U.S. Supreme Court Justices and Public Mood

Transcription

1 University of North Carolina at Chapel Hill From the SelectedWorks of Isaac Unah Spring March, 2014 U.S. Supreme Court Justices and Public Mood Isaac Unah, University of North Carolina, Chapel Hill Available at:

2 U.S. Supreme Court Justices and Public Mood Isaac Unah, University of North Carolina, Chapel Hill * Kristen Rosano, Washington University, St. Louis ** K. Dawn Milam, University of North Carolina, Chapel Hill *** * Isaac Unah, Associate Professor of Political Science and former director of the Law and Social Sciences Program at the National Science Foundation. Address all correspondence to: unah@unc.edu. ** Kristen Rosano, Medical Student at Washington University in St. Louis Medical Center *** K. Dawn Milam, JD Candidate at the University of North Carolina School of Law We would like to thank Kevin T. McGuire and Sarah Truel Roberts for comments on an earlier draft of this study.

3 TABLE OF CONTENTS I. Abstract II. Introduction III. THEORETICAL UNDERPINNINGS: PUBLIC MOOD AND SUPREME COURT VOTES A. Strategic Behavior Perspective B. The Attitudinal Change Perspective C. Adopting a Perspective...11 IV. THE CURRENT RESEARCH FRAMEWORK AND OUR APPROACH A. Prior Research on Policy Mood and Judicial Decision-Making B. Controlling for Important Independent Variables Judicial Ideology Amicus Curiae Briefs Polarization and Other Variables V. THE QUANTITATIVE ANALYSIS..18 A. The Method: Logistic Regression Analysis in Three Models B. The Results Initial Findings: Correlation and Trend Analysis Initial Findings: Applying the Three Models The Models Applied to Individual Justices Other Findings: The Influence of Court Composition and Ideological Polarization Public Mood and Court Composition Public Mood and Ideological Polarization on the Court.. 37 VI. THE QUALITATIVE ANALYSIS A. The Method: Examining the Papers of Justice Harry A. Blackmun Committee for Public Education v. Nyquist The Abortion Cases: Planned Parenthood v. Danforth Pennsylvania Central Transportation Company v. New York City Maine v. Thiboutot..49 B. Justice Thurgood Marshall.. 49 VII. CONCLUDING REMARKS

4 ABSTRACT Does public mood influence the decisions of U.S. Supreme Court Justices? Under what conditions might Justices vote against their typical ideological leanings and in favor of public opinion? We employ both quantitative and qualitative methods to address these questions. For the quantitative portion, logistic regression analysis indicates a strong relationship between public mood and Supreme Court Justices votes both in the aggregate and for a number of liberal and conservative Justices individually during the 1946 to 2011 Court terms. Justices respond less strongly to public opinion when the Court is highly polarized and when legal issues to be decided are highly politically salient. For the qualitative portion, we examine Justice Harry Blackmun s personal papers in the Library of Congress for evidence of an apparent switch from an attitudinal posture to a public mood posture. We find four cases in which he made such a switch: Planned Parenthood v. Danforth, Committee for Public Education v. Nyquist, Pennsylvania Central Transportation v. New York City, and Maine v. Thibotout. We also examine closely Justice Thurgood Marshall s normative views on judicial neutrality and the role of public opinion on judging. Our research demonstrates the enduring power of popular influences in the Supreme Court, showing that despite public and scholarly reports about growing ideological intransigence in the Court, the public holds significant sway over the decisions of individual members of the nation s Highest Court. 3

5 Introduction The attitudinal model is a well-established behavioral theory of U.S. Supreme Court decision-making. The theory rejects the idea of a value-free jurisprudence. Instead, it postulates that Justices make decisions based upon their own ideological leanings in light of case facts. 1 In recent years, empirical research demonstrates that in addition to judicial ideology, public opinion also influences Justices decision making. 2 The questions we address in this paper are (1) whether the effect of public sentiment remains significant and robust when examined over a longer time span and (2) whether the effect is strong enough to cause a Justice who is otherwise inclined to vote ideologically a priori to switch and vote against his or her own ideology and in line with public sentiment. In other words, we examine the conditions under which Justices will vote contrary to their own traditional world view or policy preferences and in favor of public sentiments. We argue that in American democracy, public mood (an aggregation of individual policy sentiments) has a statistically significant effect on the voting behavior of individual 1 See Isaac Unah & Ange-Marie Hancock, U.S. Supreme Court Decision Making, Case Salience, and the Attitudinal Model, 28 L. AND POL Y 295 (2006); Tracey E. George & Lee Epstein, On the Nature of Supreme Court Decision Making, AM. POL. SCI. REV. 86, (1992); Jeffrey A. Segal & Harold J. Spaeth, THE SUPREME COURT AND THE ATTITUDINAL MODEL (1992);.GLENDON A. SCHUBERT, THE JUDICIAL MIND (1965); C. HERMAN PRITCHETT, THE ROOSEVELT COURT: A STUDY IN JUDICIAL POLITICS AND VALUES: (1948). 2 See, e.g., Roy B. Flemming and Dan B. Wood, Individual Justice Responsiveness to American Policy Moods, 41 AM. J. POL. SCI. 468, 468 (1997) (concluding that public opinion directly affects decisions by Supreme Court Justices after controlling for the changing composition of the Court, attitudinal inertia of justices, and the strength of judicial ideologies ) (emphasis in original); Michael W. Giles, Bethany Blackstone, & Richard L. Vining, Jr., The Supreme Court in American Democracy: Unraveling the Linkages between Public Opinion and Judicial Decision Making, 70 J. POL. 293, 293 (2008) (finding preliminary evidence that public opinion may provide a mechanism by which the preferences of the Court can be aligned with those of the public ). Christopher J. Casillas, Peter K. Enns, & Patrick C. Wohlfarth, How Public Opinion Constrains the U.S. Supreme Court, 55 AM. J. POL. SCI. 74, 74 (2010) (reaching results that suggest that the influence of public opinion on Supreme Court decisions is real, substantively important, and most pronounced in nonsalient cases ). Kevin T. McGuire and James Stimson, The Least Dangerous Branch Revisited: New Evidence on Supreme Court Responsiveness to Public Preferences, 66 J. POL (2004). 4

