AGENDA SETTING, ISSUE PRIORITIES, AND ORGANIZATIONAL MAINTENANCE: THE U.S. SUPREME COURT, 1955 TO 1994

Transcription

1 University of Georgia From the SelectedWorks of Jeff L Yates 2005 AGENDA SETTING, ISSUE PRIORITIES, AND ORGANIZATIONAL MAINTENANCE: THE U.S. SUPREME COURT, 1955 TO 1994 Jeff L Yates Andrew B. Whitford, University of Georgia William Gillespie, Kennesaw State University Available at:

2 AGENDA SETTING, ISSUE PRIORITIES, AND ORGANIZATIONAL MAINTENANCE: THE U.S. SUPREME COURT, 1955 TO 1994 Jeff Yates Department of Political Science University of Georgia Athens, Georgia Andrew B. Whitford Department of Public Administration University of Georgia Athens, Georgia William Gillespie Department of Political Science and International Affairs Kennesaw State University Kennesaw, Georgia

3 In recent decades, political science has turned to the study of agenda setting as a central aspect of collective decision-making environments. The content of the public agenda and the issue agendas of political institutions makes significant social change possible. 1 Recent studies suggest that these political institutions are engaged in both competitive relationships as they identify and pursue both active and latent public issues, and more complex cue-taking relationships. 2 For separated powers, the problems of cooperation and competition with one another are entwined with internal collective decision-making dilemmas. In this study, we focus on the tension within a political institution between agenda setting as a mechanism for internal organizational maintenance, and agenda setting as a consequence of that institution s interaction with other branches of government and the general public. Specifically, we examine agenda setting by the United States Supreme Court, and ask the question of why the Court allocates more or less of its valuable agenda space to one policy issue over others. Our study environment is the policy issue composition of the Court s docket: the Court s attention to criminal justice policy issues relative to other issues. Among the most important powers of the Court is the power to apportion its agenda space among policy issues. Pacelle 3 argues that the Court s agenda-building process, its culling of cases and issues from numerous petitions, may represent the most important sequence of decisions the Court makes. But the Court s docket is a finite agenda space on which some issues are provided with a larger proportion of the Court s attention 4. Allocating space to an issue can promote the issue s national visibility and legitimacy as an important public concern. 5 2

4 We model the Court s allocation of this agenda space as a function of internal organizational demands and external political signals. We find that this agenda responds to the issue priorities of the other branches of the federal government and the public. We also find that the Court s internal ideological balance influences issue prioritisation. In contrast, organizational maintenance considerations have no impact on the Court s allocation of its agenda. In the next section of this study we outline the process of Supreme Court agenda setting as an institutional and organizational problem. We then model the Court s agenda as a function of both internal and external factors. Last, we present the model s results and offer conclusions. AGENDA SETTING IN THE U.S. SUPREME COURT The Supreme Court has great discretion in choosing cases and issues. 6 It serves as the arena for setting national legal policy and the cases that it chooses to decide have far reaching policy implications. The Court also sits at the apex of the federal judicial system and in its error correction capacity helps to ensure that the policies set by lower federal courts are broadly consistent with the Court s policy goals. Studies on Court case selection find that justices base granting certiorari on cues involving case characteristics, including the government s involvement as a petitioner, 7 conflicting decisions between lower courts, 8 and the filing of amicus briefs by interest groups. 9 Scholars also argue that certiorari decisions depend on justices preferences and strategic action to further those preferences. 10 Collectively, these explanations are fundamentally limited in that they speak only to the Court s or a justice s decision to hear a case, and not the broader question of which policy issues merit the Court s time and attention. 3

5 The topic of issue agenda setting by the Court in the certiorari process has received less attention. Like the other two branches of government, the Court engages in issue agenda setting both to manage its internal decision-making process and to attain its public policy goals. Studies of the Court rarely capture these dual purposes in agenda setting and instead concentrate on one to the other s exclusion. Caldeira examines a series of bivariate relationships between the Court s issue agenda, its organizational and ideological concerns, and external influences, but does not address the relative importance of these two broad motivations. 11 Pacelle details the Court s changing attention following the Great Depression and New Deal reforms from economic matters to the preferred position of becoming a dominant player in civil liberties. 12 More recently, Flemming, et al. find that in certain circumstances the Court is responsive to the issue agenda priorities of Congress in setting its own agenda; 13 Flemming, et al. recognize that the Court is engaged in competitive and collaborative relations with the other branches of government, so agenda setting is part of a complex process of cuetaking and consolidation of power. However, they do not address or control for the potential effects of organizational concerns or internal preferences on the Court s issue agenda allocation. Our study combines both internal and external motivations to assess the weight of the Court s caseload, its own collective preferences, and the effect of external signals in determining its policy agenda. The Court may be attentive to the agenda priorities of the other branches and the general public while, simultaneously, its internal motivations constrain its issue attention. In doing so, this study bridges the micro-level analysis of the decision to grant certiorari to broader external constraints the other branches and the 4

