Attacking Judges: How Campaign Advertising Influences State Supreme Court Elections

Stanford University Press

Stanford, California

© 2015 by the Board of Trustees of the Leland Stanford Junior University.

All rights reserved.

No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying and recording, or in any information storage or retrieval system without the prior written permission of Stanford University Press.

Printed in the United States of America on acid-free, archival-quality paper

Library of Congress Cataloging-in-Publication Data

Hall, Melinda Gann.

Attacking judges : how campaign advertising influences state supreme court elections / Melinda Gann Hall.

p. cm. — (Stanford series in law and politics)

Includes bibliographical references and index.

ISBN 978-0-8047-8795-6 (cloth : alk. paper) — ISBN 978-0-8047-9308-7 (pbk. : alk. paper)

1. Judges—United States—Election.   2. Advertising, Political—Law and legislation—United States.   3. Courts of last resort—United States—States.   4. Political questions and judicial power—United States—States.   I. Title.

KF8776.H35 2014

324.973—dc23

2014012610

ISBN 978-0-8047-9309-4 (electronic)

Typeset by Thompson Type in 10/14 Minion

ATTACKING JUDGES

How Campaign Advertising Influences State Supreme Court Elections

Melinda Gann Hall

Stanford Law Books

An Imprint of Stanford University Press

Stanford, California

STANFORD STUDIES IN LAW AND POLITICS

Edited by Keith J. Bybee

To Rose Marie Corley Gann and James Earl Gann Sr.

Contents

Tables and Figures

Preface and Acknowledgments

1. Attacking Judges: Another Dimension of Campaign Negativity in American Politics

2. State Supreme Court Elections in Contemporary Democracy

3. Campaign Advertising in State Supreme Court Elections

4. Attack Advertising and Electoral Support for State Supreme Court Justices

5. Attack Advertising and Citizen Participation in State Supreme Court Elections

6. State Supreme Court Elections Are Different—by Design

Notes

References

Index

Tables and Figures

Tables

Table 2-1. Selection Systems and Terms of Office for State Supreme Courts, 2013

Table 2-2. Electoral Competition for Incumbents in State Supreme Courts, 2000–2010

Table 2-3. Features of State Supreme Court Elections, Pre-White and Post-White

Table 3-1. Televised Campaign Advertising by State in Supreme Court Elections, 2002–2008

Table 4-1. Variable Means, Overall and by Election Type, in Campaign-Based Models of Incumbent Vote Shares

Table 4-2. Televised Campaign Advertising and the Electoral Performance of State Supreme Court Incumbents, 2002–2008

Table 4-3. Televised Campaign Advertising and the Electoral Performance of State Supreme Court Incumbents, 2002–2008, with Interaction Terms for All Ad Types

Table 4-4. Televised Campaign Advertising as Net Effects and the Electoral Performance of State Supreme Court Incumbents, 2002–2008

Table 4-5. Televised Campaign Advertising and the Electoral Performance of State Supreme Court Incumbents, 2002–2008, with Robustness Checks for Third-Party Challengers and Hybrid Election Systems (Michigan and Ohio)

Table 5-1. Variable Means, Overall and by Election Type, in Campaign-Based Models of Ballot Roll-Off

Table 5-2. Televised Campaign Advertising and Ballot Roll-Off in State Supreme Court Elections, 2002–2008, with Controls for Presidential Election Years

Table 5-3. Televised Campaign Advertising and Ballot Roll-Off in State Supreme Court Elections, 2002–2008, with Controls for the Salience of Top-Ballot Races and Election Year

Table 5-4. Televised Campaign Advertising and Ballot Roll-Off in State Supreme Court Elections, 2002–2008, with Interaction Terms for All Ad Types

Table 5-5. Televised Campaign Advertising and Ballot Roll-Off in State Supreme Court Elections, 2002–2008, in Nonpartisan Elections Only

Table 5-6. Televised Campaign Advertising and Ballot Roll-Off in State Supreme Court Elections, 2002–2008, with Robustness Checks for Third-Party Challengers and Hybrid Election Systems (Michigan and Ohio)

