Donate
Independent, objective, nonpartisan research

R 1004MBR

Authors

R 1004MBR

Tagged with:

Publication PDFs

Database

This is the content currently stored in the post and postmeta tables.

View live version

object(Timber\Post)#3742 (44) { ["ImageClass"]=> string(12) "Timber\Image" ["PostClass"]=> string(11) "Timber\Post" ["TermClass"]=> string(11) "Timber\Term" ["object_type"]=> string(4) "post" ["custom"]=> array(5) { ["_wp_attached_file"]=> string(13) "R_1004MBR.pdf" ["wpmf_size"]=> string(6) "714180" ["wpmf_filetype"]=> string(3) "pdf" ["wpmf_order"]=> string(1) "0" ["searchwp_content"]=> string(168176) "The Season of Our Discontent: Voters’Views on California Elections ••• Mark Baldassare Bruce E. Cain D. E. Apollonio Jonathan Cohen 2004 Library of Congress Cataloging-in-Publication Data The season of our discontent : voter’s views on California elections / Mark Baldassare ... [et al.] p. cm. ISBN: 1-58213-086-8 1. Elections—California—Public opinion. 2. Governors— California—Elections—Public opinion. 3. California—Politics and government—1951- 4. Public opinion—California. I. Baldassare, Mark. II. Title. JK8790.S42 2004 324.9794’053—dc22 2004023317 Copyright © 2004 by Public Policy Institute of California All rights reserved San Francisco, CA Short sections of text, not to exceed three paragraphs, may be quoted without written permission provided that full attribution is given to the source and the above copyright notice is included. PPIC does not take or support positions on any ballot measure or on any local, state, or federal legislation, nor does it endorse, support, or oppose any political parties or candidates for public office. Research publications reflect the views of the authors and do not necessarily reflect the views of the staff, officers, or Board of Directors of the Public Policy Institute of California. Foreword In 1998, a decision was made to launch the PPIC Statewide Survey. The decision was made for two reasons. First, it was readily apparent that the initiative process in California had come to dominate some of the most important public policy decisions made in recent decades. If the ballot box was becoming a key source of policy decisionmaking, it was obvious that we all needed to understand the public’s use of this important vehicle for complementing the representative process in Sacramento. Second, with a state as big and complex as California, a careful monitoring of public opinion seemed essential to anticipating issues important to voters and to identifying feasible options for resolving recurring and contentious issues that often result in gridlock and hastily launched initiatives. Over 100,000 Californians have been interviewed in the six years since the first PPIC Statewide Survey was fielded. As we anticipated, the results are often surprising. The desire for greater government benefits but the unwillingness to pay for them through conventional taxes is strong and consistent. The preference for a single-family detached home and the willingness to pay for it with ever-longer commutes is surprising, even by California standards of commute times. Active resident involvement with California’s parks and recreation areas and a preference for maintaining high environmental standards suggest that, for the voters at least, more could be done on this front without endangering the political standing of our representatives. Surely one of the most surprising findings of all is the consistently high marks respondents give to the initiative process. And now we learn from our latest round of PPIC survey findings that voter distrust of government has grown even more intense as a result of the gubernatorial election of 2002. In this report, Mark Baldassare, founder and director of the PPIC Statewide Survey, and his colleagues focus on an issue that should be of concern to every elected official in California—negative campaigning— which damages not only both opponents but also the democratic process iii itself by sowing disillusion, distrust, and cynicism among voters, many of whom then decide not to vote for either candidate. The authors also find a steady and dramatic decline in turnout in gubernatorial elections—from 70 percent of registered voters in 1982 to just 50 percent in 2002. Although there are a number of reasons for this decline in voter participation, only three in 10 Californians said that they were satisfied with the discussion of issues facing the state during the 2002 campaign. Voters wanted debates on the issues by the candidates—and they did not get what they wanted. So, they stayed away from the polls. At the same time, gubernatorial campaign expenditures reached an alltime high—mostly for the purchase of ad time on TV. And yet, only one in four likely voters just before the election in 2002 said that the campaign advertisements were helpful in making a decision about voting. Respondents expressed strong support for such election options as campaign finance reform, candidate debates, public disclosure, and voluntarily adopted codes and pledges of campaign conduct. These are straightforward suggestions and the ones most likely to survive constitutional challenges. Will representatives in Sacramento, along with the governor, take the lead in thinking through these options? Will elected officials listen to the message being sent by the state’s eligible voters? How low will trust in state government have to drift before action is taken? And will the action come from Sacramento or from the ballot box once again? All of these questions are raised by the authors in one of the most discouraging set of findings yet to emerge from the PPIC Statewide Survey. As in many other areas of public policy, there are practical solutions to the problems outlined above. The authors suggest that if nothing is done in Sacramento, voters will once again take it into their own hands at the ballot box. This latest set of findings from the PPIC Statewide Survey should serve as a high-profile warning that something needs to be done to turn the tide on a level of distrust in government that is doing serious damage to the democratic process in California. David W. Lyon President and CEO Public Policy Institute of California iv Summary The California governor’s election in November 2002 is widely regarded by scholars, political consultants, media commentators, and leaders of the political parties as one of the low points in the state’s political history. However, the qualities of the political campaigning that led to this assessment—high expenditures, negative advertising, and low voter turnout—are all indicative of recent trends in California elections. This is the backdrop against which in this report we examine the public’s views on campaigning and elections, analyze the ways that the conduct of campaigns is related to voter dissatisfaction and government distrust, and discuss what Californians would like to see changed about their state’s campaigns and elections. In 2002, a Democratic governor who won a landslide victory four years earlier, Gray Davis, defeated a Republican political novice, Bill Simon, in a surprisingly close contest that will be best remembered for its nasty tone and lack of substance. The governor’s popularity had suffered as the state faced a number of fiscal and economic calamities—including the electricity crisis that resulted from a flawed deregulation law, a dramatic slowdown in the state’s booming economy, and a multibilliondollar deficit in the state government’s budget. But the Davis-Simon matchup may never have occurred without the Democratic incumbent having spent millions of dollars in the GOP primary to help defeat Richard Riordan, a politically moderate candidate who was considered more “electable.” Davis and Simon waged a campaign of harsh words from the outset that focused on attacks and counterattacks revolving around their personal character—including allegations that there were links between the governor’s campaign fund-raising and his decisions in office, and that his GOP opponent had been involved in illegal and incompetent business dealings. They faced off in a live televised debate on the issues only once during the fall campaign. The two candidates spent their multimillion-dollar war chests on negative television v advertisements, beamed into the homes of millions of Californians, which were carefully crafted to destroy the public’s image of their opponents. In the end, the November 2002 election generated the highest campaign expenditures and the lowest voter turnout in state history. The exit polls found that most of those who voted had unfavorable images of both candidates. Governor Davis won the election by a slim five-point margin over Simon (47% to 42%). Almost immediately, Davis’s political opponents seized on the governor’s low approval ratings and the low vote threshold for qualifying a recall made possible by the record low turnout in the November election. With an infusion of cash from a wealthy supporter, the 900,000-plus signatures needed to force a special election in October 2003 were collected, and the state’s voters took the unprecedented action of removing a governor from office. The voters replaced Davis with Arnold Schwarzenegger less than a year after the incumbent’s reelection. Although voters are currently happy with their new governor, surveys indicate that they still remain dissatisfied with political campaigns and elections and are distrustful of state government. The public opinion information presented in this report is based largely on the Public Policy Institute of California (PPIC) Statewide Surveys conducted during the 2002 elections. In all, we conducted 10 surveys of over 20,000 Californians. No preelection survey of ours has ever been more meaningful and timely than the special survey of 2,000 Californians on political campaigns during the last week before the November election. That project is the centerpiece of this report, which seeks to answer the following questions about the relationships between voters and political campaigns in the state of California: • What are the recent trends in California’s voter registration, voter turnout, and campaign expenditures in statewide and legislative elections? What is the relationship between money spent, voter turnout, and winning state elections? • What were the public’s perceptions of the campaigns for the governor’s race in 2002, what are their views on negative vi advertising by candidates, and what are the factors related to dissatisfaction with the state’s elections? • What are the recent trends in the public’s trust in state and federal government, and to what extent is the conduct of political campaigns a contributing factor in the public’s distrust of state government today? • What, if anything, does the public want to do to change the nature of political campaigns and elections in California? How much support is there for campaign finance reform, public debates, increasing public disclosure by the candidates, and adopting voluntary codes of campaign conduct? We summarize our major findings below. Voting and Elections in California California’s elections over the past two decades have occurred in the context of rapid population growth and a fundamental shift in the state’s demography. The addition of over 10 million people increased California’s congressional delegation from 43 to 53 seats between 1980 and 2000. In the meantime, the ethnic and racial mix of the potential electorate has shifted significantly as the state has been transformed through immigration. Registered voters are predominantly white, yet a growing share are Latinos, blacks, and Asians. Lower levels of total voter eligibility have significantly decreased the share of registered voters as a percentage of the adult population to 60 percent, although the share of eligible voters who are registered to vote has remained fairly stable at about 70 percent. Similar to the rest of the nation, California is showing signs of declining voter turnout. In presidential election years since 1980, turnout has declined overall, but it has not declined steadily. In contrast, turnout in California’s gubernatorial elections—which occur in the nonpresidential national election years—decreased steadily and dramatically between 1982 and 2002. Over the course of the past 20 years, turnout in presidential races has dropped 6 percentage points, whereas turnout in California’s gubernatorial elections has dropped almost 20 percentage points, from 70 percent in 1982 to 51 percent in 2002. vii Historically low turnout in the 2002 governor’s race occurred in the context of another ongoing trend—steadily increasing campaign expenditures since the early 1980s, as shown in Figure S.1. The last six gubernatorial races in California have all been marked by new record highs in expenditures. The most dramatic increase was in 2002, when expenditures increased by almost $40 million from the previous election year—to nearly $98 million. Once again, the candidate who spent the most money won, which has been the trend in every governor’s election since 1986. Much campaign funding is steered toward television commercials and, following another trend in recent state elections, toward negative advertising aimed at undermining the credibility of the political opponent. Such political campaigns are not designed to encourage higher voter turnout because their focus is on winning. We hypothesize that these trends have taken a toll on public perceptions of candidates, campaigns, and elections in California and on trust in state government. 120 80 Turnout 100 Expenditures 70 80 60 60 40 50 20 0 1982 1986 1990 1994 1998 2002 40 SOURCE: California Secretary of State (2002). NOTE: Data are presented in 2002 real dollars using the Consumer Price Index provided by the U.S. Department of Labor, Bureau of Labor Statistics. Figure S.1—Gubernatorial Campaign Expenditures and Voter Turnout in California, 1982–2002 Spending ($ millions) Turnout (%) viii Voters’ Perceptions of Political Campaigns In recent years, Californians have grown increasingly dissatisfied with political campaigns in their state. In the PPIC Statewide Surveys we have conducted since the 1998 election, Californians have expressed their dissatisfaction with their candidate choices and their unhappiness with the way campaigns are run. Earlier polls reveal similar trends in public opinion about campaigns and elections. During the fall 2002 election, only about four in 10 residents said that they were satisfied with the choice of candidates for governor. Similarly, almost four in 10 Californians said that, in general, elections had gotten worse in the past 10 years, and 42 percent said that the ethics and values in campaigns had grown worse over the past decade. Likely voters were even more negative about the current state of campaigns, with half (49%) saying that elections had generally gotten worse and 54 percent saying that election ethics and values had gotten worse. Although about eight in 10 Californians said that they had been exposed to campaign advertising in the course of the fall 2002 campaign, only 33 percent said that the advertisements were helpful in making a decision about voting. Moreover, only one in four likely voters said that the candidate advertisements they had seen were helpful. Much of the dissatisfaction that Californians say they feel toward elections is focused on negative campaigning. In the PPIC Statewide Survey, 58 percent of those who were asked whether Gray Davis and Bill Simon should be critical of each other said that they should not, because campaigns today are too negative. Indeed, our preelection survey used an approach that asked about the preferred campaign behavior of Davis and Simon and then compared the responses of the candidates’ supporters and opponents. Here are some of the key results from that survey: (1) Candidates who attack their opponent run the risk of negative effects on their own image among their supporters; (2) all of the candidates’ responses to being attacked were equally unpopular, even including arguing that the basis for the attack was untrue; (3) the only responses to candidate attacks that were correlated with positive rankings of the candidates were not making a counterattack or ignoring an attack; and (4) the ability of negative ix advertising to decrease the popularity of the candidate who is being attacked must be weighed against the likely outcome of decreasing the popularity of the attacker. Potential voters said that they did not receive all of the information they needed to make a decision in the 2002 California election. In repeated preelection surveys, seven in 10 Californians said that above all else, they would like to know candidates’ stands on the issues. Notably, however, this demand for information about issues was not met in the 2002 campaign. In the same preelection surveys, only three in 10 said that they were satisfied with the discussion of issues facing the state during the campaign. On top of this frustration with the content of the campaign was disappointment with the means that candidates use to deliver campaign messages. When asked how they prefer to learn about candidates, voters chose debates more than any other means of communication. However, in the 2002 campaign, there was only one debate between the candidates. Moreover, although the gubernatorial candidates relied heavily on television advertising, about four in 10 Californians said that they would view candidates less favorably when they primarily used television advertising to get their message to voters. In sum, candidates hurt their own images by campaigning negatively, by providing inadequate information to voters, and by relying too heavily on television advertising to connect with the voters. This strategy may still result in victory, but it could depress the turnout for the winner. Thus, there are political risks in the short run and the long run for the political candidate. Equally important, attack campaigns, even beyond their effects on turnout, may engender lasting feelings of distrust in government. Political Campaigns and Distrust in Government Political observers have long suspected that recent trends in campaign practices such as negative advertising may be linked to steadily declining trust in government. The evidence from our November 2002 PPIC Statewide Survey, as well as past studies of trust in government in California, suggests that campaigns do have an effect on political trust. Californians’ trust in state government declined during the 2002 election x campaign and has not returned to previous levels in the wake of the recall of Governor Davis and his replacement by Arnold Schwarzenegger, who currently enjoys high overall job approval ratings. Negative perceptions of campaigns appear to be a contributing factor for declining trust, and confidence seems to be difficult to restore once the public loses faith in government. The level of trust in the federal government expressed by California residents actually increased since our first PPIC Statewide Survey in April 1998, up to and including the November 2002 election—in part a reaction to the September 11, 2001, terrorist attacks. As of this writing, trust in the federal government has dropped somewhat below the levels we recorded before the 2002 elections. In contrast, trust in state government was moving in a downward direction over time. Trust in state government reached its peak in 2001 and early 2002, with nearly half of all Californians (47%) saying that they trusted state government to do what is right just about always or most of the time. By August 2002, about one in three Californians (37%) said that they trusted the state government. In the months following the November 2002 election, trust in the state government drifted to significantly lower levels and, as of this writing, it has not improved to the levels recorded in the summer and fall leading up to the 2002 election. We conducted a multivariate analysis to determine the factors most highly correlated with distrust in state government. For both the public at large and likely voters, negative assessments of incumbent Governor Gray Davis, negative perceptions of the campaign, and Independent voter status were all significantly correlated with more distrust in state government. For all adults, strong Republicans had more distrust in state government than others did at the time, whereas Latinos had more trust than other racial/ethnic groups in state government. Greater interest in politics was also associated with more trust in state government. Public Support for Political Campaign Reforms We looked at a wide variety of political reforms that were chosen because they are most likely to survive constitutional challenges and because they are the types of policy proposals that have been considered xi in state contexts and discussed in the national political arena in recent years. We tested public support for the following: campaign finance reforms, candidate debates, public disclosure, and voluntarily adopted codes and pledges of campaign conduct by the candidates. We found considerable interest in changing the status quo of California’s campaigns and elections on all of these dimensions. The issue of campaign finance was prominent in the 2002 governor’s race, given the record spending of nearly $100 million by the two major party candidates. Perhaps most surprising of all, given the fiscally conservative nature of Californians, 57 percent of likely voters thought that public, rather than private, financing of campaigns was a good idea. This finding suggests that recent negative experiences with state campaigns, such as the 2002 governor’s race, had made many voters receptive to the public paying the tab for the election process. Californians have been telling us in surveys that they are frustrated with the emphasis of statewide political campaigns on 30-second television commercials. What would they prefer instead? Two in three likely voters said that more debates in the 2002 gubernatorial contest would have made the election campaign better. And when asked whether they would favor or oppose an initiative that would require five prime-time broadcast gubernatorial debates, 60 percent were in favor of the proposal. Voter support for election reforms that provide more personal contact with the candidates suggests another area for improvement. Disclosure is a key ingredient in elections because voters need access to accurate information when making ballot choices. Among likely voters, 71 percent said that the immediate disclosure of campaign contributions would make them more favorably disposed toward a candidate. By contrast, fewer say that they would have a positive impression of candidates who released their tax forms or medical records. The candidates’ sources and amounts of financial support are important to voters, and their preferences for more information point to a potential avenue for political reform. Candidates cannot be compelled to avoid personal attacks and negative campaigns for first amendment reasons. But what if candidates can be induced to do so by signing pledges that held them accountable xii during the course of a political campaign? Our survey suggests that California voters are open to this idea: Seventy-seven percent of likely voters said that they would view candidates more favorably if they signed a pledge or a code of conduct to “run a truthful, fair, and clean campaign,” and over 70 percent of likely voters said that they would view a candidate more favorably if he or she signed a code of conduct to run “an issue-oriented campaign.” The support among likely voters was almost equally high for candidates who would sign a code of conduct “not to use race, gender, religion, sexual orientation, or age as a basis for attack.” Despite the practical limitations of codes of conduct, many voters are keen on the idea of them because they offer a greater likelihood of a campaign focusing on the issues and a break from the negative tone of today’s California campaigns. The California public expressed a high level of dissatisfaction with the negative tone of the 2002 governor’s election, the limited access they had to the candidates, and the lack of attention to the issues confronting the state. The most expensive governor’s race in the state’s history resulted in the lowest voter turnout and one of the nastiest political campaigns on record—all following trends in recent state elections and likely to continue unless there is a concerted effort to change the nature of elections. In the end, the campaigns’ focus on negative advertising appeared to harm both the candidate doing the attacking and the candidate who was attacked. Although the candidate who spent the most money campaigning in this manner won, we found that there are political risks to spending so much money on negative advertising. The aftermath of Davis’s narrow victory in a low turnout election—a recall in which the incumbent found it difficult to mobilize his support base to fight off the GOP’s attempt to remove him from office—highlights the vulnerabilities of this political strategy. If the current trends of public dissatisfaction and voter disengagement continue, there could be more serious consequences—not the least of which is declining trust in state government and its elected officials—endangering the future of democratic society in California. The public’s frustration with state politics was evident in the course of the 2002 governor’s race, and it peaked in 2003 when California xiii voters removed the winner from office. But the recall did not address the underlying problem of negative campaigns and declining turnout in gubernatorial elections. California voters, deeply dissatisfied, have indicated their preferences for a new type of campaign behavior—increased financial disclosure, more public debates, campaign finance reform, and candidate codes of conduct—most of which could be implemented in time for the 2006 governor’s election. Given their frustration with the status quo and their support for change, the voters may take the matter of election reform into their own hands through the initiative process. xiv Contents Foreword......................................... iii Summary......................................... v Figures .......................................... xvii Tables ........................................... xix Acknowledgments................................... xxi 1. RECENT TRENDS IN CALIFORNIA ELECTIONS ..... 1 2. POLITICAL CAMPAIGNS AND VOTER DISSATISFACTION ............................. Population, Voting, and Public Opinion Surveys .......... Voters’ Views on the 2002 Governor’s Election ........... Perceptions of Political Campaigns .................... Low Turnout and More Initiatives: Confirming Evidence of Campaign Dissatisfaction....................... What Happens When Candidates Focus on Negative Campaigning?............................... Measuring the Influence of Attack Advertising: Results of a Survey Experiment ........................... Correlates of Voter Dissatisfaction with Candidate Choices... Information on Election Campaigns ................... Conclusions .................................... 13 14 15 19 21 24 26 31 31 34 3. POLITICAL CAMPAIGNS AND VOTER DISTRUST .... The Reasons for Public Distrust ...................... Survey Evidence on Distrust in Government ............. Trends in Distrust in State Government ................ Correlates of Distrust in State Government .............. Conclusions .................................... 37 38 40 44 48 50 4. CONCLUSIONS ON PUBLIC SUPPORT FOR CAMPAIGN REFORMS .......................... Campaign Finance Reform ......................... Candidate Debates ............................... 53 54 56 xv Public Disclosure ................................ 58 Candidate Codes of Conduct........................ 60 Appendix A. Dissatisfaction with the Candidate Choices for Governor .... B. Trust in State Government, All Adults ................. C. Trust in State Government, Likely Voters ............... 63 65 67 Bibliography ...................................... 69 About the Authors .................................. 77 Related PPIC Publications............................. 79 xvi Figures S.1. Gubernatorial Campaign Expenditures and Voter Turnout in California, 1982–2002 ................. viii 2.1. Gubernatorial Campaign Expenditures and Voter Turnout in California, 1982–2002 ................. 16 xvii Tables 1.1. 1.2. 1.3. 1.4. 1.5. 1.6. 1.7. 2.1. 2.2. 2.3. 2.4. 2.5. 2.6. 2.7. 2.8. 2.9. 2.10. Population Change in California, 1980–2002 ......... Total Population, Eligible Voters, and Registered Voters in California, 1980–2002 ....................... Voter Turnout in Presidential Elections in California, 1980–2000 ................................. Voter Turnout in Gubernatorial Elections in California, 1982–2002 ................................. Spending on Gubernatorial Elections in California, 1982–2002 ................................. Cost per Voter of Gubernatorial Elections in California, 1982–2002 ................................. Political Party Affiliation of Registered Voters in California, 2002 .............................. Californians’ Satisfaction with Candidate Choices for Governor ................................... Californians’ Perceptions of Election Trends .......... Californians’ Attitudes Toward Politics and Politicians, 1983 and 1999............................... Californians’ Perceptions of Advertising by Gubernatorial Candidates ....................... Californians’ Attention to Campaigns and Voting Uncertainty ................................. Citizens’ Campaign Reform Initiatives in California, 1980–2002 ................................. Californians’ Perceptions of Negative Campaigning ..... Perceptions of Negative Campaigning by Supporters and Nonsupporters ............................... Californians’ Preferences and Satisfaction on Attention to the Issues ................................. Californians’ Preferred Means of Candidate Communication .............................. 2 3 5 6 8 9 11 17 18 18 19 21 23 25 29 32 33 xix 3.1. Californians’ Trust in Federal Government, 1998–2004 ................................. 45 3.2. Californians’ Trust in State Government, 2001–2004.... 46 4.1. Californians’ Support for Funding Political Campaigns... 55 4.2. Californians’ Support for Gubernatorial Debates ....... 57 4.3. Californians’ Support for Public Disclosure ........... 59 4.4. Californians’ Support for Codes of Conduct .......... 60 A.1. Dissatisfaction with the Candidate Choices for Governor................................... 63 B.1. Trust in State Government, All Adults .............. 65 C.1. Trust in State Government, Likely Voters ............ 67 xx Acknowledgments We wish to thank The Pew Charitable Trusts for their generous support of the public opinion surveys and focus groups conducted during the 2002 California election and the 2003 governor’s recall election that made this report possible. Dorie Apollonio did the analysis and wrote the first draft of Chapters 1, 2, and 3. She also did the analysis for Chapter 4. Bruce E. Cain wrote the first draft of Chapter 4. Mark Baldassare rewrote the draft chapters, incorporated the draft chapters and other materials into a unified report, and took the report through several rounds of revisions. Jonathan Cohen was involved in the survey design, questionnaire development, data collection, and analysis. The authors were assisted in their design and implementation of the public opinion research by staff of the Center for Campaign Leadership at the Institute of Governmental Studies, University of California, Berkeley, including Keena Lipsitz, Julia Mitchell, Sunetti Shah, Christine Trost, Dan Schnur, and Darry Sragow, and by the PPIC Statewide Survey staff at the Public Policy Institute of California, including Eliana Kaimowitz and Renatta De Fever. We also wish to thank the staff at Schulman, Ronca & Bucuvalas, Inc., for their assistance and advice in the design of the survey questions. We also benefited from feedback offered by many scholars, journalists, political consultants, elected officials, and students at the events and seminars sponsored by the Center for Campaign Leadership in Berkeley and Sacramento. We would like to thank the reviewers of an early draft of this report—Shaun Bowler, John Ellwood, David Lesher, Paul Lewis, Joyce Peterson, Karthick Ramakrishnan, and Susan Rasky—for their comments and suggestions on the substance of the report. xxi 1. Recent Trends in California Elections Once upon a time, California voters stood in line to shake hands at political rallies with would-be officeholders and waited at train stations for their whistle-stop tours. Now, they are most accustomed to seeing candidates through the lens of 30-second television commercials beamed into their living rooms. Elections in California have become more professionalized in recent decades—and more expensive. Consultants have become more knowledgeable about how to run winning campaigns. Candidates are more informed than ever by polls and focus groups about what the voters want. Their messages are carefully tested before they are ever seen or heard by voters. Yet, for all this expertise and money devoted to figuring out how to win elections, Californians tell us in our public opinion surveys that they are dissatisfied with the state’s election process and the negative tone of campaigns. Many Californians are also forgoing a benefit and a responsibility of U.S. citizenship—registering to vote—and many of those who are registered do not turn out to cast their ballots on Election Day. This report explores the level and sources of discontent with political campaigns among Californians, examines some of the critical implications of this discontent for the democratic process, and considers the public’s views on how to improve the state’s elections in the future. California’s elections have become more expensive as the state’s population has grown and its demography has been changing. The defining characteristics of California’s elections are that they encompass a large geographic area and a racially and ethnically diverse electorate. These factors set the stage for media driven campaigns that focus on reaching voters through television commercials, adding high costs to all 1 statewide campaigns and leading to complex decisions that all statewide candidates face in trying to reach diverse audiences. Since 1980, California’s population has increased by almost 50 percent—from less than 24 million, to over 35 million people in the year 2002 (Table 1.1). In the late 1980s, the population grew by over 5 percent in each election cycle. In the recession of the early 1990s, it slowed to approximately half a million new people per election cycle. But in the two years preceding both the 2000 and the 2002 election, the California population grew by more than one million people. Meanwhile, decades of foreign immigration and racial and ethnic change have transformed California into the first large “majorityminority” state where whites make up less than half of the total population. As of this writing, about one in three residents is Latino in ethnic origin, and the remaining one in six adults is Asian, black, or of another racial category. Moreover, substantial proportions of Asians and Latinos in California are either immigrants or children of immigrants (Baldassare, 2000, 2002). Although not all of this new population was eligible to vote— because of either age disqualification or noncitizenship—the addition of over 10 million people in 20 years increased California’s Congressional Table 1.1 Population Change in California, 1980–2002 Population Population Increase Percentage Year (thousands) (thousands) Increase 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 23,782 24,805 25,816 27,052 28,393 29,828 30,987 31,523 31,962 32,862 34,088 35,336 525 1,023 1,011 1,236 1,341 1,435 1,159 536 439 900 1,226 1,248 2.3 4.3 4.1 4.8 5.0 5.1 3.9 1.7 1.4 2.8 3.7 3.7 SOURCE: California Department of Finance (2003). 2 delegation from 43 to 53 seats between 1980 and 2000. Moreover, despite the voter eligibility obstacles resulting from immigration that slow the translation from population growth to electoral change, the ethnic and racial mix of the electorate has been shifting significantly over the past decade. In the most recent state elections, about one in four voters was Latino, Asian, or black (Baldassare, 2002). These demographic and racial/ethnic shifts in the population are important to acknowledge because it would be unfair to claim that trends in political campaigning alone account for all of the decline in voter participation in recent years. Irrespective of voter registration, lower levels of total voter eligibility have decreased the share of voters as a fraction of the total adult population. In the 1980s, the percentage of adult residents eligible to vote in elections remained at a nearly constant rate of 65 percent (Table 1.2). By the early 1990s, the share of the adult population eligible to vote in California dropped sharply, to 58 percent. Table 1.2 Total Population, Eligible Voters, and Registered Voters in California, 1980–2002 Total Eligible Registered Population Voters % Voters % Year (thousands) (thousands) Eligible (thousands) Registered 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 23,669 15,384 65 11,362 24,500 15,984 65 11,559 25,550 16,582 65 13,074 26,444 17,561 66 12,834 28,592 19,052 67 14,005 29,800 19,245 65 13,478 31,000 20,864 67 15,101 32,000 18,496 58 14,724 32,344 19,527 60 15,602 33,252 20,806 63 14,969 34,336 21,461 63 15,707 35,802 21,451 60 15,303 74 72 79 73 74 70 73 78 80 72 73 71 SOURCE: Field Poll (2002). NOTE: Population estimates are slightly different from those presented in Table 1.1. 3 The share has increased slightly since that time, with a little more than 60 percent of the adult population eligible to vote in California elections by 2002. Whatever the changes in numbers eligible to vote in the adult population, however, the proportion of eligible voters who register to vote has remained fairly stable, with only a few exceptions. Since 1980, an estimated 70 percent or more of those eligible to vote have registered. During the last 20 years, the share of eligible voters registering to vote has increased only slightly in a few election years. A much larger share than usual—79 percent of those eligible to vote—registered during the 1984 presidential election cycle, which pitted Ronald Reagan against Walter Mondale. In the 1994 gubernatorial election cycle, where Kathleen Brown faced Pete Wilson, 78 percent of those eligible to vote were registered. And in 1996—a presidential year that featured not only the two major party candidates, Bill Clinton and Bob Dole—but also a surge in new Latino voters in response to voter registration drives and the passage in 1994 of an anti-illegal immigrant initiative—eight in 10 of those eligible to vote were registered. With the exception of these years, the levels of voter registration have remained stable, despite many efforts to increase voter registration among the legions of young adults and new U.S. citizens. Not all of the registered voters in California actually go to the polls on Election Day. Similar to the rest of the nation, California has shown signs of decreased voter turnout in terms of the share of registered voters going to the polls. The national enfranchisement of 18-to-20 year olds in 1972 led to a decline in the percentage of registered voters turning out nationwide, because young adults are less likely to vote than any other age group. The extensive research by Wolfinger and Rosenstone (1980) notes that some of the causes of low voter turnout among young adults may be residential mobility, which makes registration more difficult, or lower levels of education; individuals with college degrees, which are rarely acquired before the early twenties, are much more likely to vote. However, although voter turnout in California has declined overall in presidential election years since 1980, it has not declined steadily (Table 1.3). Whereas only two in three registered voters turned out in 4 Table 1.3 Voter Turnout in Presidential Elections in California, 1980–2000 % of Registered Year Voters 1980 1984 1988 1992 1996 2000 77 75 73 75 66 71 SOURCE: California Secretary of State (2000). 