A new measure for understanding the tenure of U.S. supreme court justices, 1789-2009

Judicature

Volumn 93, Issue 6, 2010, Pages 247-252

  Grofman, Bernard ^a   Kline, Reuben ^a  

^a UNIVERSITY OF CALIFORNIA   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 77955253311     PISSN: 00225800     EISSN: None     Source Type: Journal    
DOI: None     Document Type: Article

References (50)

1. See the essays in Paul D. Carrington and Roger C. Cramton, eds., REFORMING THE COURT: TERM LIMITS FOR SUPREME COURT JUSTICES (Durham: Carolina Press, 2006).
2. See especially Steven G. Calabresi and James Lindgren, Term Limits for the Supreme Court: Life Tenure Reconsidered, pp. 15-98.
3. Carrington and Cramton, Reforming the Supreme Court: An Introduction, pp. 1-14.
4. See also Steven G. Calabresi and James Lindgren, Term Limits for the Supreme Court: Life Tenure Reconsidered, 29 HARV. J. L. & PUB. POL. 669-877 (2006).
5. Ward Farnsworth, The Regulation of Turnover on the Supreme Court, U. ILL. L. REV. 407-454 (2005).
6. Farnsworth, The Ideological Stakes of Eliminating Life Tenure, 29 HARV. J. L. & PUB. POL. 879-889 (2006).
7. Kevin T. McGuire, Are the Justices Serving Too Long? An Assessment of Tenure on the U.S. Supreme Court, 89 JUDICATURE 8-15 (2005).
8. David R. Stras and Ryan W. Scott, Retaining Life Tenure: The Case for a Golden Parachute, 83 WASH. U. L. Q. 1-72 (2006).
9. Stras and Scott, An Empirical Analysis of Life Tenure: A Response to Professors Calabresi & Lindgren, 30 HARV. J. L. & - PUB. POL. 792-830 (2007).
10. Stras and Scott, An Empirical Analysis, supra n. 1.
11. "We believe the post-1970 Supreme Court has become, if anything, too insulated from public opinion, because Justices stay on the Court for an average of twenty-six years and because vacancies have opened up only once every three years or so," Calabresi and Lindgren, supra n. 1, at 823.
12. Calabresi and Lindgren, supra n. 1, at 819-822 observe that the U.S. is essentially the only democracy to give the members of its chief constitutional court life tenure, and that, at the state level, only Rhode Island allows its high court justices to serve for life. They then go on to review alternative mechanisms for constraining judicial terms, such as term limits and mandatory retirement ages.
13. Linda Greenhouse, How Long is Too Long for the Court's Justices? N.Y. Times, January 16, 2005, reports selected statistics that support the conclusions of Calabresi and Lindgren, supra n. 1 and Carrington and Cramton, supra n. 1 (among others) who see the current patterns of tenure and retirement on the Supreme Court as both distinctive and undesirable. She asserts: "The trend is clear. From 1789 to 1970, the average Supreme Court justice served for 15.2 years and retired at 68.5. But since 1970, the average tenure has risen to 25.5 years and the average age at departure to 78.8." As we will see, however, while these numbers are not wrong, they can also be said to be misleading because the impression of great changes will depend to a substantial extent upon the particular periods being compared.
14. There is a relatively small literature directly studying the length of tenure of individual legislators or judges in different institutional settings or in terms of over time variation, e.g., see Rein Taagepera and Robert D. Chapman, A Note on the Ageing of the Politburo, SOVIET STUD. 296-305 (1977).
15. Amihai Glazer and Bernard Grofman, Two plus two plus two equals six: Term lengths of representatives and senators, 12 LEG. STUD. Q. 555-563 (1987).
16. A. Wuffle, Thomas Brunell and William Koetzle, Death, where is thy sting? The Senate as a Ponce (de Leon) scheme, 1 PS: POL. SCI. & POL. 58-59 (1997).
17. But there is a larger literature on the effects produced by differences in tenure rules on the behavior of members of legislative bodies or multijudge courts: See recent work in the U.S. focusing on the consequences of legislative term limits, e.g., Bernard Grofman, ed., LEGISLATIVE TERM LIMITS: PUBLIC CHOICE PERSPECTIVES (Kluwer Academic Publishers, 1996).
18. Bruce E. Cain and Marc A. Levin, Term limits, 2 ANN. REV. POL. SCI. 163-188 (1999).
19. Rick Farmer, John David Rausch Jr. and John C. Green, eds., THE TEST OF TIME: COPING WITH LEGISLATIVE TERM LIMITS (Lexington Books, 2003).
20. Eric Prier and Kevin Wagner, Running Unopposed: Assessing the Impact of Term limits on Competition in Florida and Maine, 37 POL. & POL'Y 101-125 (2009).
21. See Carrington and Cramton, supra n. 1 and Stras and Scottt, An Empirical Analysis, supra n. 1.
22. See Carrington and Cramton, supra n. 1.
23. See id.
24. See e.g., Stras and Scott, An Empirical Analysis, supra n. 1.
25. Gary King, Adding Systematic Explanation to Probabilistic Description: Presidential Appointments to the Supreme Court, 15 AM. POL. RES. 373-386 (1987).
26. We should also note that Justin Crowe and Christopher F. Karpowitz, Where Have You Gone, Sherman Minton? The Decline of the Short-Term Supreme Court Justice, 5 PERSP. ON POL. 425-445 (2007), rather than explaining increased average tenure by a growing number of justices serving increasingly long terms, identify the "decline of the short-term justice" as the theretofore overlooked cause of increased average tenure.
27. We will not be discussing the important debate about whether judges leave the U.S. Supreme Court for political reasons. Some studies have concluded that justices do take strategic (i.e., political) considerations into account when making retirement decisions, e.g., King, supra n. 10.
28. Timothy M. Hagle, Strategic Retirements: A Political Model of Turnover on the United States Supreme Court, 15 POL. BEHAV. 25-48 (1993).