6 public s attention itself. APPROACH AND RESEARCH DESIGN We assess the weight of these key determinants of the Supreme Court s policy issue agenda by a multivariate statistical analysis of one dependent variable and two sets of independent variables. Our dependent variable (Court Issue Agenda) is the percentage of the Court s workload composed of cases concerning criminal justice issues as defined by Spaeth s Supreme Court Judicial Data Base codebook, 14 for (see Technical Appendix for measurement details) 15. Criminal justice concerns are traditionally a high profile issue. Caldeira notes that the salience of criminal justice issues, coupled with their prevalence on the Court s docket, make this issue area an especially appropriate target of inquiry for analysing agenda setting. 16 We examine the aggregate policy issue agenda for each Court term from 1955 to Our two sets of theoretically based factors are external factors (institutional signalling and public sentiment), and the Court s ideological and organizational concerns. We recognize the difficulty in assessing the role of internal demands by focusing on those internal demands for which explicit aggregatelevel representations are available. 18 Political institutions, such as the Supreme Court, give credence and legitimacy to a policy issue by addressing it in formal proceedings. However, as Stimson, et al. note, Supreme Court justices share policymaking authority with the other branches of government and must be cognizant of possible opposition and adjust accordingly in order to retain the institution s implicit authority and policy-making viability. 19 While members of the Court enjoy life tenure, they are public servants, and have motivations to 5

7 serve the public and provide fair governance. Ultimately, the Court s institutional legitimacy and authority, as well as the justices historical legacies, depend on this imperative. We argue that the Court may be attentive to the preferences of other political elites and the general public to accomplish a perception of fair governance. This premise, that members of the Court might take agenda priority cues from elite political actors and the public, is supported by studies that suggest the justices reference external concerns in their decisions. 20 The premise that members of the Court act in a strategic fashion and take stock of the preferences and priorities of external actors fits well with Kingdon s general conception of government officials as being politically-sensitive agenda setters who are exceedingly cognizant of the political stream of matters that are deemed politically relevant and worth addressing by other elite actors and the public. 21 We believe that the Court may be sensitive to the policy issue attentions of the president, Congress, and the public when formulating its own agenda. Our measure of the presidents policy issue priority (Presidential Agenda) is the percentage of a president s yearly State of the Union Address devoted to the issue of criminal justice policy. 22 Light asserts that these are the primary indicators of presidential policy priorities. 23 A president spending more of his Address on an issue indicates that the president is going public and signalling other branches and the public that it should be carefully considered. 24 Our measure of Congress attention to criminal justice issues (Congress Agenda) is the percentage of congressional hearings devoted to criminal justice issues in a given year. Our measure is derived from Baumgartner and Jones Congressional Hearings Data project, which summarizes information on all congressional hearings over time. 25 Our 6

8 third external variable (Public Opinion) is the percentage of the public believing that crime is an important policy issue, as measured in public opinion surveys. Over the past five decades, scholars have debated whether the Court acts as a majoritarian or countermajoritarian institution. 26 Flemming and Wood make a persuasive case that the justices respond directly to public opinion, finding that justices respond to public opinion across numerous issues and within a relatively short time frame. 27 These external variables are lagged by one year to provide the necessary time for justices to process this information and consider it in forming the Court s agenda. We counterbalance these external influences by accounting for the internal and organizational dynamics of the Court s agenda formation process. First, we account for the possibility that the Court sets its own agenda. In this view, the collective policy preferences of the justices dictate how much of the Court s attention will be spent on criminal justice issues. However, for aggregate-level analysis, this proposition is not a simple one. 28 Studies of justices certiorari voting suggest that justices are interested in advancing their sincere preferences at the certiorari stage, but are strategic and want to be certain that the Court s decision on the merits will turn their way if a case is taken on for review. We suggest that while justices (whether liberal or conservative) may wish to make policy in the high profile issue area of criminal justice, they will be reluctant to take on cases when they are uncertain about the vote on the merits. A moderate Court, on which one or two swing votes might decide a case, provides less certainty for either liberals or conservatives. Hence, we propose that when the Court s composition is more ideologically extreme (or immoderate) there will be more certainty of outcomes, and it will apportion more agenda space to criminal justice issues. 7

9 We offer two measurement strategies here. Our first measure (Immoderate Court) is the absolute value of the mean of the justices ideological scores for a given year, constructed from the scores developed Segal and Cover and Segal, Epstein, Cameron, and Spaeth. 29 We expect that the Court will hear more criminal justice issue cases when its mean ideology is farther from moderate levels. Our second measure (Critical Ideological Balance) is a dichotomous representation of the first. A critical value of ideological immoderation determines the Court s agenda formation; beyond that level, there is reasonable certainty about outcomes on the merits, and the Court will take on more criminal justice cases and make policy. In addition to the Court s internal preferences, we address how organizational pressures affect its allocation of issue space. First, because the Court may respond to increasing demands for criminal justice review, 30 we include the percentage of the petitions for review before the Court from the U.S. Circuit Courts of Appeals involving criminal justice concerns (CJ Adjudication Demand). 31 Second, petitions compete for the Court s agenda space, so an increase in the total number of petitions means more issues competing for that space. As the number of issues vying for the Court s attention increases, criminal justice cases may get pushed off the agenda. Our second organizational variable (General Adjudication Demand) is the total number of petitions before the Court in the year. 32 ESTIMATION AND RESULTS Our model for this multivariate time series analysis is: Court Issue Agenda = β 0 + β 1 Presidential Agenda t-1 + β 2 Congress Agenda t-1 + β 3 Public Opinion t-1 + β 4 Immoderate Court + β 5 CJ Adjudication Demand β 6 General Adjudication Demand + ε 8