1996, the share of registered adults voting in the 2000 election increased by 5 percentage points to 71 percent, which is near the levels of turnout seen in 1988 (73%) and only slightly lower than turnout rates in 1984 (75%) and 1992, when three in four of those registered actually voted. Overall, turnout rates in presidential elections in California have been inconsistent across election cycles but have not dropped significantly in the past two decades. In contrast, turnout in California’s gubernatorial elections—which occur in the nonpresidential national election years—has decreased dramatically between 1982 and 2002. California, like 36 other states and territories, elects its governor in a nonpresidential election year—that is, two years after the presidential race. Seven other states and territories also elect governors in a nonpresidential election year but not at the two-year Congressional election interval. Without the national attention that comes with the race for president, registered voters are less likely to participate in off-year elections. Turnout in the gubernatorial election years in California has been lower than turnout in presidential years in every election cycle since 1912, the first election cycle for which the California Secretary of State collected election data. Moreover, the gap in voting between presidential election years and gubernatorial election years has increased in recent 5 decades because the turnout in presidential years has changed little whereas turnout in gubernatorial years has declined steadily. Over the past 20 years, voter turnout in presidential races has dropped 6 percent, whereas voter turnout in gubernatorial election years has dropped almost 20 percent, from 70 percent in 1982 to 51 percent in 2002 (Table 1.4). The declining turnout in gubernatorial elections happened in two stages. After the 1982 elections, turnout dropped 11 percentage points to 59 percent and remained at about the 60 percent level until 1998. And then, in 2002, at the end of a gubernatorial race between Gray Davis and Bill Simon that received national recognition for its high expenditures and the negative tone of the campaigning, turnout dropped again, to only about half of all registered voters. The decline in voter turnout in 1986 and 2002 does not reflect dramatic changes in the underlying adult population. As discussed above, during these years, California was experiencing steady population growth and was integrating immigrants into the voting rolls at a fairly constant rate. The drop in turnout in 2002 is noteworthy, since it represents a large percentage decline in voter participation from the previous general election. The level of voter turnout in November 2002 is comparable to that of traditionally poorly attended special elections; the closest comparable race was the 1973 special election in California, which had a similarly low turnout of 48 percent. Table 1.4 Voter Turnout in Gubernatorial Elections in California, 1982–2002 % of Registered Year Voters 1982 1986 1990 1994 1998 2002 70 59 59 60 58 51 SOURCE: California Secretary of State (2002). 6 Although the last six gubernatorial races in California have nearly all been marked by new record highs in expenditures, no such pattern is evident in reviewing aggregate spending in state legislative races between 1980 and 1998 (California Secretary of State, 1998; note: Data after 1998 were not available). California has 80 Assembly seats, which means that candidates must run every two years; California’s 40 Senate seats have four-year terms, with approximately half of the seats in contention every two years. For instance, total spending in primary and general elections increased between 1982 ($84 million) and 1986 ($99 million); however, total spending in 1994 ($106 million) was higher than in 1998 ($101 million) for all state legislative elections (all figures are in 2002 real dollars). Since the passage of legislative term limits in 1990 took effect, the cost of state legislative races has been over $100 million in every election but, again, the available data suggest that total spending has fluctuated rather than increased over time. Federal legislative elections in California in this same time frame also do not indicate a consistent upward trend in spending (U.S. Federal Election Commission, 2004). For example, the total amount spent in California actually declined between 1982 ($52 million) and 1986 ($38 million), then increased sharply in 1992 ($80 million), and declined from that level in 1994 ($58 million) when we look at all Congressional elections in the state (all figures are in 2002 real dollars). Congressional seats are not term-limited like state seats. As a result, spending has fluctuated from year to year, largely in response to contentious individual races rather than to any structural changes. In recent years, there have been only a handful of competitive races as a result of a redistricting process after 2000 that favored safe seats for both Democrats and Republicans in office. As contentious individual races surface in the future, we expect to see the trend of fluctuations continue. Candidates running for governor since 1982, meanwhile, have increased their spending in every election year except 1994 (Table 1.5). The most dramatic leap came in 2002, when overall expenditures increased by almost $40 million from the previous election year. In earlier elections, the difference in expenditures from one election to the 7 Table 1.5 Spending on Gubernatorial Elections in California, 1982–2002 Year 1982 Tom Bradley (Dem) George Deukmejian (Rep) Total Spending 6,803,633 4,972,389 11,776,022 2002 Dollars 22,602,969 1986 Tom Bradley (Dem) George Deukmejian (Rep)a Total 6,137,522 9,565,125 15,702,647 27,891,845 1990 Dianne Feinstein (Dem) Pete Wilson (Rep) Total 13,227,930 16,028,590 29,256,520 45,768,626 1994 Kathleen Brown (Dem) Pete Wilson (Rep)a Total 12,251,634 19,555,243 31,806,877 39,377,870 1998 Gray Davis (Dem) Dan Lungren (Rep) Total 28,642,125 23,845,008 52,487,133 58,788,186 2002 Gray Davis (Dem)a Bill Simon (Rep) Total 64,261,067 33,595,244 97,856,311 97,856,311 SOURCE: California Secretary of State (2002). NOTES: Data on 2002 dollars are presented in 2002 real dollars using the Consumer Price Index provided by the U.S. Department of Labor, Bureau of Labor Statistics. Winner is shown in boldface. aIndicates incumbent. next had always been less than $20 million (all figures are in 2002 real dollars). This change in expenditures over time does not reflect structural changes; California governors were limited to two terms of office long before the 1990 California initiative establishing legislative term limits. Although candidates for governor must reach increasing numbers of people because of California’s increasing population, the 8 growth in population between 1998 and 2002 was not large enough to fully explain the dramatic increase in gubernatorial election spending. Moreover, other statewide races, such as recent races for U.S. senator, have not been marked by similar steady increases in spending. For example, the 1982 U.S. Senate race between Pete Wilson and Jerry Brown, at a cost of $39 million (in 2002 dollars), was more expensive than later Senate races with the exception of the 1994 race between Dianne Feinstein and Michael Huffington, which was clearly an outlier because it involved a Democratic incumbent and a millionaire GOP challenger able to call upon personal wealth to finance this race. It is noteworthy that the increased spending on gubernatorial campaigns in recent years has been consistent even when considered as spending on potential or actual voters. The cost per vote cast has increased substantially in 20 years, from $2.80 in 1982 to $12.64 in 2002 (Table 1.6). Although the cost per eligible voter has increased at a slower rate—a fact that reflects the increasing state population—the Table 1.6 Cost per Voter of Gubernatorial Elections in California, 1982–2002 $ Cost per $ Cost per $ Cost per Year Vote Cast Registered Voter Eligible Voter 1982 1986 1990 1994 1998 2002 2.80 3.66 5.79 4.42 6.82 12.64 1.96 2.17 3.40 2.76 3.93 6.39 1.41 1.59 2.38 2.13 2.83 4.56 SOURCES: California Secretary of State (2002); Field Poll (2002). NOTES: The number of votes cast includes all ballots cast in the election; however, in each election, a small fraction of ballots did not include a vote for governor. Data on cost per voter are presented in 2002 real dollars using the Consumer Price Index provided by the U.S. Department of Labor, Bureau of Labor Statistics. 9 significant increase from $1.41 in 1982 to $4.56 in 2002 suggests that population growth is not the only consideration. The largest increase in per capita costs, unsurprisingly, occurred between 1998 and 2002, when the cost per eligible, registered, and actual voter in the election increased respectively by 61 percent, 62 percent, and 85 percent. What are the implications of the fact that continuing increases in gubernatorial candidate spending coincide with decreases in voter turnout during the same respective years in California? First, just because candidates spend money communicating to voters and voters are more likely to be informed about the upcoming election and the candidate choices does not necessarily lead to more political participation. Indeed, the purpose of campaign advertising may be to encourage voting for your candidate and discourage voting for their candidate. As for the latter, acceptable outcomes for the political consultant could include vote-switching or simply staying home on Election Day. In addition, more spending could mean more negative advertising, which might dishearten potential voters and result in their simply not voting. Our analysis revealed a striking correlation between campaign spending and winning a gubernatorial election in California: Since 1986, the candidate who spent the most money won the election. Gubernatorial elections have been distinctive from all other kinds of statewide elections in California since the 1980s. Unlike other kinds of elections, the costs of gubernatorial races are steadily increasing, and constituents are increasingly unwilling to vote for the gubernatorial candidates running for office. One factor that may explain the trend of lower voter turnout may be the growing numbers of decline-to-state voters in the California electorate, now accounting for about one in six voters (Table 1.7). The growth of Independent voters may be another symptom of the deep disillusionment that California voters have for the campaigns of major party candidates running for statewide offices (Baldassare, 2002). This low turnout is troubling for the state’s system of representative democracy. The governor is the most powerful officeholder in California state politics and an important and high-profile symbol of the state’s 10 Table 1.7 Political Party Affiliation of Registered Voters in California, 2002 Affiliation % of Registered Voters Democrat Republican Decline to state (Independent) Other parties 45 35 15 5 SOURCE: California Secretary of State (2002). executive branch as well as of state government policymaking, yet the statistics indicate that a smaller portion of the electorate is choosing the governor over time. Moreover, with the state’s changing demographics, the voting population does not reflect the racial and ethnic mix of the state’s adult population. Results for 2002 indicate an extension if not an acceleration of recent trends. The candidates for governor convinced barely half of registered voters to cast ballots, and they spent nearly $100 million in the campaign. Once again, the candidate who spent the most money was the winner. Less than a year after the 2002 election, however, Gray Davis faced a first-ever recall, and the voters replaced him with Arnold Schwarzenegger. Has the experience of the 2003 recall changed the political climate? There was clearly excitement about the recall and signs of greater interest in state elections—61 percent of registered voters participated in the 2003 recall—but this increase in voter participation may be a short-lived phenomenon driven by the unique nature of this election. The March 2004 primary, with two state fiscal ballot measures that were heavily promoted by Governor Arnold Schwarzenegger, attracted just 44 percent of registered voters to the polls (California Secretary of State, 2003, 2004). Of course, voter response in the 2006 governor’s election will provide a more definitive answer. In this report, we argue that the low voter turnout in gubernatorial elections reflects, at least in part, an increasing dissatisfaction with political campaigning in California. This trend of increasing 11 dissatisfaction, in turn, is marked by a steady decline in the public’s trust in state government and elected officials that has not been reversed by the recent recall election. In the remainder of this report, we look at voters’ attitudes during the 2002 election—which offer the most recent evidence from a long-term trend that has been evident in California for some time—to understand why political campaigns have fallen out of favor, and we discuss ways to reverse these troubling trends through a variety of political and election reforms. 12 2. Political Campaigns and Voter Dissatisfaction In recent years, Californians have grown dissatisfied with political campaigns and elections in their state. As we noted in Chapter 1, this trend seems most evident in gubernatorial elections. In the 2002 gubernatorial race, turnout declined to 51 percent of registered voters, down from 71 percent in the 2000 presidential elections. Of course, turnout in presidential election years typically exceeds that in gubernatorial election years; however, the 2002 turnout of about half of registered voters represents a considerable decline from the nearly 60 percent of voters who turned out in the 1998 gubernatorial race in California. In successive surveys that we have conducted since the 1998 election, the voters have sent a clear and consistent message: They are dissatisfied with their candidate choices, they are unhappy with the way that campaigns are run, and they have become increasingly distrustful of state government. Although the 2003 recall was mostly about unhappiness with the incumbent, Governor Gray Davis, it was fought against a backdrop of widespread dissatisfaction with electoral politics that has been in evidence for some time. Moreover, our public opinion survey conducted during the 2003 recall election indicated that voters remained unhappy about the state of campaigns and elections with almost half of likely voters saying that the experience of the recall election actually made them feel worse about California politics (see PPIC Statewide Survey, October 2003). It is important to note that Californians were not always so cynical about their government and elected officials, so negative about the campaigns and candidates, and so apparently disconnected from the voting process as they are today. If we turn back the clock to the 1960s, for instance, nearly eight out of 10 registered California voters appeared 13 at the polls during those election years. With the state’s residents apparently buying into the concept that government operated in an efficient and an effective manner and could make their lives better, their support for the large public investments that their leaders proposed (e.g., public universities, freeway systems, state aqueducts) was all but guaranteed at the polls. The citizens’ initiative process was rarely used in this era to overrule government or to set the course for public policy. But over the past 40 years, Californians’ faith in state government and their elected officials has soured and their interest in participating in the state’s elections has diminished (Baldassare, 2000). In recent election cycles leading up to and beyond the 2003 recall, the trends of voter distrust and disinterest appear to have accelerated. These negative changes in perceptions of state government, of the campaigns of gubernatorial candidates, and of the election process itself have occurred in the context of both dramatic increases in the costs of political campaigning and heightened professionalism in campaigns. Candidates have steadily increased their investments in state races over the years in an effort to win over California’s expanding electorate, and since the 1986 contest, the gubernatorial candidate who spent the most won the race. We hypothesize in this chapter that voter dissatisfaction with the political process is related to the public’s dissatisfaction with the way political campaigns are being run. Increased spending on state campaigns may have exacerbated this trend, as campaigns tend to focus on the “dark side” of politicians. Recent California elections have been marked by a heavy reliance on relatively new forms of political communication, particularly negative campaigning and television advertising, and these efforts come under special scrutiny in our analysis. Moreover, investment in these tactics has changed the nature of political campaigning in California today, and such tactics may explain why so many voters stay home on Election Day. Population, Voting, and Public Opinion Surveys In this analysis, we draw on such sources as the PPIC Statewide Survey series and nonpartisan preelection surveys, media exit polls, and government data on state elections. We rely in particular on a special 14 survey conducted in the week leading up to the November 2002 California statewide elections. This preelection survey of 2,000 adult residents was developed by PPIC survey staff, in conjunction with the Center for Campaign Leadership (CCL) at the University of California, Berkeley, and was funded by The Pew Charitable Trusts. Our survey of Californians’ perceptions and attitudes in the closing days of an election cycle attempted to measure both the level of satisfaction with current campaigns, what Californians ideally wanted to take place in their elections, and the kinds of campaigning that they particularly did not appreciate. The survey questions encompassed both elections in general and, more specifically, the 2002 gubernatorial race. Other public opinion data are drawn from the PPIC Statewide Surveys, which have been conducted regularly since early 1998. Other time-series data before 1998 come largely from the Field Poll and the Los Angeles Times Poll. These survey sources have spent much of the last two decades measuring public opinion on ballot issues but have sometimes focused on broader political issues such as knowledge about candidates and campaigns, thus providing data for our comparisons over time. Finally, the California Secretary of State routinely provides information on population and voting trends in California, as well as detailed information about the nature of campaigns. These data include measures of candidate spending, voter turnout, and demographic changes in the population in general as well as in the population of eligible and registered voters. These data provide an objective baseline against which we can consider the public’s perceptions in surveys. Voters’ Views on the 2002 Governor’s Election To place voters’ views in perspective, it is important to note that the 2002 governor’s election reflected trends in evidence for the past 20 years (Figure 2.1). In general, spending on state elections has increased dramatically, while the turnout of registered voters has steadily declined. As we noted above, the trend is most evident in gubernatorial elections rather than in presidential election years. The 2002 governor’s race offers the most dramatic example of these inverse trends. It is important to note that spending on political campaigns is not designed to increase overall voter turnout. The purpose of campaign 15 Spending ($ millions) Turnout (%) 120 80 Turnout 100 Expenditures 70 80 60 60 40 50 20 0 1982 1986 1990 1994 1998 2002 40 SOURCE: California Secretary of State (2002). NOTE: Data are presented in 2002 real dollars using the Consumer Price Index provided by the U.S. Department of Labor, Bureau of Labor Statistics. Figure 2.1—Gubernatorial Campaign Expenditures and Voter Turnout in California, 1982–2002 spending is to win an election. Many political consultants are thus interested in designing campaigns that increase voter turnout for their candidates and depress voter turnout for their opponents. Moreover, individual decisions about whether to vote tend to be related to personal characteristics, such as education and length of residence at the same address rather than typically being a decision to vote for a specific candidate (Wolfinger and Rosenstone, 1980). But even statewide campaigns that devote money to getting out the vote without reference to specific candidates or political parties, as in the 2002 election in California, may not motivate more voters to go to the polls. Indeed, turnout in November 2002 was at a historic low. This is noteworthy because it suggests that marketing and advertising are ineffective in mobilizing voters when the public is unhappy with the campaigns and their choice of candidates, as they apparently were in the 2002 governor’s race. What did the PPIC Statewide Survey tell us about voters’ attitudes toward this election? First, only about four in 10 California adults said that they were satisfied with the choice of candidates for governor (Table 2.1). Notably, both registered voters and likely voters (i.e., voters with a 16 Table 2.1 Californians’ Satisfaction with Candidate Choices for Governor % Satisfied with Choices for Governor All adults Registered voters Likely voters 42 41 37 SOURCE: PPIC Statewide Survey (November 2002). NOTE: N = 2,000. history of participating regularly in elections) were less likely than residents overall to say that they were satisfied with the candidates. For instance, 42 percent of adults stated that they were satisfied with the choice of candidates in the week before the election, but only 37 percent of likely voters said so. The unhappiness with candidates may explain voters’ reluctance to go to the polls in 2002, despite the fact that many of them had voted regularly in the past. Similarly, when people were asked whether they felt that, in general, elections in California had gotten better or worse in the last 10 years, a significant proportion said that elections had gotten worse (Table 2.2). Nearly four in 10 adults (37%) claimed that this was the case, or nearly the same percentage that was dissatisfied with the current choice of candidates. Moreover, 42 percent of adults felt that election campaigns in California were getting worse in terms of ethics and values. It is also noteworthy that only about one in 10 adults thought that campaigns in California were getting better. In addition, the most informed residents—that is, those most likely to vote—were even more negative about the quality of campaigns. Nearly half (49%) of likely voters said that elections in general had gotten worse in California, and over half (54%) said that elections had gotten worse in terms of ethics and values. Evidence of growing dissatisfaction with politics was apparent well before the 2002 elections. Large and increasing proportions of Californians have reported that partisan politics and politicians were problematic for the state (Table 2.3). In 1983, seven in 10 respondents 17 Table 2.2 Californians’ Perceptions of Election Trends California elections have gotten better, worse, or stayed the same in the last 10 years. % Saying Better 13 % Saying Worse 37 % Saying Same or Don’t Know 50 California elections have gotten better, worse, or stayed the same in terms of ethics and values in the last 10 years. 12 42 SOURCE: PPIC Statewide Survey (November 2002). NOTE: N = 2,000. 56 Table 2.3 Californians’ Attitudes Toward Politics and Politicians, 1983 and 1999 Politics today is too partisan; the political parties are continually at odds to such an extent that government and society suffer. % Who % Who Agreed in Agreed in 1983 1999 70 76 Politicians generally are self-serving and only looking out for themselves. SOURCE: Field Poll (1999). NOTE: N = 1,005. 34 41 said that politics was too partisan and that the political partisanship was damaging government and society. In 1999, more than three in four respondents (76%) agreed with this view. Similarly, although only a third of respondents in 1983 said that politicians are self-serving and look out for themselves, over four in 10 Californians agreed with this assessment in 1999. Overall, the current results and recent trends paint a bleak picture of California voters, who are expressing a general discontent with their 18 state’s election process. Voters’ unhappiness is reflected in both public opinion and public action, as registered voters not only complain about the political process but also choose not to participate in statewide elections. The sharp decline in voter turnout in November 2002, in combination with the highly critical statements by residents about the election then in progress, suggests that something has gone noticeably wrong with elections in California. Perceptions of Political Campaigns Political campaigns play an important role in a democratic society. They serve the dual purpose of informing voters about the candidates and motivating voters to go to the polls to cast a ballot for that candidate. When candidates spend more money on their campaigns—assuming that they do so effectively—they should become more familiar to voters. However, just because a campaign reminds people of an impending election does not necessarily mean that voters will turn out to vote. Some voters may take in this campaign information and then decide not to vote for anyone. The 2002 California election is a good case on this latter point. The results of the survey we conducted in the week before the November 2002 election suggest that Californians did not lack exposure to the messages of the candidates running for the governor’s office (Table 2.4). Three in four people said that they had seen the candidates’ campaign Table 2.4 Californians’ Perceptions of Advertising by Gubernatorial Candidates % Who Saw an Ad by a % Who Found Ads Candidate for Governor Helpful in Making in the Last Month a Voting Decisiona All adults 75 33 Registered voters 78 30 Likely voters 80 26 SOURCE: PPIC Statewide Survey (November 2002). NOTE: N = 2,000. aIncludes respondents who saw ads by a candidate for governor. 19 advertisements on television. Moreover, eight in 10 of those who were most likely to vote indicated that they had seen such advertisements. Although Californians may have been exposed to campaign advertising, the advertisements did not appear to provide them with the information they wanted the most—information that would help them decide how to vote. Only one in three Californians (33%) thought that the advertisements by the candidates were helpful in making a decision about how to vote in the November election. Even fewer among those who were most likely to vote in the upcoming election—and who tend to be more politically informed—found the campaign advertisements helpful. Just one in four likely voters said that the television advertisements run by the candidates for governor helped in making a decision about how to vote. Although voters may not receive all of their political information from campaign advertisements, evidence from the 2002 election suggests that there could be a link between the lack of useful information in the political advertisements and voters’ indecision about the candidates in this election. Certainly, the evidence suggests that the quantity of information is not a determining factor in the size of the undecided vote. In a survey taken the month before the 2002 gubernatorial election, over half of respondents said that they were paying close attention to the campaign, yet nearly one in four said that they did not yet know how they would vote (Table 2.5). Moreover, in our survey conducted during the week before Election Day, nearly six in 10 Californians said that they were paying close attention to the campaign. However, despite this attention, nearly one in four voters were still unsure of how they would vote in the election. On Election Day itself, a Los Angeles Times exit poll reported that 20 percent of those leaving the polls had made a decision about how they would vote on the day of the election itself. Many others who are not reflected in exit polls apparently decided not to vote at all. By way of contrast, in a PPIC Statewide Survey taken the month before the November 1998 gubernatorial election, fewer than half of the respondents in that survey said that they were paying close attention to the race for governor. But less than 10 percent of Californians said that they did not know how they would vote for governor in the November 20 Table 2.5 Californians’ Attention to Campaigns and Voting Uncertainty % Who Reported % Who Reported Not Paying Close Attention Knowing How They to the Campaign Would Vote for Governor October 1998a October 2002b November 2002c 47 53 59 9 24 23 SOURCES: PPIC Statewide Survey (October 1998, October 2002, November 2002). aN = 2,005. bN = 2,007. cN = 2,000. 1998 election. Despite the fact that fewer people were following the last gubernatorial election than the 2002 election, fewer people were undecided in 1998 than in the more recent matchup. Complaints about campaign advertising are not new to politics, but our findings are noteworthy because they reflect a relationship between increased political information and increased voting indecision. Some campaign advertising, rather than providing potential voters with information that leads to ballot choices, may instead provide them with information about candidates that may lead to difficulty in choosing a candidate that appeals to them. For some, the decision may be to not vote, given the candidate information that is provided. In sum, low turnout in the 2002 California gubernatorial election cannot be explained by lack of public exposure to the candidates’ campaigns or lack of knowledge about the upcoming elections. Rather, low turnout may reflect a response to the information about the candidates that was provided during the governor’s race. Low Turnout and More Initiatives: Confirming Evidence of Campaign Dissatisfaction The public’s pessimism about California politics has been increasing steadily over the last several years. We argue that this growing 21 dissatisfaction has led to declining voter participation at the polls on Election Day. Another change in California elections—an increasing reliance on the citizens’ initiative process to make state laws—offers further support for the hypothesis that voters’ disaffection with politicians has increased. The 2002 gubernatorial race between Gray Davis and Bill Simon was noteworthy for its high level of campaign expenses and the negative tone set by both candidates. In the final election, Davis defeated Simon, but he won with only 47 percent of the vote and despite the fact that 1.7 million fewer people voted for him in 2002 than had voted for him in 1998. Simon’s campaign failed to attract potential voters to the polls as well. In the meantime, voters have been increasingly exposed to and enthusiastic about their ability to participate in the “direct democracy” process afforded them by the state constitution—the recall, the referendum, and the citizens’ initiative. In our PPIC Statewide Surveys over the past few years, voters have consistently expressed a high level of satisfaction with the initiative process. Moreover, many voters say that they would prefer that voters make the important decisions for the state through ballot measures (Baldassare, 2002). California state politics and policymaking today rely to a large extent on the initiative process, which allows groups outside state government to make public policy by placing potential laws and constitutional amendments directly on the state ballot for voters to enact or reject. State initiatives may cover a range of topics, from cigarette taxes to restrictions on the sale of horsemeat. However, the initiative process has recently shifted its focus, with an increasing number of statewide initiatives attempting to impose external controls on state government itself (Table 2.6). These initiatives include proposals to impose term limits, to change the primary system, to identify or punish incumbents for certain behaviors, and to restrict campaign fund-raising and expenditures. In the early 1980s, these kinds of initiatives were rarely proposed and even more rarely qualified for the ballot. In the early to mid-1990s, however, the number of initiatives with this focus, both proposed and qualified, increased dramatically (see Baldassare, 2000, 2002; Schrag, 1998). 22 Table 2.6 Citizens’ Campaign Reform Initiatives in California, 1980–2002 Total Number of Initiatives Proposed for Qualified for Qualified for Year the Ballot the Ballot the Ballot 1980 0 0 3 1982 1 0 6 1984 2 1 12 1986 1 1 7 1988 2 0 16 1990 5 1 18 1992 4 1 4 1994 7 0 6 1996 3 3 16 1998 6 2 13 2000 7 2 15 2002 3 1 7 SOURCES: California Attorney General (2002); Silva (2000); University of California Hastings College of Law (2002). NOTES: The total number of initiatives proposed and qualified includes only those proposed by citizens and not those proposed by the legislature. Initiatives counted as related to campaign reform include proposals that attempted to place external controls on government—specifically, campaign finance reform, term limits, campaign advertising disclosure, changes in primary laws, conditioning reelection on balancing the budget, designation of incumbents on the ballot, changes in registration laws, and creating a “none of the above” ballot option. During the 1996 election cycle, an unprecedented three initiatives designed to control legislative behavior appeared on the ballot—one on term limits and two on campaign contribution limits. No more than one similar initiative had appeared on the ballot in any prior election cycle. Since that time, there has been no election cycle where at least one of these kinds did not appear on the ballot. Even larger numbers of related initiatives were proposed but failed to qualify for the statewide 23 ballot, with the number of proposals increasing sharply during the 1990 election cycle. Since that time, the number of proposed initiatives that attempt to control the behavior of state government has never dropped to the levels of the 1980s. When citizens feel connected to government and in control of government activities, there is little need or desire to impose political controls beyond the normal process of representative democracy—voting candidates in and out of office. However, as people become increasingly disconnected from government and their elected officials, the desire to use the tools of direct democracy to impose structural constraints on the executive branch and legislatures grows. The increasing efforts to impose these external constraints over the past decade reflect the feeling that government pays limited attention to the desires of constituents. In the week before the 2002 election, when Californians were asked whether they felt that having elections made government pay attention to what people think, over seven in 10 respondents (71%) said that they felt the elections made government pay either a great deal (31%) or a fair amount (40%) of attention to what people think. But people who were asked how much attention government actually pays to what people think were much less positive about the influence of their views. Over half (53%) said that they felt that government paid not too much attention, or no attention at all, to what people think. The increasing numbers of initiatives designed to impose restrictions on the activities of government, both in office and during campaigns, are consistent with this belief that government is inattentive to the public. What Happens When Candidates Focus on Negative Campaigning? Much of the dissatisfaction that Californians say they feel with political campaigns today is derived from their dislike of a particular type of campaign activity—referred to by political experts as “negative campaigning.” Although people recognize that campaigns involve candidates criticizing their opponents as well as advocating their own election, in the most recent election they felt that this criticism had reached unsupportable levels. 24 Our PPIC Statewide Survey in November 2002 asked half of respondents whether they felt that political candidates should be critical of their opponents and the other half of respondents whether they felt that Gray Davis and Bill Simon, the major party candidates for governor, should be critical of each other (Table 2.7). Respondents who were asked the more general question were split evenly (47% each) between agreeing that criticism was important to learn the strengths and weaknesses of each candidate and disagreeing because campaigns are already too negative. In contrast, a large majority (58%) of those who were asked whether Gray Davis and Bill Simon should be critical of each other said that they should not be, because campaigns are too negative. It is interesting to note that only slightly more than a third of the survey respondents (36%) thought that Davis and Simon should be critical of each other. Candidates may be able to criticize each other without engaging in what is perceived to be negative campaigning. But the 2002 gubernatorial candidates failed to draw the line between criticism and negativity as each sought instead to destroy the credibility of his opponent. Earlier campaigns have not always suffered from the same problem. In a 1996 poll of state voters, 49 percent of respondents said that the Clinton-Dole campaign was “not too negative,” and 46 percent said that it was either somewhat or very negative (Field Poll, 1996). Table 2.7 Californians’ Perceptions of Negative Campaigning Should political candidates be critical of their opponents? Yes, It Is Important to Know the Strengths and Weaknesses of No, Campaigns Each Candidate Are Too Negative 47% 47% Should Gray Davis and Bill Simon be critical of each other? 36% SOURCE: PPIC Statewide Survey (November 2002). NOTE: N = 2,000. 58% 25 These results are consistent with general feelings about whether political candidates should be critical of their opponents, suggesting that feelings about the 1996 campaign simply reflected underlying assessments about the value of candidates criticizing each other rather than a distinctive feature of that particular campaign. The same could not be said about the 2002 gubernatorial race. Measuring the Influence of Attack Advertising: Results of a Survey Experiment Campaign consultants argue that negative campaigning is often necessary. They claim that voters view the candidates who are attacked more negatively, allowing the attacking candidates to increase their chances of winning. It is generally perceived to be necessary as a way to salvage the reputations of candidates who have already been attacked by their opponents. However, very little effort has been made to measure how these attacks affect people’s perception of both the attacking candidates and those who are attacked. The PPIC Statewide Survey in November 2002 attempted to test constituent responses to attack ads by asking respondents to rate their feelings about candidates after hearing that they had either run attack ads or responded to attack ads. The examples used in the survey were based on the actual accusations and actual responses of both Davis and Simon during the 2002 gubernatorial campaign. Each respondent was randomly assigned to hear about a different kind of attack ad, and each could have been asked about his or her perceptions of either the attacking candidate after hearing the ad, or the attacked candidate after hearing the ad, but not about both. Both Davis and Simon attack advertisements were used as examples. This series of survey questions made it possible to measure the effect of running attack advertisements on a candidate’s popularity and to measure the effectiveness of different possible responses to attack advertisements. The candidate who was attacked could claim that the attacker was mudslinging, could make a counterclaim, or could deny the charges. At various points in the 2002 gubernatorial campaign, the candidates did all of these things. As a final follow-up question, respondents were asked 26 how they would have felt about the attacking candidate had he chosen not to run the attack ad, or the attacked candidate had he chosen not to respond to the attack ad. For instance, respondents were asked one of the following questions: Option 1: In this campaign, a Gray Davis advertisement claimed that Bill Simon had engaged in fraudulent business dealings, and in response Simon said that Davis was engaging in mudslinging and negative campaigning. Did this exchange make you feel more positive or more negative about Bill Simon? Option 2: In this campaign, a Gray Davis advertisement claimed that Bill Simon had engaged in fraudulent business dealings, and in response Simon said that Davis was engaging in mudslinging and negative campaigning. Did this exchange make you feel more positive or more negative about Gray Davis? Option 3: In this campaign, a Gray Davis advertisement claimed that Bill Simon had engaged in fraudulent business dealings, and in response Simon said that Davis is a “toll-booth” governor who makes policy only in exchange for campaign contributions. Did this exchange make you feel more positive or more negative about Bill Simon? Option 4: In this campaign, a Gray Davis advertisement claimed that Bill Simon had engaged in fraudulent business dealings, and in response Simon said that Davis is a “toll-booth” governor who makes policy only in exchange for campaign contributions. Did this exchange make you feel more positive or more negative about Gray Davis? Option 5: In this campaign, a Gray Davis advertisement claimed that Bill Simon had engaged in fraudulent business dealings, and in response Simon denied that he was guilty of any wrongdoing. Did this exchange make you feel more positive or more negative about Bill Simon? 