Figures

Figure 2-1. Trends in Selecting State Supreme Court Justices, 1960–2010

Figure 2-2. Contestation and Defeat Rates in State Supreme Court Elections, 1980–2010

Figure 2-3. Average Vote for Incumbents in State Supreme Court Elections, 1980–2010

Figure 2-4. The Electoral Insecurity of State Supreme Court Justices by State, 2000–2010

Figure 2-5. Ballot Roll-Off in State Supreme Court Elections, 1980–2010

Figure 2-6. Average Spending (in 2008 Dollars) by Candidates in Contested State Supreme Court Elections, 1996–2008

Figure 3-1. Televised Campaign Advertising by Seat and Election Cycle in State Supreme Court Elections, 2002–2008

Figure 3-2. Televised Campaign Advertising by Seat and Incumbency in State Supreme Court Elections, 2002–2008

Figure 3-3. The Tone of Televised Campaign Airings by Election Cycle and Incumbency Status in State Supreme Court Elections, 2002–2008

Figure 3-4. Shares of Traits, Values, and Issues Appeals in Televised Campaign Advertising by Tone in State Supreme Court Elections, 2002–2008

Figure 3-5. Total Televised Campaign Airings by Sponsor and Tone in State Supreme Court Elections, 2002–2008

Figure 3-6. Total Televised Campaign Airings by Sponsor and Appeals in State Supreme Court Elections, 2002–2008

Figure 3-7. The Total Cost of Airtime Purchased by Interest Groups in State Supreme Court Elections, 2002–2008

Figure 4-1. Predicted Vote Shares of State Supreme Court Incumbents in Nonpartisan Elections at Various Levels of Televised Attack Airings, Professionalization, and Campaign Spending

Preface and Acknowledgments

IN ANY ACADEMIC ENDEAVOR, THERE ARE MANY WHO CONTRIBUTE, both intellectually and personally, to a successful project. In this regard, I am indebted to numerous colleagues, friends, and family and wish to extend my sincere thanks for their generous support.

First, I thank my former student and regular collaborator Chris W. Bonneau. Over the past decade or so, Chris has been a frequent reminder of why I chose university teaching and research as a career. Many of my ideas about judicial elections have been shaped by conversations with Chris, and our collaborations have proven to be not only interesting and productive but also highly enjoyable. Particularly relevant to this project is the chapter on ballot roll-off, which began as a collaborative project with Chris to investigate the impact of Republican Party of Minnesota v. White on citizen participation in state supreme courts. Additionally, I drew on collaborative work with Chris and Matthew J. Streb in Chapter 2, which tests for differences in features of state supreme court elections between the pre-White and post-White periods. Finally, Chris generously shared his campaign spending data, which are used in the descriptions in Chapter 2 and the empirical analyses in Chapter 4 and Chapter 5.

Herbert M. Kritzer also has been an outstanding contributor to this project. Bert read Chapter 4 and Chapter 5 as they were being developed and made countless helpful suggestions for improving the manuscript. Bert also graciously included me in discussions with Chief Justice Tom Phillips and Anthony Champagne about earlier races in Texas, which are described in Chapter 3. Overall, Bert has been a wonderful source of excellent advice, data of various sorts, good humor, and encouragement during this process.

For providing intriguing information about Texas Supreme Court elections in the 1980s and 1990s, I thank Anthony Champagne and Chief Justice Tom Phillips. I am grateful for their assistance and contributions to this work.

Many others in the fields of judicial politics and state politics have made a difference not only in this project but also in this line of inquiry. Particularly important is James L. Gibson. In this endeavor and over the span of my career, I have benefited immeasurably from conversations and encouragement from Jim. I am especially indebted to him for his intriguing insights into the practice of electing judges and for his wonderful collegiality expressed in countless ways, including in the ongoing intellectual debate about state judicial selection.

Also providing fascinating, challenging, and exciting ongoing conversations (and sometimes heated arguments) about state supreme courts, empirical research, and the practice of electing judges are Brandon L. Bartels, Damon Cann, Charles Gardner Geyh, Mark S. Hurwitz, and Matthew J. Streb. Matt was particularly helpful as a collaborator in the White Noise project, and some aspects of that analysis are replicated with additional data in Chapter 2. Mark also made a number of suggestions for improving this project and frequently is a source of interesting and enjoyable discussions about the study of law and courts.

