Circuit effects: How the norm of federal judicial experience biases the supreme court

University of Pennsylvania Law Review

Volumn 157, Issue 3, 2009, Pages 833-880

  Epstein, Lee ^a   Martin, Andrew D ^b   Quinn, Kevin M ^c   Segal, J A ^d  

^a NORTHWESTERN UNIVERSITY   (United States)

^b WASHINGTON UNIVERSITY SCHOOL OF LAW   (United States)

^c HARVARD UNIVERSITY   (United States)

^d STONY BROOK UNIVERSITY   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 64549110895     PISSN: 00419907     EISSN: None     Source Type: Journal    
DOI: None     Document Type: Review

References (194)

1. See, e.g., Michael J. Gerhardt, Why the Catholic Majority on the Supreme Court May Be Unconstitutional, 4 U. ST. THOMAS L.J. 173, 174 (2006) (arguing that Justices with ideological or religious commitments devalue the importance of precedent in judicial decision making);
2. Jake Tapper & Brooke Runnette, Alito Would Create Catholic Majority on Top Court, ABC NEWS, Jan. 24, 2006, http://abcnews.go.com/Nightline/ SupremeCourt/story?id=1536354 (contrasting the historical and current public reactions to Catholics in power);
3. Robin Toner, The Supreme Court's Catholic Majority, N.Y. TIMES, Apr. 25, 2007, http://www.nytimes.com/2007/04/25/us/politics/26web-toner. html?-r=l&coref=slogin (discussing the complicated role that Catholicism has played in American politics).
4. Detailed data on the Justices religious backgrounds is available in LEE EPSTEIN ET AL., THE SUPREME COURT COMPENDIUM 280-90 (4th ed. 2007).
5. Seven of the nine Justices are Republicans (all but Justices Stephen Breyer and Ruth Bader Ginsburg). Only two of the nine (Justices Kennedy and Stevens) worked outside of the Northeast at the time of their appointment - a regional imbalance that would have been unthinkable in die Court's early years. See, e.g., Akhil Reed Amar, Marbury, Section 13, and the Original Jurisdiction of the Supreme Court, 56 U. CHI. L. REV. 443, 469 (1989) (arguing that "[g]eography preoccupied the founding generation" and influenced its decisions regarding Supreme Court);
6. Orrin G. Hatch, Save the Court from What?, 99 HARV. L. REV. 1347, 1352 (1986) (book review) (noting that seats on the Court were "allocated according to geographical considerations"). Finally, all but Justice Stevens attended law school at Harvard, Yale, or Columbia - an unprecedented nod to the Ivy League.
7. See Richard Cohen, Ivy-Covered Court, WASH. POST, Nov. 15, 2005, at A21 ("You might think that the lock the Ivy League has on the Supreme Court is long-standing. Not so. This is a radier new phenomenon ...."). Information on the Justices' backgrounds, including their party affiliations, regional ties, and education, is available in EPSTEIN ETAL., supra note 2, at 271, 280, 291, 387-88.
8. See EPSTEIN ET AL., supra note 2, at 352-53 (indicating the Justices' positions at the time of nomination). The Chief Justice and Justices Ginsburg, Scalia, and Thomas served on the D.C Circuit; Justices Breyer and Souter on the First; Justice Alito on the Third; Justice Stevens on the Seventh; and Justice Kennedy on the Ninth.
9. Mareia Coyle, A Man Comfortable in "the Box, "NAT'L L.J., July 25, 2005, at 1, 19 (quoting Sheldon Goldman).
10. SHELDON GOLDMAN, PICKING FEDERAL JUDGES: LOWER COURT SELECTION FROM ROOSEVELT THROUGH REAGAN 115 (1997).
11. For data supporting this point, see infra Part I.
12. See Anita F. Hill, Why Harriet Miers Mattered, Ms., Winter 2006, at 19, 19 ("It's certainly possible to criticize Miers' qualifications for the Supreme Court without resorting to sexism ....");
13. Robin Toner et al., Steady Erosion in Support Undercut Nomination, N.Y. TIMES, Oct. 28, 2005, at A16 (noting questions raised by senators about Miers's "constitutional mastery");
14. Patrick J. Buchanan, Miers' Qualifications Are Non-Existent, Oct. 3, 2005, http://www.humanevents.com/article.php?id=9444 ("[H]er qualifications for the Supreme Court are non-existent. She is not a brilliant jurist, indeed, has never been a judge."). On the other hand, many commentators allege that conservatives questioned her credentials because they were unsure of her ideological commitments.