27 Option 6: In this campaign, a Gray Davis advertisement claimed that Bill Simon had engaged in fraudulent business dealings, and in response Simon denied that he was guilty of any wrongdoing. Did this exchange make you feel more positive or more negative about Gray Davis? After being asked one of these questions, respondents were asked how a different strategy would have affected their opinions of the candidates: Option A: If Bill Simon had ignored Davis’s commercial instead of responding, would this have made you feel more positive or more negative about Bill Simon? Option B: If Davis had not made this claim about Simon, would this have made you feel more positive or more negative about Gray Davis? Surprisingly, all of the potential responses to attack advertisements were equally unpopular, including denying that the attack was true. The only response that was correlated with more positive rankings of the candidates was not making the attack or ignoring the attack. On average, 45 to 50 percent of respondents said that they perceived a candidate who attacked the other candidate, or who responded to the attack, more negatively. At the same time, 25 to 35 percent of the respondents said that it would make no difference to them. In contrast, approximately four in 10 respondents said they would have felt more positively about either candidate had they not attacked the other or had they ignored attack ads directed at them. Another 30 to 40 percent of respondents said it would make no difference, but this may reflect the fact that it was difficult to perceive a campaign without attack advertisements at this late stage in the campaign. It is important to note that both Davis and Simon were already perceived very negatively in the week before the election. Only 46 percent of respondents said that they viewed Davis favorably, and only 39 percent viewed Simon favorably, and any increase in favorable 28 opinion for either of these candidates would be noteworthy and significant in the final moments of the election. It may not matter to candidates whether their approval ratings decrease when they run attack ads, as long as the potential voters who disapprove are not already their supporters. If those watching the ads will not vote for the attacking candidate anyway, the cost to the campaign of alienating them is minimal. However, when we split the survey into respondents who already supported Davis, and respondents who already supported Simon, it was apparent that negative advertising was affecting the opinions of supporters as well as opponents (Table 2.8). Between 33 and 39 percent of Davis supporters viewed Davis more negatively for attacking Simon and for responding to Simon’s attacks. Between 35 and 39 percent of Simon supporters viewed Simon more negatively when he attacked Davis or responded to Davis’s attacks. Moreover, the nonsupporters of each candidate and those who did not Table 2.8 Perceptions of Negative Campaigning by Supporters and Nonsupporters Negative campaigning When Davis attacks When Davis responds He is By By perceived . . . Nonsupporters Supporters Neither More negatively More negatively 71% 72% 33% 48% 39% 50% When Simon attacks When Simon responds More negatively More negatively 58% 66% 39% 40% 35% 41% Avoid negative campaigning When Davis does not attack More positively When Davis does not respond More positively 37% 35% 47% 35% 51% 25% When Simon does not attack More positively When Simon does not respond More positively 36% 37% 45% 32% 47% 42% SOURCE: PPIC Statewide Survey (November 2002). NOTES: Respondents were asked randomized questions about how they would perceive candidates after viewing Davis and Simon advertisements with a range of characteristics. N = 2,000. 29 support either candidate were even more likely to perceive the candidates negatively for attacking and for responding to attacks. In contrast, even nonsupporters said that they would view Davis and Simon more positively had they chosen not to attack or not to respond to attacks. Between 35 and 37 percent of those who did not support Davis said that they would have viewed him more favorably if he had avoided negative campaigning. A similar percentage of those who did not support Simon said that they would have viewed Simon more favorably if he had avoided negative campaigning. In the 2002 campaign, however, neither candidate managed to profit from this potential goodwill, because both chose to run negative campaigns throughout the course of the governor’s race. The effect of attack ads on supporters indicates the political risks of running negative advertising. Many campaign consultants argue that supporters are not affected by attack ads and will not change their votes even if they dislike the tone of the campaign. But if even supporters are unhappy with candidates who engage in negative campaigning, the ability of negative advertisements to decrease the popularity of the candidate who has been attacked must be viewed against their ability to decrease the popularity of the attacker. In the 2002 gubernatorial race, both candidates concentrated their resources on attacking each other. Although voters were not able to effectively switch to a candidate who campaigned positively, they were able to skip voting altogether. In many cases, they did exactly that. One lesson learned from this survey experiment is that the winning candidate who spent more in this election—Gray Davis—took on risks in the amount of time and resources that were dedicated to attack ads. The strategy may have ensured his narrow victory. However, it came at the expense of a shrinking support base that made Davis vulnerable after the 2002 election. For one thing, a low turnout reduced the threshold for the number of signatures needed to qualify for a recall. Moreover, even some Democratic voters stayed at home or voted to remove Davis from office in 2003. 30 Correlates of Voter Dissatisfaction with Candidate Choices The public’s perceptions of the 2002 gubernatorial election were noteworthy in part because the people who were most alienated were the most knowledgeable about the political process that was taking place. We attempted to measure the factors that were related to expressions of dissatisfaction with the choice of candidates for governor. We considered a number of potential correlates of dissatisfaction such as attention to the campaign, interest in politics, several demographic variables, and such political factors as strong party identification. The results from a logit regression show that dissatisfaction increased significantly with level of attention to the campaign and with the level of interest in politics (see Appendix A). In general, increased political information is supposed to make potential voters more confident about their choices, just as providing more information through campaigns should increase voter confidence in the candidates and voter turnout on Election Day. However, the factors that seemed to make Californians unhappy with the 2002 gubernatorial race appeared to affect even the most informed and interested people. California campaigns appear to have descended to a level of communication that even active voters can no longer tolerate. Information on Election Campaigns Potential voters said that they did not receive the type of information that they needed to make a decision in the most recent California election. However, the information that they find most useful does not seem that difficult to ascertain. In all of the preelection surveys that we have conducted since 1998, Californians have said that, above all else, they would most like to know about candidates’ stands on the issues. In our survey just before the election, Californians chose learning about candidates’ stands on the issues as the most important thing that they could learn in a campaign, rather than information about candidates’ character, political parties, or past experience in office. Notably, however, this demand for information about issues was not met in the 2002 campaign. Although 69 percent of our survey respondents said 31 that they would like most to hear about the candidates’ stands on the issues, fewer than one-third were satisfied with the discussion of issues in the 2002 campaign (Table 2.9). In all of our preelection surveys, relatively few Californians said that they were satisfied with the discussion of issues during the campaign. The picture that emerges from the 2002 campaign is one of extensive voter frustration with the content of communication, which made voters less eager to support any of the candidates. Despite paying more attention to campaigns than in the past, a large share of Californians said that they failed to receive the information that they needed to choose a candidate for governor. Nearly half of registered voters and more than half of eligible voters were no-shows at the polls. On top of frustration with the content of the campaign, however, is frustration with the means that candidates used to deliver campaign messages. When asked how they prefer to learn about candidates, voters chose debates more than any other means of communication (Table 2.10). However, in the 2002 campaign, there was only one debate between the candidates for governor. Making the debate even less relevant, it was scheduled at noon on a Monday, when a large percentage of the working population was unable to watch it. How do Californians feel about the status quo of campaign communications today? Over four in 10 Californians said that they would view candidates who primarily used television advertising to get their message out to voters less favorably; another 20 percent said it would make no difference to them. Table 2.9 Californians’ Preferences and Satisfaction on Attention to the Issues Preference Percentage Would most like to know candidates’ stands on issuesa Satisfied with discussion of the issues in the campaignb 69 30 SOURCES: PPIC Statewide Survey (November 2002, October 2002). aN = 2,000. bN = 2,007. 32 Table 2.10 Californians’ Preferred Means of Candidate Communication Means of Communication Percentage Debates TV call-in shows Speeches Town hall meetings Mailers Door-to-door 34 16 16 15 8 3 SOURCE: PPIC Statewide Survey (November 2002). NOTE: N = 2,000. Californians’ feelings about the information provided by a largely ad-driven campaign were quite negative. In particular, when asked about their impressions of political campaigns in general, as well as the 2002 gubernatorial campaign in particular, an overwhelming number of respondents volunteered that the biggest problem with campaigns in general, and with the 2002 campaign specifically, was the emphasis on negative advertising. And by nearly any measure, the tone of the recent California gubernatorial campaign was overwhelmingly negative. It is important to note here that paid advertising is obviously not the only source of information on elections. The media play an important role in providing information on candidates and election events. We know from our current study that many residents closely followed the news stories about the 2002 governor’s election, but it is unclear if the attack ads had an influence on the types of election stories that the media covered. Nor is it clear if the dissatisfaction with the tone and substance of this campaign that was elicited in our surveys can be at least partly attributed to the media coverage of the attack ads and negative campaigning. Future studies should focus on the combined effects of the media stories and campaign information on dissatisfaction with campaigns and elections in California. 33 Conclusions The dissatisfaction that Californians felt with the most recent gubernatorial election is historically noteworthy and politically significant. Moreover, it appears to have extended beyond simple protestations of unhappiness. Lacking the information that they felt they needed to make a decision, and unhappy with the candidate choices and the means by which campaign messages were communicated, an unprecedented number of voters chose not to appear at the polls at all. However, it should be noted that the 2002 election is typical and not an aberration. The findings reported here are symptomatic of trends we found both before and after the Davis-Simon race—and are ever-present in the wake of the governor’s recall in the fall of 2003. The failure of political campaigns to satisfy the demands of potential voters has implications for the democratic system. Part of the purpose of any political campaign is to bring politicians seeking elected office in touch with constituents and to serve as a mandate for continuance of past policies or the creation of new ones. One expectation placed on candidates is that they will provide information that will allow voters to make those decisions, either directly through their own campaigns or indirectly by criticizing their opponents. The competition inherent in a campaign is designed to allow voters to hear at least two platforms, one of which will direct the politician who is elected. However, when both sides of a campaign provide information that voters find only irrelevant or off-putting, constituents have few options for protest other than withholding their votes. We find that candidates hurt their own images by campaigning negatively and providing inadequate information to voters. Equally important, attack campaigns, even beyond their effects on turnout, harm democratic legitimacy. Are feelings about campaigns correlated with trust in government and, specifically, does unhappiness with campaigns lead to decreased trust? Although the effects of lowered trust in government may be difficult to pin down, a government that does not enjoy the support of its constituents may have difficulty enacting policy, and may face increasing external controls such as term limits. In the case of California, there has 34 been an increasing trend for voters to circumvent the legislative process altogether through the citizens’ initiative process. Dissatisfaction with the recent gubernatorial election appears to have generated a new effort to impose external controls on officeholders, with the first successful movement in California to recall a sitting governor. We look for further evidence of a direct link between perceptions of political campaigns and feelings of political distrust in the next chapter. 35 3. Political Campaigns and Voter Distrust Political observers have long suspected that trends in campaign practices such as negative advertising might lower the public’s capacity for political trust. In the California context, the voters’ dissatisfaction with political campaigns that we noted in the previous chapter has mounted against a backdrop of declining trust in government. The assumption underlying many campaign reforms, ranging from campaign finance regulation to pledges of clean conduct, is that they will serve to regenerate trust in government. To date, however, there is no evidence that any political or election reforms that have been enacted have actually increased the level of political trust. The question that we address here is whether trust in government increases when candidates engage in more desirable campaign behavior and decreases when they engage in less desirable campaign behavior—that is, political practices that stray widely from public ideals or expectations about appropriate behavior for candidates for elected office. There are several reasons to study the link between political campaign practices and political distrust. First, given the frequency of election campaigns in recent years, there are many different types of campaigns to observe, and there is no shortage of cases that are negative in their tone and tactics. An analysis of political campaign trends thus offers a potential explanation for increasing levels of distrust in government. Next, a consideration of the effects of negative campaigns on political trust offers the opportunity to review whether some of the more practical claims about the consequences of a decline in political trust are valid. For example, are constituents who trust government less really more likely to limit legislative activities and oppose an expansion of government’s role? A range of potential consequences of low trust in 37 government has been discussed, but there is little known on whether and how much distrust in government has real policy and political implications (see Baldassare, 2002; Citrin, 1974; Miller, 1974; Moore, 2002). The decreasing trust in government inspired by negative campaigns and voter dissatisfaction with the election process have larger implications for the prospect of governance in California. Specifically, low trust in government may decrease democratic legitimacy and thus make it increasingly difficult to enact effective public policy. It may also encourage voters to make state laws on their own through the initiative process. The role of political distrust during the 2002 gubernatorial election in California is worthy of our careful attention. This election was a political event that received national attention for its record-setting spending on a campaign that was uniformly described as overwhelmingly negative and that generated critical comments about both major party candidates. In the final analysis, the record low voter turnout raises questions about the role of political distrust. Indeed, trust in the state and federal government was relatively low during the 2002 election. Moreover, trust in the state government declined further in the months after November 2002 leading up to the October 2003 recall election and, as of this writing, has yet to recover to the pre–November 2002 levels. These trends raise another issue—whether and how much trust is regained once it is lost. The Reasons for Public Distrust Trust in government depends on a variety of factors that can be loosely organized into three general categories: (1) assessments of current political leadership and the political system; (2) individual characteristics; and (3) levels of political engagement. Past research has raised questions about whether assessments of political leaders influence political trust in a significant way. Initial research on political trust found that individual characteristics, such as a person’s age and level of education, were the main determinants of trust. However, further research on perceptions of trust in government suggested that trust varies over time in response to political conditions. 38 After the Watergate scandal, Citrin (1974) argued that levels of political trust were correlated with feelings about political leaders, especially at the presidential level (see also Baldassare, 2002; Dionne, 1991; Lipset and Schneider, 1983; Pew Research Center for the People and the Press, 1998). Specifically, those who disapprove of political incumbents are less trusting than those who approve of incumbents. Researchers have found that individuals may base their trust in government at least partially on their assessments of features of the political system, such as campaigns and elections (Buchanan, 1996, 2000; Rahn, Brehm, and Carlson, 1999; Simon, 2003; Williams 1985). What does the experience of an actual election do to one’s level of political trust? Some research suggests that certain campaigns can increase political trust (Bartels, 2000; Bartels et al., 2000; Buchanan, 2000; Clark and Acock, 1989; Ginsberg and Weissberg, 1978; Rahn, Brehm, and Carlson, 1999). And equally, there is speculation that unattractive campaigns—specifically those that are negative or vacuous—depress trust in government (Buchanan, 1996; Simon, 2003). Some evidence even suggests that distasteful political campaigns lead the electorate to participate less frequently in politics (Ansolabehere and Iyengar, 1995; Rosenstone and Hansen, 1993). Certain campaign reforms, such as campaign finance reform, may thus increase political trust. In general, then, different elements of campaign and election experiences appear to be significantly related to political trust. As we have noted above, political trust varies with individual characteristics, either alone (Miller, 1974) or in conjunction with assessments of political leaders (Citrin and Green, 1986). Underlying political beliefs, sometimes related to social and economic characteristics, also influence individual perceptions about government. Some people may have learning experiences that lead to trust in government, leading to generational differences in trust that are traditionally measured by age (Hetherington, 1998; Miller and Shanks, 1996). Individuals may also feel more trusting because of their ideological or partisan agreement with political leadership (Hetherington, 1998). Finally, researchers have found that demographic characteristics such as race and ethnicity, education, income, and gender are correlated with different levels of trust in government (Hetherington, 1998). Latinos are more trusting than 39 those from other racial and ethnic backgrounds (Cohen, Baldassare, and Kaimowitz-Rodriguez, 2003), whereas those with more education may be more (Hetherington, 1998) or less (Finkel, 1985) trusting. In keeping with the assumption that those in agreement with, or profiting from, current economic and political conditions are more trusting, those with higher incomes are typically more trusting in government. Finally, some evidence suggests that individual levels of political engagement are correlated with political trust. Most of this research argues that exposure to politics and government will increase trust (Hetherington, 1998), although some researchers argue that the effect is reversed, and those who know more about government are less trusting (Bowler and Donovan, 2002). In addition, higher levels of political participation, such as voting or attending political meetings, have been correlated with higher levels of political trust (Finkel, 1985), particularly among respondents with less education. In our study, it is noteworthy that some of the assessments of trust in government were measured in the context of a heated political campaign—that is, the 2002 gubernatorial election in California. But it is important to note that this was also a time of highly negative perceptions about the state’s economy and fiscal conditions. At the same time, California and the rest of the nation were a year out from the September 11, 2001, terrorist attacks in New York City and Washington—an event that was initially followed by higher levels of reported trust in the federal government and a significant spike in presidential approval ratings. For a variety of political and economic reasons, then, despite the fact that people typically trust their state government more than they trust the federal government (Bowler and Donovan, 2002), by the fall of 2002, at the time of the gubernatorial election campaign, Californians’ trust in state government had dropped to levels below their trust in federal government. Survey Evidence on Distrust in Government In this section, we analyze the results of the PPIC Statewide Survey that was conducted in the week before the November 2002 California gubernatorial election. This random-digit-dial telephone survey of 2,000 California adults interviewed in English or Spanish included subgroups 40 of 1,526 respondents who claimed to be registered voters and 1,025 who were deemed through their responses to be likely voters in state elections. Along with a series of questions designed to measure perceptions of candidates and campaigns (discussed in Chapter 2), all of the respondents were asked about their level of trust in state and federal government, their interest in politics, and their level of political engagement. These particular survey questions are the focus of the analysis in this section. The survey asked two specific trust questions. The first referred to trust in federal government, using the standard American National Election Studies format noted below. The form of the trust in state government question was similar to that of the trust in federal government question, but it referred to the government in Sacramento rather than the government in Washington. Responses indicating trust or lack of trust were coded the same as the responses to the trust in federal government question. Federal government: “How much of the time do you trust the government in Washington to do what is right? Always, most of the time, or only some of the time?” State government: “How much of the time do you trust the government in Sacramento to do what is right? Always, most of the time, or only some of the time?” In the analysis below, those who said that they trusted government all of the time or most of the time were considered to be trusting, and those who said that they trusted government only some of the time, none of the time, or they did not know were considered not to trust government. Although a fourth category, “none of the time,” was not provided to respondents, it is included in the response categories because people asked about their trust in government in recent years frequently volunteered this answer. In the current survey, 5 percent volunteered “none of the time” for the federal government, and 4 percent volunteered “none of the time” for the state government. Only 1 percent said that they had no opinion about the state or federal governments. 41 The PPIC Statewide Survey has been asking about trust in federal government nearly every year since the survey’s inception in 1998. The trust in state government question, however, has had a shorter though more intense recent history. The trust in state government question was first introduced into a PPIC Statewide Survey in January 2001 as part of a survey series following the 2000 presidential election. The first instance of this question preceded the California energy crisis, which would be expected to decrease trust in state government, as well as the September 2001 terrorist attacks, which tended to increase trust in all levels of government, at least temporarily. In the past two years, questions about trust in both the state and federal government have been asked every few months. Two questions were used to consider respondents’ political assessments. The first asked how favorably respondents perceived the Democratic incumbent governor, Gray Davis, who was running for a second term of office against Republican challenger, Bill Simon, using the following answer categories: Very favorable, somewhat favorable, somewhat unfavorable, very unfavorable, and no opinion. A second set of questions asked whether political campaigns in California had been getting better or worse in recent years, with the following wording asked of a random half-sample of the 2,000 adults: “More generally, overall do you think that election campaigns in California have gotten better in the last 10 years, gotten worse, or stayed about the same?” “More generally, overall in terms of ethics and values, do you think that election campaigns in California have gotten better in the last 10 years, gotten worse, or stayed about the same?” These questions were recoded into two response categories for the purposes of the analysis that follows in this chapter. We contrast those with a favorable view of Davis (i.e., very favorable or somewhat favorable combined) against all other answers (i.e., somewhat unfavorable, very unfavorable, no opinion). The perception that political campaigns in California were getting worse was contrasted against all other answers 42 (i.e., getting better, staying the same, do not know). As for the two versions of the campaign quality question (i.e., overall and in terms of ethics and values), because the results were similar using the different wording, we combined the responses for our analysis into one question that measures the perceptions of campaign quality in the past 10 years. Our measures of individual characteristics—including age, sex, race, income, education, and party identification—were thus taken from a range of questions on demographic and political characteristics. Other questions in our analysis measured political engagement. These included questions about whether respondents had engaged in any of a series of political activities, whether they had seen advertisements by the candidates for governor, how much attention they paid to politics, and how much attention they paid to news about the upcoming election. Finally, respondents were asked whether they could correctly identify the position of three political figures—U.S. Senator Dianne Feinstein, Secretary of State Colin Powell, and the Prime Minister of the United Kingdom, Tony Blair. These questions were designed to measure the respondents’ levels of political knowledge and attentiveness to politics. The questions on political activities were derived from the Roper organization’s series on political participation. Specifically, we asked whether the respondents had, in the last six months, written to a political representative, attended a political rally, attended a meeting on political affairs, signed a petition, worked for a political party, given money to a political party or candidate, or been a member of a political organization. These measures were summed to an eight-point scale ranging from zero activities to all seven activities. Measures of attention paid to politics and attention paid to the news spanned a four-point range from little or no interest to a great deal or a fair amount of interest. Finally, the measure of political knowledge ranged from zero (none of the three political figures correctly identified) to three (all three correctly identified). The preelection survey questions made it possible to test expectations for the three general hypotheses that attempt to explain what factors affect trust in government. 1. Political assessments: Trust in state government is positively correlated with support for the incumbent governor, Gray Davis. Trust 43 in state government is negatively correlated with feelings that election campaigns are getting worse. 2. Individual characteristics: Trust in state government is negatively correlated with being a strong Republican or an Independent and is positively correlated with identification as a strong Democrat. Both women and Latinos are more trusting of state government, whereas younger residents are less trusting of state government. 3. Political engagement: Residents who pay attention to the news, who have a greater interest in politics, or who know more about politics are more trusting of state government. Those who have seen the paid political advertisements by the candidates for governor, Gray Davis or Bill Simon, are less trusting of state government. Trends in Distrust in State Government Californians’ trust in state government declined during the 2002 election campaign and has not returned to previous levels as of this writing. The evidence from the PPIC Statewide Survey in November 2002, as well as past studies of trust in government in California, suggests that campaign perceptions are related to attitudes toward government. Specifically, negative assessments of election campaigns are correlated with lower levels of trust in government. Moreover, political assessments of incumbents and campaigns appear to provide stronger correlations with declining trust than either individual characteristics or political engagement according to our statistical analysis below. Californians’ trust in state government has been declining over the past three years, whereas trust in the federal government has remained relatively stable. This distinction between state and federal trust suggests that although our survey questions asking about trust in these two levels of government are similar, they are effectively measuring two different phenomena rather than an underlying level of trust in all levels of government. In addition, it suggests that some phenomena specific to California may have recently changed the levels of trust in state government. It is possible that the 2002 election was one of the major factors contributing to this decline. 44 Trust in the federal government expressed by California residents has increased since the first PPIC Statewide Survey in April 1998. In the spring of 1998, only about one-quarter of Californians said that they trusted the federal government always or most of the time (as shown in Table 3.1). By September 1998, trust in the federal government had increased to approximately one-third of all respondents and remained at or near that level through the end of 2000. In January 2002, after the September 11th terrorist attacks, the percentage of those saying that they trusted the federal government leaped to 46 percent of all respondents. Although trust in the federal government declined from this high several months later, it still remained at a higher level than in the past. By August 2002, during the 2002 election campaign, almost four in 10 respondents said that they trusted the federal government (37%), and trust remained at this level both during and after the state elections. Californians’ trust in the federal government has shown a slow but steady Table 3.1 Californians’ Trust in the Federal Government, 1998–2004 % Who Trust Federal Government Always or Date Most of the Time April 1998 May 1998 September 1998 December 1999 January 2000 October 2000 January 2002 August 2002 November 2002 February 2003 February 2004 26 24 33 32 33 33 46 37 37 37 30 SOURCES: PPIC Statewide Surveys (1998–2004). NOTE: N = approximately 2,000 per survey wave. 45 increase over the past five years and into 2003, but it declined recently in February 2004. In contrast, trust in state government has been moving in the opposite direction (Table 3.2). For the years measured, state government trust was at its peak in 2001 and early 2002, with nearly half of all respondents saying that they trusted state government (47%). Notably, these years were difficult for the state government and for the reputation of the incumbent governor, Gray Davis. In 2001, California experienced a massive energy crisis, which was blamed on the state’s energy deregulation plan and subsequent mismanagement by Governor Davis. And from 2001 to 2003, California experienced economic recession, state budget problems, or both. Despite these issues, trust in Table 3.2 Californians’ Trust in State Government, 2001–2004 Preelection January 2001 January 2002 % Who Trust State Government Always or Most of the Time 47 47 November 2002 election season August 2002 November 2002 37 35 Post–November 2002 election February 2003 June 2003 September 2003 October 2003 36 34 28 27 Post–October 2003 recall January 2004 May 2004 27 32 SOURCES: PPIC Statewide Surveys (2001–2004). NOTE: N = approximately 2,000 per survey wave. 46 state government first showed a decline during the 2002 election season. By August 2002, fewer than four in 10 Californians said that they trusted the government in Sacramento (37%). In the following months, trust remained at or below this level, with only 34 percent of Californians saying they trusted state government by June 2003. During the recall election, trust in the state government declined further, and it has not rebounded to the levels seen in the November 2002 preelection survey as of this writing. Some argue that it is difficult to attribute any changes in trust in state government coincident with a political campaign to anything but the campaign itself. Rahn, Brehm, and Carlson (1999) argue that changes in political trust during campaigns should be attributed to campaigns rather than to other factors because there typically are no other equally significant political events occurring at the same time. Although California had experienced a number of political problems in recent years—notably a serious budget deficit and an electric power crisis—in the fall of 2002 the major political event was definitely the gubernatorial election. Other statewide and federal offices were in contention at this time—including a U.S. Senate race and Congressional and state legislative races—but there is no question that the gubernatorial race drew the most attention from the media. Although it could be argued that people feel more negatively about campaigns because they are less trusting, the decline in trust that occurred during the 2002 election campaign argues that the election campaign affected political trust rather than the reverse. For instance, only the ratings of trust in state government changed over time, whereas trust in the federal government remained stable. As a result, a correlation between trust in state government and negative feelings about political campaigns should largely reflect the effect of campaigns on political trust rather than the effects of low trust on feelings about political campaigns. Even if other factors may be driving the loss of confidence in state government, the causal relationship appears clear in the case of state government. 47 Correlates of Distrust in State Government To examine the potential relationships between various factors and trust in state government, we conducted a logit regression analysis for all survey respondents, using independent variables that measured political assessments, individual characteristics, and political engagement. The numerical results are reported in Appendix B. Overall, we found that political assessments, of both the incumbent governor, Gray Davis, and perceptions of the campaign, were significantly correlated with trust in state government. In addition, certain individual characteristics, such as party identification and ethnicity, were also related to political trust. In contrast, only one measure of political engagement—the level of interest in politics—was significantly related to trust in state government. We discuss these findings in the remainder of this section. It is important to note that both of our measures of political assessments were significantly correlated with trust in state government. One of the strongest predictors of trust was the perception of political leaders; individuals who had a favorable opinion of Gray Davis were much more trusting of state government. In contrast, those who felt that political campaigns in California had been getting worse were significantly less trusting. A few measures of individual characteristics were also correlated with trust in state government. When we accounted for attitudes toward the Democratic incumbent governor, Gray Davis, both Independent voters and strong Republicans expressed less trust than others in government. In contrast, measures of conservative political ideology—gender, age, education, and income—were unrelated to lower levels of trust. Many researchers have argued that there is a generation gap in trust, with younger residents trusting in government less than older generations, but this did not appear to be the case during the 2002 gubernatorial election campaign. As for race/ethnicity differences, Latinos were more trusting of state government than other residents were. The latter results are consistent with past research on Latinos in California, which argues that Latinos consistently feel more positive about both federal and state government than do residents of other racial/ethnic backgrounds (Baldassare, 2002). 