At Michigan State University, I recognize and appreciate the outstanding work of Jonah Ralston, who served as my research assistant over the course of several years. Jonah did the lion’s share of the work in organizing and merging the ads and elections data utilized throughout this volume. Without Jonah, this work would not have progressed so well. I also am grateful to Marty Jordan and Matthew Zalewski, who provided valuable assistance in the final editorial stages of this work. Finally, I appreciate the assistance of Paul Abramson, who generously shared data on elections to the U.S. House of Representatives.

At Stanford University Press, I thank Keith J. Bybee and Michelle Lipinski. Keith and Michelle have been very helpful and supportive, making this project a pleasure from start to finish. I also appreciate the hard work and careful attention of copyeditor Margaret Pinette, production manager Patricia Myers, and the entire production and artistic team on this project.

For sharing state supreme court advertising data with the public, I thank the Brennan Center for Justice at New York University. By posting CMAG storyboards and detailed information about the various ads, the Brennan Center made this work possible. Likewise, the Justice at Stake Campaign and National Institute on Money in State Politics, in collaboration with the Brennan Center for Justice, provided vital information about campaign advertising in their New Politics of Judicial Elections reports.

On a more personal note, I thank my terrific husband and love of my life Anthony R. Duce, who always is wonderfully supportive of every task I undertake. Tony is a constant source of ideas (especially reality checks), encouragement, love, and patience, as well as a very pleasant distraction. My work and life would be so much less without Tony, who always is the perfect partner in every way.

My brilliant and beautiful daughter Amber Hall Price, ever-gracious son-in-law Joe, and wildly entertaining, totally amazing grandchildren Henry and Fiona also greatly improve my life and make my work time much more pleasant.

Finally, I thank my parents, Rose Marie Corley Gann and James Earl Gann Sr., to whom this work is dedicated. Back in the early 1970s in south Louisiana, my parents could have responded with rejection and shame to a fifteen-year-old daughter’s pregnancy, particularly given their own modest education and financial means. Instead, they provided unconditional love, helped tremendously with their granddaughter without complaint, and strongly encouraged me to continue to aspire and achieve. From my parents, I learned the value of devotion, responsibility, and plain hard work. Without their support, I would not have progressed from a high-school dropout and teenage single parent to a university distinguished professor with a happy and exciting life. They are the reason for this book and all of my other achievements.

As an important matter, I emphasize that although many contributed to this project, I am solely responsible for the arguments and analyses presented in the chapters. In no way do I represent the views of others, nor is anyone else responsible for any errors I may have made in this work.

In the following pages, I have endeavored to be as transparent as possible in the empirical choices made in the analyses, including measurement and alternative specification strategies. Chapter 4 and Chapter 5 provide extensive robustness checks on the central inferences, and Chapter 2 and Chapter 3 offer a valuable context for understanding these inferences. Overall, these analyses highlight many fascinating patterns in state supreme court elections while challenging some of the conventional wisdom about these races.

I very much hope that this work will inspire others to join in the scientific enterprise of seeking to understand state supreme courts and judicial elections. The opportunities to test significant hypotheses about campaigns and elections specifically, and judicial politics generally, are considerable in this line of inquiry. In many ways, studies of state judicial elections are in their nascent stages and much remains unexplored.

I also would like to emphasize that this book does not seek to present or constitute an argument for or against the practice of electing judges. Instead, this project adds to extant findings that collectively can be used to assess the benefits and pitfalls of any selection scheme. In this same vein, this study helps to shift the debate about judicial selection away from anecdotal evidence, unsubstantiated claims, and projected fears to more systematic evidence drawn from across the American states. Rigorous scientific studies should be a central element in any state’s decisions about how to select and retain the state court bench and the cornerstone of any intelligent discussions of the practice of electing judges. At present, the nation falls considerably short of this goal.

1

Attacking Judges

Another Dimension of Campaign Negativity in American Politics

ONE OF THE MOST NOTABLE AND WORRISOME TRENDS IN contemporary American campaign politics is the rise of televised attack advertising. Going far beyond traditional measures to promote candidates or draw distinctions among them, these nasty, below-the-belt campaigns have raised concerns from some political scientists and other astute observers that such rancor may have deleterious consequences for representative democracy. Stated effectively by West (2010: 70), attack ads in contemporary democracy are thought by many to be the electronic equivalent of the plague.