15. See, e.g., Kevin P. Martin, Miers's Qualifications, BOSTON GLOBE, Oct. 7, 2005, at Al9 ("Why... rush to dismiss Miers as a mere crony? Mostly it is because conservatives have long had a dream list of nominees to the court....");
16. Toner et al., supra (indicating Senator Sam Brownback's view that "social conservatives were simply not inclined to go on faith that Ms. Miers was a reliable conservative");
17. Emily Bazelon, Let-Down Lady, SLATE, Oct. 3, 2005, http://www.slate.com/id/2127361/ ("The real fear on the right, of course, is that Miers will turn out to be another Justice David Souter, a respecter of precedent who lets her colleagues pull her to the center and then to the left.")
18. Tuan Samahon, The Judicial Vesting Option: Opting Out of Nomination and Advice and Consent, 67 OHIO ST. L.J. 783, 815 (2006).
19. For recent reviews of some of many arguments both for and against the practice of elevating circuit court judges to the Supreme Court, see Vicki C. Jackson, Packages of Judicial Independence: The Selection and Tenure of Article III Judges, 95 GEO. L.J. 965, 983-84 (2007), who argues that looking to appellate courts for nominees provides the "opportunity to evaluate judicial temperament and craftsmanship through the nominee's past judicial experience," but "also create [s] undesirable incentives for decisions made with an eye to advancement through necessarily political confirmation processes."
20. See also Lee Epstein et al, The Norm of Prior Judicial Experience and Its Consequences for Career Diversity on the U.S. Supreme Court, 91 CAL, L. REV. 903, 908 (2003) ("We argue that There now exists a norm of prior judicial experience that induces a highly problematic level of career homogeneity on the Court.");
21. Terri L. Peretti, Where Have All the Politicians Gone? Recruiting for the Modern Supreme Court, 91 JUDICATURE 112, 112 (2007) (discussing the change from a "statesmanlike" Court to a judicial one).
22. See Epstein et al., supra note 10, at 906 (calling the need for prior judicial experience a "norm");
23. Peretti, supra note 10, at 117 (agreeing that prior judicial experience is a norm, though writing that it is not "inviolable or universal");
24. see also Joel B. Grossman, Paths to the Bench: Selecting Supreme Court Justices in a "Juristocratic" World, in INSTITUTIONS OF AMERICAN DEMOCRACY: THE JUDICIAL BRANCH 142, 162 (Kermit L. Hall & Kevin T. McGuire eds., 2005) (deeming prior judicial experience "nearly a de facto qualification").
25. We describe and analyze these claims infra Part II.
26. When he served as Chief Justice, William Rehnquist argued that professionalization of the bench may lead to a decline in its independence and the respect it has always been afforded. He bemoaned the fact that while at one time his Court housed the likes of Louis Brandeis, John Harlan, and Byron White-in other words, Justices "drawn from a wide diversity of professional backgrounds"-those days are long gone. See WILLIAM H. REHNQUIST, 2001 YEAR-END REPORT ON THE FEDERAL JUDICIARY, http://www.supremecourtus.gov/publicinfo/ year-end/2001year-endreport.html (last visited Jan. 15, 2009). Others have asserted that the Court's "steady homogenization" has caused it to become "more reluctant" to hear and decide cases.
27. See, e.g., Stuart Taylor, Jr., Comment, Remote Control, ATLANTIC MONTHLY, Sept. 2005, at 37, 37-38.
28. See infra Part III (discussing data supporting the conclusion that appellate court judges promoted from below are more likely to affirm lower decisions). It is worm noting that we know of no other empirical work on the Supreme Court along these lines and seek to fill this gap.
29. See supra note 11 and accompanying text.
30. For example, in cases coming to the Supreme Court from the U.S. Court of Appeals for the First Circuit, the predicted probability of Justice Stephen Breyer - a former judge on that court - casting a vote to affirm is 0.69; for cases coming from all other circuits, that figure is 0.29. For more details on our analysis of Justice Breyer, alone with all other Justices serving since 1953, see infra Part III.
31. For the details on how we computed these figures, see the conclusion of this Article.
32. See supra note 11 and accompanying text.
33. See, e.g., Peretti, supra note 10, at 112 (noting the shift from appointing statesmen to the Supreme Court to selecting jurists) ;
34. John R. Schmidhauser, The Justices of the Supreme Court: A Collective Portrait, 3 MIDWEST J. POL. SCI. 1 (1959) (examining an earlier shift in criteria for selecting Supreme Court Justices from prior judicial experience to personal qualifications and character).
35. See Epstein et al., supra note 10, at 909-17 (presenting empirical data in support of the claim that selection of Supreme Court Justice nominees from the appellate judge pool is now the norm).
36. EPSTEIN ET AI., supra note 2, at tbl.4-12; Judiciary Act of 1869, ch. 22, § 2, 16 Stat. 44 (1869). Not until 1891 did Congress establish the U.S. Circuit Courts of Appeals in the Judiciary Act of 1891, ch. 517, 26 Stat. 826 (1891).
37. William B. Woods was the first U.S. circuit court judge nominated to the Supreme Court, in 1880.
38. 2= 23.39 (p ≤ 0.05).