10 Four variables fail the Dickey-Fuller test for unit roots (see the Technical Appendix). We model Congress Agenda in lagged first-difference form and difference the Court internal variables: Court Issue Agenda = β 0 + β 1 Presidential Agenda t-1 + β 2 ΔCongress Agenda t-1 + β 3 Public Opinion t-1 + β 4 ΔImmoderate Court + β 5 ΔCJ Adjudication Demand β 6 ΔGeneral Adjudication Demand + ε The first model is estimated by OLS. The Variance Inflation Factors and Cook-Weisberg test score suggest multicollinearity and heteroscedasticity are unlikely. However, the Durbin-Watson test for first-order autocorrelation is indeterminate (d = 1.57), as is the Breusch-Godfrey Lagrange multiplier test (2.33). Because of possible autocorrelation, we then estimate the same model by the Prais-Winsten estimator. 33 We then provide two robustness checks, one of our measurement strategy for Immoderate Court and another for reduced models. All are estimated with robust standard errors. 34 Table 1 presents the results. [Insert Table 1 about here.] While over time the Court s agenda is consistently composed of a relatively large percentage of criminal justice cases, the model reveals substantial evidence that the Supreme Court s agenda responds to the president, Congress, and public opinion. First, the coefficient for Presidential Agenda in the OLS 2 model shows that the Court reacts to the previous period s presidential signal by increasing its criminal justice agenda by 1.27 points for a one-standard deviation shift in presidential statements in the State of the Union. Statements about presidential initiatives and priorities are effective not only for influencing the public, but also shifting the policy agenda of the one national institution the president holds no direct veto power over. 35 Second, the Congress Agenda coefficient 9

11 in OLS 2 is significant and in the expected direction. In this case, a one-standard deviation shift in the change of the number of hearings (an acceleration) in the previous year increases the Court s criminal justice agenda by 1.69 points; the deceleration of Congress agenda in 1979 independently produced a decrease in the Court s criminal justice agenda of 2.37 points. In our view, hearings have an effect that goes beyond traditional internal organizational and lawmaking motivations. The Court is not necessarily ideologically responsive to the other branches, but it reacts to their agendas by adjusting the issue composition of its docket. Finally, the Court adjusts its agenda as public views on the relative importance of criminal justice issues change. A one-standard deviation shift in Public Opinion increases the Court s criminal justice agenda by 1.87 points. We believe that this finding reflects the Court s concern with maintaining institutional legitimacy with the public. An auxiliary model shows that these results are robust to changes in the amount of serious crime in the country (see Technical Appendix Table A3 R3-R7). The public opinion finding supplements and extends the first two effects: the Court s response is a function of its position in a separated power system of national governance. In our initial model, we find that the ideological balance of the Court s personnel (from moderate to less moderate) is not associated with greater attention to criminal justice issues (the coefficient for Immoderate Court is not significantly different from zero). However, in the OLS 2 model, we find that the Court s issue agenda shifts to criminal justice concerns when the ideological balance of the Court s membership reaches a critical level of ideological tilt (one standard deviation from the mean, either liberal or conservative). Essentially, justices act strategically and take on cases in order to 10

12 make ideological inroads when they are reasonably certain how cases will turn on the merits. Perhaps the most surprising result is the lack of support for organizational motivations in the Court s formal agenda-setting mechanism. The Court s agenda allocation does not respond in a technocratic manner in culling a layer of criminal justice cases from the petitions for certiorari. Neither CJ Adjudication Demand nor General Adjudication Demand is statistically significant at conventional significance levels. Instead, the Court acts in a politically-responsive manner: while its selections are limited to what petitioners bring, its choices do not reflect the overall composition of petitioners complaints. 36 Even if petitioners bring a greater criminal justice caseload to the Court, the Court does not bind its own hands by reflecting that increase. These results also tend to reinforce the robustness of our findings concerning the influence of executive and congressional signalling. If expansion of criminal laws (Congress) or enforcement (executive) acted as intervening factors, then any Court workload pressures created by such activities would be accounted for by these variables. The last two models show that our core results are robust when organizational motivations are excluded. DISCUSSION Setting a political institution s issue agenda helps maintain decision making within the organization while providing a basis for broad social change. In a system of separated powers, agenda setting emphasizes institutional competition and cooperation. But agenda setting also sets the scope of conflict, confers legitimacy to the concerns of affected populations, and in practical effect suppresses other concerns. 37 This study assesses how a national political institution balances institutional considerations, its 11