Nowhere are these misgivings being expressed more vociferously than in the context of state judiciaries by the nation’s most distinguished court reform organizations and an almost singular voice in the legal community.¹ Convinced that judges and courts are being jeopardized by the corrosive effects of money in judicial election campaigns and by attack advertising calculated to persuade a majority of the electorate that incumbent judges should be removed from office, the American Bar Association (2003: 102)² now is advocating that the thirty-eight states currently using elections³ to staff their state court benches end the practice altogether.⁴ Taking a somewhat more moderate stance is retired U.S. Supreme Court Justice Sandra Day O’Connor, who in an extraordinary level of political activism during a post-Court career, is vigorously campaigning against contestable elections.⁵ These high-profile actors are merely the tip of the iceberg of the opposition to judicial elections in their contemporary form among legal practitioners, legal scholars, and the interest groups who serve them.

Driving this movement to alter state judicial selection practices are two recent transformations in state supreme court election campaigns. First is the emergence of television campaign advertising on a nationwide scale beginning in 2002 (e.g., Brandenburg and Schotland 2008; Goldberg et al. 2005; Sample, Jones, and Weiss 2007). Although information about the scope and content of campaign advertising was not gathered in any systematic way before the Brennan Center for Justice and the University of Wisconsin Advertising Project began to capture and code televised advertising in state supreme court elections in 2000, these messages have become the norm in contested elections across the nation over the past decade and now play a central role in many judicial elections.⁶

Second are fundamental revisions in the rules of campaign engagement brought about by the U.S. Supreme Court’s decision in Republican Party of Minnesota v. White (2002). In this landmark case, the Supreme Court invalidated announce clauses in state judicial codes of conduct that prevented candidates for judgeships from expressing their views on disputed legal or political issues. Although state supreme court elections have been competitive for decades (e.g., Dubois 1980; M. G. Hall 2001a, 2007a, 2011; Kritzer 2011, forthcoming) and in some states always have been political in tone (e.g., Dubois 1980; K. Hall 1984, 2005; M. G. Hall 2001a, 2007a, 2011; Kritzer 2011), judicial campaigns in other states prior to White were issue-free, low-information events by design. However, White changed the electoral game by opening the door in all states to issue-based discourse, including attack advertising that can be part of aggressive, well-financed campaigns.⁷

These recent developments only exacerbated criticisms that initially began to heighten in the 1980s as judicial elections were becoming noisier, nastier, and costlier (Schotland 1985: 78), trends that inspired a blizzard of commentary (Gibson 2008: 60) and a multitude of law review articles over the past several decades about why judicial elections stink (Geyh 2003: 43). But as Gibson (2009: 1285) aptly observed, It is puzzling that observers are so certain of the consequences of electioneering . . . given that the scientific evidence of such effects is so scant. Indeed, while empirical scholarship on state supreme courts and judicial behavior within these institutions is advanced and complex,⁸ few studies have investigated the exact nature of these campaigns or their effects on judges, courts, and state citizenries.⁹

This project helps to address this deficiency by taking a theoretically grounded empirical look at the effects of television advertising, including harsh attacks targeting incumbents, on two key aspects of state supreme court elections: the vote shares of justices seeking reelection and the propensity of the electorate to vote. As will be discussed in considerable detail in the following chapters, a burgeoning body of political science scholarship has investigated whether campaign negativity has harmful effects on candidate preference and citizen participation in legislative and executive elections, in response to rather alarming predictions about the toxic consequences of televised attacks. In this project, I examine these contentions within the context of state supreme courts.

As a prelude to this analysis, I provide an empirical description of state supreme court elections over the past several decades, with an emphasis on trends in the competitiveness of these races. In this regard, I test the hypothesis that the White decision and other aspects of highly competitive campaigns have altered key features of state supreme court elections in recent years. Similarly, I describe televised advertising in state supreme court elections, including the content, scope, and sponsors of attacks and other issue-based appeals. As with campaign politics generally, legal scholars and judicial reform advocates are deeply skeptical of the capacity for campaigns to provide meaningful information about candidates. As Brandenburg and Schotland (2008: 1241–1242) opine, Unfortunately, TV ads are as likely to educate voters about judicial qualifications as they are to provide nutritional information about french fries. This study provides a systematic examination of this contention.