6 Justices even though Supreme Court Justices are unelected and therefore unaccountable to the people. 3 The principal motivation for this study is the Court s decision regarding the Affordable Care Act ( ACA ), 4 otherwise known as Obama Care, and in particular the actions of Chief Justice John Roberts. On June 28, 2012, the Supreme Court decided by a 5-4 vote to uphold the ACA in National Federation of Independent Business v. Sebelius, 5 providing constitutional approval for President Barack Obama s signature legislative achievement aimed at expanding affordable healthcare for millions of Americans. 6 Roberts voted with the liberal Justices and authored the majority opinion upholding the healthcare reform act, 7 a position contrary to his traditional ideological leaning. 8 Robert s majority opinion attracted widespread commentary both favorable and critical from the media and in the blogosphere. 9 Some applauded him for 3 For a detailed discussion of the concept of public mood, see JAMES A. STIMSON, PUBLIC OPINION IN AMERICA: MOODS, CYCLES, & SWINGS (2nd ed. 1999). 4 The Patient Protection and Affordable Care Act of 2010, Pub. L. No U.S., 132 S. Ct (2012). 6 Id. 7 Id. 8 Robert s credentials as a conservative jurist are evident in his biography, especially in his career progression. He clerked for Judge Henry J. Friendly (a moderate conservative on the Second Circuit) and then for Chief Justice William Rehnquist, a man of impeccable conservative credentials. See JOHN W. DEAN, THE REHNQUIST CHOICE: THE UNTOLD STORY OF THE NIXON APPOINTMENT THAT REDEFINED THE SUPREME COURT (2002). Roberts also worked in Ronald Reagan s White House and Justice Department. But most importantly, he was appointed by Republican President George H. W. Bush to the D.C. Circuit and then by W. Bush as Chief Justice of the Supreme Court. In his book, The Oath: The Obama White House and the Supreme Court, Jeffrey Toobin described Roberts vote on Obama Care as a shattering disappointment to conservatives, including his four dissenting colleagues. Toobin further described Roberts as a visionary who seeks to use the law to foster long-term constitutional change that reflects the contemporary Republican Party. Toobin, (2012). 9 See, e.g., Tom Scocca, Obama Wins the Battle, Roberts Wins the War, SLATE.COM (June 28, 2012), available at _real_reason_the_chief_justice_upheld_obamacare_.html; Avik Roy, The Inside Story on How Roberts Changes His Supreme Court Vote on Obama care, FORBES.COM (July 1, 2012), available at Noah Feldman, How John Roberts Has Undermined Obama care, BLOOMBERGVIEW.COM (Oct. 8, 2013), available at Jennifer Rubin, Robert s Obama care Decision, Reconsidered, WASHINGTONPOST.COM (Feb. 12, 2014), available at 5

7 objectively analyzing the legal issue without being bound by his typical conservative leaning. 10 That Robert s actions attracted so much attention indicates that voting against the party line in the ideologically polarized Supreme Court 11 has become relatively unusual in the modern era. 12 The questions we address in this paper have important policy ramifications, especially at a time when the balance of conservative and liberal Justices cannot grow any narrower than it currently stands in the Roberts Court. In today s Supreme Court, a single vote switch can significantly alter the trajectory of national policy. 13 Indeed, had Chief Justice Roberts decided to vote his traditional conservative position on the question of whether the ACA should survive, the law would have been struck down, with millions of Americans remaining uninsured. This analysis presents previous research on public opinion s effect on the Court and explores how our work complements that knowledge. We rely on both quantitative and qualitative methodologies to obtain a more complete picture of decision-making in the Supreme Court. In the theory section, we discuss the causal mechanisms that we believe explain the effects we postulate. 10 See Martha Minnow, Affordable Convergence: Reasonable Interpretation and the Affordable Care Act, 126 HARV. L. REV. 117, 118 (2012) (arguing that Justice Roberts stak[ed] out a third position between the camps in a sharply divided Court and, seeing the decision as one of law, not just of politics, thus transcended the polarized political debates... through analytical convergence, not political compromise ). 11 Ideally, we would expect a non-ideological court to be fairly evenly divided in its liberal and conservative decisions. However, this has not been the case in rulings handed down over the past 70 years. See Figure 2 infra. 12 See Tom S. Clark, Measuring Ideological Polarization on the United States Supreme Court, 62 POL. RES. Q. 146, (2009) (applying an axiomatic measure of polarization... to study ideological heterogeneity on the Court to develop a polarization statistic ).One study indicates that part of the reason for the growing ideological polarization is that Supreme Court justices now select law clerks that have served ideologically like-minded judges in the courts of appeals. See Corey Ditslear & Lawrence Baum, Selection of Law Clerks and Polarization in the U.S. Supreme Court, 63 J. OF POL. 3, (2001). 13 One discussion of fluidity appears in Kevin McGuire and Barbara Palmer s article, Issue Fluidity on the U.S. Supreme Court, but their discussion centers on issue fluidity rather than decisional fluidity (i.e., changes in the behavior of the justices within the process of deciding a particular case). AM. POL. SCI. REV. 89, (1995). Hagle and Spaeth better explain the decisional fluidity we describe. See Timothy M. Hagel & Harold J. Spaeth, Voting Fluidity and the Attitudinal Model of Supreme Court Decision Making, 44 W. POL. Q. 1, (1991). 6