13 collective policy preferences, and external political and popular signals in setting its policy issue agenda. The U.S. Supreme Court is sensitive to the issue priorities of the other branches and references the policy priority concerns of the public in setting its issue agenda. Our finding that organizational maintenance considerations have limited impact provides insight into the Court s internal operations that is perhaps impossible by other methods. Two themes warrant emphasis. First, the Court does not simply cull cases in a technocratic manner. This means that the Court s certiorari decisions, at both the individual petition and aggregate issue agenda levels, are choices of great organizational and political importance. While past studies examine the Court s individual petition-level choices, our aggregate-level analysis instead reveals how the Court sets a broad public policy agenda in its attention to issues in the certiorari process. Second, these findings show how the Court, instead of technocratically processing cases, reacts to the agenda choices of the president and Congress. This extraconstitutional influence provides evidence that the Court is concerned with the implementation of its rulings by definition left to the other institutions and its historical role as an arbiter between separated powers. This finding is reinforced by the Court s reaction to changing public preferences. The Court changes its allocation of its agenda space as public priorities and political attention change. Agenda setting plays a unique role in political institutions, linking organizational effectiveness, distributive politics, and the possibility for social change. At the institutional level, narrow organizational effectiveness takes a backseat to distributive politics. But broad effectiveness an institution s long-term legitimacy works in 12

14 concert with its policy agenda as institutions read the tea leaves of public and political change. While purists may rue the political machinations of the Supreme Court, the Court s reactiveness to the other institutions itself provides a basis for the Court to have real and substantial effects on the public agenda in the future. Aggregate-level reactiveness may not be responsiveness, but it goes some way to insuring the Court s role as a cogent, enduring, and thus valuable mediator in public controversies. 13

15 TECHNICAL APPENDIX Variable Construction Variable Court Issue Agenda Presidential Agenda Congress Agenda Measurement Percentage of the Court s formally decided opinions devoted to criminal justice issues, as defined by Harold Spaeth s United States Supreme Court Judicial Database, (Ann Arbor, MI: Inter-University Consortium for Political and Social Research) (Issues criminal procedure). The Spaeth Database field ISSUE identifies the case s subject matter and is based on the Court s own statements. Criminal procedure (issues ) includes an array of issues involving criminal justice concerns: involuntary confession, plea bargaining, search and seizure, cruel and unusual punishment, right to counsel, discovery, entrapment, the Federal Rules of Criminal Procedure, statutory construction of criminal laws, and double jeopardy issues. Not included are the due process rights of prisoners. Percentage of the annual State of the Union address devoted to criminal justice issues. Similar to Cohen (1995) we counted the number of lines from the president s State of the Union Address devoted to criminal justice issues. We then divided this number by the total number of lines contained in the president s State of the Union Address for a given year to discern the percentage of the speech concerning criminal justice issues. Intercoder reliability for the measure was 95 percent. This variable is logged to account for high skew. Percentage of congressional hearings devoted to criminal justice issues from the Policy Agenda Data Project (1953 to 1994) headed by Frank Baumgartner and Bryan Jones. The data used here were originally collected by Frank R. Baumgartner and Bryan D. Jones, with the support of National Science Foundation grant number SBR , and were distributed through the Center for American Politics and Public Policy at the University of Washington and/or the Department of 14

16 Political Science at Penn State University. Neither NSF nor the original collectors of the data bear any responsibility for the analysis reported here. Public Opinion Immoderate Court Critical Ideological Balance CJ Adjudication Demand General Adjudication Demand The percentage of respondents naming crime as the most important problem using the Gallup Poll s Most Important Problem facing the U.S. question (various years). This variable is logged to account for high skew. The absolute value of the mean of the justices ideological scores for a given term using the Segal et al. (1989; 1995) measure of justice ideology. This is a dichotomous variable with 1 representing that the value of Immoderate Court is one standard deviation above the mean or higher and 0 otherwise. Criminal justice petitions for review to the Supreme Court from the U.S. Circuit Courts of Appeals as a percentage of all petitions for review from the Courts of Appeals to the Supreme Court. Source: Administrative Office of the United States Courts, Annual Report of the Director (various years). The natural log of the total petitions for review before the Court in a given year. Source: Administrative Office of the United States Courts, Annual Report of the Director (various years). Table A1 provides the sample statistics for the variables examined in this study. Figure A1 provides the time series for the dependent variable. [Insert Table A1 about here.] [Insert Figure A1 about here.] Stationarity Given the time series context, we check each variable s stationarity. A time series is nonstationary (or integrated) if the value for a given point in time may be expressed as the sum of all past disturbances. 38 If a series is nonstationary, modeling it in level form 15

17 risks mistaken inferences and spurious findings. 39 A nonstationary variable tends to wander for extended periods of time before changing course (sometimes characterized as a random walk). Our solution to nonstationarity is to first difference the time series. 40 We test each variable using the augmented Dickey-Fuller test. We find that Court Issue Agenda is stationary, but that several independent variables are nonstationary. Because Court Issue Agenda is stationary, we avoid concerns about cointegration. 41 Neither is there evidence for fractional integration. 42 Because analysis by OLS could lead to problematic inferences, we difference four variables to achieve stationarity: Congress Agenda, Immoderate Court, CJ Adjudication Demand, and General Adjudication Demand; Presidential Agenda, Public Opinion, and Critical Ideological Balance are stationary without first differencing. See Table A2 for the tests. [Insert Table A2 about here.] Measures of Serious Crime Robustness Check We also consider controls for the effect of the level of crime on the dependent variable. Consistent measures of crime are difficult to obtain over such a long time series (due to changes in record keeping methods), however we managed to construct two reliable measures for the entire series: homicide per capita and violent crime per capita (Source: Crime in the United States, various years, Federal Bureau of Investigation Crime Reports). Both measures are calculated as first differences because they are nonstationary in levels form. As shown in Table A3, the model s results are robust to the inclusion of either measure of the level of crime. [Insert Table A3 about here] 16