29. Artemus Ward, DECIDING TO LEAVE (Albany: State University of New York Press, 2003)
30. though the majority of the studies have found little evidence to support such claims, e.g., Peverill Squire, Politics and Personal Factors in Retirement from the United States Supreme Court, 10 POL. BEHAV. 180-190 (1988).
31. Saul Brenner, The Myth that Justices Strategically Retire, 36 SOC. SCI. J. 431-440 (1999).
32. Christopher Zorn and Steven R. Van Winkle, A Competing Risks Model of Supreme Court Vacancies, 1789 - 1992, 22 POL. BEHAV. 145-166 (2000).
33. Richard L. Vining, Christopher Zorn, and Susan Navarro Smelcer, Judicial Tenure on the U.S. Supreme Court, 1790 - 1868: Frustration, Resignation, and Expiration on the Bench, 20 STUD. AM. POL. DEV. 198-210 (2006).
34. Albert Yoon, Pensions, Politics, and Judicial Tenure: An Empirical Study of Federal Judges, 1869-2002, 8 AM. L. & ECON. REV. 143-80 (2006).
35. We will also not be discussing Supreme Court confirmation fights or issues of judicial independence, see e.g. Rafael La Porta, Florencio Lopez-de-Silanes, Cristian Pop-Eleches and Andrei Schleifer, Judicial Checks and Balances, J. POL. ECON. 445-470 (2004). The claim is made by some legal scholars that what they characterize as greatly lengthened expected tenure for Supreme Court justices has made confirmation fights in recent decades more bitter, see Calabresi and Lindgren, supra n. 1. However, the truth of the claim about the increased bitterness of confirmation fights does not depend upon the accuracy of claims about changes in Supreme Court longevity.
36. See Calabresi and Lindgren, Term Limits for the Supreme Court, supra n. 1, at 778.
37. See also Id. at 778, chart 1.
38. See Carrington and Cramton, supra n. 1.
39. Id. at 3.
40. See Stras and Scott, Retaining life Tenure, and An Empirical Analysis, supra n. 1, Crowe and Karpowitz, supra n. 10 also take issue with some of the questionable uses of statistics by legal scholars in the context of the dispute over Supreme Court longevity.
41. Stras and Scott, An Empirical Analysis, supra n. 1, at 795.
42. Id. at 797.
43. Id. at 793.
44. Id. at 793.
45. Id. at 803.
46. Id. at 804.
47. If we are going to compare the post 1970 period (which rims through to the retirement of Justice Souter in 2009) with the immediately earlier period, good statistical practice suggests that we should make a comparison between two periods of equal duration so as to comparably smooth the influence of outliers on the means, rather than emphasizing the comparison of 1971-2006 and 1940-1970, a 35-year period and a 30-year period, as Calabresi and Lindgren do in the earlier quote above. This would imply that we compare the duration on the Court of justices who exit during 1932-1970 with those who exit during 1971-2009. Doing so, however, actually largely reinforces the Calabresi and Lindgren claim that there has been a sharp break in the data, since the mean for the 1934-1970 period is 13.3, giving rise to a not quite 13 point difference in years of service between the two periods (13.3 versus 26.1), almost as great as the 14-year difference these authors emphasize for the comparison between 1940-1970 and 1971-2006 that they do make (12.2 versus 26.1). And, while the t-statistic does go down for the equal time periods comparison as compared to that for the comparison reported by Calabresi and Lindgren, both are significant at the p < .0001 level. However if we include Souter's retirement in 2009 (which, at the time of Calabresi and Lindgren's writing would not have been predictable), then the Calabresi and Lindgren claim that the recent era has greater longevity on the Court than the period immediately prior is supported less strongly since adding Justice Souter decreases slightly the mean longevity of recently exiting members, from 26.1 to 25.6 years. But as noted earlier, the problem with these types of analyses is that they restrict their comparisons to the more immediate past and thus omit data from still earlier periods, when Court longevity was higher.
48. It is hard to see why we should pick one particular of the data over another.
49. There are competing reasons why we might prefer one of these two different ways of partitioning justices. If we focus on the longevity of exiting justices, this is the data that we might think would be most immediately visible to presidents and U.S. senators and others who are thinking of the implications of who gets put on the Court. Moreover, it is the decision to exit which is under the control of individual justices. So, if political considerations matter for the decision to leave, we are more likely to detect such differences with data that is organized around date of exit. On the other hand, to make sense of time trends, if we wish to understand the political forces that are operating in a given time period, we might be better off looking at justices grouped according to date of entry, or perhaps grouped into so-called "natural courts", e.g., see Bernard Grofman and Timothy Brazill, Identifying the median justice on the Supreme Court through multidimensional scaling: Analysis of 'natural courts' 1953-1991, 112 PUB. CHOICE 55-79 (2002). However, if there truly was no time trend in the data, then, in equilibrium, it would not matter whether we made comparisons of mean length of service among sets of justices grouped according to when they entered or according to when they exited.
50. As a result of peculiarities like the 1940s, for the entire time period, the correlation between year of entry and age at death is only. 17.