Real Choices New Voices

Real Choices/New Voices

This page intentionally left blank

Real Choices/New Voices How Proportional Representation Elections Could Revitalize American Democracy Second Edition

Douglas J. Amy

c o l u m b i a

u n i v e r s i t y

p r e s s

n e w

y o r k

columbia university press Publishers Since 1893 New York Chichester, West Sussex Copyright © 2002 Columbia University Press All rights Reserved Library of Congress Cataloging-in-Publication Data Amy, Douglas J. Real choices/new voices : how proportional representation elections could revitalize American democracy / Douglas J. Amy.—2nd ed. p. cm. Includes bibliographical references and index. ISBN 0–231-12548-8 (cl. : alk. paper) ISBN 0–231-12549-6 (pbk. : alk. paper) 1. Proportional representation—United States. 2. Elections—United States. 3. United States—Politics and government—20th century. I. Title JF1075.U6 A49 2002 328.73⬘07347—dc21

2002067425

I Columbia University Press books are printed on permanent and durable acid-free paper. Printed in the United States of America c 10 9 8 7 6 5 4 3 2 1 p 10 9 8 7 6 5 4 3 2 1 All the citations to information derived from the World Wide Web were accurate at the time of writing. Neither the author nor Columbia University Press is responsible for Web sites that have changed or expired since the time of publication.

To my son, Sam, who brings me much joy and keeps me honest.

Now if you think that proportional representation is boring, you are a very silly person because it’s about how we can run the country better. John Cleese Monty Python’s Flying Circus

This page intentionally left blank

Contents

List of Tables and Illustrations Preface to Second Edition Introduction

ix

xi

1

1. Fair Representation for All

25

2. Solving the Redistricting Problem 3. Improving Election Campaigns 4. Toward a Multiparty System 5. Electing More Women

52 68

88

109

6. Fair Representation for Racial and Ethnic Minorities 7. Encouraging Voter Turnout

151

8. PR: A Different and Better Kind of Democracy 9. Objections to Proportional Representation

186

167

125

viii

Contents

10. Versions of Electoral Reform for the United States

215

11. The Political Prospects for PR in the United States

235

Appendixes A. Seat Allocation Formulas in Proportional Representation

259

B. Vote Counting and Ballot Transfer in the Single Transferable Vote 263 C. The Forgotten History of Proportional Representation in the United States 267 Notes

275

Selected Bibliography

299

Additional Resources on the World Wide Web Index

307

305

List of Tables and Illustrations

Tables I.1 Voting Results and Seats Allocations in Mixed-Member Voting 22 1.1 Vote/Seat Disproportionality in National Legislatures in 15 Countries 34 1.2 Vote/Seat Outcomes for Parties in Different Electoral Systems 35 1.3 Comparison of Republican Votes and Republican Seats in U.S. House Elections, 1996–2000 39 1.4 The Capriciousness of Seat Allocation under Plurality Rules 49 5.1 The Representation of Women in 13 Western Democracies 115 6.1 Percentage of Black Elected Officials by State (1997) 127 7.1 Average Voter Turnout in the 1990s 152 8.1 Competing Visions of Democratic Politics: Majoritarian vs. Proportional 182 10.1 Proportional Results in a Cumulative Vote Election 222 10.2 Disproportional Results in a Cumulative Vote Election 223 10.3 Minority Rule in a Cumulative Vote Election 224 10.4 Relationship of Number of Legislative Parties to District Magnitude in PR 232 A.1 Seat Allocation Using Largest Remainder Method 260

x

List of Tables and Illustrations

A.2 Seat Allocation Using d’Hondt Formula 261 A.3 Seat Allocation Using Sainte-Lagüe Formula 262 B.1 Counting of a Single Transferable Vote Election 265

Figures I.1 I.2 I.3 2.1 2.2 4.1 5.1

50 Single-Seat Districts 17 Ten 5-Seat Districts 17 Five 10-Seat Districts 17 Democratic Gerrymander 55 Republican Gerrymander 55 Norwegian versus American Party Systems 91 Percentage of Women’s Representation Under Different Electoral Systems, 1950–1998 116 6.1 Texas Majority-Minority Districts Declared Unconstitutional 135 6.2 Texas Majority White Districts Left Unchallenged 136 B.1 Diagram of STV Vote Counting Procedures 264

Ballots 1. 2. 3. 4. 5. 6. 7.

Closed Party List System 18 Open Party List System 19 Mixed-Member Proportional Representation System Single Transferable Vote 23 Instant Runoff Voting 216 Cumulative Voting 220 Limited Voting 221

21

Preface to the Second Edition

Today, the basic case for proportional representation (PR) remains much the same as it was when the first edition of this book was published in 1993: PR is simply a much fairer, more representative, and more democratic electoral system than our current winner-take-all approach. So it might be asked: Why is a new edition of this book needed? The answer is that much has changed since the original edition of this book was published in the early 1990s. First, the political world has continued to change during that time. Many elections have come and gone and the original book was sorely in need of less dated examples and more current information, such as that concerning the levels of women’s representation in the U.S. and around the world. But there have been more substantial political changes as well. For example, the whole area of voting rights has been altered dramatically since I wrote the first edition. In several rulings, the Supreme Court has now called into question the use of majority-minority districts. This has significantly strengthened the case for alternatives like proportional representation, and this fundamental change is reflected in the extensively revised chapter on the representation of racial and ethnic minorities. Scholarship in the area of electoral systems has also progressed in the last ten years. A number of new studies have produced more data and more arguments that reinforce the case for proportional representation elections. One comparative study, for instance, found that citizens in countries with proportional representation and power-sharing multiparty legislatures tended to be substantially more satisfied with the operation of their political

xii

Preface to Second Edition

system.1 This finding has important implications for a country such as the United States that suffers from high levels of public alienation from politics and government. In another significant study, G. Bingham Powell found that there is a closer congruence between the policy views of citizens and policymakers in PR systems than in winner-take-all systems.2 Such new advances in understanding electoral systems and their political consequences find their way into every chapter of this new edition. The other change that prompted this revised edition was one involving my own understanding of the advantages of proportional representation. In the original edition, my major concern was the improved representation produced by PR. The main arguments for PR were that it would produce much fairer and more accurate representation for parties, women, racial minorities, and so on. I still believe that these are powerful arguments in the case for PR. But I have become convinced that the case for electoral reform is actually much deeper than that—that adopting PR would also have profound implications for the basic workings of democracy in the United States. These implications are explored in a new chapter (chapter 8) entitled “PR: A Different and Better Kind of Democracy,” and this new theme is reflected in the new subtitle of the book as well. The basic argument is that proportional representation is a superior election system not only because it produces better representation, but also because it produces more democracy. We cannot have democracy for all in the United States as long as we have representation for only some. Moreover, proportional representation encourages many of the other things that we value in a democracy, including political equality, power sharing, citizen participation, vigorous political dialogue, and political bargaining and negotiation. For these reasons, I see proportional representation as an important reform in the effort to revitalize democracy in the United States—the effort to empower more citizens and to make our government more responsive to the public. Finally, a careful reader may notice many other, more subtle changes in this edition. I had not initially intended to do so, but I ended up making revisions on virtually every page of the book. The original prose sometimes left much to be desired and I spent considerable time trying to hone my arguments and express them more clearly and persuasively. On the other hand, it may be the case that I am just another obsessive academic who can’t leave well enough alone.

Preface to Second Edition

xiii

Acknowledgments This book is a work of synthesis and relies heavily on the arguments and evidence developed by the many talented scholars working the field of electoral systems. I am indebted to them for doing the important scholarly work that I had neither the time nor the skills to do. And while I will not repeat all of the acknowledgements from the first version of this book, I would like to thank a number of people who over the years have been particularly supportive of my work and who have helped to shape my ideas about electoral system reform. Among those in academia are Kathleen Barber, Lani Guinier, Andrew Reynolds, and Joseph Zimmerman. I am also grateful to the numerous electoral reform activists who have discussed their ideas about proportional representation with me. These include Rob Richie, Ed Still, Steven Hill, John Anderson, John Gear, Howie Fain, Terry Bouricius, and Steve Chessin. I am particularly thankful for my encounters with Leon Weaver and Enid Lakeman, two scholars and champions of electoral reform who are no longer with us. Early on, they encouraged me to pursue my interest in proportional representation for the United States at a time when virtually no one else seemed interested in that issue. Thanks also go to my editor at Columbia University Press, Peter Dimock, who made some insightful suggestions that led to the inclusion of a new chapter in this book. Leslie Bialler copyedited the book with a rare combination of meticulousness and good humor. And Celeste Newbrough has done her usual excellent job of producing a truly useful and usable index for this volume. I also want to thank Ed Royce, a good friend who has also served as a sounding board for the ideas in this book and given me much good advice about writing. Thanks also to Tom, Joan, Kavita, Jon, and Sarah for vital last minute advice on the title. I am also grateful to Mark Monmonier for sharing with me his excellent maps of Texas congressional districts. Finally, of course, there is my wife, Susan, and my son, Sam. Both have shown remarkable patience and sympathy as I have pursued my obsession with proportional representation over the years. Because of them, I have been able to (mostly) retain my sense of perspective and to remember that there are other important things in life besides writing books about electoral reform. I am grateful to them in more ways than they will ever know. Douglas J. Amy, July 2002

This page intentionally left blank

Real Choices/New Voices

This page intentionally left blank

Introduction

The American electoral system is outmoded, unfair, and undemocratic. Worse yet, a large number of Americans remain oblivious to these disturbing problems. To be sure, the American public frequently expresses a great deal of anger and frustration about some aspects of our elections, such as poor-quality candidates, the constant reelection of incumbents, and the role of special-interest money in campaigns. But in fact, many of the problems with American elections go much deeper than that, and are located in the mechanics of the electoral system itself. Electoral systems are the methods we use to elect officials—the basic procedural rules by which votes are cast and counted, and the winners determined. These rules form the hidden infrastructure of our election process, and many Americans hardly give them a second thought. They would be surprised to learn that our electoral system is so deeply flawed that it often violates the fundamental principles of democracy and fair representation that many believe are the hallmarks of the American political system. Fortunately, some Americans are beginning to realize that electoral systems—sometimes also called voting systems—can have important political ramifications. One wake-up call came in the form of the 2000 presidential election. The result of that contest made it painfully clear that at least one part of the American electoral system—the Electoral College—could malfunction in disturbing ways. The arcane mechanics of this system resulted in the election of George W. Bush, even though his opponent, Al Gore, actually won the popular vote. Suddenly, many Americans realized that who

2

Introduction

wins an election is determined not only by the votes that are cast, but also by the workings of the electoral system itself. Not surprisingly, this seemingly undemocratic election result produced a call to change the electoral system that we use to choose our president. Unfortunately, however, most Americans have yet to turn the same skeptical eye toward the predominant electoral system in the United States—singlemember plurality elections (SMP). This voting system is commonly used to elect officials to our local, state, and federal legislatures. In SMP, officials are elected one at a time in single-member districts, with the winner being the candidate with the most votes—the plurality. This is the winner-take-all electoral system that Americans have grown up with and that most accept as the natural form of democratic elections. But as this book will demonstrate, this system is riddled with grave political problems. In fact, the problems that afflict SMP elections pose a much more serious and widespread threat to the democratic process than any flaws in the Electoral College. The Electoral College affects elections to only one office—albeit an important one— while the flaws in our legislative electoral systems affect the election of thousands of officials in hundreds of policymaking bodies around the country. Ultimately, these electoral problems make the American political system much less democratic than it could be.

The Trend Away from Plurality Elections Americans sometimes assume that our single-member plurality voting system serves as the model for the rest of the free world. But nothing could be further from the truth. In reality, SMP systems are increasingly on the wane worldwide. Among advanced industrial democracies, only Great Britain and Canada join us in clinging to this eighteenth-century method of election. Most other Western democracies have long since adopted another form of elections: proportional representation (PR). In this system, officials are elected in large, multimember districts according to the proportion of the vote their party receives. If, for example, a given district had ten legislative seats and a party received 30 percent of the votes, it would receive three of those ten seats. Countries using various forms of proportional representation include Austria, Belgium, Cyprus, Denmark, Finland, Germany, Greece, Ireland, Israel, Luxembourg, Malta, the Netherlands, New Zealand, Norway, Portugal, Spain, Sweden, and Switzerland. And it is not

Introduction

3

the case that these countries opted for proportional representation without knowledge of the American system. Virtually all these countries at one time had forms of plurality election systems but then explicitly rejected them in favor of proportional representation. Thus for most of the last 100 years, the worldwide trend clearly has been away from the U.S. form of elections. Most recently, emerging democracies around the world have followed the trend toward proportional representation. In Eastern Europe and the former Soviet Union, many new democracies rushed to embrace American political ideals and economic practices. But the one aspect of our political system that most deliberately rejected was our SMP election system. Almost three-quarters of them opted for various forms of proportional or semiproportional systems. As one Czech legislator pointed out, “We decided virtually from the beginning not to take very seriously the [single-member plurality] system.”1 PR has also been the system of choice for many new democracies in Africa. In South Africa, for instance, the African National Congress agreed to use proportional representation elections so that the white minority would be ensured of some representation in the new parliament. It seems that when countries can start from scratch and consider all the electoral alternatives, they usually opt for proportional rather than plurality elections. Thus, although we in the United States like to think of ourselves as in the vanguard of democratic political institutions, we have in reality fallen behind in the area of electoral reform, and our voting system has become increasingly outmoded. Proportional representation is now the state of the art in elections. America was once known as the laboratory of democracy, but in the area of elections we are now in danger of becoming the museum of democracy.

Growing Discontent in Countries Using Plurality Elections Even in those few long-standing democracies that still use plurality elections, discontent with this system is growing and all of them have political movements pushing for a change to proportional representation. One of the most interesting examples of this trend is New Zealand. For more than a hundred years, New Zealand used single-member district plurality elections for its parliament. But in the 1980s, after a series of election fiascoes in which the party that came in second place in the polls actually won the majority of parliamentary seats, a national commission was formed to study the

4

Introduction

electoral system. The commission recommended that a public referendum be held on the electoral system, giving voters a choice between sticking with the plurality system and changing to a proportional system.2 After a long political battle to allow this referendum to take place, it finally occurred in 1993. Voters approved changing to a proportional system by 54 percent to 46 percent. It is perhaps more significant that agitation for electoral reform is also taking place in Great Britain, the traditional home of single-member plurality elections. For years, the Electoral Reform Society and its political allies have been calling for the abandonment of plurality elections. The movement gained some important momentum when the Labour Party, led by Prime Minister Tony Blair, was first elected in the 1990s. Blair appointed a commission to study the electoral system and promised a public referendum on the issue. The Jenkins commission issued its report in 1998, which called for the adoption of a new, semiproportional electoral system.3 At this writing, because of disagreements within the Labour party, Blair has refused to make a clear commitment to holding the promised referendum. Public support for PR in Great Britain has varied over time, but it is clearly substantial. At one of its highpoints, a poll in the Economist in May of 1997 showed a four to one majority in favor of switching to proportional representation. More importantly, proportional representation has been gaining ground in other elections held in Great Britain. Both Scotland and Wales now have their own parliaments, and both decided to reject winner-take-all elections and use forms of proportional representation to elect officials to those bodies. And in elections for representatives to the European community, Britain has now switched to proportional representation elections. These developments have been important victories for the PR movement in Great Britain. The pro-PR movements in the United States and Canada are not as advanced as that of Great Britain. In Canada, SMP elections are infamous for producing distorted results, with some parties winning substantial parts of the vote, but few seats in parliament. This has begun to provoke some discussion about electoral reform. After the 1997 elections, for example, several political commentators soundly criticized their electoral system, and the editors of the national newspaper, The Globe and Mail, came out in favor of a change to proportional representation elections.4 A number of academic books and articles have been published calling for serious consideration of both proportional and semiproportional electoral systems.5 The Reform party has called for a national referendum on the electoral system, and the

Introduction

5

New Democratic party established a study committee to look at alternatives to single-member plurality elections. Several organizations have also emerged for the explicit purpose of push for PR, the most prominent being Fair Vote Canada. In the United States, interest in proportional representation has waxed and waned several times during the last 100 years. The first wave of interest in this alternative occurred in the early part of the twentieth century, when PR was adopted and used successfully by two-dozen cities, including New York, Cincinnati, and Cleveland. (For an account of this largely forgotten history of PR in the United States, see Appendix C.) Today we are seeing a second wave of interest in PR. There has been a growing discussion of this alternative, especially among academics and political commentators. Articles on this topic have appeared in wide number of prominent national media, including Time, New Republic, New York Times, Boston Globe, Atlantic, New Yorker, Christian Science Monitor, Chicago Tribune, The Nation, Washington Post, San Francisco Chronicle, and USA Today. A number of leaders in the civil rights and voting rights community, including Jesse Jackson and Lani Guinier, have been arguing that PR could solve the problem of ensuring fair representation for racial and ethnic minorities. Several third parties, including the Greens, have endorsed PR. On the national level, the Center for Voting and Democracy in Washington D.C. has emerged as the leading nonprofit organization educating politicians and the public about electoral alternatives, including proportional representation. In part because of their efforts, legislation has been introduced during several of the recent sessions of Congress to allow states to use PR for the election of representatives in the House.6 Recognizing the significance of electoral systems, the League of Women Voters decided in 2000 to have their local chapters study the issue of electoral reform. More significantly, on the local level, more than 100 cities and counties have taken a step away from winner-take-all elections and are now experimenting with semiproportional systems for their elections. Also, educational and activist activity in favor of PR has been taking place in many states, including California, Colorado, Georgia, Illinois, Massachusetts, Michigan, Minnesota, Montana, New York, North Carolina, Oregon, Rhode Island, Texas, and Washington. Several state and local groups supporting PR have sprung up around the country, including Washington Citizens for Proportional Representation, Californians for Proportional Representation, Fairvote Wisconsin, Fairvote Minnesota, the Midwest Democracy Center and Illinois

6

Introduction

Citizens for Proportional Representation. This kind of grassroots political activity is an additional sign that public interest in this reform is on the rise.

The Problems of Plurality Elections But why is there all this discontent with the single-member plurality voting system? What exactly is wrong with this system and what are the advantages of proportional representation that have made it the most popular voting system among advanced democracies? These, of course, are the main questions that will be answered in this book. But for now, let us at least get a brief sense of some of the drawbacks of our current voting system. Consider the following list of problems and complaints often associated with American elections:

Lack of Competitive Elections It is not unusual for 90 percent of the incumbents to be reelected to Congress. Many areas of the country are so dominated by one party that very little real competition exists. For example, in elections around the country for state legislatures in 2000, either the Democrats or the Republicans failed to nominate a candidate in 41 percent of those contests.

Lesser-of-Two-Evils Often voters have trouble finding candidates they are truly excited about. So they end up reluctantly supporting the least objectionable one. Winning candidates always assume that anyone who voted for them is an enthusiastic supporter, but in reality much of that support may be lukewarm at best.

Low Voter Turnout Turnout in U.S. elections continues to be abysmally low—we consistently rank far below other Western democracies. While others rou-

Introduction

7

tinely enjoy turnout rates of 80–90 percent, we get hardly more than 50 percent in a presidential election year and often less than 40 percent for Congressional elections. More than ninety million voters regularly sit out American elections. Such low figures cast serious doubt on the democratic mandate that most elected officials like to claim.

Tweedledum and Tweedledee Candidates Telling the difference between the candidates offered by the major parties is sometimes difficult. With both Republican and Democratic candidates appealing to the swing voters in the center, they often tend to talk about the same issues in much the same way. Given this situation, it is not surprising that many voters feel they have little real choice at the polls.

The Two-Party Monopoly Even when discernible differences between Democrats and Republicans exist, those two choices hardly exhaust the political alternatives. While most other democracies offer voters a wide variety of parties and ideological alternatives, our two-party system severely constrains our choices. Increasing numbers of Americans say that they would like to see alternative parties emerge to challenge the Republicans and Democrats.

Wasted Votes For voters living in districts dominated by the opposing party, voting is largely futile. For instance, Republican voters in a district in which Democrats form a large majority are merely wasting their votes, for their candidates usually have no real chance of winning. Every election year tens of millions of Americans find themselves in this frustrating position. Voters coming away from elections with no one representing them are likely to feel alienated from the political system.

8

Introduction

Spoilers Supporters of third-party candidates are often caught in a frustrating political dilemma. If they vote for their preferred candidate, that candidate might act as a “spoiler” and actually help to elect the candidate they oppose the most. In the 2000 presidential election, for example, Greens who voted for Ralph Nader inadvertently ensured the election of George W. Bush.

Poor Quality Campaigns Ideally, candidates in U.S. elections should put issues at the forefront of their campaigns. But instead, they usually rely on slick media images and negative advertising. Such campaigns cripple the public’s ability to vote intelligently. How can the public control policy through elections when the candidates refuse to discuss in any detail what policies they will pursue once in office?

Underrepresentation of Women While women constitute more than 50 percent of the population, they made up only 14 percent of the U.S. House of Representatives in 2002. This is a figure significantly lower than in other democratic national legislatures, where women hold up to 43 percent of the seats. Many critics question whether our legislatures can effectively address women’s policy concerns when they remain virtually all-male clubs.

Lack of Minority Representation Blacks, Hispanics and Asians are consistently underrepresented in our legislatures. While African Americans constitute almost 13 percent of our population, fewer than two percent of the elected offices in the United States are occupied by black Americans. Not only is such misrepresentation unfair, it also fuels the political alienation already felt in many minority communities.

Introduction

9

Gerrymandering In the United States, election results are often determined not by how people vote but by how election district lines are drawn. Through the skillful manipulation of district lines, dominant political parties in the states often steal legislative seats from their opponents and ensure the election of incumbents. For example, in the 1990s, after redistricting in Texas, the Democrats were able to win 70 percent of the U.S. House seats, even though the party only garnered 49.9 percent of the vote. The public is certainly aware of all these problems. What many are not aware of is how they are directly related to flaws in our current electoral system. In part this is because each of these problems is usually traced to a different source. The constant reelection of incumbents is blamed on the absence of term limits. Low voter turnout is blamed on voters who are lazy and apathetic. The lack of women legislators is thought to be caused by lingering sexism among voters. Gerrymandering is considered the fault of unscrupulous and power-hungry politicians. And so on. But as this book will demonstrate, all of these problems are to some extent caused by flaws in the electoral system itself. As Hendrik Hertzberg, a political commentator for The New Yorker, has explained, to appreciate the deeper causes of these problems, we must look beyond the “usual suspects”: A lot of the political pathologies we worry about in this country— things like low voter turnout, popular alienation from politics, hatred of politicians and politics per se, the undue influence of special interests, the prevalence of negative campaigning and so on—are not caused by the usual suspects. They are not caused by the low moral character of our politicians. They are not caused by the selfishness of the electorate. They are not caused by the peculiarities of the American national character and the American political culture. They are not caused by television. They are not caused by money (although money certainly makes them worse). Instead, they are artifacts of a particular political technology. They are caused by our single-member district, geographically-based, plurality winner-take-all system of representation.7

10

Introduction

Electoral Rules and the Two-Party System Let us briefly consider just one example of how our election problems are connected to our electoral system. Let’s consider some of the problems caused by our two-party system. Voters today seem increasingly frustrated by the limited range of choices presented by the two-party system. More Americans are realizing that just as it would be ludicrous to have stores that provided only two styles of shoes or two kinds of vegetables, it is no less absurd to have a party system that provides only two choices to represent the great variety of political opinions in the United States. So why does a system that so constrains voter choices persist? Americans are certainly not wildly enthusiastic about the two major parties. Only about a third of Americans strongly identify with either of the major parties. Voters also are showing increasing interest in independent and third-party candidates, such as H. Ross Perot and Ralph Nader. And polls consistently reveal that a substantial majority of the public would like to see other parties emerge to challenge the Democrats and Republicans. Yet we remain stuck in a two-party system. Few Americans suspect that our plurality voting system is largely to blame for this frustrating situation—that our voting procedures serve to unfairly protect the two major parties from competition. But this is exactly the case. Plurality rules protect the two-party system by making it virtually impossible for minor parties to flourish in the United States. In our system, only candidates who can get the majority or plurality of the vote can win office— and by definition minor party candidates can rarely attract such a large percentage of the votes. Potential minor party supporters quickly learn that there are several disadvantages to supporting their candidate in a plurality voting system. First, they are likely to waste their vote on a candidate that cannot win. Worse, their candidate might act as a spoiler and their vote might actually help to elect the candidate they least desire. Given the disincentive for voting for third-party candidates in this system, it is not surprising that many potential supporters reluctantly switch to the lesser-of-two-evils among the major party candidates. Once it becomes clear that a minor party cannot win elections, it usually fades away and leaves the political field once again to the major parties. In this way, the plurality requirement effectively discourages new parties and acts as a barrier to a truly competitive multiparty system in the United States. Proportional representation uses very different election rules—rules that allow minor parties to become viable and to compete fairly with major par-

Introduction

11

ties. In PR, a party or candidate does not have to get a majority or plurality of the vote to win office. Parties are represented according to the proportion of the vote they win. So a small party might only get 15 percent of the vote, but it would still get 15 percent of the seats in the legislature. Thus if we were to adopt proportional representation in the United States, minor parties would immediately become a realistic alternative for the American voter. For the first time we could have a viable Green party, Libertarian party, or Reform party to compete with the Democrats and Republicans. In such a multiparty system, voters would finally enjoy a wide range of political choices at the polls—not merely a choice between the lesser-of-two-evils. And these choices could actually get elected. Such a multiparty system would have other advantages as well—advantages that would serve to reinvigorate American elections. Having a wider variety of competing parties would generate much more public excitement over elections. Voters would have a greater chance of finding someone on the ballot that truly represents their particular political views. Political campaigns would be much more interesting, with a variety of candidates expressing different ideologies and offering different analyses of our problems. It would become much easier for new ideas and new voices to be heard in the political system. The press would have to pay attention to minor-party candidates because they would be realistic candidates with a good chance of being elected. Finally, with a multiparty system, our city councils and state and federal legislatures would begin to reflect the true diversity of political views in the United States. There is, then, an intimate relationship between our use of plurality elections and the political frustrations that accompany our two-party system. This is merely one example of the numerous connections that exist between our electoral rules and the problems that plague American elections. Uncovering and analyzing these largely unnoticed connections is one of the main functions of this book. Only when Americans are fully aware of the extent of the political problems caused by our current voting system will they be ready to consider alternative arrangements like proportional representation. The other main purpose of this book is to show how adopting proportional representation in the United States would help to address these problems. This reform would prevent or mitigate all of the election problems described earlier in the chapter. Besides removing unfair barriers to third-party candidates and broadening voter choices, PR would also virtually eliminate wasted votes. Spoilers would become a thing of the past. Elections would become

12

Introduction

more competitive because all parties would have a good chance of electing someone in each district. We would no longer have to put with gerrymandering. Campaigns would concentrate more on issues than on mudslinging. And it would be much easier for women and minorities to occupy their fair share of seats in our legislatures. Given all the electoral problems that PR helps to solve, it is easy to see why it has become the most common and popular voting system in advanced Western democracies.

Electoral Reform and Revitalizing American Democracy While adopting proportional representation would certainly improve American elections, this reform would also have a more profound political effect: it would improve the basic quality of democracy in the United States. As will be seen in chapter 8, a number of scholars now believe that electoral system reform raises basic questions of democratic theory and practice: Who deserves representation? Who should participate in political dialogue? Who should share in political power? Arend Lijphart, for instance, has argued that the choice of electoral system is part of a basic choice between a political system that is inclusive and disperses power among as many people as possible, and one that is exclusive and gives power to majorities only.8 Lani Guinier, a professor at the Harvard Law School, has also written extensively about the broader political implications of proportional representation. For her, electoral system reform is about much more than fair representation for different political groups. As she explains, “Election reform, including proportional representation, is not, however, primarily about electoral rules. It is not simply about getting more people of color and women in office, although that would be an important incidental benefit. It is about transforming how power itself is exercised and shared.”9 Guinier sees proportional representation as part of an effort to “breathe new life into American democracy.” There is no doubt that our democracy is in need of this kind of reinvigoration. There are many indications that our democratic process in ailing. Polls show that most people believe that public officials do not care what they think, and that many people feel that they have no say in government.10 Well-funded special interests dominate in campaign contributions and in the lobbying process. Alienation from the two major parties is common, and most people do not even bother to vote most of the time. Proportional representation can be an important step toward addressing this

Introduction

13

democratic malaise. It would not only make elections fairer and more exciting, it would also encourage more citizen participation, broaden political debate, and make government more representative and accountable to the public. PR would ensure that more Americans have a say in the policymaking process. And legislatures that are more representative of the public are much more likely to produce policies that actually reflect the will of the people. As the Canadian political scientist Henry Milner has observed: “A country’s electoral system not only affects who sits in its [legislatures] but the type of issues that enter policy debate and the policies that emerge from that debate.”11 There is, then, much more at stake in electoral reform than might first appear. In this sense, investigating electoral systems is a bit like one of those mystery novels in which a private detective takes on a routine divorce case, only to see it quickly develop into a more elaborate case of blackmail and murder. Electoral procedures might seem at first glance to be a rather mundane subject. But once you begin to investigate electoral rules and to follow the trail of political effects emanating from them, you soon begin to see that these rules not only have dramatic effects on the results of elections, but also have important implications for the quality and extent of democracy itself. However, while the adoption of PR in the United States would solve many of our election problems and make our political system more democratic, this is not to suggest that it is a political panacea. While a vital reform, it is certainly not a cure for everything that ails our political system. It won’t, for example, fix the corrosive effects that special-interest money has on our elections and policymaking process. We need comprehensive campaign finance reform in order to effectively address that problem. However, we also need to recognize that no program to revitalize American democracy will be fully effective without including a change to proportional representation elections. There is simply no other reform that guarantees fair representation in the political process for all citizens and political groups. For this reason, proportional representation should be on the political agenda of all those who want to create a fairer, more egalitarian, and more responsive political system in the United States.

An Overview of the Book The first seven chapters of this book make the case against the singlemember plurality system and for proportional representation. Each chapter

14

Introduction

examines a particular political problem caused by our current SMP system and explains how the adoption of PR could help to solve it. Chapter 1 explores how SMP systems deny representation to large numbers of voters and produce unfair and unrepresentative election results. These problems can be solved by adopting a PR system that minimizes wasted votes and allocates party seats in a more fair and equitable manner. Misrepresentation in plurality systems can also take place deliberately through gerrymandering. This persistent problem is the topic of chapter 2. By manipulating the drawing of election district lines, dominant parties routinely steal legislative seats from other parties and create safe districts that favor the reelection of incumbents. The only sure way to eliminate these problems entirely is to adopt proportional representation. The third chapter addresses how electoral systems affect the way election campaigns are waged. It focuses particularly on how SMP elections tend to encourage campaigns that neglect a serious discussion of issues, and how they give candidates incentives to pursue negative campaigns. It also shows how adopting proportional representation for American elections would help to minimize both of these problems and produce campaigns that focus more on issues and less on mudslinging. Chapter 4 addresses the problems of our two-party system, which often fails to offer voters meaningful choices at the polls. Much of the lack of party competition in the United States can be traced to our plurality voting system, which tends to artificially reinforce our current two-party oligopoly. Adopting PR would allow for a truly competitive multiparty system and ensure that a wider variety of political perspectives would be represented in our legislatures. Chapter 5 investigates the severe underrepresentation of women in our legislatures. Evidence clearly shows that countries with proportional representation consistently nominate and elect much higher percentages of female candidates. The chronic underrepresentation of racial minorities in our legislatures is the problem highlighted in the sixth chapter. Considerable evidence indicates that the U.S. electoral system bears much of the responsibility for this problem. While the creation of majority-minority districts has mitigated this problem somewhat, the Supreme Court has now declared many of these districts unconstitutional. In the end, proportional representation would be the most effective way to ensure that all racial and ethnic groups are able to elect their fair share of representatives. Chapter 7 considers the problem of low voter turnout in the United States. Many factors contribute to this problem, but some can be traced to

Introduction

15

our single-member plurality system and the various ways it discourages voters. PR could encourage increased electoral participation in several ways: (1) by ensuring that most votes are not wasted, (2) by increasing the number of parties from which to choose, (3) by increasing electoral competition, and (4) by encouraging parties to mobilize voters in all districts. Chapter 8 discusses the deeper implications of proportional representation for democratic theory and practice. It examines the work of a number of electoral scholars who argue that PR encourages a different and better kind of democracy—one that is more inclusive and egalitarian. Chapter 9 considers the other side of this debate: the case against proportional representation and for winner-take-all elections. It discusses the most common political objections to PR, including the charges that it is too complicated for voters, creates unstable coalition governments, encourages political extremism, and deprives citizens of local representatives. These and other criticisms of PR are examined in detail and largely refuted. Many are inaccurate or exaggerated, and others can be addressed by simple modifications in PR election procedures. Chapter 10 considers in more detail the various alternatives to singlemember district elections. It first discusses several electoral alternatives other than PR—such as instant runoff voting and cumulative voting—and explains why they are inferior to proportional representation. The chapter then turns to the question of which specific form of PR would be best for the United States, examining the advantages and disadvantages of each. The final chapter considers the political prospects for proportional representation in the United States. It examines the main obstacles that stand in the way of proportional representation—which include public ignorance, constitutional barriers, and the opposition of political elites—and also looks at several ways that these obstacles can be overcome.

A Brief Tutorial One last topic must be addressed. In order to follow the case for proportional representation, readers must have some familiarity with how it works. What follows is a description of the main electoral options discussed in the book. First I will very briefly go over the two main forms of winner-take-all elections we already use in the United States: single-member plurality, and at-large. I am assuming that most readers are already somewhat familiar with

16

Introduction

these systems. Next will be descriptions of the three most common systems of proportional representation. These descriptions will necessarily be more thorough, since these systems are new to most Americans. For readers interested in more details about these systems, a number of excellent works about electoral systems can be consulted for further information.12 At this point, however, the description of the systems below should be sufficient to allow readers to follow the arguments set out in the rest of the book.

Winner-Take-All Systems Single-Member Plurality This is the electoral system familiar to Americans, and also used in Great Britain and Canada. Single-member means that only a single legislative seat is filled in each district election. On the ballot, candidates are listed for each position, and voters indicate their choice of one of them—by marking an X, pulling a voting lever, etc. The votes of each candidate are totaled, and the winner is the candidate who receives the most votes—the plurality of votes.

At Large Systems The at-large system is a less common version of plurality elections but one that continues to be used in many local elections in the United States. It uses large, multimember districts in which many officials are elected at once. For example, in a city council contest for seven seats, all candidates are elected on one citywide ballot with voters casting seven votes each. The winners are those candidates who receive the highest number of votes. The results of at-large elections are similar to single-member plurality elections in that only the candidates with the majority or plurality of votes are elected.

Proportional Representation Systems While there are a variety of proportional representation systems, they all share some common features. First, they are all built on the same political principles: that all voters deserve representation and that all political groups in society deserve to be represented in our legislatures in proportion to their

17

Introduction

strength in the electorate. In other words, everyone should have the right to fair representation. In order to achieve this fair representation, all PR systems have certain basic characteristics—characteristics that set them apart from a single-member plurality system. First, they all use multi-member districts. Instead of electing one person in each district, several people are elected. These multi-member districts may be relatively small, with only three or four members, or they may be larger, with ten or more members. (See the accompanying figures, which illustrate districting maps for a hypothetical 50-person state senate. Figure 1.1 shows 50 single-seat districts, as is common with plurality systems, figure 1.2 depicts 10 five-seat PR districts, and figure 1.3 shows 5 ten-seat PR districts.) The second characteristic of all PR systems is that they divide up the seats in these multi-member districts according to the proportion of votes received by the various parties or groups running candidates. Thus if the candidates of a party win 40 percent of the vote in a 10 member district, they receive four of the ten seats—or 40 percent of the seats. If another party wins 20 percent of the vote, they get two seats, and so on. That, in a nutshell, is how proportional representation works. But while all PR systems have the same goals of ensuring that all voters receive some representation and that all groups are represented fairly, various systems do have different ways of achieving these goals. So it is helpful to see how different kinds of PR systems work in practice.

figure i.1 50 Single-Seat Districts

figure i.2 Ten 5-Seat PR districts

figure i.3 Five 10-Seat PR Districts

18

Introduction

Party List PR Party list voting systems are by far the most common form of proportional representation. More than 80 percent of the PR systems used worldwide are some form of party list voting. It remains the system used in most European democracies and in many newly democratized countries, including South Africa. In this system, legislators are elected in large, multi-member districts. Each party puts up a list or slate of candidates equal to the number of seats in the district. Independent candidates may also run, and they are listed separately on the ballot as if they were their own party (see ballot 1). On the ballot, voters indicate their preference for a particular party and the parties then receive seats in proportion to their share of the vote. So in a five-member district, if the Democrats win 40 percent of the vote, they would win two of the five seats. The top two Democratic candidates on the list would occupy those seats. (For a more detailed discussion of how votes are translated into seats, see Appendix A which describes the various formulas used for this purpose.) There are two broad types of list systems: closed list and open list. In a closed list system—the original form of party list voting—the party fixes the order in which the candidates are listed and elected, and the voter simply casts a vote for the party as a whole. This is shown in the first party list ballot, ballot 1. Closed Party List system

Introduction

19

which illustrates an election for the House of Representatives in a five-seat district. Voters are not able to indicate their preference for any candidates on the list, but must accept the list in the order presented by the party.13 Winning candidates are selected in the exact order they appear on the original list. So in the example here, if the Democrats won two seats, the first two candidates on the pre-ordered list—Paine and Kusor—would be elected. Most European democracies now use the open list form of party list voting. This approach allows voters to express a preference for particular candidates, not just parties. It is designed to give voters some say over the order of the list and thus which candidates get elected. One version of this is illustrated in the second party list ballot. Voters are presented with unordered or random lists of candidates chosen in party primaries. Voters cannot vote for a party directly, but must cast a vote for an individual candidate. This vote counts for the specific candidate as well as for the party. So the order of the final list completely depends on the number of votes won by each candidate on the list. The most popular candidates rise to the top of the list and have a better chance of being elected. In our example, if the Democrats won two seats, and Silver and Rodriguez received the highest and next highest number of individual votes, they would rise to the top of the list and be elected. This example is similar to the system used in Finland and widely considered to be the most open version of list voting. ballot 2. Open Party List System

20

Introduction

Mixed-Member PR Mixed-member proportional representation goes by a variety of other names, including “the additional member system,” “compensatory PR,” the “two-vote system,” and “the German system.” It is an attempt to combine a single-member district system with a proportional voting system. Half of the members of the legislature are elected in single-member district plurality contests. The other half are elected by a party list vote and added on to the district members so that each party has its appropriate share of seats in the legislature. Proponents claim that mixed-member proportional voting (MMP) is the best of both worlds: providing the geographical representation and close constituency ties of single-member plurality voting along with the fairness and diversity of representation that comes with PR voting. This system was originally invented in West Germany right after World War II, and more recently has been adopted by New Zealand, Scotland, and Wales. In MMP, people cast votes on a double ballot. First, on the left part of the ballot, they vote for a district representative. This part of the ballot is a single-member district plurality contest to see which person will represent the district in the legislature. The person with the most votes wins. Typically half of the seats in the legislature are filled in this way. So in a hypothetical 100-member state legislature, the winners of these district contests would occupy 50 of the seats. On the right part of the ballot—the party list portion—voters indicate their choice among the parties, and the other half of the seats in the legislature are filled from regional lists of candidates chosen by these parties. The party lists are closed in the German version. These party list votes are counted on a national basis to determine the total portion of the 100-seat legislature that each party deserves. Candidates from each party’s lists are then added to its district winners until that party achieves its appropriate share of seats. Table I.l illustrates how this process works for our hypothetical election. The Democrats won 40 percent of the party list votes in the 100-member state legislature, so they would be entitled to a total of 40 of the 100 seats. Since they already elected 28 of their candidates in district elections, they would then add 12 more from their regional party lists to come up to their quota of 40 seats.

Single Transferable Vote This system of proportional representation is known by several names. Political scientists call it “the single transferable vote” or STV. In the United

Introduction

21

ballot 3. Mixed-Member Proportional Representation System

States, electoral reform activists have taken to calling it “choice voting” or the “preference vote.” Currently this system is used to elect parliaments in Ireland and Malta. In Australia it is used to elect the federal Senate, as well as the legislatures in several states there. It is also the PR system that was used in a number of cities in the United States during the twentieth century, including New York, Cincinnati, Cleveland, Toledo, and Boulder. It continues to be used today in Cambridge, Massachusetts.

22

Introduction

table i.1 Voting Results and Seats Allocations in Mixed-Member Voting Political Parties

Number of District Seats Won

Percentage of the National Party List Vote

Total Number of Seats Deserved by Party

Number of Seats Added from Party Lists

Democratic

28

40%

40

12

Republican

18

36%

36

18

Libertarian

4

18%

18

14

New Party

0

6%

6

6

50

100%

100

50

Totals

In the STV voting process, all candidates are listed in the same place on the ballot. Instead of voting for one person, voters rank each candidate in their order of choice. So if you like Campbell best, you would mark the “1” after his name. If you liked Gomez second best, you would mark “2” by his name, and so on. You can rank as few or as many as you want. This ballot illustrates the use of the AccuVote system used in Cambridge, Massachusetts to elect its city council and school board. Voters fill in the ranking numbers as they would for standardized tests taken in school, which allows for computerized vote counting and ballot transfers. The principle behind the STV counting system is that voters get to direct their votes to maximize their effectiveness. The result is that almost all voters help to elect one of their top-ranked candidates. In order to begin the count, a threshold (or quota) is established that represents the minimum number of first-choice votes a candidate must receive to get elected. The threshold usually consists of the total number of voters divided by the number of seats to be filled plus one. Thus in a nine-seat district, a candidate would need just over 1/9⫹1 or one-tenth of the vote, to be elected on the first count. To decide who is elected, all the first place votes are counted and any candidates that receive enough votes to place them over the threshold are declared elected. If there are still seats to be filled, then a process of transferring votes begins—a process unique to this voting system and one designed to ensure that most people contribute to the election of a candidate. Two kinds of

Introduction

23

ballot 4. Single Transferable Vote

transfers take place. First, once a candidate reaches the threshold and is elected, any surplus ballots beyond what the candidate needed to elected are redistributed to the next available preferred candidate indicated on the ballot. After the redistribution of surplus ballots, the votes are recounted to see if any remaining candidates have reached the threshold. If no candidates reach the threshold during a counting round, a second kind of transfer takes place: the last place candidate is eliminated and those votes are transferred to the secondchoice candidates on the ballot and another count is made. The redistribution continues until all seats are filled. (For a more detailed description of this vote counting and ballot transfer process, see Appendix B.)

24

Introduction

Perhaps the simplest way to envision this transfer system is to imagine students in a class trying to elect a committee. Any student who wishes to run stands at the front of the class, and the other students vote for their candidates by standing beside them. Students standing almost alone next to their candidate will soon discover that this person has no chance of being elected and will choose another candidate to stand beside. Similarly, some students standing in very large groups around a candidate will decide that not all are needed to elect that candidate and so move on to help elect another candidate they like. In the end, after all the shuffling, most students would be standing next to a candidate who will be elected. The workings of STV are essentially similar, except that ballots, not bodies, are moved from one candidate to another. This transfer process is a bit complicated, so why does it exist? The transfer process was invented primarily to reduce the problem of wasted votes— votes that are cast but do not actually elect anyone. Plurality systems routinely waste large numbers of votes and this is why they are prone to such problems as party misrepresentation, and the underrepresentation of political minorities, racial minorities, and women. The transfer process in STV is designed to ensure that the fewest votes are wasted and that the maximum number of people get to elect a representative to office. It also ensures that all political groups receive the number of seats they deserve based on their portion of the vote.

Simpler Than They Look To American eyes, these various systems of proportional representation can often appear at first to be overly complex and confusing. And while the mechanics of seat allocation can sometimes be complicated, the actual voting process is not intimidating at all and can be easily used by the average citizen. Voters need not understand all the mathematics of these systems to use them effectively. To use an analogy: you don’t have to understand how all the electronic components in your car radio work in order to use it to find the kind of music you like. The party list system, the mixed-member system, and the single transferable vote have all been used for decades in other Western democracies. All of these systems have proven to be very voter-friendly. Voters in these countries have had no trouble using these systems, as indicated by the very high voter turnout rates and very low ballot-spoilage rates. Certainly we could expect that American voters would easily master the use of these systems as well.

1

Fair Representation for All

The main reason that most other Western countries have rejected the U.S. voting system would strike most Americans as quite extraordinary: They believe that our system consistently fails to provide fair representation for voters and thus violates some of the most basic principles of democratic government. The essential purpose of democratic elections is to allow as many people as possible to gain representation and to produce a legislature that faithfully reflects the various political views of the electorate. But single-member plurality (SMP) elections often fail miserably on both of these counts. The SMP system routinely denies representation to large numbers of voters and frequently produces distorted representation of political parties. At times, SMP elections are so misrepresentative that they actually allow the minority to rule over the majority—a breach of one of the most basic rules of democracy. Many Americans remain largely unaware of these undemocratic characteristics of our elections. Indeed, we are raised to believe that our election system is the epitome of representative democracy. We are so confident in that judgment that we often use our own election system as the yardstick by which we measure how fair and democratic political systems are in other countries. Thus it comes as quite a surprise to learn that many other countries fault our election system in the very area in which we believe it excels. Why is this the case? What are the flaws in our election system that seem so obvious to others but remain hidden to many Americans?