8 For the quantitative component, we apply logistic regression analysis using Justices votes as the dependent variable, and public opinion (represented here by policy mood) as the key independent variable. We add several control variables to stabilize our models. These control variables include the ideology of the Justice, involvement of the solicitor general in the case, information about amicus curiae briefs, case salience, direction of the lower court decision, and alteration of precedent. We augment this statistical analysis with a case study of Justice Blackmun s papers to examine his thinking in situations where he voted contrary to his traditional ideological leaning. We do so to discover whether concerns for public sentiment played a role in his decisions. I. Theoretical Underpinnings: Public Mood and Supreme Court Votes In the 1950s, scholars began to challenge the notion that Supreme Court Justices were uninfluenced by popular opinion, and were acting as the sage protectors of insular minorities against potentially tyrannical majoritarian impulses. Robert Dahl brought this skepticism to light in his seminal article in 1957 when he noted that "[t]he view of the Court as protector of the liberties of minorities against the tyranny of majorities is beset with... difficulties that are not so much ideological as matters of fact and logic. 14 Indeed, debate concerning this countermajoritarian difficulty (as Alexander Bickel rephrased it 15 ) came to a head in the early 1990s when the American Political Science Review published several articles in a single issue of the journal dedicated to the controversy over whether elite and popular opinion influences Justices votes and, if so, whether the influence was direct or indirect Robert A. Dahl, Decision-Making in a Democracy: The Supreme Court as National Policy-Maker, 6 J. PUB. L. 279, 283 (1957). 15 ALEXANDER M. BICKEL, THE LEAST DANGEROUS BRANCH: THE SUPREME COURT AT THE BAR OF POLITICS (1962). 16 William Mishler & Reginald S. Sheehan, The Supreme Court as a Countermajoritarian Institution? The Impact of Public Opinion on Supreme Court Decisions, 87 AM. POL. SCI. REV. 87 (1993); Helmut Norpoth & Jeffrey A. Segal, Popular Influence on Supreme Court Decisions, 88 AM. POL. SCI. REV. 711 (1994). William Mishler & Reginald S. 7

9 With significant improvement in survey and statistical methodology, scholars have repeatedly confirmed that the Court is indeed influenced by the sentiments of the people and of their elected representatives. The question is why? Scholars have not agreed on a single satisfactory explanation. Part of the reason is that scholars have long been preoccupied with establishing the nature of the linkage between public opinion and justices decisions, not with testing alternative explanations. Using historical data on appointments, Dahl proposed that it is the frequent replacement of justices (one justice every 22 months on average) that compels justices to remain attuned to public opinion. However, the recent tendency for presidents to appoint younger justices, along with justices serving longer, 17 has added to the stability and lengthening of the intervals of membership change in the Court, making Dahl s personnel replacement thesis somewhat suspect. Two alternative theories have been proposed to explain public opinion s effects on the Supreme Court: The strategic behavior perspective and the attitudinal change perspective. 18 This section briefly presents each perspective and explains why we believe the strategic behavior perspective is the stronger theory. A. Strategic Behavior Perspective Of the two theories, the strategic behavior perspective is arguably the more accurate, direct explanation. This theory maintains that the Court directly and deliberately follows public opinion for fear of losing its legitimacy as an institution. 19 As Michael Giles, Bethany Blackstone and Richard Vining noted, The logic of strategic behavior is that judges conform their voting to public opinion to avoid negative public reactions and to protect the institutional integrity of the Sheehan, Response: Popular Influence on Supreme Court Decisions. American Political Science Review, 88: See Kevin T. McGuire, Are the Justices Serving Too Long?: An Assessment of Tenure on the U.S. Supreme Court, 89 JUDICATURE 1, 8-15 (2005); See also ARTEMUS WARD, DECIDING TO LEAVE: THE POLITICS OF RETIREMENT FROM THE UNITED STATES SUPREME COURT (2003). 18 See Giles et al., supra note 2, at Id. at