48 Finally, as noted above, only one measure of political engagement— the level of interest in politics—was correlated with trust in state government. In this case, residents who were more interested in politics reported higher levels of trust. Other measures of engagement, including attention paid to the news, exposure to campaign advertising, and the level of political knowledge, were not significantly correlated with trust in state government. Despite our expectations that these other measures would be correlated with political trust, when other factors were considered they did not appear to be relevant. Although the opinions of all residents clearly matter, political campaigns often target likely voters rather than the population at large. If feelings about political campaigns are less important to likely voters than they are to the general population, candidates may not feel that the loss of trust is relevant to their decisions about how to campaign. The observation that those who are more interested in politics were more trusting suggests that likely voters may not be influenced by the same factors that affect the population at large. Were likely voters less trusting in response to negative perceptions of the gubernatorial campaign? We ran an identical logit regression model only for likely voters, and the numerical findings are reported in Appendix C. The findings indicate that political assessments had the same effect on trust for likely voters that they did for California adults in general. However, measures of individual characteristics and political engagement were less significant for likely voters’ reports of their trust in government. In sum, the political assessments of the candidates and elections were correlated with likely voters’ levels of trust in government the same way that they were with all adults. Those who felt more favorably toward Gray Davis were significantly more trusting in state government. Those likely voters who felt that campaigns were getting worse, rather than staying the same or getting better, were significantly less trusting. Individual characteristics, however, were much less relevant to political trust for the population of likely voters than they were for all adults. The only measure that was significantly correlated with political trust for likely voters was being registered to vote as an Independent or “decline to state.” As in the general population, Independents felt significantly less trust in state government than those who identified with 49 political parties. However, other measures, such as a strong Republican Party affiliation, or identifying as Latino, were not correlated with trust for likely voters. Finally, none of the measures of political engagement were significant predictors of political trust for likely voters. Some of this effect may result because likely voters in our survey sample are partially identified on the basis of the question measuring interest in politics. The effects of political engagement measures may be more evident in the general population than among voters (Finkel, 1985). However, even in the general population, only one of these measures had any noticeable effect on political trust. Overall, our expectation that assessments of political leaders and the political system would influence the levels of trust in government seems to be borne out in the analysis of our November 2002 survey. In contrast, only certain measures of individual characteristics appeared to affect political trust in a significant manner. The level of a respondent’s political engagement on a number of dimensions did not appear to affect trust in government much, although the level of interest in politics had some effect. Certainly, many other factors were driving the levels of distrust toward historical lows in the months leading up to the November 2002 election. In the California context at this particular time, the unattractive campaign for governor appears to have contributed to a decline in trust in state government. Moreover, trust in state government remained at its new low after the close of the 2002 campaign, and it was to drop even further as California voters confronted the historic events of 2003—the largest state budget deficit in history and the first recall of a governor. Conclusions Most Americans report having more trust in their state government than in the federal government, but California residents have shown a pattern of trusting their state government less than, or at least no more than, they do the federal government. This trend became evident during the 2002 governor’s race. The empirical evidence presented in this chapter demonstrates that campaign perceptions are significantly related to the public’s level of 50 trust in state government. The relationship remains after accounting for individual characteristics and various political factors. At the close of the 2002 governor’s race, Californians with more negative perceptions of the campaign also had lower levels of trust in state government. Is trust in government relevant to the complex issues involved in governing? Some say that it is not, arguing that voter apathy and disinterest can actually lead to a more politically stable environment and that low trust in government is unrelated to the ability to govern effectively (Berelson, Lazarsfeld, and McPhee, 1954; Dahl, 1956; Moore, 2002). However, others suggest that political trust and legitimacy are important factors in governance. The observable effects of political distrust can range from legislative gridlock, to difficulty in passing public policy measures at the ballot box, to efforts to impose political reforms on candidates and elected officials (Hetherington, 1998; Teixeira, 1992). Although it is difficult to link the decline in trust in state government directly to any particular set of policy outcomes, the recent history of California politics suggests that the more dramatic predictions about the effects of the loss of political trust may have some external validity. The current state of distrust in California politics may be a major contributing factor to the demand for structural change through the direct democracy process, such as citizens’ initiatives that call for legislative term limits, fiscal constraints on elected officials, and administrative change in the case of the Davis recall. At the very least, the current level of distrust provides a steady foundation for political change. What is even more striking in the California example reported here is the further drift downward in trust in state government after the November 2002 election. In other words, the decline in trust that took place before the election was not a temporary phenomenon. Moreover, the lack of recovery to earlier levels of trust in state government several months after the October 2003 recall is noteworthy, especially because Governor Arnold Schwarzenegger has high job approval ratings (64% of all adults; 69% of likely voters) according to the PPIC Statewide Survey of May 2004. This trend raises fundamental questions of what it takes to repair trust when the public has lost a high degree of confidence. 51 California residents, when asked about the current political process in the state during the course of the election, expressed strong feelings that political reform was needed. Many said that campaigns were too negative and that elections were too expensive. In the absence of desirable political campaigns and candidates, the state’s politics may suffer negative consequences from the distrust and alienation that was evident in November 2002 that extend far past a single election. For instance, this could result in less interest in registering to vote, a declining turnout of voters in elections, and voters’ attraction to citizens’ initiatives that limit the powers and authority of the state’s elected officials. In the concluding chapter, we consider the types of political and election reforms that have public support, and we assess their chances for addressing the current trends of negative campaigning, low voter turnout, and distrust in government. 52 4. Conclusions on Public Support for Campaign Reforms Given that California voters are unhappy with the way that campaigns and elections are run in the state today, the question is what, if anything, would they like to do about it? Moreover, are the state’s voters so dissatisfied with the political status quo that they would be willing to increase their taxes to pay for campaign finance reforms? These are the types of public policy preferences that we explore in this final chapter. For many of the possible campaign reforms, there could be legal resistance for government regulation because of first amendment speech and association rights. So any election reforms one might want to propose have to avoid constitutional problems and court challenges by politicians, parties, and consultants who think that they benefit from the ways in which campaigns are run today. For example, our survey indicates that voters are dissatisfied with negative advertising and personal attacks. An outright ban on negative advertising might be appealing to a majority of California voters, but it would certainly not be constitutional and would surely face legal challenges by those who make a living off political commercials. Among a smaller set of feasible political reforms, we tested public attitudes toward four categories of measures: campaign finance restrictions, promoting more spontaneous and personal forms of candidate-voter contact, public disclosure, and voluntarily adopted codes of campaign conduct. We review our findings for each of these reforms below. Most of these reforms are not new. They have been the subject of lengthy debate and careful discussion in California, in other states, and in national commissions. In many instances, the political will to 53 implement these changes has been lacking. The political establishment has doubted the public’s interest in campaign and election reforms. Our results indicate that—contrary to conventional wisdom of indifference toward the political process—the public does want to see substantive change in campaigns and elections. California voters most strongly favor more debates and would like to see the candidates adopt some codes of conduct. It is somewhat surprising that in the light of past polls and elections, Californians in the heat of the 2002 governor’s race, and especially likely voters, were even open to public financing as an option in the future. Apparently, many voters are even willing to pay more to make improvements in what they see as an election process that is broken and badly in need of change. In this chapter, we learn that many voters are eager to embrace proposals to improve California elections (see Baldassare, Cain, and Cohen, 2003a, 2003b; Public Policy Institute of California, 2003). Campaign Finance Reform The issue of campaign finance was prominent in the 2002 governor’s race. Gray Davis’s aggressive fund-raising tactics made more than a few Democrats uneasy and raised questions about his ethics in the media, eventually causing Davis to cancel some events and slow down the pace of money-raising. Against the backdrop of federal action on the Bipartisan Campaign Reform Act, Californians might question the relatively permissive regime we have in this state. Before the passage of Proposition 34, the state had only disclosure laws and no contribution limits or public financing. Proposition 34 introduced relatively high limits on individual and group contributions but not public financing. Moving to new campaign finance reforms may be difficult according to our survey just before the November 2002 election. To begin with, voters are conflicted when asked what they think contributors should get for their money. About 49 percent of likely voters believe that candidates should do nothing or merely say thank-you for the contributions they get, whereas 39 percent of likely voters think that candidates should give contributors either the policies they want or more access to themselves. To put it another way, there is a real divide 54 between the large proportion of voters who think that the current system of campaign finance is ethically acceptable and the substantial percentage of voters who think it is not. In recent years, some wealthy candidates have tried to turn the public’s distrust about candidates’ accepting contributions to their advantage by claiming that their personal wealth frees them from special interest ties. Apparently, this point of view is not shared by many of the state’s voters. Our survey indicates that fewer than one in five Californians thought that it was best for candidates to use their own money to finance their campaigns (i.e., the plutocrat’s virtue), whereas 38 percent thought that it was best to get money from supporters and 37 percent thought that money from the public was the best. And perhaps most surprising, given the fiscally conservative nature of Californians, 57 percent of likely voters thought that public financing was a good idea (Table 4.1). Conventional wisdom by the political establishment in California has been that public financing was a nonstarter. Among other reasons, why would people want to pay for speeches by candidates that they disagreed with or openly opposed in the heat of an election? But the experience of the 2002 election, and perhaps the fact that several states and some California cities have experimented with public financing Table 4.1 Californians’ Support for Funding Political Campaigns “Would you favor or oppose having a system of public funding for state and legislative campaigns in California, even if it cost taxpayers a few dollars a year to fund?” Favor Oppose Do not know % of California Adults 50 46 4 % of Likely Voters 57 39 4 SOURCE: PPIC Statewide Survey (November 2002). NOTE: N = 2,000. 55 schemes, seems to have made many voters receptive to this option for political reform.1 Candidate Debates Californians have been telling us in surveys in recent years that they are frustrated by the distance and superficiality of television-driven statewide contests. On Election Day in November 2002, many voters apparently voted with their feet by staying away from the polls. What would they prefer instead? Several questions in the survey deal with this topic. Clearly, voters would like a more intimate form of campaigning than is feasible above the local level in this state: Fifty-six percent of likely voters said that they would feel more favorably about candidates who campaigned primarily by going from door to door. Unfortunately, with Assembly, Senate, and Congressional districts of over one-half million constituents, and a statewide electorate of 15 million people, door-to-door campaigning is not a realistic option for all but the physically fittest of candidates. Although it is unlikely that voters will get to know statewide candidates more personally, it might be possible for them to witness candidates in more potentially spontaneous settings than the 30-second television commercials that they are providing to voters today. The dominant trend in modern campaigning is to program candidates to avoid mistakes and follow formulaic answers. The ideal for some political consultants would be to create the scripts, have their clients well rehearsed in private, and then have the candidates read them in public. Television advertisements rarely give much specific information about candidates and their positions. Settings where reporters get to ask unscreened questions of the candidates, or where there is the possibility that a candidate might look bad in a real-life interaction with actual voters, are avoided by many political consultants at all costs. Although some reporters get to ask unscripted questions in interview settings, _____________ 1In a more recent survey, majorities of state residents (57%) and likely voters (54%) said that they were opposed to spending a few dollars a year to support public funding for state and legislative campaigns (PPIC Statewide Survey, September 2004). 56 others are exposed to the candidates in only highly controlled environments. Even debates seem to pose “excessive” risks to some candidates. The incumbent governor agreed to only one debate, televised in a single media market in the middle of the day during the 2002 governor’s race. Although no one believes that the answers in debates are unscripted, debates do present a chance for large numbers of voters to compare the answers of the candidates on specific policy questions. When asked whether more debates in the 2002 gubernatorial contest would have made the election campaign better or worse, nearly two-thirds of likely voters said it would have made it better. When asked if they would favor or oppose an initiative that would require five prime-time broadcast gubernatorial debates, 60 percent said they favored the idea (Table 4.2). Thus, support for proposals that provide more contact with the candidates suggest yet another clear area of reform. It is unlikely that a measure that would require debate participation would survive a court test. A more defensible measure would be to set up a debate format and a commission to administer it and then hope that public expectations would induce participation in a series of highprofile debates and town hall forums by all of the major candidates. Table 4.2 Californians’ Support for Gubernatorial Debates “Would you favor or oppose an initiative that required candidates for governor to participate in a minimum of five prime-time publicly broadcasted debates?” Favor Oppose Do not know % of California Adults 60 33 7 % of Likely Voters 56 39 5 SOURCE: PPIC Statewide Survey (November 2002). NOTE: N = 2,000. 57 Public Disclosure Disclosure is a key component to a well-functioning electoral system. Voters need to have access to good information if they are going to make good choices. Although the media often ask for the fullest possible disclosure, candidates sometimes resist on the grounds that subjecting key pieces of personal information to public scrutiny seems like an unnecessary invasion of their privacy. Candidates who enter the public domain must necessarily surrender the privacy they enjoyed as private citizens. But does that mean that candidates have no right to privacy? Just because the press wants to have the material for a story does not mean that the information about a candidate necessarily serves an important democratic purpose. Where does one draw the line? The line has moved toward shrinking the zone of privacy for candidates. A good example is the controversy over Bill Simon’s tax forms during the 2002 California governor’s race. Having filled out the conflict of interest forms required by the state, Simon argued that since he did not have to release his complete tax forms, he would not. Members of the press argued that because Simon had no experience in public office, and was running for office as a businessman and outsider, the public had a right to know more about how he earned his money and how much he paid in California taxes. Moreover, a jury verdict on a business deal and an investigation by the Internal Revenue Service of his company’s affairs in a year of corporate scandals raised the media’s interest in his tax records. In the end, reporters were allowed to review Simon’s taxes in a highly restricted setting. We asked in our survey whether the respondent would feel more favorably, less favorably, or no different toward candidates who released their tax forms, their medical records, or information about campaign contributions as soon as they received such contributions. Given the furor over Simon’s taxes, the results are somewhat surprising. A majority of likely voters (53%) said that they would feel no different about candidates who released their tax forms, whereas about four in 10 said that it would make them look at the candidates more favorably (Table 4.3). An even larger percentage of likely voters (79%) said that they would feel no different about candidates who released their medical 58 Table 4.3 Californians’ Support for Public Disclosure No difference if candidate: Publicly released tax returns Publicly released medical records % of California Adults 53 77 % of Likely Voters 53 79 More favorable view if candidate: Publicly released a list of campaign contributors and amounts received as soon as candidates got them 66 SOURCE: PPIC Statewide Survey (November 2002). NOTE: N = 2,000. 71 records—an issue that came up in the 2000 presidential campaigns of John McCain and Bill Bradley—whereas only about one in five respondents said that this would make them look at the candidates more favorably. By far the more relevant piece of personal information for evaluating candidates seems to be the immediate disclosure of campaign contributions: Seventy-one percent of likely voters said that this would make them more favorably disposed toward a candidate, whereas about three in 10 said that this would make no difference. Obviously, this type of private, financial information from the candidate is judged by the public as having the most direct relevance to the current campaign and future policy directions. This survey finding points to a possible direction for political reform. California has taken some initial steps toward electronic disclosure in recent years. But there are questions about the expense and user-friendliness of the software. Also, there is no unique donor identification that would allow reporters to track contributions at several levels simultaneously. The expense of running for office in California and the wealth of groups and individuals who are would-be donors create a need to follow the money trail in a much more systematic fashion. Who supports which candidates financially and at what level are important pieces of information California voters need to understand the 59 likely influences that successful candidates will face after they are elected to office. Of course, the media would also probably need to increase its role to present such information in a form that was readily available and understandable to the state’s voters. Candidate Codes of Conduct Candidates for elected office cannot be compelled to avoid personal attacks and negative campaigns for first amendment reasons. But perhaps they can be induced to do so by signing pledges that they could be held accountable for during the course of a political campaign. Our survey suggests that California voters are open to this idea. Seventy-seven percent of likely voters said that they would view candidates more favorably (combining the much and somewhat categories) if they signed a code of conduct to “run a truthful, fair, and clean campaign” (Table 4.4). Similarly, over 70 percent of likely voters said that they would view a candidate more favorably if they signed a code of conduct to run “an issue-oriented campaign.” The support for reforms among likely voters was almost equally high for candidates who would sign a code of conduct “not to use race, gender, religion, sexual orientation, or age as a basis for attack.” Table 4.4 Californians’ Support for Codes of Conduct “If a candidate signed a campaign code of conduct pledging to run a truthful, fair, and clean campaign, would this make you view him or her much more favorably, or somewhat more favorably, or would it make no difference to you?” Much more favorably Somewhat more favorably No difference Less favorably, do not know % of California Adults 48 27 23 2 % of Likely Voters 48 29 22 1 SOURCE: PPIC Statewide Survey (November 2002). NOTE: N = 2,000. 60 Perhaps some political consultants would prefer that their candidates not sign such pledges, since it restricts their options in running a campaign. Moreover, it is difficult to predict whether these codes of conduct would actually be followed by the candidates who adopted them. From the voter’s perspective, it seems like a worthy, low-risk, and inexpensive experiment. Candidates do not like to hear that they have violated promises, so the act of getting them to sign pledges might on the margin provide incentives for better conduct. For the voters, it also offers the welcome prospect of a campaign that is geared toward the issues that matter most to them and a welcome respite from the business of attacks and counterattacks. The California public is dissatisfied with highly negative races and the limited access they have to their candidates. Their frustration was evident in the course of the 2002 elections, according to our surveys. It peaked in 2003 when California voters took matters into their own hands and removed an incumbent governor who, among many other things, epitomized many of the undesirable aspects of modern campaigning. Recall elections are now part of the statewide political landscape—not just the local election context as in the past. Whatever the merits and demerits of recall elections, they do not address the underlying sense, especially felt by better informed and more involved voters, that political dialogue is conducted at a demeaning and negative level. Indeed, we could find no evidence in our surveys since the November 2002 election that attitudes toward campaigns and trust in government changed in a positive way after the recall of Governor Gray Davis and his replacement by Arnold Schwarzenegger in 2003. The problem is that political consultants are paid to get their clients elected, not to make voters feel happier about the process. Political campaigns today are not designed to increase voter turnout, or to improve the public’s trust in government, or to shed a positive light on elected officials or their challengers. Many of the members of the political establishment seem to have concluded that a candidate who takes the high road against a candidate who takes the low one might lose the war of attrition. Moreover, his or her consultant will not be rewarded for sacrificing the candidate to the cause of higher ethics or 61 preserving civic sensibilities. With increasing campaign expenses to run for office, politicians and political consultants are under pressure to win elections at all costs. However, the survey evidence discussed in this report shows that negative tactics hurt the candidate who initiates them as well as the candidate who is the object. Although the Davis campaign outspent the Simon campaign and won the election while using negative attack ads—following a consistent pattern that has existed in gubernatorial elections since 1986—we noted several risks to this political strategy. In the short run, the candidate can alienate his supporters and turn them off to voting for anyone in the election. This could spell defeat in a close election in which the candidate needs every vote from supporters (e.g., Davis won by only 5 percentage points). In the long run, this can also leave the winner vulnerable in a recall election such as the one faced by Davis in October 2003. Some Democratic voters were disinterested in supporting a candidate that they grew to dislike during the election. More important, the elected official may find that voters have less trust in government and less respect for the elected officeholder because of the way he or she behaved in the course of the campaign. This raises problems when an elected official needs support for policies and ballot measures. As we have seen in our survey, once confidence is lost it is difficult to regain. We hope that these survey findings will serve to dampen the appetite for negativity as candidates think about their winning strategy in statewide elections in 2006 and beyond. The voters have also shown the way to a new type of political campaigning in their survey responses to a wide variety of different ideas for political and election reforms. If there is resistance to enacting such changes in the state’s political arena, then we can anticipate that the California public will incorporate its expectations about good campaigns into its evaluations in the voting booth through the citizens’ initiative process. 62 Appendix A Dissatisfaction with the Candidate Choices for Governor Independent Variable Coefficient S.E. Level of attention to the news (four-point scale) 0.47*** 0.06 Level of interest in politics (four-point scale) 0.26*** 0.07 Saw an ad for a gubernatorial candidate 0.03 0.11 Strong Democrat (dummy variable) –0.50*** 0.15 Strong Republican (dummy variable) –0.28 0.15 Independent (dummy variable) 0.49*** 0.15 Female (dummy variable) –0.05 0.10 White (dummy variable) 1.64*** 0.26 Latino (dummy variable) 0.49** 0.17 Constant –3.27*** 0.32 N Log likelihood Pseudo R2 1901 2361.68 0.18 SOURCE: PPIC Statewide Survey (November 2002). NOTES: Figures are logit regression coefficients. Dependent variable is whether a respondent is dissatisfied with the choice of candidates for governor. N = 2,000. **p < .01. ***p < .001. 63 Appendix B Trust in State Government, All Adults Coefficient S.E. Political assessments Favorable ranking of incumbent governor Gray Davis (dummy variable) 0.594** 0.184 Perception that election campaigns are getting worse (dummy variable) –0.462* 0.184 Individual characteristics Strong Democrat (dummy variable) –0.289 0.251 Strong Republican (dummy variable) –0.688* 0.297 Independent (dummy variable) –0.748** 0.271 Conservative (dummy variable) –0.004 0.210 Liberal (dummy variable) –0.059 0.210 Female (dummy variable) –0.013 0.168 White (dummy variable) –0.033 0.238 Latino (dummy variable) 0.577* 0.262 Youth, ages 18–34 (dummy variable) –0.013 0.185 Education (three-point scale) –0.033 0.125 Income (three-point scale) 0.050 0.118 Political engagement Political activities (eight-point scale) –0.013 0.063 Interest in politics (four-point scale) 0.256* 0.129 Attention paid to the news (four-point scale) –0.197 0.110 Political knowledge (four-point scale) –0.039 0.090 Exposure to advertisements (dummy variable) 0.135 0.179 Constant Na Pseudo R2 –0.907 754 0.14 0.462 Percentage correctly predicted 0.69 SOURCE: PPIC Statewide Survey (November 2002). NOTES: Figures are logit regression coefficients. Dependent variable is whether respondent trusts the state. N = 2,000. aN is less than 2,000 because some questions were asked of a randomly split sample of half of all respondents. *p < .05. **p < .01. 65 Appendix C Trust in State Government, Likely Voters Coefficient S.E. Political assessments Favorable ranking of incumbent governor Gray Davis (dummy variable) 0.976** 0.290 Perception that election campaigns are getting worse (dummy variable) –0.515* 0.261 Individual characteristics Strong Democrat (dummy variable) –0.329 0.341 Strong Republican (dummy variable) –0.743 0.381 Independent (dummy variable) –0.875* 0.390 Conservative (dummy variable) –0.204 0.328 Liberal (dummy variable) –0.077 0.307 Female (dummy variable) –0.141 0.250 White (dummy variable) –0.015 0.352 Latino (dummy variable) 0.366 0.445 Youth, ages 18–34 (dummy variable) 0.206 0.312 Education (three-point scale) –0.149 0.187 Income (three-point scale) 0.271 0.169 Political engagement Political activities (eight-point scale) 0.013 0.081 Interest in politics (four-point scale) –0.289 0.230 Attention paid to the news (four-point scale) –0.110 0.185 Political knowledge (four-point scale) 0.056 0.132 Exposure to advertisements (dummy variable) 0.368 0.278 Constant Na Pseudo R2 –0.451 404 0.17 0.806 Percentage correctly predicted 0.75 SOURCE: PPIC Statewide Survey (November 2002). NOTE: Figures are logit regression coefficients. Dependent variable is whether respondent trusts the state. N = 2,000. aNo. is less than 2,000 because some questions were asked of a randomly split sample of half of all respondents. *p < .05. **p < .01. 67 Bibliography Ansolabehere, Stephen, and Shanto Iyengar, Going Negative: How Political Advertisements Shrink and Polarize the Electorate, Free Press, New York, 1995. Baldassare, Mark, California in the New Millennium: The Changing Social and Political Landscape, University of California Press, Berkeley, California, 2000. Baldassare, Mark, A California State of Mind: The Conflicted Voter in a Changing World, University of California Press, Berkeley, California, 2002. Baldassare, Mark, Bruce Cain, and Jonathan Cohen, “Voters Nostalgic for a Good Old Fashioned Campaign,” California Journal, January, 2003a. Baldassare, Mark, Bruce Cain, and Jonathan Cohen, “California Voters Take Charge,” California Journal, November, 2003b. Bartels, Larry M., “Campaign Quality: Standards for Evaluation, Benchmarks for Reform,” in Larry M. Bartels and Lynn Vavreck, eds., Campaign Reform: Insights and Evidence, University of Michigan, Ann Arbor, Michigan, 2000, pp. 1–61. Bartels, Larry M., et al., “Campaign Reform: Insights and Evidence,” in Larry M. Bartels and Lynn Vavreck, eds., Campaign Reform: Insights and Evidence, University of Michigan, Ann Arbor, Michigan, 2000, pp. 201–247. Berelson, Bernard R., Paul F. Lazarsfeld, and William N. McPhee, Voting: A Study of Opinion Formation in a Presidential Campaign, University of Chicago Press, Chicago, Illinois, 1954. 69 Bowler, Shaun, and Todd Donovan, “Why Do People Like Their State and Local Government More Than the Federal Government?” paper presented at the annual meeting of the Midwest Political Science Association, Chicago, Illinois, April 25–27, 2002. Buchanan, Bruce, Renewing Presidential Politics: Campaigns, Media, and the Public Interest, Rowman and Littlefield, Lanham, Maryland, 1996. Buchanan, Bruce, “Regime Support and Campaign Reform,” in Larry M. Bartels and Lynn Vavreck, eds., Campaign Reform: Insights and Evidence, University of Michigan, Ann Arbor, Michigan, 2000, pp. 173–200. California Attorney General, Initiative Measures, 2000–2002, State of California, Sacramento, California, 2002, available at http://www. caag.ca.us/initiatives/index.htm. California Department of Finance, Historical California Population Estimates, 1941–2003, State of California, Sacramento, California, 2003. California Secretary of State, Total Campaign Expenditures by Candidates for the California State Legislature, 1976–1998, State of California, Sacramento, California, 1998. California Secretary of State, Statement of the Vote, State of California, Sacramento, California, November 2000. California Secretary of State, Statement of the Vote, State of California, Sacramento, California, November 2002. California Secretary of State, Statement of the Vote, State of California, Sacramento, California, October 2003. California Secretary of State, Statement of the Vote, State of California, Sacramento, California, March 2004. Citrin, Jack, “Comment: The Political Relevance of Trust in Government,” American Political Science Review, Vol. 68, 1974, pp. 973–988. 70 Citrin, Jack, and Donald Philip Green, “Presidential Leadership and the Resurgence of Trust in Government,” British Journal of Political Science, Vol. 16, 1986, pp. 431–453. Clark, Harold D., and Alan C. Acock, “National Elections and Political Attitudes: The Case of Political Efficacy,” British Journal of Political Science, Vol. 19, 1989, pp. 551–562. Cohen, Jonathan, Mark Baldassare, and Eliana Kaimowitz-Rodriguez, “Trust in Government and Political Participation Among California Latinos: 1998–2002,” paper presented at annual meeting of the American Association for Public Opinion Research, Nashville, Tennessee, May 14–18, 2003. Dahl, Robert A., A Preface to Democratic Theory, University of Chicago Press, Chicago, Illinois, 1956. Dionne, E. J., Why Americans Hate Politics, Simon and Schuster, New York, 1991. Field Poll, Survey Report, October 1996. Field Poll, Political Parties and the Political System, April 1999. Field Poll, Record Low Voter Turnout Expected, November 2002. Finkel, Steven E., “Reciprocal Effects of Participation and Political Efficacy: A Panel Analysis,” American Journal of Political Science, Vol. 29, No. 4, 1985, pp. 891–913. Ginsberg, Benjamin, and R. Weissberg, “Elections and the Mobilization of Popular Support,” American Journal of Political Science, Vol. 22, 1978, pp. 31–55. Hetherington, Marc J., “The Political Relevance of Political Trust,” American Political Science Review, December 1998. Lipset, Seymour Martin, and William Schneider, The Confidence Gap, Free Press, New York, 1983. Los Angeles Times Poll, November election exit poll, 2002. 71 Miller, Arthur H., “Political Issues and Trust in Government: 1964–1970,” American Political Science Review, Vol. 68, 1974, pp. 951–972. Miller, Warren E., and J. Merrill Shanks, The New American Voter, Harvard University Press, Cambridge, Massachusetts, 1996. Moore, David W., “The Myth and Mythology of Trust in Government,” Public Perspective, January/February 2002, pp. 7–11. The Pew Research Center for the People and the Press, Deconstructing Distrust: How Americans View Government, Washington, D.C., March 10, 1998. PPIC Statewide Survey, The Changing Political Landscape of California Series, Public Policy Institute of California, San Francisco, California, April 1998. PPIC Statewide Survey, The Changing Political Landscape of California Series, Public Policy Institute of California, San Francisco, California, May 1998. PPIC Statewide Survey, The Changing Political Landscape of California Series, Public Policy Institute of California, San Francisco, California, September 1998. PPIC Statewide Survey, The Changing Political Landscape of California Series, Public Policy Institute of California, San Francisco, California, October 1998. PPIC Statewide Survey, Californians and Their Government Series, Public Policy Institute of California, San Francisco, California, December 1999. PPIC Statewide Survey, Californians and Their Government Series, Public Policy Institute of California, San Francisco, California, January 2000. PPIC Statewide Survey, Californians and Their Government Series, Public Policy Institute of California, San Francisco, California, October 2000. 72 PPIC Statewide Survey, Californians and Their Government Series, Public Policy Institute of California, San Francisco, California, January 2001. PPIC Statewide Survey, Californians and Their Government Series, Public Policy Institute of California, San Francisco, California, January 2002. PPIC Statewide Survey, Californians and Their Government Series, Public Policy Institute of California, San Francisco, California, August 2002. PPIC Statewide Survey, Californians and Their Government Series, Public Policy Institute of California, San Francisco, California, October 2002. PPIC Statewide Survey, Special Survey on Political Campaigns, Public Policy Institute of California, San Francisco, California, November 2002. PPIC Statewide Survey, Californians and Their Government Series, Public Policy Institute of California, San Francisco, California, February 2003. PPIC Statewide Survey, Special Survey on the California State Budget, Public Policy Institute of California, San Francisco, California, June 2003. PPIC Statewide Survey, Special Survey on Political Campaigns, Public Policy Institute of California, San Francisco, California, October 2003. PPIC Statewide Survey, Special Survey on the California State Budget, Public Policy Institute of California, San Francisco, California, January 2004. PPIC Statewide Survey, Californians and Their Government Series, Public Policy Institute of California, San Francisco, California, February 2004. 73 PPIC Statewide Survey, Special Survey on the California State Budget, Public Policy Institute of California, San Francisco, California, May 2004. PPIC Statewide Survey, Californians and Their Government Series, Public Policy Institute of California, San Francisco, California, September 2004. Public Policy Institute of California, “California’s Recall Election,” Just the Facts, San Francisco, California, November 2003. Rahn, Wendy M., John Brehm, and Neil Carlson, “National Elections as Institutions for Building Social Capital,” in Morris Fiorina and Theda Skocpol, eds., Civic Engagement in American Democracy, Brookings Institution, Washington, D.C., and Russell Sage Foundation, New York, 1999, pp. 111–160. Rosenstone, Steven J., and John Mark Hansen, Mobilization, Participation, and Democracy in America, Macmillan, New York, 1993. Schrag, Peter, Paradise Lost: California’s Experience, America’s Future, The New Press, New York, 1998. Silva, Fred, The California Initiative Process: Background Material, Public Policy Institute of California, San Francisco, California, 2000. Simon, Mark, “Fiscal Crisis and Voter Resentment Led to Recall,” San Francisco Chronicle, July 25, 2003. Teixeira, Ruy A., The Disappearing American Voter, Brookings Institution, Washington, D.C., 1992. University of California Hastings College of Law, California Ballot Initiatives Database, University of California, San Francisco, California, 2002, available at http://www.holmes.hastings.edu. U.S. Federal Election Commission Campaign Finance Reports and Data (electronic resource), Washington, D.C., 2004. 74 Williams, John T., “Systemic Influences on Political Trust: The Importance of Perceived Institutional Performance,” Political Methodology, Vol. 11, 1985, pp. 125–142. Wolfinger, Raymond E., and Steven J. Rosenstone, Who Votes? Yale University Press, New Haven, Connecticut, 1980. 75 About the Authors MARK BALDASSARE Mark Baldassare is Director of Research at PPIC. He also holds the Arjay and Francis Fearing Miller Chair in Public Policy and directs the PPIC Statewide Survey—a large-scale public opinion project designed to develop an in-depth profile of the social, economic, and political forces at work in California elections and in shaping the state’s public policies. He is the author of nine books, including A California State of Mind: The Conflicted Voter in a Changing World. Before joining PPIC, he was a professor of urban and regional planning at the University of California, Irvine, where he held the Johnson Chair in Civic Governance and initiated and directed the Orange County Annual Survey. He has conducted surveys for the Los Angeles Times, the San Francisco Chronicle, and the California Business Roundtable. He holds a Ph.D. in sociology from the University of California, Berkeley. BRUCE E. CAIN Bruce E. Cain is Director of the Institute of Governmental Studies and Robson Professor of Political Science at the University of California, Berkeley. A Rhodes Scholar at Trinity College, Oxford, he received his Ph.D. in political science from Harvard University. He has been a redistricting consultant to, among others, the Los Angeles City Council, the U.S. Justice Department, the San Diego Citizens’ Commission on Redistricting and the City and County of San Francisco. His recent publications include Voting at the Political Fault Line: California’s Experiment with the Blanket Primary, co-edited with Elisabeth R. Gerber, and Democracy Transformed? Expanding Political Opportunities in Advanced Industrial Democracies, co-edited with Russell J. Dalton and Susan Scarrow. Professor Cain is a political commentator for numerous radio and television stations in the Bay Area. D. E. APOLLONIO Dorie Apollonio is a Legacy Postdoctoral Fellow at the University of California, San Francisco. She received her Ph.D. in political science from the University of California, Berkeley, and her M.P.P. from the 77 John F. Kennedy School of Government at Harvard University. Her research focuses on campaign finance, political reform, and interest group behavior, specifically the political activities of the tobacco industry and their effect on health policy. JONATHAN COHEN Jon Cohen is Associate Director of the PPIC Statewide Survey. His research interests include survey methodology, state tax policy, and electoral systems. He is the co-author of the second edition of the California Latino/Latina Demographic Databook. He has also written a recent book chapter and magazine articles on electoral politics in California. He holds a B.A. in history from the Johns Hopkins University, a graduate diploma from the Nitze School of Advanced International Studies, and an M.A. in political science from the University of California, Berkeley. 78 Related PPIC Publications A California State of Mind: The Conflicted Voter in a Changing World Mark Baldassare California in the New Millennium: The Changing Social and Political Landscape Mark Baldassare Adapting to Term Limits: Recent Experiences and New Directions Bruce Cain and Thad Kousser PPIC publications may be ordered by phone or from our website (800) 232-5343 [mainland U.S.] (415) 291-4400 [Canada, Hawaii, overseas] www.ppic.org 79" } ["___content":protected]=> string(104) "

R 1004MBR