18 Table 1: Regression Results for the U.S. Supreme Court s Issue Agenda, Variable OLS 1 PW 1 OLS 2 PW 2 PW 3a PW 3b Presidential Agenda t ** (0.2814) ΔCongress Agenda t ** (0.5384) Public Opinion t * (0.7690) ΔImmoderate Court (8.1376) Critical Ideological Balance ΔCJ Adjudication Demand ( ) ΔGeneral Adjudication Demand (7.9895) Constant ** (1.0400) ** (0.2954) ** (0.4592) * (0.7833) (8.6570) ( ) (6.4680) ** (1.1340) ** (0.2890) ** (0.4793) ** (0.7688) * (1.8667) ( ) (8.1810) ** (1.1592) * (0.2805) ** (0.4311) ** (0.7874) * (1.9207) ( ) (7.2278) ** (1.2334) * (0.2974) ** (0.4033) ** (0.7454) * (1.8673) ** (1.2435) ** (0.2882) ** (0.4349) * (0.6400) ** (1.1280) ρ R Root MSE Durbin-Watson d N **p < 0.01, *p < 0.05, one-tailed test. Notes: Models estimated using STATA 8.0. Standard errors are in parentheses. Details are available upon request. 17

19 Table A1: Sample Statistics Variable Mean Std. Dev. Minimum Maximum Court Issue Agenda Presidential Agenda t ΔCongress Agenda t Public Opinion t ΔImmoderate Court Critical Ideological Balance ΔCJ Adjudication Demand ΔGeneral Adjudication Demand

20 Table A2: Augmented Dickey-Fuller Unit Root Tests, Variable Test Statistic Solution Court Issue Agenda ** Presidential Agenda t ** Congress Agenda t First difference Public Opinion t ** Immoderate Court First difference Critical Ideological Balance * CJ Adjudication Demand First difference General Adjudication Demand First difference **p < 0.01, *p < 0.05, one-tailed test. Notes: First differencing Congress Agenda, Immoderate Court, CJ Adjudication Demand, and General Adjudication Demand produces stationarity. N =

21

22 Table A3: Robustness Checks Auxiliary Models Variable R1 (PW) R2 (PW) R3 (PW) R4 (PW) R5 (PW) R6 (PW) R7 (PW) Presidential Agenda t * (0.2778) %Crime SUA t * (0.2768) ΔCongress Agenda t ** ** (0.4349) (0.4279) Public Opinion t * (0.8826) Critical Ideological Balance ΔCJ Adjudication Demand ΔGeneral Adjudication Demand * (2.0562) ( ) (8.8449) ** (0.7910) * (1.9406) ( ) (7.2313) * (0.3209) ** (0.4252) ** (0.7877) * (1.8903) ( ) (7.4162) * (0.2929) ** (0.4106) ** (0.7887) * (1.7909) ( ) (7.2343) ΔViolent Crime (0.0244) ΔHomicide (1.2319) Post (1.7228) Post (2.0426) Constant ** (1.3637) ** (1.2193) ** (1.4374) ** (1.3449) * (0.2885) ** (0.4170) ** (0.7982) * (1.9049) ( ) (9.2343) (1.2774) (1.8056) ** (1.2435) * (0.2878) ** (0.4142) * (0.8723) * (1.9200) ( ) (8.5807) (1.2389) (1.7307) (2.0275) ** (1.4172) * (0.2846) ** (0.4116) * (0.8728) * (1.9279) ( ) (8.6516) (1.2386) (1.7430) (2.0233) ** (1.3830) ρ R Root MSE Durbin-Watson d N ** p < 0.01, *p < 0.05, one tailed test. Notes: The parameters of all models estimated using STATA 8.0. Standard errors are in parentheses. Details are available upon request. 21