39. See, e.g., Epstein et al., supra note 10, at 909 (claiming that the shift began as early as 1959);
40. Peretti, supra note 10, at 114 ("Most commentators point to the 1950s as the origin of this new norm."). We adapt some of the material in this paragraph from the former.
41. 347 U.S. 483(1954).
42. Justices Black, Frankfurter, and Warren were all members of the Court that produced Brown. Prior to their nomination to the Supreme Court, Black was a U.S. Senator, and Warren was Governor of California. Frankfurter was a professor at Harvard Law School.
43. Schmidhauser, supra note 19, at 41.
44. For a review of the proposals, see Robert J. Steamer, Statesmanship or Craftsmanship: Current Conflict over the Supreme Court, 11 W. POL. Q. 265, 270-71 (1958).
45. Peretti, supra note 10, at 117, writes that even today "Congress regularly considers requiring Supreme Court justices to have five or ten years of previous judicial experience."
46. See GOLDMAN, supra note 6, at 115 (discussing Eisenhower's selection of Justices),
47. See generally DAVID ALISTAIR YALOF, PURSUIT OF JUSTICES: PRESIDENTIAL POLITICS AND THE SELECTION OF SUPREME COURT NOMINEES (1999) (analyzing the Justiceselection process as it becomes increasingly politicized).
48. See GOLDMAN, supra note 6, at 115 (noting that Eisenhower's criteria for choosing Supreme Court Justices included an upper age limit, common sense, and lack of extreme viewpoint), Indeed, Eisenhower apparently went so far as to make appellate court service a near prerequisite for service on the Supreme Court. In his diary, Eisenhower recounted a conversation he had with Attorney General Brownwell about appointing Brownwell to the Supreme Court: I told Brownwell that if he had any ambitions to go on the Court, that we should appoint him immediately to the vacancy now existing on the Appellate Court in New York and then when and if another vacancy occurred on the Supreme Court, I could appoint him to it.
49. Id. (citation and brackets omitted). Brownwell turned down Eisenhower's offer.
50. YALOF, supra note 28, at 170 (internal quotation marks omitted).
51. Mark Murray, Hillary on the Supreme Court ?, MSNBC, May 23, 2008, http:// firstread.msnbc.msn.com/archive/2008/05/2S/1057014.aspx (quoting David Yalof).
52. See, e.g., Samahon, supra note 9, at 816 ("Federal appellate judges have previously survived ABA and FBI scrutiny during a prior confirmation.");
53. David A. Yalof, Dress Rehearsal Politics and the Case of Earmarked Judicial Nominees, 26 CARDOZO L. REV. 691 (2005) (discussing President George W. Bush's conservative nominees and the difficulty of confirming them).
54. See Lee Epstein et al, The Changing Dynamics of Senate Voting on Supreme Court Nominees, 68 J. POL. 296, 305 (2006) ("[W] hile ideological distance may 'matter' more than ever, professional merit continues to exert an important influence on senators' votes.").
55. Stewart received a rating of exceptionally well-qualified by a ten-to-one vote; Carswell, a unanimous rating of qualified; Rehnquist, a rating of qualified, with nine members voting that he was well qualified and three not opposed; Bork received ten votes of well-qualified, one not opposed, and four not qualified; and Thomas received a rating of qualified by a divided vote (twelve voted qualified, two not qualified, and one recusal). EPSTEIN ET AL., supra note 2, at 389-90.
56. The difference is not statistically significant (p≤ 0.05).
57. Jeffrey A. Segal & Albert D. Cover, Ideological Values and the Votes of U.S. Supreme Court justices, 83 AM. POL. SCI. REV. 557, 559 (1989);
58. see also Charles M. Cameron et al., Senate Voting on Supreme Court Nominees; A Neoinstitutional Model, 84 AM. POL. SCI. REV. 525 (1990) (using a statistical model to estimate Senate confirmation votes based on electorally attractive positions). The updated scores are available on Segal's website.
59. See Perceived Qualifications and Ideology of Supreme Court Nominees, 1937-2005, http://www.sunysb.edu/polsci/jsegal/qualtable.pdf (last visited Jan. 15, 2009).