26

Fair Representation for All

Denying Representation to Voters One of the most serious defects of single-member plurality systems—and the one that is the root cause of nearly all of the problems that are discussed in this book—is that large numbers of voters come away from these elections without any representation, and thus go without any say in government. This problem is not an accident; it is built into the very workings of the system. In fact, SMP elections are intentionally designed to deny representation to large numbers of voters. They are designed to represent only the majority or plurality of voters in a district. Thus the only voters who are actually represented are those who vote for the winning candidate. Those who vote for the losing candidates end up with no representation at all. This is what is meant by a “winner-take-all” system: the winners get all the representation and the losers get none. In effect, when citizens vote for a losing candidate in a single-member district, they lose the right to representation. We are so used to this result in the United States that we fail to appreciate how unfair and undemocratic it actually is. In a close two-way election, the number of voters denied representation might approach half the electorate in a district. In essence, we have democracy only for the winners. When we have only partial representation, we have only partial democracy. It is exactly for this reason that some British critics of the SMP system have labeled it “only half a democracy.”1 To better appreciate the undemocratic implications of this arrangement, imagine for a moment that for some reason a large political minority was suddenly denied the right to vote in the United States. The public protest would be immediate and vociferous. Americans have long championed the idea of universal suffrage—the right of everyone to vote and to thus exercise control over our government. Denying that right to part of the population would clearly violate this most basic tenet of democracy. Yet while such formal disenfranchisement would be clearly intolerable, every one of our singlemember plurality elections has virtually the same effect, producing a kind of de facto disenfranchisement of a substantial block of voters. Voters on the losing side in a single-member plurality election have no more political influence than if they were officially denied the right to vote. Their votes are what political scientists call wasted votes—votes that produce no representation and have no impact on policymaking. Many Americans are familiar with this problem. If you are a Democrat in a predominantly Republican district, a Republican in a Democratic district, or a third-party sup-

Fair Representation for All

27

porter in any district, you eventually realize that your vote will not win you any representation. Any clear political minority in a district is doomed to political frustration. Consider, for example, the plight of Republican voters living in Massachusetts’s eighth congressional district, where no Republican has been elected for well over a century. It would make little difference if these voters were actually disenfranchised; the single-member plurality system has rendered their votes essentially useless anyway. In short, the right to vote means little if it is not accompanied by the right to representation. Voting without representation gives only the illusion of democracy. Ideally, a democracy should not restrict representation but extend it to as many of its citizens as possible. In a fully representative democracy, legislatures would include not only politicians who represent the majority, but also policymakers who speak for any substantial political minorities. This is not a new idea; it has long been expressed by many theorists of democracy. The English political philosopher John Stuart Mill, for instance, has argued: “It is an essential part of democracy that minorities should be adequately represented. No real democracy, nothing but a false show of democracy, is possible without it.”2 This, however, is not the vision of democracy that is promoted by the single-member plurality system. In this system, political minorities are intentionally denied representation. This denial of representation is not only unjust, it also contributes to a number of other political problems in the United States. For example, it contributes to voter apathy. Voters frequently complain that voting seems to be a waste of time—which is entirely true for many voters in SMP systems. Public service get-out-the-vote campaigns try to entice voters to the polls by telling them that if they don’t vote, they won’t have any say in government. But the fact is that under our current election rules, many who do vote will still have no say in our government. The issue of low voter turnout in the United States will be discussed at length in chapter 7, but it can be said now that the tendency for single-member plurality systems to deny representation to large numbers of voters clearly contributes to this chronic problem. The epidemic of wasted votes in the United States also raises questions about the legitimacy of our elected bodies. Legislatures claim the right to make policy based on the assumption that they represent the voters. But how many voters are actually represented in these bodies? How many voters actually vote for the policymakers that create our laws? The answer is very disturbing. Consider, for example, the elections for the U.S. House of Representatives in 1998. First, among the approximately 200 million eligible voters, only

28

Fair Representation for All

about 67 million (32.9 percent) chose to vote at all. Twenty-three million of these votes were wasted, leaving only 44 million who actually voted for winning candidates—or 22 percent of the electorate. Of those, only around 26 million cast votes for the Republican candidates who actually ended up in control of the House of Representatives. Thus the passage of bills in the House in 1999 and 2000 required the support of politicians who literally represented only 13 percent of the entire electorate. It is difficult to take seriously any party’s claim to a public mandate when a mere fraction of the electorate actually voted for those elected officials.

The Myth of Representation in SMP Systems Why do Americans continue to accept an election system that denies representation and thus political power to large segments of voters? One reason is the convenient myth promoted by politicians that everyone in a district is represented by the winning candidate—that that person becomes “our” representative in the legislature. Clearly, however, this can be true only in a very limited sense: when the legislator is addressing policy issues on which everyone in the district has similar interests or opinions. Some such issues exist, such as those concerning local economic development, but they are relatively few. More typical is a situation in which interests and political viewpoints diverge, especially over important issues like abortion, defense spending, taxes, affirmative action, global warming, and so forth. It is clearly misleading to pretend that one legislator can possibly represent the wide range of political opinion found in many districts. In most cases legislators will represent only the opinions of those who elected them. This situation becomes obvious to voters who try to lobby a legislator who was elected by the opposition. They quickly find out that their legislator is not really representing their perspective at all. Indeed, they can often only look on helplessly as “their” representative promotes policies that are the very opposite of what they would prefer. Another fiction that obscures the unrepresentative nature of winner-takeall elections is the notion that even if some voters are left out in one district, they are represented by a politician elected in another. According to this reasoning, if Democrats fail to elect a representative in district A, the process “evens out” because they are represented by a Democrat elected in district B. Republicans denied representation in district B gain it back in district A.

Fair Representation for All

29

But again this argument has a number of misleading assumptions. First, all party candidates are not alike, so it is a mistake to presume that voters denied representation in one district will feel they are represented by a candidate of their party elected in another district. Can a conservative Democrat from Mississippi really represent the interests or viewpoints of liberal Democrats from Connecticut? It is highly doubtful. But even if all Democratic candidates were interchangeable, there is no assurance that enough of them would be elected in other districts to represent those Democratic voters denied representation in districts in which they are a minority. In other words, it cannot be assumed that the misrepresentation in the plurality system will somehow magically balance itself out so that each party gets its rightful share of legislative seats. In fact, as we will see later, the typical result of an SMP election is for at least one party to be underrepresented in the legislature. For several reasons, then, the notion of “evening out” is simply another myth that serves to conceal the injustices built into this system.

PR: Ensuring Greater Voter Representation Americans also tend to accept the massive denial of political representation in our single-member plurality system because many of us are unaware that elections could work in any other way. We assume that it is natural and inevitable that elections will have winners and losers, as is the case in many of the other aspects of our competitive society. And if our candidate should not win, we learn to be good losers and simply try harder next time around. But while most of us have become reconciled to this arrangement, in reality this situation is not inevitable at all. It is simply not true that large numbers of voters must be denied representation in every election. This is only the case in plurality systems in which only one candidate represents each district. Proportional representation systems are explicitly designed to maximize the number of voters who are able to elect a representative—indeed that is one of the central goals of this voting system. PR seeks to ensure that few voters lose completely in an election—that almost all voters come away winners with someone in the legislature speaking up for their interests and their political values. Proportional representation systems minimize wasted votes by using large multimember districts and by allocating seats according to the proportion of votes won by various parties. This arrangement allows even small political mi-

30

Fair Representation for All

norities to elect representatives in every district. Consider, for example, a tenmember PR district that is made up of 70 percent Democrats and 30 percent Republicans. Under PR rules, the Democrats would win 7 out of the 10 seats. The Republicans, instead of being shut out as they would be in SMP elections, would win 3 out of the 10 seats. In PR districts, voters in both the majority and minority win representation. So instead of a winner-take-all system, PR is an all-are-winners system. PR presumes that in a real democracy, everyone deserves representation and everyone deserves a say in government. Countries that use proportional representation do not eliminate wasted votes entirely, but they do have far fewer of them.3 Consider Israel, where all 120 Knesset members are elected from one nationwide district. It is common for 95 percent of Israeli voters to elect someone to represent them.4 And in the last German election, more than 90 percent of the voters elected a candidate of their choice. In stark contrast, throughout the 1990s, in elections for the U.S. House of Representatives, only about 64 percent of voters elected a candidate.5 This meant that an average of more than 29 million voters came away from these elections with no representatives in the House, a remarkable result in a political system that calls itself a democracy. Even worse, these millions of voters then had to watch “their” representatives promote the very policies they voted against. We needn’t have to look abroad to find examples of how proportional representation ensures representation for more voters. One form of proportional representation, the single transferable vote (STV), was used in two dozen American cities during the twentieth century. In this system, voters rank the candidates, giving a “1” to their first choice, a “2” to their second, etc. Vote transfers are then used to ensure the maximum number of effective votes. For example, if a voter’s first choice candidate is not elected, the ballot is transferred to the voter’s second choice to help that candidate get elected. This approach resulted in dramatic decreases in the number of wasted votes where it was used. A typical example was Cleveland, Ohio, where STV was used from 1923 to 1933 to elect the city council. Local advocates of PR had argued that this system would ensure that a larger share of the voters would cast votes that actually elected a candidate. A study by Kathleen Barber confirmed that they were right.6 During the STV years, almost 80 percent of the voters were successful, and the number of wasted votes was reduced to only 20 percent. But after Cleveland returned to using single-member plurality voting, only about 60 percent of the voters were successful in electing a representative.

Fair Representation for All

31

A similar but more recent example can be found in Cambridge, Massachusetts, the one remaining city in the United States that still uses the single transferable vote for its city council and school board elections. An analysis of one typical election there revealed that 96 percent of the electorate saw their votes contribute to the election of one of the candidates they preferred, with 80 percent having their first or second choices elected.7 One longtime resident of Cambridge, John Moot, explained some years ago how STV gave him the kind of representation he could never receive under single-member plurality arrangements: “Since becoming a resident of Cambridge in the 1950s, I have been fortunate to have always had a representative of my choice on the City Council and on the School Committee, thanks to proportional representation. In contrast, I have never had a representative of my choice in the U.S. House of Representatives because I am a Republican in what was Tip O’Neill’s and is now Joe Kennedy’s district.”8 Given the result of fewer wasted votes, one would expect that voters would feel more positive about elections in a proportional system, and there is some evidence that this is indeed the case. Banducci, Donovan, and Karp conducted a study of voter attitudes toward elections and government in New Zealand when that country switched from SMP to PR elections in the 1990s. They found that after adopting PR, more voters came to believe that their votes “really counted.” This certainly makes sense, since more votes do in fact count under PR rules. Understandably this more positive attitude toward voting was strongest among those political minorities that had had trouble winning representation in the previous SMP system.9 This same study also found that after the switch to PR fewer New Zealand voters thought that their representatives did not care or were out of touch. This may be related to the fact that PR elections can also work to increase contacts between voters and their legislators. Joseph Zimmerman noted that PR serves to maximize the accessibility of legislators after the election.10 Single-member plurality arrangements often tend to limit public contact with legislators. Elected officials need not feel obligated to be available to a constituent from a different political party. Moreover, voters may often feel reluctant to approach an official they voted against. But under proportional representation, almost all voters can approach a representative of their party that they helped to elect and who can provide a sympathetic ear. In this way PR arrangements help to make the political system more representative and more responsive to voters, both during and after elections.

32

Fair Representation for All

Unrepresentative Legislatures What is most alarming about the wasted votes in single-member plurality systems is not simply that large numbers of people are denied representation, but also that the resulting legislatures often do not accurately represent the public. In these SMP legislatures, some groups of voters get more representation than they deserve and others get less. This process of misrepresentation begins on the district level, where the group that votes for the winning candidate gets all of the representation and those who vote for the loser get none. When 51 percent of the voters win 100 percent of representation and 49 percent of the voters win 0 percent of the representation, this is hardly fair and accurate representation. This process of misrepresentation that starts on the district level often produces distorted representation on the legislative level as well. Single-member plurality elections usually produce city, state, and federal legislatures where some parties are overrepresented and other parties underrepresented. This kind of unfair representation is a serious political defect, considering that the basic purpose of elections in the United States is to produce lawmaking bodies that faithfully reflect the will of the people. The notion that legislatures should reflect the political desires of the public as accurately as possible is central to most Americans’ conception of democracy. We believe that the more representative such a body is, the more likely it is to pass laws that embody the views of the public—which is much of what democratic government is all about. Thus we possess a rule for evaluating how democratic an election system is: How well does it produce a legislature that accurately mirrors the public’s political preferences? Roughly speaking, the more representative the legislature, the more democratic the election system. This notion that elections should produce legislatures that mirror the political preferences of the public has some deep roots in American political history. In the early days of our republic, for example, some founding fathers used the idea of a portrait to express the ideal relationship between a legislature and the public. John Adams argued that legislatures in the United States “should be an exact portrait, in miniature, of the people at large, as it should think, feel, reason, and act like them.”11 Using a similar analogy, James Wilson argued at the Constitutional Convention that “the portrait is excellent in proportion to its being a good likeness,” so “the legislature ought to be the most exact transcript of the whole society . . . the faithful echo of the voices of the people.”12

Fair Representation for All

33

Unfortunately, however, our voting system consistently violates this basic democratic principle; it often produces legislatures that fail to accurately mirror the political preferences of the public. Indeed, a single-member plurality electoral system often functions more like a fun-house mirror, enlarging some features of our body politic and shrinking others, until what is left is often only a distorted image of the voting public. Considerable evidence of this problem exists. In repeated studies of the accuracy of representation in the legislatures of various Western democracies, the United States consistently ranks near the bottom.13 Political scientists usually measure how representative a legislature is by comparing the share of the vote the public has given to various parties to the share of legislative seats allocated to those parties by the election. Ideally, a party’s strength in the nation should correlate closely with its strength in the legislature. If a party receives 40 percent of the vote, it should receive 40 percent of the seats. The larger the deviation of seats from votes, the more unrepresentative the legislature. Table 1.1 ranks 20 democracies according to the average percentage by which political parties’ shares of seats deviate from their voting shares—using a widely accepted formula developed by Michael Gallagher. What is most interesting about this table is not simply the low position of the United States, but that all of the countries ranking lowest on the list use winner-take-all elections—either single-member plurality or single-member majority voting systems. (A single-member majority system is essentially the same as the plurality system except that it requires that the winning candidate receive a majority of the vote in the district. This is usually achieved by a runoff.) Note also that all of those countries that have the most representative legislatures have some form of proportional representation. This is a strong indication that the electoral arrangements are a major cause of misrepresentation. You may notice that the U.S. ranks somewhat better than the other plurality/majority countries. This is because of the absence of viable third parties in this country. Other winner-take-all countries, such as Great Britain and France, have relatively strong minor parties that attract respectable numbers of votes but are severely underrepresented in the national legislature, and this is one element that increases their disproportionality. The barriers to minor parties are much higher in the United States.14 For example, most states require third parties to gather substantial numbers of petition signatures to appear on the ballot, whereas no such requirement exists in Great Britain. In any case, the general pattern of disproportionality remains clear: legislatures tend to be much less representative in plurality/majority countries than in PR countries.

34

Fair Representation for All

table 1.1 Vote/Seat Disproportionality in National Legislatures in 15 Countries Country

Electoral System

Level of Disproportionality

Netherlands

PR

1.4

Austria

PR

1.5

Germany

PR

1.6

Denmark

PR

1.7

Sweden

PR

1.9

Italy

PR

2.6

Belgium

PR

3.0

Finland

PR

3.2

Ireland

PR

3.3

Norway

PR

3.7

United States

Plurality

5.9

Australia

Majority

9.6

Canada

Plurality

12.1

France

Majority

14.0

United Kingdom

Plurality

14.4

Source: G. Bingham Powell, Elections as Instruments of Democracy (New Haven: Yale University Press, 2000), p. 96.

Political scientists have long known why single-member district systems tend to produce unrepresentative legislatures: these systems have a built-in bias in favor of the largest political party.15 As numerous empirical studies in this country and abroad have confirmed, winner-take-all systems almost inevitably create legislatures that overrepresent the largest political party and underrepresent the smaller ones. This effect can be seen clearly in a study done by G. Bingham Powell. He compared the percentage of seats to the percentage of votes for various sized parties in plurality/majority electoral systems and proportional representation systems. The results are shown in table 1.2. On average, plurality/majority systems produced significant overrepresentation of the largest party, giving them 55 percent of the legislative

35

Fair Representation for All

table 1.2 Vote/Seat Outcomes for Parties in Different Electoral Systems Electoral System

Party Rank

Vote %

Seat %

Largest party

45

55

2nd party

38

39

All other parties

18

7

Largest party

39

41

2nd party

30

32

All other parties

31

28

Plurality /Majority

Proportional Representation

Source: G. Bingham Powell, Elections as Instruments of Democracy (New Haven: Yale University Press, 2000), p. 138. Note: Scores for plurality/majority systems based on 45 elections in six countries; scores for PR systems based on 80 elections in nine countries.

seats based on 45 percent of the vote. The second largest parties tended to receive accurate representation.16 The other small parties were severely disadvantaged, averaging only 7 percent of the legislative seats based on 18 percent of the votes. In contrast, the distortions found in proportional representation systems were relatively minimal for all parties—which is, of course, the intention of these voting systems. How this bias toward the largest party comes about is easy to see. Imagine a state that elects ten members to the U.S. House of Representatives, each from one of ten single-member districts. Assume that 60 percent of the voters are Democrats and 40 percent are Republicans. Ideally, the state would send six Democrats and four Republicans to Congress, a distribution that would accurately mirror the political preferences of that state’s citizens. And this would indeed be the result under a proportional representation system. But under plurality rules the results could be radically different. Assuming that the 60/40 proportion of Democrats to Republicans held in each district, the election would send ten Democrats and no Republicans to the House.

36

Fair Representation for All

One hundred percent of the seats for 60 percent of the vote is nowhere close to a fair or representative outcome. Such outcomes are not merely theoretical possibilities. This kind of blatant misrepresentation has often been seen in recent elections for the Massachusetts delegation to the U.S. House of Representatives. Republican candidates consistently win about one-third of the votes cast in these contests— but they usually win no seats at all. The Democratic majority wins every one of the state’s ten seats, leaving hundreds of thousands of Massachusetts Republicans with no voice whatsoever in the U.S. House of Representatives. Of course in reality the results of SMP elections are usually not quite this exaggerated—with one party taking all the seats. This is because party supporters are usually not distributed equally among all districts in a state, so that even if the Republicans are in a minority statewide, they may constitute a majority in several districts. But the fact remains that our SMP system has an inherent and strong tendency to overrepresent the largest party and to underrepresent or completely shut the smaller parties out of the legislature. Any election for the House of Representatives will demonstrate this problem at work in the real world. Consider the 2000 U.S. House election in Iowa, which sends five members to the House. The Democrats won 42 percent of the vote, but only 20 percent of the seats. That same year, in Oregon, it was the Republicans who were the victims of underrepresentation. They won almost half of the votes (44 percent), but received only one of the five (20 percent) U.S. House seats. Such results are not unusual. The Center for Voting and Democracy calculated the discrepancy between the parties’ percentage of the vote and their percentage of the seats for each state during the 2000 elections for the U.S. House. On average, the distortion was 26.46 percent.17 Such examples of exaggerated majorities clearly demonstrate the tendency of SMP elections to violate basic principles of fair and accurate representation. This system unfairly diminishes the power of minority parties and artificially enhances the power of the largest party. In short, in singlemember plurality elections, the politically rich get richer and the politically poor get poorer.

Manufactured Majorities The distortions in representation produced by single-member plurality elections can create a number of serious political problems. Probably the

Fair Representation for All

37

least objectionable result is the production of “exaggerated majorities” in the legislature. If a party wins 55 percent of the vote, and yet receives 65 percent of the seats, this is an exaggerated majority. This result is clearly unrepresentative and does give an artificially inflated political mandate to the party in power. However, these exaggerated majorities usually have little political impact in the legislature, because even if the representation had been more fair, the same party would have won a majority of the seats and thus still controlled policymaking. However, SMP elections can produce much more unfair and undemocratic outcomes in the legislature. In particular, they can create “manufactured majorities.” Manufactured majorities are produced when election procedures give a party that receives less than 50 percent of the vote more than 50 percent of the seats in the legislature. For example, a party might win on 45 percent of the vote, but be awarded 55 percent of the legislative seats. This is in contrast to an “earned majority” in which a party wins a majority of both votes and seats. Manufactured majorities can easily happen when more than two parties contest an election. The process of misrepresentation begins on the district level. When candidates for three or more parties run in a single-member district, the vote is often so divided that the winning candidate does not receive a majority. Party A’s candidate might receive 40 percent of the vote, while parties B and C’s candidates receive 35 percent and 25 percent respectively. Thus party A wins 100 percent of the representation with less than a majority of the vote. If this process occurs in enough districts, party A can easily receive a total of considerably less than 50 percent of the votes, but garner more than 50 percent of the seats in the legislature. Great Britain, the traditional home of single-member plurality elections, has often been plagued with manufactured majorities. For instance, the Conservative party ruled with large parliamentary majorities from 1979 to 1997 despite never having won more than 45 percent of the vote in any election. Typical was the 1987 election where the Conservative party won 42.3 percent of the vote, the Labour party won 30.8 percent, and the Alliance of Social Democrats and Liberals 22.8 percent. However, because Labour and the Alliance split the center-left vote, the Conservatives were able to win 58 percent of the seats in Parliament. The large majority of seats garnered by Thatcher’s party gave the illusion of a strong public mandate for the prime minister and her conservative ideology. This was evident in the newspaper headlines in the United States. The New York Times declared, “British Voters Fashion a Virtual Coronet for Thatcher.” The Washington Post trum-

38

Fair Representation for All

peted, “Thatcher Basks in Huge Win.” But while such interpretations gave the impression of a ringing public endorsement of the Conservative party, the opposite was in fact the case. The clear majority of British voters (53.6 percent) supported parties considerably to the left of the Conservative party—either liberal or socialist. But the SMP system allocated those parties substantially fewer seats than they deserved, particularly the Social Democrats and Liberals who received 22.8 percent of the vote but were given only a 3.4 percent of the seats. Thus it was the election system, not the voters, that put the Conservatives in control of the government. The main problem with manufactured majorities is that they violate one of the most sacred tenets of democratic politics: majority rule. Manufactured majorities give the representatives of a minority of voters unchecked domination over the majority of citizens. A single party rules in the government even though it does not have a mandate from the majority of the voters. In Great Britain, this meant that the Conservative party was able to pursue a radically right-wing policy agenda for almost two decades, even though a majority of the British voters had endorsed parties on the political left—either the Alliance or the Labour party. Whether one sympathizes with the Conservative party and its policy agenda is beside the point; the central question here is whether it is legitimate for any party supported by only a minority of the voters to impose its policies on the public in this way. It is difficult to see how such an arrangement can be justified in any society claiming to be democratic. Not surprisingly, surveys have shown that many voters are uncomfortable with manufactured majorities. One of the most interesting of these surveys was conducted in New Zealand. New Zealand voters are familiar with both SMP and PR election systems, having switched from the former to the latter in the 1990s. The New Zealand SMP election system had also produced a number of manufactured majorities in their national parliament—so many voters were familiar with the problem. In the survey, voters were asked to imagine that one party wins the most votes in an election with about 40 percent of the vote. When asked if that party should get more than half of the parliamentary seats, so that it could govern easily on its own, only 19 percent the respondents said “yes.”18

A Common Problem One might think that an electoral travesty like a manufactured majority would be a relatively rare occurrence. After all, who would put up with a sys-

39

Fair Representation for All

table 1.3 Comparison of Republican Votes and Republican Seats in U.S. House Elections, 1996-2000 % of Republican Votes

% of Republican House Seats

1996

48.9

52.2

1998

48.9

51.3

2000

47.9

50.8

Source: Statistical Abstract of the United States (Washington, DC: U.S. Printing Office, 2001).

tem that often allowed the minority to rule? But in fact manufactured majorities are quite common in winner-take-all elections. Arend Lijphart did a study of fourteen countries that used either plurality or majority voting systems and found that manufactured majorities occurred in 43.7 percent of the elections between 1945 and 1996.19 This is a stunning level of electoral malfunctioning. It is one reason that another of our leading electoral scholars, Matthew Shugart, has referred to manufactured majorities as the “worst pathology” of plurality elections.20 Manufacturing majorities do tend to occur at a lower rate in the United States than in countries Great Britain or Canada, again because of the absence of substantial minor parties here. But it is still easy to find instances of this problem in this country. Consider, for example, the results of the 1996, 1998, and 2000 elections for the U.S. House of Representatives, as shown in table 1.3. In those years, the Republicans never won a majority of the vote, and yet they won a majority of the seats in the House every time. The Republican party was able to push its policies through the House in those years not because a majority of voters endorsed that party, but because our peculiar election system awarded that party an unfair portion of House seats. It might be argued that their share of the vote was so close to a majority that the distortion does not really matter. But the principle of majority rule is not suspended simply because a party gets close to that level of support. In 2000, the Democrats also came close to a majority of the votes, with 47.94 percent versus the 47.96 percent of the Republicans, but it seems doubtful that Republicans would have considered it legitimate for the Democrats to be awarded a majority of seats in that election. The basic principle remains: it is unfair to allow a party to rule that most people voted against.

40

Fair Representation for All

PR Discourages Manufactured Majorities Some critics of proportional representation have argued it would do little good to abandon SMP elections because PR is also prone to manufactured majorities. Mark Rush, for instance, has suggested that manufactured majorities are “not an uncommon occurrence” in PR countries.21 In fact, however, the record shows that PR strongly discourages manufactured majorities. While they are possible in PR systems, they are much less likely to occur than in SMP systems. Lijphart’s extensive comparative study of elections between 1945 and 1996 found that such outcomes were relatively rare in the proportional representation countries studied. They occurred in only 9 percent of all PR elections—a dramatic contrast to the 44 percent rate found in plurality and majority countries.22 PR arrangements discourage manufactured majorities because they are explicitly designed to eliminate the distortions in party representation and to ensure that parties receive their fair share of legislative seats. Thus is it highly likely that if a party receives less than a majority of the vote, it will also receive less than a majority of the seats. In PR systems, it is virtually impossible for a party that receives 42 percent of the vote to receive 55 percent of the seats—as can happen in SMP systems. Also, it is nearly always assured that the majority of legislative seats will go to the party or parties that received the majority of the vote. If no one party garners the majority of the votes, a coalition of parties representing a majority of the voters rules the legislature. In either case, policymakers actually have a mandate from the majority of the voters—which is how it is supposed to be in a democracy. An intriguing study conducted by G. Bingham Powell has confirmed that proportional representation elections do a better job of ensuring that the legislators who make policy represent the policy positions of the majority of the electorate. In his book, Elections as Instruments of Democracy, Powell studied 36 democracies that used either PR elections or majority/plurality elections. One aim of the study was to measure the similarity of political views of policymakers and citizens. As Powell argues, the purpose of elections is to ensure that “policymakers do what their citizens want them to do,”23 so it makes sense to examine how closely the general policy positions of elected policymakers actually correspond to those of the most voters. Interestingly, Powell had expected to find a close congruence between the political views of policymakers and citizens in both majority/plurality and PR countries. But what he found was that, in general, policymakers in

Fair Representation for All

41

PR countries had policy positions that corresponded much more closely to that of the majority of citizens. The policymakers in majority/plurality countries tended to have less similar views to the majority. He attributed much of the inferiority of the majority/plurality governments to the tendency of these electoral systems to produce manufactured majorities. Indeed, it is not surprising at all that a legislature controlled by one party—a party that may not even represent a majority of the voters—would sometimes adopt political positions that are not in accord with the preferences of most citizens. In contrast, PR legislatures are usually ruled by coalitions—coalitions that do in fact collectively represent the political positions of the largest part of the electorate. Powell also believes that it is significant that in these coalitions, several parties—not one—have an influence over policymaking. The bargaining and negotiation that takes place between the parties is much more likely to produce policy positions that more closely correspond to the views of most of the public. British politics provide a good illustration of Powell’s points. As seen earlier, plurality elections at one time allowed the Conservative party to rule for almost two decades, without ever having received close to a majority of the votes. It pursued economic and social policies that moved that country sharply in a conservative direction—policies that did not reflect the ideological positions of most British citizens. If Britain had used PR elections, however, the political and policy results would have been quite different. One likely possibility would have been a center-left coalition of the Liberal/ Democratic Alliance and the Labour party—parties that were in fact supported by a majority of voters. Such a coalition would have been more likely to produce policies closer to the views of most British voters. A less likely, but possible, outcome of PR elections would have been a center-right coalition of the Liberal/Democrats and the Conservative party. But even this situation would have produced more representative policies than those passed by the Conservatives alone. The Conservatives would have had to bargain and compromise with the Liberal/Democrats, which would have forced them to agree on some policies closer to the political center and thus pleasing to more voters. Either way, a PR coalition government would have produced more representative policies than those passed by a manufactured majority of Conservative policymakers. The political lesson from this example and Powell’s study is clear: if we want policymakers who are more likely to accurately represent the policy positions of the majority of the public, we are better off using proportional representation voting systems.

42

Fair Representation for All

The Hidden Problem So why has there not been more political agitation about the travesty of manufactured majorities in the United States, and more interest in the PR solution to this problem? For one thing, as mentioned earlier, these problems do not occur as often in the U.S. as they do in some other SMP countries like Great Britain. But another reason is that most Americans are kept in the dark about this problem—they are not told about it when it does occur. The U.S. media rarely report the occurrence of a manufactured majority. Indeed, they rarely even provide the public with the crucial statistic—the total vote that the parties win in local, state, and federal legislative elections—that people need to determine whether a party has won an earned or a manufactured majority. Nor are these figures readily available even if one is looking for them. To find the national party totals in elections for the U.S. House of Representatives, you would have to compile them yourself from state returns; or you could wait a year and then consult the next edition of the Statistical Abstract of the United States— a book that is rarely on anyone’s bookshelf. For these reasons, then, the antidemocratic results of SMP elections often go unseen by the American public.

Spoilers However, some Americans are becoming more aware that the singlemember plurality system can frustrate majority rule—when it does so on the district level. As noted earlier, when there are more than two candidates running in a single-member district, our election system allows a candidate with less than a majority of the vote to win office—a so-called “plurality winner.” People tend to notice this result particularly when one candidate acts as a socalled spoiler. Spoilers are third-party or independent candidates who take away enough votes from one major party candidate to ensure the victory of another candidate who would not have won otherwise. This situation strikes many people as unfair because it frustrates the will of the majority of the voters. For instance, if a Republican candidate wins 47 percent of the vote, a Democrat wins 48 percent of the vote, and an independent conservative candidate wins 5 percent of the vote, the winner is the Democrat even though most voters (52 percent) cast ballots for conservative candidates. One of the most famous spoilers in recent memory was Ralph Nader, the Green party candidate for president in 2000. In Florida, he took enough votes away from the Democratic nominee, Al Gore, to ensure that that state’s crucial

Fair Representation for All

43

Electoral College votes went to George Bush. Therefore, while most Florida voters cast ballots for leftist candidates (Gore and Nader combined for 50.4 percent of the state vote), Bush won the state and thus the presidency. As interest grows in third parties in the United States, the possibility for spoilers grows. For example, in New Mexico the Green party has experienced growing support. This has had the effect of weakening the election chances of some Democratic candidates and thus working to the advantage of Republicans. A typical example of this occurred in a special election held in 1997 to fill a U.S. House seat. A popular Green candidate, Carol Miller, won 17 percent of the vote and thus threw the election to the Republican candidate, who won with only 42 percent of the vote. In the absence of the Green candidate, the Democrats would have won what had been traditionally their seat. Sometimes it is Republican candidates that are hurt by thirdparty candidates. In the 2000 elections for the U.S. House of Representatives, it appears that at least two Republican candidates owed their defeat to the presence of Libertarian candidates on the ballot.24 But while political commentators often vilify particular candidates, like Ralph Nader, for being spoilers in elections, it is important to realize that it is the system, not the candidates, that is the real spoiler. Spoilers are created by the single-member plurality system itself. And if we want to avoid this problem, we must change this system.25 A change to proportional representation for legislative elections would eliminate spoilers. Without winnertake-all districts, spoilers simply cannot exist. Under PR rules, the majority of voters cannot be shut out of representation—indeed they almost inevitably win the majority of the seats in the district. Moreover, votes for minor-party candidates usually result in their election, and do not inadvertently aid in the election of a major party candidate. Proportional representation allows voters to support minor-party and independent candidates without fearing that they will inadvertently aid the candidates or parties that they most dislike. In SMP, the only way to avoid spoilers is to not vote for third parties and instead embrace the lesser-of-two-evils among the major parties. PR eliminates this difficult choice and allows voters to vote their conscience and to be assured that all parties will still achieve their fair share of representation.

The Losing Party Wins If plurality winners and manufactured majorities were the only anti-majoritarian and undemocratic results produced by single-member plurality elections,

44

Fair Representation for All

that would be more than enough reason to abandon this system for proportional representation. But there are actually even more disturbing results possible. Sometimes an SMP election even violates the principle of plurality rule and allows the losing party to gain a majority of legislative seats. In these cases, SMP does not give an artificial boost to the largest party but instead gives it to the second-place party. When this occurs, it results in the majority of legislative seats going to the party that came in second in the balloting—the losing party. Incidents of the second-place party winning the majority of seats in the national legislature have taken place in almost all SMP countries, including Canada, Great Britain, and New Zealand.26 Typical of these was the Canadian election of 1979. The Liberal party came in first place with 40.1 percent of the votes and won 114 seats in parliament, but the Progressive Conservative party came in second with only 35.9 percent of the votes and yet still won a majority of 136 seats. When the second-place party wins the election and control of the legislature in this way, this is an even more extreme violation of majority rule than in manufactured majorities. Not surprisingly, Arend Lijphart has called this kind of result, the “gravest democratic defect” of SMP elections, and G. Bingham Powell has described it as a “disaster” and an “unmitigated failure” of this electoral system.27 Admittedly, this problem is not a common occurrence with SMP elections—it happens much less often than manufactured majorities. Nevertheless, it is such an outrageous result that it tends to prompt many people to question whether a system that produces such blatantly undemocratic results is acceptable. One could draw an analogy to the public response to the victory of George W. Bush over Albert Gore in the 2000 U.S. presidential elections. When the distortions built into the Electoral College system allowed the person who placed second in the popular vote to win the presidency, many people called for the abolition of that electoral system. Polls at the time showed that clear majority preferred a change to electing the president by popular vote.28 Similar public reactions have been found in countries where the second-place party won the right to rule. In 1974, in Great Britain, the Labour party won 37.2 percent of the vote and garnered 301 seats, the Conservatives 38.2 percent and 297 seats. After that election a London Times survey showed that 70 percent of those polled supported a change to a new system of voting in which “the number of seats a party wins reflects the number of votes it gets in the election.”29 Public outrage also accompanied the occurrence of this problem in New Zealand. That country was unlucky enough to have had two of these disas-

Fair Representation for All

45

trous elections in a row. In 1978 the Labour party won 40.4 percent of the vote, but was allocated only 40 seats in the 92 seat parliament. The National party came in second with 39.8 percent of the vote, but won 51 seats—a clear majority. This problem recurred in 1981 when the National party again garnered second place with 38.8 percent to the Labour party’s 40 percent and yet again received a majority of seats (47) in the parliament. Naturally, these outcomes produced a great deal of public resentment, and a royal commission was convened to study the electoral system. The commission eventually recommended the replacement of the SMP system with a form of proportional representation.30 The commission’s report led to a national referendum on the electoral system in 1993 in which the public voted to switch from SMP elections to the mixed-member form of proportional representation. Instances of this electoral malfunction can be found in the United States as well—though again they almost always go unreported by the media. In the 1982 Indiana state elections the Democratic candidates for the Indiana House received 51.9 percent of the votes, but were given a minority—43 percent—of the seats. The Democrats were understandably upset with this unfair outcome and challenged the results in court, but lost that fight as well. Notice, that unlike manufactured majorities, this problem does not require the presence of more than two parties on the ballot. Even if there are only the two major parties vying for office, the SMP process can give the second-place party the majority of legislative seats. Many examples of this electoral failure can also be found in elections for state delegations to the U.S. House of Representatives. To mention but a few examples: in the 1994 elections for the U.S. House of Representatives in California, the Democrats won 52 percent of the seats even though they came in second place with 47 percent of the statewide vote. That same year, the Democrats pulled off the same political coup in Louisiana and Michigan as well. In the 1996 House races in Washington State, the Republicans trailed the Democrats with only 47 percent of the vote, but won 67 percent of the seats. In 1998, Florida Republicans came in second in the vote to the Democrats, but won 65 percent of the U.S. House seats. In Pennsylvania and Virginia that year it was the Democrats who benefited from this unfair representation. And in 2000, Democrats in Texas, Wisconsin, and Arkansas all placed second to the Republicans in terms of votes, and yet they won the majority of their states’ U.S. House seats. That same year in New Mexico, the Republican’s second place finish in the

46

Fair Representation for All

vote gave them a majority of that state’s House seats. The main point, of course, is that in all these cases the majority voters in those states were forced to watch helplessly as most of their representatives in the U.S. House pursued policies at odds with their wishes. This is not the way democracy is supposed to work.

SMP Fails on Its Own Terms Because single-member plurality elections often violate the principle of majority rule, this system fails on its own terms. Much of the rhetoric supporting the SMP system repeatedly refers to the importance and primacy of majority rule. For example, many defenders of SMP elections will acknowledge that winner-take-all elections do not necessarily provide representation for everyone in the district, but they argue that this system is still desirable because it does provide the most important thing in elections: representation for the majority of the district. As G. Bingham Powell has observed, SMP elections are part of a majoritarian vision of politics. In this vision, representation of minorities is less important than assuring that political power rests in the hands of the majority of the voters. As Powell explains it: “The majoritarian vision seeks connections between voter choices and concentrated political power, so that voters’ choices will provide effective mandates from the citizen majority and so that the voting majority can hold incumbents accountable for their performance.”31 But as this chapter has shown, single-member plurality elections often fail to produce majority rule. Problems like manufactured majorities and the second-place party winning the majority of seats mean that this system can actually violate the principle of majority rule that it values so much. Ironically, it is proportional representation that does a much better job of assuring majority mandates and majority rule. It also does a better job of ensuring that incumbents are held accountable to the majority of voters—another point mentioned by Powell. As seen earlier, SMP allowed the incumbent Conservatives to be continually reelected for almost two decades in Great Britain, even though the majority of British citizens wanted them out and voted against them in each election. So even if we judge voting systems solely on the central political value of the majoritarian vision of politics, proportional representation systems come out as superior to single-member plurality elections.

Fair Representation for All

47

Lack of Responsiveness in SMP Elections An additional problem with single-member plurality elections is that not only do they often give a distorted picture of the political views of the public at any given time; they also tend to poorly represent any changes in those views over time. A good electoral system should accurately translate any changes in party preferences among the public to changes in the share of party seats in the legislature. In other words, if a party loses favor among the public, it should also lose seats in the legislatures to a corresponding degree. This is sometimes called the “responsiveness” of an electoral system. A good democratic voting system is responsive to political changes. But the responsiveness of SMP systems is typically poor. They often either under-respond or over-respond to changes in voter preferences. This is most evident on the district level. For instance, assume that in one election 75 percent of the voters cast ballots for the Republican candidate and 25 percent for the Democrat. But in the next election, there is large change in voter sentiment with 20 percent of the voters switching from the Republicans to the Democrats. The Republican candidate would still get 55 percent of the vote—and there would be no change in representation. This is underresponsiveness, where a large change in voter sentiment produces little or no change in representation. But single-member districts can also be hugely over-responsive as well, where a very small change in voter preferences translates into a very large change in representation. In this case, assume that the Democrats win in the first election by 51 percent to 49 percent. In the next election, only 2 percent of the voters change their mind and vote Republican, but this assures a Republican victory. This 2 percent change in voters’ party preference produces a 100 percent change in representation. This kind of inappropriate responsiveness is an inevitable result of winnertake-all districts. Not surprisingly, the poor responsiveness of single-member plurality elections on the district level also can produce unresponsiveness on the legislative level—where there is a poor correspondence between changes in voter sentiments and changes in number of party seats. A classic example of overresponsiveness can be found in the results of elections for the U.S. House of Representatives in Washington State, where representation of the two parties has often seesawed violently despite relatively small changes in voter sentiment. For instance, between 1992 and 1994 there was a small shift in party preferences with the Democrats losing about 7 percent of the vote

48

Fair Representation for All

(from 56 to 49 percent). But this produced a sea change in representation with the Democrats’ share of House seats plunging from 89 to 22 percent. By 2000, the Democrats had won back 4 percent of that vote (from 49 to 53 percent), and moved up from 22 to 67 percent of the House seats—another wild swing in representation. In contrast to SMP elections, proportional representation elections are very responsive. Indeed, PR systems are designed to accurately translate changes in voters’ views to changes in seats. This characteristic is a product of the large, multimember districts used in PR. In a ten-member district, a 20 percent change in voter sentiment will usually result in a shift of two seats—not four seats or zero seats as can happen in SMP systems. The analogy of light switches serves to illustrate the differences between the responsiveness of SMP and PR systems. Single-member plurality systems are like conventional light switches, where the initial movement of the switch from bottom to top at first produces no change at all in the light, but when the midpoint is passed, suddenly there is a dramatic change from 0 percent light to 100 percent light. In contrast, PR systems are like dimmer switches where the movement of the switch is smoothly translated in corresponding changes in the strength of the light. In this same way, PR elections are superior in assuring accurate representation for citizens as their views change over time.

A Capricious Election System Another important value in electoral systems is reliability. We want a system that consistently provides for fair representation and majority rule. But SMP elections are disturbingly inconsistent. They sometimes provide fair representation for parties, but they more often produce distorted representation. Which party will rule is often determined not simply by how many votes party candidates receive, but by how many of those votes happen to be wasted. And it is the concentration and distribution of party supporters in various districts that determines which parties end up being helped or hurt by the problem of wasted votes. The full extent of this problem is shown clearly in table 1.4. In this series of hypothetical examples, a state is divided into ten districts each with 100,000 voters. Of the one million state voters, 58 percent (580,000) are Democratic and 42 percent (420,000) are Republican. The fair result would be six seats for Democrats and four for Republicans. But as the table demonstrates, how the

table 1.4 The Capriciousness of Seat Allocation under Plurality Rules (10 districts, 1,000,000 voters) Example A

Example B

Example C

Example D

Example E

Election Districts 1

58,000 – 42,000

65,000 – 35,000

71,000 – 29,000

77,500 – 22,500

94,000 – 6,000

2

58,000 – 42,000

65,000 – 35,000

71,000 – 29,000

77,500 – 22,500

94,000 – 6,000

3

58,000 – 42,000

65,000 – 35,000

71,000 – 29,000

77,500 – 22,500

49,000 – 51,000

4

58,000 – 42,000

65,000 – 35,000

71,000 – 29,000

77,500 – 22,500

49,000 – 51,000

5

58,000 – 42,000

65,000 – 35,000

71,000 – 29,000

45,000 – 55,000

49,000 – 51,000

6

58,000 – 42,000

65,000 – 35,000

45,000 – 55,000

45,000 – 55,000

49,000 – 51,000

7

58,000 – 42,000

47,500 – 52,500

45,000 – 55,000

45,000 – 55,000

49,000 – 51,000

8

58,000 – 42,000

47,500 – 52,500

45,000 – 55,000

45,000 – 55,000

49,000 – 51,000

9

58,000 – 42,000

47,500 – 52,500

45,000 – 55,000

45,000 – 55,000

49,000 – 51,000

10

58,000 – 42,000

47,500 – 52,500

45,000 – 55,000

45,000 – 55,000

49,000 – 51,000

Dem. Rep. 580,000 420,000

Dem. Rep. 580,000 420,000

Dem. Rep. 580,000 420,000

Dem Rep. 580,000 420,000

Dem Rep. 580,000 420,000

Vote Totals Allocation of Seats Between Parties

10

0

6

4

5

5

6

4

2

8

50

Fair Representation for All

seats are divided in our plurality, winner-take-all system depends entirely on how the party members are distributed among the districts and which party wastes the most votes. As example A demonstrates, if the party members are divided equally among the districts (58,000 and 42,000 in each), the Republicans are forced to waste all of their votes and the Democrats win all the seats. But as the other examples demonstrate, practically any other allocation of seats in the legislature is possible, depending entirely on how party supporters happen to be distributed throughout the districts. In examples C, D, and E, the Democrats are forced to waste many of their votes and fail to receive the majority they deserve. In the last two examples the Republicans rule the legislature despite getting only a minority of the votes. In other words, in a single-member plurality system, who governs can often be determined by where the voters happen to live—what one political commentator calls the “accident of sleeping place.”32 It hardly seems sensible for a matter as important as who rules our country to be determined by something as capricious as where people happen to live and vote. Of course what makes this absurd situation even more unacceptable is that it is completely unnecessary. Using proportional representation in the United States would eliminate this electoral capriciousness and ensure that all parties are given their fair share of seats and political power. In the example in table 1.4, the use of a PR voting system, with all the candidates elected in one tenmember district, would automatically give the Democrats six seats and the Republicans four, accurately reflecting their 58 percent and 42 percent shares of the vote.

PR: The More Democratic Alternative Centuries ago, when winner-take-all voting was used in the first elections in many European countries, it was an important step away from monarchy and rule by an aristocracy and toward a more democratic political system. But today SMP elections have become an obstacle to more democracy. If we want democracy for all in the United States, we need fair representation for all. But that is exactly what winner-take-all elections cannot provide. As this chapter has shown, the evidence is overwhelming that proportional representation is a more representative and a more democratic electoral system. Proportional representation elections are simply much better at doing what democratic elections are supposed to do. PR is better at ensuring that the

Fair Representation for All

51

maximum number of citizens are represented, that both majorities and minorities have a say in government, that all parties have their fair share of seats in legislatures, that seats change in response to changes in voters views, that the majority rules, and that policymakers faithfully reflect the political views of the majority of the electorate. There is absolutely no reason why Americans should continue to put up with an outmoded voting system that routinely wastes large number of votes, denies fair representation to political groups, and even frustrates majority rule. Advances in voting systems, such as proportional representation, have made such problems easily avoidable. The residents of most other advanced Western countries enjoy the advantages of this fairer and more democratic electoral system. It is time that American voters had the benefits of this superior electoral system as well.