23 Figure A1: The Court s Criminal Justice Issue Agenda,

24 1 E.g. see John W. Kingdon, Agendas, Alternatives, and Public Policies (Boston: Little, Brown, 1984). 2 See, inter alia, Roy B. Fleming, B. Dan Wood, and John Bohte, Attention to Issues in a System of Separated Powers: The Macrodynamics of American Policy Agendas, Journal of Politics 61 (1999), ; Jeffrey E. Cohen, Presidential Rhetoric and the Public Agenda, American Journal of Political Science 39 (1995), ; Jeffrey E. Cohen, Presidential Responsiveness and Public Policy-Making (Ann Arbor: University of Michigan Press, 1997); Kim Quaile Hill, The Policy Agendas of the President and the Mass Public: A Research Validation and Extension, American Journal of Political Science 42 (1998), ; George C. Edwards and B. Dan Wood, Who Influences Whom? The President, Congress, and the Media, American Political Science Review 87 (1999), Richard L. Pacelle, The Dynamics and Determinants of Agenda Change in the Rehnquist Court in Lee Epstein, ed., Contemplating Courts (Washington, D.C.: Congressional Quarterly Press, 1995), Richard L. Pacelle, The Transformation of the Supreme Court s Agenda: From the New Deal to the Reagan Administration (Boulder, Colorado: Westview Press, 1991); Roy B. Flemming, John Bohte, and B. Dan Wood, One Voice Among Many: The Supreme Court s Influence on Attentiveness to Issues in the United States, , American Journal of Political Science 41 (1997), As Harper and Rosenthal note, When it [the Supreme Court] decides a case involving important matters of public interest, it is making national policy. When it declines to review a case involving important issues, is it not also determining policy?, Fowler Harper and Alan Rosenthal, What the Supreme Court Did Not Do in the 1949 Term University of Pennsylvania Law Review 99 (1950), p. 293). In similar fashion, Justice Brandeis explained that, the most important thing the Court does is not doing as quoted in Gregory A. Caldeira and John R. Wright, Organized Interests and Agenda Setting in the U.S. Supreme Court, American Political Science Review 82 (1988), For an extended discussion of how the U.S. Supreme Court uses its discretion to select cases for review, see H.W. Perry, Jr., Deciding to Decide: Agenda Setting in the United States Supreme Court (Cambridge, Mass.: Harvard University Press, 1991). 7 See Joseph Tanenhaus, Marvin Schick, Matthew Muraskin, and Daniel Rosen, The Supreme Court s Certiorari Jurisdiction: Cue Theory, in Judicial Decisionmaking, Glendon Schubert, ed., (New York: Free Press, 1963), ; S. Sidney Ulmer, The Decision to Grant Certiorari as an Indicator to Decision On the Merits, Polity 4 (1972), ; Marie Provine Case Selection in the United States Supreme Court, (Chicago, Ill.: University of Chicago Press, 1980); Stuart H.Teger and Douglas Kosinski, The Cue Theory of Supreme Court Certiorari Jurisdiction: A Reconsideration, Journal of Politics 42 (1980), See S. Sidney Ulmer, The Supreme Court s Certiorari Decisions: Conflict as a Predictive Variable, American Political Science Review 78 (1984), ; Caldeira and Wright, Organized Interests and Agenda Setting in the U.S. Supreme Court ; and Perry, Deciding to Decide: Agenda Setting in the United States Supreme Court. 9 Caldeira and Wright, Organized Interests and Agenda Setting in the U.S. Supreme Court. 10 Glendon A. Schubert, Quantitative Analysis of Judicial Behavior (Glencoe, Ill.: Free Press, 1959); Saul Brenner, The New Certiorari Game, Journal of Politics 41 (1979), ; Jan Palmer, An Econometric Analysis of the U.S. Supreme Court s Certiorari Decisions, Public Choice 39 (1982), ; Gregory A. Caldeira and John R. Wright, The Discuss List: Agenda Building in the Supreme Court, Law and Society 24 (1990), ; Robert L. Boucher and Jeffrey A. Segal, Supreme Court Justices as Strategic Decision Makers: Aggressive Grants and Defensive Denials on the Vinson Court, Journal of Politics 57 (1995), ; Gregory A. Caldeira, John R. Wright, and Christopher J. W. Zorn, Strategic Voting and Gatekeeping in the Supreme Court, Paper presented at the 1996 annual meeting of the American Political Science Meeting (1996) ; Lee Epstein and Jack Knight, The Choices Justices Make, (Washington, D.C.: Congressional Quarterly Press, 1998). 11 Gregory A. Caldeira, The United States Supreme Court and Criminal Cases, : Alternative Models of Agenda Building, British Journal of Political Science 11 (1981), Richard L. Pacelle, The Transformation of the Supreme Court s Agenda: From the New Deal to the Reagan Administration; and Richard L. Pacelle, The Dynamics and Determinants of Agenda Change in the Rehnquist Court. 13 Roy B. Flemming, B. Dan Wood, and John Bohte, Attention to Issues in a System of Separated Powers: The Macrodynamics of American Policy Agendas. 23