60. More specifically, Segal and Cover identified every editorial in four leading newspapers that offered an opinion on a candidate's qualifications. With the editorials in hand, Segal and Cover evaluated their content on the basis of claims about the nominee's acceptability from a professional standpoint. For example, the following, which appeared in the New York Times, would be evaluated as a negative statement about Clarence Thomas's credentials: "Believe him or not, nothing in this bizarre episode enhances Judge Thomas's qualifications, which were slim to start.... If Judge Thomas were a brilliant jurist, a Holmes or a Brandeis, the gamble might be justified. But Clarence Thomas offers no such brilliance ...." Editorial, Against Clarence Thomas: Even "Don't Know" Calls for a "No" Vote, N.Y. TIMES, Oct. 15, 1991, at A24. On the other hand, this sentence,
61. also appearing in the generally liberal New York Times, would be counted as a positive claim about Antonin Scalia's and William Rehnquist's qualifications: "Even liberal critics acknowledge the impressive legal credentials of the Supreme Court nominees. Justice Rehnquist was first in his Stanford Law School class;
62. Judge Scalia was a Harvard Law Review editor; both have written scholarly articles and learned, if combative, judicial opinions." Editorial, Presidential Insults: On Manion, N.Y. TIMES, June 25, 1986, at A26. After analyzing all of the editorials, Segal and Cover created a scale of the lack of qualifications for each nominee that ranges from 0 (most qualified) to 1 (least qualified).
63. Segal & Cover, supra note 36, at 562 fig. 1.
64. Several analysts have accused the ABA of a liberal ideological bias in its rankings. See, e.g., James Lindgren, Examining the American Bar Association's Ratings of Nominees to the U.S. Courts of Appeals for Political Bias, 1989-2000, 17J.L. & POL. 1, 28 (2001) ("For those without prior judicial experience, just having been nominated by Clinton instead of Bush is a stronger positive variable [for earning an ABA 'Well Qualified' rating] than any other credential or than all other credentials put together.").
65. See, e.g., Epstein et al., supra note 33, at 297 (stating that the Segal-Cover calculation "figure[s] prominently into many (if not most) essays" on the topic).
66. We include Thurgood Marshall, who had served as a federal appellate court judge, but was Solicitor General at the time of his appointment. The results are the same if we exclude him or do not treat him as a former circuit court judge.
67. The standard deviations are, respectively, 0.27 and 0.29. (t= 0.2943, p= 0.77).
68. See, e.g., YALOF, supra note 28, at 171.
69. Id.
70. For example, when asked by Senator Kohl about his dissent in Planned Parenthood of Southeastern Pennsylvania v. Casey, 947 F.2d 682, 720-25 (3d Cir. 1991)
71. (Alito, J., concurring in part and dissenting in part), in which he argued that the requirement that a woman notify her husband prior to obtaining an abortion did not impose an undue burden upon a woman, Samuel Alito responded: Trying to apply the undue burden test at that time to the provisions of the Pennsylvania statute that were before the court in Casey was extremely difficult, and I can really remember wrestling with the problem and I took it very seriously and I mentioned that in my opinion and it presented some really difficult issues. Part of the problem was that the law just was not very clear at that time. The undue burden standard had been articulated by Justice O'Connor in several of her own opinions and there were just a few hints in those opinions about what she meant by it. Confirmation Hearing on the Nomination of Samuel A. Alito, Jr. to Be an Associate Justice of the Supreme Court of the United States: Hearing Before the S. Comm. on the Judiciary, 109th Cong. 382-83 (2006) (statement of then-Judge Samuel Alito).
72. YALOF, supra note 28, at 171.
73. See, e.g., id. (suggesting that circuit court judges are preferable to other professions due to their moderate positions on policy issues).
74. The difference is not statistically significant (t = 0.02; p = 0.98). Again, we include Thurgood Marshall who had served as a federal appellate court judge but was Solicitor General at the time of his appointment. The substantive results are the same if we exclude him (t= 0.55; p= 0.59).
75. Yalof, supra note 32, at 697.
76. Samahon, supra note 9, at 816.
77. Again, we include Thurgood Marshall as an appellate court judge. The results are the same if we exclude him.
78. The same holds whether we consider confirmations before or after 1953. Of the 62 nominees prior to 1953, 76.9% of those who lacked service on the circuits were confirmed; that figure for the appellate court judges was 90.0%. After 1953, the Senate confirmed 71.4% (15 out of 21) of the judges and 81.8% (9 out of 11) of the nonjudges. In neither period is the difference statistically significant.
79. Cameron et al., supra note 36, at 528-29;
80. see also Epstein et al., supra note 33, at 298 describing the model's variables).
81. See Cameron et al., supra note 36, at 530-31 ("Overwhelmingly ... it is the interaction of qualifications and ideology that determines the votes of Senators.").
82. This analysis codes Thurgood Marshall as an appellate court judge. The results are substantively and statistically identical if we treat him as a nonfederal judge.
83. The standard errors in this table are very likely too small since the implicit assumption that the confirmation votes are independent of one anoter given the covariates seems unlikely to hold. We simply note this fact raher than attempting a complicated correction because the main purpose of the table is to summarize the observed data and not to test formal hypotheses.
84. We generated this figure via SPost. See generally J, SCOTT LONG & JEREMY FREESE, REGRESSION MODELS FOR CATEGORICAL DEPENDENT VARIABLES USING STATA (2d ed. 2006).