10 Court 20 Legitimacy is a belief with significant behavioral consequences for institutions: It induces citizens to defer to legal authorities and encourages them to engage in voluntary cooperation with judicial edicts and the commands of other government officials. 21 The public perception that the Court s decisions are principled forms the basis of the Court s legitimacy. If the Supreme Court constantly voted contrary to the public s beliefs or to the beliefs of other members of government, the Court would risk losing the public s confidence in its decisions and potentially forfeit its legitimacy as an institution. 22 Public opinion thus serves as a direct constraint on justices decisions as was acknowledge in Justice Felix Frankfurter s dissent in Baker v. Carr: The [Supreme] Court s authority possessed of neither the purse nor the sword ultimately rests on sustained public confidence in its moral sanction. 23 Reaffirming that sentiment in Planned Parenthood v. Casey, 24 the Court acknowledged that public perception was essential to the Court s survival and future policy effectiveness. In Casey, the Justices clearly considered public viewpoint in crafting their holding, which reaffirmed a woman s right to terminate her pregnancy in its early stages. 25 Writing for the plurality, Justice O Connor explained that the Court s authority derived from public acceptance of Supreme Court decisions: Americans of each succeeding generation are rightly told [that] the Court cannot buy support for its decisions by spending money and, except to a minor degree, it 20 Id. at See Tom Tyler, Psychological Perspectives on Legitimacy and Legitimation, 57 ANN. REV. OF PSYCHOL., (2006); TOM TYLER, WHY PEOPLE OBEY THE LAW (2006); James Gibson. Institutional Legitimacy, Procedural Justice and Compliance with Supreme Court Decisions: A Question of Causality, 25 LAW AND SOC. REV., (1991). 22 This is quite consistent with Alexander Hamilton s statement in The Federalist Papers #78, that judges have neither Force nor Will, but merely judgment. The public is more likely to have confidence in the Court s judgment if decisions do not deviate too far from the public s median position. Justices decisions will not deviate too far insofar as justices anticipate the reaction of citizens to the Court s decisions. See Kevin T. McGuire & James A. Stimson, The Least Dangerous Branch Revisited: New Evidence on Supreme Court Responsiveness to Public Preferences J. POL. Vol. 66, No. 4 (Nov., 2004), pp U.S. 186, 268 (1962) U.S. 833 (1992). 25 Id. at

11 cannot independently coerce obedience to its decrees. The Court s power lies, rather, in its legitimacy, a product of substance and perception that shows itself in the people s acceptance of the Judiciary as fit to determine what the Nation s law means and to declare what it demands. 26 This acknowledgment of public acceptance as the key to the Court s legitimacy and power makes it clear that the Justices do care about public opinion. Indeed, while Justices O Connor, Kennedy, and Souter s opinion in Casey cautioned that their decision did not reflect acquiescence to social and political pressure, 27 their discussion indicates that at least some Justices may be persuaded by changes in public opinion. B. The Attitudinal Change Perspective This perspective on public opinion and judicial voting is less direct and, in our opinion, weaker than the strategic behavior perspective. It maintains that Justices do not respond directly to public opinion, but are instead influenced by the same social forces that sway public opinion. 28 The underlying logic of the attitudinal change theory is that because Justices are members of the general society, social forces affect their beliefs in the same manner that those forces affect other citizens. In other words, as human decision makers, Justices are influenced by external forces in the same way that the public is influenced by external forces when they make political choices. 29 Before he came to the Supreme Court Benjamin Cardozo gave us a hint at the potential for this kind of influence when he warned judges not to ignore the sweep of perturbing and deflecting 26 Id. at 865 (emphasis added). 27 Id. at See Giles et al., supra note 2, at 295. It should be noted that the attitudinal change theory is different from the attitudinal model. The former emphasizes the potential for external social and political forces to influence Justices votes whereas the latter emphasizes the potential for internal forces (i.e., justices attitudes, values, and preferences) to influence vote choice. 29 See id. One paper that attempted to test for this effect defined social forces as the nation s political currents and changes in the economy, inequality, and the crime rate. See Casillas et al., supra note 2, at 78. For example, as unemployment increases, the public tends to more strongly support an improvement in the jobs situation, and Supreme Court Justices may be more likely to make decisions to alleviate this problem. Casillas et al. found that after controlling for these social forces, the effect of public opinion on the Justices was real. Casillas et al. at

12 forces in their decision making. According to Cardozo, The great tides and currents which engulf the rest of men do not turn aside in their course and pass the judge by. 30 C. Adopting a Perspective In many ways, conducting empirical research is about making choices. Which of the two theories discussed above is more accurate? We have difficulty accepting the attitudinal change perspective because as a matter of pure logic, the postulated relationship is merely associational rather than causal. Moreover, the concept of social forces is too amorphous to pin down, even though we believe the theory does have some explanatory value. For one thing, it is unclear whether the proponents of the attitudinal change explanation equate social forces directly with public opinion as traditionally reflected in opinion polling or if social forces represent something much larger, including the substance of public policy. 31 We believe that the strategic behavior model is more germane because it connotes a direct effect with a powerful logic behind it: Justices rely on the public for its moral sanction. Furthermore, Justices need to listen to public opinion in order to maintain the legitimacy of their institution in the public s mind. II. THE CURRENT RESEARCH FRAMEWORK AND OUR APPROACH A. Prior Research on Policy Mood and Judicial Decision-Making While there is no consensus on the causal mechanism, other researchers agree that there is a relationship between public opinion and judicial voting. Relying on data from 1956 to 1989, Flemming and Wood used Stimson s policy mood index a widely accepted measure of the liberalness of the American public over time and across economic, social and political issue 30 BENJAMIN CARDOZO, THE NATURE OF THE JUDICIAL PROCESS (1921). Chief Justice William Rehnquist adopted a similar position on public opinion. See William Rehnquist, Constitutional Law and Public Opinion, 20 Suffolk U. L. Rev. 751, 768 (1986). 31 For example, Giles et al. argue that the linkage between public opinion and the behavior of justices arises from the force of mutually experienced events and ideas in shaping and reshaping the preferences of both the public and the justices. Giles et al., supra note 2, at