2

Solving the Redistricting Problem

Most Americans believe that who wins political races is decided on election day by the voters. But in a single-member district electoral system that is frequently not true. Who wins is often determined before voters even go to the polls—sometimes many years before. The outcome is decided by those who draw the district lines. If they decide to create a district that is 70 percent Republican, there is little chance the Democratic candidate will win. And Republican candidates will usually lose if a district is drawn so that it is predominantly Democratic. Voters go to the polls confident in the illusion that they control the fate of the candidates. But in reality they are often only participating in the last act of political play whose ending has already been written. In a single-member plurality system, the power of the vote pales in comparison to the power to draw district lines. The districting process not only can determine which candidates will win in specific districts, but also which party ultimately controls our local, state, and federal legislatures. In a very real way, then, the political manipulation of district lines devalues the vote and undermines the democratic process. Describing the redistricting process in unusually candid terms, North Carolina State Senator Mark McDaniel has said: “We are in the business of rigging elections.”1 This process of drawing district lines to rig elections is known as “gerrymandering,” and it is one of the great curses of the singlemember plurality electoral system. The United States could rid itself of this unscrupulous and undemocratic practice; but eliminating gerrymandering

Solving the Redistricting Problem

53

in all of its forms can only be accomplished by abandoning the singlemember districts that make it possible and by adopting proportional representation elections.

Gerrymandering After each census, a process of reapportionment and redistricting begins in the states. The federal courts require that every district contain approximately the same number of voters, so shifts in population require that U.S. House seats be reapportioned between the states. States that lose or gain seats must necessarily redraw their congressional districts to take this into account. Even in states that do not gain or lose Congressional seats, shifts of populations within regions of the state can also require the redrawing of districts. In this redistricting process, gerrymandering—the manipulation of district lines for political advantage—has become the political weapon of choice. The two main purposes of gerrymandering are to protect the seats of incumbents and to allow the dominant party in a state to win more seats that it deserves. Gerrymandering in nothing new. It is a form of political fraud that has plagued American elections from the early days of the Republic. Patrick Henry is said to have led an effort to gerrymander the home district of James Madison in order to prevent his election to Congress. The term gerrymander itself did not come into use until early in the nineteenth century. It was coined in 1812 when the Massachusetts legislature carved out an oddly shaped congressional district in Essex County. When the newspaper artist Gilbert Stuart saw the misshapen district, he drew it with a head, wings, and claws and exclaimed: “That will do for a salamander!” His editor, Benjamin Russell, replied, “Better say a Gerrymander”—after Elbridge Gerry, who was governor of Massachusetts at the time.2 Unfortunately, gerrymandering continues unabated today, largely because our single-member plurality election system makes it so temptingly easy. In its most unscrupulous and undemocratic form, gerrymandering is used by the dominant party in the state legislature to ensure that it wins more seats that it deserves based on its proportion of the vote. This is called “partisan” gerrymandering. As seen in the previous chapter, the large numbers of wasted votes that are inevitable in SMP systems often produce distortions in representation that benefit some parties and hurt others. In partisan gerrymandering, the dominant party takes advantage of this flaw in our electoral system by

54

Solving the Redistricting Problem

intentionally manipulating the district lines to force the rival party to waste many of its votes and thus be underrepresented. The two main strategies that are used in partisan gerrymandering are called “cracking” and “packing.” In cracking, district lines are redrawn to ensure that the opposing party’s pockets of voting strength are divided up so that they become permanent minorities in most districts. For example, Democrats may eliminate a previous district where Republicans had 60 percent of the vote, and divide that party’s supporters into two new districts, where they make up only 30 percent of the voters in each—thus making it impossible for them to win a seat. The packing strategy is used when the opposing party’s strength is too large to be completely divided. Instead, lines are redrawn to concentrate most of that party’s voters in one district so that many of its votes will be wasted in an overly large majority. If, for example, the Democrats put most of the Republicans in a region into one district in which they make up 85 percent of the voters, they will win only one seat; whereas if those voters were spread around several districts, the Republicans might have won more seats. The two techniques are illustrated in figures 2.1 and 2.2, both of which represent the same central city with surrounding suburbs. (In these figures D ⫽ 1,000 Democrats and R ⫽ 1,000 Republicans; each grouping of six letters represents a neighborhood.) As is often the case, the central city is predominantly Democratic, while the suburbs are predominantly Republican. Given that the Democratic and Republican parties have approximately even numbers of supporters (52,000 and 44,000), it would seem fair that they each receive two of the four seats to be filled. However, gerrymandering can produce quite different results. In figure 2.1 the Democratic party draws the district lines in a way that fragments the Republican vote. Each district includes part of the Democrat-dominated inner city, and the Republicans become minorities in all of them. As a result, the Democrats take all four seats. All 44,000 Republican votes are wasted—producing no representation. In figure 2.2 the Republicans draw the lines and create a central city district into which they pack the Democratic vote, where it wins one seat by a lopsided margin, thus wasting much of its voting strength. This allows the Republicans to win the other three seats. As these figures illustrate, partisan gerrymandering can be a very powerful political tool. Which party wins the most seats may not be determined by how many votes that party gets, but instead by how the district lines are drawn. A real life example of this can be found in the results of districting in New York

Solving the Redistricting Problem

figure 2.1 Democratic Gerrymander

55

figure 2.2 Republican Gerrymander

State. As shown by presidential votes, about 60 percent of the voters in New York State consistently vote Democratic. This is reflected in the State Assembly where Democrats routinely win about 64 percent of the seats. But the election results are dramatically different in the State Senate, where the Republicans consistently dominate with a large majority of about 57 percent of the seats.3 Same voters, but very different results. The obvious suspect is the districting process. The districts for the Senate races are different from those

56

Solving the Redistricting Problem

for the Assembly races and they are largely drawn by the Republican dominated Senate, which makes every effort to favor GOP candidates. This seems to be a clear example of where districting, not the voters, decides who wins control of legislative bodies.4 The intended result of partisan gerrymandering is misrepresentation of the public. The party controlling the redistricting artificially inflates the number of seats it gets. In some cases, this distortion of representation is so gross that it allows a party that gets a minority of the votes to win a majority of the seats—a clear violation of the basic principles of democracy and majority rule. In the 2000 elections for the U.S. House of Representatives, for example, gerrymandering in Texas allowed the Democrats to win 57 percent of those seats, even though they received only 47 percent of the vote statewide. It is important to note that the shape of districts often tells us little about whether they are gerrymandered or not. It is often thought that gerrymanders can be identified by the strange shapes of the districts. And it is true that the political mapmakers must sometimes severely contort the sides of districts to accomplish their purposes. But this is not always the case. The districts in the figures above have normal shapes, and they are quite compact and reasonable. So the shape of a district does not necessarily tell us whether it has been gerrymandered or not. Examples of partisan gerrymandering abound in the United States. Indeed, some states, such as California, Texas, and Indiana, have become infamous for their blatant gerrymanders. After one reapportionment, Republican legislators in Indiana did everything they could to use district lines to their advantage. Over the protests of the Democratic minority, the Republicans adopted a plan that took two districts with Democratic incumbents and packed them with additional Democratic voters so that GOP candidates in other districts would have less opposition. The plan also relocated three Democratic incumbents into the same district, and separated another Democrat from his hometown, placing him in a traditionally Republican area. In case anyone had doubts about the political intentions underlying this plan, one of its principal authors, Republican state Senator Charles Bosma, told reporters that the Democrats “are going to have to face the political reality that we are going to do everything we can to hurt them within the restraints of the court rulings.”5 These political shenanigans were much in evidence again in the redistricting efforts following the 2000 census. In Kentucky, Albert Robinson, the

Solving the Redistricting Problem

57

chairman of the committee charged with redrawing their Senate’s district map vowed revenge on the Democrats now that the Republicans were in control of that body. “I’ve been a victim of partisanship,” he complained, pointing out that the Democrats “tried to destroy the Republican party” in previous redistricting efforts. He cheerfully admitted that the Republicans eagerness to use their advantage “goes without saying. Any party that’s in control, charity begins at home.”6 The Democrats are hardly any better when they are in control. In California, the Democratic controlled legislature has a history of creating district lines that meander around California, throwing Republicans incumbents together, and carving out new districts that favor Democratic candidates. In one famous instance of excessive creativity, they designed a district for a Democratic incumbent that had 385 sides.7 In the past, the electoral power grab has been so effective that Democrats have been able to win a majority of the U.S. House seats from that state, even though they sometimes have earned less than a majority of the votes.

Redistricting: Politics at Its Worst The legislative battles over redistricting can be incredibly passionate. The political combat has been variously described in the press as “bare-knuckle brawling,” “cut-throat politics,” and “political street fights.” Speaking of the 2001 redistricting process in Pennsylvania, Professor Terry Madonna said: “I’ve studied reapportionment and redistricting for 30 years, and I’ve never seen an atmosphere so poisonous as it is today in Pennsylvania.”8 There are several reasons for this political intensity. First, the very jobs of the state legislators are at stake. Redrawing boundaries can make them vulnerable to defeat and they naturally will fight desperately to protect their careers. It’s a matter of political survival. Second, redistricting has profound and wideranging political impacts on who has political power in the United States. These decisions affect not only who controls state governments, but the federal government as well. Officials in the U.S. House of Representatives are very aware of the implications of redistricting. After the last census, U.S. House Majority Whip Thomas DeLay gave a speech to the Texas state legislature urging them to use redistricting to dramatically increase the number of their Republican House seats. “I hope each of you will also remember that your vote will not simply count for Texas,” he said. “The decisions will

58

Solving the Redistricting Problem

carry implications for our entire nation.”9 The political effects of redistricting also extend over time. Once districts are drawn, the political landscape is set until the next census and reapportionment. As political scientist Thad Beyle, has observed, “Whoever sets up the legislative districts for the 2002 election basically sets up the elections of the whole decade.”10 Given the enormous political stakes, the political struggles over redistricting now start many years before the actual process of redrawing the lines takes place. For example, the two major political parties have poured increasing amounts of resources into state legislative races, hoping to eventually gain the upper hand during the redistricting process. Speaking of the 1998 elections, one Republican consultant said: “Long term, redistricting may be the most important issue of this campaign.”11 In the key prereapportionment election in 2000, the two parties spent a record-breaking sum—over a billion dollars—on state contests. In recent years, the Republicans have gotten the best return on this investment. Between 1990 and 2000 they were able to increase to 17 the number of state legislatures they controlled, thus bringing them into parity with the Democrats (the other state legislatures were split between the parties). One reason why redistricting political battles have become worse in recent years is the ease with which more groups can now join in the battles. In the past, redistricting was done by a few groups of experts and politicians working largely in secret. They worked over maps by hand and revealed their final plans only after the process was over. These few players were the only ones with the information and expertise to accomplish this difficult task. But with the advent of the personal computer, internet access to demographic data from the census, and new user-friendly redistricting software, nearly any organized group can develop its own redistricting plan. As Bruce Cain, director of the Institute of Governmental Studies as the University of California at Berkeley has observed: “There used to be two or three groups that were able to draw their own (redistricting) lines. Now there could be hundreds.”12 Redistricting battles have not only become more complicated, but now they usually do not end with the passage of a final plan by the state legislature. Increasingly, redistricting ends up in the courts, as various groups challenge the plans offered by the legislatures. This means that the redistricting mess can drag on for years and states can spend large sums of money in court fights. In Florida, Alabama, Louisiana, and North Carolina, litigation over the districts drawn in response to the 1990 census was still going on in 1999.

Solving the Redistricting Problem

59

In Chicago, in the 1990s, the city government was forced to pay more than $8 million in legal fees for litigation resulting from its redistricting process. Increasingly, race has become an issue in these court challenges. In the 1980s, the U.S. Department of Justice began to pressure states to create more black and Hispanic majority districts (so called majority-minority districts), so that minorities could begin to win their fair share of legislative seats. But in the 1990s, white voters challenged those districts and the Supreme Court began to declare some of them unconstitutional. (A more detailed analysis of the issue of race and redistricting can be found in chapter 6.) But the Court has been vague about when such districts are acceptable and this has only invited more legal challenges by the various sides in these disputes. Looking at this complicated legal situation, Rep. Ken Bentsen, chairman of the House Democratic Caucus’ watchdog group on redistricting warned in 2001: “Litigation can be expected on [redistricting] plans in virtually every state.”13 To sum up, redistricting is a political disaster—an enormous political train wreck that occurs every ten years. It is a process full of partisan rancor, naked power grabs, unfair representation, and millions of dollars in legal costs. It is an example of American politics at its worst—and all made possible by our single-member district electoral system.

Gerrymandering as Incumbent Protection If there is one bright spot in this redistricting mess, it is that brazen partisan gerrymandering is sometimes difficult to accomplish. It works best in states where a single party controls both houses of the state government and the governorship. When control over state government is split between the major parties, this gives each party a veto over any redistricting scheme and makes it difficult for one party to grossly take advantage of another. Often, when one party can’t dominate the process, the end result is no large change of seats from one party to another. However, each party gets something out of the process: the protection of its incumbents. This is sometimes called a “sweetheart gerrymander”—where the two major parties make sweetheart deals intended to create safe districts for both Democratic and Republican incumbents. Safe districts ensure incumbents’ reelection by packing their party supporters into those areas. So in these situations, the redrawing of district lines becomes mainly an exercise in incumbent protection, with politicians of all parties

60

Solving the Redistricting Problem

scrambling to make sure that their districts are not eliminated or that any new district lines will encompass as many of their supporters as possible. Legislators from the minor party in a state often tend not to object to such sweetheart gerrymanders even if these arrangements result in some underrepresentation of their party. While these legislators do care about how many seats are won by their party, their first priority in redistricting is to preserve their own seats. Focusing on these bipartisan arrangements, defenders of single-member plurality elections sometimes argue that the problem of gerrymandering has been exaggerated. They maintain that it is not really a serious political problem because often it does not result in gross distortions in representation. But this sanguine view mistakenly assumes that incumbent-protection gerrymanders are politically innocuous and it overlooks the serious political damage produced by this approach to redistricting. The main problem is that these gerrymanders create districts that are uncompetitive—districts in which voters are denied any meaningful choice. When legislators manipulate the district lines to create safe seats for Republican and Democratic incumbents, any real competition in these elections disappears. Commenting on the latest redistricting effort in California that protected incumbents of both parties, one Republican consultant concluded: “This new plan basically does away with the need for elections. These seats shouldn’t change (parties) over the decade.”14 While such arrangements clearly benefit incumbents, they mean that voters are faced with uninspiring, no-contest elections. Instead of a healthy competition between two parties, we end up with a series of one-party fiefdoms in which a single party rules and voters feel helpless to change it. The New York Times once editorialized that as a result of gerrymandering, New Yorkers had “no more voting options than North Koreans have.”15 The tendency of legislatures to create safe districts is well known and well documented. After every election, the Center for Voting and Democracy analyzes the competitiveness of races for the U.S. House, and they inevitably find that an enormous number of the seats are no-contest elections. In 2000, for example, 78 percent of the U.S. House seats were won by landslides—defined as victory by a margin of greater than 20 percent. Only 9 percent of House races were defined as competitive—having been won by a margin of less than 10 percent.16 Naturally, most of the landslide wins took place in districts where one party had a lopsided advantage in voters. The study identified “two major culprits for our non-competitive elections: a winner-takeall electoral system that turns natural majorities into no-choice elections and typically partisan methods of redistricting that allow legislators to craft dis-

Solving the Redistricting Problem

61

tricts “safe” from competition—legislators literally choose their constituents before their constituents choose them.”17 Another one of the Center’s studies, Monopoly Politics, concluded that the partisan make-up of districts is even more important than disparities in campaign funds in explaining who wins elections. “Campaign contributors respond to safe seats much more than cause them. Their contributions generally are a way to buy influence, not to buy elections. Party control of districts and winners’ victory margins correlate far more strongly to the partisan nature of the district than to how much money is spent.”18

Uncontested Contests In safe districts, candidates of the minority party are often little more than sacrificial lambs being led to the slaughter. But it gets worse. These safe districts are often so one-sided that the disadvantaged party does not even bother to put up a candidate. There is no contest at all. The number of uncontested U.S. House seats varies considerably from state to state. Florida and Texas are two of the worst. In the 2000 U.S. House elections in Florida, 43 percent of the races were uncontested. That year in Texas, 30 percent of their House races were uncontested. The existence of so many uncontested races is a political disgrace, but it is also a perfectly understandable result in a winner-take-all system. Why should the smaller party bother to compete in districts dominated by the larger party? It would be a waste of valuable political resources. As the former head of the Republican campaign effort in the House, Rep. John Linder of Georgia, explained: “The GOP did not recruit candidates in districts where Democrats are shoo-ins for re-election.” And according to his counterpart on the Democratic side, Rep. Martin Frost of Texas: “Both parties, I believe, had made a decision to concentrate on really competitive districts.”19 So in an SMP system, it is a rational political strategy for a party to not contest seats in many districts. On the level of state legislative contests, the problem of unopposed candidates has reached epidemic levels. In races for state legislatures in 2000, either the Democrats or the Republicans refused to nominate a candidate in 40.6 percent of the districts.20 In 1998 that figure was 41.1 percent. Thus in a large number of cases, the creation of safe seats for incumbents means that voters literally have no choice on election day. These are not so much races as prolonged victory laps by the preordained winners.

62

Solving the Redistricting Problem

These safe districts created by gerrymandering not only deny voters a meaningful choice, they contribute to other political problems as well. For one thing, safe districts clearly encourage political apathy and low voter participation. Why bother to participate if the outcome has been predetermined by how the district lines have been drawn. As one San Francisco resident complained: “Vote? Why vote? I know who’s going to win, everybody knows who’s going to win. [Nancy] Pelosi always wins, with 80 percent of the vote. Nobody else has a chance.”21 Those who support the party that always loses are not the only ones discouraged from voting in safe districts. Even those who support the dominant party have little reason to vote because they know that their candidate will win anyway.

Lack of Responsiveness Another major political problem of uncompetitive districts is that they make the political system much less responsive to the public—a problem that was touched on in chapter 1. One expert on redistricting, David Wells, explains this problem in the following way: Gerrymandering invariably inflates the number of safe districts. Barring a successful primary challenge, the individual incumbent is virtually assured of continued reelection for as long as he or she cares to hold the seat. This has the effect of insulating the legislative body against the consequences of changing sentiments and circumstances, for gerrymandering has provided the individual legislator, the legislative leadership, and the legislature as a whole with rather strong guarantees of continued office and power. The political . . . composition of the legislature has been effectively frozen for a decade, and changes are possible only within a limited, narrow range. The representation system, because it has been made less politically sensitive and therefore less responsive, has thus been rendered less able to perform its most fundamental task—the translation of public sentiment into public policy as accurately as possible.22 A hypothetical example may help to illustrate this point. Assume that gerrymandering creates a safe district packed with Democratic voters and the Democratic candidate is elected with 70 percent of the vote to 30 percent

Solving the Redistricting Problem

63

for the Republican. Before the next election, however, voter sentiment turns strongly toward the Republican party—with 16 percent of the former Democrats now favoring the GOP. Despite the dramatic change in political opinion, the Democratic candidate will likely continue to represent this district—and be reelected this time by 54 percent to 46 percent. So despite a large change in political preferences by the public, no resulting change of representation takes place in this safe district. The protection of legislators from both parties in this way severely restricts the responsiveness of the political system. Contrast this result with the effect of changing voter sentiments in the multimember districts used in proportional representation. Assuming that ten representatives are elected in a district, a swing of 16 percent from one party to another would result in change of at least one and possibly two seats in favor of the party that is growing in popularity. In this sense, PR elections are much more sensitive to changes in voter preferences, and these changes are much more likely to be reflected in the makeup of legislatures. In PR systems, the total number of a party’s seats in a district cannot be made safe. Parties always can lose seats in proportion to their drop in popularity— which may not be true in single-member plurality systems. To sum up, there are two different kinds of political evils created by gerrymandering. In its most egregious form, gerrymandering is used by parties to steal seats and political power away from one another. Thus it violates basic notions of fair representation. But it also acts as an incumbent protection process, where instead of voters choosing their politicians, politicians actually choose their voters and thus ensure their reelection. Clearly redistricting and gerrymandering are serious threats to fair and competitive elections in the United States. But how can this problem be solved? We turn to that question next.

The Difficulty of Reform within the SMP System Given the pernicious and clearly undemocratic results of gerrymandering, its elimination is obviously high on the list of those who believe in fair elections. But reform in this area has been slow in coming, and the problem continues to plague our political system. Attempts by some reformers to challenge this practice in court have been consistently unsuccessful.23 While the Supreme Court has said that redistricting plans can be challenged in federal

64

Solving the Redistricting Problem

court, they have consistently refused to overturn them. Strangely, while the Supreme Court has been eager to label as unconstitutional the practice of “racial gerrymandering”—the manipulation of district lines to increase the representation of minorities—it has been reluctant to criticize partisan gerrymandering even in its most blatant and undemocratic forms. Reformers have had some success in transferring the redistricting process out of the hands of state legislatures and into bipartisan commissions. Hawaii, Idaho, New Jersey, and Washington have given the task of redistricting to commissions consisting of equal numbers of Democrats and Republicans, with an additional member approved by both parties. But this approach has proved far from ideal. Members of these commissions have often proved to be as rabid partisans as the legislators themselves. And some critics have pointed out that, unlike legislators, these political mapmakers are not directly answerable to the public if they draw questionable district lines. Tim Storey, a redistricting expert for the National Conference of State Legislatures, has observed that the redistricting commissions “concentrate a fair amount of power in small unaccountable groups, rather then in the legislatures, where members have to stand for elections.”24 But the most serious problem with bipartisan commissions is the possibility of sweetheart gerrymanders. With equal numbers of Democrats and Republicans, there is a good possibility of deals being struck that serve to protect the incumbents of both parties. The result would inevitably be the kinds of uncompetitive districts that were discussed earlier. The most radical approach to the problem of redistricting has been taken by Iowa, which has empowered civil servants to draw the district maps, allowing them to take into account population figures only. Some reformers, including Common Cause, have backed similar apolitical approaches that would require the drawing of districts using only politically neutral criteria, such as avoiding splitting up counties and cities, ensuring compact districts with the shortest boundary lines possible, and minimizing population differences between districts.25 (Some have even suggested leaving redistricting to computers programmed to use only nonpolitical criteria.) But while this approach would eliminate the most obvious examples of packing and cracking and thus would be a step forward, a number of problems remain. First, most legislatures are unlikely to agree to give up their redistricting power entirely to such bodies. Second, and more importantly, while this approach would clearly be an improvement, it would not succeed in abolishing gerrymandering altogether. By eliminating political discretion from the redistricting

Solving the Redistricting Problem

65

process, this apolitical approach would eliminate the most severe cases of intentional gerrymandering, but it would not eliminate the problem of unintentional gerrymandering. Unintentional gerrymandering is less well known than the deliberate variety, but it can also pose very serious political problems. Even districting that is done with apolitical criteria will likely produce many of the symptoms of gerrymandering—such as the creation of many uncompetitive districts where one party dominates. The reason for this is simple. Intentional gerrymandering is made possible by the geographic concentrations of Republicans and Democrats—concentrations that can be split and combined to provide advantages for a particular candidate or party. Apolitical districting does not change that geographical fact. Thus many districts created by neutral criteria will inevitably have concentrations of partisan voters that work in one party’s favor. Take, for example, the situation illustrated earlier in figure 2.2. In an urban area reasonable and apolitical district lines could be drawn that respect the integrity of the boundaries between cities and suburbs. But in creating a central city district, Democratic voters are unintentionally packed into that district, forcing them to waste much of their voting power with excessively large majorities. The unintentional result is the creation of districts that are uncompetitive and that unfairly distort the representation of the parties—the hallmarks of gerrymandering. As such examples illustrate, just because apolitical districting ignores geographic partisan concentrations does not mean that these concentrations no longer have any political effects. The concentrations remain and create unintentional gerrymanders. In short, even when mapmakers use apolitical criteria to guide their work, they are bound to produce political effects. This is exactly what has happened in Britain, whose independent redistricting commissions have sometimes been offered as models for what we need here in the United States. A number of studies have shown that although these commissions have used neutral criteria to draw district boundary lines, they have often created unintentional gerrymanders that biased the results of national elections.26 One of Britain’s leading political geographers, R. J. Johnston, found that the “review of recent redistricting in Britain, undertaken by independent commissioners within the constraints of apparently neutral rules, cast doubt as to the likelihood that fair redistricting would be achieved in the U.S.A. or indeed anywhere that the plurality single-member constituency electoral system is used.” He concluded that “the concept of non-partisan cartography is a myth.”27

66

Solving the Redistricting Problem

This is not to say that apolitical districting with its unintentional gerrymanders may not be better than the flagrant intentional gerrymandering that occurs now. But it does mean that we should not delude ourselves into thinking that this approach will eliminate the problem of gerrymandering. Indeed it cannot, for it does not address the underlying cause of the problem: the single-member plurality system itself. Gerrymandering is a necessary, built-in characteristic of any plurality voting system that uses singlemember districts. As long as we draw district lines where only one candidate wins and where large numbers of wasted votes are unavoidable, some gerrymandering is inevitable, if only unintentionally. As Robert Dixon, a longtime student of redistricting in the United States, concluded: “To be brutally frank, whether or not there is a gerrymander in design, there normally will be some gerrymandering in result as a concomitant of all district systems of legislative elections . . . . In a functional sense it thus may be said that districting is gerrymandering.”28

PR: The Real Solution The only sure way to eliminate gerrymandering—both intentional and unintentional—from American elections is to abandon single-member plurality arrangements and adopt proportional representation. Indeed, the whole purpose of PR is to minimize wasted votes and ensure that the parties are represented in proportion to the votes they receive. This eliminates the possibilities of unfair representation produced by gerrymandering. The key to eliminating partisan gerrymandering is the large multimember districts used in PR systems. As numerous studies have shown, as long as a PR system has at least five seats in every district, it is effectively immune from gerrymandering.29 These districts largely eliminate the wasted votes that make gerrymandering possible. In such districts, even small political minorities do not waste their votes and are able to elect their fair share of representatives. Thus, under PR arrangements, where voters live or how district lines are drawn makes no difference—fair representation will result. Imagine, for example, that we have a region in a state that is 60 percent Republican and 40 percent Democratic and that it must be divided into two 10-member PR election districts. No matter how the district lines are drawn and no matter how party voters are distributed between the districts, each party will be able to elect its fair share of representatives. If all the Demo-

Solving the Redistricting Problem

67

crats are packed into one district, they will constitute 80 percent of the voters there and elect eight of the ten representatives in that district and none in the other—40 percent of the total seats. If the Democratic voters are fragmented and make up 40 percent minorities in the each of the two districts, they will be able to elect four representatives in each—and still receive 40 percent of the total seats. That gerrymandering and unfair representation can be truly eliminated only under PR arrangements is so indisputable that even political scientists who continue to support our single-member plurality system are forced to acknowledge it. Nelson Polsby of the University of California at Berkeley is no supporter of proportional representation. Yet he testified in a court case on gerrymandering that “proportionality cannot be guaranteed in a system of voting in which the winner of the most votes wins the election regardless of how many candidates run in each race. To ensure proportionality it is necessary to have a proportional representation system of elections.”30 Proportional representation would also eliminate the uncompetitive districts created to protect incumbents. Under PR rules, it is impossible for one party to have a monopoly on the seats in a district and to prevent the others from winning any representation. Instead of a winner-take-all system we would have an all-are-winners system, where all parties have a reasonable chance to win some district seats. Also, as noted earlier, PR districts are much more responsive to changes in voter sentiment than one-party dominated districts. Even relatively small changes in partisan preferences are translated into changes in seats in the large multimember districts used by PR. Thus the adoption of proportional representation in the United States would in one stroke cut through the Gordian knot of redistricting and gerrymandering. It would eliminate all the partisan brawling, endless and expensive legal challenges, stolen seats, and uncompetitive districts that plague our single-member plurality system. Rarely do we find such a clear and simple solution to such a difficult set of political problems, but in the case of gerrymandering, PR is exactly that. Indeed, this ability to effectively eliminate both intentional and unintentional gerrymandering has been one of PR’s main attractions in countries that have adopted it. In these countries, redistricting is a trivial and boring political activity because it has no real political impact. We could certainly use some of that boredom here in the United States.

3

Improving Election Campaigns

Clearly, electoral systems can have dramatic impacts on who wins elections and whether citizens are represented fairly. However, they also have political impacts that go far beyond issues of representation. In particular, electoral systems can also have a significant effect on the style of political campaigns—for better or for worse. In many people’s minds, campaigns could not be much worse than they are now in the United States. Every election year disenchanted citizens complain to pollsters and write irate letters to the editor about lackluster races and mindless campaigns. Instead of opportunities to seriously debate the issues the public cares about, campaigns have become an endless parade of slick commercials and stage-managed appearances. Candidates seem to spend most of their time mouthing simplistic sound bites and slinging mud at their opponents. As rule, U.S. election campaigns have become a triumph of image over substance. But what few Americans realize is that some of these problems with our election campaigns are linked to the particular nature of our current electoral system. Traditionally, we tend to blame candidates and their campaign managers for boring or nasty election campaigns. The role that single-member plurality voting arrangements play in these fiascos has remained largely overlooked. But in many ways, candidates are merely responding to the rules of the electoral game as they find them. In any game, the rules help to determine the competitive conditions, which in turn influence how the players behave. It is no different in the political game of elections. If we want to

Improving Election Campaigns

69

understand why candidates campaign in a certain way, we have to look at the electoral rules that establish the conditions of competition. To take just one example, consider how electoral rules affect how much effort politicians put into campaigns. As seen in the last chapter, singlemember district plurality voting rules tend to foster the creation of many safe districts that are dominated by one party. In those districts, the election contest is basically over. Since entrenched incumbents usually have little trouble getting reelected in these districts, they need not put a great deal of effort into their campaigns. And the rival party in these districts often puts up only a token campaign against the incumbent. As a result, voters are forced to put up with lackadaisical and uninformative campaigns. However, lazy campaigns are eliminated in electoral systems that routinely produce more competitive districts. For example, Bowler, Donovan, and Brockington studied campaigns in local areas in the U.S. that are now using semiproportional voting systems. These systems utilize large multimember districts that encourage many different candidates to run for office, and allow many of them to win. The researchers found that when many candidates run in these more competitive districts they are forced to conduct more vigorous campaigns and to have more contact with voters.1 In general, in semiproportional and proportional election districts, both large and small parties always have good incentives to run vigorous campaigns, because they all can win seats and even a small swing in votes could translate into an additional seat. In this way, switching to PR elections in the United States would ensure that all voters in all districts had the benefit of enthusiastic campaigns. This chapter will examine in some detail two other ways that electoral systems affect political campaigns. In particular, it will focus on (1) how SMP elections tend to encourage campaigns that neglect a serious discussion of issues, and (2) how they give candidates incentives to pursue negative campaigns. We will also consider how adopting proportional representation for American elections would help to minimize both of these problems and produce campaigns that focused more on issues and less on mudslinging.

Neglecting the Issues Ideally, election campaigns should be opportunities for the voters to learn about the candidates’ policy positions and the social and economic theories

70

Improving Election Campaigns

that underlie them. Candidates should offer informative political advertisements that publicize their stands on the issues, especially the tough ones. And campaign speeches should be opportunities for candidates to systematically analyze our nation’s problems and offer a detailed set of proposals to address them. This would allow the public to assess what candidates stand for, evaluate the strengths and weaknesses of those positions, understand what candidates are likely to do in office, and choose the ones that best reflect their own policy preferences. But these are hardly the campaigns we get today, when issues and policies rarely play a central role in political campaigns. Campaigns avoid any meaningful engagement of the issues of the day in several ways. Often campaigns neglect policy issues in favor of creating a pleasing and appealing image for the candidate. The best examples of this approach are found in campaign advertisements, particularly television commercials. These ads seldom attempt to address issues systematically and often have no policy content whatsoever. The classic example is the “It’s morning again in America” television spots that ran during Ronald Reagan’s 1984 presidential campaign. They were full of soft music and comforting images of a mythical America: neighborhoods of freshly painted houses and immaculate lawns, a fire fighter raising a flag, a wedding, workers arriving at their factory, and a parade down Main Street. Today, one finds similar “all style/no content” commercials in virtually all congressional campaigns. Typically these will show the candidates in a variety of settings, walking in a parade, playing basketball with their children, walking their dogs, talking with senior citizens—all accompanied by a cheerful musical theme. At best, voters learn that the candidate loves his family, likes animals, enjoys talking to people—but nothing about what the candidate will actually do once in office. This approach gives voters no intelligent basis for making a decision. Campaign ads that avoid clear issues stands have become the norm. This was clearly shown in a study done by Richard Joslyn. He analyzed 131 television spots run by 24 liberal U.S. House and Senate incumbents. He found that almost half the ads (47 percent) had virtually no policy or ideological content at all. Fifteen percent had ambiguous or confusing policy content, and 8 percent of the ads even endorsed traditionally conservative stands. Fourteen percent of the ads could be identified as containing possibly liberal stands, and 16 percent contained clearly liberal stands on issues. Joslyn concluded that “it was only the 16 percent of the ads in [the last category] that provided citizens with the kind of information that would encourage the accurate perception of the policy preferences of congressional candidates.”2

Improving Election Campaigns

71

Policy Stands: Vagueness and Ambiguity Preferred Obviously, campaigns cannot ignore policy issues entirely. But even when candidates do address issues, they usually do so in a way that is so vague and ambiguous it is of little use to voters. This was shown in a study conducted by Paul Bradford Raymond in which he examined the campaign brochures used by 137 candidates for the U.S. House of Representatives.3 About half the brochures were devoted to issues, a finding that might seem encouraging. But he also found that candidates often simply mentioned issues without offering any detailed solutions. For example, almost half of those who mentioned government spending as a problem offered no solution at all or only a general pledge to cut spending. And even among those who proposed to cut spending by a specific amount, none mentioned any program or service that should be eliminated. Raymond found that, as a rule, candidates only offered a specific policy suggestion when it would be considered uncontroversial by most voters—such as proposing the elimination of the “marriage penalty” in the Social Security program.4 Vagueness and ambiguity in policy stands also fill campaign speeches. Unlike ads, which may be too short for detailed policy discussions, speeches would seem the ideal vehicle for informing voters of specific policy plans. Again, however, candidates usually fail to use speeches in this way. Studies show that most campaign speeches contain remarkably little useful policy information.5 Even so-called policy speeches consist primarily of general descriptions of problems (recessions, foreign threats, etc.) and promises to attain vague and uncontroversial goals (growth, peace, etc.). Incumbents typically devote most of their speeches to boasting about how good times are. In contrast, challengers focus on the problems afflicting society and try to pin the blame on incumbents. Candidates spend little speech time analyzing policy alternatives or offering specific proposals that spell out exactly what they intend to do once in office. Rarely do we hear candidates offering specific proposals about environmental problems like toxic waste or the greenhouse effect or detailed plans about how to address the deterioration of our education system. In his study of campaign speeches, Benjamin Page concludes: The most striking feature of candidates’ rhetoric about policy is its extreme vagueness. The typical campaign speech says virtually nothing specific about policy alternatives; discussions of the issues are hidden away in little-publicized statements and position papers. Even the most

72

Improving Election Campaigns

extended discussions leave many questions unanswered. In short, policy stands are infrequent, inconspicuous, and unspecific . . . . Candidates are skilled at appearing to say much while actually saying little.6

Undermining Democratic Control of Officials The main problem with deemphasizing issues in campaigns is that it undermines the democratic purpose of elections. Elections are one of the main mechanisms through which the public attempts to control policymaking. In democracies, the actions of government officials are supposed to correspond to the policy preferences of the public. Voters have two ways to use elections to ensure this occurs. First, voters can try to use campaign promises to predict which candidates will best promote their own policy preferences once in office, a process called prospective voting. Second, voters can use retrospective voting, evaluating how officials have performed in office and voting against those who have pursued policies with which they disagree. Both serve to keep officials democratically accountable. But they work only if voters have access to information about what candidates promise to do if elected to the legislature and the policy records of incumbent legislators. Campaigns that neglect issues deny voters both crucial types of information. As Richard Joslyn concludes from his study of congressional campaigns, “The policy appeals of candidates are too sparse to support the prospective policy choice approach, and the policy performance appeals are too infrequent and district case workoriented to support the retrospective policy satisfaction approach.”7 In other words, campaigns that neglect issues tend to sever the lines of control between the public and elected officials. When campaigns rarely mention specific policy stands and encourage people to vote for candidates because they have pleasing manners, good looks, or reassuring slogans, they undermine the process of electoral accountability vital to democracies. When voters support candidates for other than policy-oriented reasons, they have no guarantee that those candidates will enact the policies voters’ desire. Once in office, winning candidates inevitably claim that they have a mandate from the voters for their policy agenda. But as we have seen, voters often are not given enough relevant information to give candidates such a mandate. The result is that the public could easily find its elected officials pursuing policies that it does not support.

Improving Election Campaigns

73

The election of Ronald Reagan in 1980 was a good example of this problem. As noted earlier, he favored campaigns that were long on image and short on policy specifics. Nevertheless, he believed that he entered office with a clear mandate from the public to pursue a strongly conservative policy agenda. But surveys of public opinion during that time indicate that the public had not shifted to the political right on most policy issues. In fact, most voters actually supported increased government spending in many areas, such as aid to the poor, health care, environmental protection, and aid to big cities— all areas that Reagan targeted for drastic budget cuts once he was in office.8 A more recent example was the election of George W. Bush in 2000. During his campaign he kept his comments about the environment very general, claiming that “I am for clean air and clean water, and I have a record in Texas to prove it.” He also said he believed that global warming was a serious problem and that he would fight it through joint action with other nations.9 Given such assertions, it was easy for voters to get the impression that he was supportive of environmental protection measures. But once in office, he quickly revealed a policy agenda largely hostile to the environment. Within a few months he summarily rejected the Kyoto treaty on global warming, approved the loosening of arsenic standards in drinking water, proposed opening up the Arctic National Wildlife Refuge to oil drilling, proposed slashing the budget for alternative energy, announced his support for increased use of nuclear power, eased the way for increased mining on public lands, and made it easier for developers to eliminate wetlands. Clearly, the voters should have had these controversial policy initiatives in front of them if they were to make a rational choice between the candidates. But equally clearly, the Bush camp knew that some voters would strenuously object to many of these measures. So they made the politically expedient choice to keep these inherently controversial policies hidden from the public. As a result, we had a kind of pig-in-a-poke election, in which voters were forced to buy something they could not see. One should not get the impression from these examples that only conservatives are guilty of hiding their true policy intentions. Liberal candidates also pursue a similar strategy when they fear that their policy positions may actually be farther to the left than many of their constituents’. But keeping important policy information hidden from the public in this way is nothing less than political deceit—deceit that ultimately undermines the public’s ability to control the actions of the government. Campaigns that neglect issues weaken the democratic process, and this should be a matter of serious concern to American voters.

74

Improving Election Campaigns

The Overlooked Suspect: The Voting System Who or what is responsible for the avoidance of issues in campaigns? There are of course the usual suspects. Undoubtedly the media contribute to this problem. In recent years, reporters and editors have had less interest in reporting on issues in campaigns. Most of the coverage is of the race itself—who has momentum, who has made a gaffe, etc.—rather than the substantive differences between the candidates. Television news has been particularly bad in this regard. With its severe time restrictions, it tends to reduce speeches and debates to a series of ten-second sound bites, and the in-depth analysis of serious issues is rare. The rise of television as the main conduit of campaign news is clearly one reason why coverage of issues in campaigns has declined during the last several decades. And certainly some of the blame must rest on the candidates and their campaign advisors. After all, they are the main architects of this avoid-theissues approach to campaigning. But why do the candidates adopt this approach? The obvious answer is that it works—it helps them win the election. Or to be more precise, it helps them to win in a single-member plurality election. And this is the crucial point: what constitutes a successful campaign strategy can differ from one electoral system to another. It is similar to sports. As any coach will acknowledge, the rules of the games affect the strategy of the players. Professional golfers, for instance, play differently when they are trying to win a tournament conducted under match-play rules, than one using stroke-play rules. It is the same for elections. Electoral systems define the rules of the game—the manner in which votes are cast and the method for determining the winners—and this inevitably has an effect on how politicians play this game. To be more specific: under plurality rules, it is easier to win elections if candidates downplay issues. In contrast, the rules of proportional representation elections actually encourage candidates and parties to emphasize detailed policy stands. Let us see exactly why this is so.

Plurality Rules Make Issues Risky At the heart of our voting system is the plurality requirement: The only candidate who wins is the one who gets the most votes. In our two-party system, this usually means that candidates must receive the support of a majority of voters in their districts. So the basic overriding principle in any cam-

Improving Election Campaigns

75

paign is the need to garner that majority support. In practice, this is often a difficult problem, for candidates must try to please most of the people most of the time. Doing so is not an easy task with an electorate that often holds a wide range of conflicting opinions on such issues as abortion, taxes, global warming, affirmative action, gay rights, economic globalization, churchstate relations, and so on. In areas that are politically homogeneous—as in a town that is overwhelmingly liberal or conservative—this may be less of a problem. But even those areas often have real conflicts over the priorities given to certain community goals, such as economic development and environmental protection. In most cases, then, candidates are faced with the problem of how to please a majority or plurality of the voters in a heterogeneous political environment. Over the years candidates have developed several campaign strategies to deal with this problem—all of which contribute significantly to the tendency toward issueless campaigns. In a single-member plurality system, the first rule for building a majority is to wage a campaign that offends as few people as possible. Many Americans vote against one candidate rather than for another, so candidates must try not to alienate potential supporters. One of the best ways to do this is to avoid taking explicit stands on issues—because no matter what position a candidate takes, some portion of the electorate will take offense. If candidates support abortion rights, they alienate the antiabortion contingent, and if they oppose abortions, they alienate the other side. If they take a clear stand for strong measures to slow global warming, they will probably offend business; but if they do not, they offend environmentalists. Voter alienation multiplies as the candidates broadcast more and more of their stands on issues—eventually threatening their ability to put together a majority coalition. For example, if a candidate takes six positions on issues and each stance alienates a different 10 percent of the voters, the candidate will have succeeded in losing 60 percent of the electorate. The danger in taking a position is not merely theoretical. In a study of incumbents running for the U.S. House of Representatives, Robert O. Simmons, Jr. found that “incumbents actually were seriously hurt by knowledge of their [specific policy] votes when it was widespread in the district.”10 Thus successful legislative candidates in SMP elections quickly learn to keep their policy stands and legislative votes to themselves, a tendency noted in several studies cited earlier. Indeed, candidates can become so sensitive to alienating voters with their issue stands that they tie themselves in knots trying to be inoffensive. Consider, for example, several classic examples of can-

76

Improving Election Campaigns

didates desperately hedging on difficult issues. In his 1988 election campaign, George Bush found himself in an awkward situation while visiting a nonunion furniture factory in North Carolina. Faced with a potentially damaging question concerning his position on unionization, Bush confidently declared: “I believe in unions and I believe in non-unions.”11 Bill Clinton was also a master of appearing to be on both sides of an issue at once. In his 1992 campaign he tried to be all things to all people by repeatedly declaring that he was “pro-business and pro-labor,” “for economic growth and for protecting the environment,” and “for affirmative action and against quotas.”12 Such verbal contortions may appear comical, but they are in fact a very rational strategy in an SMP system where offending people can be the kiss of death. Instead of emphasizing issue stands—which is risky—it is much safer to campaign on character and personality. While voters hardly agree on specific policy issues, they generally agree on the personal qualities that make a good political leader. Campaigns that depict a candidate as strong, decisive, experienced, warm, and energetic will offend few voters. For a candidate in a single-member plurality system, an emotional or psychological appeal is much better than a policy appeal. Campaigns that focus on appealing personalities and comforting images are much more likely to help the candidates not offend potential supporters and build the majorities they need. Once we begin to see campaign behavior as a response to the plurality requirement, many of the most undesirable aspects of American election campaigns become much more understandable. Ambiguity and vagueness in campaigning suddenly begin to make more sense, because they are useful tools for building majorities in an SMP contest. A candidate strongly in favor of such vague ideals as peace, progress, family values, or a clean environment will disturb few voters. Such vagueness not only lessens the risk of alienating voters, it also increases the chance that some voters will mistakenly project their own views onto the candidate. As long as a candidate’s position remains vague, there is a chance that voters will attribute their own position to the candidate, especially if they already identify with the candidate’s image or personality. This process seemed to have helped Ross Perot’s independent bid for the presidency in 1992. Early on in his campaign he offered very few specific policy positions, thus enabling both conservatives and liberals to identify with him. Finally, if candidates have to pick some specific issues to emphasize in their campaigns, they are better off picking ones that are not divisive and on

Improving Election Campaigns

77

which many agree. This practice explains why during every election year in recent memory, all potential candidates for national office have gone out of their way to publicly advocate stronger anti-drug policies. Being against drugs is a no-risk policy stand, and it usually enhances the tough-on-crime image that candidates like to project. Candidates may risk losing the drug-pusher vote, but virtually everyone else will be on their side. This may not be the most important issue facing the nation, but it is relatively risk-free and thus exactly what candidates are likely to discuss in their ads and in their speeches.

Blame the System, Not the Candidates Given the role that plurality voting rules play in encouraging issueless campaigns, we should be blaming the voting system, not politicians, for this problem. Candidates are simply adapting to the political environment in which they find themselves. The biological analogy is quite appropriate. Evolution, to put it simply, is about who survives and who perishes in a particular environment. Similarly, elections are literally about which politicians will survive in our particular political environment and which will not. An important part of the political environment to which politicians must adapt is our election system with its plurality requirement. In a very real way, the plurality requirement sets the terms of political survival—it decides how the winners will be determined. Politicians must adapt or suffer political death. So candidates have an extremely powerful motivation to adopt strategies that win in a plurality system. Those who ignore this requirement and are foolish enough to try to emphasize substance over image will soon die off. As Patrick Caddell, a Democratic pollster, once noted: “Too many good people have been defeated because they tried to substitute substance for style.”13 Under the rules—the plurality requirement—the pursuit of a campaign heavy on image and light on policy is the most intelligent and rational strategy. The system is misguided, not the politicians. Therefore it is foolish to think we can solve this problem by simply encouraging candidates to put greater emphasis on policy stands. That is like asking them to commit political suicide. If we want them to move to more issue-oriented campaigns, we must change the rules of the electoral game. We must consider alternative electoral systems, such as proportional representation, that do not penalize those candidates who focus on issues.