85. The 95% confidence intervals are, respectively, [0.89, 0.93] and [0.96, 0.98].
86. The 95% confidence intervals are, respectively, [0.73, 0.84] and [0.51,0.62].
87. Justice Scalia was confirmed by a vote of 98-0; Justice Kennedy, 97-0;
88. Justice Souter, 90-9;
89. Justice Ginsburg, 96-3;
90. and Justice Breyer, 87-9. EPSTEIN ET AL., supra note 2, at 387-88 tbl.4-15.
91. By contentious nomination, we mean a nomination garnering twenty-five or more nay votes.
92. EPSTEIN ETAL., supra note 2, at 387 tbl.4-15.
93. Id.
94. This argument also occasionally arises in the context of lower court nominations. For example, in considering Jerome A. Holmes's nomination for the Tenth Circuit, Senator Feingold proclaimed, President Bush originally nominated Mr. Holmes to be a Federal district judge in Oklahoma earlier this year. But for some reason Mr. Holmes' nomination was upgraded to the U.S. Court of Appeals for the Tenth Circuit. Placing a nominee with no judicial experience on an appellate court makes it hard to evaluate the nominee's judicial temperament - his capacity to be fair and impartial. 152 CONG. REC. S8149 (daily ed. July 25, 2006) (statement of Sen. Feingold).
95. For a list of proposals and dieir sponsors, see Steamer, supra note 27, at 270-71.
96. 152 CONG. REC. S5200 (daily ed. May 25, 2006) (statement of Sen. Hatch).
97. Id. S347 (daily ed. Jan. 31, 2006) (statement of Sen. Reid);
98. see also supra note 8 (citing media references to Miers's lack of experience),
99. When he nominated Samuel Alito, the President emphasized Alito's prior judicial experience. See Remarks Announcing the Nomination of Samuel A. Alito, Jr., to Be an Associate Justice of the Supreme Court of the United States, 44 WEEKLY COMP. PRES. DOC 1625 (Oct. 31,2005).
100. 152 CONG. REC. S92 (daily ed, Jan. 25,2006) (statement of Sen. Allard).
101. Id. S197 (daily ed. Jan. 26, 2006) (statement of Sen. Dole).
102. David E. Rosenbaum, If Approved, a First-Time Judge, Yes, but Hardly the First in Court's History, N.Y. TIMES, Oct. 4, 2005, at A24.
103. Id.
104. STEPHEN L. CARTER, THE CONFIRMATION MESS: CLEANING UP THE FEDERAL APPOINTMENT PROCESS 161 (1994).
105. John R. Schmidhauser, Stare Decisis, Dissent, and the Background of the Justices of the Supreme Court of the United States, 14 U. TORONTO L.J. 194, 204 (1962).
106. 102 CONG. REG 7277 (1956) (statement of Sen. Stennis).
107. Walter P. Armstrong, Mr. Justice Douglas on Stare Decisis: A Condensation of the Eighth Cardozo Lecture, 35 A.B.A.J. 541, 543 (1949).
108. William G. Ross, The Ratings Game: Factors that Influence Judicial Reputation, 79 MARQ. L. REV. 401, 420 (1996).
109. Schmidhauser, supra note 73, at 202.
110. Id.
111. We estimated a logistic regression in which the outcome variable was whether the Supreme Court overturned one of its own precedents and the number of former federal appellate court judges was entered as a covariate. The coefficient on the number of former judges was not statistically significant (p = 0.83). As a check on the analysis, we also considered whether the Court was less likely to overturn precedent as the number of all former judges (state and federal) increased. Again, the coefficient was not statistically significant (p = 0.86). Data are from Harold J. Spaeth, Documentation, The Original United States Supreme Court Judicial Database: 1953-2007 Terms, http://www.cas.sc.edu/poli/juri/sctdata.htm (select "Documentation" file format under The Original U.S. Supreme Court Judicial Database (nickname: ALLCOURT) header, with analu = 0 and dec-type = 1, 6, or 7) (last visited Jan. 15, 2009).
112. For this analysis we considered only previously decided cases that the Court overruled. We coded the overruled precedent as liberal or conservative based on Spaeth's "dir" variable (indicating whether a decision was liberal or conservative), supra note 79. We then estimated a logistic regression of the number of former federal appellate court judges on whether the Court overturned a liberal (coded 1) or conservative (coded 0) precedent. The coefficient on the number of former federal judges is +0.41 (with a standard error of .10), which tells us that a statistically significant relationship exists between the number of former federal judges and the Court's propensity to overturn liberal precedent. The same result holds if we consider the number of all judges, state and federal.
113. See 393 U.S. 410, 415 (1969) (laying out a strict standard for evaluating a finding of probable cause and rejecting a "totality of the circumstances" approach),
114. overruled by Illinois v. Gates, 462 U.S. 213, 238 (1983).