Generally, there are two compelling reasons to think that the concerns expressed by the American Bar Association, other reform organizations, and the legal academy about negativity in judicial elections may be merited. First, the legitimacy of judges may depend to some extent on an image of impartiality (e.g., Gibson 2008, 2009). Thus, negative advertising explicitly designed to disparage judges and their choices may have especially adverse effects in judicial elections. Second, many states do not list the partisan affiliations of judicial candidates on the ballot, thereby removing the most valuable heuristic in American elections. In these races, campaign messages, including scathing and repeated attacks, may constitute much of the information available to voters and thus may be a strong force in shaping the electoral fates of incumbents and the willingness of citizens to vote.

Alternatively, a theoretically sophisticated body of political science scholarship on campaign negativity in congressional and presidential elections predicts neutral or positive effects of these controversial messages. Though certainly not without contradiction (e.g., Ansolabehere et al. 1994; Ansolabehere and Iyengar 1995; Fridkin and Kenney 2011; Kahn and Kenney 1999), empirical evidence largely discounts the effectiveness of attack advertising in swaying voters (e.g., Lau et al. 1999; Lau, Sigelman, and Rovner 2007) or demobilizing the electorate (e.g., Brooks and Geer 2007; Finkel and Geer 1998; Jackson and Carsey 2007). In fact, quite a few studies (e.g., Finkel and Geer 1998; Jackson and Carsey 2007; Lau and Pomper 2001) show that negative ads increase voter turnout in some races.

Though currently lacking assessments of the direct effects of televised campaign messages in these races, empirical scholarship on state supreme courts likewise has documented that rather than being alienated by aggressive campaigns, the electorate is mobilized to vote by the very factors that intensify these races, especially partisan elections, hotly contested seats, and over-the-top spending (e.g., Baum and Klein 2007; Bonneau and M. G. Hall 2009; M. G. Hall 2007b; M. G. Hall and Bonneau 2008, 2013; Hojnacki and Baum 1992; Klein and Baum 2001). Generally, extant scholarship has failed to identify any observable behavioral manifestations of a disaffected electorate stemming from highly contentious races, at least in the form of the willingness to vote.

In short, a deep disjunction is evidenced between the assumptions underlying legal advocacy and much of the empirical evidence generated in studies of elections, including state supreme court elections. To bridge this gap, this research evaluates these two competing perspectives by capitalizing on the solid theoretical foundations of the scholarship on U.S. campaigns and elections and the analytical leverage of comparative research designs. Specifically, I examine all supreme court elections from 2002 through 2008 in the twenty-two states using partisan and nonpartisan elections to staff their highest courts. Although this study will provide a wealth of information about state supreme court elections and the campaigns of incumbents and challengers, five primary sets of questions are explored:

1. How competitive are state supreme court elections, and have the fundamental features of these races been transformed in the post-White era?

2. What is the exact nature and extent of televised campaign broadcasting?

3. Does the tone of broadcast advertising sway voters about candidates?

4. Does campaign negativity dampen citizen participation?

5. Are any impacts of campaign advertising contingent on the presence or absence of partisan ballots?

The Comparative Advantage of State Supreme Court Elections

State supreme court elections present an outstanding opportunity for systematic comparative inquiry into the question of how political institutions shape the impact of campaigns and mass electoral behavior, two of the discipline’s most abiding concerns. Across the states, both partisan and nonpartisan ballots are used in supreme court elections, reflecting institutional variation that typically does not exist for other national or statewide offices in the United States. Likewise, judicial elections are low-salience, low-information events relative to many other important elections, including presidential, senatorial, and gubernatorial races. Assessing the effects of attack advertising and other campaign messages in these alternative contexts using single models that control for the wide range of other forces affecting elections can provide much needed insight into the role played by institutional arrangements and other contextual forces in democratic politics. This seems especially important given the relative inattention in the scientific literature to nonpartisan elections, despite their widespread use in many state and local elections across the nation (Schaffner, Streb, and Wright 2001).