" ["_permalink":protected]=> string(109) "https://www.ppic.org/publication/the-season-of-our-discontent-voters-views-on-california-elections/r_1004mbr/" ["_next":protected]=> array(0) { } ["_prev":protected]=> array(0) { } ["_css_class":protected]=> NULL ["id"]=> int(8321) ["ID"]=> int(8321) ["post_author"]=> string(1) "1" ["post_content"]=> string(0) "" ["post_date"]=> string(19) "2017-05-20 02:36:36" ["post_excerpt"]=> string(0) "" ["post_parent"]=> int(3502) ["post_status"]=> string(7) "inherit" ["post_title"]=> string(9) "R 1004MBR" ["post_type"]=> string(10) "attachment" ["slug"]=> string(9) "r_1004mbr" ["__type":protected]=> NULL ["_wp_attached_file"]=> string(13) "R_1004MBR.pdf" ["wpmf_size"]=> string(6) "714180" ["wpmf_filetype"]=> string(3) "pdf" ["wpmf_order"]=> string(1) "0" ["searchwp_content"]=> string(168176) "The Season of Our Discontent: Voters’Views on California Elections ••• Mark Baldassare Bruce E. Cain D. E. Apollonio Jonathan Cohen 2004 Library of Congress Cataloging-in-Publication Data The season of our discontent : voter’s views on California elections / Mark Baldassare ... [et al.] p. cm. ISBN: 1-58213-086-8 1. Elections—California—Public opinion. 2. Governors— California—Elections—Public opinion. 3. California—Politics and government—1951- 4. Public opinion—California. I. Baldassare, Mark. II. Title. JK8790.S42 2004 324.9794’053—dc22 2004023317 Copyright © 2004 by Public Policy Institute of California All rights reserved San Francisco, CA Short sections of text, not to exceed three paragraphs, may be quoted without written permission provided that full attribution is given to the source and the above copyright notice is included. PPIC does not take or support positions on any ballot measure or on any local, state, or federal legislation, nor does it endorse, support, or oppose any political parties or candidates for public office. Research publications reflect the views of the authors and do not necessarily reflect the views of the staff, officers, or Board of Directors of the Public Policy Institute of California. Foreword In 1998, a decision was made to launch the PPIC Statewide Survey. The decision was made for two reasons. First, it was readily apparent that the initiative process in California had come to dominate some of the most important public policy decisions made in recent decades. If the ballot box was becoming a key source of policy decisionmaking, it was obvious that we all needed to understand the public’s use of this important vehicle for complementing the representative process in Sacramento. Second, with a state as big and complex as California, a careful monitoring of public opinion seemed essential to anticipating issues important to voters and to identifying feasible options for resolving recurring and contentious issues that often result in gridlock and hastily launched initiatives. Over 100,000 Californians have been interviewed in the six years since the first PPIC Statewide Survey was fielded. As we anticipated, the results are often surprising. The desire for greater government benefits but the unwillingness to pay for them through conventional taxes is strong and consistent. The preference for a single-family detached home and the willingness to pay for it with ever-longer commutes is surprising, even by California standards of commute times. Active resident involvement with California’s parks and recreation areas and a preference for maintaining high environmental standards suggest that, for the voters at least, more could be done on this front without endangering the political standing of our representatives. Surely one of the most surprising findings of all is the consistently high marks respondents give to the initiative process. And now we learn from our latest round of PPIC survey findings that voter distrust of government has grown even more intense as a result of the gubernatorial election of 2002. In this report, Mark Baldassare, founder and director of the PPIC Statewide Survey, and his colleagues focus on an issue that should be of concern to every elected official in California—negative campaigning— which damages not only both opponents but also the democratic process iii itself by sowing disillusion, distrust, and cynicism among voters, many of whom then decide not to vote for either candidate. The authors also find a steady and dramatic decline in turnout in gubernatorial elections—from 70 percent of registered voters in 1982 to just 50 percent in 2002. Although there are a number of reasons for this decline in voter participation, only three in 10 Californians said that they were satisfied with the discussion of issues facing the state during the 2002 campaign. Voters wanted debates on the issues by the candidates—and they did not get what they wanted. So, they stayed away from the polls. At the same time, gubernatorial campaign expenditures reached an alltime high—mostly for the purchase of ad time on TV. And yet, only one in four likely voters just before the election in 2002 said that the campaign advertisements were helpful in making a decision about voting. Respondents expressed strong support for such election options as campaign finance reform, candidate debates, public disclosure, and voluntarily adopted codes and pledges of campaign conduct. These are straightforward suggestions and the ones most likely to survive constitutional challenges. Will representatives in Sacramento, along with the governor, take the lead in thinking through these options? Will elected officials listen to the message being sent by the state’s eligible voters? How low will trust in state government have to drift before action is taken? And will the action come from Sacramento or from the ballot box once again? All of these questions are raised by the authors in one of the most discouraging set of findings yet to emerge from the PPIC Statewide Survey. As in many other areas of public policy, there are practical solutions to the problems outlined above. The authors suggest that if nothing is done in Sacramento, voters will once again take it into their own hands at the ballot box. This latest set of findings from the PPIC Statewide Survey should serve as a high-profile warning that something needs to be done to turn the tide on a level of distrust in government that is doing serious damage to the democratic process in California. David W. Lyon President and CEO Public Policy Institute of California iv Summary The California governor’s election in November 2002 is widely regarded by scholars, political consultants, media commentators, and leaders of the political parties as one of the low points in the state’s political history. However, the qualities of the political campaigning that led to this assessment—high expenditures, negative advertising, and low voter turnout—are all indicative of recent trends in California elections. This is the backdrop against which in this report we examine the public’s views on campaigning and elections, analyze the ways that the conduct of campaigns is related to voter dissatisfaction and government distrust, and discuss what Californians would like to see changed about their state’s campaigns and elections. In 2002, a Democratic governor who won a landslide victory four years earlier, Gray Davis, defeated a Republican political novice, Bill Simon, in a surprisingly close contest that will be best remembered for its nasty tone and lack of substance. The governor’s popularity had suffered as the state faced a number of fiscal and economic calamities—including the electricity crisis that resulted from a flawed deregulation law, a dramatic slowdown in the state’s booming economy, and a multibilliondollar deficit in the state government’s budget. But the Davis-Simon matchup may never have occurred without the Democratic incumbent having spent millions of dollars in the GOP primary to help defeat Richard Riordan, a politically moderate candidate who was considered more “electable.” Davis and Simon waged a campaign of harsh words from the outset that focused on attacks and counterattacks revolving around their personal character—including allegations that there were links between the governor’s campaign fund-raising and his decisions in office, and that his GOP opponent had been involved in illegal and incompetent business dealings. They faced off in a live televised debate on the issues only once during the fall campaign. The two candidates spent their multimillion-dollar war chests on negative television v advertisements, beamed into the homes of millions of Californians, which were carefully crafted to destroy the public’s image of their opponents. In the end, the November 2002 election generated the highest campaign expenditures and the lowest voter turnout in state history. The exit polls found that most of those who voted had unfavorable images of both candidates. Governor Davis won the election by a slim five-point margin over Simon (47% to 42%). Almost immediately, Davis’s political opponents seized on the governor’s low approval ratings and the low vote threshold for qualifying a recall made possible by the record low turnout in the November election. With an infusion of cash from a wealthy supporter, the 900,000-plus signatures needed to force a special election in October 2003 were collected, and the state’s voters took the unprecedented action of removing a governor from office. The voters replaced Davis with Arnold Schwarzenegger less than a year after the incumbent’s reelection. Although voters are currently happy with their new governor, surveys indicate that they still remain dissatisfied with political campaigns and elections and are distrustful of state government. The public opinion information presented in this report is based largely on the Public Policy Institute of California (PPIC) Statewide Surveys conducted during the 2002 elections. In all, we conducted 10 surveys of over 20,000 Californians. No preelection survey of ours has ever been more meaningful and timely than the special survey of 2,000 Californians on political campaigns during the last week before the November election. That project is the centerpiece of this report, which seeks to answer the following questions about the relationships between voters and political campaigns in the state of California: • What are the recent trends in California’s voter registration, voter turnout, and campaign expenditures in statewide and legislative elections? What is the relationship between money spent, voter turnout, and winning state elections? • What were the public’s perceptions of the campaigns for the governor’s race in 2002, what are their views on negative vi advertising by candidates, and what are the factors related to dissatisfaction with the state’s elections? • What are the recent trends in the public’s trust in state and federal government, and to what extent is the conduct of political campaigns a contributing factor in the public’s distrust of state government today? • What, if anything, does the public want to do to change the nature of political campaigns and elections in California? How much support is there for campaign finance reform, public debates, increasing public disclosure by the candidates, and adopting voluntary codes of campaign conduct? We summarize our major findings below. Voting and Elections in California California’s elections over the past two decades have occurred in the context of rapid population growth and a fundamental shift in the state’s demography. The addition of over 10 million people increased California’s congressional delegation from 43 to 53 seats between 1980 and 2000. In the meantime, the ethnic and racial mix of the potential electorate has shifted significantly as the state has been transformed through immigration. Registered voters are predominantly white, yet a growing share are Latinos, blacks, and Asians. Lower levels of total voter eligibility have significantly decreased the share of registered voters as a percentage of the adult population to 60 percent, although the share of eligible voters who are registered to vote has remained fairly stable at about 70 percent. Similar to the rest of the nation, California is showing signs of declining voter turnout. In presidential election years since 1980, turnout has declined overall, but it has not declined steadily. In contrast, turnout in California’s gubernatorial elections—which occur in the nonpresidential national election years—decreased steadily and dramatically between 1982 and 2002. Over the course of the past 20 years, turnout in presidential races has dropped 6 percentage points, whereas turnout in California’s gubernatorial elections has dropped almost 20 percentage points, from 70 percent in 1982 to 51 percent in 2002. vii Historically low turnout in the 2002 governor’s race occurred in the context of another ongoing trend—steadily increasing campaign expenditures since the early 1980s, as shown in Figure S.1. The last six gubernatorial races in California have all been marked by new record highs in expenditures. The most dramatic increase was in 2002, when expenditures increased by almost $40 million from the previous election year—to nearly $98 million. Once again, the candidate who spent the most money won, which has been the trend in every governor’s election since 1986. Much campaign funding is steered toward television commercials and, following another trend in recent state elections, toward negative advertising aimed at undermining the credibility of the political opponent. Such political campaigns are not designed to encourage higher voter turnout because their focus is on winning. We hypothesize that these trends have taken a toll on public perceptions of candidates, campaigns, and elections in California and on trust in state government. 120 80 Turnout 100 Expenditures 70 80 60 60 40 50 20 0 1982 1986 1990 1994 1998 2002 40 SOURCE: California Secretary of State (2002). NOTE: Data are presented in 2002 real dollars using the Consumer Price Index provided by the U.S. Department of Labor, Bureau of Labor Statistics. Figure S.1—Gubernatorial Campaign Expenditures and Voter Turnout in California, 1982–2002 Spending ($ millions) Turnout (%) viii Voters’ Perceptions of Political Campaigns In recent years, Californians have grown increasingly dissatisfied with political campaigns in their state. In the PPIC Statewide Surveys we have conducted since the 1998 election, Californians have expressed their dissatisfaction with their candidate choices and their unhappiness with the way campaigns are run. Earlier polls reveal similar trends in public opinion about campaigns and elections. During the fall 2002 election, only about four in 10 residents said that they were satisfied with the choice of candidates for governor. Similarly, almost four in 10 Californians said that, in general, elections had gotten worse in the past 10 years, and 42 percent said that the ethics and values in campaigns had grown worse over the past decade. Likely voters were even more negative about the current state of campaigns, with half (49%) saying that elections had generally gotten worse and 54 percent saying that election ethics and values had gotten worse. Although about eight in 10 Californians said that they had been exposed to campaign advertising in the course of the fall 2002 campaign, only 33 percent said that the advertisements were helpful in making a decision about voting. Moreover, only one in four likely voters said that the candidate advertisements they had seen were helpful. Much of the dissatisfaction that Californians say they feel toward elections is focused on negative campaigning. In the PPIC Statewide Survey, 58 percent of those who were asked whether Gray Davis and Bill Simon should be critical of each other said that they should not, because campaigns today are too negative. Indeed, our preelection survey used an approach that asked about the preferred campaign behavior of Davis and Simon and then compared the responses of the candidates’ supporters and opponents. Here are some of the key results from that survey: (1) Candidates who attack their opponent run the risk of negative effects on their own image among their supporters; (2) all of the candidates’ responses to being attacked were equally unpopular, even including arguing that the basis for the attack was untrue; (3) the only responses to candidate attacks that were correlated with positive rankings of the candidates were not making a counterattack or ignoring an attack; and (4) the ability of negative ix advertising to decrease the popularity of the candidate who is being attacked must be weighed against the likely outcome of decreasing the popularity of the attacker. Potential voters said that they did not receive all of the information they needed to make a decision in the 2002 California election. In repeated preelection surveys, seven in 10 Californians said that above all else, they would like to know candidates’ stands on the issues. Notably, however, this demand for information about issues was not met in the 2002 campaign. In the same preelection surveys, only three in 10 said that they were satisfied with the discussion of issues facing the state during the campaign. On top of this frustration with the content of the campaign was disappointment with the means that candidates use to deliver campaign messages. When asked how they prefer to learn about candidates, voters chose debates more than any other means of communication. However, in the 2002 campaign, there was only one debate between the candidates. Moreover, although the gubernatorial candidates relied heavily on television advertising, about four in 10 Californians said that they would view candidates less favorably when they primarily used television advertising to get their message to voters. In sum, candidates hurt their own images by campaigning negatively, by providing inadequate information to voters, and by relying too heavily on television advertising to connect with the voters. This strategy may still result in victory, but it could depress the turnout for the winner. Thus, there are political risks in the short run and the long run for the political candidate. Equally important, attack campaigns, even beyond their effects on turnout, may engender lasting feelings of distrust in government. Political Campaigns and Distrust in Government Political observers have long suspected that recent trends in campaign practices such as negative advertising may be linked to steadily declining trust in government. The evidence from our November 2002 PPIC Statewide Survey, as well as past studies of trust in government in California, suggests that campaigns do have an effect on political trust. Californians’ trust in state government declined during the 2002 election x campaign and has not returned to previous levels in the wake of the recall of Governor Davis and his replacement by Arnold Schwarzenegger, who currently enjoys high overall job approval ratings. Negative perceptions of campaigns appear to be a contributing factor for declining trust, and confidence seems to be difficult to restore once the public loses faith in government. The level of trust in the federal government expressed by California residents actually increased since our first PPIC Statewide Survey in April 1998, up to and including the November 2002 election—in part a reaction to the September 11, 2001, terrorist attacks. As of this writing, trust in the federal government has dropped somewhat below the levels we recorded before the 2002 elections. In contrast, trust in state government was moving in a downward direction over time. Trust in state government reached its peak in 2001 and early 2002, with nearly half of all Californians (47%) saying that they trusted state government to do what is right just about always or most of the time. By August 2002, about one in three Californians (37%) said that they trusted the state government. In the months following the November 2002 election, trust in the state government drifted to significantly lower levels and, as of this writing, it has not improved to the levels recorded in the summer and fall leading up to the 2002 election. We conducted a multivariate analysis to determine the factors most highly correlated with distrust in state government. For both the public at large and likely voters, negative assessments of incumbent Governor Gray Davis, negative perceptions of the campaign, and Independent voter status were all significantly correlated with more distrust in state government. For all adults, strong Republicans had more distrust in state government than others did at the time, whereas Latinos had more trust than other racial/ethnic groups in state government. Greater interest in politics was also associated with more trust in state government. Public Support for Political Campaign Reforms We looked at a wide variety of political reforms that were chosen because they are most likely to survive constitutional challenges and because they are the types of policy proposals that have been considered xi in state contexts and discussed in the national political arena in recent years. We tested public support for the following: campaign finance reforms, candidate debates, public disclosure, and voluntarily adopted codes and pledges of campaign conduct by the candidates. We found considerable interest in changing the status quo of California’s campaigns and elections on all of these dimensions. The issue of campaign finance was prominent in the 2002 governor’s race, given the record spending of nearly $100 million by the two major party candidates. Perhaps most surprising of all, given the fiscally conservative nature of Californians, 57 percent of likely voters thought that public, rather than private, financing of campaigns was a good idea. This finding suggests that recent negative experiences with state campaigns, such as the 2002 governor’s race, had made many voters receptive to the public paying the tab for the election process. Californians have been telling us in surveys that they are frustrated with the emphasis of statewide political campaigns on 30-second television commercials. What would they prefer instead? Two in three likely voters said that more debates in the 2002 gubernatorial contest would have made the election campaign better. And when asked whether they would favor or oppose an initiative that would require five prime-time broadcast gubernatorial debates, 60 percent were in favor of the proposal. Voter support for election reforms that provide more personal contact with the candidates suggests another area for improvement. Disclosure is a key ingredient in elections because voters need access to accurate information when making ballot choices. Among likely voters, 71 percent said that the immediate disclosure of campaign contributions would make them more favorably disposed toward a candidate. By contrast, fewer say that they would have a positive impression of candidates who released their tax forms or medical records. The candidates’ sources and amounts of financial support are important to voters, and their preferences for more information point to a potential avenue for political reform. Candidates cannot be compelled to avoid personal attacks and negative campaigns for first amendment reasons. But what if candidates can be induced to do so by signing pledges that held them accountable xii during the course of a political campaign? Our survey suggests that California voters are open to this idea: Seventy-seven percent of likely voters said that they would view candidates more favorably if they signed a pledge or a code of conduct to “run a truthful, fair, and clean campaign,” and over 70 percent of likely voters said that they would view a candidate more favorably if he or she signed a code of conduct to run “an issue-oriented campaign.” The support among likely voters was almost equally high for candidates who would sign a code of conduct “not to use race, gender, religion, sexual orientation, or age as a basis for attack.” Despite the practical limitations of codes of conduct, many voters are keen on the idea of them because they offer a greater likelihood of a campaign focusing on the issues and a break from the negative tone of today’s California campaigns. The California public expressed a high level of dissatisfaction with the negative tone of the 2002 governor’s election, the limited access they had to the candidates, and the lack of attention to the issues confronting the state. The most expensive governor’s race in the state’s history resulted in the lowest voter turnout and one of the nastiest political campaigns on record—all following trends in recent state elections and likely to continue unless there is a concerted effort to change the nature of elections. In the end, the campaigns’ focus on negative advertising appeared to harm both the candidate doing the attacking and the candidate who was attacked. Although the candidate who spent the most money campaigning in this manner won, we found that there are political risks to spending so much money on negative advertising. The aftermath of Davis’s narrow victory in a low turnout election—a recall in which the incumbent found it difficult to mobilize his support base to fight off the GOP’s attempt to remove him from office—highlights the vulnerabilities of this political strategy. If the current trends of public dissatisfaction and voter disengagement continue, there could be more serious consequences—not the least of which is declining trust in state government and its elected officials—endangering the future of democratic society in California. The public’s frustration with state politics was evident in the course of the 2002 governor’s race, and it peaked in 2003 when California xiii voters removed the winner from office. But the recall did not address the underlying problem of negative campaigns and declining turnout in gubernatorial elections. California voters, deeply dissatisfied, have indicated their preferences for a new type of campaign behavior—increased financial disclosure, more public debates, campaign finance reform, and candidate codes of conduct—most of which could be implemented in time for the 2006 governor’s election. Given their frustration with the status quo and their support for change, the voters may take the matter of election reform into their own hands through the initiative process. xiv Contents Foreword......................................... iii Summary......................................... v Figures .......................................... xvii Tables ........................................... xix Acknowledgments................................... xxi 1. RECENT TRENDS IN CALIFORNIA ELECTIONS ..... 1 2. POLITICAL CAMPAIGNS AND VOTER DISSATISFACTION ............................. Population, Voting, and Public Opinion Surveys .......... Voters’ Views on the 2002 Governor’s Election ........... Perceptions of Political Campaigns .................... Low Turnout and More Initiatives: Confirming Evidence of Campaign Dissatisfaction....................... What Happens When Candidates Focus on Negative Campaigning?............................... Measuring the Influence of Attack Advertising: Results of a Survey Experiment ........................... Correlates of Voter Dissatisfaction with Candidate Choices... Information on Election Campaigns ................... Conclusions .................................... 13 14 15 19 21 24 26 31 31 34 3. POLITICAL CAMPAIGNS AND VOTER DISTRUST .... The Reasons for Public Distrust ...................... Survey Evidence on Distrust in Government ............. Trends in Distrust in State Government ................ Correlates of Distrust in State Government .............. Conclusions .................................... 37 38 40 44 48 50 4. CONCLUSIONS ON PUBLIC SUPPORT FOR CAMPAIGN REFORMS .......................... Campaign Finance Reform ......................... Candidate Debates ............................... 53 54 56 xv Public Disclosure ................................ 58 Candidate Codes of Conduct........................ 60 Appendix A. Dissatisfaction with the Candidate Choices for Governor .... B. Trust in State Government, All Adults ................. C. Trust in State Government, Likely Voters ............... 63 65 67 Bibliography ...................................... 69 About the Authors .................................. 77 Related PPIC Publications............................. 79 xvi Figures S.1. Gubernatorial Campaign Expenditures and Voter Turnout in California, 1982–2002 ................. viii 2.1. Gubernatorial Campaign Expenditures and Voter Turnout in California, 1982–2002 ................. 16 xvii Tables 1.1. 1.2. 1.3. 1.4. 1.5. 1.6. 1.7. 2.1. 2.2. 2.3. 2.4. 2.5. 2.6. 2.7. 2.8. 2.9. 2.10. Population Change in California, 1980–2002 ......... Total Population, Eligible Voters, and Registered Voters in California, 1980–2002 ....................... Voter Turnout in Presidential Elections in California, 1980–2000 ................................. Voter Turnout in Gubernatorial Elections in California, 1982–2002 ................................. Spending on Gubernatorial Elections in California, 1982–2002 ................................. Cost per Voter of Gubernatorial Elections in California, 1982–2002 ................................. Political Party Affiliation of Registered Voters in California, 2002 .............................. Californians’ Satisfaction with Candidate Choices for Governor ................................... Californians’ Perceptions of Election Trends .......... Californians’ Attitudes Toward Politics and Politicians, 1983 and 1999............................... Californians’ Perceptions of Advertising by Gubernatorial Candidates ....................... Californians’ Attention to Campaigns and Voting Uncertainty ................................. Citizens’ Campaign Reform Initiatives in California, 1980–2002 ................................. Californians’ Perceptions of Negative Campaigning ..... Perceptions of Negative Campaigning by Supporters and Nonsupporters ............................... Californians’ Preferences and Satisfaction on Attention to the Issues ................................. Californians’ Preferred Means of Candidate Communication .............................. 2 3 5 6 8 9 11 17 18 18 19 21 23 25 29 32 33 xix 3.1. Californians’ Trust in Federal Government, 1998–2004 ................................. 45 3.2. Californians’ Trust in State Government, 2001–2004.... 46 4.1. Californians’ Support for Funding Political Campaigns... 55 4.2. Californians’ Support for Gubernatorial Debates ....... 57 4.3. Californians’ Support for Public Disclosure ........... 59 4.4. Californians’ Support for Codes of Conduct .......... 60 A.1. Dissatisfaction with the Candidate Choices for Governor................................... 63 B.1. Trust in State Government, All Adults .............. 65 C.1. Trust in State Government, Likely Voters ............ 67 xx Acknowledgments We wish to thank The Pew Charitable Trusts for their generous support of the public opinion surveys and focus groups conducted during the 2002 California election and the 2003 governor’s recall election that made this report possible. Dorie Apollonio did the analysis and wrote the first draft of Chapters 1, 2, and 3. She also did the analysis for Chapter 4. Bruce E. Cain wrote the first draft of Chapter 4. Mark Baldassare rewrote the draft chapters, incorporated the draft chapters and other materials into a unified report, and took the report through several rounds of revisions. Jonathan Cohen was involved in the survey design, questionnaire development, data collection, and analysis. The authors were assisted in their design and implementation of the public opinion research by staff of the Center for Campaign Leadership at the Institute of Governmental Studies, University of California, Berkeley, including Keena Lipsitz, Julia Mitchell, Sunetti Shah, Christine Trost, Dan Schnur, and Darry Sragow, and by the PPIC Statewide Survey staff at the Public Policy Institute of California, including Eliana Kaimowitz and Renatta De Fever. We also wish to thank the staff at Schulman, Ronca & Bucuvalas, Inc., for their assistance and advice in the design of the survey questions. We also benefited from feedback offered by many scholars, journalists, political consultants, elected officials, and students at the events and seminars sponsored by the Center for Campaign Leadership in Berkeley and Sacramento. We would like to thank the reviewers of an early draft of this report—Shaun Bowler, John Ellwood, David Lesher, Paul Lewis, Joyce Peterson, Karthick Ramakrishnan, and Susan Rasky—for their comments and suggestions on the substance of the report. xxi 1. Recent Trends in California Elections Once upon a time, California voters stood in line to shake hands at political rallies with would-be officeholders and waited at train stations for their whistle-stop tours. Now, they are most accustomed to seeing candidates through the lens of 30-second television commercials beamed into their living rooms. Elections in California have become more professionalized in recent decades—and more expensive. Consultants have become more knowledgeable about how to run winning campaigns. Candidates are more informed than ever by polls and focus groups about what the voters want. Their messages are carefully tested before they are ever seen or heard by voters. Yet, for all this expertise and money devoted to figuring out how to win elections, Californians tell us in our public opinion surveys that they are dissatisfied with the state’s election process and the negative tone of campaigns. Many Californians are also forgoing a benefit and a responsibility of U.S. citizenship—registering to vote—and many of those who are registered do not turn out to cast their ballots on Election Day. This report explores the level and sources of discontent with political campaigns among Californians, examines some of the critical implications of this discontent for the democratic process, and considers the public’s views on how to improve the state’s elections in the future. California’s elections have become more expensive as the state’s population has grown and its demography has been changing. The defining characteristics of California’s elections are that they encompass a large geographic area and a racially and ethnically diverse electorate. These factors set the stage for media driven campaigns that focus on reaching voters through television commercials, adding high costs to all 1 statewide campaigns and leading to complex decisions that all statewide candidates face in trying to reach diverse audiences. Since 1980, California’s population has increased by almost 50 percent—from less than 24 million, to over 35 million people in the year 2002 (Table 1.1). In the late 1980s, the population grew by over 5 percent in each election cycle. In the recession of the early 1990s, it slowed to approximately half a million new people per election cycle. But in the two years preceding both the 2000 and the 2002 election, the California population grew by more than one million people. Meanwhile, decades of foreign immigration and racial and ethnic change have transformed California into the first large “majorityminority” state where whites make up less than half of the total population. As of this writing, about one in three residents is Latino in ethnic origin, and the remaining one in six adults is Asian, black, or of another racial category. Moreover, substantial proportions of Asians and Latinos in California are either immigrants or children of immigrants (Baldassare, 2000, 2002). Although not all of this new population was eligible to vote— because of either age disqualification or noncitizenship—the addition of over 10 million people in 20 years increased California’s Congressional Table 1.1 Population Change in California, 1980–2002 Population Population Increase Percentage Year (thousands) (thousands) Increase 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 23,782 24,805 25,816 27,052 28,393 29,828 30,987 31,523 31,962 32,862 34,088 35,336 525 1,023 1,011 1,236 1,341 1,435 1,159 536 439 900 1,226 1,248 2.3 4.3 4.1 4.8 5.0 5.1 3.9 1.7 1.4 2.8 3.7 3.7 SOURCE: California Department of Finance (2003). 2 delegation from 43 to 53 seats between 1980 and 2000. Moreover, despite the voter eligibility obstacles resulting from immigration that slow the translation from population growth to electoral change, the ethnic and racial mix of the electorate has been shifting significantly over the past decade. In the most recent state elections, about one in four voters was Latino, Asian, or black (Baldassare, 2002). These demographic and racial/ethnic shifts in the population are important to acknowledge because it would be unfair to claim that trends in political campaigning alone account for all of the decline in voter participation in recent years. Irrespective of voter registration, lower levels of total voter eligibility have decreased the share of voters as a fraction of the total adult population. In the 1980s, the percentage of adult residents eligible to vote in elections remained at a nearly constant rate of 65 percent (Table 1.2). By the early 1990s, the share of the adult population eligible to vote in California dropped sharply, to 58 percent. Table 1.2 Total Population, Eligible Voters, and Registered Voters in California, 1980–2002 Total Eligible Registered Population Voters % Voters % Year (thousands) (thousands) Eligible (thousands) Registered 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 23,669 15,384 65 11,362 24,500 15,984 65 11,559 25,550 16,582 65 13,074 26,444 17,561 66 12,834 28,592 19,052 67 14,005 29,800 19,245 65 13,478 31,000 20,864 67 15,101 32,000 18,496 58 14,724 32,344 19,527 60 15,602 33,252 20,806 63 14,969 34,336 21,461 63 15,707 35,802 21,451 60 15,303 74 72 79 73 74 70 73 78 80 72 73 71 SOURCE: Field Poll (2002). NOTE: Population estimates are slightly different from those presented in Table 1.1. 3 The share has increased slightly since that time, with a little more than 60 percent of the adult population eligible to vote in California elections by 2002. Whatever the changes in numbers eligible to vote in the adult population, however, the proportion of eligible voters who register to vote has remained fairly stable, with only a few exceptions. Since 1980, an estimated 70 percent or more of those eligible to vote have registered. During the last 20 years, the share of eligible voters registering to vote has increased only slightly in a few election years. A much larger share than usual—79 percent of those eligible to vote—registered during the 1984 presidential election cycle, which pitted Ronald Reagan against Walter Mondale. In the 1994 gubernatorial election cycle, where Kathleen Brown faced Pete Wilson, 78 percent of those eligible to vote were registered. And in 1996—a presidential year that featured not only the two major party candidates, Bill Clinton and Bob Dole—but also a surge in new Latino voters in response to voter registration drives and the passage in 1994 of an anti-illegal immigrant initiative—eight in 10 of those eligible to vote were registered. With the exception of these years, the levels of voter registration have remained stable, despite many efforts to increase voter registration among the legions of young adults and new U.S. citizens. Not all of the registered voters in California actually go to the polls on Election Day. Similar to the rest of the nation, California has shown signs of decreased voter turnout in terms of the share of registered voters going to the polls. The national enfranchisement of 18-to-20 year olds in 1972 led to a decline in the percentage of registered voters turning out nationwide, because young adults are less likely to vote than any other age group. The extensive research by Wolfinger and Rosenstone (1980) notes that some of the causes of low voter turnout among young adults may be residential mobility, which makes registration more difficult, or lower levels of education; individuals with college degrees, which are rarely acquired before the early twenties, are much more likely to vote. However, although voter turnout in California has declined overall in presidential election years since 1980, it has not declined steadily (Table 1.3). Whereas only two in three registered voters turned out in 4 Table 1.3 Voter Turnout in Presidential Elections in California, 1980–2000 % of Registered Year Voters 1980 1984 1988 1992 1996 2000 77 75 73 75 66 71 SOURCE: California Secretary of State (2000). 