78

Improving Election Campaigns

PR Allows for Issue-Oriented Campaigns Under systems of proportional representation, the rules for winning (and for political survival) are quite different, at least in the legislative arena. A legislative candidate or party does not have to garner a majority or plurality of votes to get elected. A party may receive only 20 percent of the vote in a PR system, yet its candidates will win 20 percent of the legislative seats. This simple procedural difference has a number of important effects on the style of political campaigns. First, candidates need not be obsessed with pleasing most of the people most of the time to get elected. As a result, their campaigns need not be confined to the vague generalities, undivisive issues, or character appeals that are unlikely to offend potential supporters. Second, and more important, candidates can be honest about their policy positions and can run issue-oriented campaigns because that approach is much less risky in PR systems. Parties and candidates can take clear stands on controversial issues—stands that inevitably alienate some voters—because as long as their positions appeal to some portion of the electorate, they will be able to place some candidates in office. For example, proportional representation has allowed viable Green parties to emerge in Germany and other European countries. Instead of merely invoking the vague label of environmentalist, Green candidates have run on policy platforms that specifically emphasize environmental and other issues. They have not shrunk from making controversial proposals that call for economic restructuring and the rethinking of basic societal values. The Greens have been well aware that this approach may initially disturb large numbers of mainstream voters, but they have considered that a fair price to pay in order to faithfully promote the issues and principles they believe are most important. Despite their highly issueoriented campaigns, the Greens have enjoyed some electoral success, particularly in Germany where they have been able to elect several dozen members to the national legislature, and have even been part of a ruling coalition. Equally important, by constantly hammering away at environmental issues, the Green party has been able to force the other political parties to address those issues in their campaigns. Clearly voters are more likely to find parties with such issue-oriented campaigns in PR systems where parties do not have to garner a majority of the votes to survive.

Improving Election Campaigns

79

Elections as Party Contests Many large European parties do run campaigns that seek to maximize their voter support and their number of seats in parliament. Today these parties also engage in sloganeering and slick media campaigns. But in comparison to American parties, even these mass-oriented parties tend to run highly issue-oriented campaigns. This is in large part due to another feature of most PR elections: Candidates run not simply as individuals but as part of a party slate. Elections are usually seen primarily as contests between parties, not individuals, which significantly changes the nature of campaigns. American campaigns emphasize the personality and character of individual candidates not only because such appeals will not offend voters but also because American candidates run as individuals. Personal appeals are natural when voters perceive elections as choices between individuals. But when voters see elections as primarily contests between party slates and party ideologies, the characteristics of individual candidates become less important, and the focus of campaigns shifts to the competing party platforms.14 In most PR countries, for example, campaigns do pay some attention to the personal characteristics of the party candidates running for prime minister or chancellor. But otherwise the focus is on the principles and policy positions of the parties, not simply on the individuals seeking election to the legislature.15 Indeed, most legislative candidates serve primarily as conduits for communicating the party platform to the public.16 Their campaign speeches are highly issue-oriented and less concerned with making personal appeals to voters. This is especially true in those few countries that use closed party list PR, where voters cannot vote for individuals at all, but must simply choose one party slate over another. However, most PR systems allow voters to do both: to cast a vote for a party and for a particular candidate on the slate. But this arrangement still avoids the extreme emphasis on individuals seen in American elections. There is more of a balance. Candidates in these PR elections are not simply saying “Vote for me,” they are also saying “Vote for my party” and “This is what my party stands for.” And this serves to make party platforms a key part of the campaign. Again, this kind of campaign behavior is very much a matter of political survival. In PR systems, the fate of candidates is much more closely tied to the fate of their parties than in single-member plurality systems. Even in PR systems that allow votes for specific candidates on the party slates, the individual

80

Improving Election Campaigns

candidates have a direct stake in how the party as a whole does. All the party’s candidates pool their votes and this determines how many seats the party wins. If the party does poorly, fewer of them will get elected. However, in the United States, candidates can run successful personal campaigns that largely ignore their party affiliations. Paul Bradford Raymond found in his study of candidates for the U.S. House that “most candidates avoid partisan appeals.”17 The tendency to run individual campaigns has steadily increased during the last thirty years as the role of parties in U.S. elections has decreased. American candidates no longer have to rely so much on their parties for campaign help. They can raise their own funds from political action committees and communicate directly with the public through their own media campaigns, appearances on talk shows, and so forth. In most PR systems, the centralized parties, not individual candidates, fund and run political campaigns. As a result, these campaigns are heavily party-oriented. The tendency of the media in PR countries is to follow the lead of the parties and to concentrate their coverage on the party platforms and policies also increases the focus on the parties.

Multiple Parties Encourage Detailed Positions Candidates and parties in PR systems tend not only to be more issueoriented but also to go into more detail in their discussions of their policy positions. Another characteristic of PR systems—the presence of multiple parties contesting the elections—encourages this. As chapter 4 will show, single-member plurality elections tend to produce two-party systems, while PR arrangements encourage a larger number of parties. This difference can affect campaign strategies. As Richard Katz explains, candidates and parties in multiparty systems are forced to give detailed explanations of their policy stands in order to distinguish themselves from the wide variety of stands offered by the other parties: If redistribution of income, for example, is an important issue, each candidate competing under a plurality formula would be expected to stress only that he favored more or less redistribution than his optimal enemy in the district. Candidates in PR systems, however, ought to stress exactly how much redistribution they favor since, in general, they will have to defend their share of the electorate against parties that favor more distribution as well as against parties that favor less.18

Improving Election Campaigns

81

The platforms of parties in PR systems, which are likely to be much more detailed and informative than what we find in the United States, provide evidence of this tendency. As a rule, European party platforms contain not only descriptions of the political and economic principles guiding the party but also detailed information about specific policies the party candidates will pursue once in office. U.S. party platforms are typically filled with lofty rhetoric and vague promises—which of course are appropriate in an electoral environmental created by plurality rules. Explaining the lack of specific policy proposals in the Democratic platform, one party activist notes: “The platform is only something that can be used against you.”19 PR countries, however, see the party platform as a campaign asset to be used for voter education. As a result, voters in these countries are more readily exposed to detailed policy information about the parties. Thus several unique characteristics of proportional representation elections encourage more informative campaigns: the ability to win without a majority of the votes, competition between party slates instead of individuals, and the presence of multiple parties. These features change the rules of political survival for candidates and parties and combine to push them toward more issue-oriented campaigns. This in turn makes it easier for voters to engage in issue voting, rather than simply responding to personality or image. In an issue-oriented campaign environment, voters have a much better chance of finding and electing candidates that reflect their own particular policy views. Moreover, candidates are elected with valid policy mandates, because voters know the policies that candidates will pursue in office. In this sense, the kinds of campaigns encouraged by proportional representation rules would increase the American public’s ability to exercise democratic control over policymaking.

Encouraging Candid Politicians Proportional representation not only encourages different kinds of campaigns, it also encourages different kinds of candidates. It fosters what some have called “principled politicians”—intelligent candidates who are committed to a set of political principles and policies. Instead of rewarding only bland candidates with vague appeals, PR makes it possible for principled politicians and principled parties with well thought-out ideologies and policy programs to gain office too.

82

Improving Election Campaigns

Principled politicians are also more honest with the public about their policy stands. As we saw earlier, candidates in plurality contests who are frank about their positions on controversial issues are likely to have a difficult time. SMP elections thus tend to encourage candidates to hide their true positions and retreat into vague and ambiguous stances. In a sense, this system rewards dishonesty. In contrast, PR rewards honesty. Under PR rules, candidates can be candid with the voters and still win office. They can voice all of their positions, even those that might alienate some voters, and still be elected with 10 or 20 percent of the vote. There is some reason to believe that principled politicians also make better leaders. These officials are more willing to face difficult issues and are unafraid of being decisive. One problem with candidates who run safe, issueavoiding campaigns is that this timidity may carry over into their actions in office. A voting system that rewards risk-averse candidates naturally creates a pool of legislators with an instinctive tendency to waffle on issues. One of the most frequent complaints about Congress is its lack of political courage and decisiveness. In a cover story on Congress, Business Week once lamented what it called “legislative cowardice” and traced part of the problem to the nature of American political campaigns: “Nothing motivates members of Congress like the fear of doing something that might be criticized. So all too often they do nothing . . . . With the rise of negative campaigning, many members are concluding that it’s safer to duck hard issues than to make unpopular choices.”20 American voters often suspect that candidates are not being straight with them—and they are usually right. This can only increase cynicism about politicians and politics. It is probably no coincidence that in countries that use PR, politicians are viewed with greater respect than in our country.21 To the extent, then, that proportional representation encourages the emergence of politicians who are candid with the public and who act on their principles, it would be a useful step toward improving both our election campaigns and the quality of political leadership in this country.

The Problem of Negative Campaigns Besides issue-avoidance, another major problem with American election campaigns is the use of negative advertising. Voters frequently complain about candidates slinging mud at their opponents. Polls show that 85 percent of the public believes that negative, attack-oriented campaigns are un-

Improving Election Campaigns

83

ethical and that candidates should spend more time discussing the issues.22 Unfortunately, it seems that in recent years candidates are spending more time “going negative.” It has become common for campaigns to hire consultants to do “opposition research”—a euphemism for digging up dirt on one’s opponent. Some independent political action committees also specialize in attack ads—often of the most aggressive kind. The characteristics of these negative ads have become all too familiar. They often use ominous music and unflattering pictures of their targets— sometimes grainy, black and white photos. The more disreputable ads may have footage that appears to be from newscasts, or may use actors posing as “citizens” or “experts.” All of them tend to seize on a single statement made by an opponent (or a single vote on a bill) and exaggerate its importance. Occasionally the ads contain falsehoods, but more often they specialize in innuendo, subtle deceptions, or half-truths. And they are usually designed to play on some fear or prejudice on the part of the voter. The hard-fought contests for the U.S. Senate in New York often provide good examples of how candidates can get nasty in their campaigns. In their 1992 race, Alphonse D’Amato and Robert Abrams used terms such as “sleazeball,” “wimp,” “sissy,” “fascist,” “bigot,” and “shakedown artist” to describe each other. The 1998 Senate race between Alphonse D’Amato and Charles Schumer was little better. The Schumer campaign portrayed D’Amato as a liar who cut Medicare, raised property taxes, acted “just like a bully taking lunch money from hungry kids,” and “doesn’t really care about you or your family.” In turn, the D’Amato campaign characterized Schumer as a “New York City liberal” who was soft on child pornographers, and wanted less jail time for violent criminals. Naturally, most of these charges were oversimplified and distorted the record of these two politicians. The “raising property taxes” charge against D’Amato was based solely on actions taken back in the 1970s when he was a Long Island town supervisor. The notion that Schumer was soft on pornography came from a vote against a 1977 measure in the state Assembly that was also opposed by libraries and publishers as being much too broad—and was supported by only 13 of 131 lawmakers. Of course, these particular attacks, while irresponsible, pale in comparison to some of more vicious negative campaigns, which sometimes attack a candidate’s family members, or exploit past problems with alcohol, or play on racist, homophobic, or xenophobic sentiments among voters. Negative campaigning creates a number of political problems. It degrades the tenor of political discussion—making it something more akin to

84

Improving Election Campaigns

the ranting found in professional wrestling than reasoned political debate. It also manipulates voting decisions by providing voters with misleading information. Attack ads often do contain references to the political stances of the candidates, but the “facts” are often unreliable or distorted—and so are not of much use in making an intelligent choice between the candidates. Some evidence indicates that excessively negative campaigning also depresses voter turnout. In their book, Going Negative: How Political Advertising Shrinks and Polarizes the Electorate, Ansolabehere and Iyengar stated that “our most troubling finding is that negative or attack advertising actually suppresses turnout.”23 They argue that after several weeks of negative campaigning, voters become convinced that both candidates are undeserving of political office and this makes them less likely to vote. A study by Kahn and Kenney came to similar conclusions. Interestingly, they found that voters could distinguish between legitimate negative information and mudslinging. Voters believed that critical campaign messages could be useful as long as they were relevant and delivered in an appropriate manner. But “when candidates hurl unsubstantiated and unjustified attacks at each other, [people] are ‘turned off’ to the campaign and are more likely to stay home.”24 Given the already dismally low levels of turnout that we have in American elections, we can ill afford a campaign style that depresses participation even further. However, there is an even more troubling effect of negative campaigning: it feeds the political cynicism that already eats away at American politics. Voters exposed to relentless negative attacks can become disaffected from all political candidates. And this alienation may spread to the political system itself. In their book, Ansolabehere and Iyengar argue that voters are unlikely to trust governmental institutions, like legislatures, that are comprised of the untrustworthy individuals portrayed in negative ads. For this reason, they fear that “negative advertisements pose a serious antidemocratic threat.”25 Former U.S. Representative Matthew McHugh, president of an association of former members of Congress, has reached a similar conclusion. “Public confidence and participation are critical to a smooth-running democracy. Unfortunately, they have declined in recent years, in part because of increasingly negative campaigns that have contributed to public cynicism about the political process.”26 Thus it would seem that the political costs of negative campaigning are felt beyond the elections themselves; this approach can also contribute to the deterioration of the quality of our nation’s political life.

Improving Election Campaigns

85

The Electoral Connection Of course the main reason that candidates so often “go negative” is that this strategy works—it helps them get elected. But it is important to note that negative campaigning is much more effective in the two-way contests that are typical of an SMP voting system. If there are only two choices of candidates— usually Republican and Democrat—then attacks ads that drive voters away from one candidate will often drive them into the arms of the other. Or, as we’ve seen, it may drive some voters to not vote at all. For the campaign launching the attacks, either of these outcomes is advantageous. It is best if your opponent’s supporters come over to you, but it is also beneficial if they simply choose not to vote. Lani Guinier has explained well the rationale behind such efforts to discourage voting. Winner-take-all systems are also responsible for at least part of the problem with the way campaigns are conducted in America. When all that counts is winning, its not surprising that candidates who fall behind resort to negative campaigns. If the point is to win more votes than your opponent—not to win as many votes as you can—then it makes just as much tactical sense to suppress the votes for your opponent as to raise your won vote totals. So candidates try to discourage at least some people from voting.27 However, negative campaigning is not such a sure-fire strategy in proportional representation elections. This is true for several reasons. First, as discussed earlier, PR election contests are as much about parties as they are about individual candidates. Instead of running as lone individuals, candidates run as members of party slates in each district. It is much more difficult to smear a party than to smear an individual candidate. Parties are rarely arrested for drunk driving or caught cheating on their wives. However, the main factor in PR elections that discourages negative campaigning is the fact that these elections typically involve more than two parties. Usually several major and minor parties offer slates of candidates in these multimember district contests. These multiparty contests mean that candidates cannot be sure that they will benefit greatly from attack campaigns. Consider, for example, an election where four parties are vying for votes. On the right are the Republicans, and a smaller, more conservative party. There is also a small centrist party, and the Democratic party on the left. Even if a Republican at-

86

Improving Election Campaigns

tack were to succeed in driving voters away from the Democratic party, there is no guarantee that they would end up in the Republican camp. More likely, they would opt for the centrist party. So in these multiparty contests, there is much less incentive for campaigns to be attack oriented. One particular type of proportional representation, the single transferable vote (STV) also discourages negative campaigning in another way. In STV, voters rank order the candidates, putting a 1 by their first choice, a 2 by their second choice, and so on. If your first candidate has too little support to get elected, then your vote is transferred to your number two choice to help elect that candidate. This potential for vote transfer between candidates serves to discourage negative campaigning. The reason is simple: If candidates can benefit from being the second or third choice of a voter (and thus get a transferred vote) they are less likely to throw mud at the voter’s first choice. If they engage in vicious personal attacks against their opponents, they risk alienating these possible supporters. So while candidates and parties do criticize their opponents, they tend to do so in polite terms, and they also highlight areas of agreement—all in hopes of attracting those valuable second-choice votes. We need not merely speculate about the effects of STV on campaigns, because we actually have some experience with this system in the United States when it was used in several dozen cities in the earlier part of the twentieth century. There is some anecdotal evidence from that time that STV did have a beneficial influence. Writing in the early 1940s, George Hallett, an early proponent of PR in the United States, observed: Not the least of the advantages of PR is that it promotes decency and good feeling in elections and afterward. Instead of trying to beat a particular person, with the temptation to belittle his ability and blacken his character, each candidate is trying to win a group of supporters for himself out of the whole field. He knows that he cannot defeat the leaders of the opposition, and he knows further that unfairness to any candidate may alienate second choices that he might otherwise receive from that candidate’s friends . . . . It does not pay to throw mud. His best course is to make a vigorous statement of what he himself stands for, without gratuitous attacks on anyone else.28 In his studies of the use of the single transferable vote in Cambridge, Massachusetts (where it is still used), and other American cities, Leon

Improving Election Campaigns

87

Weaver found at least partial corroboration for Hallett’s claim. He concluded that PR does have an effect on campaign styles and that in particular “competition for transfer votes provides an incentive for candidates to mute personal attacks on other candidates (including those on opposing slates) lest they alienate those candidates’ first choice supporters.”29 To the extent that STV could add this kind of civility to election contests, it would certainly be welcome.

Stifling the Great God Ballyhoo During the first proportional representation election in Cleveland in 1924, a reporter for the Cleveland Plain Dealer noted the improved nature of the political campaigns: “It is a strange campaign, the strangest Cleveland has ever seen. The great god Ballyhoo, the patron of party strategists in all other campaigns, is dead, the victim of P.R.”30 This is an encouraging observation, but probably a bit overly optimistic. While the adoption of proportional representation would have a number of beneficial effects on campaigns in the United States, it is no panacea. The great god Ballyhoo is a durable deity. Even under PR some candidates and parties would downplay difficult issues in order to broaden their appeal. And some candidates always will decide to run on personality or to use negative campaigning to manipulate the public. But while PR would not be a complete cure for our campaign ills, it would certainly be a great improvement. As Lani Guinier has concluded, proportional representation would be a “constructive reform [that] would help to undermine the culture of negative campaigning and candidatecentered politics.”31 Proportional representation would make it possible to run positive, issue-based campaigns and for principled politicians to be honest with voters and still win office. Issue-oriented candidates and campaigns would produce better-educated voters and encourage more issue-based voting. This in turn would make politicians more accountable and more likely to reflect the policy preferences of the public. In this way, the adoption of proportional representation would not only help to improve political campaigns, it would also contribute to making our political institutions more accountable and more democratic.

4

Toward a Multiparty System

Imagine going to the supermarket and turning down the breakfast cereal aisle to find that there are only two kinds of cereal for sale: corn flakes and granola. Or imagine going into a shoe store to find that, instead of dozens of styles to choose from, there are only two. Undoubtedly most Americans would be disappointed and upset about this lack of choice. But a similar lack of choice occurs in most U.S. elections. When Americans walk into the voting booth, they usually find only two viable choices: Republicans and Democrats. While Americans come in a wide variety of political stripes—from ultra-conservative to ultra-liberal and everything in between—on election day they are forced to try to fit themselves into one of only two political categories. What makes this restriction of choice even more peculiar is that it does not exist in most other Western democracies. If you lived in Sweden, you would have seven parties in the legislature from which to choose on election day. In Germany, you would have five parties and in New Zealand, six. So why are the political choices so limited in this country? The answer, it turns out, can be traced back to our particular kind of electoral system. In fact, political scientists have long known that there is an intimate connection between a country’s electoral system and the size and nature of the party system that develops there. This chapter will explore in some detail this connection between electoral systems and party systems. First, we will consider many of the common complaints about our two-party system, and then see how those problems are caused by our single-member plurality voting sys-

Toward a Multiparty System

89

tem. We will then examine how proportional representation would encourage a healthy and dynamic multiparty system in the United States, and consider all the political benefits that would come from this change.

Dissatisfaction with the Two-Party System Political pundits sometimes maintain that we have a two-party system in the United States because that is exactly what the American people want— that voters are quite satisfied with the choices offered by the Democratic and Republican parties. But the evidence suggests otherwise. First of all, most Americans do not feel a passionate connection to either major party. Surveys show that only 16 percent of Americans strongly identify with the Republican party, and a similar number feel a strong attachment to the Democrats. Another 30 percent feel only a weak connection to one of the parties. More importantly, the largest group—35 percent—reject both parties and call themselves independents.1 So fully two-thirds of Americans feel little or no connection to the major parties. Dissatisfaction with our two-party system can also be seen in the growing interest in third parties and independent candidates. The presidential candidacy of Ross Perot in the 1990s demonstrated that there was considerable voter interest in alternatives to the major-party candidates. In recent years, millions of Americans have also supported a variety of third-party efforts, including the Libertarian party, the Greens, the Reform party, the Natural Law party, and the New party. Moreover, these efforts to create new parties to rival the two major parties have considerable sympathy among most Americans. Polls consistently show that a majority of Americans now would like to see a third party emerge to challenge the Republicans and Democrats. It was not always this way. In 1944, most Americans were satisfied with the two-party system and only 14 percent said they would like to see a new party. But by the mid-1990s, the polls revealed a striking change of heart. For example, a CNN/USA Today poll found that only 26 percent of Americans wanted the party system to continue as it was, and 53 percent wanted a new party to exist along with the two traditional parties. Additional polls have shown that 51 percent, 57 percent and 59 percent agreed with the view that the nation should have a third party competing for people’s votes.2 Today less than a quarter of Americans (23 percent) believes the two-party system “works fairly well,” and almost half (49 percent) think that our two-party system is basically unsound.3

90

Toward a Multiparty System

The Parties as Fraternal Twins Americans are dissatisfied with the two-party system for a number of reasons. One of the most frequent complaints is that the two parties are often too similar in their stands, and so don’t offer Americans a real choice at the polls. This view is found not only among the public, but among political experts as well. Political scientists have been known to comment that “the major parties [tend] to cluster around the middle of every issue”4 and that “the two major parties try to be all things to all people and are therefore indistinguishable from each other.”5 A favorite exercise in some college textbooks on American politics is to quote passages on various policy issues from Democratic and Republican party platforms and to challenge students to identify which is which. They inevitably do poorly, in large part because the parties often sound similar in some of their basic ideological and issue positions. Indeed, the overlap between the parties is so great that it is often virtually impossible to tell the difference between conservative Democrats and liberal Republicans. It has reached the point where party labels are sometimes useless as guides to a politician’s behavior—it is not unusual to find Democrats and Republicans in our legislatures voting for each other’s bills. Of course, this is not to say that the major parties are indistinguishable. In fact, they do offer distinct policy options on some very important issues, such as abortion, environmental protection, taxes, and immigration. So clearly the parties are not identical twins. But it would not be inaccurate to call them fraternal twins who sometimes bear more than a passing resemblance to each other. This similarity restricts the choices available and can be very frustrating for some voters. A classic example occurred in the 1968 election, when those against the Vietnam War were confronted with a choice between two presidential candidates and two parties that both supported that war. During the 1980s, liberal Democrats could have voted for a Democratic candidate for Congress as a protest of Ronald Reagan’s conservative policies, only to see their winning candidate side with the Republicans on cutting social programs, increasing military spending, dismantling environmental regulation, and abandoning progressive taxation. And in the 1990s, voters who opposed the Gulf War or the drastic cuts in welfare spending found that neither major party supported their positions. The similarity of our two major parties is perhaps easiest to see if we step out of the American context, which tends to amplify the differences between

Toward a Multiparty System

91

figure 4.1 Norwegian versus American Party Systems

the Democrats and Republicans, and compare our party system with the multiparty systems common to most other democratic countries. Figure 4.1 compares our parties with the parties available to Norwegian voters. Several things can be observed in this illustration. First, while there is a difference between our two parties, they are actually relatively close together on a larger ideological scale. The Democrats are really a centrist party—not a leftist party—which brings them much closer to the Republicans than we might first think. Second, it is clear that there is a much wider range of parties in the Norwegian system and that those parties offer a broader variety of distinctive choices for voters. The Labor party is a moderate leftist organization that represents the interests of union members, fishing crews, and other workers. Further to the left are the Left Socialists, strong supporters of social and economic equality and extensive welfare state programs. In the political center is (unsurprisingly) the Center party, which largely represents rural interests. The Christian People’s party is also a centrist party and it appeals primarily to Lutherans who support sexual and social equality but also favor a prohibition on liquor, religious instruction in schools, and a ban on abortion. Further to the right are the Conservatives who represent the urban middle class voter interested in lower taxes, free markets, and freedom from moralistic government constraints. The Progress party is a far-right party vehemently opposed to immigration and the welfare state. This multiparty system offers Norwegian voters a wide variety of parties that are clearly different from one another. In the United States, the Republicans and Democrats huddle together in the middle-right of the political spectrum; but in Norway, the parties offer options all the way from the far right to the far left.

92

Toward a Multiparty System

The tendency of the two parties to crowd together in the middle of the political spectrum is a product of our single-member plurality electoral system. As Lowi and Ginsberg have explained, the “effect of the plurality system is that over time it contributes to the development not only of a two-party system but of two moderate parties that generally differ only slightly in their programs and policies.”6 The source of this tendency is the plurality requirement itself—only the candidate with the majority or plurality of votes can win. The problem here is essentially the same one discussed in the earlier chapter on campaigns: In a plurality system, parties must try to appeal to the largest number of voters possible. That is the only way to succeed. If most voters tend to congregate in the middle of the political spectrum—as many believe is the case in the United States—both parties will also tend to migrate there in search of a majority.7 In contrast, in multiparty PR systems, parties tend to spread out along the political spectrum in order to represent a wide variety of voting groups.8 The similarity of the parties also helps to explain another frequent complaint about the two-party systems: that the parties are out of touch with some portions of the public. Although they both claim to be umbrella parties that represent a wide variety of groups and interests, in reality they do not pursue policy agendas that please all of their constituents. In particular, as the two parties compete over the same middle-of-the-road swing voters, they tend to ignore those voters who are farther out on the political spectrum. In its march to the middle, for instance, the Democratic party has ignored large portions of its more progressive constituency—labor unionists, left liberals, staunch environmentalists, feminists, minority activists, etc. On the Republican side, some of the more right-wing elements of the party, such as fundamentalist Christians, feel that the party has not been promoting their major moral and social concerns. In a two-party system, these kinds of voters can easily be ignored or taken for granted by the parties because they have nowhere else to go. Feminists will not vote for Republicans, and members of the Christian Coalition will rarely support a Democratic candidate. So the parties realize that they can afford to be out of touch with some of their supporters while they concentrate on courting the swing voters that could make the difference in a plurality election contest.

The Inadequacy of Only Two Parties Even if the two parties were not overly similar—even if they were to develop into parties offering detailed and distinctly different policy options to

Toward a Multiparty System

93

the voters—this would not entirely eliminate the problems surrounding our two-party system. Having only two parties—no matter how different—is in itself a very limiting and problematic political situation. Having only two choices unreasonably restricts the political options available to the electorate. Many combinations of positions can be taken on the pressing issues of the day, but in our party system they are automatically reduced to two. Say, for example, that you were very liberal on social issues (in favor of gay rights, the decriminalization of drugs, open immigration, etc.) but very conservative on economic issues (opposed to property and income taxes, government regulation of business, etc.). You would not fit neatly into either major party. But this is exactly the political stance of the Libertarian party, and enough Americans sympathize with this stance to make this party the largest third party in the U.S. Another kind of example illustrates the full extent of this problem. Assume that we face only five important political issues in an election—say, defense, education, environmental protection, farm subsidies, and health policy— and that we must choose to increase or decrease expenditures in each area. Even this simplified situation presents 32 combinations of positions that could be offered by parties. In a two-party system, one party might advocate increasing defense spending and cutting the others, and the other party might advocate the opposite. But where does that leave all the voters who desire any of the other combinations? They are left with no choice they can enthusiastically endorse and with the task of deciding which is the lesser of two evils. And that is the basic problem with a two-party system: It simply cannot offer anything approaching a reasonable variety of positions on the issues. Again, the inherent limitations of choice in our party system become even more obvious when compared with European party systems—systems that offer not only a larger number of parties but also a wider variety of parties with distinct ideologies and policy programs. Consider again the party system in Norway. Voters there have the option of not one, but two moderate parties in the center. They also have a labor and a socialist party on the left and an ultra-conservative party on the far right. Many European democracies also have Green parties that claim to be neither left nor right. In these multiparty systems, voters have a much better chance of finding parties and candidates with policy positions close to their own. Supporters of the two-party system sometimes suggest that limiting the public to two choices may in fact be an advantage—that reducing our options conveniently minimizes the complexity and difficulty of election choices. The assumption is that any more than two options would strain the

94

Toward a Multiparty System

intellectual capacity of most voters. But this is an insulting idea. There is no reason to believe that European voters are capable of choosing among multiple parties but American voters are not. Indeed, American consumers have become quite good at choosing between multiple versions of a product— such as the many different models of cars, each with a variety of optional features. Choosing their favorite among four or five competing political parties should not be a problem for American voters.

One-Party Systems Having only two options in the election booth is bad enough, but our choices are often even more restricted than that. The reality is that in many areas of the country we do not even have a two-party system; we have a oneparty system. In cities, counties, and states in which one party has a reliable majority of the voters, that dominant party is usually the only viable option. Indeed, for most of the last one hundred years, one-party systems have been the rule in most areas of the United States. Today, in many areas of the Northeast, Democrats dominate politics; while in the South and West, it is the Republicans who usually rule. Even in states such as Illinois, where no one party is in control, one part of the state is dominated by the Democrats and another by the Republicans. We can get an idea of the extent of one-party rule in the United States by looking at the number of uncontested seats in state legislative elections. Major parties will sometimes give up and not even mount a challenge in a district that is clearly controlled by another party. In 2000, for example, either the Democrats or the Republicans failed to nominate a candidate in 40.6 percent of the contests for seats in state legislatures in the United States.9 In Massachusetts that year, an incredible 71 percent of the state legislative races had only one major-party candidate. It is not an exaggeration to call these “no-choice” elections. Our current party system often closely resembles a corporate oligopoly in which the two dominant companies divide up their territories and agree not to compete with each other in them. Indeed, Mayhew and others present evidence that Republicans and Democrats sometimes collude in exactly this way, with legislators agreeing on gerrymandering schemes that ensure safe districts for the representatives of each party.10 So it is misleading to call ours a competitive two-party system; often it is more accurately described as a pair

Toward a Multiparty System

95

of one-party monopolies. The big losers in this situation are the same ones who suffer in a one-company territory—the public. Such arrangements severely curtail the choices of American voters and ultimately undermine their power to control the political system. Americans have long been aware of the evils of economic monopolies and oligopolies, but we have been slow to awaken to the dangers of the same arrangements in our party system.

The Electoral Connection If our two-party system is so frustrating to many Americans, why does it persist? Why haven’t we developed a multiparty system that offers a set of genuine political choices? Why don’t our legislatures reflect the true variety of political views present in the American electorate? One reason is the power of tradition. Most Americans are socialized into our two-party system and learn to view our political universe in those limited terms. We come to think of having only two parties as natural and desirable. The media contribute to this view by giving little coverage to any minor-party candidates who do happen to run, making it more difficult for these challengers to get their messages out and to build larger bases of public support.11 The two parties have also devised numerous election procedures that discourage minor parties. Many states, for example, still require excessively large numbers of voter signatures on petitions before minor parties can even get access to the ballot.12 But even if third parties had automatic access to the ballot, it is unlikely that they would flourish in the United States. Today, where these parties do appear on the ballot, they rarely elect anyone or become viable alternatives. That is because the biggest obstacle to these parties still remains: our singlemember plurality election system. This is the main reason why we continue to have a two-party system when there is so much public dissatisfaction with it. Plurality rules tend to foster two-party systems by systematically discriminating against minor parties and making it extremely difficult for them to achieve any electoral success. But how exactly do SMP systems foster two-party systems? In the 1950s the French political scientist Maurice Duverger was the first to describe the relationship between voting rules and party systems, and it remains one of the most famous propositions in political science.13 Duverger argued that plurality voting tends to favor two-party systems and proportional representation tends to favor multiparty systems. In particular, he noted that two effects of

96

Toward a Multiparty System

SMP elections serve to encourage and maintain two-party systems: the “mechanical” effect and the “psychological” effect. The mechanical effect is the built-in tendency of plurality voting to give the largest party more seats than it deserves and smaller parties fewer seats than they deserve. This problem of distorted representation was discussed extensively in the first chapter. In general, the second-place party is underrepresented in the legislature; but the fate of even smaller parties can be much worse—they are usually shut out of representation entirely. The culprit is the requirement that a candidate receive at least a plurality of the vote to win office. Usually only candidates from the two largest parties stand a realistic chance of winning the large number of votes that is needed. When minor-party candidates are denied representation in this way, the inevitable result is a two-party legislature. The only way for minor parties to enjoy any kind of electoral success in plurality systems is by being concentrated in local or regional enclaves, where they can sometimes muster a plurality of the votes. This is the case with the small Welsh and Scottish parties in Great Britain; regional popularity allows them to send several members to Parliament. In the United States some third parties have been concentrated in particular states. In the 1930s the Progressive party in Wisconsin was able to elect a governor and many state legislators; during that same period the Farmer-Labor party in Minnesota captured the governorship for three successive terms. As a rule, however, minor parties have not been able to develop this kind of regional dominance in the United States, and so it has been difficult, if not impossible, for minor-party candidates to win office. The tendency of plurality systems to prevent the election of minor-party candidates is compounded by the “psychological” effect of those rules. Duverger noted that potential supporters will hesitate to vote for a minor-party candidate if they believe that the candidate has little chance of winning a plurality or majority of the vote. They fear wasting their votes. It is much more rational for voters to support a candidate who stands a chance of winning—usually one from the two major parties. Thus even though minor parties and their candidates might enjoy some support among the electorate, supporters will often realize that the only realistic choice is to vote for the “lesser-of-two-evils”—the major-party candidate they believe will do the least damage to them politically. Comparative studies have confirmed the existence of this problem, finding that voters often will abandon a preferred minor-party candidate to reluctantly cast a vote for a major-party candidate with a better chance of being elected.14

Toward a Multiparty System

97

Thus plurality voting subjects minor parties to a double disadvantage: it first ensures that these parties will have little or no representation in the legislature, and this in turn discourages people from voting for these candidates in the first place. As if this were not bad enough, parties that cannot elect candidates for office usually also have a very difficult time recruiting experienced and talented politicians. Such politicians are inevitably attracted to the two mainstream parties where the career opportunities are dramatically better. Moreover, minor parties usually have difficulty attracting financial contributors, who are understandably hesitant to invest money in quixotic campaigns. Thus minor parties are caught in a vicious circle: Plurality rules discourage voter support, which makes potential candidates and contributors reluctant to join up, which further erodes the ability of these parties to conduct effective campaigns and to attract voters, and so on. These effects can quickly seal the fate of a minor party. But the plight of minor parties under plurality rules is actually even worse than this. Minor-party voters may actually make themselves worse off politically by contributing to the election of the very candidate they oppose the most. Imagine, for instance, a voter faced with a choice of a liberal Democrat, a moderate Republican, and a candidate of the Right to Life party. A strongly conservative voter might be tempted to support the Right to Life candidate, if only as a protest vote, but doing so would take that vote away from the moderate Republican and thus boost the chances of the conservative’s least preferable candidate, the liberal Democrat. This is the infamous “spoiler” problem. A spoiler is a minor-party candidate who takes away enough votes from one major-party candidate to ensure the election of another one who would not have won otherwise. The classic example took place in the 2000 election for president. Disgruntled leftists who left the Democratic party to support Ralph Nader, the Green party candidate, learned a very harsh lesson. Their votes hurt Al Gore in Florida and several other key states and directly resulted in the election of George W. Bush, whose policy agenda quickly proved to be a political nightmare for these voters. The spoiler phenomenon is the ultimate punishment for anyone who would dare support a minor-party candidate in an SMP system. Supporters of the two-party system often cite the low number of votes cast for minor-party candidates as evidence that there is little public interest in such parties. However, this argument is misleading since our SMP system actively discourages potential supporters from casting votes for these minor parties. Because of this problem, we cannot rely on final vote counts as an accurate indicator of voter interest. Consider, for example, the case of John

98

Toward a Multiparty System

Anderson, who ran as an independent candidate in the 1980 presidential elections. Opinion polls indicated that a substantial number of Americans— up to 24 percent—supported Anderson. However, only 7 percent cast ballots for him on election day. A University of Michigan national survey indicated that of those voters who rated Anderson the highest among the three candidates, only 39 percent actually voted for him.15 Clearly his final vote tally was not an accurate measure of his support among the electorate. As the survey cited earlier indicated, most Americans would now like to see additional parties competing in elections. And this interest in third parties is nothing new. The long history of third-party efforts in the United States—including the Populists, Socialists, Progressives, American Independents, and others—demonstrates that millions of Americans have frequently been interested in a wider range of political options. In fact, more than 1,000 minor parties have been created since the founding of the Republic. The problem is not that there is no interest in such parties, but that our particular voting system has doomed these efforts to failure. Given the multiple ways that single-member plurality elections disadvantage and discourage minor parties, it is hardly surprising that the two-party system persists in the United States. The longevity and stability of this system is not an indication of its popularity, but a testament to its ability to rig the election rules in its favor. Plurality voting rules discriminate against minor parties and provide artificial and unfair support for the Democratic and Republican parties. Instead of creating an open electoral market in which all parties compete freely for the support of voters, plurality rules put minority parties at a huge competitive disadvantage and virtually ensure the continued dominance of the two major parties.

Toward a Multiparty System The only effective way to move beyond the limitations and problems of the two-party system is to adopt proportional representation in the United States. Under PR rules, minor parties would immediately become viable alternatives. They would no longer be discriminated against, and they would finally be able to win their fair share of legislative seats. The plurality voting requirement would no longer prevent minor parties from being represented. A small party might only receive 10 percent of the votes, but it would still win 10 percent of the legislative seats.

Toward a Multiparty System

99

In short, proportional representation would be an antitrust law for the party system. It would discourage party monopolies and oligopolies and allow for free competition among parties. It would create a level playing field on which all parties could vie fairly for public support. Numerous studies have confirmed the ability of proportional representation to provide a more welcome setting for minor parties and thus to foster a multiparty system. One study found that more than half of the new parties formed in PR countries eventually won seats in the national legislatures—something that is extremely rare in our system.16 Another comparative study found that PR countries average four to five parties in their national legislatures, while countries using plurality/majority systems average only a bit over two parties.17 A more hospitable political environment for minor parties under PR would thus probably result in the expansion of the party system in the United States. Voter support for minor parties would increase as voters realize that their votes will not be wasted and their candidates will not act as spoilers. Talented politicians would be more attracted to these parties. They could run for office on those tickets without fearing that they are throwing their careers away. Donations to these parties would probably increase as the contributors realize that these investments could actually produce some electoral dividends. What kind of minor parties would be likely to thrive in this new political environment? Third parties that already have some presence, such as the Libertarians or the Greens, would be likely to expand under PR rules. A centrist, nonideological party like the Reform party could also garner widespread support from many Americans who now think of themselves as independents. It is also possible that new parties might break off from the major parties. A coalition of progressives—including unionists, feminists, and left liberals—might defect from the Democratic party. A conservative party— perhaps appealing to fundamentalist Christians—could split off from the Republican party. In areas with concentrations of racial minorities, we might see the emergence of an African-American party or a Latino party. Which minor parties eventually emerged would undoubtedly vary from region to region depending on the political preferences of local voters. Would the major parties disappear entirely, perhaps fracturing into many small parties? There is little chance of that. It is much more likely that these parties would remain the largest parties in the U.S. Sixty million voters still strongly identify with these parties and that situation is unlikely to change very rapidly. Moreover, the presence and importance of presidential elec-

100

Toward a Multiparty System

tions in our political system will continue to aid the major parties. Unlike parliamentary systems, the chief executive in our presidential system is elected separately by a plurality vote. The winning presidential candidate must garner a majority or substantial plurality of the vote—something that is also true of governors and mayors—and this requirement gives an advantage to large political parties like the Democrats and Republicans. These broadbased parties are best equipped to muster the wide voter support required. In fact, the presidential election may be much of why two-party dominance has been stronger in the United States than in other plurality countries, like Great Britain and Canada, which have parliamentary systems. In any case, the most likely scenario for the United States would be for the Democratic and Republican parties to remain the dominant parties, with a number of minor parties emerging to join them in the legislatures. Moreover, it is important to recognize that the adoption of PR in the United States would not force us to have a multiparty system; it simply would allow such a system to develop if it reflected the wishes of the American voters. As political scientists have often observed, many factors other than electoral systems help determine the number of parties in a political system— such as the number and depth of political cleavages in a society.18 Thus if American voters choose to support only the two major parties, PR would produce a two-party system, as has happened often in Malta and Austria. In this sense, PR does not mandate any particular kind of party system; it simply does not inhibit the development of a multiparty system the way plurality rules do. With proportional representation, what the public wants in a party system, it gets. This principle was evident in the experiments with proportional representation in U.S. cities. The effect of PR on party systems varied from city to city, depending on local political conditions and public preferences. In some cities that adopted PR, such as Cincinnati, essentially two parties still contested local elections, though PR produced a much more accurate representation of those parties in the city council. In cities with more heterogeneous political populations, like New York, a vigorous multiparty system emerged. Before the adoption of proportional representation, New York City was dominated by the Democratic machine, which elected virtually the entire city council. The onset of PR broke the political monopoly of the Democrats, and what was a one-party system became a multiparty system. The PR city council in 1947 reflected the wide variety of political persuasions among the New York City electorate at that time and consisted of twelve Demo-

Toward a Multiparty System

101

crats, five Republicans, two Liberals, two Communists, and two American Laborites.

Multiparty Systems Provide More Choices for Voters Development of an authentic multiparty system in the United States would have a number of significant political advantages. First it would greatly benefit voters, because they would now have a larger number of genuine political choices. Instead of having only one viable party option in a district—the dominant party—voters could choose among four or five parties that would have the ability to elect candidates to office. And with a wider variety of choices, voters would be more likely to find a party with which they could truly identify. Instead of having to choose between the lesser-of-twoevils of the major-party candidates, they could choose any candidate and party that they felt accurately reflected their political philosophies and policy preferences. And if they opted for a minor party, they would not have to worry about wasting their vote or inadvertently aiding the party they liked the least. Given this wider range of viable choices and fewer wasted votes, voters would find participating in elections a much more rewarding and satisfying activity. (This would tend to increase voter turnout—an advantage that will be addressed in more detail in chapter 7.) Proportional representation would help not only those who want a wider range of parties to choose from but also those who do not like parties at all and prefer independent candidates. Right now voters who support independent candidates who are not affiliated with any party often find themselves in the very same situation as those who support minor-party candidates: They have little chance of electing their candidates in a single-member plurality system. In most cases, such candidates have trouble amassing the plurality of the vote needed to win office. In contrast, independent candidates can easily be included on PR ballots and can get elected with relatively small amounts of the vote, just like minor-party candidates. Ironically, then, even if voters are turned off completely to the idea of parties, a multiparty PR system is much better from them than an SMP system. A multiparty PR system would also solve the political dilemma faced by members of the major parties who do not feel effectively represented by these organizations—particularly those on the far left of the Democratic party and the far right of the Republican party. Currently, discontented

102

Toward a Multiparty System

members of these parties are faced with a no-win situation. If they stay in the parties, they will continue to be ignored because it is taken for granted that they will not defect to the other party. But if they try to break off from the major party and start their own party, our SMP voting system will ensure that this effort fails. For example, small right-wing parties that have come from the Republican ranks, such as the Right to Life party and the U.S. Taxpayers party, have had almost no electoral success. However, proportional representation would allow dissatisfied members of the major parties to start their own viable parties and ensure that their points of view are faithfully represented in the political process. Research has also shown that because PR voting systems are more open to the creation of new parties, they are more effective in channeling alienated and disgruntled citizens back into the political process. In their study, Arthur Miller and Ola Listhaug note that a party system that has few options and that does not effectively address some people’s problems can create dissatisfaction with parties and elections. Democracies rely upon the institution of elections to maintain satisfaction with government as an arena of authoritative decision-making. But if the same social problems persist across a series of governments that are led by different parties, an increasing number of people come to believe that none of the parties reflect a particularly viable alternative.19 They also found that when citizens become alienated from parties in this way, these feelings tend to spill over and create distrust of government in general. Thus when Americans feel that the Republican and Democratic parties are not addressing their grievances, they are likely to become turned off to elections and to feel more cynical about government. A good example of this situation may be the poor in the United States. Neither party actively courts the support of this interest group; and in the last several decades, neither party has been effective in lowering the rate of poverty or reducing the growing gap between the rich and poor. So it is not surprising that many of the poor have given up on electoral politics and have a very low rate of voter turnout. Such situations are ripe for the creation of new political parties to address these neglected concerns and lure these citizens back into the political process. But as Miller and Listhaug found, countries like the United States have had less success in channeling these discontented citizens back into the normal political process because they have a rigid party system that makes it vir-

Toward a Multiparty System

103

tually impossible for these protest parties to emerge.20 Without viable alternative parties, these citizens tend to drop out of the political process or pursue other less desirable means to make their political points, such as demonstrations or riots. However, proportional representation makes it easier for viable new parties to start up and serve as routes back into the political process. A multiparty PR system ensures that virtually all political groups have parties that espouse their views, and that the election system can be responsive to many alternative points of view. At a time when there is growing concern about the high levels of political alienation and apathy in the United States, it seems clear that a multiparty system that offers all citizens an avenue to at least some political power could help disaffected citizens to become re-engaged in the political process.

Could PR Help Major Parties? Certainly adopting proportional representation would aid the prospects of minor parties, but it might have some beneficial effects for major parties as well. This may not seem obvious. Indeed, some fear that PR would actually hurt these parties by encouraging some of their weaker supporters to defect to some of the alternative parties. However, while some defections would certainly occur, their results may not be as damaging as they first appear. First, the major parties might actually benefit from losing some of their more radical members to minor parties. The exodus of disgruntled party members would significantly cut down on fractious internal conflicts. The Democratic party, for instance, is often racked by intense and divisive battles between its centrist and leftist elements. When the leftist groups lose these fights, they feel resentful and may offer little active support for the party’s nominees. On the other hand, when the party adopts the position of these leftist groups, the mainstream of the party feels betrayed. As was evident in the case of presidential candidate George McGovern in 1972, the victory of leftist groups in the Democratic party can drive away some of its traditional supporters. In this sense, losing some of these more radical voters to other parties would allow the major parties to become more moderate, unified, and cohesive. It is also important to note that the loss of these splinter groups is, in a sense, only temporary. Minor parties that split off from major ones would

104

Toward a Multiparty System

most likely join up with the main party in a legislative coalition. Say, for example, that a fundamentalist Christian party broke off from the Republican party. In the legislature, these parties would be natural allies on many issues and would join together their conservative forces in most legislative battles with the left. Moreover, proportional representation could actually benefit many of those voters who support the major parties by finally assuring them of representation. No such assurances exist in our SMP system. In every congressional election tens of millions of major-party supporters come away from the polls with no one to represent them. Republicans living in predominantly Democratic districts—and vice versa—stand little chance of electing their candidate. In fact, currently almost all of the wasted votes in the United States are cast by supporters of the two major parties, not third-party supporters. Republicans in Massachusetts and Democrats in Idaho would be much better off under PR than under SMP. Indeed, all major-party supporters who are minorities in their districts would be beneficiaries of a proportional representation system and it would be in their interest to support this reform.