As the results of this analysis will show, although the political controversy over electing judges is enormously complex and largely normative, the theoretical and empirical story about the impact of negativity in judicial campaigns is straightforward. Partisan state supreme court elections in many respects resemble their legislative and executive counterparts and present a striking challenge to the notion that this new era of intense televised campaigning necessarily threatens incumbents or weakens the participatory proclivities of the electorate. Generally speaking, partisan ballots and other institutional arrangements insulate supreme court justices and state citizenries from any adverse effects of short-lived events like televised attack ads.

However, when partisan labels are removed from ballots, campaigns have pronounced consequences. On the positive side, attack advertising and other factors related to aggressive competition increase voter participation in nonpartisan elections. In fact, in nonpartisan and partisan state supreme court elections, there is no evidence whatsoever that attack advertising is a demobilizing force.

On the negative side from the perspective of the advocacy literature,¹⁰ attack advertising serves to attenuate the incumbency advantage in nonpartisan elections. Although nonpartisan elections were adopted as a reform to insulate judges from external political events, removing partisan labels creates a strategic contingency within which some of the most damaging consequences of negative advertising can manifest. In short, nonpartisan elections render some of the most serious concerns about caustic campaigns into self-fulfilling prophecies.

A theoretical perspective drawn from neoinstitutionalism explains these results.¹¹ Nonpartisan judicial elections are influenced to a greater extent by hard-fought campaigns not because courts are intrinsically different from the other branches but because nonpartisan elections alter the rules of the game. These deliberate choices made by states about selection and retention mechanisms not only define the fundamental rules under which elections operate but also create alternative strategic contingencies that structure the manner in which voters receive and use information and the extent to which incumbents are insulated from external political forces. With regard to legal advocacy, court reformers are partially right about the harmful effects of negativity but for the wrong reasons. Through the lens of political science, those predicting the pernicious effects of negativity (e.g., Ansolabehere et al. 1994; Ansolabehere and Iyengar 1995) missed the institutions and the processes to which their arguments best apply.

The implications of this inquiry are significant. Understanding linkages between citizens and government is basic to a science of politics. By assessing judicial elections comparatively with a focus on the exact nature of the campaign messages broadcast to voters, this analysis helps to improve the current state of knowledge about judicial elections while facilitating the development of theory that captures the realities of these contests. Through the lens of democratic theory and the science of judging, ascertaining how citizens are drawn into the electoral arena and how voters select among candidates is essential for building theories of judicial choice that accurately reflect the complex task of balancing pressures from democratic processes with norms of judicial independence. Looking beyond the judiciary, this study helps to refine existing accounts of electoral politics, particularly with respect to ways in which institutional arrangements, especially the presence or absence of partisan ballots, shape citizen behavior.

The Practical Politics of Judicial Selection

From a practical perspective, this work helps to inform the debate about how best to select judges. Since 1960, twelve states have abandoned partisan elections for selecting their highest courts in favor of gubernatorial appointment plans, nonpartisan elections, or the Missouri Plan, which combines initial executive appointment with subsequent retention elections.¹² Another six states jettisoned nonpartisan elections for the Missouri Plan. In addition to the current claims about the harmful effects of campaign politics, reform advocates have asserted for decades that partisan elections are highly undesirable and should be replaced with other selection schemes.

Especially central to the initial case against partisan elections but still mentioned in the contemporary debate is the claim that judicial elections fail to achieve their goal of accountability, evidenced in part by purportedly incompetent voters and a widespread lack of participation.¹³ The conventional wisdom (e.g., Dubois 1980; Schotland 1985), based largely on anecdotal evidence, is that voters know nothing and care less (Dubois 1980: 36), are plagued by ignorance, apathy, and incapacity (Geyh 2003: 63), are only slightly affected by close contests (Adamany and Dubois 1976: 743), and attach limited importance to the work of the judicial branch of government and thus decline to vote (National Center for State Courts 2002: 38). Geyh (2003: 76) summarizes the overall argument succinctly: "Judicial elections promote accountability so