25 14 We examine cases with signed opinions and orally-argued per curiam decisions. These cases are used by the Court to make public stances on legal issues and dominate its attention and efforts. 15 Harold J. Spaeth, United States Supreme Court Judicial Data Base Codebook: Term, 9 th ICPSR Version No. 9422, (Ann Arbor, Mich.: Inter-university Consortium for Political and Social Research, 1999). 16 Gregory A. Caldeira, The United States Supreme Court and Criminal Cases, : Alternative Models of Agenda Building, p See Figure A1 in the Technical Appendix 18 Historically, both micro and macro approaches have had their relative strengths. An example is found in the comparison of aggregate-level studies in Caldeira and Wright s study, Organized Interests and Agenda Setting in the U.S. Supreme Court, which investigates certiorari decisions. They found that the likelihood that the Court grants certiorari in a given case depends on whether interest groups filed amicus briefs and whether the petition involved a real or square conflict in the lower courts. Caldeira and Wright note that amicus and real conflict petitions composed only 7.8 and 5.7 percent, respectively, of the petitions for that term, Caldeira and Wright Organized Interests and Agenda Setting in the U.S. Supreme Court, p While such case factors, which are impractical to obtain at the aggregate level over a long time frame, may influence certiorari decisions on a specific petition, systematic effects for the aggregate issue agenda are unlikely. On balance, we expect that our macro-level approach offers insights not possible at the case level. 19 James A. Stimson, Michael B. MacKuen, and Robert S. Erikson, Dynamic Representation, American Political Science Review 78 (1995), , p. 555). 20 See William N. Eskridge, Jr., Reneging on History? Playing the Court/Congress/President Civil Rights Game California Law Review 79 (1991), ; William Mishler and Reginald S. Sheehan, The Supreme Court as a Countermajoritarian Institution? The Impact of Public Opinion on Supreme Court Decisions, American Political Science Review 87 (1993), ; Stimson, MacKuen, and Erikson, Dynamic Representation ; Michael W. Link, Tracking Public Mood in the Supreme Court: Cross-Time Analysis of Criminal Procedure and Civil Rights Cases, Political Research Quarterly 51 (1995), 61-78; William Mishler and Reginald S. Sheehan, Public Opinion, the Attitudinal Model, and Supreme Court Decision Making: A Micro-Analytic Perspective, Journal of Politics 58 (1996), ; Lee Epstein and Jeffrey A. Segal, Assessing Cross-Institutional Constraints on Supreme Court Agenda Setting, Presented at the annual meeting of the Conference on the Scientific Study of Judicial Politics (1997); Roy B. Flemming and B. Dan Wood, The Public and the Supreme Court: Individual Justice Responsiveness to American Public Moods, American Journal of Political Science 41 (1997), ; Epstein and Knight, The Choices Justices Make; Thomas Hansford and David Damore, Congressional Preferences, Perceptions of Threat, and Supreme Court Decision Making, American Politics Quarterly 28 (2000), ; Lee Epstein, Jack Knight, and Andrew Martin, The Supreme Court as a Strategic National Policymaker, Emory Law Journal 50 (2001), ; Frank B. Cross and Blake J. Nelson, Strategic Institutional Effects on Supreme Court Decisionmaking, Northwestern University Law Review 95 (2001), Kingdon, Agendas, Alternatives, and Public Policies, p See the Technical Appendix for measurement details for all included variables. 23 Paul Light, The President s Agenda: Domestic Policy Choice From Kennedy to Reagan, (Baltimore, Md.: The Johns Hopkins University Press, 1991), p. 160; see also Cohen, Presidential Responsiveness and Public Policy-Making. 24 See Samuel Kernell, Going Public: New Strategies of Presidential Leadership (Washington, D.C.: Congressional Quarterly Press, 1986); Light, The President s Agenda: Domestic Policy Choice From Kennedy to Reagan; Cohen, Presidential Rhetoric and the Public Agenda. Another avenue for presidential policy signaling is through the executive s legal arm, the Solicitor General s office. Studies have shown that the Solicitor General s office is highly successful before the Court in getting cases heard and winning on the merits (e.g., Rebecca Mae Salokar, The Solicitor General: The Politics of Law (Philadelphia, Penn.: Temple University Press, 1992)). Ideally, we would employ a measure of Solicitor General s issue emphasis in its appeals to the Court. However, the office has only produced internal statistics on this since 1984, making analysis for our time series impractical (based on personal communication with the Solicitor General s office). Stephen Meinhold and Steven Shull find that presidential policy preferences (measured by the State of the Union address) are congruent with Solicitor General policy position, Stephen Meinhold and Steven Shull, Policy Congruence Between the President 24