1996, the share of registered adults voting in the 2000 election increased by 5 percentage points to 71 percent, which is near the levels of turnout seen in 1988 (73%) and only slightly lower than turnout rates in 1984 (75%) and 1992, when three in four of those registered actually voted. Overall, turnout rates in presidential elections in California have been inconsistent across election cycles but have not dropped significantly in the past two decades. In contrast, turnout in California’s gubernatorial elections—which occur in the nonpresidential national election years—has decreased dramatically between 1982 and 2002. California, like 36 other states and territories, elects its governor in a nonpresidential election year—that is, two years after the presidential race. Seven other states and territories also elect governors in a nonpresidential election year but not at the two-year Congressional election interval. Without the national attention that comes with the race for president, registered voters are less likely to participate in off-year elections. Turnout in the gubernatorial election years in California has been lower than turnout in presidential years in every election cycle since 1912, the first election cycle for which the California Secretary of State collected election data. Moreover, the gap in voting between presidential election years and gubernatorial election years has increased in recent 5 decades because the turnout in presidential years has changed little whereas turnout in gubernatorial years has declined steadily. Over the past 20 years, voter turnout in presidential races has dropped 6 percent, whereas voter turnout in gubernatorial election years has dropped almost 20 percent, from 70 percent in 1982 to 51 percent in 2002 (Table 1.4). The declining turnout in gubernatorial elections happened in two stages. After the 1982 elections, turnout dropped 11 percentage points to 59 percent and remained at about the 60 percent level until 1998. And then, in 2002, at the end of a gubernatorial race between Gray Davis and Bill Simon that received national recognition for its high expenditures and the negative tone of the campaigning, turnout dropped again, to only about half of all registered voters. The decline in voter turnout in 1986 and 2002 does not reflect dramatic changes in the underlying adult population. As discussed above, during these years, California was experiencing steady population growth and was integrating immigrants into the voting rolls at a fairly constant rate. The drop in turnout in 2002 is noteworthy, since it represents a large percentage decline in voter participation from the previous general election. The level of voter turnout in November 2002 is comparable to that of traditionally poorly attended special elections; the closest comparable race was the 1973 special election in California, which had a similarly low turnout of 48 percent. Table 1.4 Voter Turnout in Gubernatorial Elections in California, 1982–2002 % of Registered Year Voters 1982 1986 1990 1994 1998 2002 70 59 59 60 58 51 SOURCE: California Secretary of State (2002). 6 Although the last six gubernatorial races in California have nearly all been marked by new record highs in expenditures, no such pattern is evident in reviewing aggregate spending in state legislative races between 1980 and 1998 (California Secretary of State, 1998; note: Data after 1998 were not available). California has 80 Assembly seats, which means that candidates must run every two years; California’s 40 Senate seats have four-year terms, with approximately half of the seats in contention every two years. For instance, total spending in primary and general elections increased between 1982 ($84 million) and 1986 ($99 million); however, total spending in 1994 ($106 million) was higher than in 1998 ($101 million) for all state legislative elections (all figures are in 2002 real dollars). Since the passage of legislative term limits in 1990 took effect, the cost of state legislative races has been over $100 million in every election but, again, the available data suggest that total spending has fluctuated rather than increased over time. Federal legislative elections in California in this same time frame also do not indicate a consistent upward trend in spending (U.S. Federal Election Commission, 2004). For example, the total amount spent in California actually declined between 1982 ($52 million) and 1986 ($38 million), then increased sharply in 1992 ($80 million), and declined from that level in 1994 ($58 million) when we look at all Congressional elections in the state (all figures are in 2002 real dollars). Congressional seats are not term-limited like state seats. As a result, spending has fluctuated from year to year, largely in response to contentious individual races rather than to any structural changes. In recent years, there have been only a handful of competitive races as a result of a redistricting process after 2000 that favored safe seats for both Democrats and Republicans in office. As contentious individual races surface in the future, we expect to see the trend of fluctuations continue. Candidates running for governor since 1982, meanwhile, have increased their spending in every election year except 1994 (Table 1.5). The most dramatic leap came in 2002, when overall expenditures increased by almost $40 million from the previous election year. In earlier elections, the difference in expenditures from one election to the 7 Table 1.5 Spending on Gubernatorial Elections in California, 1982–2002 Year 1982 Tom Bradley (Dem) George Deukmejian (Rep) Total Spending 6,803,633 4,972,389 11,776,022 2002 Dollars 22,602,969 1986 Tom Bradley (Dem) George Deukmejian (Rep)a Total 6,137,522 9,565,125 15,702,647 27,891,845 1990 Dianne Feinstein (Dem) Pete Wilson (Rep) Total 13,227,930 16,028,590 29,256,520 45,768,626 1994 Kathleen Brown (Dem) Pete Wilson (Rep)a Total 12,251,634 19,555,243 31,806,877 39,377,870 1998 Gray Davis (Dem) Dan Lungren (Rep) Total 28,642,125 23,845,008 52,487,133 58,788,186 2002 Gray Davis (Dem)a Bill Simon (Rep) Total 64,261,067 33,595,244 97,856,311 97,856,311 SOURCE: California Secretary of State (2002). NOTES: Data on 2002 dollars are presented in 2002 real dollars using the Consumer Price Index provided by the U.S. Department of Labor, Bureau of Labor Statistics. Winner is shown in boldface. aIndicates incumbent. next had always been less than $20 million (all figures are in 2002 real dollars). This change in expenditures over time does not reflect structural changes; California governors were limited to two terms of office long before the 1990 California initiative establishing legislative term limits. Although candidates for governor must reach increasing numbers of people because of California’s increasing population, the 8 growth in population between 1998 and 2002 was not large enough to fully explain the dramatic increase in gubernatorial election spending. Moreover, other statewide races, such as recent races for U.S. senator, have not been marked by similar steady increases in spending. For example, the 1982 U.S. Senate race between Pete Wilson and Jerry Brown, at a cost of $39 million (in 2002 dollars), was more expensive than later Senate races with the exception of the 1994 race between Dianne Feinstein and Michael Huffington, which was clearly an outlier because it involved a Democratic incumbent and a millionaire GOP challenger able to call upon personal wealth to finance this race. It is noteworthy that the increased spending on gubernatorial campaigns in recent years has been consistent even when considered as spending on potential or actual voters. The cost per vote cast has increased substantially in 20 years, from $2.80 in 1982 to $12.64 in 2002 (Table 1.6). Although the cost per eligible voter has increased at a slower rate—a fact that reflects the increasing state population—the Table 1.6 Cost per Voter of Gubernatorial Elections in California, 1982–2002 $ Cost per $ Cost per $ Cost per Year Vote Cast Registered Voter Eligible Voter 1982 1986 1990 1994 1998 2002 2.80 3.66 5.79 4.42 6.82 12.64 1.96 2.17 3.40 2.76 3.93 6.39 1.41 1.59 2.38 2.13 2.83 4.56 SOURCES: California Secretary of State (2002); Field Poll (2002). NOTES: The number of votes cast includes all ballots cast in the election; however, in each election, a small fraction of ballots did not include a vote for governor. Data on cost per voter are presented in 2002 real dollars using the Consumer Price Index provided by the U.S. Department of Labor, Bureau of Labor Statistics. 9 significant increase from $1.41 in 1982 to $4.56 in 2002 suggests that population growth is not the only consideration. The largest increase in per capita costs, unsurprisingly, occurred between 1998 and 2002, when the cost per eligible, registered, and actual voter in the election increased respectively by 61 percent, 62 percent, and 85 percent. What are the implications of the fact that continuing increases in gubernatorial candidate spending coincide with decreases in voter turnout during the same respective years in California? First, just because candidates spend money communicating to voters and voters are more likely to be informed about the upcoming election and the candidate choices does not necessarily lead to more political participation. Indeed, the purpose of campaign advertising may be to encourage voting for your candidate and discourage voting for their candidate. As for the latter, acceptable outcomes for the political consultant could include vote-switching or simply staying home on Election Day. In addition, more spending could mean more negative advertising, which might dishearten potential voters and result in their simply not voting. Our analysis revealed a striking correlation between campaign spending and winning a gubernatorial election in California: Since 1986, the candidate who spent the most money won the election. Gubernatorial elections have been distinctive from all other kinds of statewide elections in California since the 1980s. Unlike other kinds of elections, the costs of gubernatorial races are steadily increasing, and constituents are increasingly unwilling to vote for the gubernatorial candidates running for office. One factor that may explain the trend of lower voter turnout may be the growing numbers of decline-to-state voters in the California electorate, now accounting for about one in six voters (Table 1.7). The growth of Independent voters may be another symptom of the deep disillusionment that California voters have for the campaigns of major party candidates running for statewide offices (Baldassare, 2002). This low turnout is troubling for the state’s system of representative democracy. The governor is the most powerful officeholder in California state politics and an important and high-profile symbol of the state’s 10 Table 1.7 Political Party Affiliation of Registered Voters in California, 2002 Affiliation % of Registered Voters Democrat Republican Decline to state (Independent) Other parties 45 35 15 5 SOURCE: California Secretary of State (2002). executive branch as well as of state government policymaking, yet the statistics indicate that a smaller portion of the electorate is choosing the governor over time. Moreover, with the state’s changing demographics, the voting population does not reflect the racial and ethnic mix of the state’s adult population. Results for 2002 indicate an extension if not an acceleration of recent trends. The candidates for governor convinced barely half of registered voters to cast ballots, and they spent nearly $100 million in the campaign. Once again, the candidate who spent the most money was the winner. Less than a year after the 2002 election, however, Gray Davis faced a first-ever recall, and the voters replaced him with Arnold Schwarzenegger. Has the experience of the 2003 recall changed the political climate? There was clearly excitement about the recall and signs of greater interest in state elections—61 percent of registered voters participated in the 2003 recall—but this increase in voter participation may be a short-lived phenomenon driven by the unique nature of this election. The March 2004 primary, with two state fiscal ballot measures that were heavily promoted by Governor Arnold Schwarzenegger, attracted just 44 percent of registered voters to the polls (California Secretary of State, 2003, 2004). Of course, voter response in the 2006 governor’s election will provide a more definitive answer. In this report, we argue that the low voter turnout in gubernatorial elections reflects, at least in part, an increasing dissatisfaction with political campaigning in California. This trend of increasing 11 dissatisfaction, in turn, is marked by a steady decline in the public’s trust in state government and elected officials that has not been reversed by the recent recall election. In the remainder of this report, we look at voters’ attitudes during the 2002 election—which offer the most recent evidence from a long-term trend that has been evident in California for some time—to understand why political campaigns have fallen out of favor, and we discuss ways to reverse these troubling trends through a variety of political and election reforms. 12 2. Political Campaigns and Voter Dissatisfaction In recent years, Californians have grown dissatisfied with political campaigns and elections in their state. As we noted in Chapter 1, this trend seems most evident in gubernatorial elections. In the 2002 gubernatorial race, turnout declined to 51 percent of registered voters, down from 71 percent in the 2000 presidential elections. Of course, turnout in presidential election years typically exceeds that in gubernatorial election years; however, the 2002 turnout of about half of registered voters represents a considerable decline from the nearly 60 percent of voters who turned out in the 1998 gubernatorial race in California. In successive surveys that we have conducted since the 1998 election, the voters have sent a clear and consistent message: They are dissatisfied with their candidate choices, they are unhappy with the way that campaigns are run, and they have become increasingly distrustful of state government. Although the 2003 recall was mostly about unhappiness with the incumbent, Governor Gray Davis, it was fought against a backdrop of widespread dissatisfaction with electoral politics that has been in evidence for some time. Moreover, our public opinion survey conducted during the 2003 recall election indicated that voters remained unhappy about the state of campaigns and elections with almost half of likely voters saying that the experience of the recall election actually made them feel worse about California politics (see PPIC Statewide Survey, October 2003). It is important to note that Californians were not always so cynical about their government and elected officials, so negative about the campaigns and candidates, and so apparently disconnected from the voting process as they are today. If we turn back the clock to the 1960s, for instance, nearly eight out of 10 registered California voters appeared 13 at the polls during those election years. With the state’s residents apparently buying into the concept that government operated in an efficient and an effective manner and could make their lives better, their support for the large public investments that their leaders proposed (e.g., public universities, freeway systems, state aqueducts) was all but guaranteed at the polls. The citizens’ initiative process was rarely used in this era to overrule government or to set the course for public policy. But over the past 40 years, Californians’ faith in state government and their elected officials has soured and their interest in participating in the state’s elections has diminished (Baldassare, 2000). In recent election cycles leading up to and beyond the 2003 recall, the trends of voter distrust and disinterest appear to have accelerated. These negative changes in perceptions of state government, of the campaigns of gubernatorial candidates, and of the election process itself have occurred in the context of both dramatic increases in the costs of political campaigning and heightened professionalism in campaigns. Candidates have steadily increased their investments in state races over the years in an effort to win over California’s expanding electorate, and since the 1986 contest, the gubernatorial candidate who spent the most won the race. We hypothesize in this chapter that voter dissatisfaction with the political process is related to the public’s dissatisfaction with the way political campaigns are being run. Increased spending on state campaigns may have exacerbated this trend, as campaigns tend to focus on the “dark side” of politicians. Recent California elections have been marked by a heavy reliance on relatively new forms of political communication, particularly negative campaigning and television advertising, and these efforts come under special scrutiny in our analysis. Moreover, investment in these tactics has changed the nature of political campaigning in California today, and such tactics may explain why so many voters stay home on Election Day. Population, Voting, and Public Opinion Surveys In this analysis, we draw on such sources as the PPIC Statewide Survey series and nonpartisan preelection surveys, media exit polls, and government data on state elections. We rely in particular on a special 14 survey conducted in the week leading up to the November 2002 California statewide elections. This preelection survey of 2,000 adult residents was developed by PPIC survey staff, in conjunction with the Center for Campaign Leadership (CCL) at the University of California, Berkeley, and was funded by The Pew Charitable Trusts. Our survey of Californians’ perceptions and attitudes in the closing days of an election cycle attempted to measure both the level of satisfaction with current campaigns, what Californians ideally wanted to take place in their elections, and the kinds of campaigning that they particularly did not appreciate. The survey questions encompassed both elections in general and, more specifically, the 2002 gubernatorial race. Other public opinion data are drawn from the PPIC Statewide Surveys, which have been conducted regularly since early 1998. Other time-series data before 1998 come largely from the Field Poll and the Los Angeles Times Poll. These survey sources have spent much of the last two decades measuring public opinion on ballot issues but have sometimes focused on broader political issues such as knowledge about candidates and campaigns, thus providing data for our comparisons over time. Finally, the California Secretary of State routinely provides information on population and voting trends in California, as well as detailed information about the nature of campaigns. These data include measures of candidate spending, voter turnout, and demographic changes in the population in general as well as in the population of eligible and registered voters. These data provide an objective baseline against which we can consider the public’s perceptions in surveys. Voters’ Views on the 2002 Governor’s Election To place voters’ views in perspective, it is important to note that the 2002 governor’s election reflected trends in evidence for the past 20 years (Figure 2.1). In general, spending on state elections has increased dramatically, while the turnout of registered voters has steadily declined. As we noted above, the trend is most evident in gubernatorial elections rather than in presidential election years. The 2002 governor’s race offers the most dramatic example of these inverse trends. It is important to note that spending on political campaigns is not designed to increase overall voter turnout. The purpose of campaign 15 Spending ($ millions) Turnout (%) 120 80 Turnout 100 Expenditures 70 80 60 60 40 50 20 0 1982 1986 1990 1994 1998 2002 40 SOURCE: California Secretary of State (2002). NOTE: Data are presented in 2002 real dollars using the Consumer Price Index provided by the U.S. Department of Labor, Bureau of Labor Statistics. Figure 2.1—Gubernatorial Campaign Expenditures and Voter Turnout in California, 1982–2002 spending is to win an election. Many political consultants are thus interested in designing campaigns that increase voter turnout for their candidates and depress voter turnout for their opponents. Moreover, individual decisions about whether to vote tend to be related to personal characteristics, such as education and length of residence at the same address rather than typically being a decision to vote for a specific candidate (Wolfinger and Rosenstone, 1980). But even statewide campaigns that devote money to getting out the vote without reference to specific candidates or political parties, as in the 2002 election in California, may not motivate more voters to go to the polls. Indeed, turnout in November 2002 was at a historic low. This is noteworthy because it suggests that marketing and advertising are ineffective in mobilizing voters when the public is unhappy with the campaigns and their choice of candidates, as they apparently were in the 2002 governor’s race. What did the PPIC Statewide Survey tell us about voters’ attitudes toward this election? First, only about four in 10 California adults said that they were satisfied with the choice of candidates for governor (Table 2.1). Notably, both registered voters and likely voters (i.e., voters with a 16 Table 2.1 Californians’ Satisfaction with Candidate Choices for Governor % Satisfied with Choices for Governor All adults Registered voters Likely voters 42 41 37 SOURCE: PPIC Statewide Survey (November 2002). NOTE: N = 2,000. history of participating regularly in elections) were less likely than residents overall to say that they were satisfied with the candidates. For instance, 42 percent of adults stated that they were satisfied with the choice of candidates in the week before the election, but only 37 percent of likely voters said so. The unhappiness with candidates may explain voters’ reluctance to go to the polls in 2002, despite the fact that many of them had voted regularly in the past. Similarly, when people were asked whether they felt that, in general, elections in California had gotten better or worse in the last 10 years, a significant proportion said that elections had gotten worse (Table 2.2). Nearly four in 10 adults (37%) claimed that this was the case, or nearly the same percentage that was dissatisfied with the current choice of candidates. Moreover, 42 percent of adults felt that election campaigns in California were getting worse in terms of ethics and values. It is also noteworthy that only about one in 10 adults thought that campaigns in California were getting better. In addition, the most informed residents—that is, those most likely to vote—were even more negative about the quality of campaigns. Nearly half (49%) of likely voters said that elections in general had gotten worse in California, and over half (54%) said that elections had gotten worse in terms of ethics and values. Evidence of growing dissatisfaction with politics was apparent well before the 2002 elections. Large and increasing proportions of Californians have reported that partisan politics and politicians were problematic for the state (Table 2.3). In 1983, seven in 10 respondents 17 Table 2.2 Californians’ Perceptions of Election Trends California elections have gotten better, worse, or stayed the same in the last 10 years. % Saying Better 13 % Saying Worse 37 % Saying Same or Don’t Know 50 California elections have gotten better, worse, or stayed the same in terms of ethics and values in the last 10 years. 12 42 SOURCE: PPIC Statewide Survey (November 2002). NOTE: N = 2,000. 56 Table 2.3 Californians’ Attitudes Toward Politics and Politicians, 1983 and 1999 Politics today is too partisan; the political parties are continually at odds to such an extent that government and society suffer. % Who % Who Agreed in Agreed in 1983 1999 70 76 Politicians generally are self-serving and only looking out for themselves. SOURCE: Field Poll (1999). NOTE: N = 1,005. 34 41 said that politics was too partisan and that the political partisanship was damaging government and society. In 1999, more than three in four respondents (76%) agreed with this view. Similarly, although only a third of respondents in 1983 said that politicians are self-serving and look out for themselves, over four in 10 Californians agreed with this assessment in 1999. Overall, the current results and recent trends paint a bleak picture of California voters, who are expressing a general discontent with their 18 state’s election process. Voters’ unhappiness is reflected in both public opinion and public action, as registered voters not only complain about the political process but also choose not to participate in statewide elections. The sharp decline in voter turnout in November 2002, in combination with the highly critical statements by residents about the election then in progress, suggests that something has gone noticeably wrong with elections in California. Perceptions of Political Campaigns Political campaigns play an important role in a democratic society. They serve the dual purpose of informing voters about the candidates and motivating voters to go to the polls to cast a ballot for that candidate. When candidates spend more money on their campaigns—assuming that they do so effectively—they should become more familiar to voters. However, just because a campaign reminds people of an impending election does not necessarily mean that voters will turn out to vote. Some voters may take in this campaign information and then decide not to vote for anyone. The 2002 California election is a good case on this latter point. The results of the survey we conducted in the week before the November 2002 election suggest that Californians did not lack exposure to the messages of the candidates running for the governor’s office (Table 2.4). Three in four people said that they had seen the candidates’ campaign Table 2.4 Californians’ Perceptions of Advertising by Gubernatorial Candidates % Who Saw an Ad by a % Who Found Ads Candidate for Governor Helpful in Making in the Last Month a Voting Decisiona All adults 75 33 Registered voters 78 30 Likely voters 80 26 SOURCE: PPIC Statewide Survey (November 2002). NOTE: N = 2,000. aIncludes respondents who saw ads by a candidate for governor. 19 advertisements on television. Moreover, eight in 10 of those who were most likely to vote indicated that they had seen such advertisements. Although Californians may have been exposed to campaign advertising, the advertisements did not appear to provide them with the information they wanted the most—information that would help them decide how to vote. Only one in three Californians (33%) thought that the advertisements by the candidates were helpful in making a decision about how to vote in the November election. Even fewer among those who were most likely to vote in the upcoming election—and who tend to be more politically informed—found the campaign advertisements helpful. Just one in four likely voters said that the television advertisements run by the candidates for governor helped in making a decision about how to vote. Although voters may not receive all of their political information from campaign advertisements, evidence from the 2002 election suggests that there could be a link between the lack of useful information in the political advertisements and voters’ indecision about the candidates in this election. Certainly, the evidence suggests that the quantity of information is not a determining factor in the size of the undecided vote. In a survey taken the month before the 2002 gubernatorial election, over half of respondents said that they were paying close attention to the campaign, yet nearly one in four said that they did not yet know how they would vote (Table 2.5). Moreover, in our survey conducted during the week before Election Day, nearly six in 10 Californians said that they were paying close attention to the campaign. However, despite this attention, nearly one in four voters were still unsure of how they would vote in the election. On Election Day itself, a Los Angeles Times exit poll reported that 20 percent of those leaving the polls had made a decision about how they would vote on the day of the election itself. Many others who are not reflected in exit polls apparently decided not to vote at all. By way of contrast, in a PPIC Statewide Survey taken the month before the November 1998 gubernatorial election, fewer than half of the respondents in that survey said that they were paying close attention to the race for governor. But less than 10 percent of Californians said that they did not know how they would vote for governor in the November 20 Table 2.5 Californians’ Attention to Campaigns and Voting Uncertainty % Who Reported % Who Reported Not Paying Close Attention Knowing How They to the Campaign Would Vote for Governor October 1998a October 2002b November 2002c 47 53 59 9 24 23 SOURCES: PPIC Statewide Survey (October 1998, October 2002, November 2002). aN = 2,005. bN = 2,007. cN = 2,000. 1998 election. Despite the fact that fewer people were following the last gubernatorial election than the 2002 election, fewer people were undecided in 1998 than in the more recent matchup. Complaints about campaign advertising are not new to politics, but our findings are noteworthy because they reflect a relationship between increased political information and increased voting indecision. Some campaign advertising, rather than providing potential voters with information that leads to ballot choices, may instead provide them with information about candidates that may lead to difficulty in choosing a candidate that appeals to them. For some, the decision may be to not vote, given the candidate information that is provided. In sum, low turnout in the 2002 California gubernatorial election cannot be explained by lack of public exposure to the candidates’ campaigns or lack of knowledge about the upcoming elections. Rather, low turnout may reflect a response to the information about the candidates that was provided during the governor’s race. Low Turnout and More Initiatives: Confirming Evidence of Campaign Dissatisfaction The public’s pessimism about California politics has been increasing steadily over the last several years. We argue that this growing 21 dissatisfaction has led to declining voter participation at the polls on Election Day. Another change in California elections—an increasing reliance on the citizens’ initiative process to make state laws—offers further support for the hypothesis that voters’ disaffection with politicians has increased. The 2002 gubernatorial race between Gray Davis and Bill Simon was noteworthy for its high level of campaign expenses and the negative tone set by both candidates. In the final election, Davis defeated Simon, but he won with only 47 percent of the vote and despite the fact that 1.7 million fewer people voted for him in 2002 than had voted for him in 1998. Simon’s campaign failed to attract potential voters to the polls as well. In the meantime, voters have been increasingly exposed to and enthusiastic about their ability to participate in the “direct democracy” process afforded them by the state constitution—the recall, the referendum, and the citizens’ initiative. In our PPIC Statewide Surveys over the past few years, voters have consistently expressed a high level of satisfaction with the initiative process. Moreover, many voters say that they would prefer that voters make the important decisions for the state through ballot measures (Baldassare, 2002). California state politics and policymaking today rely to a large extent on the initiative process, which allows groups outside state government to make public policy by placing potential laws and constitutional amendments directly on the state ballot for voters to enact or reject. State initiatives may cover a range of topics, from cigarette taxes to restrictions on the sale of horsemeat. However, the initiative process has recently shifted its focus, with an increasing number of statewide initiatives attempting to impose external controls on state government itself (Table 2.6). These initiatives include proposals to impose term limits, to change the primary system, to identify or punish incumbents for certain behaviors, and to restrict campaign fund-raising and expenditures. In the early 1980s, these kinds of initiatives were rarely proposed and even more rarely qualified for the ballot. In the early to mid-1990s, however, the number of initiatives with this focus, both proposed and qualified, increased dramatically (see Baldassare, 2000, 2002; Schrag, 1998). 22 Table 2.6 Citizens’ Campaign Reform Initiatives in California, 1980–2002 Total Number of Initiatives Proposed for Qualified for Qualified for Year the Ballot the Ballot the Ballot 1980 0 0 3 1982 1 0 6 1984 2 1 12 1986 1 1 7 1988 2 0 16 1990 5 1 18 1992 4 1 4 1994 7 0 6 1996 3 3 16 1998 6 2 13 2000 7 2 15 2002 3 1 7 SOURCES: California Attorney General (2002); Silva (2000); University of California Hastings College of Law (2002). NOTES: The total number of initiatives proposed and qualified includes only those proposed by citizens and not those proposed by the legislature. Initiatives counted as related to campaign reform include proposals that attempted to place external controls on government—specifically, campaign finance reform, term limits, campaign advertising disclosure, changes in primary laws, conditioning reelection on balancing the budget, designation of incumbents on the ballot, changes in registration laws, and creating a “none of the above” ballot option. During the 1996 election cycle, an unprecedented three initiatives designed to control legislative behavior appeared on the ballot—one on term limits and two on campaign contribution limits. No more than one similar initiative had appeared on the ballot in any prior election cycle. Since that time, there has been no election cycle where at least one of these kinds did not appear on the ballot. Even larger numbers of related initiatives were proposed but failed to qualify for the statewide 23 ballot, with the number of proposals increasing sharply during the 1990 election cycle. Since that time, the number of proposed initiatives that attempt to control the behavior of state government has never dropped to the levels of the 1980s. When citizens feel connected to government and in control of government activities, there is little need or desire to impose political controls beyond the normal process of representative democracy—voting candidates in and out of office. However, as people become increasingly disconnected from government and their elected officials, the desire to use the tools of direct democracy to impose structural constraints on the executive branch and legislatures grows. The increasing efforts to impose these external constraints over the past decade reflect the feeling that government pays limited attention to the desires of constituents. In the week before the 2002 election, when Californians were asked whether they felt that having elections made government pay attention to what people think, over seven in 10 respondents (71%) said that they felt the elections made government pay either a great deal (31%) or a fair amount (40%) of attention to what people think. But people who were asked how much attention government actually pays to what people think were much less positive about the influence of their views. Over half (53%) said that they felt that government paid not too much attention, or no attention at all, to what people think. The increasing numbers of initiatives designed to impose restrictions on the activities of government, both in office and during campaigns, are consistent with this belief that government is inattentive to the public. What Happens When Candidates Focus on Negative Campaigning? Much of the dissatisfaction that Californians say they feel with political campaigns today is derived from their dislike of a particular type of campaign activity—referred to by political experts as “negative campaigning.” Although people recognize that campaigns involve candidates criticizing their opponents as well as advocating their own election, in the most recent election they felt that this criticism had reached unsupportable levels. 24 Our PPIC Statewide Survey in November 2002 asked half of respondents whether they felt that political candidates should be critical of their opponents and the other half of respondents whether they felt that Gray Davis and Bill Simon, the major party candidates for governor, should be critical of each other (Table 2.7). Respondents who were asked the more general question were split evenly (47% each) between agreeing that criticism was important to learn the strengths and weaknesses of each candidate and disagreeing because campaigns are already too negative. In contrast, a large majority (58%) of those who were asked whether Gray Davis and Bill Simon should be critical of each other said that they should not be, because campaigns are too negative. It is interesting to note that only slightly more than a third of the survey respondents (36%) thought that Davis and Simon should be critical of each other. Candidates may be able to criticize each other without engaging in what is perceived to be negative campaigning. But the 2002 gubernatorial candidates failed to draw the line between criticism and negativity as each sought instead to destroy the credibility of his opponent. Earlier campaigns have not always suffered from the same problem. In a 1996 poll of state voters, 49 percent of respondents said that the Clinton-Dole campaign was “not too negative,” and 46 percent said that it was either somewhat or very negative (Field Poll, 1996). Table 2.7 Californians’ Perceptions of Negative Campaigning Should political candidates be critical of their opponents? Yes, It Is Important to Know the Strengths and Weaknesses of No, Campaigns Each Candidate Are Too Negative 47% 47% Should Gray Davis and Bill Simon be critical of each other? 36% SOURCE: PPIC Statewide Survey (November 2002). NOTE: N = 2,000. 58% 25 These results are consistent with general feelings about whether political candidates should be critical of their opponents, suggesting that feelings about the 1996 campaign simply reflected underlying assessments about the value of candidates criticizing each other rather than a distinctive feature of that particular campaign. The same could not be said about the 2002 gubernatorial race. Measuring the Influence of Attack Advertising: Results of a Survey Experiment Campaign consultants argue that negative campaigning is often necessary. They claim that voters view the candidates who are attacked more negatively, allowing the attacking candidates to increase their chances of winning. It is generally perceived to be necessary as a way to salvage the reputations of candidates who have already been attacked by their opponents. However, very little effort has been made to measure how these attacks affect people’s perception of both the attacking candidates and those who are attacked. The PPIC Statewide Survey in November 2002 attempted to test constituent responses to attack ads by asking respondents to rate their feelings about candidates after hearing that they had either run attack ads or responded to attack ads. The examples used in the survey were based on the actual accusations and actual responses of both Davis and Simon during the 2002 gubernatorial campaign. Each respondent was randomly assigned to hear about a different kind of attack ad, and each could have been asked about his or her perceptions of either the attacking candidate after hearing the ad, or the attacked candidate after hearing the ad, but not about both. Both Davis and Simon attack advertisements were used as examples. This series of survey questions made it possible to measure the effect of running attack advertisements on a candidate’s popularity and to measure the effectiveness of different possible responses to attack advertisements. The candidate who was attacked could claim that the attacker was mudslinging, could make a counterclaim, or could deny the charges. At various points in the 2002 gubernatorial campaign, the candidates did all of these things. As a final follow-up question, respondents were asked 26 how they would have felt about the attacking candidate had he chosen not to run the attack ad, or the attacked candidate had he chosen not to respond to the attack ad. For instance, respondents were asked one of the following questions: Option 1: In this campaign, a Gray Davis advertisement claimed that Bill Simon had engaged in fraudulent business dealings, and in response Simon said that Davis was engaging in mudslinging and negative campaigning. Did this exchange make you feel more positive or more negative about Bill Simon? Option 2: In this campaign, a Gray Davis advertisement claimed that Bill Simon had engaged in fraudulent business dealings, and in response Simon said that Davis was engaging in mudslinging and negative campaigning. Did this exchange make you feel more positive or more negative about Gray Davis? Option 3: In this campaign, a Gray Davis advertisement claimed that Bill Simon had engaged in fraudulent business dealings, and in response Simon said that Davis is a “toll-booth” governor who makes policy only in exchange for campaign contributions. Did this exchange make you feel more positive or more negative about Bill Simon? Option 4: In this campaign, a Gray Davis advertisement claimed that Bill Simon had engaged in fraudulent business dealings, and in response Simon said that Davis is a “toll-booth” governor who makes policy only in exchange for campaign contributions. Did this exchange make you feel more positive or more negative about Gray Davis? Option 5: In this campaign, a Gray Davis advertisement claimed that Bill Simon had engaged in fraudulent business dealings, and in response Simon denied that he was guilty of any wrongdoing. Did this exchange make you feel more positive or more negative about Bill Simon? 