the incumbent president is not of the same party as the president who appointed the justice (compared with when the incumbent president is of the same party).
To illustrate that our analyses are not sensitive to different ways of conceptualizing the problem, we also show that if one views death and retirement as competing risks, an approach that we do not favor, the results of a multinomial probit show much the same effects as we show for separate analyses of retirement and death in office.
We find that the odds that justices will retire or take senior status in a year when they are eligible for their pension are more than eight times larger than the odds of retirement in years in which they are not eligible for their pension. Also, contrary to researchers who have hypothesized tenure on the bench as a linear predictor or those who find that increasing tenure always increases the estimated odds of retirement, we find that until judges have been on the bench for 25 or more years, each additional year of tenure makes them less likely to retire, rather than more likely to retire. We also document the secular increases since 1789 in the ages at which United States Supreme Court justices leave the Court, the ages at which they ultimately die, their length of tenure on the Court, and the probability that they will leave office by retirement, rather than by death.
In general, our empirical account supports the rich historical literature that documented individual cases in which justices resigned, retired, or took senior status to perpetuate party influence on the Court. We find that Supreme Court Justices act more or less as one would expect sophisticated people to act regarding their careers. While personal factors, such as pension eligibility, are more important predictors of retirement than political variables, the data are nonetheless consistent with a hypothesis of politicized departure: Whether leaving by retirement, resignation, or death, justices tend to time their departures from the Court based in part on a president’s party and the years remaining in a president’s term in office.
Keywords: Supreme Court, politics, political science, demography, retirement, death, aging, life course, rational choice
JEL Classification: J14, J18, J26
Suggested Citation: Suggested Citation