13 areas to quantify public opinion. 32 Specifically, the policy mood index is based on 139 questions tapping the policy views of the American people over several decades involving over 1,700 administrations of the questions. 33 These questions cover a wide range of issues dealing with race, the economy, health care, "big government," taxation, welfare, and defense spending. Flemming and Wood then applied Harold Spaeth s operationalization of liberalism to assign liberalism scores to Justices by term. 34 The results demonstrated that public opinion had a direct effect on the decisions made by individual members of the Supreme Court. 35 This effect holds across issue areas and is not restricted to only a few Justices, but is rather a widespread phenomenon. 36 Although the effect was not large, it was statistically significant. 37 Casillas, Enns, and Wohlfarth, using an instrumental variable to account for social currents that together influence the variation in mood, similarly concluded that public opinion has a direct effect on Supreme Court voting. 38 Relying on data from 1957 to 2000, their study controlled for Justices attitudinal changes resulting from the same forces that affected the public, and still found that public opinion had a statistically significant effect on judicial votes. 39 Giles, Blackstone, and Vining, on the other hand, have a different view about the causal relationship between public opinion and Justices votes. Their study reveals strong correlation 32 Flemming & Wood, supra note 2, at See Stimson, supra note 3, at 34. With multiple indicators, Stimson's measure has greater validity than any single question tracked over time. Furthermore, Stimson's multiple recursive methodology for correcting cross time differences in questions and question wording produces a measure that is consistent with corresponding measures of respondent self-identification. See id. at Flemming & Wood, supra note 2, at 474. Generally speaking, when a Supreme Court decision favors the underdog the party with fewer resources that decision is codified as a liberal decision. Flemming and Wood then performed a pooled time series cross-sectional analysis, with the individual Justice-term as the unit of analysis, to determine the relationship between policy mood and judicial voting. They controlled for the changing composition of the Court, the attitudinal inertia of Justices, and the strength of judicial ideologies to rule out the possibility of a confounding effect. Id. at Id. at Id. 37 Id. 38 See Casillas et al., supra note 2, at Id. at 77 78,

14 between public opinion and Supreme Court voting. 40 However, despite employing virtually the same experimental design as Fleming and Wood, they concluded that there is not a direct relationship, but an indirect one: public opinion does not directly affect Court decisions, and so the causal mechanism is not real, but simply an artifact of social forces influencing the Justices along with the public. 41 Epstein and Martin also showed a relationship between public opinion and Justices votes, but noted the difficulty in discerning between the strategic behavior and the attitudinal change models and so did not offer an opinion. 42 B. Controlling for Important Independent Variables Our hypothesis is that public opinion has a real effect on how individual Supreme Court Justices vote. In order to gauge this effect, we must control for several other variables deemed in the literature as important correlates of judicial decision-making. This section presents the research establishing the key influences of judicial ideology, amicus curiae briefs, and the polarization of the Court. Additional factors controlled for in our study include the role of the Solicitor General; the salience of the case; the direction of the lower court decision; and the opinion s result whether it altered precedent or declared a law unconstitutional. 1. Judicial Ideology Because we are interested in the additional effect of public opinion, the individual ideological leanings of the Justices are an important control factor in our analysis. It is important to understand the role that ideology of the Justices plays in their decision-making in order to control for it. Ideology is the centrifugal force governing the behavior of Justices in the modern Court. Landes and Posner have ranked the ideology of Justices and looked at their voting patterns 40 See Giles, supra note 2, at Id. 42 Lee Epstein & Andrew D. Martin, Does Public Opinion Influence the Supreme Court? Possibly Yes (But We re Not Sure Why), 13 U. PA. CONST. L. 263, (2010). 13