115. See 448 U.S. 448, 480 (1980) (describing a separate standard of review for Congress's remedial use of racial and emnic criteria),
116. overruled by Adarand Constructors, Inc. v. Pena, 515 U.S. 200, 235 (1995).
117. See 355 U.S. 41, 45-46 (1957) (setting form the rule that "a complaint should not be dismissed for failure to state a claim unless it appears beyond doubt that the plaintiff can prove no set of facts in support of his claim which would entitle him to relief),
118. abrogated by Bell AtI. Corp. v. Twombly, 127 S. Ct. 1955, 1969 (2007).
119. The 95% confidence interval is [0.05, 0.24].
120. The 95% confidence interval is [0.36, 0.57].
121. In our dataset are seventeen former federal circuit court judges: Chief Justices Roberts and Burger; and Justices Alito, Blackmun, Breyer, Burger, Ginsburg, Harlan, Kennedy, Marshall, Minton, Scalia, Souter, Stevens, Stewart, Thomas, and Whittaker. Of these, eight (Roberts, Burger, Alito, Harlan, Minton, Scalia, Thomas, and Whittaker) were to the right of the median Justice for most, if not all, of their service on the Court. We base this claim on Andrew Martin and Kevin Quinn's ideal point estimates. See Andrew D. Martin & Kevin M. Quinn, Martin-Quinn Scores: Measures, http://mqscores.wustl.edu/measures.php (last visited Jan. 15, 2009) [hereinafter Martin-Quinn Scores].
122. See generally Andrew D. Martin & Kevin M. Quinn, Dynamic Ideal Point Estimation via Markov Chain Monte Carlo for the U.S. Supreme Court, 1953-1999, 10 POL. ANALYSIS 134 (2002) (explaining the methodology behind the scores).
123. (n = 45). Data are from JEFFREY A. SEGAL ET AL., THE SUPREME COURT IN THE AMERICAN LEGAL SYSTEM 317 tbl.12-3 (2005).
124. See Cambridge Featured Tides: The Supreme Court in the American Legal System, http://www.cambridge.org/ us/features/0521785081/ (last visited Jan. 15, 2009), for the raw data sets Segal et al. used in creating their Table 12-3.
125. To measure ideology, we use the mean (1986-2000 terms) of the Justices' Martin-Quinn scores.
126. See Martin-Quinn Scores, supra note 86.
127. See id.
128. The coefficient on whether the judge had served on a circuit is -11.82 (with a standard error of 7.29); on the Justices' ideology it is 10.96 (with a standard error of 1.67).
129. We estimated a logit model with the percentage of votes to overturn conservative precedent as the dependent variable, and both the Justices' ideology and whether they had served as a federal circuit court judge as independent variables. The coefficient on ideology is -6.41 (with a standard error of 1.93); on federal circuit court experience, it is 18.7 (8.43). Both are statistically significant (p ≤ 0.05).
130. The 95% confidence intervals are, respectively, [66.76, 86.41] and [44.66, 71.09]. 93 These analyses include only the forty-five cases in which at least one Justice stated his or her desire to overturn precedent and only the fourteen Justices sitting between 1986 and 2000.
131. See Jackson, supra note 10, at 984 (discussing the potential increase in temptation for self-interested decision making, especially when confirmation battles focus on ideology).
132. See Judith Resnik, Judicial Selection and Democratic Theory: Demand, Supply, and Life Tenure, 26 CARDOZO L. REV. 579, 609 (2005) (expressing concern that "the possibility of promotion may undercut the ability of judges to feel unfettered by personal interest when rendering judgments" (footnote omitted)).
133. Jackson, supra note 10, at 983-84.
134. Id. at 984.
135. Yalof, supra note 32, at 697.
136. Segal & Cover, supra note 36, at 559-60 (justifying their methodology on the grounds that newspaper editorials can provide data with (1) ideological content, (2) comparable data, (3) independence from prior votes, and (4) no systematic errors);
137. see also supra note 37, describing the procedures Segal and Cover used to categorize the qualifications of nominees. They used the same approach to characterize the nominees' ideology. 100 For former federal judges the median perceived ideology at the time of nomination is 0.21; for those without prior service on the federal bench it is 0.73. 101 The interquartile range is 0.58 for the former federal judges and 0.48 for the others.
138. Under this definition, 60% of the former federal judges were extremists; the figure for those lacking federal judicial experience is 39.1%.
139. Peretti, supra note 10, at 118.
140. Samahon, supra note 9, at 816;
141. see also Resnik, supra note 95, at 609 ("To the extent we value independent judges, unafraid of encountering popular disapproval and free from needing collegial approval, the possibility of promotion may undercut the ability of judges to feel unfettered by personal interest when rendering judgments." (footnote omitted)).