Multiparty Systems Provide Better Representation A multiparty PR system in the United States would not only give voters more choices, but would also for the first time produce local, state, and federal legislatures that accurately reflected the true strength and variety of political opinions among the electorate. This is currently not the case in our twoparty system. As political commentator Michael Lind has observed: “Because of our peculiar election law, the American government is divided between two parties. The American people are not.”21 We need a government that faithfully mirrors the political diversity that exists in American society—that is, after all, one of the main purposes of elections. Unquestionably a multiparty PR system is more likely to accomplish that basic democratic goal. Once our legislatures reflect the views of all significant political factions, these institutions will be much more likely to produce policies that actually reflect the public will. This is the ultimate political payoff of having a multiparty legislature that is a true microcosm of society. As was seen in chapter 1, one of the main advantages of multiparty PR legislatures is that their positions on policy issues more closely resemble those of the public. In his study,

Toward a Multiparty System

105

G. Bingham Powell measured the political views of policymakers in PR legislatures and majority/plurality legislatures, and then compared those with the general views of the public in those countries.22 He found that there was typically a closer correspondence between the views of the public and the views of policymakers in the PR legislatures. So if the purpose of elections is to ensure that governments enact policies that are in keeping with public opinion, then clearly this is more likely to happen in multiparty PR systems.

Encouraging Broader Debate and New Ideas The emergence of a multiparty system in the United States would also encourage a much more stimulating and wide-ranging discussion of political ideas. It would do so in several ways. First, multiple competing parties would certainly broaden and enliven the political debates that take place during campaigns. Currently, most major media ignore minor-party candidates and their ideas. Their typical excuse is that it makes little sense to cover parties and candidates that have no chance of winning. This same reason is also invoked to justify excluding independent and minor-party candidates from campaign debates—resulting in a two-party discussion of issues that is often limited and unimaginative. But under PR rules, this excuse would no longer be valid. Minor-party candidates can win office in PR elections, and thus they would have to be included in regular media coverage and in the candidate debates. At the very least, it would make these campaigns less predictable and boring. It would certainly expose the public to a broader range of political and policy ideas, and thus contribute considerably to public education about these issues. As was explained earlier in the chapter on campaigns, a multiparty contest not only ensures the expression of more points of view, but also forces the parties to be much more detailed about their policy stands so that they may differentiate themselves from all of their competitors. Having many parties in our legislatures would also raise the quality of political debate in these institutions. In a democracy, legislatures are not merely decision-making bodies; they are also deliberative institutions that serve as forums for political discussion and debate. A multiparty legislature would encourage a truly pluralistic debate over the issues and would ensure a forum for the views of minority political groups. As Lani Guinier explains, such wide-ranging, multi-perspective discussions would be very useful in

106

Toward a Multiparty System

Congress, city councils, and local school boards. “Indeed, the more representative such a body, the more it can perform its deliberative function. A legislator or council member needs to consider a full range of views before he or she deliberates to reach consensus or to participate in a public conversation about important policy issues.”23 Having independent and minor-party voices in our city, state, and federal legislatures would improve political discussion in several specific ways. First, it would force the dominant parties to acknowledge and defend the assumptions underlying their policy programs. A two-party system in which the parties share some basic social and economic assumptions often discourages discussion of these central ideas. As a result, political debates are less vigorous and enlightening than they should be. When the two major parties agree on some important issues—such as pursuing the Vietnam War in the 1960s, cutting social programs in the 1980s, or reducing welfare spending in the 1990s—then it is very useful to have other parties that are willing to vigorously question this consensus. Challenging the conventional political wisdom offered by the two parties enlivens debate within legislatures, while serving to educate the public about deeper implications of public policy decisions. The presence of minor parties in legislatures also ensures that new political ideas and new political voices are heard in a timely fashion. One of the most frequent complaints about the two parties is that they are slow to respond to new issues and new political perspectives. Indeed, their dominance of the legislative and media forums often makes it difficult for those new voices to be heard and to gain some legitimacy. In the United States, third parties have often served as alternative forums for new ideas and policies. Parties like the Socialists, Progressives, Populists, and Farmer-Laborites often publicized and promoted new policy ideas that were being ignored by the major parties. Many of the social, economic, and political reforms that we now take for granted—such as free public education, abolition of slavery, regulation of railroads and business monopolies, progressive income taxes, child labor laws, women’s suffrage, direct election of senators, Social Security, and unemployment insurance—were first proposed and championed by minor parties. Many of these issues were quite controversial when the smaller parties first embraced them, and it is not surprising that the major parties shied away from them, viewing them as too divisive for the large electoral coalitions they sought to build. Minor parties helped to develop popular interest and support for reform in these areas. Along with other pressure

Toward a Multiparty System

107

groups, these parties played an important role in forcing many of these issues out into the open and ensured that they received a place on the political agenda of the nation.24 All of this is not to imply that the ideas promoted by minor parties are always better than those of the major parties. We need only consider a policy like Prohibition to see that this is not true. However, our two-party system is also quite capable of producing enormous policy fiascoes of its own—the Vietnam War, the massive budget deficits of the 1980s, and the banking deregulation that necessitated a $500 billion federal bailout of failed savings and loan institutions in the 1990s are but a few examples. Indeed, such fiascoes may be more likely in a two-party system in which the options are few and the debate is limited. The more challenging and wide-ranging debate of multiparty legislatures could help protect against ill-conceived policies. Unsound ideas—from whatever part of the political spectrum—are more likely to wilt when exposed to the heat and light of an intense public debate. Americans now live in a world of accelerating technological, economic, and social change. We are constantly facing new political and economic challenges both at home and abroad. Faced with these challenges, we can ill afford a party system that restricts debate and is slow to recognize new ideas. Instead, we need a multiparty system that can serve as an incubator for a wide variety of creative policy approaches. We are more likely to find innovative and effective policy approaches to our current and emerging problems if we have a free marketplace of ideas—and that is exactly what a multiparty system can encourage.

PR—The Only Route to a Multiparty Democracy Given all the political advantages of multiparty elections and multiparty legislatures, it is not surprising that these systems are found in most Western democracies. Nor is it surprising that a variety of efforts continue to be made to encourage a multiparty system in the United States. For example, in recent decades people dissatisfied with the two major parties have created several new parties, such as the Reform party and the Green party, to better represent their views. So far, however, this strategy has failed to produce electoral payoffs. In every national election, several million voters do cast their ballots for various third-party candidates, but those votes are almost inevitably wasted. A few Libertarians and a few Greens have been elected on

108

Toward a Multiparty System

the local level, but by and large our SMP voting system has been very effective in blocking third-party candidates from office. It is evident that merely voting for these candidates will not make a multiparty system a reality in the United States. Some reformers have attempted to address some of the institutional obstacles to third parties in the hopes that this will smooth the way to a multiparty system. Third-party activists have been trying to remove ballot access restrictions that make it difficult for these organizations to take part in elections. In many states, lowering the number of signatures needed on petitions would make it much easier for minor parties to get on the ballot. Campaign finance reformers have been promoting public financing of campaigns and this could also aid minor parties. Equal public funding of all candidates would provide much needed money to minor-party contenders and allow them to compete with major-party candidates on a more even playing field. Some minor-party activists believe that providing their party with sufficient funds to get their ideas in front of the public will finally allow them to begin to win elections. However, while these institutional reforms are certainly desirable, they are not sufficient to create a multiparty political system in the United States. Even if third-party candidates could easily appear on ballots and had funds to run adequate campaigns, they would still stand little chance of election in a single-member plurality system. They would still be unlikely to get the plurality or majority of the vote needed to win office. In the end it is clear that the only way that we can have a true multiparty system in our legislatures is to abandon single-member plurality elections and adopt PR. A multiparty PR system would not only make our party system and legislatures more representative, it would also help to revitalize public interest in elections. Americans would see a more interesting variety of candidates, much livelier debates, more competitive districts, and a broader discussion of new ideas—a refreshing contrast to the dull contests that are all too typical of today’s two-party elections. So if we want American elections to be both more representative and more engaging, then we need the kind of multiparty system that only proportional representation can provide.

5

Electing More Women

Imagine a group photograph of 100 people standing on the steps of the U.S. Capitol. At first glance, the group appears to be entirely female, but a closer look reveals a dozen or so male faces among the sea of women. The caption below the picture: “Members of the United States Senate.” Undoubtedly most people would find such a picture very strange. Indeed, if the Senate were to actually become 88 percent female, this would certainly make headlines in newspapers around the country. Political pundits and talk show hosts would be wondering how we got into this extraordinary situation and whether this severe gender imbalance was good for the country. Of course, that is not the situation today. In recent memory, the Senate has always been a decidedly male institution. In 2002, for example, it was 87 percent male. But if it is shocking to think of the Senate being almost entirely female, shouldn’t we be equally shocked that it is almost entirely male? The U.S Senate is merely one example of the persistent and gross underrepresentation of women in U.S legislative bodies. Although women constitute more than half the voting population, they make up only a small fraction of our elected officials. In 2002 the U.S. House was little better than the Senate, with a mere 14 percent of its members being female. As a result, the most important and powerful policymaking bodies in the United States are virtually all-male clubs. Women do somewhat better in state legislatures, where they hold an average of about 21 percent of the seats. But this is still far from an equitable proportion of the legislative offices. America’s poor record of representation for women looks even more abysmal when we realize that

110

Electing More Women

many other countries do better—in fact, a great deal better. For example, in 2001 women held 43 percent of the seats in the national legislature of Sweden, 36 percent in Norway, 36 percent in the Netherlands, and 31 percent in Germany.1 The contrast is striking. This chapter explores the problem of women’s underrepresentation in the United States, how it is tied to our voting system, and how it could be improved by proportional representation elections. But before addressing the causes and potential solutions, it is useful to first consider exactly why this situation is problematic—why it is so undesirable. To many it may seem obvious that the severe underrepresentation of women in political office is unfair and even undemocratic. But there are a number of different reasons to condemn it.

What’s Wrong with Underrepresentation The first reason the underrepresentation of women in American political life is undesirable has more to do with efficiency than fairness: Underrepresenting women means that a large pool of potential political leaders is underused. Women clearly have enormous and varied talents to contribute to political life, and limiting their participation in leadership positions is wasteful. This is not a new argument. It goes all the way back to John Stuart Mill, one of the earliest supporters of the movement to increase women’s participation in legislatures. In On the Subjection of Women, published in 1869, he asked: Is there so great superfluity of men fit for high duties, that society can afford to reject the service of any competent person? Are we so certain of always finding a man made to our hands for any duty of function of social importance which falls vacant, that we lose nothing by putting a ban upon one-half of mankind, and refusing beforehand to make their faculties available, however distinguished they may be?2 Such arguments are just as applicable today when we find our nation faced with a great many difficult challenges and problems and what often seems to be a lack of effective political leadership. As it stands, the political skills and expertise of more than half the population are being wasted. The second argument is more policy oriented: Male-dominated legislatures have not always satisfactorily represented women’s views and interests.

Electing More Women

111

Often these legislatures have not taken women’s issues seriously and have been very slow to pass policies that adequately address these problems. If more women were elected to positions of power in legislatures, those bodies would be more likely to pursue a more women-friendly policy agenda, particularly in such areas as affirmative action, pay equity, spousal abuse, parental leave, and day care.3 Changing the gender mix of lawmakers may also have policy impacts beyond the realm of “women’s issues.” Polls reveal significant differences between men and women in a diverse range of important policy areas. For example, 70 percent of women want stricter gun control versus 50 percent of men. And nearly twice as many women (44 percent vs. 29 percent) believe that poverty and homelessness should be extremely important government priorities.4 Studies of female state legislators also reveal that they differ significantly from their male colleagues on such issues as the death penalty, mandatory prayer in school, and tax cuts.5 Clearly, any large increase in the number of women officeholders could make a big difference on a variety of political issues. Of course not all women officials are likely to share the same political opinions. Female Republican lawmakers, for instance, tend to reject feminist positions on some issues. Their policy positions are sometimes closer to their male Republican colleagues than to female Democrats. However, studies have consistently found a gender gap among legislators, with most women being significantly more liberal and feminist than men. In one study of state legislators, 39 percent of female lawmakers identified themselves as liberal, while only 19 percent of males did so.6 Also, voting studies of women in the U.S. House of Representatives have revealed a strong tendency to support women’s issues and concerns.7 Several studies also give us a glimpse of how the policy output of legislatures would change if substantially more women were elected. A study by Sue Thomas found that state legislatures with higher number of women pass more bills dealing with women, children, and families.8 Studies on the international level have found that national legislatures with higher proportions of women tend to have more innovative and expansive family policies and pass more laws that benefit children.9 The final argument against the underrepresentation of women has to do with political legitimacy. Irrespective of what policies it pursues, a government that grossly underrepresents women can undermine its own legitimacy. The representativeness of policymaking bodies is one of the main ways that democratic governments establish legitimacy and authority for their policies.

112

Electing More Women

When any large segment of the population, such as women, is severely underrepresented, it tends to erode this legitimacy. Policies handed down by an exclusively or predominantly male legislature might easily have a problem garnering the respect and support of the entire population. This would certainly seem to be the case in the area of women’s issues. One reason that antiabortion legislation has been so controversial in the United States is the perception that it is being made primarily by males. Male-dominated legislatures are always open to the accusations that they do not (or cannot) really understand what is at stake in women’s issues and that they do not fully appreciate the way their policies affect women. Questions naturally arise about whether people who cannot become pregnant should make the crucial decisions about birth control and abortion policies or whether the very people who have never suffered from sexual discrimination—who in fact have been the main beneficiaries—should decide whether it is a serious problem. Thus in many policy areas the decisions of a virtually all-male legislature may be automatically viewed with suspicion by some women, and those decisions may therefore lack the authority they need to be fully effective.

The Conventional Wisdom The conventional wisdom about women’s poor political representation in the United States usually blames other factors instead of the electoral system. It has often been suggested that we have few women representatives because (1) women run less aggressive and less competent campaigns, (2) they have more trouble raising funds for their campaigns, (3) they are disproportionately slated in “hopeless” races against incumbents, and (4) that many voters are sexist and prefer male candidates. But the conventional wisdom is wrong. Studies have shown that none of these factors play a significant role in the underrepresentation of women today. For example, Darcy, Welch, and Clark found no evidence that women were poor campaigners. Indeed, they found that voters reported significantly more contacts from campaigning women than men and that female candidates were often better known and had higher name recognition than similar male candidates.10 While at one time female candidates did have trouble raising money, this is no longer the case. In fact, beginning in the 1990s, the evidence shows that women have on average been more successful in raising funds than male candidates. A recent study by Burrell found that “current women can-

Electing More Women

113

didates now raise and spend as much or more than their male counterparts . . . . When women run, they can raise the money they need to win.”11 In the 1994 election cycle, for instance, women incumbents, women challengers, and women open-seat contenders all raised more money from political action committees than male counterparts in comparable situations.12 The charge that women are most often recruited to run against incumbents in difficult races appears at first to be true: Most new female candidates for Congress have had to battle incumbents. But this fact is misleading because most new candidates, irrespective of gender, must run against incumbents. Given the longevity of most incumbents, women have little alternative but to challenge them if they desire to win office. Similar proportions of new male candidates for Congress also have had to try to unseat an incumbent. Thus it does not appear that women are being systematically and selectively recruited for hopeless races. Another frequently cited cause of women’s poor representation levels is voter discrimination against female candidates. At one time, this was clearly true. Many voters in the 1950s and 1960s did believe that women were less qualified for office than men. But recent evidence suggests that sexism does not present a substantial barrier to female candidates. Darcy, Welch, and Clark argue that women now suffer little electoral disadvantage in the United States. In their study of U.S. House of Representatives races, they found that after adjusting for party affiliation and whether the candidate was an incumbent, voters showed little evidence of discrimination—men and women did equally well in attracting votes.13 More recently, Burrell found that female candidates are now often more successful than men in attracting votes.14 This is not to suggest that sexism among the American electorate has disappeared completely. It has not. Surveys continue to show that one in five voters agree that women are less well suited to politics than men.15 But, as scholars point out, the effect of these attitudes is undermined by several other factors. First, much of this anti-women bias is offset by a pro-women bias on the part of other voters. In a survey that asked voters whether they would prefer a generic male or female candidate, more voters indicated a preference for female candidates over males.16 Second, even if some people have a sexist attitude toward female candidates, it often does not translate into anti-female votes because that prejudice is outweighed by more important political considerations, such as the candidate’s party or incumbency status.17 So a conservative voter with some anti-female bias could easily end up voting for a woman if she is a Republican running against a male liberal

114

Electing More Women

Democrat. In the end, then, it seems clear that sexism among voters does little to explain why the U.S. does so poorly in the representation of women.

The Electoral Connection If conventional explanations for the poor representation of women in the United States do not hold up under scrutiny, what does explain this situation? One major cause of this problem is the under-nomination of female candidates. When women run, they tend to win, but they simply are not nominated enough to make large inroads in our legislatures. Why is this the case? There are several reasons. First, women continue to be underrepresented in the occupational pool from which candidates are typically drawn. For instance, many legislators begin their careers as lawyers. And while law schools are now quite open to women, their numbers as a proportion of the entire law profession have only been growing slowly.18 But while the continuing lack of women in the eligibility pool does help to explain their underrepresentation in American legislatures, it does not explain why the levels of women’s representation in the United States are so low compared to many other Western democracies. As table 5.1 shows, many other European democracies are far ahead of us in terms of women’s representation in their national legislatures. Why is that the case? It is certainly not because women in these countries constitute much higher proportions of the occupations from which politicians are recruited. There are not three times as many female lawyers in Sweden than in the United States. So what does explain this difference? It turns out to be the electoral system. As table 5.1 illustrates, countries that use proportional representation tend to have much better representation of women. Countries like the U.S., which use single-member plurality or majority electoral systems, tend to cluster near the bottom of the list. Women tend to be nominated in much higher numbers in PR countries, and so they are elected in higher numbers as well. This relationship between electoral systems and women’s representation is one of the most well documented tendencies in the field of electoral systems. Numerous studies have shown a clear propensity for single-member plurality systems to do much worse than PR systems in representing women. After reviewing the considerable literature in this area, Francis Castles observed that

115

Electing More Women

table 5.1 The Representation of Women in 13 Western Democracies Country

Voting System

Percentage of Women Elected to Lower House

Sweden

Party List PR

42.7% (1998)

Denmark

Party List PR

37.4% (1998)

Finland

Party List PR

36.5% (1999)

Norway

Party List PR

36.4% (1997)

Netherlands

Party List PR

36.0% (1998)

Germany

Mixed PR Member

30.9% (1998)

Austria

PR Party List

26.8% (1999)

Belgium

PR Party List

23.5% (1999)

Canada

Single-Member District Plurality

20.6% (1997)

United Kingdom

Single-Member District Plurality

17.9% (2001)

United States

Single-Member District Plurality

14.0% (2000)

Ireland

Single-Transferable Vote PR

12.0% (1997)

France

Single-Member District Majority

10.9% (1997)

Source: Inter-Parliamentary Union, “Women in National Parliaments,” 2001. http://www.ipu. org/wmn-e/classif.htm (August, 2001).

“it is the unanimous finding of all those studies which have been explicitly concerned with the linkage between women’s legislative representation and the nature of the electoral system that systems of proportional representation appear to favor higher levels of female representation.”19 In one early study, Wilma Rule conducted a regression analysis of 23 countries in which she tested the correlation between the levels of women’s representation in those countries and a number of political and socioeconomic factors. She found that the strongest correlation was with the type of voting system, which was the best predictor of how well women did in the various legislatures. Pointing to the poor level of representation in majority-plurality countries, she concluded: “the primary reason for this

116

Electing More Women

figure 5.1 Percentage of Women’s Representation Under Different Electoral Systems, 1950–1998

Source: Richard Matland, “Enhancing Women’s Political Participation: Legislative Recruitment and Electoral Systems,” in Azza Karam, ed., Women in Parliament: Beyond the Numbers (Stockholm: International IDEA, 1998), p. 77.

is the electoral system, rather than inhibiting social, economic, and cultural factors.”20 In a more recent study of 24 advanced democracies, Richard Matland found that proportional representation countries consistently outpaced winner-take-all countries in women’s representation during the last 50 years—see figure 5.1.21 He found that the differences between the systems tended to be relatively small until the 1970s, when organized feminist groups began to push for the election of more women. PR systems responded with dramatic increases in the number of female legislators, while the progress in single-member district countries has been much slower. This has meant an increasing gap between the two electoral systems. In 1998, for instance, women averaged 23.03 percent of the seats in the national legislatures in PR countries, but only about half as many—11.64 percent—in winner-take-all countries.

Electing More Women

117

The Best Evidence: Germany and New Zealand While these comparative studies consistently point to electoral systems playing an important role in the problem of women’s underrepresentation, by far the most clear and convincing evidence of this connection can be found by examining the results of elections for the German national legislature, the Bundestag. Some people are skeptical of comparative studies because they suspect that differences in representation may be due to differences in the political cultures among countries, rather than a result of electoral differences. For example, some countries may have less sexism and this could account for their better record of women’s representation. However, examining the situation in Germany allows the influence of the voting systems to be assessed in isolation from other causal factors, such as political culture. This is because Germany uses the mixed-member electoral system, an election system that includes both SMP and PR on a double ballot. On the first part of the ballot, voters cast votes for individual candidates in districts, as we do in the United States. All the winning candidates from these SMP races are seated in the Bundestag. On the second part of the ballot, citizens vote for a list of party candidates. This ballot determines the percentage of seats that each party will ultimately receive in the Bundestag. If, for example, a party received 30 percent of the vote on the party-list ballot, it would receive 30 percent of the seats. This is achieved by adding candidates from the party-list ballots to the candidates who won seats in the SMP elections until the party reaches its proper percentage of seats. The double ballot arrangement makes Germany an ideal case to test the influence of voting systems alone since other potential factors, such as sexist attitudes or differences in eligibility pools, are at exactly the same levels for both voting systems. Thus any significant differences in women elected to office in the two ballots would have to be primarily attributable to differences in the voting systems. It is revealing, then, that women have consistently done much better in the PR portion of the German elections. In a typical example in 1994, the percentage of women elected in the single-member district contests was 13 percent—about the same as what we find in U.S. congressional elections—while the percentage elected from the party list PR contests was 39 percent. It is significant that the same phenomenon is also seen in elections in New Zealand. In 1993, New Zealand switched from SMP to the German style mixed-member version of PR. In the 1999 national elections there, women won 24 percent of the single-member district

118

Electing More Women

seats, but 40 percent of the party list seats. Such results are convincing evidence that electoral systems can make a dramatic difference in the way women are represented in legislatures.

Electoral Systems and Party Self-Interest Women are better represented in proportional representation systems because they are nominated more often. But why is this the case? The answer is that PR systems change the incentives for parties to nominate women. Under PR rules, it is in a party’s interest to nominate many female candidates, and it is easier to do so than in SMP systems. The secret lies in the fact that parties in PR systems do not run individual candidates in single-member districts, but slates of candidates in multimember districts. This situation puts pressure on parties to construct lists that represent the broad electorate—including women—so that their slates will have wide appeal. Any party that ran a slate entirely of men would be open to accusations of sexism and jeopardize its election chances. In addition, if there is a women’s group within a party, having gender balance in the party slate is a way of maintaining internal peace and assuring the continued support of this faction. So for several reasons, it is in the self-interest of parties in a PR system to include women on their slates in fair numbers and positions. The same political pressures on parties are not built into single-member plurality systems. It doesn’t necessarily appear sexist if the lone party candidate is a man. On a practical level, there is no way to have gender balance on a party ticket if only one candidate is nominated. Also, trying to nominate more women in a SMP system can have political costs for a party. It is a zerosum game: nominating more women often requires a party to explicitly deny the seat to an incumbent or to a man, which could create internal tensions in the party.22 Such problems do not exist in PR systems, where party slates easily allow for the nomination of candidates from both genders. Studies have also shown that there is often a process of what Matland calls “contagion” at work in the nomination of female candidates in PR countries. Once one party goes ahead and includes large numbers of women on their slates, other parties feel pressure to do the same. In Norway, for example, the Socialist Left party was the first to have an official policy of including a larger proportion of women on their slates. The Norwegian Labor Party reacted by increasing the number of women candidates in

Electing More Women

119

winnable positions in exactly those districts where they faced a serious challenge by the Socialists.23

Quotas—Best in Conjunction with PR Another factor that contributes to the better representation of women in some PR countries is the use of quotas by political parties. In Sweden, for example, several parties now require that women occupy no less than 40 percent of the positions on the party slate. The Swedish Social Democratic Party also requires that men and women alternate on their lists, so that if a man is listed first, a woman must be second, and so on. In South Africa, the African National Congress has adopted a quota of 33.3 percent women on their party list. In 34 democracies, some or all of the political parties now have quotas for female candidates in national legislative elections. Some controversy surrounds the use of quotas and not all parties support them; but there is no denying that quotas have increased the nomination and election of women where they are used. PR by itself increases women’s representation, but the effect is even stronger in countries where parties use these quotas. Viewing the successful use of quotas in other Western democracies, some feminists in the United States have called for their adoption here. But quotas do not work very well in single-member district electoral systems where candidates run one at a time and, as in the U.S., are chosen in primaries. It is difficult for a party to impose a gender quota in this situation. Quotas clearly work best in multimember district PR systems that use slates of candidates. For this reason, activists interested in using party quotas to increase women’s representation in the United States should also be working for the adoption of PR as well.

The American Experience Interestingly, we need not rely entirely on evidence from abroad to show that using slates of candidates can increase women’s representation. There are several examples in the United States that illustrate this relationship. For example, we have one election in the U.S. that utilizes the equivalent of a party-list system—the presidential election. We do not vote directly for presidential candidates, but instead for slates of electors that will cast their votes

120

Electing More Women

in the Electoral College. So when votes are cast for a party’s candidate, they are really cast for that party’s list of electors. As in party-list PR, pressure is on the state parties to include women on the lists, if only to avoid accusations of sexism. The results are striking. In the 2000 presidential election only one state, Missouri, elected a slate of electors that was all male. Of the 538 electors chosen, more than 38 percent were women—certainly a large improvement over the fate of women in single-member plurality elections. The American experience with proportional representation—in the form of the single transferable vote (STV)—also provides some evidence to support the notion that this system is good for female candidates. Kathleen Barber studied the use of STV in Cleveland, Ohio from 1921 to 1931. She found that women were elected to the city council for the first time under STV. At least one and often two women were elected in every STV election. In contrast, no women were elected before PR, and women disappeared from the council for several decades after the abandonment of PR. Barber concludes that the election of women was due in part to the political mobilization of women for the suffrage campaigns but also notes that “the multimember districts of the PR period provided a structural opportunity facilitating [women’s] representation.”24

Undermining the Incumbency Factor Besides increasing the nomination of female candidates, the adoption of proportional representation in the United States could help to increase women’s chances of being elected in another way. One factor slowing the growth in the number of women in legislatures is incumbency. In the United States, particularly in Congress, incumbents are almost always reelected, in part due to advantages in fund-raising, name recognition, and so on. This means that the turnover in legislative seats is infrequent. The less turnover, the smaller the chance that women can increase their numbers in Congress. The importance of incumbency seems to be one of the main reasons why women are better represented in state legislatures (averaging 21 percent) than in the U.S. Congress (13 percent). In the states, the rate of voluntary retirement from legislatures is higher, as more lawmakers run for higher office or return to private-sector jobs. This means that fewer incumbents are running and more races have open seats, which increases the opportunities for female candidates.

Electing More Women

121

Incumbency advantages can be lessened in a number of ways. Term limits is one obvious approach, though that has been ruled unconstitutional on the federal level. Another option is full public financing of campaigns, which would even out the gross imbalances between the campaign funds available to incumbents and challengers. But the adoption of proportional representation systems also could lessen the incumbency factor in several ways and thus increase women’s chances of elections. First, gerrymandering in single-member plurality systems is often used to protect incumbents and ensure that they have safe districts in which to run. PR systems largely eliminate gerrymandering, thus undermining this method of protecting incumbents. Second, PR systems encourage more competition within parties, thereby lessening the number of incumbents. For example, incumbents in single-member plurality districts often have virtually no competition for their party’s nomination, but in STV and open party list systems the voters can effectively rearrange the party list to put newcomers in a better position to be elected. The net effect of heightened intraparty competition is to lessen the importance of the incumbency factor, which can increase the openings for female candidates.25

Which PR System Is Best for Women? While all proportional representation systems give women a better chance to get elected than SMP systems, some are better at it than others. Two factors that tend to affect how well PR systems represent women are the district magnitude and the degree of voter choice among candidates. PR systems that have large district magnitudes—the number of seats elected in each district— seem to work better for women. If many seats are elected in a district—say 7 to 10—then there is a better chance that parties will elect several people on their lists. The more people elected from a list, the further the party goes down on its list, and the more this will increase women’s chances of being elected. The party list versions of PR tend to have the highest district magnitudes and so this form of PR tends to be the friendliest to female candidates. Large districts may also aid female candidates in single transferable vote systems. The more seats contested in a district, the lower the percentage of votes needed by a female candidate to win office. A woman in a ten-member STV district would need to receive only a little more than 9 percent of the vote, while she would need more than 25 percent in a three-member district.26 It is not surprising then that Ireland, which uses STV and small three to five seat

122

Electing More Women

districts, does poorly in electing women compared to many other PR countries. While the STV system is very favorable to women, in Ireland the low district magnitudes tend to offset that advantage. Low district magnitude is also the reason that mixed-member PR systems tend to do more poorly at women’s representation than pure party list systems. Recall that in this system half of the legislature is still elected in single-member districts—which have the lowest possible district magnitude. As seen earlier, women tend to be nominated and elected in lower proportions in these districts. So while MMP systems still far outpace U.S.-style SMP systems in the representation of women, they do tend to lag behind party list forms of PR. Another factor that affects how well proportional representation systems represent women is the degree of choice among candidates that is given to the voters. For example, there are two kinds of party list system: open and closed. In closed party list systems, the parties choose the order of the list of candidates and it remains unchanged. So if a party ends up winning two seats in a district, the first two people on the party-determined list are elected. In open list systems, voters are able to express preferences for particular candidates on the list and this is used to change the order of the candidates. So if many voters expressed support for a particular female candidate, this would move her up on the list, thus increasing her chances of being one of the candidates to win a seat. In societies where there is substantial support for female candidates, open lists would seem to work in the favor of greater representation for women.27 The single transferable vote is another form of PR that allows voters to pick individual candidates, rather than endorsing a set list. Voters who prefer female candidates can rank them higher in STV systems, thus increasing their chances of election. In general, then, if one is designing a proportional representation system for United States elections that would encourage fairer representation of women, several strategies should be adopted. First, one should choose a version of PR that uses large multimember districts, and avoid versions that elect only a small number of seats in each district—say three to five. Second, it is better to have a PR system that allows voters to cast their ballots for specific female candidates, such as open party list PR or the single transferable vote.

Growing Support Among Women’s Organizations Clearly, the adoption of proportional representation in the United States could greatly increase the number of women elected to office. Indeed, PR is

Electing More Women

123

the one political reform most likely to quickly and dramatically increase the number of female officials elected in this country. Given this, we would expect the adoption of PR to be high on the political agenda of progressive women’s organizations. Yet feminists have begun to promote this reform only relatively recently. In the 1960s and 1970s, many feminists assumed that the emergence of strong feminist organizations and an increased public focus on feminist concerns would naturally lead to more women being elected. But this did not occur. In the 1980s, it became clear that a much more active approach was needed and greater efforts were made to recruit and fund female candidates. Organizations like the National Women’s Political Caucus, the Women’s Campaign Fund, the Fund for the Feminist Majority, and Emily’s List began to put considerable energy into electing more women. This produced ever-increasing numbers of women in office during the 1980s and 1990s, but progress has been frustratingly slow and the total number of women remains far below desirable levels. At the current rate of growth, for example, it will take well into the twenty-second century for women to finally occupy 50 percent of the seats in Congress. This frustrating situation has lead some feminists to explore new approaches to the problems of women’s underrepresentation—approaches that clearly lead in the direction of proportional representation. For example, some feminists have proposed that the Republican and Democratic parties adopt minimum quotas for female candidates—such as the 30 percent or 40 percent levels required by some parties in Europe. Such quotas could indeed be useful; but as seen earlier, they are difficult to impose if candidates continue to be chosen individually in local areas, as is now the case with our SMP elections. Quotas work best in proportional representation systems where parties offer slates of candidates in multimember districts, so reformers pushing for quotas should also be pushing for PR. Other feminists have given up on the major parties altogether and have urged the formation of a new party that would be more sensitive to women’s issues and would offer more women as candidates. Perhaps even the threat of feminists dropping out of the Democratic party would be enough to finally get the Democrats to nominate more female candidates. But again, this third-party approach has little chance of success within the framework of single-member plurality elections. A feminist party, like any minor party, would rarely attract enough votes to win any seats. And as long as we have a two-party system, leaders of the Democratic party know that feminists really have nowhere else to go. A third-party strategy could only be successful in a proportional representation system, where minor parties are viable and have

124

Electing More Women

a good chance of consistently winning seats. When a women’s party formed in Iceland, for instance, PR rules allowed it to begin immediately to elect women to national office. Increasingly, those who are interested in increasing women’s representation have begun to take a careful look at the impact of electoral systems and the need for electoral reform. For instance, several women’s organizations in the U.S. have begun to openly explore the PR option. The Feminist Majority has been studying the political impacts of electoral systems, and its president, Eleanor Smeal has been an active supporter of PR for years. The National Organization for Women has also acknowledged the importance of electoral reform and has endorsed the use of PR for U.S. elections. According to NOW President Patricia Ireland, “We are tired of a government that purports to represent all of the people, but resembles a corporate boardroom. That is why NOW is initiating an education campaign on reforming our winner-take-all voting system in order to elect more feminist candidates to office.”28 NOW also allows its regional caucuses to use proportional representation to elect their representatives to its national board. This growing interest in proportional representation by women’s groups in the U.S. mirrors a similar development on the international level. The United Nation’s Fourth World Conference on Women in 1995 approved a platform plank urging all governments to “review the differential impact of electoral systems on the political representation of women in elected bodies and consider, where appropriate, the adjustment or reform of those systems.” More recently, June Zeitlin, executive director of the Women’s Environment and Development Organization, has pointed out that “women are grossly underrepresented at all levels of government worldwide. However, women fare significantly better in proportional representation electoral systems. . . . The 10 countries with the highest percentage of women in parliament have systems that include proportional representation.”29 Such developments signal a growing realization that proportional representation should be included as part of the overall strategy to increase women’s representation and that it needs to become a higher priority on the political reform agenda of women’s organizations. Clearly, the removal of the unfair barriers to female candidates contained in our single-member plurality system would be a crucial step in the struggle to ensure that women move quickly to occupy their rightful places in the halls of power in America.

6

Fair Representation for Racial and Ethnic Minorities

The lack of fair political representation for racial and ethnic minorities is one of the most persistent problems plaguing American democracy. We are an increasingly multiracial and polyethnic society, with nearly one third of the country now being made up of African Americans, Hispanics, and Asian Americans. But how to ensure these groups their fair share of representation has remained a vexing political problem. It has been the topic of numerous legislative battles and a seemingly endless series of state and federal court cases. Some progress has been made during the last 50 years, but the fact remains that these minorities are still underrepresented in most of our local, state, and federal legislatures. For many years, the main problem was racist barriers to voting that kept many minorities from exercising their right to vote. Today, however, much of the problem lies in the plurality voting systems that we use in the United States—systems that create another barrier to fair representation for minorities. As we will see in this chapter, trying to ensure fair minority representation within the constraints of plurality electoral systems has proven to be an almost unsolvable political puzzle, and it is this frustrating situation that has prompted increasing numbers of people to realize that proportional representation may ultimately be the simplest and most effective solution to this lingering problem.

126

Fair Representation for Minorities

The Problem If African Americans had representation in the U.S. House of Representatives in 2002 equal to their 12.9 percent of the population, they would have had 56 seats, instead of only 37. African Americans held no U.S. Senate seat in 2002, when they should have held 12 seats. Hispanics accounted for almost 12.5 percent of the total U.S. population in 2002, yet they occupied only 4.4 percent of the seats in the House and none in the Senate. Asians were also underrepresented in the 2002 Congress, occupying only four seats in the U.S. House, when they deserved 16 seats based on their portion of the population. The situation on the state and local levels is often not much better. For instance, as table 6.1 illustrates, the number of black elected officials in states with large African-American populations typically falls far short of the number of the positions they deserve based on their share of the voting age population. Increased representation of racial minorities would be desirable for several obvious reasons. First, increased representation means minority communities can better promote their political and economic interests and focus more attention on what they see as pressing issues. Historically, all-white or predominantly white legislatures in the United States have often contributed to the oppression of racial minorities or at least paid little attention to their problems. Larger numbers of minority representatives would amplify their political voices and increase the chances that our laws would reflect the needs of all citizens, not just a white majority. Imagine, for example, if there were 25 African-American and Latino members of the U.S. Senate. It would clearly have an affect on the way that institution approaches many policy issues, including such things as civil rights, immigration, urban renewal, affirmative action, and social welfare spending. The notion that fair levels of minority representation can further minority interests is supported by a number of studies that show that increased black representation in city governments has made these institutions “more attentive and responsive to their needs for basic community services.”1 AfricanAmerican elected officials have also been particularly effective in increasing the number and quality of municipal jobs held by blacks.2 Increased minority representation not only allows minorities to push for larger portions of the public pie but also serves as a check on prejudice and racism on the part of public employees.3 And as Grofman and Davidson have observed: “Quite simply, the presence of minority officeholders makes it harder for racism to persist inside a legislature.”4

127

Fair Representation for Minorities

table 6.1 Percentage of Black Elected Officials by State (1997) State

Blacks as a % of Voting Age Population

% of Black Elected Officials

Alabama

23.9

16.6

Arkansas

14.3

5.8

Delaware

18.0

2.1

Florida

13.4

3.9

Georgia

26.5

8.9

Illinois

13.9

1.3

Louisiana

29.6

12.8

Maryland

26.6

9.2

Michigan

13.2

1.8

Mississippi

33.3

16.9

Missouri

10.1

1.1

New Jersey

13.6

2.5

New York

16.8

1.2

North Carolina

20.5

8.7

Ohio

10.5

1.2

South Carolina

27.8

13.7

Tennessee

15.0

2.5

Texas

11.8

1.6

Virginia

19.0

10.7

Source: David Bositis, Black Elected Officials: A Statistical Summary 1993-1997 (Washington, D.C.: Joint Center for Political and Economic Studies, 1998), p. 10.

The election of more minority representatives probably would also serve to increase minority political awareness and political participation. Election campaigns often are important avenues for political organizing, and the presence of minority candidates can increase the level of political debate within minority communities. Successful minority candidates can

128

Fair Representation for Minorities

have other beneficial effects on the civic life of minority communities. In their study of American cities, Bobo and Gilliam found that “where blacks holds positions of political power, they are more active and participate at higher rates than whites of comparative socioeconomic status. . . . [Also this] empowerment leads to higher levels of political knowledge and it leads to a more engaged (i.e. trusting and efficacious) orientation to politics.”5 Another study, by Banducci, Donovan, and Karp, produced similar findings. They found that in areas with African-American elected officials, minority citizens were more likely to see government as responsive to their interests.6 Fair levels of minority representation, then, could help to address the high levels of political alienation that exist in many minority communities today. This alienation has played a significant role in many incidents of urban unrest and rioting. Clearly, the election of more minority representatives could help to increase the legitimacy of governmental bodies in the eyes of these communities. The need to increase the perceived legitimacy of our legislative institutions may be even more acute today, when one traditional avenue for minority redress of racial grievances, the judicial system, grows increasingly unwilling to aggressively pursue civil rights remedies. If a hostile judiciary is combined with legislatures that continue to grossly underrepresent racial minorities, this will cut off these legitimate avenues of political participation and only serve to increase the feelings of frustration and political alienation felt by many minority citizens.

The Electoral Connection Unquestionably, the consistent underrepresentation of racial and ethnic minorities in American politics is historically connected to a series of deliberate efforts to diminish their voting power. Often those efforts took the form of disenfranchisement: attempts to discourage or prevent minorities from voting. Before the Civil War most African Americans were slaves and prevented by law from voting. After the war whites used terrorism and violence to keep blacks away from the polls. Around the turn of the twentieth century, more subtle discriminatory measures became popular, including poll taxes and literacy tests. More recently, techniques such as difficult registration procedures, changing polling places on short notice, and dropping people from the registration rolls have been used to discourage minority electoral participation.

Fair Representation for Minorities

129

During the last thirty years, as many of the barriers to registration and voting have been gradually dismantled, the number of minorities registering and voting has increased. At first, the number of minority representatives also increased dramatically. The total number of black elected officials in the United States grew an average of 16.7 percent a year between 1970 and 1977, from 1,469 to 4,311. In 1981, the number reached 5,038, a little over 1 percent of the elected positions in the United States.7 But those increases have slowed since then. In 1999, despite yearly increases, the total number of black officials had grown to only 1.7 percent—in stark contrast to their 12.9 percent of the population.8 But if the electoral barriers are down, why do African Americans continue to be so underrepresented? The answer is that not all the barriers are down. Our current electoral system has built-in barriers that ensure that while minorities can vote, they often cannot cast an effective vote—one that elects a minority representative. Minorities have the right to vote, but our voting procedures undermine their right to representation. So while minorities have largely won the first stage of the struggle for political power—winning the right to vote—the second stage of that struggle continues today: the need to create a voting system that allows them a fair chance to elect their own representatives.

At-Large Elections and Racial Representation Two kinds of plurality voting systems are used in the United States—atlarge and single-member district—and both can serve to underrepresent minorities. In at-large elections, which are used almost exclusively at the local level, officials are elected in multimember districts with voters having as many votes as there are seats. Thus in a typical at-large election of nine city council members, the entire city is one district in which all the candidates run, and each voter is able to cast nine votes. (Though this type of election uses multimember districts, it should not be confused with proportional representation systems that also have multimember districts. At-large elections are essentially just a combination of single winner-take-all elections.) Such elections discourage the election of minority representatives in several ways. First, the necessity for candidates to run campaigns in large areas works to the disadvantage of candidates from low-income and working-class constituencies. In an urban area, a candidate would need the funds and organization to run a citywide campaign in order to have any hope of winning.

130

Fair Representation for Minorities

Often minority candidates do not have large campaign war chests and so are handicapped in such races. Far more important is that the system of voting in at-large elections often makes it virtually impossible for minority candidates to get elected. At-large elections require candidates to get a majority or plurality of the vote to win, which usually allows a white majority to take all the seats and to deny minorities any representation at all. For example, even if African Americans were a substantial minority—say 30 percent of the electorate—the whites could cast their 70 percent of the votes for each of the white candidates and sweep the election. A classic example of this problem occurred in a municipal election in Fairfield, Alabama. The city had a version of the at-large system: It was divided into six districts, with two council members elected from each, but all were voted on at large. One year the contests for nine of the twelve seats for city council had both white and black candidates. In those races, 42 percent of the votes went to the black candidates, but none was elected. The 58 percent white majority took 100 percent of the seats.9 Thus at-large elections are a subtle but effective form of political discrimination. This system allows minorities unimpeded access to voting and a clear road for minority candidates to run for office, but it stacks the election process so that they can rarely win. At-large elections became popular in American cities around the beginning of the twentieth century. Cities often adopted them as part of a set of political reforms championed by municipal reformers and later the Progressives. The reform packages usually also included the adoption of the city manager form of government and the use of the nonpartisan ballot. These reformers claimed that council members elected at large would be better qualified and would make decisions based on what was good for the city as a whole, rather than what would benefit one district or group. It was also seen as a way to undermine the power of corrupt urban machines. Most historians now agree, however, that another purpose underlay the support for at-large elections: to minimize the power of racial and ethnic minorities in urban politics.10 Many northern Progressives had become increasingly anxious as millions of immigrants flooded their urban centers, eager to exercise their newfound opportunity to vote. The Progressives, typically from the ranks of professionals and the upper-income levels, began looking for ways to minimize the role of what they termed “uneducated” and “morally inferior” ethnic groups and their political representatives. At-large elections seemed to fit the bill. In the South, Progressivism was heavily tinged with racist appeals. Many of the movement’s leaders advocated the disenfranchisement of blacks, and some saw at-large districts as a second line of defense against potential black voting power.11

Fair Representation for Minorities

131

The second great wave of cities and states adopting the at-large system was in the 1950s and 1960s, another period of heightened racial tensions. The civil rights movement was growing in power and influence, and increasing numbers of African Americans were registering and voting—trends that were undoubtedly seen as threatening by some white political elites. One response to the enactment of the federal Voting Rights Act of 1965, a law designed to eliminate racial barriers to registration and voting, was an especially large surge in adoptions of at-large election systems. In Georgia, for instance, 20 county governments and county boards of education switched to at-large election systems between 1964 and 1975. In its first session after enactment of the Voting Rights Act, the Mississippi legislature passed a law that required or allowed county boards of supervisors and school boards to be elected in at-large contests.12 Considerable evidence exists that these at-large electoral arrangements work to the disadvantage of minorities. Numerous studies of cities that use at-large elections have shown that they have the worst rates of minority representation.13 Berry and Dye, for example, found in their analysis that the use of at-large elections was “the single most important variable” in explaining the underrepresentation of minorities on city councils in the United States.14 The blatantly discriminatory nature of these elections eventually prompted the federal government and voting rights lawyers to mount political and legal challenges to them. In 1982, Congress amended the Voting Rights Act to prevent what it called “minority vote dilution.” Among the kinds of vote dilution mentioned was an electoral arrangement where minority voters have “less opportunity than other members of the electorate . . . to elect representatives of their choice.”15 In the decades after the passage of these amendments, hundreds of federal court cases were brought against at-large election systems. In one that eventually made its way to the Supreme Court—Thornburg v. Gingles—the court ruled that atlarge systems did indeed produce minority vote dilution and ordered as a remedy a change to single-member district plurality elections.16 But turning to single-member districts was to prove to be far from a final solution to this problem.