15 to see how the two are correlated in both courts of appeals judges and Supreme Court Justices. 43 The results of their study indicated that ideology does indeed play a role in decision-making, and that this effect is stronger in the Supreme Court than in the courts of appeals. 44 In explaining their findings, Landes and Posner noted that Justices often encounter novel areas of law that do not yet have clear legal answers; thus under such indeterminacies, Justices end up voting in a way that aligns with their own ideological beliefs. 45 In essence, the Justice s beliefs and preferences become a heuristic to the vote. In another study of how ideology affects voting in the Supreme Court, Richards and Kritzer studied jurisprudential regimes and the role they play in Justices decision-making. 46 A jurisprudential regime is a set of legal guidelines based on precedents that the Court has created to guide their future decisions on certain issues. However, Richards and Kritzer s argument is that these regimes are often constructed by the Justices to further their own policy goals. 47 Thus, even though it appears that the Justices are strictly following the law guided by these jurisprudential regimes, these guidelines may in themselves be ideological in nature. Another indication of this phenomenon is that these regimes tend to change with changing composition of the Court and 43 William M. Landes & Richard A. Posner, Rational Judicial Behavior: A Statistical Study, 1 J. LEGAL ANALYSIS 775 (2009). Their dependent variable was the ideological classification of votes and their key independent variable was a measure of the ideology of a Justice based on the party of the president who appointed the jurist. Id. at The other variables in their analysis were the partisan make-up of the Senate at the time the Justice was confirmed, demographic characteristics of the Justice, and the ideological make-up of the other Justices on the Court. Id. at Id. at Id. at 789. Landes and Posner reached additional conclusions from their analysis. First, they demonstrated that the ideologies of Justices are not constant, but change over time. Id. at Second, they found no evidence of the conformity effect, which states that in a small group decision making context, as the number of individuals in the minority gets smaller, people in that minority tend to vote with the majority. Id. They also found no evidence of the group-polarization effect, which states that as people of different ideologies are allowed to have a discussion, everyone tends to develop more extreme views. Id. They did find evidence of the political-polarization effect among Democratic but not Republican appointees. Id. They observed that the fewer the number of judges appointed by Democratic presidents, the more liberally these appointees voted. Id. 46 Mark J. Richards & Herbert M. Kritzer, Jurisprudential Regimes in Supreme Court Decision Making, 96 AM. POL. SCI. REV. 305 (2002). 47 Id. at

16 with the political climate of the country, so in many ways judicial decisions are not completely objective readings of the law Amicus Curiae Briefs Another independent variable on judicial voting analysis is the influence of outside parties who file amicus curiae briefs. Amicus curiae briefs are informational vessels that alert Supreme Court Justices to the perspectives of special interests groups and individuals on matters before the Court. These documents present alternative views of the case and signal to Justices the impact that their decision may have on these groups and on other social elements. Research suggests that groups filing amicus briefs do influence the Justices votes. Kearney and Merrill found that amicus curiae briefs, over the 1946 to1995 period, moderately affected the Court. 49 They further discovered that the most successful filers were the United States Solicitor General, the States, and repeat-player organizations such as the ACLU and AFL-CIO. 50 Paul Collins set out to explain why this is true. 51 He proposed and tested two hypotheses: the affected group hypothesis and the information hypothesis. 52 The affected group hypothesis states that an amicus brief is successful because it signals to the Court how many people or groups will be affected by their decision, a relevant consideration to Justices who do not wish to damage their 48 See, e.g., id. at 315 ( Our jurisprudential regime construct moves beyond the dominant model of Supreme Court decision making, the attitudinal model, which posits that legal considerations make little difference in the way the justices vote. We theorize and observe that both justices policy goals and legal considerations matter in Supreme Court decision making. ). 49 Joseph D. Kearney & Thomas W. Merrill, The Influence of Amicus Curiae Briefs on the Supreme Court, 148 U. PA. L. REV. 743 (2000). To measure this, they first recognized that, in a case with no amicus briefs, petitioners won 60 percent of the time, respondents won 37 percent of the time, and a mixed result occurred 3 percent of the time. Id. at They then analyzed how these percentages changed when briefs were added to one or both sides of the case. Id. at Their results indicated that a few briefs may help a case but that too many become redundant and duplicative, and may even be counterproductive as the law of diminishing returns sets in. Id. at 830. Kearney and Merrill also found that amicus filers supporting respondents had more success than those supporting petitioners. Id. at Id. at See generally Paul M. Collins, Jr., Friends of the Court: Examining the Influence of Amicus Curiae Participation in U.S. Supreme Court Litigation, 38 L. & SOC Y. REV. 807 (2004). 52 See id. at

17 legitimacy by having their decisions rejected or ignored by the public. 53 The information hypothesis states that these briefs provide Justices with additional information that they may not have known otherwise, which could ultimately change their decision. 54 Collins s results showed that amicus participation increases litigation success, albeit moderately, and that this influence is best explained by the information hypothesis. 55 This study, like Kearney and Merrill s, also revealed that the prestige of amicus participants is vital to success in the Supreme Court. 56 Unah and Hancock similarly reached this conclusion after examining the role of the ACLU and NAACP in civil rights cases Polarization and Other Variables A polarized court is one that has significantly conflicting views of the law. We consider the possibility that the degree of polarization on the Court during a particular term determines how responsive the Justices are to public opinion. A straightforward mechanism for assessing polarization in the Supreme Court is the proportion of cases decided by one-vote margin each term. Using cases listed in the Supreme Court Compendium, we have determined that polarization in the Court was extremely low throughout the 1800s and early 1900s (see Figure 1). However, despite significant volatility in the incidence of polarization, it is important to note that polarization in the Court has steadily grown since One researcher, Tom Clark, has measured the ideological polarization of the Supreme Court from approximately 1950 to 2012 via a more complex methodology. 58 Clark s results show that the Supreme Court was particularly 53 See id. at See id. 55 Id. at Id; Kearney & Merrill, supra note 44, at See Unah & Hancock, supra note 1, at Tom S. Clark, Measuring Ideological Polarization on the United States Supreme Court, 62 POL. RES. Q. 146 (2009). Clark used the polarization measures developed by Esteban and Ray and applied them to the Supreme Court. Id. at 146. Two methods of measuring the polarization of a population are (1) to measure the distance between the medians of two parties, or (2) to determine the bimodality or high variance of a population. Both indicate high 16