142. But see Segal & Cover, supra note 36 (showing that data on the perceptions of nominees is no better a predictor of ideology for lower court judges than for nominees coming from other walks of life). This seems to suggest that "[j] udges and scholars perpetuate the myth of merit. The reality, however, is that every appointment is political." DAVID M. O'BRIEN, STORM CENTER: THE SUPREME COURT IN AMERICAN POLITICS 34 (7th ed. 2005).
143. Suzanna Sherry, Logic Without Experience: The Problem of Federal Appellate Courts, 82 NOTRE DAME L. REV. 97, 148-49 (2006) (extending the claim by arguing that the President should appoint more federal appellate court judges and Supreme Court Justices with service on a district court).
144. See, e.g., Stephen Choi & Mitu Gulati, A Tournament of judges ?, 92 CAL. L. REV. 299, 300, 303 (2004) (proposing to subject federal judges to a tournament "where the reward to the winner is elevation to the Supreme Court," using the "norm" of including only federal appellate judges as the "starting point for the tournament").
145. See, e.g., Steven Goldberg, Federal Judges and the Heisman Trophy, 32 FLA. ST. U. L. REV. 1237, 1238-43 (2005) (discussing the underrepresentation of former federal judges amongst consensus "great" Supreme Court Justices).
146. Montgomery N. Kosma, Measuring the Influence of Supreme Court Justices, 27 J. LEGAL STUD. 333 (1998).
147. Id. at 370 (footnote omitted).
148. See Thomas G. Walker & William E. Hulbary, Selection of Capable Justices: Factors to Consider, in THE FIRST ONE HUNDRED JUSTICES 52, 66 (Albert P. Blaustein & Roy M. Mersky eds., 1978) ("[T]he group of justices with the highest performance scores were those that had no judicial service prior to assuming a position on the Court.").
149. Gregory A. Caldeira, In the Mirror of the Justices: Sources of Greatness on the Supreme Court, 10 POL. BEHAV. 247, 258 (1988). Using an updated version of the same data,
150. Robert C Bradley, Who Are the Great Justices and What Criteria Did They Meet ?, in GREAT JUSTICES OF THE U.S. SUPREME COURT 1, 9
151. (William D. Pederson & Norman W. Provizer eds., 2d prtg. 1993), did not need to update the conclusion: "In considering future Court appointees, presidents should heed the message that prior judicial experience is not related, and is possibly an adverse influence, to superior Court performance."
152. Goldberg, supranote 108, at 1240-42.
153. Id. at 1241.
154. Workshop on Empirical Research on the Law (WERL), On Tournaments for Appointing Great Justices to the U.S. Supreme Court, 78 S. CAL. L. REV. 157, 166 (2004).
155. See DAVID R. MAYHEW, DIVIDED WE GOVERN: PARTY CONTROL, LAWMAKING AND INVESTIGATIONS, 1946-2002, at 9 (2d ed. 2005) (measuring the relevance of congressional investigations by whether an investigation was a front page story in the New York Times "on at least twenty days ... during any Congress").
156. See Lee Epstein & Jeffrey A. Segal, Measuring Issue Salience, 44 AM. J. POL. SCI. 66, 72 (2000) (assessing media coverage to determine issue salience).
157. The relationship could also work the omer way: when a superior craftsman writes, the opinion is more important. To assess this, we use a model developed by Lee Epstein, Barry Friedman, and Nancy Staudt, On the Capacity of the Roberts Court to Generate Consequential Precedent, 86 N.C. L. REV. 1299, 1309-13 (2008), designed to predict the circumstances leading to a consequential decision (we exclude cases in which the number of participating former circuit court judges was 0 or 9). Incorporating into this model a variable to indicate whether a former appellate court judge wrote the majority opinion leads to the same conclusion we discuss in the text: majority opinions written by former appellate court judges are significantly less likely to result in important decisions than those written by other Justices. Holding all other variables at their mean, when a former judge writes for the majority opinion, the odds are only 0.12 (with a 95% confidence interval of [0.11, 0.14]) mat it will result in an important decision; the figure increases to 0.18 [0.16, 0.19] for all other Justices.
158. Specifically, we estimated a logistic regression, in which the dependent variable was whether or not the majority opinion was a former federal circuit court judge and the independent variable was whether the case was consequential or not, as measured by Epstein & Segal's approach, supra note 117. The coefficient on the independent variable is -0.47 (with a standard error of 0.07).
159. The 95% confidence intervals are, respectively, [0.35, 0.41] and [0.59, 0.65].
160. Goldberg, supra note 108, at 1241-42 (citations omitted).
161. Harry T. Edwards, The Effects of Coltegiality on Judicial Decision Making, 151 U. PA. L. REV. 1639,1644(2003).
162. Felix Frankfurter, The Supreme Court in the Mirror of Justices, 105 U. PA. L, REV. 781,795(1957).
163. Lee C. Bollinger, the Mind in the Major American Law School, 91 MLCH. L. REV. 2167, 2176(1993).
164. Jackson, supra note 10, at 983,
165. Kevin M. Scott, Understanding Judicial Hierarchy: Reversals and the Behavior of Intermediate Appellate Judges, 40 LAW& SOC'Y REV. 163, 170 (2006).