Single-Member Districts and Racial Representation At first glance, the use of single-member districts would seem to offer the potential for some minority representation. If minorities make up the majority

132

Fair Representation for Minorities

of citizens in such a district, they stand a very good chance of electing one of their own to office. But in reality, most minority voters found themselves submerged in predominately white districts, and thus unable to elect their own representatives. Worse yet, the use of single-member districts allowed politicians to manipulate district lines to ensure the underrepresentation of minorities. As seen in chapter 2, gerrymandering has often been used for partisan purposes, to hurt particular parties, but it has also been used to hurt particular racial groups as well—a process known as racist gerrymandering. Racist gerrymandering can be accomplished in several ways, all of which are essentially similar to the forms of partisan gerrymandering seen earlier. In one form, called “cracking,” lines are drawn through areas of minority concentration in order to break the population up among several districts. This ensures that minorities will be unable to elect any representatives. One particularly clear example occurred in Mississippi, where the black population is concentrated in the northwestern section of the state, an area known as “the Delta.” Between 1882 and 1966, when blacks were unable to register and vote, the Delta was always contained in one congressional district that was 66 percent black. In 1965, after the enactment of the Voting Rights Act, the Mississippi legislature redrew the congressional district lines in a way that broke the Delta area up among four other districts, all with white majorities—an arrangement that existed into the 1990s.17 Another common form of racist gerrymandering is “packing.” Minorities who might constitute a majority in several districts are all put into one district in which they make up 80 to 90 percent of the vote and thus get only one seat. Because of the growing population in south Texas, the state legislature had to redraw the congressional district lines after one census to add one district there. Mexican Americans enjoyed a 65 percent majority in two of the preexisting districts and consistently elected Mexican Americans to Congress. The legislature could easily have created a third district in which Mexican Americans had a substantial majority. Instead, legislators drew the new lines so that Mexican Americans were packed into one district in which they made up 80 percent of the population, in order to prevent them from having a viable majority in the new district. After the U.S. attorney general objected to the scheme, a U.S. district court ordered the implementation of another redistricting plan that created three districts in south Texas with Mexican-American majorities of 72 percent, 71 percent, and 64 percent.18

Fair Representation for Minorities

133

Even more subtle forms of racist gerrymandering can take place, such as when lines are drawn to create the illusion of districts with black majorities. Sometimes lines are drawn to create black voting majorities that are very thin—say, 54 percent. While such districts are considered black districts, in practice white voters may control them. Consider Hinds County, Mississippi. In one case, the board of supervisors used cracking to create five districts, all with white majorities. When that was found to be discriminatory, they redrew the lines so that African Americans had a slight majority in two districts, 54 percent and 53 percent. The board argued that this plan was not discriminatory because it provided for the two black majority districts. However, a closer look at the census data revealed that in those two districts 68 percent of the white population was of voting age, compared to only 55 percent of the blacks. Thus while blacks had a majority of the total population, whites had a majority of the voting age population and therefore retained effective political control.19 As these examples illustrate, single-member districts are hardly an automatic solution to the problem of minority underrepresentation. The only way this electoral system can help minorities is if an explicit effort is made to create districts in which they are truly in the voting majority, and so are able to win some seats. This process is called “race-based districting” or “raceconscious districting” and it aims to create “majority-minority” districts. For a while it seemed that this approach might be the answer to the problem of minority representation.

The Rise and Fall of Race-Based Districting Beginning in the 1980s and accelerating through the early 1990s, the U.S. Department of Justice began to push majority-minority districts as the preferred remedy in voting rights cases. They required a number of states to redraw their district lines to produce several districts in which minorities would have the majority of the votes. This effort proved successful in increasing the number of minorities elected to office. For example, between 1990 and 1992, the number of Latino and black members of the U.S. House of Representatives jumped from 35 to 55. It appeared to those in the voting rights community that the country was finally heading down a path toward fair representation for minorities. But this enthusiasm was short lived.

134

Fair Representation for Minorities

Considerable political opposition had been mounting to race-based districting. Some whites labeled it a form of reverse discrimination, and saw it as an unfair form of political affirmative action. This political backlash prompted a flood of court challenges to this practice after the 1992 elections, including suits in North Carolina, Texas, Louisiana, Florida, New York, Virginia, Georgia, Illinois, South Carolina, and Alabama. The first such case to reach the Supreme Court was Shaw v. Reno.20 That case concerned redistricting in North Carolina, which had a 22 percent black population but had not sent a black to Congress since 1901. In redistricting following the 1990 census, North Carolina created two black majority-minority districts. One of these, the 12th district, was challenged in the Shaw case. In its decision, a sharply split Supreme Court ruled that this district was unconstitutional. Much of the court’s criticism focused on the “bizarre” shape of the district—a strangely elongated district that wound through 10 counties connecting various small black communities. Justice Sandra Day O’Connor, writing for the majority, ruled that this district violated the equal protection clause of the constitution. She argued that by including in this district “individuals who belonged to the same race, but who are otherwise widely separated by geographical and political boundaries, and who may have little in common with one another but the color of their skin bears an uncomfortable resemblance to political apartheid.”21 There followed a number of further rulings that increasingly questioned the constitutionality of race-based districting.22 Various federal courts struck down majority-minority districts in New York, Georgia, Texas, Virginia, and Louisiana. However, the Supreme Court seemed unable to offer a consistent rationale for why these districts were a problem. In Miller, for instance, the Court moved away from a focus on the shapes of these districts to a concern that race should not be the “predominant” factor in the drawing of district lines. It also ruled that there is not a compelling state interest to force states to comply with the amendments to the Voting Rights Act. In 2001, in Easley v. Cromartie, the Supreme Court shifted its reasoning once again and ruled that a North Carolina district that was 47 percent black was constitutional because race was not the “dominant or controlling” factor in designing that district. This seemed to signal that while the Court remained hostile to intentionally created majority-minority districts, it would be willing to uphold some minority-dominated districts that were created for other reasons—such as partisan advantage. However, this approach appeared to merely invite more legal challenges over the exact extent that race plays a factor in drawing district lines.

Fair Representation for Minorities

135

The Weak Case Against Majority-Minority Districts Upon closer examination, it is clear that most of the political and legal arguments against majority-minority districts are questionable. For example, in the past the Court has not questioned the legitimacy of oddly shaped districts created by partisan gerrymandering. District shape seems to be a constitutional issue only when minorities are involved. In Bush v. Vera, for instance, the Court ruled unconstitutional several majority-minority districts in Texas— shown in figure 6.1. But left unchallenged, and tacitly accepted by the Court, were similar bizarrely shaped districts designed to protect white incumbents— shown in figure 6.2. If strangely shaped districts are acceptable for white majorities, why are they not acceptable for African-American or Latino majorities? Also, the Court was simply mistaken to liken majority-minority districts to apartheid, where blacks were segregated off from whites. In reality these districts were not segregated at all, but were the most racially integrated districts in the country. Far from being all black, the majority-minority districts created in the South after the 1990 census contained an average of 45 percent nonblack voters—much more integrated in comparison to the white-majority districts.23 For this reason, Justice Stevens concluded that black-majority districts were in fact a form of “racial integration.”24 Equally suspect are contentions that race-based districting presumes that only minorities can represent minorities or that it is a form of affirmative action. The ultimate point of race-based districting is simply to ensure that minority voters can elect whomever they choose—whether that candidate be white, black, liberal, or conservative. This approach says nothing about who should or should not be elected. Neither is race-based districting a kind of

figure 6.1 Texas Majority-Minority Districts Declared Unconstitutional

136

Fair Representation for Minorities

figure 6.2 Texas Majority White Districts Left Unchallenged

political affirmative action plan that gives preference to black politicians over whites. No candidates are screened out based on race—all are free to run; and voters are free to elect anyone they want. So there is virtually no real resemblance to affirmative action. As Justice White concluded in his dissent in Shaw, “efforts to remedy minority vote dilution are wholly unlike what typically has been labeled ‘affirmative action.’ ”25 Critics of race-based districting also often argue that we no longer need such measures, that we have reached a stage where many whites will vote for black candidates, and they cite a few examples of where blacks have been elected in white districts. But in many of those cases, those candidates were first elected in majority-black districts. This gave them the opportunity to establish a track record of service to constituents of both races, and to use the considerable advantages of incumbency to continue to be elected.26 As Handley and Grofman have concluded, “the increase in the number of blacks elected to office in the South is a product of the increase in the number of black-majority districts and not of blacks winning in majority-white districts.”27 Moreover, research has consistently found that in many regions, racially polarized voting continues and that whites are still very unlikely to elect blacks to office.28 Nationwide, only about 1 percent of black elected officials come from white districts. There is little evidence then that race-based districting isn’t needed. However, the ultimate problem with these attacks on majority-minority districts is that the alternative offered by these critics would be a large step backwards in the struggle for fair representation. Critics advocate a return to “raceneutral” districting, where districts are drawn without any consideration of race. Race, they argue, should have no place in districting decisions.29 But the results of this approach are anything but race-neutral. Given the continued ex-

Fair Representation for Minorities

137

istence of racially polarized voting, this approach would inevitably re-create the predominantly white districts in which minorities have little or no chance of winning any representation. Bernard Grofman has shown that even if we were to draw districts with no racial or political criteria whatsoever—if we had computers draw districts completely at random—many minorities would end up submerged in white districts and this would inevitably result in their underrepresentation.30 Ironically, then, if we follow the Court’s prescription to be “color-blind” in districting, this will discriminate against minority voters and allow whites to once again maximize their political control of legislatures. So even if race-based districting has some problems, a return to whitedominated districts would be much worse. And if the choice were just between those two approaches, race-based districting is clearly superior. Even if flawed, it will work to achieve a crucial political goal: to improve the chances of minorities to elect some of their own candidates. Unfortunately, this approach seems to be meeting a dead-end, given the Supreme Court’s hostility to it. And so as a nation we now face a series of seemingly intractable dilemmas. For example, the Department of Justice requires many state governments to abide by the Voting Rights Act and create majorityminority districts—and yet the Supreme Court seems inclined to strike many of them down. And it remains unclear whether and to what extent state legislatures can take race into account in districting. What is a state government to do in this situation? And there is also the more basic dilemma: We are increasingly blocked from creating race-based districts, and yet we would not want to return to the old days of virtually all white legislatures elected from predominantly white districts. Fortunately, there is a way out of these quandaries. But the solution requires a realization that these dilemmas exist only because we are trying to work within the single-member district system. If we begin to “think outside the box,” and consider other electoral alternatives, like proportional representation, then we can escape these problems. Proportional representation can ensure that minorities achieve fair representation without having to fashion the special majority-minority districts that have drawn so much political opposition.

PR—The Simplest Solution Proportional representation is the best solution to the problem of minority representation. This doesn’t mean that race-based districting is a bad

138

Fair Representation for Minorities

idea, but only that PR is a better one. It is not surprising that PR would solve this problem. It was designed explicitly to ensure fair representation for both majorities and minorities. In fact, one of the primary reasons PR was adopted by most European countries in the early part of the twentieth century was to ensure a fair share of power for their ethnic minorities. As Arend Lijphart has observed, a key factor in this process of electoral reform “was the problem of ethnic and religious minorities; PR was designed to provide minority representation and thereby to counteract potential threats to national unity and political stability. ‘It was no accident,’ Stein Rokkan writes, ‘that the earliest moves toward proportional representation came in the ethnically most heterogeneous countries.’ ”31 In proportional representation, the key to ensuring fair representation for minorities is the use of large multimember districts instead of single-member districts. In multimember PR districts, a racial or ethnic minority would need only a small portion of the vote—say 10 to15 percent—to ensure the election of at least one of their candidates. In many urban areas in the United States the entire city would be one large district for the purposes of council elections. In state or federal elections several large multimember districts would elect legislators. These multimember districts allow representation of racial and ethnic minorities without the need for race-based districting. Consider, for example, a city in which all nine city council members are elected by single transferable vote PR in one citywide district. Under this arrangement, it would take only about 10 percent of the vote for a group to elect a representative to the council. While African Americans might constitute only 15 percent of the city’s population, they would be able to elect a representative easily—even if they were dispersed widely throughout the city. In Cambridge, Massachusetts, which uses proportional representation to elect its city council, a black has won office in every election in the last twenty years, even though this minority makes up only about 12 percent of the city’s population. Minority candidates can run in PR elections in several ways. They can be part of the major parties’ slates. As in the case of women candidates, these parties would find it in their interests to include members of significant racial and ethnic groups in order to increase the appeal of their slates. Any party that refused to include such candidates would be open to charges of racism. Also, PR would give minority communities the opportunity to create their own parties and run their own slates of candidates. We could imagine, for example, an African-American party or a Latino party or an Asian-American

Fair Representation for Minorities

139

party in areas with large concentrations of these minorities. Finally, minority candidates can also run as independents in single transferable vote PR elections, and succeed with a relatively small percentage of the vote. Proportional representation also ensures fair representation by eliminating the possibility of the racist gerrymanders that have disempowered minority communities in the past. As we saw in the discussion of partisan gerrymandering in chapter 2, if multimember PR districts are large enough—at least five or six seats—gerrymandering ceases to be a real possibility. Thus, under PR rules, any effort to divide minority populations to diminish their voting power would be useless. For example, if a state used ten-seat districts to elect its legislators, it would make little difference whether a minority population was packed into one district with an 80 percent majority, or whether it was split into two districts in each of which it would have only 40 percent of the electorate. In both cases, the minority voters would be able to elect eight legislators, either eight of the ten representatives in the one district or four of the ten representatives in each of the two districts. Thus if the United States adopted PR, we would no longer have to worry about who was drawing the district lines and what their intentions were—because they would have little effect on racial representation. Proportional representation would also eliminate the political grievances that have fueled the intense political conflicts and legal battles surrounding race-based districting. Accusations by whites that majority-minority districts are reverse discrimination would become moot. Also, whites would no longer feel the frustration of wasting their votes when they vote in majorityminority districts. Indeed, PR has the advantage of assuring fair representation for both minorities and whites. Single-member districts always divide the population by producing winners (the majority) and losers (the minority). And the losers, whether they are racial minorities or whites, always feel like political victims. In the large multimember districts used in PR, no groups are losers—all groups can win their fair share of representation. PR is a “win-win” political game, and thus should garner the support of whites and minorities who believe in fair representation for all. Given all these advantages of proportional representation, it is not a surprise that many electoral scholars now recommend these systems for multiracial or polyethnic societies. Andrew Reynolds is an electoral systems expert who has studied the racially and ethnically divided societies of southern Africa and concluded that the ideal election system for these countries would be some form of proportional representation. He further argued that

140

Fair Representation for Minorities

“straightforward majoritarian institutions are, on the whole, inappropriate for any plural society in the developing world.”32 Arend Lijphart has also concluded that PR systems can be particularly useful in societies experiencing deep racial and ethnic conflicts: “Divided societies, both in the West and elsewhere, need peaceful coexistence among the contending ethnic groups. This requires conciliation and compromise, goals that in turn require the greatest possible inclusion of representatives of these groups in the decisionmaking process. Such power sharing can be arranged much more easily in . . . PR systems than in . . . plurality systems.”33

The Advantages of PR over Majority-Minority Districts Even if the Supreme Court were to reverse itself and come to clearly endorse race-based districting, there would still be many reasons to see proportional representation as the preferred approach. Proportional representation has several important advantages over majority-minority districts. As noted above, since everyone gets fair representation, PR generates less political resentment. Also, the use of PR would avoid many of the practical problems associated with majority-minority districts. For example, one major limitation of race-based districting is that it works well only where minority populations are concentrated in small geographic areas—such as particular neighborhoods in a city. Such geographical concentrations are necessary to create districts in which racial and ethnic minorities have clear majorities. But it becomes much harder to find such areas above the city level. At the level of state and federal elections, with their larger election districts, it becomes increasingly difficult to create districts in which minorities dominate.34 That is why single-member districts have worked much better for increasing black representation on city councils than in state legislatures and Congress. Even on the city level affirmative gerrymandering does not always work. It only functions well in cities in which the minority population is geographically segregated from the white population. Only then is it possible to construct districts in which minorities constitute a majority. If that population were to be integrated throughout the city, its members would become minorities in all districts, and their voting strength would be greatly diluted and largely ineffective. That is one reason why the use of single-member districts has not helped Asian and Hispanic voters as much as it has helped

Fair Representation for Minorities

141

African Americans.35 Asian populations in the Midwest and Latino populations in the Southwest and West are distributed more evenly than the black population. This dispersed residential pattern often makes it difficult to draw majority-minority districts for these groups. Ironically, then, race-based districting seems to require housing segregation in order to produce better representation. In contrast, in a system of proportional representation we could have both housing integration and a fair representation of racial minorities. The whole point of PR is to allow representation of political minorities, no matter where they might live. For example, in a single transferable vote system in which nine city council members were elected, it would take only about 10 percent of the vote to elect a representative. Thus while Latinos might constitute only 15 percent of the population of a city and might be dispersed throughout a variety of neighborhoods, they could still vote together for a Latino representative and get that person elected. Raced-based districting is also difficult in areas with more than one minority. New York City, for instance, hosts a number of different racial and ethnic groups. Many are living in such close proximity that it becomes problematic to draw districts that represent them all. Inevitably, some groups get left out. In 1974 Hasidic Jews in the Williamsburg section of Brooklyn challenged in court a state redistricting plan designed to create several predominantly African American and Hispanic assembly districts. Before the redistricting the Hasidim had all been located in one district. The new plan divided them between two different districts. They argued that the situation unfairly lessened their voting strength and political influence and that it was unjust to disempower one minority in order to empower another.36 A similar situation arose after the 1991 redistricting in New York City. This time it was the Hispanic community that claimed that the new districts created AsianAmerican districts at the expense of Latino voters. These cases illustrate how intractable the redistricting process becomes when an area is racially and ethnically heterogeneous. This is not a rare situation: 50 cities with populations over 100,000 now have at least two ethnic groups that make up more than 10 percent of the population. When many different minority groups overlap in a city, it becomes extremely difficult for even the most wellintentioned political cartographer to ensure that all are represented fairly. Such problems are easily solved with the large multimember districts used by proportional representation. PR allows all significant minority groups in a city to win representation.

142

Fair Representation for Minorities

Proportional representation also avoids several other practical problems of race-based districting. For instance, how large must the minority majority be to ensure political control over a district? Some say just over 50 percent, while other insist that a figure of well over 60 percent is necessary because of the low turnout of minorities.37 PR eliminates the need to make such problematic calculations. In addition, in race-based districting, it is often difficult to decide whether it would be more in the interest of minorities to create a single district in which they have a clear majority or to divide them between two districts where they could have a substantial influence on who gets elected. It may not be clear which would best maximize minority political power. Creating one majority-minority district might guarantee some representation, but it could drain so many minorities from surrounding districts that it makes it more likely that the others will elect more conservative white candidates. Indeed, there is a debate over whether creating majority-minority districts has inadvertently helped the Republican party in this way.38 Because of this possibility, some argue that minorities would be better off being large minorities in several districts where they could affect the balance of power and help determine who gets elected. In any case, the use of PR eliminates the need to make this difficult political calculation. PR allows minorities to have their fair share of political power irrespective of whether they are the majority or the minority in the multimember PR districts. On the other hand, one possible advantage of black majority districts over proportional representation is that those districts are “safe” ones where the incumbent is usually assured of reelection. This means that even if the turnout of minorities is low, they will still elect a representative. In contrast, under PR rules, low turnout by minorities would hurt their chances of winning a seat in a multimember district. But as Lani Guinier points out, the absence of safe seats would actually be an advantage of PR because minority incumbents would no longer be able take their constituents for granted. Proportional representation, she argues, encourages a more active and healthy interrelationship between politicians and their constituents. As she explains: Because [proportional] representation, unlike the districting strategy, depends on high voter turnout, incumbents will find it necessary to mobilize voter interest and participation in the election, a task that will require incumbents to develop substantive programs and proposals. For their part, minorities may participate in election campaigns in

Fair Representation for Minorities

143

greater numbers than they do in single-member district, because election will no longer be zero-sum solutions for minority interests . . . . In addition, after the election, [they] may be sufficiently mobilized to monitor the legislative activity of their elected representatives . . . . Therefore, under [proportional] representation, voters may be more involved in the political process, at both the pre- and post-election stages, than are voters under a single-member district scheme.39 Another advantage of PR is that it allows voters to ultimately decide how they will be represented. All single-member district systems, including racebased districting, give politicians the power to decide how voters will be represented. When districting, politicians must decide whether they will emphasize geographical representation, party representation, or racial representation. Some critics, including Justice Clarence Thomas, disapprove of race-based districting because it presumes that all minorities have similar interests and want to be represented on the basis of race, and so districts are drawn with that in mind. However, in many forms of PR it is the voters themselves who decide how they want to be represented. For example, the single transferable vote allows voters in a multimember district to group themselves together into “voluntary constituencies” based on their common political interests and to be represented on that basis.40 Some voters, for instance, might band together to elect a minority candidate because they all believe that kind of representation is important. Other voters might put a higher priority on geographical representation and so vote into office a candidate from their local area. A black woman might feel it’s most important for women to be better represented, and join with similarly minded voters to pursue that form of representation. In any case, the point is that in singlemember district systems, politicians must decide which kind of representation is going to be privileged; in PR, voters make that crucial decision for themselves.41

Race and PR in Cincinnati The case for PR as the solution to our voting rights problems is not simply a theoretical one; a great deal of information supports the contention that PR assures fair representation for all racial and ethnic groups. In fact, we need only look to how proportional and semiproportional systems have been

144

Fair Representation for Minorities

used in the United States to see its beneficial effects on racial representation. One of the most interesting cases is that of Cincinnati, which used the single transferable vote form of PR for its city council elections from 1925 to 1957.42 Before the adoption of PR, Cincinnati had no black city council members. For sixty years blacks had been a loyal part of the Republican party machine—a party that steadfastly refused to help elect black candidates to city office. After the adoption of PR in 1925, however, the black community realized that the rules of the game had changed and began to come awake as an independent political power. Blacks represented 11 percent of the population of the city, which meant that under PR arrangements they might be able to elect their own member to the nine-member city council. As the Times-Star noted in an editorial: “It cannot be forgotten that we have nine members of Council and about one-ninth of the people of Cincinnati are Negroes. . . . Under the old system, it was almost impossible to elect a Negro to public office in Cincinnati. He would get many votes, but a larger number under normal conditions could be counted on to go against him for racial reasons. PR tends to a different result.”43 An independent black candidate ran for the council in 1925, but many African Americans continued to vote Republican as they always had, and he was easily defeated. In 1927, however, the independent black candidate came very close to being elected, and black political leaders served notice to the Republican party that they could no longer be counted as Republican loyalists. By 1931 Republican leaders had decided that in order to keep the black vote they would have to include a black candidate on their party slate. The support of black voters, along with the help of some white Republican voters, made that candidate, Frank A. B. Hall, the first black council member elected in Cincinnati. Hall was defeated for reelection in 1933, but another Republican black candidate, R. P. McClain, won a council seat in 1935. The trend continued, with blacks serving on 10 of the 15 councils elected during the PR period. Ralph Straetz, who documented Cincinnati politics under PR, noted that while African Americans sat on the city council, Cincinnati initiated a number of reforms in racial policies. By 1945 public park facilities were open to all races. All public hospital and clinic facilities were integrated as well. In the early 1950s black council members led the successful fight for legislation that made it illegal for the Cincinnati Street Railway Company to discriminate against blacks in hiring. A black council member also introduced

Fair Representation for Minorities

145

an amendment to the city plumbers’ code that allowed African Americans to take the examination to become licensed plumbers, an option previously denied them. A black was appointed by the city manager to the Recreation Commission for the first time. All segregated swimming in the junior and senior high schools was discontinued. In 1954 the Cincinnati Civil Service Commission announced that it would drop the racial designation on its application forms and cease to keep dual eligibility lists. While such reforms cannot be solely attributed to the rise of black council members, they were certainly a reflection of the growing political power of the African-American community—a growth that was significantly aided by the adoption of PR. Several attempts were made to repeal PR in Cincinnati. Black political leaders played an increasing role in defending PR from its attackers. In the 1947 repeal attempt the Cincinnati branch of the National Association for the Advancement of Colored People (NAACP) came out strongly in support of retaining PR, noting that without it, black candidates found it nearly impossible to be elected. In that same year a young independent black candidate for council, Theodore Berry, spent much of his campaign defending PR. In his pamphlets he noted that PR allowed black voters to wield the balance of power: The politicians know this and want to destroy the group’s political bargaining power and control the Negro vote. We have only received recognition when it was forced from the political bosses . . . . Without PR our jobs, businesses, unions, homes, and community welfare would be controlled by political bosses. [Blacks would] suffer from poor housing, high rents, low wages, police abuse, vice, segregation, and discrimination.44 The repeal effort that year was defeated. Berry lost the election in 1947, but in 1949 he was nominated as part of the slate of the other major party in Cincinnati, the Charterites, who had finally decided that they had to court the black vote more vigorously. The Republicans also renominated their current black council member, Jesse Locker. Both black candidates won, to the consternation of some white Cincinnatians. Berry and Locker won reelection in 1951. In 1953 President Dwight D. Eisenhower appointed Locker ambassador to Liberia, but Locker’s black law partner, A. Bruce McClure, was elected along with Berry. That election created something of a political crisis. Berry received the largest number of first-ballot

146

Fair Representation for Minorities

choices—more than any of the white candidates. And it was the tradition in Cincinnati that the council member with the largest vote would be elected by the council to serve as mayor. As soon as the election was over, some in the black community demanded that Berry be made mayor—the first black mayor of a major U.S. city. Predictably, many in the white community did not find that an appealing idea. After much political maneuvering, the white council member with the most seniority was chosen mayor, and Berry accepted the chairmanship of the powerful finance committee. Clearly, the political position of the black community and black politicians had changed dramatically between 1924 and 1953. PR created a hospitable environment for the political organization of the black community and made it possible for black leaders to gain positions of power in city government. But for some in the white community PR worked too well in increasing black political influence. The increasing power of black politicians in Cincinnati eventually became an issue to be exploited by those interested in repealing PR. Ralph Straetz describes what eventually happened: On September 30, 1957, after thirty-two years of existence, proportional representation, a strategic bulwark of the continuing reform movement in Cincinnati, was defeated by a heavy majority after an extremely bitter campaign. By a vote of 65,593 to 54,004, PR in Cincinnati has been eliminated from the political scene. The unusually heavy anti-PR vote can be accounted for in large part by a renewed, expanded and effective use of an old ingredient in Cincinnati politics—racial hostility. In Cincinnati as in many other communities the Negro is breaking out of his ghetto. PR has helped him break out politically. Symbolic of the Negro’s effective use of PR to achieve political recognition has been the successful career of Ted Berry, Negro councilman and vice-mayor of Cincinnati. Throughout the recent repealer campaign racial bigots repeatedly raised two questions: Do you want Ted Berry to be mayor of Cincinnati? Do you want Ted Berry for your next-door neighbor? Rumors had Berry buying a home in almost every white neighborhood in the city. In a minor fashion Cincinnati thus shares some of the tragedy of Little Rock, the tragedy lying not in the defeat of PR but in the extensive use of hate to achieve political power.45 The story of proportional representation in Cincinnati is fascinating for several reasons. First, it illustrates how this system can work to the benefit of

Fair Representation for Minorities

147

racial minorities. In this case, the major parties began to see the necessity of including minority candidates on their election slates. The history of PR in Cincinnati also sheds some important light on the mystery of what happened to some of the early experiments with PR in the United States. Almost two-dozen American cities used PR elections in the early part of the twentieth century. But it was eventually abandoned in all but one—Cambridge, Massachusetts. This record is sometimes used to argue that PR has been tried in the United States, but simply did not work. But as the Cincinnati case illustrates, proportional representation was sometimes abandoned not because it failed but because it worked. PR worked to give Cincinnati blacks their first real taste of political power, which eventually created a political backlash that sealed the fate of PR in that city.

Semiproportional Representation and Minority Representation in the U.S. The use of semiproportional systems in the United States provides more contemporary evidence of how proportional representation can produce more racially balanced political institutions. Currently, more than 100 local jurisdictions use either the cumulative vote or the limited vote for such things as city councils, county councils, and school boards. One particularly revealing example is that of Chilton County, Alabama.46 In the 1980s Chilton County was sued on the grounds that its use of at-large plurality elections for the county commission and the county Board of Education discriminated against black voters and denied them fair representation. The county settled the suit by agreeing to use a method known as the cumulative vote (CV) to elect members of these bodies. Cumulative voting is usually classified as a semiproportional form of elections. It usually yields more proportional results than regular at-large elections, but may fall short of the high degree of proportionality found in most other forms of PR. In cumulative voting, representatives are elected in multimember districts, with voters allowed to cast all their votes for one candidate or to split them among several candidates. Thus if five members were being elected, a voter could cast all five votes for a favored candidate, or one vote for each of five candidates, or two votes for one candidate and three for another, or any other combination. The unique arrangement makes it possible for relatively small minorities of voters to elect representatives by concentrating all their

148

Fair Representation for Minorities

votes on one candidate. For example, in a five-member district, an AfricanAmerican candidate would need the support of only 17 percent of the voters to be elected if all supporters cast all five of their votes for that candidate. In contrast, in a normal at-large plurality election only candidates who received more than 50 percent of the vote would be elected. In effect, cumulative voting dramatically lowers the threshold of votes necessary to get elected, which increases the opportunity for minority candidates to win office.47 Since African Americans make up only 12 percent of the voters in Chilton County, they never were able to elect any representatives under the old at-large elections. However, when CV rules were used in the 1988 elections, black voters were able for the first time to elect a black member to the seven-member county commission. The first black was also elected to the seven-member county Board of Education. (Interestingly, the use of CV not only broke the monopoly of whites in these bodies, it also broke the party domination of the Democrats. Traditionally, Democrats occupied all seven seats in both bodies. Under CV rules, Republican voters were able to finally gain representation, electing three members to each body in 1988. This is a good example of how cumulative voting encourages fair representation for all political minorities, whether racial or partisan.) In their study of cumulative voting in Chilton County, Richard Pildes and Kristin Donoghue discovered that some whites accepted it only begrudgingly, having preferred the creation of a single black majority district. Nevertheless, they found in the community a general belief that “cumulative voting was effective [and] necessary to achieve minority representation.” They also found that election of minorities produced several beneficial political changes. Decisions about road paving were made more equitably, and more minority appointments were made to administrative boards. And perhaps most significant, black voters felt more connected to local government.48 Chilton County is only one of the many successful examples of the use of semiproportional systems to ensure fair racial representation. In their recent book, Electoral Reform and Minority Representation, Bowler, Donovan, and Brockington studied 32 uses of these systems where African Americans were the minority political group.49 They found that in 97 percent of the cases where African American candidates sought office, at least one was elected. This is particularly significant given that in nearly all of these jurisdictions, no minority candidate had been able to win office under plurality rules.50 They also found that cumulative voting has been useful in some communities with Latino minorities. They cite one particularly interesting example of

Fair Representation for Minorities

149

the 2000 elections for the school board in Amarillo, Texas. Only one minority member had ever been elected to the school board there even though more than 20 percent of the residents were minority. The first use of cumulative voting resulted in the election of two minority members, one African American and the other Latina. The local head of the NAACP explained: “I think you have to say that cumulative voting was a plus. It’s the first time in the history of this area that two minorities were elected to the school board . . . . and I think cumulative voting had an effect on that.”51 Bowler and his colleagues conclude that their findings “should be encouraging to those interested in facilitating minority representation without relying upon the acrimonious and litigious process of drawing districts on the basis of race.”52

Justice for All The struggle to ensure fair elections and fair representation for racial and ethnic minorities in the United States is far from over. It is not enough that minorities have won the right to register and to vote. That is only the first step. They also need the right to cast effective votes—votes that enable them to elect the representatives of their choice. That means that we must look carefully at how the electoral process itself can disempower minority communities. It is encouraging that in the wake of the Supreme Court’s rulings against race-based districting, an increasing number of scholars, politicians, voting rights lawyers, and civil rights organizations are looking more seriously at alternative systems like proportional representation. Lani Guinier of the Harvard Law School has been one of the most vocal proponents of PR. As she has concluded: “winner-take-all elections are ill-suited to the multiracial and polyethnic society of today because they are unable to fill the need for diverse debate and broad representation.”53 The demand for increased minority representation is not likely to go away; indeed it is likely to grow even louder in the years ahead. America is becoming increasingly heterogeneous racially and ethnically. We have rapidly expanding Latino and Asian populations, and the demand for increased political power and representation is likely to intensify correspondingly. That demand will also grow if court rulings preventing the use of majorityminority districts result in the loss of representation for minorities. It should become increasingly clear that a PR system of elections is the most simple and effective way to fulfill minority demands for a fair share of political

150

Fair Representation for Minorities

power. This is not to suggest that the adoption of PR would be a panacea for all of the political problems of racial minorities in the United States. It is simply to observe that the election of their fair share of representatives is a vital political tool for minorities, and for that reason minority political leaders should embrace proportional representation as a part of their agenda for political reform in the United States. Ironically, the need for a truly fair election system is now becoming increasingly evident to whites in some sections of the United States. In Texas and the Southwest, for example, fast-growing Latino populations will soon make whites a minority in many cities and towns. Many whites in these towns have expressed the fear that they may soon be the ones locked out of municipal government. They wonder who will promote their interests or listen sympathetically to their problems “when the Mexicans take over.”54 Interestingly, it is exactly this same circumstance—a declining political majority trying to protect its position—that helped to usher PR in to many European countries in the early twentieth century. In those countries the old established parties embraced proportional representation as a way of retaining some power in the face of the large numbers of new voters brought into those political systems by the advent of universal suffrage.55 In the American situation, declining white majorities will find it in their interest to promote PR as a way of preserving their fair share of seats and power in local legislatures. Such situations make it evident that the proper representation of racial and ethnic groups in American politics is no longer simply a concern of African Americans, Asian Americans, or Latinos. Ultimately, it is an issue of basic political justice—one that should be of concern to all Americans who believe in the principles and values of democracy. As Lani Guinier explains, the issue of electoral reform goes far beyond the concerns of racial minorities. “The real value in exploring alternative election systems is that engaging with and trying to remedy the experience of racial underrepresentation can provide useful lessons about democracy for all groups, not just blacks.”56 So this is not just about fairness for racial and ethnic minorities, but fairness for us all. Fair elections and fair representation are in everyone’s interest, and proportional representation is the best way to achieve this kind of electoral justice.

7

Encouraging Voter Turnout

One of the most curious aspects of elections in the United States is that many Americans simply choose to not take part in them. Despite all the rhetoric about voting being the essence of democracy and the sacred duty of all citizens, despite the arduous political struggles waged by women and minorities to win the vote, and despite all the American lives lost defending democracy and the right to vote, most Americans usually do not exercise that right. In presidential election years, barely half the eligible electorate bothers to use the ballot box and so roughly 100 million Americans routinely sit out our most important elections. In off-year elections the turnout is even lower. In 1998 only 32.9 percent of the electorate cast votes in our congressional elections. That meant that nonvoters totaled 134,896,000, and outnumbered voters by almost two to one. This abysmally low level of voter turnout is almost uniquely American. As table 7.1 illustrates, many other Western democracies routinely have voter turnout for parliamentary elections of 70 and 80 percent. The United States ranks dramatically lower than all of these other countries. As the table suggests, there seems to be a connection between voting systems and turnout. In general, voting is much higher in PR systems than in single-member district systems. But as we will see in this chapter, low turnout is a very complex phenomenon with a number of factors—including electoral systems—contributing to this problem. But before we begin to

152

Encouraging Voter Turnout

table 7.1 Average Voter Turnout in the 1990s Country

Voting System

Average Turnout

Italy

Mixed Member PR

90.2%

Iceland

Party List PR

88.3%

Belgium

Party List PR

84.1%

Sweden

Party List PR

82.6%

Denmark

Party List PR

81.7%

Austria

Party List PR

79.6%

Spain

Party List PR

79.0%

Norway

Party List PR

75.7%

Netherlands

Party List PR

75.2%

Germany

Mixed Member PR

72.7%

United Kingdom

Single-Member District Plurality

72.4%

France

Single-Member District Majority

60.6%

Canada

Single-Member District Plurality

60.1%

United States

Single-Member District Plurality

44.9%

Source: International Institute for Democracy and Electoral Assistance, “Voter Turnout from 1990s.” http://www.idea.int/voter_turnout/voter_turnout_pop9.html (November 27, 1999).

sort out these causes and examine the role played by electoral systems, let’s first consider exactly why this is such a serious political problem.

Is Low Turnout Really a Problem? For most observers of American politics, the chronically low turnout rate indicates something amiss in our political system. A study of voter apathy by the Commission on the Media and the Electorate found “a dangerous and widening disconnection between the American electorate and its own political process . . . with broad, perhaps dangerous implications for democracy.”1 But not everyone agrees with such conclusions. In the 1950s the political scientist Hans Eulau argued that a lack of voter participation might actually be a good sign in a democracy. In the aptly titled “The Politics of Happiness,” Eulau suggested

Encouraging Voter Turnout

153

that citizens who do not vote are expressing satisfaction with political and social arrangements.2 Similar arguments resurface every decade or so. In 1990, writing in Time, Charles Krauthammer, praised low voter turnout as a “healthy development.” He argued that it “is a leading indicator of contentment” and that those who refused to vote were able to make “more room for the things that really count in life: science, art, religion, family and play.”3 But while such upbeat interpretations of nonvoting are comforting, they fly in the face of reality. If nonvoting were truly a function of personal satisfaction, we would expect low turnout rates among those who are well off financially and high turnout rates among those who are poor. But the opposite is true: Voter turnout is lowest among the poor in our society. Among Americans whose annual incomes fall below the poverty line, the voting rate is 25 percent, while those above the poverty line vote at a 65 percent rate.4 It hardly seems likely that the poor are simply too content to vote—that they are too busy crowding into our science and art museums to find their way to the voting booth. It is more likely that the poor abstain from voting because they see few politicians of either party offering policies that would significantly address their problems. In another effort to put an optimistic spin on our low turnout rate, George Will has argued that high turnout rates may actually be undesirable.5 He notes that the German election that swept the Nazis into power in 1933 had an 88.8 percent turnout. In his view, high voter participation indicates an unstable political system—a society in turmoil. But again little evidence supports this conclusion. Countries that routinely have high turnout rates, such as Belgium, Sweden, and the Netherlands, have also been remarkably stable politically. They are hardly hotbeds of political turmoil where democratic government hangs by a thread. Systematic studies of this question also indicate that high voter turnout rates are associated with more stable, rather than less stable, political regimes. In G. Bingham Powell’s study of 29 democracies, he found a strong association between higher voter turnout and less citizen turmoil and violence. After controlling for various political and economic variables, he found that those countries with higher turnout rates experienced fewer protests and riots.6

Why Low Voter Turnout Is a Problem In contrast to the silver lining interpretations of massive nonvoting in the United States, most political commentators believe that it raises serious

154

Encouraging Voter Turnout

questions about political alienation, the absence of political equality, and the legitimacy of our elected officials. Many consider nonvoting to be a symptom of widespread political alienation in the American electorate. As William Crotty observes, “A principal cause of non-voting clearly is the psychological malaise that dominates contemporary politics.”7 Attitudinal studies reveal that large numbers of Americans distrust government and feel they exercise little influence over what it does. For many, parties, candidates, and elections are largely irrelevant—and for such people, voting becomes a meaningless act.8 In his book Why Americans Don’t Vote, Ruy Teixeira argues that declining confidence in government and government officials helps to explain why voter turnout (already low by comparative standards) has been declining since the 1950s—falling in presidential elections from 62.8 percent in 1960 to near 50 percent today.9 Arend Lijphart has noted another serious problem with low turnout. In his presidential address to the American Political Science Association, he argued that voter turnout serves as one important indicator of political equality in a society, and that low turnout serves to undermine that equality.10 The principle of political equality lies at the very heart of democracy. Much of what separates democracy from other more authoritarian forms of government is the notion that all citizens should be able to participate equally in determining the actions of government—that power should not be concentrated in the hands of a few groups. Unfortunately, in most forms of political participation in the U.S., citizens are not equal. For example, some citizens can afford to donate much more money to political campaigns than others, and some interest groups have the advantage of being able to fund much more extensive lobbying efforts. But voting can serve as a counterbalance to this political inequality. It is the one form of political participation in which everyone has equal power: one vote. However, this political equality breaks down if many people don’t participate in elections. And this unequal participation can have serious political consequences. As Lijphart observes, “who votes and who doesn’t has important consequences for who gets elected and for the content of public policies.”11 Walter Dean Burnham made a similar point when he argued: “The blunt truth is that politicians and officials are under no compulsion to pay much heed to classes and groups of citizens that do not vote.”12 One reason this political inequality is disturbing is that we have long known that there is a class bias to voting in the United States; that lower class voters tend to turn out in much lower numbers than middle and upper class

Encouraging Voter Turnout

155

voters. This means that the views of the lower class may have less clout with policymakers and so their interests are less likely to be manifested in public policies.13 Piven and Cloward, in Why Americans Still Don’t Vote, argue that this class bias in the American electorate has had substantial public policy effects during the last 100 years, especially in recent efforts to cut back on social welfare policies.14 Because of this class bias in turnout, Curtis Gans of the Committee for the Study of the American Electorate fears that “the nation that prides itself as being the best example of government of, for, and by the people, will continue to drift towards a government of, for, and by the interested few.”15 Finally, low voter turnout can also undermine the political legitimacy of government itself—the right of officials to rule. Elected officials must ultimately rest their claims to power on the fact that they represent the public— as evidenced by election results. But when large numbers of voters sit out elections, this claim is seriously eroded. In recent memory, most presidents have been elected with around 50 percent of the vote or less. But half of the electorate usually stays home in these elections, which means that the winning candidates are actually endorsed by only about 25 percent of the eligible voters. The problem is much worse for Congress. When the turnout for congressional elections drops to 37 percent, the party that controls the House of Representatives only garners the votes of about 20 percent of the eligible voters. As Kay Lawson has observed, such situations make a travesty of the basic democratic notion of majority rule.16 Those who are making the laws in Congress may claim to have mandate to rule, but in reality they represent barely a fifth of the public—a clear situation of minority rule. When the democratic link between citizens and their political leaders is weakened by low voter participation, political leaders feel freer to follow their own policy preferences. Citizens, in turn, become less inclined to identify with the decisions made by these leaders. That those countries with low voter turnout are precisely those countries that have more riots and protests is not coincidental. Such activities are classic expressions of citizens who no longer feel an effective political link with elected officials.

Explaining Low Voter Turnout Political scientists have taken several approaches to explaining the problem of low voter turnout in the United States. One focuses on the individual

156

Encouraging Voter Turnout

characteristics of citizens, and seeks to identify which factors best explain the tendency to vote. These studies have found that income and education are best correlated with voting.17 Those who have less income and less education are less liable to vote. Such explanations seem to make sense until we start to compare turnout in the United States with that in other Western democracies. As indicated in table 7.1, we tend to lag far behind many of these countries. Is this because we also lag behind in terms of the levels of income and education of our citizens? No—in fact the U.S. is one of the richest and most highly educated of Western democracies. This points to the limitation of focusing on individual characteristics to explain low turnout. In his study of voter turnout among 37 countries, Mark Franklin noted the basic problem with this kind of analysis. The trouble with this approach is that it cannot explain the large differences in turnout that exist between countries . . . . Differences in turnout by levels of education and income are less than differences in turnout by country. Moreover, high turnout countries do not have richer or more educated people than low turnout countries.18 Franklin goes on to conclude that not only do individual characteristics explain little about turnout differences between countries, but also that changing those characteristics would do little to change comparative turnout rates. For instance, even if everyone in the U.S. were to become college educated, this would still not bring our turnout even up to the level found in medium turnout countries like Spain or Ireland. Problems also afflict explanations that focus on political culture. For example, conventional wisdom has it that U.S. turnout is low because our citizens are simply more apathetic and less likely to participate in political activities. But studies have shown that in fact Americans express the highest level of interest in politics and are more apt to discuss politics than citizens in other democracies.19 Americans also show the highest levels of political participation for activities other than voting, such as being active in community problem-solving organizations and contacting officials about social problems.20 So political culture seems to explain little about differences in turnout. These problems with individualistic and cultural explanations have led many political scientists to another approach to explaining voter turnout— the institutional approach. This approach focuses primarily on how coun-

Encouraging Voter Turnout

157

tries’ electoral institutions differ and how these arrangements encourage or discourage voter turnout. Studies based on this comparative approach have found that many institutional factors have an effect on rates of election participation, including registration procedures, when elections are held, whether postal voting is allowed, whether voting is mandatory, and what electoral systems are used. Studies using statistical analysis to compare large numbers of countries have been able to isolate the effects of each of these variables. Franklin, for instance, found that compulsory voting increases turnout by 7 percent, postal voting increases it by 4 percent, and Sunday voting by 6 percent. For our purposes, his most interesting finding is that using proportional representation rather than winner-take-all voting increases turnout by 12 percent.21 So a complex picture emerges from this institutional analysis, one that shows that many factors affect voting turnout. But it is clear that electoral systems do play a significant role. Different scholars have found somewhat different levels of effects for proportional representation. Lijphart estimates that PR increases voting by somewhere between 9 and 12 percent.22 A study by Blais and Carty found a 7 percent boost from PR.23 If we average the figures from Franklin and these others, we come out with an estimate of about 10 percent increased turnout due to using proportional representation. This may sound like a relatively modest increase, but in the U.S., this would translate into millions of more voters going to the polls. Bowler, Donovan, and Brockington did a revealing study that compared turnout in U.S. cities that are now using a semiproportional system, the cumulative vote, to those same cities when they were using plurality voting. By looking at the same cities, they could be sure that any differences in turnout were due to the change in electoral system, and not to other demographic or cultural factors. They found a turnout increase of about 5 percent with the adoption of cumulative voting. This figure is more significant than it might appear at first. These local elections only averaged about a 19 percent turnout, so a 5 percent boost translates into an increase of 24 percent. They concluded: “If political reformers are interested in increasing participation in U.S. elections, these findings could be taken as evidence that proportional representation could stimulate greater participation, at least in local elections where the potential problem of low turnout is most severe.”24 But while it is clear that proportional representation plays some role in increasing voter turnout, it may not be obvious at first why this is so. In fact there are several potential explanations of why PR tends to increase voter

158

Encouraging Voter Turnout

participation, including its ability to waste fewer votes, create more competitive districts, encourage parties to mobilize voters, offer more parties to choose from, and provide clearer choices for voters.