27 Option 6: In this campaign, a Gray Davis advertisement claimed that Bill Simon had engaged in fraudulent business dealings, and in response Simon denied that he was guilty of any wrongdoing. Did this exchange make you feel more positive or more negative about Gray Davis? After being asked one of these questions, respondents were asked how a different strategy would have affected their opinions of the candidates: Option A: If Bill Simon had ignored Davis’s commercial instead of responding, would this have made you feel more positive or more negative about Bill Simon? Option B: If Davis had not made this claim about Simon, would this have made you feel more positive or more negative about Gray Davis? Surprisingly, all of the potential responses to attack advertisements were equally unpopular, including denying that the attack was true. The only response that was correlated with more positive rankings of the candidates was not making the attack or ignoring the attack. On average, 45 to 50 percent of respondents said that they perceived a candidate who attacked the other candidate, or who responded to the attack, more negatively. At the same time, 25 to 35 percent of the respondents said that it would make no difference to them. In contrast, approximately four in 10 respondents said they would have felt more positively about either candidate had they not attacked the other or had they ignored attack ads directed at them. Another 30 to 40 percent of respondents said it would make no difference, but this may reflect the fact that it was difficult to perceive a campaign without attack advertisements at this late stage in the campaign. It is important to note that both Davis and Simon were already perceived very negatively in the week before the election. Only 46 percent of respondents said that they viewed Davis favorably, and only 39 percent viewed Simon favorably, and any increase in favorable 28 opinion for either of these candidates would be noteworthy and significant in the final moments of the election. It may not matter to candidates whether their approval ratings decrease when they run attack ads, as long as the potential voters who disapprove are not already their supporters. If those watching the ads will not vote for the attacking candidate anyway, the cost to the campaign of alienating them is minimal. However, when we split the survey into respondents who already supported Davis, and respondents who already supported Simon, it was apparent that negative advertising was affecting the opinions of supporters as well as opponents (Table 2.8). Between 33 and 39 percent of Davis supporters viewed Davis more negatively for attacking Simon and for responding to Simon’s attacks. Between 35 and 39 percent of Simon supporters viewed Simon more negatively when he attacked Davis or responded to Davis’s attacks. Moreover, the nonsupporters of each candidate and those who did not Table 2.8 Perceptions of Negative Campaigning by Supporters and Nonsupporters Negative campaigning When Davis attacks When Davis responds He is By By perceived . . . Nonsupporters Supporters Neither More negatively More negatively 71% 72% 33% 48% 39% 50% When Simon attacks When Simon responds More negatively More negatively 58% 66% 39% 40% 35% 41% Avoid negative campaigning When Davis does not attack More positively When Davis does not respond More positively 37% 35% 47% 35% 51% 25% When Simon does not attack More positively When Simon does not respond More positively 36% 37% 45% 32% 47% 42% SOURCE: PPIC Statewide Survey (November 2002). NOTES: Respondents were asked randomized questions about how they would perceive candidates after viewing Davis and Simon advertisements with a range of characteristics. N = 2,000. 29 support either candidate were even more likely to perceive the candidates negatively for attacking and for responding to attacks. In contrast, even nonsupporters said that they would view Davis and Simon more positively had they chosen not to attack or not to respond to attacks. Between 35 and 37 percent of those who did not support Davis said that they would have viewed him more favorably if he had avoided negative campaigning. A similar percentage of those who did not support Simon said that they would have viewed Simon more favorably if he had avoided negative campaigning. In the 2002 campaign, however, neither candidate managed to profit from this potential goodwill, because both chose to run negative campaigns throughout the course of the governor’s race. The effect of attack ads on supporters indicates the political risks of running negative advertising. Many campaign consultants argue that supporters are not affected by attack ads and will not change their votes even if they dislike the tone of the campaign. But if even supporters are unhappy with candidates who engage in negative campaigning, the ability of negative advertisements to decrease the popularity of the candidate who has been attacked must be viewed against their ability to decrease the popularity of the attacker. In the 2002 gubernatorial race, both candidates concentrated their resources on attacking each other. Although voters were not able to effectively switch to a candidate who campaigned positively, they were able to skip voting altogether. In many cases, they did exactly that. One lesson learned from this survey experiment is that the winning candidate who spent more in this election—Gray Davis—took on risks in the amount of time and resources that were dedicated to attack ads. The strategy may have ensured his narrow victory. However, it came at the expense of a shrinking support base that made Davis vulnerable after the 2002 election. For one thing, a low turnout reduced the threshold for the number of signatures needed to qualify for a recall. Moreover, even some Democratic voters stayed at home or voted to remove Davis from office in 2003. 30 Correlates of Voter Dissatisfaction with Candidate Choices The public’s perceptions of the 2002 gubernatorial election were noteworthy in part because the people who were most alienated were the most knowledgeable about the political process that was taking place. We attempted to measure the factors that were related to expressions of dissatisfaction with the choice of candidates for governor. We considered a number of potential correlates of dissatisfaction such as attention to the campaign, interest in politics, several demographic variables, and such political factors as strong party identification. The results from a logit regression show that dissatisfaction increased significantly with level of attention to the campaign and with the level of interest in politics (see Appendix A). In general, increased political information is supposed to make potential voters more confident about their choices, just as providing more information through campaigns should increase voter confidence in the candidates and voter turnout on Election Day. However, the factors that seemed to make Californians unhappy with the 2002 gubernatorial race appeared to affect even the most informed and interested people. California campaigns appear to have descended to a level of communication that even active voters can no longer tolerate. Information on Election Campaigns Potential voters said that they did not receive the type of information that they needed to make a decision in the most recent California election. However, the information that they find most useful does not seem that difficult to ascertain. In all of the preelection surveys that we have conducted since 1998, Californians have said that, above all else, they would most like to know about candidates’ stands on the issues. In our survey just before the election, Californians chose learning about candidates’ stands on the issues as the most important thing that they could learn in a campaign, rather than information about candidates’ character, political parties, or past experience in office. Notably, however, this demand for information about issues was not met in the 2002 campaign. Although 69 percent of our survey respondents said 31 that they would like most to hear about the candidates’ stands on the issues, fewer than one-third were satisfied with the discussion of issues in the 2002 campaign (Table 2.9). In all of our preelection surveys, relatively few Californians said that they were satisfied with the discussion of issues during the campaign. The picture that emerges from the 2002 campaign is one of extensive voter frustration with the content of communication, which made voters less eager to support any of the candidates. Despite paying more attention to campaigns than in the past, a large share of Californians said that they failed to receive the information that they needed to choose a candidate for governor. Nearly half of registered voters and more than half of eligible voters were no-shows at the polls. On top of frustration with the content of the campaign, however, is frustration with the means that candidates used to deliver campaign messages. When asked how they prefer to learn about candidates, voters chose debates more than any other means of communication (Table 2.10). However, in the 2002 campaign, there was only one debate between the candidates for governor. Making the debate even less relevant, it was scheduled at noon on a Monday, when a large percentage of the working population was unable to watch it. How do Californians feel about the status quo of campaign communications today? Over four in 10 Californians said that they would view candidates who primarily used television advertising to get their message out to voters less favorably; another 20 percent said it would make no difference to them. Table 2.9 Californians’ Preferences and Satisfaction on Attention to the Issues Preference Percentage Would most like to know candidates’ stands on issuesa Satisfied with discussion of the issues in the campaignb 69 30 SOURCES: PPIC Statewide Survey (November 2002, October 2002). aN = 2,000. bN = 2,007. 32 Table 2.10 Californians’ Preferred Means of Candidate Communication Means of Communication Percentage Debates TV call-in shows Speeches Town hall meetings Mailers Door-to-door 34 16 16 15 8 3 SOURCE: PPIC Statewide Survey (November 2002). NOTE: N = 2,000. Californians’ feelings about the information provided by a largely ad-driven campaign were quite negative. In particular, when asked about their impressions of political campaigns in general, as well as the 2002 gubernatorial campaign in particular, an overwhelming number of respondents volunteered that the biggest problem with campaigns in general, and with the 2002 campaign specifically, was the emphasis on negative advertising. And by nearly any measure, the tone of the recent California gubernatorial campaign was overwhelmingly negative. It is important to note here that paid advertising is obviously not the only source of information on elections. The media play an important role in providing information on candidates and election events. We know from our current study that many residents closely followed the news stories about the 2002 governor’s election, but it is unclear if the attack ads had an influence on the types of election stories that the media covered. Nor is it clear if the dissatisfaction with the tone and substance of this campaign that was elicited in our surveys can be at least partly attributed to the media coverage of the attack ads and negative campaigning. Future studies should focus on the combined effects of the media stories and campaign information on dissatisfaction with campaigns and elections in California. 33 Conclusions The dissatisfaction that Californians felt with the most recent gubernatorial election is historically noteworthy and politically significant. Moreover, it appears to have extended beyond simple protestations of unhappiness. Lacking the information that they felt they needed to make a decision, and unhappy with the candidate choices and the means by which campaign messages were communicated, an unprecedented number of voters chose not to appear at the polls at all. However, it should be noted that the 2002 election is typical and not an aberration. The findings reported here are symptomatic of trends we found both before and after the Davis-Simon race—and are ever-present in the wake of the governor’s recall in the fall of 2003. The failure of political campaigns to satisfy the demands of potential voters has implications for the democratic system. Part of the purpose of any political campaign is to bring politicians seeking elected office in touch with constituents and to serve as a mandate for continuance of past policies or the creation of new ones. One expectation placed on candidates is that they will provide information that will allow voters to make those decisions, either directly through their own campaigns or indirectly by criticizing their opponents. The competition inherent in a campaign is designed to allow voters to hear at least two platforms, one of which will direct the politician who is elected. However, when both sides of a campaign provide information that voters find only irrelevant or off-putting, constituents have few options for protest other than withholding their votes. We find that candidates hurt their own images by campaigning negatively and providing inadequate information to voters. Equally important, attack campaigns, even beyond their effects on turnout, harm democratic legitimacy. Are feelings about campaigns correlated with trust in government and, specifically, does unhappiness with campaigns lead to decreased trust? Although the effects of lowered trust in government may be difficult to pin down, a government that does not enjoy the support of its constituents may have difficulty enacting policy, and may face increasing external controls such as term limits. In the case of California, there has 34 been an increasing trend for voters to circumvent the legislative process altogether through the citizens’ initiative process. Dissatisfaction with the recent gubernatorial election appears to have generated a new effort to impose external controls on officeholders, with the first successful movement in California to recall a sitting governor. We look for further evidence of a direct link between perceptions of political campaigns and feelings of political distrust in the next chapter. 35 3. Political Campaigns and Voter Distrust Political observers have long suspected that trends in campaign practices such as negative advertising might lower the public’s capacity for political trust. In the California context, the voters’ dissatisfaction with political campaigns that we noted in the previous chapter has mounted against a backdrop of declining trust in government. The assumption underlying many campaign reforms, ranging from campaign finance regulation to pledges of clean conduct, is that they will serve to regenerate trust in government. To date, however, there is no evidence that any political or election reforms that have been enacted have actually increased the level of political trust. The question that we address here is whether trust in government increases when candidates engage in more desirable campaign behavior and decreases when they engage in less desirable campaign behavior—that is, political practices that stray widely from public ideals or expectations about appropriate behavior for candidates for elected office. There are several reasons to study the link between political campaign practices and political distrust. First, given the frequency of election campaigns in recent years, there are many different types of campaigns to observe, and there is no shortage of cases that are negative in their tone and tactics. An analysis of political campaign trends thus offers a potential explanation for increasing levels of distrust in government. Next, a consideration of the effects of negative campaigns on political trust offers the opportunity to review whether some of the more practical claims about the consequences of a decline in political trust are valid. For example, are constituents who trust government less really more likely to limit legislative activities and oppose an expansion of government’s role? A range of potential consequences of low trust in 37 government has been discussed, but there is little known on whether and how much distrust in government has real policy and political implications (see Baldassare, 2002; Citrin, 1974; Miller, 1974; Moore, 2002). The decreasing trust in government inspired by negative campaigns and voter dissatisfaction with the election process have larger implications for the prospect of governance in California. Specifically, low trust in government may decrease democratic legitimacy and thus make it increasingly difficult to enact effective public policy. It may also encourage voters to make state laws on their own through the initiative process. The role of political distrust during the 2002 gubernatorial election in California is worthy of our careful attention. This election was a political event that received national attention for its record-setting spending on a campaign that was uniformly described as overwhelmingly negative and that generated critical comments about both major party candidates. In the final analysis, the record low voter turnout raises questions about the role of political distrust. Indeed, trust in the state and federal government was relatively low during the 2002 election. Moreover, trust in the state government declined further in the months after November 2002 leading up to the October 2003 recall election and, as of this writing, has yet to recover to the pre–November 2002 levels. These trends raise another issue—whether and how much trust is regained once it is lost. The Reasons for Public Distrust Trust in government depends on a variety of factors that can be loosely organized into three general categories: (1) assessments of current political leadership and the political system; (2) individual characteristics; and (3) levels of political engagement. Past research has raised questions about whether assessments of political leaders influence political trust in a significant way. Initial research on political trust found that individual characteristics, such as a person’s age and level of education, were the main determinants of trust. However, further research on perceptions of trust in government suggested that trust varies over time in response to political conditions. 38 After the Watergate scandal, Citrin (1974) argued that levels of political trust were correlated with feelings about political leaders, especially at the presidential level (see also Baldassare, 2002; Dionne, 1991; Lipset and Schneider, 1983; Pew Research Center for the People and the Press, 1998). Specifically, those who disapprove of political incumbents are less trusting than those who approve of incumbents. Researchers have found that individuals may base their trust in government at least partially on their assessments of features of the political system, such as campaigns and elections (Buchanan, 1996, 2000; Rahn, Brehm, and Carlson, 1999; Simon, 2003; Williams 1985). What does the experience of an actual election do to one’s level of political trust? Some research suggests that certain campaigns can increase political trust (Bartels, 2000; Bartels et al., 2000; Buchanan, 2000; Clark and Acock, 1989; Ginsberg and Weissberg, 1978; Rahn, Brehm, and Carlson, 1999). And equally, there is speculation that unattractive campaigns—specifically those that are negative or vacuous—depress trust in government (Buchanan, 1996; Simon, 2003). Some evidence even suggests that distasteful political campaigns lead the electorate to participate less frequently in politics (Ansolabehere and Iyengar, 1995; Rosenstone and Hansen, 1993). Certain campaign reforms, such as campaign finance reform, may thus increase political trust. In general, then, different elements of campaign and election experiences appear to be significantly related to political trust. As we have noted above, political trust varies with individual characteristics, either alone (Miller, 1974) or in conjunction with assessments of political leaders (Citrin and Green, 1986). Underlying political beliefs, sometimes related to social and economic characteristics, also influence individual perceptions about government. Some people may have learning experiences that lead to trust in government, leading to generational differences in trust that are traditionally measured by age (Hetherington, 1998; Miller and Shanks, 1996). Individuals may also feel more trusting because of their ideological or partisan agreement with political leadership (Hetherington, 1998). Finally, researchers have found that demographic characteristics such as race and ethnicity, education, income, and gender are correlated with different levels of trust in government (Hetherington, 1998). Latinos are more trusting than 39 those from other racial and ethnic backgrounds (Cohen, Baldassare, and Kaimowitz-Rodriguez, 2003), whereas those with more education may be more (Hetherington, 1998) or less (Finkel, 1985) trusting. In keeping with the assumption that those in agreement with, or profiting from, current economic and political conditions are more trusting, those with higher incomes are typically more trusting in government. Finally, some evidence suggests that individual levels of political engagement are correlated with political trust. Most of this research argues that exposure to politics and government will increase trust (Hetherington, 1998), although some researchers argue that the effect is reversed, and those who know more about government are less trusting (Bowler and Donovan, 2002). In addition, higher levels of political participation, such as voting or attending political meetings, have been correlated with higher levels of political trust (Finkel, 1985), particularly among respondents with less education. In our study, it is noteworthy that some of the assessments of trust in government were measured in the context of a heated political campaign—that is, the 2002 gubernatorial election in California. But it is important to note that this was also a time of highly negative perceptions about the state’s economy and fiscal conditions. At the same time, California and the rest of the nation were a year out from the September 11, 2001, terrorist attacks in New York City and Washington—an event that was initially followed by higher levels of reported trust in the federal government and a significant spike in presidential approval ratings. For a variety of political and economic reasons, then, despite the fact that people typically trust their state government more than they trust the federal government (Bowler and Donovan, 2002), by the fall of 2002, at the time of the gubernatorial election campaign, Californians’ trust in state government had dropped to levels below their trust in federal government. Survey Evidence on Distrust in Government In this section, we analyze the results of the PPIC Statewide Survey that was conducted in the week before the November 2002 California gubernatorial election. This random-digit-dial telephone survey of 2,000 California adults interviewed in English or Spanish included subgroups 40 of 1,526 respondents who claimed to be registered voters and 1,025 who were deemed through their responses to be likely voters in state elections. Along with a series of questions designed to measure perceptions of candidates and campaigns (discussed in Chapter 2), all of the respondents were asked about their level of trust in state and federal government, their interest in politics, and their level of political engagement. These particular survey questions are the focus of the analysis in this section. The survey asked two specific trust questions. The first referred to trust in federal government, using the standard American National Election Studies format noted below. The form of the trust in state government question was similar to that of the trust in federal government question, but it referred to the government in Sacramento rather than the government in Washington. Responses indicating trust or lack of trust were coded the same as the responses to the trust in federal government question. Federal government: “How much of the time do you trust the government in Washington to do what is right? Always, most of the time, or only some of the time?” State government: “How much of the time do you trust the government in Sacramento to do what is right? Always, most of the time, or only some of the time?” In the analysis below, those who said that they trusted government all of the time or most of the time were considered to be trusting, and those who said that they trusted government only some of the time, none of the time, or they did not know were considered not to trust government. Although a fourth category, “none of the time,” was not provided to respondents, it is included in the response categories because people asked about their trust in government in recent years frequently volunteered this answer. In the current survey, 5 percent volunteered “none of the time” for the federal government, and 4 percent volunteered “none of the time” for the state government. Only 1 percent said that they had no opinion about the state or federal governments. 41 The PPIC Statewide Survey has been asking about trust in federal government nearly every year since the survey’s inception in 1998. The trust in state government question, however, has had a shorter though more intense recent history. The trust in state government question was first introduced into a PPIC Statewide Survey in January 2001 as part of a survey series following the 2000 presidential election. The first instance of this question preceded the California energy crisis, which would be expected to decrease trust in state government, as well as the September 2001 terrorist attacks, which tended to increase trust in all levels of government, at least temporarily. In the past two years, questions about trust in both the state and federal government have been asked every few months. Two questions were used to consider respondents’ political assessments. The first asked how favorably respondents perceived the Democratic incumbent governor, Gray Davis, who was running for a second term of office against Republican challenger, Bill Simon, using the following answer categories: Very favorable, somewhat favorable, somewhat unfavorable, very unfavorable, and no opinion. A second set of questions asked whether political campaigns in California had been getting better or worse in recent years, with the following wording asked of a random half-sample of the 2,000 adults: “More generally, overall do you think that election campaigns in California have gotten better in the last 10 years, gotten worse, or stayed about the same?” “More generally, overall in terms of ethics and values, do you think that election campaigns in California have gotten better in the last 10 years, gotten worse, or stayed about the same?” These questions were recoded into two response categories for the purposes of the analysis that follows in this chapter. We contrast those with a favorable view of Davis (i.e., very favorable or somewhat favorable combined) against all other answers (i.e., somewhat unfavorable, very unfavorable, no opinion). The perception that political campaigns in California were getting worse was contrasted against all other answers 42 (i.e., getting better, staying the same, do not know). As for the two versions of the campaign quality question (i.e., overall and in terms of ethics and values), because the results were similar using the different wording, we combined the responses for our analysis into one question that measures the perceptions of campaign quality in the past 10 years. Our measures of individual characteristics—including age, sex, race, income, education, and party identification—were thus taken from a range of questions on demographic and political characteristics. Other questions in our analysis measured political engagement. These included questions about whether respondents had engaged in any of a series of political activities, whether they had seen advertisements by the candidates for governor, how much attention they paid to politics, and how much attention they paid to news about the upcoming election. Finally, respondents were asked whether they could correctly identify the position of three political figures—U.S. Senator Dianne Feinstein, Secretary of State Colin Powell, and the Prime Minister of the United Kingdom, Tony Blair. These questions were designed to measure the respondents’ levels of political knowledge and attentiveness to politics. The questions on political activities were derived from the Roper organization’s series on political participation. Specifically, we asked whether the respondents had, in the last six months, written to a political representative, attended a political rally, attended a meeting on political affairs, signed a petition, worked for a political party, given money to a political party or candidate, or been a member of a political organization. These measures were summed to an eight-point scale ranging from zero activities to all seven activities. Measures of attention paid to politics and attention paid to the news spanned a four-point range from little or no interest to a great deal or a fair amount of interest. Finally, the measure of political knowledge ranged from zero (none of the three political figures correctly identified) to three (all three correctly identified). The preelection survey questions made it possible to test expectations for the three general hypotheses that attempt to explain what factors affect trust in government. 1. Political assessments: Trust in state government is positively correlated with support for the incumbent governor, Gray Davis. Trust 43 in state government is negatively correlated with feelings that election campaigns are getting worse. 2. Individual characteristics: Trust in state government is negatively correlated with being a strong Republican or an Independent and is positively correlated with identification as a strong Democrat. Both women and Latinos are more trusting of state government, whereas younger residents are less trusting of state government. 3. Political engagement: Residents who pay attention to the news, who have a greater interest in politics, or who know more about politics are more trusting of state government. Those who have seen the paid political advertisements by the candidates for governor, Gray Davis or Bill Simon, are less trusting of state government. Trends in Distrust in State Government Californians’ trust in state government declined during the 2002 election campaign and has not returned to previous levels as of this writing. The evidence from the PPIC Statewide Survey in November 2002, as well as past studies of trust in government in California, suggests that campaign perceptions are related to attitudes toward government. Specifically, negative assessments of election campaigns are correlated with lower levels of trust in government. Moreover, political assessments of incumbents and campaigns appear to provide stronger correlations with declining trust than either individual characteristics or political engagement according to our statistical analysis below. Californians’ trust in state government has been declining over the past three years, whereas trust in the federal government has remained relatively stable. This distinction between state and federal trust suggests that although our survey questions asking about trust in these two levels of government are similar, they are effectively measuring two different phenomena rather than an underlying level of trust in all levels of government. In addition, it suggests that some phenomena specific to California may have recently changed the levels of trust in state government. It is possible that the 2002 election was one of the major factors contributing to this decline. 44 Trust in the federal government expressed by California residents has increased since the first PPIC Statewide Survey in April 1998. In the spring of 1998, only about one-quarter of Californians said that they trusted the federal government always or most of the time (as shown in Table 3.1). By September 1998, trust in the federal government had increased to approximately one-third of all respondents and remained at or near that level through the end of 2000. In January 2002, after the September 11th terrorist attacks, the percentage of those saying that they trusted the federal government leaped to 46 percent of all respondents. Although trust in the federal government declined from this high several months later, it still remained at a higher level than in the past. By August 2002, during the 2002 election campaign, almost four in 10 respondents said that they trusted the federal government (37%), and trust remained at this level both during and after the state elections. Californians’ trust in the federal government has shown a slow but steady Table 3.1 Californians’ Trust in the Federal Government, 1998–2004 % Who Trust Federal Government Always or Date Most of the Time April 1998 May 1998 September 1998 December 1999 January 2000 October 2000 January 2002 August 2002 November 2002 February 2003 February 2004 26 24 33 32 33 33 46 37 37 37 30 SOURCES: PPIC Statewide Surveys (1998–2004). NOTE: N = approximately 2,000 per survey wave. 45 increase over the past five years and into 2003, but it declined recently in February 2004. In contrast, trust in state government has been moving in the opposite direction (Table 3.2). For the years measured, state government trust was at its peak in 2001 and early 2002, with nearly half of all respondents saying that they trusted state government (47%). Notably, these years were difficult for the state government and for the reputation of the incumbent governor, Gray Davis. In 2001, California experienced a massive energy crisis, which was blamed on the state’s energy deregulation plan and subsequent mismanagement by Governor Davis. And from 2001 to 2003, California experienced economic recession, state budget problems, or both. Despite these issues, trust in Table 3.2 Californians’ Trust in State Government, 2001–2004 Preelection January 2001 January 2002 % Who Trust State Government Always or Most of the Time 47 47 November 2002 election season August 2002 November 2002 37 35 Post–November 2002 election February 2003 June 2003 September 2003 October 2003 36 34 28 27 Post–October 2003 recall January 2004 May 2004 27 32 SOURCES: PPIC Statewide Surveys (2001–2004). NOTE: N = approximately 2,000 per survey wave. 46 state government first showed a decline during the 2002 election season. By August 2002, fewer than four in 10 Californians said that they trusted the government in Sacramento (37%). In the following months, trust remained at or below this level, with only 34 percent of Californians saying they trusted state government by June 2003. During the recall election, trust in the state government declined further, and it has not rebounded to the levels seen in the November 2002 preelection survey as of this writing. Some argue that it is difficult to attribute any changes in trust in state government coincident with a political campaign to anything but the campaign itself. Rahn, Brehm, and Carlson (1999) argue that changes in political trust during campaigns should be attributed to campaigns rather than to other factors because there typically are no other equally significant political events occurring at the same time. Although California had experienced a number of political problems in recent years—notably a serious budget deficit and an electric power crisis—in the fall of 2002 the major political event was definitely the gubernatorial election. Other statewide and federal offices were in contention at this time—including a U.S. Senate race and Congressional and state legislative races—but there is no question that the gubernatorial race drew the most attention from the media. Although it could be argued that people feel more negatively about campaigns because they are less trusting, the decline in trust that occurred during the 2002 election campaign argues that the election campaign affected political trust rather than the reverse. For instance, only the ratings of trust in state government changed over time, whereas trust in the federal government remained stable. As a result, a correlation between trust in state government and negative feelings about political campaigns should largely reflect the effect of campaigns on political trust rather than the effects of low trust on feelings about political campaigns. Even if other factors may be driving the loss of confidence in state government, the causal relationship appears clear in the case of state government. 47 Correlates of Distrust in State Government To examine the potential relationships between various factors and trust in state government, we conducted a logit regression analysis for all survey respondents, using independent variables that measured political assessments, individual characteristics, and political engagement. The numerical results are reported in Appendix B. Overall, we found that political assessments, of both the incumbent governor, Gray Davis, and perceptions of the campaign, were significantly correlated with trust in state government. In addition, certain individual characteristics, such as party identification and ethnicity, were also related to political trust. In contrast, only one measure of political engagement—the level of interest in politics—was significantly related to trust in state government. We discuss these findings in the remainder of this section. It is important to note that both of our measures of political assessments were significantly correlated with trust in state government. One of the strongest predictors of trust was the perception of political leaders; individuals who had a favorable opinion of Gray Davis were much more trusting of state government. In contrast, those who felt that political campaigns in California had been getting worse were significantly less trusting. A few measures of individual characteristics were also correlated with trust in state government. When we accounted for attitudes toward the Democratic incumbent governor, Gray Davis, both Independent voters and strong Republicans expressed less trust than others in government. In contrast, measures of conservative political ideology—gender, age, education, and income—were unrelated to lower levels of trust. Many researchers have argued that there is a generation gap in trust, with younger residents trusting in government less than older generations, but this did not appear to be the case during the 2002 gubernatorial election campaign. As for race/ethnicity differences, Latinos were more trusting of state government than other residents were. The latter results are consistent with past research on Latinos in California, which argues that Latinos consistently feel more positive about both federal and state government than do residents of other racial/ethnic backgrounds (Baldassare, 2002). 