166. RICHARD A. POSNER, HOW JUDGES THINK 74 (2008).
167. Nick Linder & John Dorsett Niles, The Effect of Trial-Judge Experience on Appellate Decisionmaking Behavior (Apr. 25, 2008) (unpublished manuscript) available at http://ssrn.com/abstract=l 126655 (empirically analyzing the relative likelihood of former state and federal trial judges affirming trial court decisions following promotion to a federal circuit court).
168. See id. at 33 (attributing such tendencies to state judges' familiarity with the trial court's advantage of hearing evidence directly). But Linder and Niles also find that prior experience as a federal trial judge does not have a statistically significant effect on decisions to reverse or affirm lower court rulings.
169. Id.
170. Charles Babington, Sitting Judges to Speak on Alito's Behalf, WASH. POST, Jan. 7, 2006, at A5.
171. Id. The possibility of ethical violations was also raised. Democrats on the Senate Judiciary Committee asserted that the judges, if allowed to testify, would be in violation of Canon 2B of the Code of Conduct for United States Judges. Editors at the New York Times and other commentators echoed the Democrats' concern.
172. See Editorial, Fairness in the Alito Hearings, N.Y. TIMES, Jan. 11, 2006, at A28 ("It is extraordinary for judges to thrust themselves into a controversial Supreme Court nomination in this way, a move that could reasonably be construed as a partisan gesture. The judges will be doing harm to the federal bench.").
173. Babington, supra note 130, at A5.
174. See, e.g., LINDA GREENHOUSE, BECOMING JUSTICE BLACKMUN 23-24 (2005) (noting that when Burger joined the D.C. Circuit, he "threw himself into the ideological combat. His nemesis was the equally combative [Judge David] Bazelon")
175. For an interesting analysis of the propensity of these courts to reverse, see FRANK B. CROSS, DECISION MAKING IN THE U.S. COURTS OF APPEALS (2007), analyzing the statistical effects that an array of factors have on circuit court judges' decisions.
176. Data are from Spaeth, supra note 79.
177. Their reversal rates, respectively, are 0.48 and 0.49.
178. Analysts focusing on the internal calculations of the Justices reach a similar conclusion. They show that Justices who agree with the lower court's decision are better off denying certiorari. If they vote to grant and the Court ultimately reverses, the cost is substantial: the establishment of unfavorable precedent across all the circuits. But if they vote to deny certiorari, "there [is] a small but certain gain." JAN PALMER, THE VINSON COURT ERA 59 (1990). That is, they can be assured that at least one circuit will maintain favorable precedent. Now consider a Justice who disagrees with the lower court's decision. "[T] here [is] a small but certain loss from denying cert. Thus, justices [are] more likely to vote to grant when they want[] to reverse the lower court."
179. Id.
180. The results of the regressions are available at http://epstein.law. northwestern. edu/ research/ circui teffects.htm.
181. The prediction is 0.53 with a 95% confidence interval of [0.39, 0.68].
182. The prediction is 0.78 with a 95% confidence interval of [0.68, 0.89].
183. The predictions are 0.68 [0.64, 0.72] versus 0.50 [0.45, 0.54].
184. Justices Blackmun and Marshall were favorably and significantly biased toward the circuits.
185. The predictions are 0.41 [0.37, 0.45] versus 0.34 [0.28, 0.40].
186. Due to an insufficient number of cases from dieir home circuits, we exclude Justice Alito ( n = 1 ) and Chief Justice Roberts ( n = 1 ).
187. In this analysis, which does not control for the ideological direction of the circuit court's decision, Justice Stevens's behavior toward his former circuit and all others is statistically indistiguishable. Including the control for the ideological direction of the decision, he is positively biased toward his circuit, (p = 0.54).
188. Of the 254 cases coming from the Second Circuit, Justice Marshall voted to affirm in 162, or 64%.
189. The 95% confidence interval is [0.48, 0.91].
190. Babington, supra note 130.
191. Along with an interaction of ideology and lower court direction.
192. See, e.g., BARBARA A. PERRY, A "REPRESENTATIVE" SUPREME COURT ? 5-6 (1991) ("Presidents were particularly scrupulous in maintaining a balance among the country's regions prior to the Civil War.");
193. Amar, supra note 3, at 472 (discussing the importance of geographical concerns for the founders in differentiating between the jurisdiction of the Supreme Court and the lower federal courts) ;
194. Hatch, supra note 3, at 1351-52 (describing how nominations during the Reconstruction Era "were considered political patronage and were allocated according to geographic considerations").