Reducing Wasted Votes Encourages Voting One of the main reasons why proportional representation encourages voting is that it minimizes wasted votes. Plurality elections always have large numbers of wasted votes and this takes away much of the incentive to vote. In many contests, voters simply see no political payoff for voting. SMP elections have two kinds of wasted votes. First are all the votes cast for the candidates who lose. In a district controlled by a large majority of Democrats, any votes for a Republican candidate are wasted. In such situations Republican voters quickly learn that their votes are useless and that they have little reason to vote. Second are all the votes in excess of what the winning candidate needs for victory—anything over 51 percent. If a candidate gets 80 percent of the vote, 29 percent of those votes are not needed and have no effect. Thus many Democrats in Democrat-dominated districts also have little reason to turn out—their candidates will win anyway. The frustration produced by this kind of situation was expressed well by the complaint of one San Francisco resident after a recent election: “Vote? Why vote? I know who’s going to win, everybody knows who’s going to win. Pelosi always wins, with 80 percent of the vote. Nobody else has a chance.”25 When both kinds of wasted votes are added together, they always equal at least 49 percent of the votes cast in a plurality election. For example, in a twoway race in which one candidate wins 60 percent and the other 40 percent of the vote, the excess votes for the winner—9 percent—added to the votes for the loser equal 49 percent. When a race has more than two candidates, the proportion of wasted votes can be even larger. If the winner garners only a plurality of the vote—say, 42 percent—a majority of the votes (58 percent) are wasted. Given this situation, it is hardly surprising that many American voters complain that their votes have no impact—in many cases that is literally true. And when votes do not count, it makes little sense to vote. Proportional representation systems produce far fewer wasted votes and far more effective votes, which naturally increases the incentives for people to participate in elections. If you think that your vote has a very good chance of electing someone, you are more likely to go to the polls. With PR, wasted

Encouraging Voter Turnout

159

votes may be reduced to a little as 10–15 percent, a stark contrast with SMP elections. In the single transferable vote (STV) form of PR, votes are not wasted if they are cast for a candidate who has too little support to win or for a candidate who will win by an excessive amount. In those cases the votes are simply transferred to the voter’s next choice—thus still serving to elect a representative. When STV has been used in U.S. cities, the number of wasted votes has been reduced significantly—usually to about 15 percent. Party-list forms of PR also provide a strong incentive to vote because a voter’s party of choice may need only 5 to 10 percent of the vote for its candidate to get elected. A voter rarely is in the hopeless position of many supporters of minor parties in a winner-take-all system. In this regard, it is significant that in a study of 18 democracies, Karp and Banducci found evidence that “those with small party preferences are discouraged from voting in plurality systems, but not in PR systems.”26 It is just common sense that if your vote is more likely to have an impact, you are more likely to cast it.

Increased Competitiveness Encourages Participation Proportional representation also encourages turnout because it increases the competitiveness of elections. Political scientists have long known that when elections are competitive, more people turn out to vote.27 But plurality elections tend to dampen competitiveness. Many single-member district seats are “safe” seats, where the dominance of one party virtually ensures the victory of their candidate. Indeed, some seats are so safe that the other party doesn’t even bother to put up a candidate. In the 2000 state legislative races, for example, either the Democrats or the Republicans failed to nominate a candidate in 40.6 percent of the races.28 When elections are “no-contest,” voters have little reason to go to the polls. PR increases competitiveness in several ways. First, it virtually eliminates gerrymandering, one of the main causes of safe districts. More importantly, in the large multimember districts used in PR, all parties are competitive. All parties—even minor ones—have a good chance of winning some seats, so their supporters have a reason to turn out. Moreover, representation in proportion to the vote provides an additional incentive to turn out: the more votes a party gets, the more seats it wins. As Hendrik Hertzberg has pointed out, in many single-member district elections, “your side can turn out ten percent more voters than last time and it makes no practical difference. But

160

Encouraging Voter Turnout

under proportional representation ten percent more votes means ten percent more victories . . . . Under PR, all races are close and every vote counts—another reason turnout would undoubtedly be higher.”29

Party Mobilization and Voting The inherent competitiveness of PR elections increases turnout in another way. The decision to vote is not merely a product of individual motivations; it is affected by party efforts to get out the vote. When parties conduct strong mobilization programs, turnout increases. However, because SMP systems boast large numbers of safe seats, parties often have little incentive to mobilize. The minority party usually finds it most strategic to write off the safe districts as hopeless and to concentrate its efforts on districts in which it has a chance to win. Parties with guaranteed seats also have little incentive to organize voters. It makes little difference if their candidate garners 75 percent rather than 65 percent of the vote. However, in large multimember PR districts, parties normally have a chance to win at least one seat, and as we have seen, the larger their vote the more seats they will win. This opportunity creates a powerful incentive for parties to mobilize everywhere.30 When all parties are mobilizing their supporters in all districts statewide or nationwide, turnout levels inevitably increase. The increased competitiveness of proportional representation elections thus has a double effect, increasing incentives for the individual to vote and for the parties to mobilize voters.

More Choices and Higher Turnout Fewer wasted votes, increased competitiveness, and greater party mobilization are usually cited as the main reasons why PR increases voter turnout. But this system also encourages higher turnout in several other indirect ways. As we saw in chapter 4, PR systems tend to have a larger number of candidates and parties, which could encourage turnout by increasing the options available to voters.31 The limited choice of our two-party system often makes it difficult for some voters to find a candidate or party with whom they strongly identify, and this increases the tendency to abstain from voting. This frustrating experience is much less typical under PR. The wide

Encouraging Voter Turnout

161

variety of parties makes it much easier for voters to find parties and candidates who reflect their particular interests and policy orientations. Voting becomes a more attractive and satisfying experience. For instance, under PR in the United States, voters who are fiscally conservative and socially liberal would be able to find a Libertarian party candidate on the ballot who espoused this same combination of views, and thus come away from the voting booth knowing that their views have a good chance of being represented in the legislature. That is not possible today. Common sense indicates that voter participation rises when most voters can find candidates with whom they agree—candidates they can get truly excited about. Research supports this notion as well. John Zipp conducted a study of presidential elections in which he measured how similar or different the candidates’ positions were from voters’ positions on a variety of issues and correlated that data with the propensity of those voters to go to the polls in those elections. He found that “in general, if one has a clear choice among the candidates and one’s policy preferences are close to at least one candidate, one is much more likely to vote.” He concluded that “one reason that individuals do not vote is that their interests are not represented by any of the major candidates.”32 Studies of the relationship between party identification and voting also support the notion that the greater choices offered by PR increase turnout. We have long known that lower levels of party identification correlate with low voting rates.33 And we also know that in our plurality system only about 30 percent of voters strongly identify with either major party. In their study of electoral systems and participation, Karp and Banducci conclude that PR systems tend to foster stronger party identifications, because the wider choice of viable parties means that a larger number of voters can feel more in sync with the party of their choice. In their view, these stronger identifications help to explain the higher turnout found in PR systems.34

More Information and Clearer Choices Proportional representation may also boost turnout because this system provides voters with better information about their choices. Anthony Downs and other rational choice theorists argue that voters tend to abstain primarily when they find it difficult to tell the difference between the parties or have trouble determining which candidate will maximize their political benefits.35

162

Encouraging Voter Turnout

When voters face great uncertainty and the available information about parties and candidates is ambiguous, abstaining can become rational. To use an economic analogy, when the electoral market provides little useful product information to the voters, they have trouble rationally choosing which product best meets their needs and may decide not to choose at all. This problem is more likely to occur in two-party plurality systems than in multiparty PR systems. As chapter 3 showed, plurality systems encourage election campaigns that downplay issues. In order not to offend the voters needed for a majority, candidates and parties tend to emphasize personality and image-making and avoid taking stands on controversial issues. Also, when issue positions are discussed, candidates often try to occupy similar positions near the middle of the political spectrum in order to woo the allimportant swing voters. But when parties and candidates fail to offer clear choices, voters are more likely to abstain. In contrast, several aspects of campaign in PR systems would tend to help clarify the policy positions of candidates and thus encourage higher turnout. First, there is more emphasis on policies and party platforms in these campaigns than on individual candidates. Second, parties are likely to spread themselves out along the ideological spectrum rather than converging in the center.36 As a result, more policy positions are discussed and parties must make a greater effort to educate the public about how their positions differ from those of their rivals. As Karp and Banducci explain this point: “PR is likely to increase the number of parties and the likelihood that parties will strive to distinguish themselves ideologically . . . . The increase in parties and a tendency not to converge in the ideological center should increase the options from which voters can choose and result in fewer abstentions.”37 Walter Dean Burnham has reached a similar conclusion, arguing that campaign strategies in PR systems produce “a maximum clarity on issues for voters during the election period and thus, one would anticipate, extremely high normal turnouts—as indeed one finds in such electoral systems.”38

Blaming the Voting System, Not the Voters Once we fully appreciate the role played by electoral systems in turnout and all the ways that our single-member plurality system discourages voting, the low voter turnout in the United States begins to make much more sense. Nonvoting may actually be a rational response to this flawed electoral system.

Encouraging Voter Turnout

163

It makes little sense to vote when elections are not competitive, when the choices are not clear, when you cannot find a candidate in which you believe, and when your vote is likely to be wasted. In this situation, citizens should not be faulted for concluding that voting is largely meaningless and literally a waste of time—they are right. As Mark Franklin has observed: “Voters are not fools, and an unresponsive system will motivate many fewer of them to vote.”39 So we should be blaming our faulty voting system, not the voters who refuse to participate in it. And instead of trying to harangue people into voting, it would be much more effective to simply switch to an electoral system where casting a vote becomes a much more meaningful and effective political act. That system, as we have seen, is proportional representation.

Electoral Systems and Class Bias in Voting Understanding the discouraging effects of single-member plurality rules not only helps to explain low voter turnout, but also sheds light on the other puzzle in U.S. voting patterns: Why the poor and working classes tend to vote at substantially lower rates than others. This problem is in many ways as disturbing as the general state of low voter participation. Blue-collar workers have only a 48 percent turnout rate, while white-collar workers have a turnout rate of 71 percent. Thus although working-class voters make up 32.9 percent of the population, they make up only 26.1 percent of the voters.40 And as seen earlier, turnout rates are even lower for those living in poverty. This means that the lower classes tend to be underrepresented among voters and therefore in the political system as well. Common sense would suggest that this class bias in the electorate would result in policies that favor wealthier classes over poorer ones, but there is also some empirical evidence of this as well. Hill and Leighley conducted a study to test whether class bias in the electorate translated into policies that favored richer Americans and hurt poorer ones. They looked at the degree of class bias in state electorates and found that this correlated highly with the levels of social welfare spending in these states. States where the poor were most underrepresented in the electorate also tended to allocate much less funds to various social welfare programs. They concluded that this was compelling evidence in support of the conclusion that “an electorate disproportionately representative of high-class citizens will be rewarded with public policies in favor of its interests and at the expense of lower-class citizens.”41

164

Encouraging Voter Turnout

In their book, Why Americans Still Don’t Vote, Piven and Cloward argue that a class bias in turnout helped to promote the “dismantling of the welfare state” that occurred in the 1980s and 1990s under Republican governments. They point out that the elections of anti-welfare state politicians like Ronald Reagan and Newt Gingrich were aided by the abstention of lower-class voters who tend to vote Democratic. When the Republicans took over Congress in 1994, for instance, the Democrats won only 47 percent of the actual vote, but they scored 58 percent among nonvoters.42 What causes the class bias in voter turnout? It is often assumed that the correlation between low income and low turnout means that something about being poor discourages participation—that the poor may, for example, lack the civic orientation that encourages voting. But the argument is problematic, not in the least because it does not jibe with the fact that the poor in most other industrial democracies tend to vote at respectable rates. In European countries with 80–90 percent turnout rates, the poor and working classes are obviously turning out in very high numbers. A better explanation for the poor turnout among lower-income Americans is that the institutional factors that discourage voting in the United States tend to disproportionately discourage those with lower incomes. For example, Piven and Cloward argue persuasively that voter registration rules have historically worked to disproportionately discourage lower-income voters.43 Similarly, SMP election rules may work in several ways to selectively discourage these same voters. For example, one aspect of plurality elections that may discourage lowincome voters more than others is the relative paucity of useful information coming from political campaigns—a point mentioned earlier. When voters must make the effort to dig out the information necessary to make a rational voting choice, the burden will be most strongly felt by the lowerincome voters. Low-income voters are less likely to have the time and resources to assemble the relevant information. As Anthony Downs concludes, one of the main “reasons for higher abstention among low-income classes [is] greater uncertainty caused by the inability to bear information costs.”44 Another characteristic of single-member plurality election systems that plays an even more important role in selectively discouraging lower class voters is the lack of a variety of parties. In particular, what is missing in our two-party system is a party that speaks directly for the interests of those at the bottom of the economic ladder. According to Curtis Gans, director of the

Encouraging Voter Turnout

165

Committee for the Study of the American Electorate, this helps to explain why less well-off Americans tend to abstain from voting: “The basic cause is essentially that neither party is speaking to the interests of the lower-income brackets of Americans.”45 In most PR countries, lower class voters can find candidates on the ballot from a labor party, a socialist party, or a social democratic party—parties that have traditionally acted as genuine representatives of their interests. Such parties, however, do not appear on the ballot in the United States. Despite the Democratic party’s rhetoric about being the party of the working people, it does not have the same kind of strong working-class oriented policies as the other true leftist parties nor does it have the same kind of intimate organizational connections with the working-class groups that these other parties have in their countries. And as a rule, the Democrats do not make the same massive efforts to mobilize the poor and working class during elections as occurs in many other countries. Indeed, many Democrats have eschewed this electoral strategy in favor of concentrating on raising more money from corporate political action committees and wealthy contributors.46 Canadian elections provide some interesting evidence of how the presence of a real leftist party can encourage turnout among the working class. Zipp and Smith studied the relationship between the New Democratic party (NDP) and turnout among working-class voters. The NDP is classified as a mildly socialist or social democratic party and thus could be expected to appeal to many of these voters. Zipp and Smith compared workingclass turnout for those regions of the country where the NDP was strong and their candidates had a good chance of being elected and regions where the party had little chance of electoral success. They found that voting among the working class was higher where NDP candidates were realistic choices.47 If we were to adopt PR in the United States, a party addressing the needs and concerns of lower income and working-class voters would likely emerge. Indeed, more than one party might actively court this part of the electorate. We could imagine that some working-class voters would find the appearance of a socially conservative, religion-based party attractive. On the left, a Labor party could also serve to mobilize and represent many working class and poor Americans. In any case, the probable emergence of new parties under a PR system in the United States would undoubtedly assist lower income Americans of all political stripes in their efforts to win their fair shares of representatives and political power.

166

Encouraging Voter Turnout

PR as an Important Part of the Solution Massive abstention from voting in the United States and the class bias among those who do vote are serious problems that compromise the fairness and responsiveness of our political system. Low voter turnout can only continue to undermine the kind of political equality we need to make democracy work in everyone’s interest. Proportional representation by itself will not completely solve this complex problem. Other reforms, including such things as postal voting and voting on a weekend or holiday, would help as well. Nevertheless, as the evidence in this chapter has shown, a large part of this problem is undoubtedly caused by our reliance on plurality elections, and so part of the solution has to be a change to PR. Proportional representation would give Americans much better reasons to show up at the polls. We would be much more confident that our votes would help to elect someone in whom we believe, and the wider variety of parties would make it much easier to find candidates that we were actually excited about. These changes could not help but increase the turnout of American voters and make our elections into a more truly democratic and representative process.

8

PR: A Different and Better Kind of Democracy

When thinking about proportional representation, it is easy to get focused on its specific advantages over single-member plurality elections: the fact that it wastes fewer votes, creates more accurate representation of parties, encourages higher voter turnout, and so on. These are valuable features of this voting system, but it is important to not lose sight of the larger political goals of electoral reform. Adopting PR is not simply about making elections fairer or even about encouraging better representation—it is ultimately about creating a better democracy. The point of proportional representation is to create a political system in which we empower as many people as possible. It aims to create a situation of greater political equality where everyone has a voice in the process of making the laws that govern our country. In this sense, proportional representation and single-member plurality systems represent not only different kinds of voting systems, but also different visions of politics—different visions of who should have power in a democracy and how it should be wielded. Growing up in a given society, we often tend to assume that there is only one way to run a democracy. But in fact there are many ways. This notion that there can be different kinds of democracies—some better than others— has gained attention only relatively recently. During much of the twentieth century, people were not so much concerned with different forms of democracy, but with whether democracy would survive or not. In the first half of

168

PR: A Different and Better Kind of Democracy

the century, the political struggle was between democracy and fascism; and in the second half, it was between democracy and communism. Democracy was seen as one political system among others, and often it was largely defined by the presence of free elections. What seemed to matter most was whether countries had free elections or not—not the kind of elections they had or how those electoral systems affected the quality of democracy in those countries. However, today democracy seems safely ensconced in much of the world. And so a different set of political questions is being raised in these countries: What kind of democracy is best? What political arrangements are best at empowering citizens and holding officials accountable? Increasingly it is being realized that democracy is an unfinished project—one that can and should be open to improvement. Citizens discontented with a variety of political problems—including corruption, abuse of power, domination by special interests, and poor policy decisions—have been demanding reforms in their political systems. And electoral systems have become part of this discussion about how to make democracy work better. As Kathleen Barber has explained: “Election systems in the United States and around the world are focal points for discussion of democracy, its meaning, and its practice in a fast-changing world.”1 In many countries, citizens and politicians have begun to take a critical look at their voting systems and consider how they affect the quality of democracy in those nations. In recent years, for instance, public debates about electoral systems and their political implications have taken place in Japan, New Zealand, Mexico, Israel, and Italy—and all of these nations have changed or modified their voting systems as a result of these discussions. Increasingly, electoral scholars have also begun to examine the democratic implications of electoral systems. In this chapter we will consider the work of several scholars who have explored how different voting systems contribute to different theories and practices of democracy. They all argue that proportional representation encourages a different and better kind of democracy. PR does this by encouraging the things that make democracy work better, such as political equality, power sharing, citizen participation, vigorous political dialogue, and political bargaining and negotiation. In the end, proportional representation is desirable not simply because it is a better electoral system, but because it also helps to create a more pluralistic and egalitarian political system. As Barber has put it, “If democracy in the twenty-first century is finally to mean the full and substantive participation of all groups

PR: A Different and Better Kind of Democracy

169

in political life, the electoral foundations for such a change can be found in proportional systems.”2

Consensus versus Majoritarian Democracies Undoubtedly, the scholar who has done the most to illuminate the effects that electoral systems have on promoting different kinds of democracies is Arend Lijphart. Through his books, he is largely responsible for introducing an important distinction between basic types of modern democracies: majoritarian versus consensus democracies. Some of his work in this area has already been referred to in earlier chapters in this book, but it is time to explore in more detail his typology of democracies and the role that electoral systems play in encouraging one type or another. To distinguish between consensus and majoritarian systems, Lijphart sets out several criteria—institutional characteristics that are typical of these two approaches to democracy. There are a wide range of these criteria, including some that apply to the judicial, legislative, and interest group systems.3 However, for our purposes the most important distinguishing characteristics are those related to electoral and party systems. In particular, consensus democracies are typified by proportional representation electoral systems, multiparty systems, and coalition governments; while majoritarian democracies are characterized by plurality/majority elections, two-party systems, and singleparty governments. Thus while the SMP/PR distinction is not exactly the same as the majoritarian/consensus distinction, there is a great deal of similarity, and electoral systems play a key role in distinguishing between these two models. As Lijphart himself argues, proportional representation is one element of “crucial indirect importance” in providing an “impetus towards consensus democracy.”4 So clearly PR has a central role in promoting this kind of democracy. But what exactly is consensus democracy and how does it differ from the majoritarian model? One way to see the difference between majoritarian and consensus democracy is to see how each model responds to one of the most basic dilemmas of democracy. Democratic governance would be easy if all citizens were in agreement. But of course this is not the case. So the issue becomes this: who will do the governing and to whose interest should the government be responsive when the people are in disagreement and have divergent preferences? As Lijphart explains: “One answer to this dilemma is the majority of the people.

170

PR: A Different and Better Kind of Democracy

This is the essence of the majoritarian model of democracy.” However, as he goes on to observe, there is another possible answer to this crucial question. The alternative answer to the dilemma is: as many people as possible. This is the crux of the consensus model. It does not differ from the majoritarian model in accepting that majority rule is better than minority rule, but accepts majority rule only as a minimum requirement: instead of being satisfied with narrow decision-making majorities, it seeks to maximize the size of these majorities. Its rules and institutions aim at broad participation in government and broad agreement on the policies that the government should pursue. The majoritarian model concentrates political power in the hands of a bare majority—and often even merely a plurality instead of a majority—whereas the consensus model tries to share, disperse, and limit power in a variety of ways. A closely related difference is that the majoritarian model of democracy is exclusive, competitive, and adversarial, whereas the consensus model is characterized by inclusiveness, bargaining, and compromise; for this reason, consensus democracy could also be termed “negotiation democracy.”5 In some ways the term “negotiation democracy” does give a better sense of this style of democratic politics. The term “consensus” can give the misleading impression that this model fosters or requires unanimous agreement—which is not the case. The essence of this kind of democracy is, as Lijphart suggests, to include as many people in the policymaking process as possible, and in this way encourage more negotiation and compromise. After all, real give-and-take between different political, economic, racial, and religious groups is possible only if all of them are at the political table. Including as many people as possible in the political process has always been one of the basic appeals of democracy. Indeed, the reason that democracy has become the preferred means of government around much of the world is that it benefits more people—it allows more people to wield power in the political system compared to, say, a dictatorship or rule by an aristocracy. In this sense, the basic case for democracy is a utilitarian one: the best form of government is the one that includes and benefits the most people. At a minimum, this means that a minority should not rule—that the majority should have their way in the policymaking process. But, as Lijphart asks, why not raise the bar? Why stop with majority rule? Why not allow as many

PR: A Different and Better Kind of Democracy

171

people as possible to have a voice in our legislatures and the opportunity to see their values and needs reflected in the laws that are passed? The work of another electoral scholar, G. Bingham Powell, has reinforced some of the points made by Lijphart. Powell draws a similar distinction between what he calls two visions of democracy: the “majoritarian vision” and the “proportional influence vision.”6 In the majoritarian vision, voters use winner-take-all elections to choose between two competing parties, and the policymaking power is concentrated in the single winning party. In the proportional influence vision, voters use PR elections to create coalition governments where power is more dispersed. In Powell’s words, “The essence of the [proportional influence] vision is that the election brings representative agents of all the factions in the society into the policymaking arena. These agents then bargain with each other in a flexible and accommodative fashion.”7 For both Lijphart and Powell, the key element is the degree to which power is shared or dispersed in these types of democracies. Majoritarian democracies seek to concentrate power by excluding people who are not in the majority, and by creating single-party governments that can dictate policy without having to bargain with any other political groups. In a consensus democracy, a wider variety of political groups—both majorities and minorities—gain access to legislatures. This allows these bodies to be more inclusive and to represent a broader range of citizens’ political views. Importantly, however, the sharing of power among “as many people as possible” continues on in the policymaking process itself. Multiparty PR legislatures almost always require coalitions to form a ruling majority, and this in turn requires negotiation and compromise between parties. Often this power-sharing approach even goes beyond the creation of a ruling coalition. In parliamentary PR countries, the cabinet positions are allocated not to just one party, but to all the parties in the coalition, roughly in proportion to their political strength. These diverse multiparty cabinets stand in sharp contrast to the single-party cabinets typical in the United States. But perhaps the most surprising example of power sharing in consensus/PR systems takes place in the legislative committee system. Committees and their chairs often play key roles in promoting and shaping legislation. Majoritarian systems acknowledge their importance by insisting that all committee chairs be from the ruling party—to better enhance its ability to rule unchecked. In most PR countries, however, the committee chairs are shared in a rather remarkable way. Not only are the committee chairs divided up among the par-

172

PR: A Different and Better Kind of Democracy

ties in the ruling coalition, but some chairs are also given to the parties in the opposition. In countries such as Belgium, Denmark, Germany, Spain, and the Netherlands, it is considered fair to give committee chairs to all parties in the legislature in proportion to their number of seats. Thus while the majority of committee chairs go to the ruling coalition parties, a substantial minority go to parties in the opposition.8 There is perhaps no better example of how consensus democracy embodies an approach to politics that emphasizes inclusiveness, negotiation, and the sharing of power—its effort to ensure that the interests of “as many people as possible” are represented in government. Such an approach is, of course, very alien to the adversarial and majoritarian style of democracy that currently reigns in the United States. Can you imagine a ruling Republican party allowing some committee chairs to go to Democrats? In our majoritarian version of democracy, the ideal is to have one party form a virtual dictatorship in the legislature—for it to monopolize power and refuse to share it with any other parties. These are strikingly different visions of how legislative politics should work in a democratic society. Lijphart’s and Powell’s conclusions about the democratic benefits of proportional representation become even more interesting when it is realized that they did not always believe in the superiority of that electoral system. As they both have explained, they were at first “reluctant” to acknowledge the advantages of PR.9 Both were trained as political scientists at a time when it was largely assumed in the profession that single-member plurality was the preferred electoral system. As the New Zealand political scientist Jack Vowles has explained: “The American hegemony in the literature of political science and, to a lesser extent, that of Britain, contributed to teaching and scholarship which privileged [SMP] systems, turning out graduates who moved into teaching, business or government to whom the superiority of [SMP] systems was virtually unquestionable.”10 But eventually, after years of careful and systematic research, the evidence of the advantages of PR become so overwhelming that both Lijphart and Powell were forced to acknowledge that this electoral system was superior in many important ways to SMP.

Lani Guinier: Beyond Majority Rule and Zero-Sum Politics Lani Guinier is another scholar who has studied the implications of proportional representation for the theory and practice of democracy. She is a professor at Harvard Law School and her work has often focused on the problem of

PR: A Different and Better Kind of Democracy

173

majority rule—one of the central tenets of democratic politics. She has, however, approached this issue from a somewhat different direction than many others. Most PR advocates focus on how single-member district systems routinely violate majority rule by producing plurality winners and manufactured majorities, and they argue that PR more reliably ensures majority rule. But Guinier has a different interest in the issue of majority rule: She is concerned about its limitations and how PR can move us beyond those limitations. Guinier argues that while majority rule is an important democratic principle, it can have some significant drawbacks. In particular, majority rule can turn into a “tyranny of the majority:” the ability of a fixed majority to rule entirely in its own interests, without having to take into account the interests or concerns of the minority.11 Guinier is primarily concerned about racial politics and the ability of fixed white majorities to outvote and overpower racial and ethnic minorities in the political arena. But she also notes that this problem transcends race and has been a perennial question for democratic theorists. Even the founding fathers wrestled with this issue when they were creating the Constitution. Madison and others warned of the dangers of a tyrannical majority and many of the basic characteristics of our federal government, such as checks and balances and the separation of powers, were explicitly designed to make it difficult for these problematic majorities to form. Guinier observes that the problem of tyrannical majorities is often encouraged by plurality elections rules. She illustrates the problems that can come from this winner-take-all approach by describing a situation that developed at Brother Rice High School in Chicago. The students on the senior prom committee had to decide what kind of music would be played. They decided to let each senior list his or her three favorite songs, and the band would play the songs that appeared most frequently on the lists. Seems attractively democratic. But Brother Rice is predominately white, and the prom committee was all white. That’s how they ended up with two proms. The black seniors at Brother Rice felt so shut out by the “democratic process” that they organized their own prom. As one black student put it: “For every vote we had, there were eight votes for what they wanted. . . . With us being in the minority we’re always outvoted. It’s as if we don’t count.” Some embittered white seniors saw things differently. They complained that the black students should have gone along with the majority: “The majority makes the decisions.

174

PR: A Different and Better Kind of Democracy

That’s how it works.” In a way, both groups were right. To the black students, majority rule sent the message: “we don’t count” is the “way it works.” In a racially divided society, majority rule may be perceived as majority tyranny.12 For Guinier, such divisive and polarizing results are typical of majority rule decision systems. The prom incident is an apt metaphor for the kind of unfair election results that have occurred for years in the United States because of winner-take-all voting systems. Blacks and Latinos have often been denied representation by majority/plurality systems that ensure that whites win all the representation in a district and minorities get none. But for Guinier, the SMP system is merely another example of the larger winnertake-all approach to politics that is found in American society. In this winnertake-all atmosphere, “politics becomes a battle for total victory rather than a method of governing open to all significant groups.”13 For Guinier, as for Lijphart, democracy should be a cooperative enterprise that allows for as many people as possible to participate in decisionmaking—rather than a political battle in which all the spoils go to the victor. “In the end, I do not believe that democracy should encourage rule by the powerful—even a powerful majority. Instead, the ideal of democracy promises a fair discussion among self-defined equals about how to achieve our common aspirations . . . . My life’s work has been to try to find the rules that can best bring us together as a democratic society.”14 It is this search for alternative political rules that led Guinier to proportional representation. She notes that if the high school students had used proportional methods to vote on songs, that whites would have chosen the majority of the songs, but blacks would have had some of their music as well, thus enabling everyone to enjoy the dance. In the same way, proportional representation elections allow all groups, both majorities and minorities, to achieve their fair share of representation. Elections do not have to be a zerosum, winner-take-all affairs—we can have elections in which all citizens win. She advocates PR elections, not just to ensure fair representation for racial minorities, but for all political minorities. And for her, proportional representation represents more than just a way of including more people in the political process, it embodies a new vision of democratic politics. The principle of proportionality is molded by the hope that a more cooperative political style of deliberation and ultimately a more equal

PR: A Different and Better Kind of Democracy

175

basis of preference satisfaction is possible when authentic minority representatives are reinforced by electoral structures to empower them at every stage of the political process. Ultimately, however, representation and participation based on principles of proportionality are also an attempt to reconceptualize the ideal of political equality, and so the ideal of democracy itself.15 Reconceptualizing democracy in this way does not necessarily come easily for everyone. For example, the notion of zero-sum, winner-take-all politics is so embedded in American culture that some people have trouble imagining or accepting any other approach to politics. The inevitability of there being winners and losers is built into our capitalist business system and into our competitive sports culture—so some people naturally assume that this result is unavoidable in elections as well. This attitude has been expressed well by Lauren Bain, a Libertarian political writer. She has criticized proportional representation because it promises to give all citizens representation. “PR’s champions most likely mean to be fair-minded. But we live in a republic founded on the principle of majority rule; there must be winners and losers. . . . The vexing assumption behind PR is that everyone has a right to win. That’s a concept belonging to the Special Olympics, not a republican form of government.”16 Bain seems unable to understand that minority representation in legislatures does not violate the principle of majority rule. But even more disturbing is her insistence that some people should be losers in the electoral process—that some people should be denied representation—even when PR makes that political injustice unnecessary.

Taking Turns in Politics Unlike Bain, Guinier is clearly not trapped in a zero-sum vision of politics. She has no trouble understanding or embracing an approach to democracy that gives representation to both majorities and minorities. She has even come up with another way that majorities and minorities can share power—something she calls “taking turns.” It is an idea we often teach children. When asked which game a group of six children should play—four preferred tag and two wanted to play hide-and-seek—my then four-year-old son Nikolas got it right away.

176

PR: A Different and Better Kind of Democracy

“They’ll play both,” Nico advised. “First, they’ll play tag and then they’ll play hide-and-seek.” By contrast, the winner-take-all system . . . awards all the power to those with the most votes who then play “tag” over and over.17 This notion of “taking turns” as a way of allowing more people to fulfill their needs is an intriguing one, but how could it be applied to politics? For example, how could it work in the policymaking process? It would seem to make little sense to have the majority make some policy decisions and then let the minority take its turn and make other decisions. Few would support the idea of allowing a legislative minority to make policy over the objections of the majority. But that doesn’t mean that we are necessarily stuck with a single majority ruling over all others in the legislature. What if we had different majorities taking turns making policy? That is actually possible in multiparty PR legislatures. The ruling multiparty coalitions in PR legislatures are not always perfectly stable between elections. At times they may break apart and a new ruling coalition of parties forms, usually consisting of some parties from the old coalition along with some new ones. As will be seen in the next chapter, some critics see this process of changing coalitions as a drawback to PR. But in fact, as G. Bingham Powell has argued, it can actually make these legislatures more responsive to a broader range of the public. Powell observes that changes in ruling coalitions between elections actually serve to expand the diversity of citizens whose representatives participate in making policy decisions. In SMP elections, one party rules between elections and the others are left out. But between-election changes in the parties that make up the ruling coalitions in PR systems allow new parties to share in the policymaking process. It is in this sense that the parties are taking turns wielding power in the legislatures. There is nothing anti-majoritarian about this, because at any one time the ruling coalition represents the majority of the voters. Powell found that in some PR countries, changing coalitions “ensured that the voters on both sides of the political spectrum had powerful representation in policy making at least part of the time.”18 In this sense, coalition instability may actually be a political blessing because it allows “as many people as possible” to participate in the policymaking process—the hallmark of the proportional influence vision of democracy. The idea of a sequence of different majority coalitions has another appeal as well. One of the other problems with plurality elections is the assumption that there is a single majority—that the winning majority of citizens agree

PR: A Different and Better Kind of Democracy

177

on all important issues. In reality, however, the majority position often differs from issue to issue. On the environment, the majority may agree with the Democratic position, while on taxes they may agree with the Republicans, and so on. This diversity of majorities cannot be embodied in a legislature dominated by one party. But it can in a multiparty PR legislature with changing ruling coalitions. In an extreme form, one could see different coalitions forming around different bills as they come up for consideration—depending on where the true majority views lie. Powell concludes that an “argument in favor of proportional representation is that policymakers should choose the policy desired by the citizen majority on each issue. Because many issues will be considered by the national government between every election and different sets of citizens will form the majority on different issues, it is important that the policymaking coalition not be locked into place by the immediate election outcome.”19

A Test Case: The Political Influence of the Poor The political situation of the poor makes a good test case for the arguments of Lijphart, Guinier, and Powell that PR can facilitate power sharing and help to create a more inclusive and egalitarian political system. The poor tend to have little political power and are one of the political groups most likely to be left out of the political process. So we can ask the question: Can the use of proportional representation elections enable the poor to promote their interests more effectively in the political system? The answer, it turns out, is yes. Clearly the poor have not been able to participate very effectively in the current U.S. political system. This is undoubtedly because they have few traditional sources of political power. They have no well-funded political action committees. Their voting power is dwarfed by the more numerous voters of the middle class. Both of the major parties naturally spend most of their time wooing these middle-class voters and thus pay relatively little attention to the poor. The poor also have no expensive lobbying firms arguing their case in Congress. Given this lack of political influence, it is not surprising that the poor benefit relatively little from public policies in the United States. Compared to many other Western democracies, the United States spends much less on social welfare programs, particularly on those that aid the poor, and is much

178

PR: A Different and Better Kind of Democracy

less effective in reducing poverty. Recent studies have found that in the United States, after taking into account the effects of government social welfare programs such as food stamps, welfare, social security, etc., almost 11 percent of the population still remains in poverty. However, in Germany, Sweden, and the Netherlands, more extensive social welfare programs succeed in reducing the number in poverty to below 4 percent. To look at it another way, these countries’ policies raise an average of 83 percent of their poor out of poverty, while our antipoverty efforts raise a mere 36 percent of the U.S. poor out of poverty—a very substantial difference.20 Why do the poor do better in these other countries? Could the egalitarian, power-sharing style of politics encouraged by proportional representation help to account for this situation? Unquestionably there are many reasons why the poor fare better in some other democracies, but studies indicate that proportional representation contributes to this difference. Several studies have found that consensus/PR countries tend to have more extensive and better-funded social welfare programs than majority/SMP countries.21 It is not difficult to understand why this is the case. PR allows for the representation of a much wider variety of parties. These include leftist parties (Socialist, Labor, etc.) that are much more likely to take up the cause of the poor and the working class than centrist parties, such as the Democrats. As was seen in chapter 7, the presence of these parties in PR countries is undoubtedly one of the reasons why the poor turn out to vote in higher numbers there. More importantly, even though most of these parties represent only a minority of the citizenry, PR allows them to win seats in the legislature— something less likely to occur in an SMP system. In addition, although these leftist parties may have a relatively small presence in legislatures, they at times have the opportunity to participate in ruling coalitions and to occupy some places in the cabinet. This puts them in a position to effectively champion social welfare policies that better take into account the needs of those at the bottom of the economic ladder. There is evidence that this characteristic of PR systems helps to explain some of the differences in social welfare programs across countries. In a study of social welfare spending in 17 democracies between 1970 and 1990, Woojin Moon found that the presence of proportional representation was one important factor in explaining the variations in this spending. Moon concluded that PR is a factor because it encourages multiparty coalitions rather than one-party governments and this has allowed leftist parties to sometimes join in ruling legislative coalitions with centrist parties and to participate in government cabinets.22

PR: A Different and Better Kind of Democracy

179

In other words, PR enables parties strongly representing the interests of the poor to sometimes get their turn at making policy, and the poor have directly benefited from that opportunity in the form of more extensive social welfare programs. The case of the poor, then, seems to give concrete support to the notion that proportional representation allows more citizens to share power in the political process and encourages the creation of governments that take into account the interests of a broader range of citizens—even political minorities with relatively few political resources.

Enhancing Political Dialogue Proportional representation elections have another beneficial effect on democracy besides increased power sharing—they also enhance the quality and range of political discussion in society. Political theorists have long noted that democracies are defined in part by the scope and vigor of their public political discussions, just as authoritarian regimes are characterized by the absence of such discussions. In political theory, a whole literature has developed exploring the advantages of “deliberative democracy”—an approach to revitalizing democratic politics that emphasizes institutional arrangements which encourage a heightened level of political discussion and argument.23 It is argued that a high level of public political dialogue in a society serves a number of important purposes. It produces more knowledgeable and thoughtful citizens. It also encourages more rational policy decisions, ones that take into account a wide range of arguments and points of view. And when all interests have a political voice, it helps to ensure that laws are more just—that decisionmakers are more aware of and sensitive to the interests of a broader portion of the population. Electoral systems have a major impact on the process of political dialogue in a society. In general, proportional representation systems celebrate and encourage a wide-ranging political discussion that includes all political points of view. In contrast, plurality elections often work to stifle public dialogue—restricting both the topics of political debate and who participates in these discussions. The different effects of these two electoral systems are obvious in political campaigns. Ideally, election campaigns serve as a good opportunity for a vigorous public debate about what values and policies a society should be pursuing. They allow competing candidates and parties to persuade voters that their political goals are the best ones. But electoral systems

180

PR: A Different and Better Kind of Democracy

differ greatly in their ability to encourage campaigns that serve as a basis for public education and public discussion. As shown in chapter 3, single-member plurality elections tend to encourage candidates to avoid discussion of controversial issues for fear that they will alienate the voters needed to form a majority or plurality of support. In this voting system, the best way to win is for candidates to adopt vague policy positions and to emphasize personality over issues in their campaigns. This decreases the quality of political debate, but it increases their chances of victory. In contrast, in PR systems, candidates can be candid about their policy stands—even adopt controversial positions—and still get elected. They do not have to try to please most of the people most of the time. In addition, candidates do not run simply as individuals but as part of a party slate, which puts more emphasis on party platforms in campaigns. In general, issues play a more central role in proportional representation elections and this tends to facilitate public discussion and to give voters a much better idea of what they are voting for. The multiparty systems encouraged by proportional representation also tend to enhance the quality of political debate in campaigns. More parties mean more points of view included in campaign ads and in candidate debates. The public gets to see the Libertarian and the Green view of the issues, not just the Republican and Democratic. And more parties mean that they must spend more time detailing how their positions on the issues differ from all the other parties. The Democrats must not only explain how their environmental policies differ from the Republicans, but also how they differ from that of the Greens. In this way, multiparty PR campaigns do much more to enhance the quality and range of public debate and to further the political education of the public. Thomas Christiano, a political theorist who has done much work on the importance of debate and deliberation in democratic societies, has concluded that proportional representation elections are much better than singlemember plurality elections at promoting this kind of political dialogue in campaigns, and “in promoting rational social deliberation on the overall aims of society.” The system of single-member district representation puts a straightjacket on the kinds of issues that are discussed by citizens and parties. In general, the discussion in electoral campaigns tends to take place on a one-dimensional issue space. Discussion in proportional representation elections tends to involve more issues in a number of different di-

PR: A Different and Better Kind of Democracy

181

mensions. The single-member district system simplifies unnecessarily the process of social discussion . . . . [Proportional representation] has the beneficial effect of getting all citizens to understand the diverse interests and points of view that exist in their society and deepening their understanding of how to fairly accommodate these interests.24 The beneficial effects of proportional representation on democratizing political discussion also extend into the legislature. Like campaigns, legislatures also function as important forums for public discussion of issues. They are not simply places where laws are made, but places were issues are debated and where policy decisions are explained and justified to the public. And again, PR does a better job of encouraging this kind of legislative political discussion than SMP does. As was shown in chapter 4, the more parties present in the legislature, the more points of views injected into policy debates and the easier it is for new ideas to get a hearing. The best way for legislators to ensure that they are passing the best policies is for these bodies to be true marketplaces of ideas where officials consider seriously a wide range of policy alternatives and the arguments for and against them. Legislatures that represent a greater diversity of political views also have a better chance of passing fairer and more just policies—policies that do not neglect important interests in society. An incident that took place in the U.S. Senate in 1993 illustrates how valuable political diversity can be in encouraging more thorough debate and more thoughtful policies. The Senate was about to vote on what was considered a routine measure to approve the request of the Daughters of the Confederacy to renew their patent on the Confederate flag insignia. The measure had been passed overwhelmingly in an earlier test vote, with practically no discussion. Suddenly, the lone black member of the Senate, Carol Mosely Braun rose up and gave an impassioned speech against the measure and threatened a filibuster. She argued that it was insult to African Americans to give official approval to that symbol of slavery. Her outburst provoked a three-hour debate on the floor in which 27 Senators eventually reversed their earlier votes, and the measure was ultimately defeated.25 If no African Americans had been in the Senate that day, it is unlikely that that political debate would have taken place. Democracies are based on the political equality of all citizens. That political equality must take several forms, one of which is an equal chance to participate in the political discussions taking place in a society. As Christiano

182

PR: A Different and Better Kind of Democracy

table 8.1 Competing Visions of Democratic Politics: Majoritarian vs. Proportional Majoritarian Vision

Proportional Vision

Who Should Rule

Only the Majority/Plurality As Many People as Possible

Power

Concentrated

Shared/Dispersed

Who Makes Policy

Single Party

Broad Coalition of Parties

Legislative Power, Chairs, Etc.

Monopolized by One Party Shared Among Parties

Legislative Policymaking Style

One Party Dictatorship

Negotiation and Compromise

Political Minorities

Denied Representation and Power

Given Representation and Power

Purpose of Elections

Choose One Party to Rule

Choose Representatives to Bargain

Political Agreement

As Narrow as Feasible

As Broad as Possible

Majority Rule

Sufficient

A Minimum to be Improved Upon

Types of Majorities

Single

Multiple

Manufactured Majorities

Desirable

Undesirable

Party System

Limited to Two Parties

Open to Many Parties

Party Relationships

Adversarial

Cooperative

Participation in Political Dialogue

Minimized

Maximized

Range of Political Debate

Narrow

Broad

has pointed out, the advantage of proportional representation is not simply that it gives citizens equal power to be represented in the political system, but also that it gives political groups and parties an equal opportunity to participate in a society’s political dialogue. In his words, PR promotes “equality in the process of social deliberation.”26 So if democracy is not simply about voting in elections, but also about people taking part in a public dialogue about the goals that a society should be pursuing, then proportional representation elections can serve to enhance that democratic experience.