48 Finally, as noted above, only one measure of political engagement— the level of interest in politics—was correlated with trust in state government. In this case, residents who were more interested in politics reported higher levels of trust. Other measures of engagement, including attention paid to the news, exposure to campaign advertising, and the level of political knowledge, were not significantly correlated with trust in state government. Despite our expectations that these other measures would be correlated with political trust, when other factors were considered they did not appear to be relevant. Although the opinions of all residents clearly matter, political campaigns often target likely voters rather than the population at large. If feelings about political campaigns are less important to likely voters than they are to the general population, candidates may not feel that the loss of trust is relevant to their decisions about how to campaign. The observation that those who are more interested in politics were more trusting suggests that likely voters may not be influenced by the same factors that affect the population at large. Were likely voters less trusting in response to negative perceptions of the gubernatorial campaign? We ran an identical logit regression model only for likely voters, and the numerical findings are reported in Appendix C. The findings indicate that political assessments had the same effect on trust for likely voters that they did for California adults in general. However, measures of individual characteristics and political engagement were less significant for likely voters’ reports of their trust in government. In sum, the political assessments of the candidates and elections were correlated with likely voters’ levels of trust in government the same way that they were with all adults. Those who felt more favorably toward Gray Davis were significantly more trusting in state government. Those likely voters who felt that campaigns were getting worse, rather than staying the same or getting better, were significantly less trusting. Individual characteristics, however, were much less relevant to political trust for the population of likely voters than they were for all adults. The only measure that was significantly correlated with political trust for likely voters was being registered to vote as an Independent or “decline to state.” As in the general population, Independents felt significantly less trust in state government than those who identified with 49 political parties. However, other measures, such as a strong Republican Party affiliation, or identifying as Latino, were not correlated with trust for likely voters. Finally, none of the measures of political engagement were significant predictors of political trust for likely voters. Some of this effect may result because likely voters in our survey sample are partially identified on the basis of the question measuring interest in politics. The effects of political engagement measures may be more evident in the general population than among voters (Finkel, 1985). However, even in the general population, only one of these measures had any noticeable effect on political trust. Overall, our expectation that assessments of political leaders and the political system would influence the levels of trust in government seems to be borne out in the analysis of our November 2002 survey. In contrast, only certain measures of individual characteristics appeared to affect political trust in a significant manner. The level of a respondent’s political engagement on a number of dimensions did not appear to affect trust in government much, although the level of interest in politics had some effect. Certainly, many other factors were driving the levels of distrust toward historical lows in the months leading up to the November 2002 election. In the California context at this particular time, the unattractive campaign for governor appears to have contributed to a decline in trust in state government. Moreover, trust in state government remained at its new low after the close of the 2002 campaign, and it was to drop even further as California voters confronted the historic events of 2003—the largest state budget deficit in history and the first recall of a governor. Conclusions Most Americans report having more trust in their state government than in the federal government, but California residents have shown a pattern of trusting their state government less than, or at least no more than, they do the federal government. This trend became evident during the 2002 governor’s race. The empirical evidence presented in this chapter demonstrates that campaign perceptions are significantly related to the public’s level of 50 trust in state government. The relationship remains after accounting for individual characteristics and various political factors. At the close of the 2002 governor’s race, Californians with more negative perceptions of the campaign also had lower levels of trust in state government. Is trust in government relevant to the complex issues involved in governing? Some say that it is not, arguing that voter apathy and disinterest can actually lead to a more politically stable environment and that low trust in government is unrelated to the ability to govern effectively (Berelson, Lazarsfeld, and McPhee, 1954; Dahl, 1956; Moore, 2002). However, others suggest that political trust and legitimacy are important factors in governance. The observable effects of political distrust can range from legislative gridlock, to difficulty in passing public policy measures at the ballot box, to efforts to impose political reforms on candidates and elected officials (Hetherington, 1998; Teixeira, 1992). Although it is difficult to link the decline in trust in state government directly to any particular set of policy outcomes, the recent history of California politics suggests that the more dramatic predictions about the effects of the loss of political trust may have some external validity. The current state of distrust in California politics may be a major contributing factor to the demand for structural change through the direct democracy process, such as citizens’ initiatives that call for legislative term limits, fiscal constraints on elected officials, and administrative change in the case of the Davis recall. At the very least, the current level of distrust provides a steady foundation for political change. What is even more striking in the California example reported here is the further drift downward in trust in state government after the November 2002 election. In other words, the decline in trust that took place before the election was not a temporary phenomenon. Moreover, the lack of recovery to earlier levels of trust in state government several months after the October 2003 recall is noteworthy, especially because Governor Arnold Schwarzenegger has high job approval ratings (64% of all adults; 69% of likely voters) according to the PPIC Statewide Survey of May 2004. This trend raises fundamental questions of what it takes to repair trust when the public has lost a high degree of confidence. 51 California residents, when asked about the current political process in the state during the course of the election, expressed strong feelings that political reform was needed. Many said that campaigns were too negative and that elections were too expensive. In the absence of desirable political campaigns and candidates, the state’s politics may suffer negative consequences from the distrust and alienation that was evident in November 2002 that extend far past a single election. For instance, this could result in less interest in registering to vote, a declining turnout of voters in elections, and voters’ attraction to citizens’ initiatives that limit the powers and authority of the state’s elected officials. In the concluding chapter, we consider the types of political and election reforms that have public support, and we assess their chances for addressing the current trends of negative campaigning, low voter turnout, and distrust in government. 52 4. Conclusions on Public Support for Campaign Reforms Given that California voters are unhappy with the way that campaigns and elections are run in the state today, the question is what, if anything, would they like to do about it? Moreover, are the state’s voters so dissatisfied with the political status quo that they would be willing to increase their taxes to pay for campaign finance reforms? These are the types of public policy preferences that we explore in this final chapter. For many of the possible campaign reforms, there could be legal resistance for government regulation because of first amendment speech and association rights. So any election reforms one might want to propose have to avoid constitutional problems and court challenges by politicians, parties, and consultants who think that they benefit from the ways in which campaigns are run today. For example, our survey indicates that voters are dissatisfied with negative advertising and personal attacks. An outright ban on negative advertising might be appealing to a majority of California voters, but it would certainly not be constitutional and would surely face legal challenges by those who make a living off political commercials. Among a smaller set of feasible political reforms, we tested public attitudes toward four categories of measures: campaign finance restrictions, promoting more spontaneous and personal forms of candidate-voter contact, public disclosure, and voluntarily adopted codes of campaign conduct. We review our findings for each of these reforms below. Most of these reforms are not new. They have been the subject of lengthy debate and careful discussion in California, in other states, and in national commissions. In many instances, the political will to 53 implement these changes has been lacking. The political establishment has doubted the public’s interest in campaign and election reforms. Our results indicate that—contrary to conventional wisdom of indifference toward the political process—the public does want to see substantive change in campaigns and elections. California voters most strongly favor more debates and would like to see the candidates adopt some codes of conduct. It is somewhat surprising that in the light of past polls and elections, Californians in the heat of the 2002 governor’s race, and especially likely voters, were even open to public financing as an option in the future. Apparently, many voters are even willing to pay more to make improvements in what they see as an election process that is broken and badly in need of change. In this chapter, we learn that many voters are eager to embrace proposals to improve California elections (see Baldassare, Cain, and Cohen, 2003a, 2003b; Public Policy Institute of California, 2003). Campaign Finance Reform The issue of campaign finance was prominent in the 2002 governor’s race. Gray Davis’s aggressive fund-raising tactics made more than a few Democrats uneasy and raised questions about his ethics in the media, eventually causing Davis to cancel some events and slow down the pace of money-raising. Against the backdrop of federal action on the Bipartisan Campaign Reform Act, Californians might question the relatively permissive regime we have in this state. Before the passage of Proposition 34, the state had only disclosure laws and no contribution limits or public financing. Proposition 34 introduced relatively high limits on individual and group contributions but not public financing. Moving to new campaign finance reforms may be difficult according to our survey just before the November 2002 election. To begin with, voters are conflicted when asked what they think contributors should get for their money. About 49 percent of likely voters believe that candidates should do nothing or merely say thank-you for the contributions they get, whereas 39 percent of likely voters think that candidates should give contributors either the policies they want or more access to themselves. To put it another way, there is a real divide 54 between the large proportion of voters who think that the current system of campaign finance is ethically acceptable and the substantial percentage of voters who think it is not. In recent years, some wealthy candidates have tried to turn the public’s distrust about candidates’ accepting contributions to their advantage by claiming that their personal wealth frees them from special interest ties. Apparently, this point of view is not shared by many of the state’s voters. Our survey indicates that fewer than one in five Californians thought that it was best for candidates to use their own money to finance their campaigns (i.e., the plutocrat’s virtue), whereas 38 percent thought that it was best to get money from supporters and 37 percent thought that money from the public was the best. And perhaps most surprising, given the fiscally conservative nature of Californians, 57 percent of likely voters thought that public financing was a good idea (Table 4.1). Conventional wisdom by the political establishment in California has been that public financing was a nonstarter. Among other reasons, why would people want to pay for speeches by candidates that they disagreed with or openly opposed in the heat of an election? But the experience of the 2002 election, and perhaps the fact that several states and some California cities have experimented with public financing Table 4.1 Californians’ Support for Funding Political Campaigns “Would you favor or oppose having a system of public funding for state and legislative campaigns in California, even if it cost taxpayers a few dollars a year to fund?” Favor Oppose Do not know % of California Adults 50 46 4 % of Likely Voters 57 39 4 SOURCE: PPIC Statewide Survey (November 2002). NOTE: N = 2,000. 55 schemes, seems to have made many voters receptive to this option for political reform.1 Candidate Debates Californians have been telling us in surveys in recent years that they are frustrated by the distance and superficiality of television-driven statewide contests. On Election Day in November 2002, many voters apparently voted with their feet by staying away from the polls. What would they prefer instead? Several questions in the survey deal with this topic. Clearly, voters would like a more intimate form of campaigning than is feasible above the local level in this state: Fifty-six percent of likely voters said that they would feel more favorably about candidates who campaigned primarily by going from door to door. Unfortunately, with Assembly, Senate, and Congressional districts of over one-half million constituents, and a statewide electorate of 15 million people, door-to-door campaigning is not a realistic option for all but the physically fittest of candidates. Although it is unlikely that voters will get to know statewide candidates more personally, it might be possible for them to witness candidates in more potentially spontaneous settings than the 30-second television commercials that they are providing to voters today. The dominant trend in modern campaigning is to program candidates to avoid mistakes and follow formulaic answers. The ideal for some political consultants would be to create the scripts, have their clients well rehearsed in private, and then have the candidates read them in public. Television advertisements rarely give much specific information about candidates and their positions. Settings where reporters get to ask unscreened questions of the candidates, or where there is the possibility that a candidate might look bad in a real-life interaction with actual voters, are avoided by many political consultants at all costs. Although some reporters get to ask unscripted questions in interview settings, _____________ 1In a more recent survey, majorities of state residents (57%) and likely voters (54%) said that they were opposed to spending a few dollars a year to support public funding for state and legislative campaigns (PPIC Statewide Survey, September 2004). 56 others are exposed to the candidates in only highly controlled environments. Even debates seem to pose “excessive” risks to some candidates. The incumbent governor agreed to only one debate, televised in a single media market in the middle of the day during the 2002 governor’s race. Although no one believes that the answers in debates are unscripted, debates do present a chance for large numbers of voters to compare the answers of the candidates on specific policy questions. When asked whether more debates in the 2002 gubernatorial contest would have made the election campaign better or worse, nearly two-thirds of likely voters said it would have made it better. When asked if they would favor or oppose an initiative that would require five prime-time broadcast gubernatorial debates, 60 percent said they favored the idea (Table 4.2). Thus, support for proposals that provide more contact with the candidates suggest yet another clear area of reform. It is unlikely that a measure that would require debate participation would survive a court test. A more defensible measure would be to set up a debate format and a commission to administer it and then hope that public expectations would induce participation in a series of highprofile debates and town hall forums by all of the major candidates. Table 4.2 Californians’ Support for Gubernatorial Debates “Would you favor or oppose an initiative that required candidates for governor to participate in a minimum of five prime-time publicly broadcasted debates?” Favor Oppose Do not know % of California Adults 60 33 7 % of Likely Voters 56 39 5 SOURCE: PPIC Statewide Survey (November 2002). NOTE: N = 2,000. 57 Public Disclosure Disclosure is a key component to a well-functioning electoral system. Voters need to have access to good information if they are going to make good choices. Although the media often ask for the fullest possible disclosure, candidates sometimes resist on the grounds that subjecting key pieces of personal information to public scrutiny seems like an unnecessary invasion of their privacy. Candidates who enter the public domain must necessarily surrender the privacy they enjoyed as private citizens. But does that mean that candidates have no right to privacy? Just because the press wants to have the material for a story does not mean that the information about a candidate necessarily serves an important democratic purpose. Where does one draw the line? The line has moved toward shrinking the zone of privacy for candidates. A good example is the controversy over Bill Simon’s tax forms during the 2002 California governor’s race. Having filled out the conflict of interest forms required by the state, Simon argued that since he did not have to release his complete tax forms, he would not. Members of the press argued that because Simon had no experience in public office, and was running for office as a businessman and outsider, the public had a right to know more about how he earned his money and how much he paid in California taxes. Moreover, a jury verdict on a business deal and an investigation by the Internal Revenue Service of his company’s affairs in a year of corporate scandals raised the media’s interest in his tax records. In the end, reporters were allowed to review Simon’s taxes in a highly restricted setting. We asked in our survey whether the respondent would feel more favorably, less favorably, or no different toward candidates who released their tax forms, their medical records, or information about campaign contributions as soon as they received such contributions. Given the furor over Simon’s taxes, the results are somewhat surprising. A majority of likely voters (53%) said that they would feel no different about candidates who released their tax forms, whereas about four in 10 said that it would make them look at the candidates more favorably (Table 4.3). An even larger percentage of likely voters (79%) said that they would feel no different about candidates who released their medical 58 Table 4.3 Californians’ Support for Public Disclosure No difference if candidate: Publicly released tax returns Publicly released medical records % of California Adults 53 77 % of Likely Voters 53 79 More favorable view if candidate: Publicly released a list of campaign contributors and amounts received as soon as candidates got them 66 SOURCE: PPIC Statewide Survey (November 2002). NOTE: N = 2,000. 71 records—an issue that came up in the 2000 presidential campaigns of John McCain and Bill Bradley—whereas only about one in five respondents said that this would make them look at the candidates more favorably. By far the more relevant piece of personal information for evaluating candidates seems to be the immediate disclosure of campaign contributions: Seventy-one percent of likely voters said that this would make them more favorably disposed toward a candidate, whereas about three in 10 said that this would make no difference. Obviously, this type of private, financial information from the candidate is judged by the public as having the most direct relevance to the current campaign and future policy directions. This survey finding points to a possible direction for political reform. California has taken some initial steps toward electronic disclosure in recent years. But there are questions about the expense and user-friendliness of the software. Also, there is no unique donor identification that would allow reporters to track contributions at several levels simultaneously. The expense of running for office in California and the wealth of groups and individuals who are would-be donors create a need to follow the money trail in a much more systematic fashion. Who supports which candidates financially and at what level are important pieces of information California voters need to understand the 59 likely influences that successful candidates will face after they are elected to office. Of course, the media would also probably need to increase its role to present such information in a form that was readily available and understandable to the state’s voters. Candidate Codes of Conduct Candidates for elected office cannot be compelled to avoid personal attacks and negative campaigns for first amendment reasons. But perhaps they can be induced to do so by signing pledges that they could be held accountable for during the course of a political campaign. Our survey suggests that California voters are open to this idea. Seventy-seven percent of likely voters said that they would view candidates more favorably (combining the much and somewhat categories) if they signed a code of conduct to “run a truthful, fair, and clean campaign” (Table 4.4). Similarly, over 70 percent of likely voters said that they would view a candidate more favorably if they signed a code of conduct to run “an issue-oriented campaign.” The support for reforms among likely voters was almost equally high for candidates who would sign a code of conduct “not to use race, gender, religion, sexual orientation, or age as a basis for attack.” Table 4.4 Californians’ Support for Codes of Conduct “If a candidate signed a campaign code of conduct pledging to run a truthful, fair, and clean campaign, would this make you view him or her much more favorably, or somewhat more favorably, or would it make no difference to you?” Much more favorably Somewhat more favorably No difference Less favorably, do not know % of California Adults 48 27 23 2 % of Likely Voters 48 29 22 1 SOURCE: PPIC Statewide Survey (November 2002). NOTE: N = 2,000. 60 Perhaps some political consultants would prefer that their candidates not sign such pledges, since it restricts their options in running a campaign. Moreover, it is difficult to predict whether these codes of conduct would actually be followed by the candidates who adopted them. From the voter’s perspective, it seems like a worthy, low-risk, and inexpensive experiment. Candidates do not like to hear that they have violated promises, so the act of getting them to sign pledges might on the margin provide incentives for better conduct. For the voters, it also offers the welcome prospect of a campaign that is geared toward the issues that matter most to them and a welcome respite from the business of attacks and counterattacks. The California public is dissatisfied with highly negative races and the limited access they have to their candidates. Their frustration was evident in the course of the 2002 elections, according to our surveys. It peaked in 2003 when California voters took matters into their own hands and removed an incumbent governor who, among many other things, epitomized many of the undesirable aspects of modern campaigning. Recall elections are now part of the statewide political landscape—not just the local election context as in the past. Whatever the merits and demerits of recall elections, they do not address the underlying sense, especially felt by better informed and more involved voters, that political dialogue is conducted at a demeaning and negative level. Indeed, we could find no evidence in our surveys since the November 2002 election that attitudes toward campaigns and trust in government changed in a positive way after the recall of Governor Gray Davis and his replacement by Arnold Schwarzenegger in 2003. The problem is that political consultants are paid to get their clients elected, not to make voters feel happier about the process. Political campaigns today are not designed to increase voter turnout, or to improve the public’s trust in government, or to shed a positive light on elected officials or their challengers. Many of the members of the political establishment seem to have concluded that a candidate who takes the high road against a candidate who takes the low one might lose the war of attrition. Moreover, his or her consultant will not be rewarded for sacrificing the candidate to the cause of higher ethics or 61 preserving civic sensibilities. With increasing campaign expenses to run for office, politicians and political consultants are under pressure to win elections at all costs. However, the survey evidence discussed in this report shows that negative tactics hurt the candidate who initiates them as well as the candidate who is the object. Although the Davis campaign outspent the Simon campaign and won the election while using negative attack ads—following a consistent pattern that has existed in gubernatorial elections since 1986—we noted several risks to this political strategy. In the short run, the candidate can alienate his supporters and turn them off to voting for anyone in the election. This could spell defeat in a close election in which the candidate needs every vote from supporters (e.g., Davis won by only 5 percentage points). In the long run, this can also leave the winner vulnerable in a recall election such as the one faced by Davis in October 2003. Some Democratic voters were disinterested in supporting a candidate that they grew to dislike during the election. More important, the elected official may find that voters have less trust in government and less respect for the elected officeholder because of the way he or she behaved in the course of the campaign. This raises problems when an elected official needs support for policies and ballot measures. As we have seen in our survey, once confidence is lost it is difficult to regain. We hope that these survey findings will serve to dampen the appetite for negativity as candidates think about their winning strategy in statewide elections in 2006 and beyond. The voters have also shown the way to a new type of political campaigning in their survey responses to a wide variety of different ideas for political and election reforms. If there is resistance to enacting such changes in the state’s political arena, then we can anticipate that the California public will incorporate its expectations about good campaigns into its evaluations in the voting booth through the citizens’ initiative process. 62 Appendix A Dissatisfaction with the Candidate Choices for Governor Independent Variable Coefficient S.E. Level of attention to the news (four-point scale) 0.47*** 0.06 Level of interest in politics (four-point scale) 0.26*** 0.07 Saw an ad for a gubernatorial candidate 0.03 0.11 Strong Democrat (dummy variable) –0.50*** 0.15 Strong Republican (dummy variable) –0.28 0.15 Independent (dummy variable) 0.49*** 0.15 Female (dummy variable) –0.05 0.10 White (dummy variable) 1.64*** 0.26 Latino (dummy variable) 0.49** 0.17 Constant –3.27*** 0.32 N Log likelihood Pseudo R2 1901 2361.68 0.18 SOURCE: PPIC Statewide Survey (November 2002). NOTES: Figures are logit regression coefficients. Dependent variable is whether a respondent is dissatisfied with the choice of candidates for governor. N = 2,000. **p < .01. ***p < .001. 63 Appendix B Trust in State Government, All Adults Coefficient S.E. Political assessments Favorable ranking of incumbent governor Gray Davis (dummy variable) 0.594** 0.184 Perception that election campaigns are getting worse (dummy variable) –0.462* 0.184 Individual characteristics Strong Democrat (dummy variable) –0.289 0.251 Strong Republican (dummy variable) –0.688* 0.297 Independent (dummy variable) –0.748** 0.271 Conservative (dummy variable) –0.004 0.210 Liberal (dummy variable) –0.059 0.210 Female (dummy variable) –0.013 0.168 White (dummy variable) –0.033 0.238 Latino (dummy variable) 0.577* 0.262 Youth, ages 18–34 (dummy variable) –0.013 0.185 Education (three-point scale) –0.033 0.125 Income (three-point scale) 0.050 0.118 Political engagement Political activities (eight-point scale) –0.013 0.063 Interest in politics (four-point scale) 0.256* 0.129 Attention paid to the news (four-point scale) –0.197 0.110 Political knowledge (four-point scale) –0.039 0.090 Exposure to advertisements (dummy variable) 0.135 0.179 Constant Na Pseudo R2 –0.907 754 0.14 0.462 Percentage correctly predicted 0.69 SOURCE: PPIC Statewide Survey (November 2002). NOTES: Figures are logit regression coefficients. Dependent variable is whether respondent trusts the state. N = 2,000. aN is less than 2,000 because some questions were asked of a randomly split sample of half of all respondents. *p < .05. **p < .01. 65 Appendix C Trust in State Government, Likely Voters Coefficient S.E. Political assessments Favorable ranking of incumbent governor Gray Davis (dummy variable) 0.976** 0.290 Perception that election campaigns are getting worse (dummy variable) –0.515* 0.261 Individual characteristics Strong Democrat (dummy variable) –0.329 0.341 Strong Republican (dummy variable) –0.743 0.381 Independent (dummy variable) –0.875* 0.390 Conservative (dummy variable) –0.204 0.328 Liberal (dummy variable) –0.077 0.307 Female (dummy variable) –0.141 0.250 White (dummy variable) –0.015 0.352 Latino (dummy variable) 0.366 0.445 Youth, ages 18–34 (dummy variable) 0.206 0.312 Education (three-point scale) –0.149 0.187 Income (three-point scale) 0.271 0.169 Political engagement Political activities (eight-point scale) 0.013 0.081 Interest in politics (four-point scale) –0.289 0.230 Attention paid to the news (four-point scale) –0.110 0.185 Political knowledge (four-point scale) 0.056 0.132 Exposure to advertisements (dummy variable) 0.368 0.278 Constant Na Pseudo R2 –0.451 404 0.17 0.806 Percentage correctly predicted 0.75 SOURCE: PPIC Statewide Survey (November 2002). NOTE: Figures are logit regression coefficients. Dependent variable is whether respondent trusts the state. N = 2,000. aNo. is less than 2,000 because some questions were asked of a randomly split sample of half of all respondents. *p < .05. **p < .01. 67 Bibliography Ansolabehere, Stephen, and Shanto Iyengar, Going Negative: How Political Advertisements Shrink and Polarize the Electorate, Free Press, New York, 1995. Baldassare, Mark, California in the New Millennium: The Changing Social and Political Landscape, University of California Press, Berkeley, California, 2000. Baldassare, Mark, A California State of Mind: The Conflicted Voter in a Changing World, University of California Press, Berkeley, California, 2002. Baldassare, Mark, Bruce Cain, and Jonathan Cohen, “Voters Nostalgic for a Good Old Fashioned Campaign,” California Journal, January, 2003a. Baldassare, Mark, Bruce Cain, and Jonathan Cohen, “California Voters Take Charge,” California Journal, November, 2003b. Bartels, Larry M., “Campaign Quality: Standards for Evaluation, Benchmarks for Reform,” in Larry M. Bartels and Lynn Vavreck, eds., Campaign Reform: Insights and Evidence, University of Michigan, Ann Arbor, Michigan, 2000, pp. 1–61. Bartels, Larry M., et al., “Campaign Reform: Insights and Evidence,” in Larry M. Bartels and Lynn Vavreck, eds., Campaign Reform: Insights and Evidence, University of Michigan, Ann Arbor, Michigan, 2000, pp. 201–247. Berelson, Bernard R., Paul F. Lazarsfeld, and William N. McPhee, Voting: A Study of Opinion Formation in a Presidential Campaign, University of Chicago Press, Chicago, Illinois, 1954. 69 Bowler, Shaun, and Todd Donovan, “Why Do People Like Their State and Local Government More Than the Federal Government?” paper presented at the annual meeting of the Midwest Political Science Association, Chicago, Illinois, April 25–27, 2002. Buchanan, Bruce, Renewing Presidential Politics: Campaigns, Media, and the Public Interest, Rowman and Littlefield, Lanham, Maryland, 1996. Buchanan, Bruce, “Regime Support and Campaign Reform,” in Larry M. Bartels and Lynn Vavreck, eds., Campaign Reform: Insights and Evidence, University of Michigan, Ann Arbor, Michigan, 2000, pp. 173–200. California Attorney General, Initiative Measures, 2000–2002, State of California, Sacramento, California, 2002, available at http://www. caag.ca.us/initiatives/index.htm. California Department of Finance, Historical California Population Estimates, 1941–2003, State of California, Sacramento, California, 2003. California Secretary of State, Total Campaign Expenditures by Candidates for the California State Legislature, 1976–1998, State of California, Sacramento, California, 1998. California Secretary of State, Statement of the Vote, State of California, Sacramento, California, November 2000. California Secretary of State, Statement of the Vote, State of California, Sacramento, California, November 2002. California Secretary of State, Statement of the Vote, State of California, Sacramento, California, October 2003. California Secretary of State, Statement of the Vote, State of California, Sacramento, California, March 2004. Citrin, Jack, “Comment: The Political Relevance of Trust in Government,” American Political Science Review, Vol. 68, 1974, pp. 973–988. 70 Citrin, Jack, and Donald Philip Green, “Presidential Leadership and the Resurgence of Trust in Government,” British Journal of Political Science, Vol. 16, 1986, pp. 431–453. Clark, Harold D., and Alan C. Acock, “National Elections and Political Attitudes: The Case of Political Efficacy,” British Journal of Political Science, Vol. 19, 1989, pp. 551–562. Cohen, Jonathan, Mark Baldassare, and Eliana Kaimowitz-Rodriguez, “Trust in Government and Political Participation Among California Latinos: 1998–2002,” paper presented at annual meeting of the American Association for Public Opinion Research, Nashville, Tennessee, May 14–18, 2003. Dahl, Robert A., A Preface to Democratic Theory, University of Chicago Press, Chicago, Illinois, 1956. Dionne, E. J., Why Americans Hate Politics, Simon and Schuster, New York, 1991. Field Poll, Survey Report, October 1996. Field Poll, Political Parties and the Political System, April 1999. Field Poll, Record Low Voter Turnout Expected, November 2002. Finkel, Steven E., “Reciprocal Effects of Participation and Political Efficacy: A Panel Analysis,” American Journal of Political Science, Vol. 29, No. 4, 1985, pp. 891–913. Ginsberg, Benjamin, and R. Weissberg, “Elections and the Mobilization of Popular Support,” American Journal of Political Science, Vol. 22, 1978, pp. 31–55. Hetherington, Marc J., “The Political Relevance of Political Trust,” American Political Science Review, December 1998. Lipset, Seymour Martin, and William Schneider, The Confidence Gap, Free Press, New York, 1983. Los Angeles Times Poll, November election exit poll, 2002. 71 Miller, Arthur H., “Political Issues and Trust in Government: 1964–1970,” American Political Science Review, Vol. 68, 1974, pp. 951–972. Miller, Warren E., and J. Merrill Shanks, The New American Voter, Harvard University Press, Cambridge, Massachusetts, 1996. Moore, David W., “The Myth and Mythology of Trust in Government,” Public Perspective, January/February 2002, pp. 7–11. The Pew Research Center for the People and the Press, Deconstructing Distrust: How Americans View Government, Washington, D.C., March 10, 1998. PPIC Statewide Survey, The Changing Political Landscape of California Series, Public Policy Institute of California, San Francisco, California, April 1998. PPIC Statewide Survey, The Changing Political Landscape of California Series, Public Policy Institute of California, San Francisco, California, May 1998. PPIC Statewide Survey, The Changing Political Landscape of California Series, Public Policy Institute of California, San Francisco, California, September 1998. PPIC Statewide Survey, The Changing Political Landscape of California Series, Public Policy Institute of California, San Francisco, California, October 1998. PPIC Statewide Survey, Californians and Their Government Series, Public Policy Institute of California, San Francisco, California, December 1999. PPIC Statewide Survey, Californians and Their Government Series, Public Policy Institute of California, San Francisco, California, January 2000. PPIC Statewide Survey, Californians and Their Government Series, Public Policy Institute of California, San Francisco, California, October 2000. 72 PPIC Statewide Survey, Californians and Their Government Series, Public Policy Institute of California, San Francisco, California, January 2001. PPIC Statewide Survey, Californians and Their Government Series, Public Policy Institute of California, San Francisco, California, January 2002. PPIC Statewide Survey, Californians and Their Government Series, Public Policy Institute of California, San Francisco, California, August 2002. PPIC Statewide Survey, Californians and Their Government Series, Public Policy Institute of California, San Francisco, California, October 2002. PPIC Statewide Survey, Special Survey on Political Campaigns, Public Policy Institute of California, San Francisco, California, November 2002. PPIC Statewide Survey, Californians and Their Government Series, Public Policy Institute of California, San Francisco, California, February 2003. PPIC Statewide Survey, Special Survey on the California State Budget, Public Policy Institute of California, San Francisco, California, June 2003. PPIC Statewide Survey, Special Survey on Political Campaigns, Public Policy Institute of California, San Francisco, California, October 2003. PPIC Statewide Survey, Special Survey on the California State Budget, Public Policy Institute of California, San Francisco, California, January 2004. PPIC Statewide Survey, Californians and Their Government Series, Public Policy Institute of California, San Francisco, California, February 2004. 73 PPIC Statewide Survey, Special Survey on the California State Budget, Public Policy Institute of California, San Francisco, California, May 2004. PPIC Statewide Survey, Californians and Their Government Series, Public Policy Institute of California, San Francisco, California, September 2004. Public Policy Institute of California, “California’s Recall Election,” Just the Facts, San Francisco, California, November 2003. Rahn, Wendy M., John Brehm, and Neil Carlson, “National Elections as Institutions for Building Social Capital,” in Morris Fiorina and Theda Skocpol, eds., Civic Engagement in American Democracy, Brookings Institution, Washington, D.C., and Russell Sage Foundation, New York, 1999, pp. 111–160. Rosenstone, Steven J., and John Mark Hansen, Mobilization, Participation, and Democracy in America, Macmillan, New York, 1993. Schrag, Peter, Paradise Lost: California’s Experience, America’s Future, The New Press, New York, 1998. Silva, Fred, The California Initiative Process: Background Material, Public Policy Institute of California, San Francisco, California, 2000. Simon, Mark, “Fiscal Crisis and Voter Resentment Led to Recall,” San Francisco Chronicle, July 25, 2003. Teixeira, Ruy A., The Disappearing American Voter, Brookings Institution, Washington, D.C., 1992. University of California Hastings College of Law, California Ballot Initiatives Database, University of California, San Francisco, California, 2002, available at http://www.holmes.hastings.edu. U.S. Federal Election Commission Campaign Finance Reports and Data (electronic resource), Washington, D.C., 2004. 74 Williams, John T., “Systemic Influences on Political Trust: The Importance of Perceived Institutional Performance,” Political Methodology, Vol. 11, 1985, pp. 125–142. Wolfinger, Raymond E., and Steven J. Rosenstone, Who Votes? Yale University Press, New Haven, Connecticut, 1980. 75 About the Authors MARK BALDASSARE Mark Baldassare is Director of Research at PPIC. He also holds the Arjay and Francis Fearing Miller Chair in Public Policy and directs the PPIC Statewide Survey—a large-scale public opinion project designed to develop an in-depth profile of the social, economic, and political forces at work in California elections and in shaping the state’s public policies. He is the author of nine books, including A California State of Mind: The Conflicted Voter in a Changing World. Before joining PPIC, he was a professor of urban and regional planning at the University of California, Irvine, where he held the Johnson Chair in Civic Governance and initiated and directed the Orange County Annual Survey. He has conducted surveys for the Los Angeles Times, the San Francisco Chronicle, and the California Business Roundtable. He holds a Ph.D. in sociology from the University of California, Berkeley. BRUCE E. CAIN Bruce E. Cain is Director of the Institute of Governmental Studies and Robson Professor of Political Science at the University of California, Berkeley. A Rhodes Scholar at Trinity College, Oxford, he received his Ph.D. in political science from Harvard University. He has been a redistricting consultant to, among others, the Los Angeles City Council, the U.S. Justice Department, the San Diego Citizens’ Commission on Redistricting and the City and County of San Francisco. His recent publications include Voting at the Political Fault Line: California’s Experiment with the Blanket Primary, co-edited with Elisabeth R. Gerber, and Democracy Transformed? Expanding Political Opportunities in Advanced Industrial Democracies, co-edited with Russell J. Dalton and Susan Scarrow. Professor Cain is a political commentator for numerous radio and television stations in the Bay Area. D. E. APOLLONIO Dorie Apollonio is a Legacy Postdoctoral Fellow at the University of California, San Francisco. She received her Ph.D. in political science from the University of California, Berkeley, and her M.P.P. from the 77 John F. Kennedy School of Government at Harvard University. Her research focuses on campaign finance, political reform, and interest group behavior, specifically the political activities of the tobacco industry and their effect on health policy. JONATHAN COHEN Jon Cohen is Associate Director of the PPIC Statewide Survey. His research interests include survey methodology, state tax policy, and electoral systems. He is the co-author of the second edition of the California Latino/Latina Demographic Databook. He has also written a recent book chapter and magazine articles on electoral politics in California. He holds a B.A. in history from the Johns Hopkins University, a graduate diploma from the Nitze School of Advanced International Studies, and an M.A. in political science from the University of California, Berkeley. 78 Related PPIC Publications A California State of Mind: The Conflicted Voter in a Changing World Mark Baldassare California in the New Millennium: The Changing Social and Political Landscape Mark Baldassare Adapting to Term Limits: Recent Experiences and New Directions Bruce Cain and Thad Kousser PPIC publications may be ordered by phone or from our website (800) 232-5343 [mainland U.S.] (415) 291-4400 [Canada, Hawaii, overseas] www.ppic.org 79" ["post_date_gmt"]=> string(19) "2017-05-20 09:36:36" ["comment_status"]=> string(4) "open" ["ping_status"]=> string(6) "closed" ["post_password"]=> string(0) "" ["post_name"]=> string(9) "r_1004mbr" ["to_ping"]=> string(0) "" ["pinged"]=> string(0) "" ["post_modified"]=> string(19) "2017-05-20 02:36:36" ["post_modified_gmt"]=> string(19) "2017-05-20 09:36:36" ["post_content_filtered"]=> string(0) "" ["guid"]=> string(51) "http://148.62.4.17/wp-content/uploads/R_1004MBR.pdf" ["menu_order"]=> int(0) ["post_mime_type"]=> string(15) "application/pdf" ["comment_count"]=> string(1) "0" ["filter"]=> string(3) "raw" ["status"]=> string(7) "inherit" ["attachment_authors"]=> bool(false) }