18 polarized between 1970 and 1990, with a relatively constant and lower level of polarization before and after this range. In addition to controlling for these key variables, our study also controls for the role of the Solicitor General; the number of amicus curiae briefs filed on either side of the case; the salience of the case; the direction of the lower court decision; and the opinion s result whether it altered precedent or declared a law unconstitutional. Figure 1: Data from the Supreme Court Compendium (2009) demonstrates the growth of legal disagreement in the Supreme Court as indicated by cases decided with a one-vote margin. polarization, but they cannot be applied to the Supreme Court because there are not two distinct groups, nor are there enough Justices to perform meaningful statistical analysis. Id. at 148. Esteban and Ray included characteristics such as homogeneity within a group, heterogeneity among groups, and the number of groups to quantitatively measure the polarization. Id. at 148. Clark used the Segal-Cover Judicial Common Space scores and applied this algorithm to measure polarization on the Court. Id. at 149. As Clark explains, the estimate of polarization is an additive representation of the distance from each Justice to each other Justice, weighted by how many Justices fall at any given point. Id. at

19 III. THE QUANTITATIVE ANALYSIS A. The Method: Logistic Regression Analysis in Three Models To examine the role of policy mood on voting of individual Justices, we rely on data from 1946 to 2011, a 65year time period. This time span dwarfs in length the range employed in previous studies. By examining a longer time span, we hope to gain greater confidence in the observed relationship between public mood and Justices votes. Furthermore, we hope to account for inter-temporal variations concerning judicial ideology, patterns of amicus curiae brief filings, polarization in the Court and other factors. Our data are derived from the Supreme Court judicial database compiled by Harold Spaeth, with case citation as the unit of analysis. 59 We then augmented these data with information from a variety of sources, including the Segal-Cover ideology scores, 60 policy mood scores, 61 amicus curiae information, 62 Solicitor General data, 63 and case salience measures. 64 Since we are primarily interested in analyzing individual Justice votes, the structure of the data is such that each case has multiple entries corresponding to the number of Justices sitting in judgment of that case. This resulted in a total of 75,326 entries for the entire period examined. 65 However, the public mood data are only available from 1952 to 2011 and this effectively limited our analysis to the 69,297 votes cast during that period. Table 1 summarizes the variable descriptions, coding, and data sources. Table 1: Description of Variables Used in the Logistic Regression Analysis Variable Meaning Coding Source Role 59 Harold J. Spaeth, United States Supreme Court Judicial Database. ICPSR. 60 See Jeffery A. Segal & Albert D. Cover, Ideological Values and the Votes of U.S. Supreme Justices, 83 AM. POL. SCI. REV. 557 (1989). 61 See Stimson, supra note See Collins, supra note 46; Kearney & Merrill, supra note See Chris Nicholson & Paul M. Collins, Jr., The Solicitor General s Amicus Curiae Strategies in the Supreme Court, 36 AM. POL. RES. 382 (2008). 64 See Lee Epstein & Jeffrey A. Segal, Measuring Issue Salience, 44 AM. J. POL. SCI. 66 (2000). 65 This accounts for instances where a Justice might not participate in the case due to recusals, illness, or other forms of unavailability. 18

20 Vote Direction Vote direction 1=liberal, 0=conservative Spaeth Dependent variable Ideology of Justice Predicted ideology of Justice prior to position on Supreme Court based on newspaper editorials Range from 0 (most conservative) to 1 (most liberal) Segal- Cover Control Policy Mood Liberalness of the United States public based on aggregation of public opinion Range from 0 (most conservative) to 100 (most liberal) Stimson Independent variable Solicitor General as Liberal Party Solicitor General as Conservative Party Solicitor General, Liberal Amicus Curiae Brief Solicitor General, Conservative Amicus Curiae Brief Altered Precedent Declared Unconstitutional Number of Liberal Amicus Curiae Briefs Number of Conservative Amicus Curiae Briefs Whether Solicitor General was a party on the liberal side of the case Whether Solicitor General was a party on the conservative side of the case Whether Solicitor General wrote an amicus curiae brief for the liberal party Whether Solicitor General wrote an amicus curiae brief for the conservative party Whether the case altered a Supreme Court precedent Whether the decision declared a law unconstitutional The number of amicus curiae briefs on the liberal side of the case The number of amicus curiae briefs on the conservative side of the case 0 = no, 1 = yes Collins Control 0 = no, 1 = yes Collins Control 0 = no, 1 = yes Collins Control 0 = no, 1 = yes Collins Control 0 = no, 1 = yes Spaeth Control 0 = no, 1 = yes Spaeth Control Count Collins Control Count Collins Control 19