26 and Solicitor General, Political Research Quarterly 51 (1998), Frank R. Baumgartner and Bryan D. Jones, Agendas and Instability in American Politics (Chicago, Ill.: University of Chicago Press, 1993); see also Edwards and Wood, Who Influences Whom? The President, Congress and the Media. 26 Robert Dahl, Decisionmaking in a Democracy: The Supreme Court as a National Policymaker, Journal of Public Law 6 (1957), ; Jonathan D. Casper, The Supreme Court and National Policymaking, American Political Science Review 70 (1976), 50-66; Mishler and Sheehan, The Supreme Court as a Countermajoritarian Institution? The Impact of Public Opinion on Supreme Court Decisions ; Helmut Norpoth and Jeffrey A. Segal, Popular Influence on Supreme Court Decisions, American Political Science Review 88 (1994) ; Link, Tracking Public Mood in the Supreme Court: Cross-Time Analysis of Criminal Procedure and Civil Rights Cases ; Mishler and Sheehan, Public Opinion, the Attitudinal Model, and Supreme Court Decision Making: A Micro-Analytic Perspective. 27 Flemming and Wood, The Public and the Supreme Court: Individual Justice Responsiveness to American Public Moods. 28 As Pacelle notes, change in the Court s ideological composition does not necessarily alter its issue agenda dynamics because it may use its agenda in the next time period to refine or create exceptions to doctrines set by a prior, ideologically-distinct Court, as well as to change doctrine and provide lower court guidance as to such changes, Richard L. Pacelle, The Dynamics and Determinants of Agenda Change in the Rehnquist Court, p For example, if the Warren Court apportioned space to create liberal doctrinal change, then the Burger and Rehnquist Courts might also apportion space to scale back prior liberal advances. 29 See Jeffrey A. Segal and Albert Cover, Ideological Values and the Votes of Supreme Court Justices, American Political Science Review 83 (1989), ; and Jeffrey A. Segal, Lee Epstein, Charles M. Cameron, and Harold J. Spaeth, Ideological Values and the Votes of Supreme Court Justices, Journal of Politics 57 (1995), See Gregory A. Caldeira, The United States Supreme Court and Criminal Cases, : Alternative Models of Agenda Building. 31 Ideally, we would use a measure that captured both federal and state petitions for certiorari to the U.S. Supreme Court. However, while records were available over the time frame of this study for the General Adjudication Demand variable, we could find no records on petitions before the Court that incorporated state cases while denoting case issue type (criminal vs. civil) over the time frame of our study. Personal correspondence with the Administrative Office of the Federal Courts, the Federal Judicial Center, and the U.S. Supreme Court Clerk s Office indicates that such records have not been systematically kept. A Federal Judicial Center report on the Court s docket that was used by Casper provided this information, but only available for a limited time frame ( ). See Casper, The Supreme Court and National Policymaking. 32 Both of our Adjudication Demand variables incorporate in forma pauperis (indigent) petitions. Given that these petitions constitute a substantial portion of both the petitions the Court considers and the cases it ultimately chooses to decide and that they have produced some of the Court s most celebrated cases (e.g. Gideon v. Wainwright, 372 U.S. 335 (1963)), we have chosen to include them. 33 The Prais-Winsten estimator is a Feasible Generalized Least Squares (FGLS) method that allows estimation of ρ, the autocorrelation parameter. See S. Prais and C. Winsten, Trend Estimation and Serial Correlation, (Chicago, Ill.: Cowles Commission Discussion Paper No. 383, 1954). If the sample is small and the value of ρ less than 0.3, OLS is at least as efficient as FGLS. For this comparison, see Potluri Rao and Zvi Griliches, Small Sample Properties of Several Two-Stage Regression Methods in the Context of Auto-Correlated Errors, Journal of the American Statistical Association 64, 325 (1969): The Prais-Winsten models help address our results robustness. 34 Halbert White, A Heteroscedasticity-Consistent Covariance Matrix Estimator and a Direct Test for Heteroscedasticity, Econometrica 48 (1980), ; P.J. Huber, The Behavior of Maximum Likelihood Estimates under Non-standard Conditions, In Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability (Berkeley, Calif.: University of California Press, 1967). 35 To further assess the robustness of this finding, we tested an alternative measure of presidential State of the Union issue emphasis, Crime SUA the percentage of all domestic concerns in the Address (as opposed to both domestic and international concerns) that were on criminal justice concerns (as in the original 25

27 model, this measure is logged to account for high skew). Our findings are robust to this alternative measure specification (see Technical Appendix, Table A3 R2 and R7). 36 Generally, the Court has enjoyed enormous discretion in setting its docket over the time frame of this study. In an auxiliary model we controlled for the effects of the two times when Congress amended the Court s mandatory jurisdiction, (a series of amendments) and 1988 (in both instances curtailing its mandatory jurisdiction and hence, enhancing the discretionary docket See Elder Witt, Congressional Quarterly s Guide to the U.S. Supreme Court, Second Edition, (Washington, D.C.: Congressional Quarterly Press, 1990), p We included dummy variables denoting these two regime shifts, one for 1971 forward and a second for 1988 forward. If these changes affected the Court s issue prioritization, we would expect either that these regime shifts would have independent significant effects in our model or that including these regimes would change the significance of the other coefficients. The auxiliary analysis revealed that the model is robust to the inclusion of these regime markers: neither coefficient is significant and none of the effects presented in the main model change (See Technical Appendix, Table A3 R1, R5, R6 and R7). 37 E.E. Schattschneider, The Semisovereign People (Hinsdale, Ill.: Dyden Press, 1960); Roger Cobb and Charles Elder, Participation in American Politics: The Dynamics of Agenda-Building (Baltimore, Md.: The Johns Hopkins Press, 1971); and Kingdon, Agendas, Alternatives, and Public Policies. 38 Robert H. Durr, What Moves Policy Sentiment?, American Political Science Review 93 (1993), Harold D. Clarke and Marianne C. Stewart, Prospections, Retrospections, and Rationality: The Bankers Model of Presidential Approval Reconsidered, American Journal of Political Science 38 (1994), , p For potential solutions to the stationarity problem, see James Hamilton, Time Series Analysis (Princeton, NJ: Princeton University Press, 1994); William H. Greene, Econometric Analysis (Upper Saddle River, NJ: Prentice-Hall). 41 Clarke and Stewart, Prospections, Retrospections, and Rationality: The Bankers Model of Presidential Approval Reconsidered ; Harold D. Clarke, Karl Ho, and Marianne C. Stewart, Major s Lesser (Not Minor) Effects: Prime Ministerial Approval and Governing Party Support in Britain since 1979, unpublished manuscript, University of North Texas (1998); Robert F. Engle and Clive Granger, Cointegration and Error Correction: Representation, Estimation, and Testing, in Robert F. Engle and Clive Granger, eds., Long-Run Economic Relationships: Readings in Cointegration (New York, NY: Oxford University Press, 1991), James MacKinnon, Critical Values for Cointegration Tests, in Robert F. Engle and Clive Granger, eds., Long-Run Economic Relationships: Readings in Cointegration (New York, NY: Oxford University Press, 1991). 42 J. Geweke and S. Porter-Hudak, The Estimation and Application of Long Memory Time Series Models, Journal of Time Series Analysis 4 (1983),