21 Case Salience Index Whether the case appeared on the front page of the New York Times, indicating salience 0 = no, 1 = yes Epstein- Segal Control Lower Court Decision Whether the lower court represented a liberal decision 0 = no, 1 = yes Spaeth Control We start the analysis be taking an aggregate look at the correlation between public mood and the liberalism of Supreme Court decisions. Correlation tells us the extent to which mood and Supreme Court decisions are associated with each other. Correlation is not causation but causation cannot exist without correlation. Then, we used three models to analyze the data via logistic regression, and each model was applied to both the aggregate dataset and to each Justice separately. Vote direction is the dependent variable throughout and policy mood is the key independent variable of interest. In Table 2, policy mood is not normalized. That is, we are agnostic about whether the time at which policy mood is measured links up nicely with the time the Court s decision was made, so long as both events occurred during the same year. In Model 1, no control variables are used, making it a simple bivariate analysis. Model 2 introduces a more complex analysis because we add three controls variables: ideology of the Justice, solicitor general s liberal amicus curiae briefs, and case salience. Model 3 is the most complex and complete model displayed in Table 2, with eleven control variables employed. We add these variables to address the possibility of confounding influences. In Table 3, public mood is made contemporaneous with the outcome of Supreme Court decisions. 20

22 We are also interested in knowing whether the polarization of the Supreme Court has an effect on how Justices respond to public opinion. Therefore, we performed an analysis using the three models for the period , which Tom Clark designated as a particularly polarized period in the Court s storied history. 66 We then compared the findings to the periods before and after this range. B. The Results 1. Initial Findings: Correlation and Trend Analysis Our data indicate that there is a strong and robust association between public mood and the direction of Supreme Court decisions. A Pearson correlation coefficient {r} of.47, which is also statistically significant with nearly 100% confidence, was returned by the analysis as indicated in Figure 2. This figure shows a definite trend in which Supreme Court liberalism moves in a similar pattern as policy mood. In the early 1950s when U.S. policy mood was moving in a liberal direction, liberalism increased in the Supreme Court as well. Mood started to move in a more conservative direction around 1968, which is about the time that liberalism retreated sharply in the Court as conservative Warren Burger was brought in to lead the Court, following the retirement of Chief Justice Earl Warren. Since the early 1990s, both mood and Supreme Court liberalism have been moving in a downward (i.e., conservative) direction. We now turn to the causal analysis. 66 See Clark, supra note 53, at

23 Figure 2 2. Initial Findings: Applying the Three Models Table 2 shows results from the aggregate analysis, emphasizing the three models described above. Model 1, the simple bivariate analysis, shows that without controlling for any other independent variables, policy mood does indeed have a statistically significant effect on Supreme Court decisions. The model performs moderately well, correctly predicting about 55 percent of the cases. This performance is five percentage points better than chance. 22

24 Table 2. Effect of Public Mood on Supreme Court Decision Making, Variable Model 1 β (S.E.) Policy Mood.047*** (.002) Odds Model 2 Ratio β (S.E.) *** (.002) Odds Model 3 Ratio β (S.E.) *** (0.003) Odds Ratio Judicial Ideology 1.404*** (.029) *** (0.042) Case Salience (New York Times Measure).184*** (.026) (0.038) SG, Liberal Amicus Brief.809*** (.034) SG, Conservative Amicus Brief *** (0.044) *** (0.045) SG as Liberal Party (0.184) SG as Conservative Party (0.189) Alteration of Precedent 0.187** (0.084) Declaration of Unconstitutionality Number of Liberal Amicus Briefs Number of Conservative Amicus Briefs 0.402*** (0.021) 0.092*** (0.009) *** (0.008) Lower Court Decision *** (0.027) Constant *** ***.105 (.121) (.113) (.178) Number of Cases Pseudo R *** p <.01; ** p <.05, (all one-tailed tests) 51, , The odds ratio measures the impact of the independent variable (in this case, policy mood) on the dependent variables (Supreme Court vote in a liberal direction). An odds ratio describes the ratio of two odds: In this case, the odds of a liberal decision divided by the odds of a conservative decision. An odds ratio higher than one means that the Supreme Court is more likely to decide in favor of the higher value of the dependent variable (liberalism). An odds ratio that is 23

25 lower than 1 indicates that the Supreme Court is less likely to favor the liberal view. Finally, an odds ratio of 1 means the Court is neutral on the effect of the independent variable in question. In this first model, the regression produced an odds ratio of 1.048, which means that the odds are approximately 5 percent higher the Supreme Court will issue a liberal decision as policy mood becomes more liberal. In Model 2, we add three important independent variables into the mix and find that policy mood remains statistically significant at the.05 level or better. Overall, this model performs better than Model 1, predicting 61 percent of the cases correctly. Specifically, when we vary the conditions by adding ideology, case salience, and the Solicitor General s liberal amicus briefs, the odds ratio becomes Although this impact is statistically significant, it is slightly lower than in the previous model. This indicates that the effect of public mood on Justices weakens as more independent conditions are introduced into the Court s decisional calculus, suggesting that these other variables may counter-balance policy mood s effect. The Model 2 analysis also indicates that ideology of the Justices has the greatest relative impact on their decisions. The odds are four times higher that the Court will issue a liberal decision when the Justices are ideologically liberal. In addition, when the case is politically salient as indicated by the opinion s publication on the front page of the New York Times the odds are about 20 percent higher the Court s decision will be a liberal decision. Model 3, which employs eleven control variables, produces even more nuanced results. Justices must explicitly or implicitly consider a variety of factors when deciding cases in the U.S. Supreme Court. Though any number or combination of factors could be competing for influence, Justices typically consider the lower court s decision; the arguments in the legal briefs submitted by the parties, their supporters, and by the Solicitor General; the density of these briefs; legal 24