PR: A Different and Better Kind of Democracy

183

Signs of a Better Democracy There are many reasons, then, to believe that using proportional representation elections helps to create a better and more fully realized democracy. (See table 8.1 for a summary of the comparison of the majoritarian and PR visions of democracy.) However, if PR actually helps democracies to work better, then we should be able to see that in countries where this system is used. There should be signs that PR countries enjoy a higher quality democracy. And there are. One such sign is higher rates of voter participation. Voter turnout is one indication of citizen satisfaction with the democratic process. Low levels of voter turnout suggest that many citizens feel alienated from the democratic process and do not see it as worth their while to participate in elections. In contrast, a high level of electoral participation indicates that most citizens feel that the democratic process—at least as it is manifest in elections—works well enough that they want to take part in it. It is revealing, then, that voter turnout is appreciably higher in PR countries than in SMP countries—a point demonstrated in chapter 7. Political scientists have also studied directly the level of citizen satisfaction associated with various types of democracies. In general they have found that the design of political institutions—such as elections—has a strong effect on citizens’ political satisfaction. Several studies have taken Lijphart’s typology of consensus vs. majoritarian democracies and used surveys to gauge the level of citizen satisfaction with these political arrangements. For example, Hans-Dieter Klingemann asked citizens in 18 democracies whether they were very satisfied, fairly satisfied, not very satisfied, or not at all satisfied with the way democracy works in their countries. His results show that people in consensus/PR democracies are substantially more satisfied with the performance of their democracy than people in majoritarian democracies—the difference being about 19 percentage points.27 Consensus/PR countries also came out better in another study done of citizen attitudes in eleven European democracies, this one conducted by Christopher Anderson and Christine Guillory. They looked at how well different groups of voters were satisfied with the workings of democracy in their country. In particular, they compared the satisfaction levels of those who voted for the winning parties in the election to the levels of those who voted for the losing parties. Not surprisingly, they found that those whose parties won were always more satisfied than those whose parties lost. However, there was an interesting difference in the responses of voters in consensus

184

PR: A Different and Better Kind of Democracy

and majoritarian countries. In general, the losing voters in consensus/PR countries were more satisfied with how democracy worked than those in majoritarian countries, and the difference between the levels of satisfaction for winners and losers was much smaller in consensus countries. This result makes sense given the workings of SMP and PR systems. It is logical that winner-take-all/losers-get-nothing systems like SMP would produce greater political dissatisfaction on the part of the losers, and that PR systems that are much fairer to political minorities would mitigate their political discontent. As Anderson and Guillory explain: When it comes to citizen access to and participation in the political process, it is important to note that some countries’ institutions are designed to afford greater opportunities for both winners and losers of democratic competition to be represented in the political arena and to implement their preferred policies. Given that consensual systems provide the political minority with a voice in the decision-making process, we expect that the more consensual the set of political institutions in a country, the greater is the extent to which negative consequences of losing elections are muted. Conversely, the more majoritarian the countries’ institutions, the more winners get to have a say and impose their will on the minority.28 Unlike single-member plurality elections, where the losing voters in a district usually receive no representation at all, PR elections ensure that supporters of all parties—even those that lose the election—receive fair representation. Moreover, as seen earlier, power is usually more widely shared in multiparty coalition governments, with even the losing parties sometimes being allotted chairs of legislative committees. A democracy that attempts to include as many people as possible in the governing process would seem to naturally be more likely to have fewer citizens with high levels of political dissatisfaction. Finally, another way to gauge the quality of democracy is to move beyond citizen attitudes and to directly measure how well the policy positions of policymakers reflect those of the public. After all, the ultimate aim of democracy is to ensure that officials pass policies that reflect the political preferences of the citizens. As G. Bingham Powell has put it: “Citizens’ participation in a democracy should influence policy making, pressing it closer to the citizens’ preferences than it otherwise might be.”29 In a study of twenty countries de-

PR: A Different and Better Kind of Democracy

185

scribed earlier in this book, Powell set out to see whether the majoritarian or the proportional influence vision of elections and politics produced policymakers whose political positions were most congruent with those of the public. He found, somewhat unexpectedly, that the advantage went to proportional representation. As he explains: “Each vision expects its ideal electoral process to create good congruence between citizens and policymakers. [However,] they are not equally successful. The proportional influence designs enjoy a surprising advantage.”30 So if what democracy and democratic elections are primarily about is ensuring that policymakers follow the will of the people, PR is the preferable electoral system and produces more democratic results. Or as Powell concluded: “In these countries at this time, the proportional vision and its designs enjoyed a clear advantage over their majoritarian counterparts in using elections as instruments of democracy.”31

PR: A Step Toward True Democracy In the end, the choice between proportional representation and winnertake-all elections is not simply about deciding between two different voting systems, nor even about choosing between two different visions of democracy—it is about choosing between more democracy or less democracy. Single-member plurality elections produce a less democratic political system. When only some people are represented in the political process, we have democracy for only some. When we have partial representation, we have partial democracy. What we have now in the United States is an incomplete democracy. Proportional representation would be an important step toward empowering more people and making our democracy more complete. It would breathe new life into the democratic process. It would give American citizens more of what democracy is all about: greater political equality, more power sharing, fairer representation, increased participation, broader political debate, more political bargaining, and policies that better reflect the public will. Thus if we want true democracy in the United States, if we want democracy for all, we need proportional representation.

9

Objections to Proportional Representation

Clearly, proportional representation has a number of important political advantages over our single-member plurality system. Adopting PR in the United States would dramatically limit the number of wasted votes, ensure that all political perspectives are represented fairly in our legislatures, encourage more issue-oriented election campaigns, create a level playing field for all political parties, ensure the election of more women and minorities to office, and foster a higher level of voter participation. And as was seen in the last chapter, PR would help to reinvigorate our democracy by increasing political equality, encouraging power sharing, and enriching political dialogue. Taken together these advantages make an impressive case for switching into a proportional representation system. However, the argument to this point has been largely one-sided. It has primarily focused on what is wrong with single-member plurality arrangements and what is right with proportional representation. The other side is a set of arguments against PR and in favor of retaining our current system. These arguments deserve careful consideration because any decision about the kind of electoral system that would be best for United States must be based on a thorough appreciation of all the advantages and disadvantages of these rival systems. In that spirit, this chapter will consider the case against PR that has been made by critics of this system. It will describe these criticisms, examine the evidence on which they are based, consider the counterarguments offered by proponents of PR, and draw some conclusions about how serious these concerns actually are.

Objections to Proportional Representation

187

The Case Against PR In examining the case against proportional representation, one feature of critics’ arguments immediately stands out. They rarely attack the most central claim made by PR proponents: that it is more representative and thus has more democratic legitimacy than the SMP system. That PR legislatures tend to mirror the political opinions of the electorate much more accurately than plurality systems is so easily demonstrated that even PR critics accept it. As the Hansard Society noted when it studied the issue of electoral reform in Great Britain, “There can be no doubt that the [plurality] system does not produce a fair representation of the views of people in Parliament. Indeed, supporters of the present system do not attempt to defend it on grounds of fairness.”1 In fact, the empirical evidence for most of the advantages of PR—such as minimizing wasted votes, eliminating gerrymandering, and ensuring fair representation for minorities—is so overwhelming that there can be no question of its validity. Admitting to these important advantages of PR would seem to put defenders of the single-member plurality system at a great disadvantage. But it does leave them with one other plausible line of attack: that the disadvantages of PR elections are so substantial that they outweigh its considerable advantages. PR should be rejected, they argue, because it creates more problems than it solves. The specific charges commonly leveled at proportional representation are: •

• •

•

Proportional representation elections are too complex. Most American voters would not understand the formulas by which PR systems work, and many would find the complicated ballots intimidating. Election administrators also might have difficulty coping with the complexity of the vote-counting and seat allocation procedures. PR systems require people to cast votes for parties. Most Americans, however, prefer to vote for the person rather than the party. PR legislatures are typically ruled by coalitions formed by a number of parties instead of the single-party legislative majorities usually found in SMP systems. These coalitions are inherently unstable and are often unable to govern effectively. Multiparty coalitions are less accountable than one-party majorities. In particular, PR elections make it difficult to throw a party out of power.

188

•

•

•

•

•

Objections to Proportional Representation

The single-party majorities typical of SMP systems have a clear mandate from the voters. Multiparty coalition governments lack such clear guiding messages from the electorate. The proliferation of different political parties under PR would “balkanize” American politics, fragmenting our society into warring political factions. Small parties wield too much power in PR systems. They are sometimes in a position to decide which parties will be in the ruling coalition, and they may also be able to force their own policies onto larger coalition partners. PR encourages political extremism. Because it allows for the election of minor parties, PR also allows radical and extremist parties to elect candidates and thus increase their power and legitimacy. With its large multimember districts, proportional representation undermines geographical representation and the close relationship that exists between representatives and their constituents in single-member plurality systems.

Let us consider each of these criticisms in more detail and see how proponents of PR respond to them.

PR Is Too Complex Critics of proportional representation often complain about the complicated features of these voting systems. When the single transferable vote was used in some U.S. cities earlier in this century, local detractors frequently criticized it for being overly complex.2 Unquestionably, a single-member plurality system is simpler to understand than most PR systems. Under plurality voting, the person with the most votes wins—and that is that. PR systems are more complicated. Counting the votes and determining the winners often involve mathematical manipulations that are not easily understood by most citizens. In party-list and additional member systems, for instance, a formula must be used to determine exactly how votes are translated into legislative seats. In single transferable vote PR, the transfer of ballots from one candidate to another can also become a complicated process. But even if the vote-counting systems in proportional representation are relatively complex, is this really a problem for most voters? The answer is

Objections to Proportional Representation

189

no. Voters need not understand the formulas underlying PR systems in order to use these systems effectively; they need only be able to accomplish such simple tasks as casting a vote for a party or ranking the candidates they prefer. In industrial societies, citizens are not required to master the technical details of all the complex mechanisms they use. Otherwise we would soon have to stop using our computers, DVD players, and most other modern conveniences. Similarly, a full comprehension of the inner workings of a PR election system is certainly not necessary in order to use it easily and effectively. The New Zealand Royal Commission on the Electoral System arrived at a similar conclusion in its investigation of PR alternatives to that country’s plurality system. It examined the charge that PR counting procedures were too complex and concluded that it had little validity: [PR systems] use methods of allocating seats the full details of which may not be universally understood. We do not, however, believe that it is necessary for every voter to understand all the intricacies of any voting system so long as voters can be confident that the system is fair, that counting is carried out by impartial officials under the scrutiny of candidates and parties, and that there are effective and impartial avenues to deal with any allegations of malpractice or unfairness.3 A related complaint is that the ballots in PR systems may be too complex for voters. There would be many more parties and candidates on the ballot and voters would need to cast their votes differently—such as the ranking process required in the single transferable vote. SMP ballots offer voters a much simpler choice between two parties and they need only mark the candidate they prefer. However, the claim that PR ballots would intimidate Americans is not supported by the evidence. Cambridge, Massachusetts, which uses the single transferable vote for its municipal elections, has had no such problems. Few citizens seem to have been deterred by the many choices or by the difficulties in voting by numerical preference. Cambridge has enjoyed an enviably high level of voter turnout during the last decade—49 to 60 percent of the registered voters. And the number of invalid ballots—caused by such voter errors as marking two number 1s instead of 1, 2, 3, and so on—has remained a constant 2 percent of the total votes cast, which is comparable to SMP voting systems.4

190

Objections to Proportional Representation

In addition, a study by Engstrom and Brischetto of the use of cumulative voting in Texas found that Americans adapt quite readily to new voting systems.5 Cumulative voting is widely considered to be one of the most complicated alternative voting systems to use because voters can combine their votes for candidates in a number of different ways. Yet exit polls revealed that only 16 percent of voters thought that cumulative voting was more difficult to understand than traditional American voting systems. Almost twice as many—28 percent—said that the new system was actually easier to understand. Based on this and other studies, Richard Engstrom has concluded: “The notion that cumulative voting is so confusing to voters that they will not be able to cast their ballots effectively can be safely rejected.”6 A British electoral systems expert, David Farrell, has completed a more extensive comparative study that also sheds some light on this issue of complexity. He compared the rates of invalid votes in majority/plurality countries to those in PR countries. If PR critics were correct, then we would expect to find much higher rates of spoiled ballots where “complex” PR systems are used. But in fact, Farrell found only a very small difference of 0.7 percent—which he termed “hardly earth-shattering.” He concluded: “It seems pretty clear that the complexity of the electoral system has a very limited role in determining the number of invalid votes.”7 Ultimately, many of the arguments about the daunting complexity of PR systems are insulting to American voters. They assume that American voters are somehow less intelligent or less sophisticated than voters in other countries. Voters in PR countries have shown no difficulty in casting ballots or choosing among multiple parties. The very high turnout rates in these countries are another clear indication that voters have not been intimidated by the supposed complexities of PR voting. Thus this criticism of PR has no validity. With the proper programs to educate voters about the new voting system and a short period of transition, American voters would quickly adapt to the use of proportional representation.

PR Elections Are Difficult to Administer Sometimes critics suggest that adopting PR in the United States would make administering elections much more difficult and expensive. Election officials would have to learn new vote counting methods and seat allocation

Objections to Proportional Representation

191

formulas. Some PR systems might also be incompatible with current voting machines and new ones would then have to be purchased to accommodate these new voting systems. However, experience with proportional representation systems in other Western democracies has shown that the election administrators there have had no problems mastering the vote-counting and seat allocation procedures. So there is no reason to believe that this should be a problem in the United States. Also, most PR systems can be used with current voting machines—with the single transferable vote being the possible exception.

Voting for Parties Instead of Individuals One common myth about proportional representation is that voters in this system can only endorse a party—that they cannot vote for individual candidates. U.S. Representative Charles T. Canady voiced this concern when he noted, “proportional representation systems have been criticized for turning the focus away from individual candidates and toward conformity to party, as voters are no longer choosing between candidates but between parties.”8 Since many American prefer to vote for the person, not the party, this would seem to be a big drawback for PR. However, there is only one form of PR—closed party-list voting—in which voters cannot express a preference for a particular candidate. In this form, the parties provide a list of candidates and voters simply endorse one party slate or another. But only a very few countries use this form of PR. All the most popular forms of PR allow voters to choose a particular candidate. For instance, in open party list voting, voters can vote for a specific candidate—with the vote counting both for that candidate and for the party. The candidates with the most votes move to the top of the party lists and thus have a better chance of being elected. The single transferable vote was specifically designed as a candidatecentered voting system. Voters do not cast votes for parties, but for individual candidates. The mixed-member system of proportional representation also requires voting for specific candidates. In this system, half of the legislative seats are filled by single-member district elections in which voters choose the particular candidate they want to represent their local area. Thus there is nothing incompatible about wanting to vote for specific candidates and wanting to use proportional representation.

192

Objections to Proportional Representation

Multiparty Legislatures Undermine Stable and Effective Government Undoubtedly one of the most serious concerns that many people have about proportional representation is whether it will undermine effective government. The worry is that legislatures with many parties will find it difficult to pass legislation and thus become mired in gridlock. Since Americans have no experience with multiparty legislatures, they often have difficulty imagining how they could work efficiently. The views of John Bibby are typical: It is hard to imagine how introducing more third- or fourth-party representatives and senators into the mix would facilitate more timely or effective policymaking . . . . How could a three- or four-party system with small and more cohesive parties than the existing congressional parties and with more polarized activists as their support base contribute much to the policymaking process other than delay, intensified conflicts, greater divisiveness, and gridlock?9 This concern is understandable. If we have a multiparty system in which all of the parties are going their own way, how can there possibly be any agreement on policy issues? The answer, however, is that the parties do not all go their own way in PR legislatures. Experience shows that parties invariably form into coalitions. There are often two broad coalitions—one on the left and one on the right—and so these multiparty legislatures often end up operating much like a two-party legislature. The ruling coalition, which controls a majority of the seats in the legislature, usually has no trouble passing policies. In fact, PR countries with these multiparty coalitions often have a better record of efficiently passing legislation than our own Congress. So we do not have to rely on our imaginations to see how a multiparty legislature could work efficiently, we need only look at how PR has worked in practice in other Western democracies. PR critics tend to ignore this record of policymaking efficiency and insist that multiparty coalition governments are inherently unstable. They maintain that there is always a danger of these coalitions falling apart and undermining effective government. Italy and Israel are inevitably cited as typical examples of this problem. Italy in particular has often been plagued by unstable coalitions, which sometimes have lasted only one or two years, or

Objections to Proportional Representation

193

even less. The potential for this kind of instability is the most frequent and serious complaint that critics raise about PR. They argue that there is an inherent tradeoff with proportional representation: that it may make legislatures more representative, but it also makes them more unstable and less able to govern effectively. As one critic explained: “proportional representation promises instability just as surely as it promises fairness.”10 Former Republican Senator Slade Gorton has had the same misgivings and feels that it would be a disastrous mistake for the United States to adopt “the worst system of all, proportional representation. All we need is to be like Israel does [sic]. Boy, everybody gets represented that way and no decisions can ever be made.”11 These PR critics believe that we would be much better off sticking to our SMP elections and our two-party legislatures. This arrangement is thought to produce much more stable and effective government. Typically one party wins the majority of seats—so there is little worry about disintegrating coalitions. Also, one-party ruling majorities have a much easier time passing legislation because they needn’t negotiate with coalition partners. However, this criticism of proportional representation misses the mark for several reasons. First of all, critics fail to recognize that the structure of American political institutions makes the problem of unstable legislative coalitions much less of a concern. Even if ruling coalitions were not perfectly stable, this would not be as large a problem here is it would be in a European country. Most European democracies have parliamentary systems where the breakup of a ruling coalition has serious political ramifications. At the very least it means a reshuffling of the cabinet, but it may require a prime minister to resign and new elections to be held. However, this is not true in the United States. Unlike prime ministers, presidents and governors are not elected by legislatures, and their rule is not dependent on the presence of a stable ruling coalition. Thus unstable coalitions would be much less of a problem in our political system. Moreover, even if we were to experience some shifting of coalition partners in our legislatures, this would not be all that different than what occurs now in many of these bodies. The Republican and Democratic parties in Congress are not monolithic, but are made up of coalitions of various political groups. And these coalitions do not always stay together. It is not at all unusual, for instance, to find conservative Democrats sometimes joining with Republicans to form a legislative majority—or moderate Republicans temporarily allying themselves with the Democrats on some issues. Thus

194

Objections to Proportional Representation

the presence in PR legislatures of changing coalitions would in some ways simply be business as usual in American politics, and certainly not the political disaster that critics predict. In fact, as we saw in chapter 8, some scholars argue that shifting coalitions are actually an advantage because they facilitate power sharing and allow more parties to participate in the policymaking process. However, the main problem with the charge that multiparty PR coalitions are inherently and intensely unstable is that it is simply not true. If the critics were right about extreme instability being inherent to proportional representation, then we would expect to see this problem occurring in most countries using this electoral system. But we do not. As Matthew Shugart argues, it is simply inaccurate to suggest that the experience of Italy is typical of PR systems. Anti-PR polemicists take as given that all PR systems can be damned by stating one simple word: Italy. While there are indeed PR systems that have been characterized by frequent inter-election changes of government . . . such characterizations do not apply to PR systems as a class.12 The record of PR use in dozens of European countries over many decades shows only a few instances—Italy being one of them—where instability has been a serious problem. The vast majority of PR countries have enjoyed stable and efficient governments. Once ruling coalitions are formed, there is a great deal of incentive for them to stay together. These coalitions usually last many years, and in some cases, have gone on for decades. Studies done on the durability of governments under various party systems can help to shed some light on this issue. This research has indeed shown a tendency for one-party governments to last somewhat longer than multiparty coalitions—which is not particularly surprising or alarming. What is more significant is that this research has found no systematic evidence of widespread or persistent government instability in PR countries. The vast majority of multiparty coalitions are reasonably stable and govern quite effectively. In fact, David Farrell found that several PR countries, including Austria, Iceland, Ireland, Luxembourg, and the Netherlands, have actually had more stable governments than Great Britain, a leading example of steady one-party government.13 If this is true, how can the problem of countries such as Israel or Italy be explained? As many electoral scholars have noted, Italy and Israel have un-

Objections to Proportional Representation

195

usual and “extreme” forms of proportional representation.14 Both of these countries have had unusually low electoral thresholds, which have contributed to their political problems. The threshold is the minimum portion of the vote that a party must receive in order to be awarded any seats at all in the legislature. Very low thresholds allow many small parties into the legislature and this can complicate the process of creating stable ruling coalitions. For example, in Israel that threshold is only a little over 1 percent, which allows numerous parties, sometimes over a dozen, to elect members of the Knesset. This can make it difficult to put together stable coalitions. But such low thresholds are the exception, not the rule in PR systems. The vast majority of PR countries use a more moderate form of this voting system, with a much higher threshold. As a result, the average number of legislative parties in Western democracies using PR is four to five—a very workable number. In Germany, for example, a party must receive at least 5 percent of the national vote to be eligible for seats in their parliament. While this may still seem like a low figure, in practice very small parties are usually unable to reach this threshold and the German parliament typically has only four or five parties. Party systems of this size allow sizable political minorities to be represented, but also allow for stable coalitions to be built of two or three parties. This moderate form of PR is what reformers have been proposing for the United States. For example, the PR system used in Cambridge, Massachusetts has a threshold level of about 10 percent, which has worked well for decades to allow for both diverse representation and workable local government.15 In fact, research on other U.S. cities that have used proportional representation elections has found these city councils worked very effectively and that gridlock was rarely a problem. Reporting on studies of PR use in five cities in Ohio, Kathleen Barber concluded: “Little evidence can be found to support the fear that PR would cause fragmentation of council behavior; indeed, improved consensus building was observed and appears to be linked to this electoral system.”16 For all of these reasons, many leading electoral scholars have concluded that proportional representation is not a threat to political stability and that it can produce effective and efficient government. David Farrell of the University of Manchester has observed that “on the basis of the evidence . . . the supposed trade-off between the proportionality of the electoral system and measures of governmental or system stability appears, for the most part, conspicuous by its absence.”17 Arend Lijphart is even more emphatic about this

196

Objections to Proportional Representation

point. He studied the effectiveness of PR and SMP governments in managing the economy and concluded: The conventional wisdom is wrong in positing a trade-off between the advantages of plurality and PR systems. The superior performance of PR with regard to political representation is not counterbalanced by an inferior record on governmental effectiveness; if anything, the record of the PR countries on macro-economic management appears to be a bit better than that of the plurality systems—but not to the extent that the differences are statistically significant. The practical conclusion is that PR is to be preferred over plurality since it offers both better representation and at least as effective public policymaking.18

SMP Systems Are Prone to Their Own Form of Instability Critics of proportional representation are not only wrong about its being plagued by unstable coalitions, they also ignore the fact that two-party SMP systems tend to foster their own form of political instability. These systems are characterized by complete changeovers in power, with one party replacing the other as the majority in the legislature. When Republicans take over from Democrats, or vice versa, this often leads to dramatic political changes as the new party reverses many of the policies of the previous party. A typical example of this occurred in 1994 when the Republicans took over the House of Representatives. Even though the change in voter sentiment was relatively modest—the Republicans only won 7 percent more of the vote than in the previous election—there was a 100 percent changeover in power. And the new majority party immediately declared a “Republican Revolution” and attempted to effect vast changes in tax policy, foreign policy, environmental regulation, abortion policy, etc. Such abrupt reversals of policy direction can be very disruptive. People who have become used to one set of tax policies or environmental regulations may suddenly find themselves having to respond to a very different set of requirements. And when the previous party returns to power, another reversal in political direction is likely. In contrast, multiparty PR legislatures are less prone to complete turnovers of the parties in power. A more typical result is a change in the ruling coalition with some parties leaving and some joining to form a new ma-

Objections to Proportional Representation

197

jority. Incremental changes in these coalitions tend to produce incremental policy changes as well, instead of the wild swings in policy seen in SMP countries. S. E. Finer has argued that policy continuity is especially important in the area of economic policy, and that PR governments are better at producing steady centrist economic policies.19 In this regard, it is interesting to note that in Great Britain during the 1970s, businesspeople complained that it was difficult to plan effectively when government taxing and regulatory policies varied dramatically depending on whether Labour or the Conservatives were in power. Some parts of the business community supported a change to proportional representation in the hopes that a multiparty coalition legislature would create a more stable and predictable business climate.

Lack of Accountability Ironically, some critics have now abandoned the notion that multiparty coalitions are inherently unstable and have instead started arguing that they are too stable. They maintain that these coalitions are so persistent that it becomes difficult to throw a party out of power. The issue is one of accountability. Some believe that the main purpose of elections is to hold the ruling party or parties accountable for their actions in office. If the public is dissatisfied with their performance, they can hold them accountable by removing them. As Michael Pinto-Duschinsky has explained: “What is really important is not to select a new government, but to be able to oust an existing one.”20 Critics of PR maintain that this is much easier to do in a two-party SMP system where a single party usually has the legislative majority. Disgruntled voters can simply replace one party with the other. This is harder to do in a multiparty PR system. In these systems a party may lose a great deal of public support but still continue in power as part of a coalition of parties. For example, voter support for the largest party might fall from 47 to 35 percent in the next election, and yet this party could remain in power by forming a coalition with another party that received 20 percent of the vote. Germany is often cited as an example of this problem, where the larger Christian Democratic Party and the smaller Free Democrats formed a ruling coalition that lasted almost two decades, from 1982 to 1998.21 It is true that a complete turnover in government is less likely to occur in a multiparty PR system than in a two-party SMP system. But as Arend Lijphart has explained, this is not necessarily a disadvantage.

198

Objections to Proportional Representation

In the Netherlands, for instance, there has never been a complete cabinet turnover since 1945, but that does not mean that the same parties were in power all the time: center-left cabinets have alternated with center-right cabinets, generally in response to electoral verdicts. It is not clear to me why less frequent complete replacements should be preferable to more frequent partial replacements.22 In fact, as seen earlier, one could argue that the partial changes in governments that are the typical result of PR elections are more desirable than the complete turnovers found in SMP elections. Switching one ruling party for another creates a great deal of policy instability and may not accurately represent voter sentiments. Voters, for instance, may be dissatisfied with the Democrats’ economic policy, but agree with their social and environmental policies. But replacing them with a Republican majority creates large changes in all of these policy areas. The incremental changes in PR coalitions are more likely to allow for some continuity in policies and more accurately represent the policy preferences of most votes. But there is an even more serious problem with this criticism of PR. Critics neglect to point out that SMP elections are also guilty of making it difficult to throw the ruling party out of power. As was seen in chapter 1, SMP systems are prone to the problem of manufactured majorities, where a single party receives less than a majority of the vote but a majority of the legislative seats. This problem makes it difficult for voters to hold the ruling party accountable. In Great Britain, for instance, in eight of the fourteen national elections between 1950 and 2000, the ruling party remained in power even though the majority of voters were dissatisfied with it and voted against it. The same thing occurred in New Zealand when it used SMP elections. Between 1954 and 1993, the majority of New Zealand voters cast ballots against the ruling party fourteen times, but that party was removed only six of those times.23 Thus when we apply the standard of accountability to SMP elections, it turns out that they often fail this test as well.

PR Coalitions Lack Mandates Another alleged advantage of the single-party legislative majorities common to SMP systems is that they have a clear and unambiguous mandate from the voters. If a majority of the voters favor the Republican party, for in-

Objections to Proportional Representation

199

stance, that party has an unmistakable mandate to institute its policies. And since that party has a majority in the legislature, it can easily keep its promises to the voters. In contrast, multiparty coalition governments are thought to lack such clear guiding messages from the electorate. If two or three parties with differing policy platforms make up the ruling coalition, what exactly is their mandate? Obviously, the policies eventually pursued by such a coalition will depend on negotiations among the parties. But these negotiations will require some parties to compromise on the policy positions in their platforms—thus deviating from the mandate of their supporters. However, this criticism of PR rests on several inaccurate assumptions. First, it assumes that voters in a PR system are not voting for a coalition— that they have no idea of what the ruling coalition might look like after the election. But in fact this is often not true. In many PR countries, such as Germany, parties announce ahead of time which other parties they will join with to form a ruling coalition. And even where there are no formal preelection agreements, voters often have a very good idea of the likely coalition partners resulting from an election.24 In addition, in PR systems where people have multiple votes, they can actually vote for a particular coalition. Germany, for instance, uses the mixed-member form of PR where voters cast one vote on the party list portion of the ballot and another vote for a candidate in a single-member plurality race in their local district. Voters may cast their party list vote for a minor party and their district vote for a major party, thus indicating their preference for coalition partners. Even when there are not clear pre-election coalitions, there is still good reason to believe that the multiparty coalitions that end up ruling in PR countries have valid mandates from the public. Indeed, a coalition of two or more parties may actually have a broader mandate—one that more accurately represents the political sentiments of a wider range of the electorate— than a single party. This notion is supported by the work of G. Bingham Powell, who found that the policy positions of policymakers in these multiparty governments more closely resemble the views of most of the public than the positions of those in single-party SMP governments.25 The other questionable assumption in this criticism is the notion that singlemember plurality elections inevitably give a single party a clear political mandate. To produce a true mandate, voters must be voting sincerely—that is, they must be casting a vote for a party that truly represents their own specific ideological and policy preferences. But this may not happen in singlemember plurality elections, for two reasons. First, as we saw in earlier chap-

200

Objections to Proportional Representation

ters, SMP encourages two-party systems—systems in which voters may identify only weakly with one of the two major parties. Indeed, they may embrace one party only because they see it as the lesser of two evils. They may even prefer a minor party candidate, but support a major party candidate merely in order to not waste their vote. This kind of unenthusiastic voting is hardly likely to produce the ringing endorsement described by advocates of SMP systems. Second, it is misleading to talk of mandates in an SMP system that encourages candidates to avoid specific policy stands and to campaign on personality, negative ads, and ambiguous promises. To give a true mandate, voters must have policy-oriented campaigns and a variety of different parties from which to choose—both of which are more likely in PR systems than in SMP systems. A more important flaw in the mandate argument is that the single-party legislative majorities produced by SMP elections often do not even represent the majority of voters. Again the problem here is manufactured majorities, where a party wins a majority of seats with fewer than 50 percent of the votes. These manufactured majorities are quite common in SMP elections. In six plurality countries, 43.7 percent of the elections held between 1945 and 1996 produced manufactured majorities in the national legislatures.26 In Great Britain the Conservative party ruled for almost two decades but never once won a majority of the vote. Thus the claim of a majority mandate is patently false in many cases. Instead, we have a party with the support of only a minority of the voters pursuing policies opposed by the majority.27 If the Conservatives had been forced to rule in a coalition with the centrist Liberal Democrats, their policies would have undoubtedly been more moderate and more in keeping with the views of the majority of the electorate.

PR Would Balkanize American Politics Some critics of proportional representation are concerned that it will increase political conflict. The focus here is not on the legislatures, but on society as a whole. The contention is that the proliferation of different political parties would “balkanize” American politics, further fragmenting our society into warring political factions. John Bibby, in defending our current twoparty system, has warned that a multiparty system “would in all likelihood further fragment society and heighten divisiveness.”28 Others, such as Abigail Thernstrom, worry that PR will increase racial polarization: “There will be a

Objections to Proportional Representation

201

David Duke party and a black nationalist party.”29 Echoing these sentiments, Andrew Busch has argued that result of PR would be “increased ethnic divisions” and “the deliberate cultivation of ethnic identity or separatism.”30 We already have serious political, racial, religious, and economic divisions in our society, and the concern is that candidates and parties in a PR system would seek to exploit those divisions, waging campaigns that stressed group differences and made divisive appeals to narrow interest groups. We are better off, critics argue, with a party system that brings people together, rather than splitting them apart. The benefit of a two-party system is that these parties are large, umbrella organizations that bring different political factions together to negotiate and compromise. In contrast, a multiparty PR system would simply partition people into smaller parties and further entrench our political divisions. Given the political diversity that exists in the United States, it certainly makes sense to encourage a political system that facilitates dialogue and negotiation between these political factions. But the critics of PR are mistaken in their assumption that winner-take-all elections and a two-party system are the only, or the best, way to manage and reduce political conflict. In fact, as seen in chapter 8, one of the strengths of proportional representation is its ability to promote negotiation and compromise. In practice, both two-party and multiparty systems encourage negotiation among political factions; they simply do it at different times in the election process. In a two-party system, negotiations usually take place before the election. During conventions and primaries, the various groups in our large umbrella parties try to settle their differences and build an electoral coalition. In a multiparty PR system, that process of political bargaining takes place in the legislatures after the election. After the various groups elect their representatives, then they engage in negotiation and coalition building. SMP legislatures are usually dominated by one party that has no need to work with other parties; but in multiparty PR legislatures, parties are forced to bargain and cooperate with each other to form a ruling coalition. This process of negotiation and consensus building may even extend into the administrative branch where several different parties may occupy seats in the cabinet. However, while both party systems attempt to mitigate conflict, a multiparty system is often more effective at doing so in the long run. This is because the two-party system tends to ignore or paper over many conflicts. For instance, real negotiation often does not take place in party conventions because the mainstream of the party knows that it will be able to get its way

202

Objections to Proportional Representation

most of the time. White mainstream Democrats, for instance, may not have to take the demands of African-American Democrats seriously because they know that this group has no choice but to support Democratic candidates. In addition, winner-take-all elections usually produce legislatures with less diversity than in society, and this hampers their ability to manage political conflict. Only majorities are represented and minority political groups are either left out or underrepresented. This approach gives the appearance of political consensus, but it simply masks the underlying political tensions. In the 1950s and 1960s, for instance, the presence of all-white city councils in multiracial cities gave the illusion of a calm and stable political situation—but it did nothing to address the deep racial tensions simmering beneath the surface. Eventually these tensions exploded into political protests and urban riots. Experience abroad also shows how winner-take-all elections can give the impression of a tranquil and undivided society, while actually serving to exacerbate political tensions. Rein Taagepera and Matthew Shugart point out that this is exactly what happened for many years in Northern Ireland. Under single-member plurality voting rules, the Protestant majority completely dominated the Parliament. The election system “reduced the Catholic representation much below their PR share of seats, denying them even the role of a meaningful parliamentary opposition. The situation looked stable for fifty years—until all too many Catholics replaced their meaningless ballots with bullets.”31 A better way to address political conflicts is to acknowledge them and bring them into our legislative bodies. Only when all of the opposing groups are at the political table can a serious effort at negotiation and conciliation take place. This is the approach embodied by a multiparty PR system. Racial and ethnic minorities also have a much better chance of having their concerns addressed if they actually have representatives in government. Some political observers in Cambridge, Massachusetts believe that the ability of various racial and ethnic groups (which include Italian Americans, Irish Americans, Jewish Americans, and African Americans) to use proportional representation to elect representatives to the city council and school committee has helped to smooth relationships among these groups. As one former member of the school committee notes: “[Proportional representation] is the reason Cambridge didn’t burn during the years of demonstrations, the reason desegregation of the schools was achieved without any significant disruption.”32 In this way, PR attempts to build consensus not by excluding

Objections to Proportional Representation

203

some interests and parties from the legislature but by including as many as possible. In the long run, there is a better chance of resolving conflicts if all political groups participate in the governing process. While some Americans worry that proportional representation will worsen group conflict, PR is widely considered by many election experts to be the best way to mitigate intense political factionalism and to bolster political stability. Arend Lijphart, for instance, argues that the inclusive approach of PR can be particularly useful in societies divided by deep racial and ethnic conflicts: Divided societies, both in the West and elsewhere, need peaceful coexistence among the contending ethnic groups. This requires conciliation and compromise, goals that in turn require the greatest possible inclusion of representatives of these groups in the decision-making process. Such power sharing can be arranged much more easily in . . . PR systems than in . . . plurality systems.33 Consider, for instance, the case of South Africa—a country that was torn apart by violent political and racial conflicts. When it emerged from apartheid and adopted a democratic form of government, all sides eventually agreed that proportional representation would be the best electoral system for their diverse society. The African National Congress explicitly rejected Americanstyle elections, even though a plurality system would have worked to their political advantage by overrepresenting their party in the assembly. They realized that this system would seriously underrepresent whites and other minorities, and in this way actually exacerbate political conflict.

Small Parties Wield Too Much Power Another fear of critics of proportional representation is that it allows minor parties to gain excessive amounts of power in the legislatures. Their most serious complaint is that a small party may be in position to arbitrarily determine the composition of the ruling coalition. For example, if a large leftist party wins 45 percent of the vote, another large party on the right wins 40 percent, and a small centrist party wins 15 percent of the vote, that gives the small party the balance of power and puts it in the position of “king-maker.” It can decide which party it joins with to form a legislative majority. Critics

204

Objections to Proportional Representation

complain that this allows a small party, not the voters, to decide who wins the election. This situation is particularly troubling if the small party bypasses the party that won the most votes and forms a coalition with the second-place party. Such outcomes are possible, but in reality several factors usually minimize the likelihood of such a problem arising. First, many small parties are not politically positioned to play this role. They are either on the political left or right and would not feel comfortable forming a coalition with a party on the other side. In this country, for example, the Greens would be unlikely to form a coalition with the Republicans. Second, in PR countries where coalition governments are common, most voters have a good sense of what the probable coalition arrangements will be. Indeed, in many of these countries, small parties announce their likely coalition partners before the election, which greatly reduces the importance of post-election bargaining in creating a government. Finally, studies have shown that the worst-case scenario—in which a small party forms a coalition with a party that came in second in the polls— is a highly unusual occurrence. Jack Vowles of the University of Waikato in New Zealand found that this happens only about 12 percent of the time— which happens to be the same frequency that SMP elections allow the second place party to gain a majority of the seats.34 So in this regard, PR is no worse than plurality elections. Nine times out of ten, the smaller party forms a coalition with the most popular large party; and in this sense it is following, not frustrating, the public will. Another concern about small parties is that they may be able to exert undue policymaking power in ruling coalitions. Critics fear that they can use their leverage to wring concessions from their larger coalition partners, perhaps even causing the adoption of radical policies not supported by most voters. William Kimberling, deputy director of the National Clearinghouse on Election Administration in Washington, D.C., expressed his concern this way: “Any scheme of proportional representation has built into it the ability to create small factional parties. These groups try to win enough seats to become swing votes. It gives a disproportionate advantage to small, vocal committed or ideological groups.”35 It is certainly true that proportional representation is apt to give some power to minor parties—power that is exercised by taking part in coalitions to pass legislation. After all, proportional representation is intentionally designed to broaden power-sharing in government; it aims to allow minor par-

Objections to Proportional Representation

205

ties and the people they represent some say in government. As noted in the previous chapter, the political philosophy underlying PR is that as many groups as possible should participate in the governing process and that even minor parties should have a turn at helping to shape public policy. Is sharing power in this way politically undesirable? Not usually. In fact, including minor parties in legislative coalitions may have a number of positive political benefits. The bargaining in coalitions encourages compromise policies that are acceptable to a wider variety of parties and their supporters. And as the studies of G. Bingham Powell have shown, the result is that the political views of policymakers in these ruling coalitions are much more likely to reflect the views of most of the citizens.36 Moreover, the record of PR use in European democracies provides very few examples of small parties’ acting as “the tail that wags the dog.” It is actually rare for a small party to wring inappropriate concessions for a large party, in part because it would fear that this kind of political extortion could provoke a voter backlash. This was the conclusion reached in the report of New Zealand’s Royal Commission on the Electoral System: “Overseas experience indicates that the extent to which a minor party can exact an excessive price for its support is limited. If voters consider a small party to be demanding unreasonable concessions for its support, or to have been irresponsible in changing the Government, that party can be heavily penalized in the next election.”37 In the end, then, the ability of small parties to wield disproportionate power in proportional representation systems should not be a major concern.

PR Encourages Political Extremism Another concern about proportional representation is that it might encourage political extremism in the United States. Ferdinand Hermens, for instance, contended that because PR allows the representation of small parties, extremist political groups could be able to gain a political foothold in the legislature. They could then use this forum to increase their appeal and legitimacy. The result is that “democratic government can easily be perverted from within.”38 In his 1941 book arguing against the use of PR in the United States, Hermens described the rise of Hitler and the Nazi party in Germany as a typical example of this problem.39 He suggested that under plurality rules, the Nazis would have had difficulty winning seats in the German parliament. But

206

Objections to Proportional Representation

under PR rules, Nazis were able to elect representatives and ultimately build their power to a point where they could abolish democratic government. Critics of PR also cite another example, this one closer to home—the single transferable vote elections held in New York City in the 1930s and 40s. In 1941 a Communist was elected to the city council in these PR elections, and two were elected in 1943 and 1945. Many mainstream politicians found this a disturbing development. Both Democratic and Republican leaders often cited the threat of communism as a major concern during the effort to repeal PR in 1947. The leader of Tammany Hall, Frank Sampson, urged voters to “throw out this Stalin Frankenstein” and predicted that “PR, a foreign political theory that has created confusion with the blessing of the Kremlin, will meet its Waterloo on the sidewalks of New York.” Robert Moses, a prominent Republican, also saw PR as a threat to democracy: “No democratic system can exist in any form of government save under the twoparty system. I want to see the two-party system restored in our city, instead of the three, four, and five party system which now exists.”40 In urging a vote to repeal PR, the New York Times argued that “it has fostered the growth of splinter parties, seating Communists and other radicals who could not, by normal majority and district voting methods, have hoped to become members and giving them an official sounding board for views shared by only a meager fraction in the electorate.”41 Given the strength of such sentiments, Zeller and Bone concluded in their study of PR in New York City that “there is no doubt that the one issue above all others responsible for the repeal of PR in 1947 was Communism.”42 As these examples illustrate, proponents of SMP elections celebrate rather than condemn the tendency of this system to discourage the election of minorparty candidates. For them the bias against minor parties is a way of preventing undesirable and undemocratic political forces from gaining any political foothold. The resulting legislatures may not be the most representative, but they are more likely to resist the threat of extremist elements in society. This criticism of PR is problematic in several ways. First, using the example of the rise of the Nazis in Germany to condemn PR is simplistic and misleading. The rapid growth of the Nazi party in the 1930s is widely acknowledged to have been due to a complex combination of factors, primary among them the economic depression of that time and the association of the Weimar Republic with Germany’s defeat in World War I. Given these conditions, the appeal of the Nazi party was strong, and its power would probably have grown irrespective of the electoral system in place. Indeed, if Ger-

Objections to Proportional Representation

207

many had had single-member plurality elections in 1932, Hitler probably would have taken power a year earlier than he did under PR. In 1932 the Nazis had become the largest party in Germany, and the tendency of SMP elections to overrepresent the largest parties would probably have given the Nazis an absolute majority in the national legislature. In fact, at Nuremberg Hermann Goering claimed that if Germany had had SMP elections, the Nazis might have won every single seat in the parliament that year.43 Furthermore, if PR opens the door to antidemocratic extremism, how do we explain that numerous other European countries also had PR election systems during the 1930s but nevertheless enjoyed continued stable and democratic government? For example, proportional representation elections in Scandinavia consistently produced moderate, social democratic governments. This was probably because these countries had more stable political and economic environments than Germany and more deeply rooted democratic traditions. In short, the success of the Nazis had more to do with the specific political, economic, and social conditions in Germany at that time and was not a function of the presence of PR systems.44 Jack Vowles, commenting on the tumultuous course of European politics in the 1920s and 1930s, has concluded: “There is an intensifying consensus that the role of electoral systems in the collapse or survival of democracy was at most quite marginal.”45 This criticism of PR also ignores the entire postwar record of proportional representation. For the last fifty years Germany and other PR countries in Europe have enjoyed moderate and democratic governments—despite the presence in several countries of small extremist groups. Part of this success is clearly due to the use of voting thresholds, which serve as barriers to these small extremist parties. Only parties that can garner at least 3 to 5 percent of the vote nationwide can gain seats, which tends to filter out most radical political sects. Thus if we consider the entire record of how PR has worked in practice, the fear that it encourages the subversion of democracy by extremist groups is clearly unsubstantiated.

Are Fringe Parties a Danger? To be fair to the critics, however, it is true that PR does make it more likely that fringe political parties will gain representation. In a PR system, small radical parties on the far right or far left might be able to overcome

208

Objections to Proportional Representation

vote thresholds and elect members to government—particularly in cities and counties in which such parties may have a concentrated local following. However, this may not be as alarming as it first appears. Keep in mind that most fringe political parties are not extremist and do not threaten democracy. An unfortunate but persistent tendency in American politics is to label nonmainstream parties as subversive and extremist—irrespective of their actual philosophies and policies. Certainly the Libertarian party, with its radical political philosophy, is on the fringe of American politics. Libertarian policy positions—such as the legalization of prostitution and drugs, the prevention of virtually any government intervention in the economy, the privatization of nearly all government services, and the abolition of the income tax—are certainly not mainstream, but the party is clearly committed to democratic norms and processes. The same is also true of most fringe parties on the American left. A revealing example of how extremist labels can obscure political reality is the case of Pete Cacchione, the Communist elected to the New York city council in 1941. Contrary to the popular cold war image of the wild-eyed, antidemocratic communist demagogue, Cacchione was a soft-spoken man who proved to be a diligent and widely respected council member. He saw his role in the council as being to fight “for better housing, lower taxes for low-income groups, control of prices, a curb on excessive profits and an end to all practices of discrimination whether directed against the Negro people, the Italian people, the Jewish people, or the foreign born.”46 His record won him a wide following among working-class, minority, and liberal voters in New York. A mere one-tenth of his votes came from members of the Communist party. The Consumers Union, a middle-class professional good government group, noted after his first term in office that he was “a conscientious councilman with a very good voting record.”47 In 1943 Cacchione received the most first-place votes of any person running for city council in New York City—hardly the “meager fraction in the electorate” described in the New York Times editorial noted earlier. Looking back, many of the so-called extremist parties in American history were simply offering ideas that were ahead of their times. Politicians from the two major parties often accused fringe parties like the Populists and the Socialists of being subversive, yet many of the policies championed by these parties—such as women’s suffrage, child labor laws, and social insurance—are now considered quite reasonable and have become law. What is “extremist” today may well be mainstream in several decades. In seeking to protect us from those relatively rare fringe parties that truly are threats to democratic

Objections to Proportional Representation

209

norms, the advocates of single-member plurality elections prefer rules that eliminate virtually all non-mainstream parties—including those that represent responsible political minorities and those with potentially valuable ideas to contribute. This is a classic case of throwing the baby out with the bath water.

A Worst Case Scenario Clearly, most fringe parties pose no threat to democracy and indeed may even enhance it. Nevertheless, it cannot be denied that some elements in American politics truly are extremist—their goals and policies violate our most basic notions of democracy, equality, and human rights. It is fair to ask whether the adoption of PR would aid these groups. Consider, for instance, a representative of the neo-Nazi movement or the Ku Klux Klan running for election. Candidates with ties to such racist organizations, such as David Duke in Louisiana, have actively sought office in the United States. Such a candidate certainly would stand a better chance of getting elected under proportional representation rules. But should such possibilities be a sufficient reason for rejecting proportional representation? In answering that question, two points must be considered; one is a matter of political principle, the other a matter of practical politics. The matter of principle has to do with the need to have a consistent commitment to the right of fair representation. Do not all citizens in a democracy, even those whose views are considered loathsome by many in society, deserve representation in the political system—so long as these citizens are sufficiently numerous to elect a representative? When we begin to deny representation to political parties based on their particular views, we set a dangerous political precedent. If representation is a basic right in a democracy, it should be extended to as many political groups as possible, even to those whose ideas we find bizarre or abhorrent. In the same way that the Constitution protects the free speech rights of all political groups, we should protect the right to representation for all groups as well. The second set of considerations is more practical. If a few such extremists were to be elected to office, would this have a significant effect on our political system? Probably not. For example, even if PR were to allow for the election of a few white supremacists, it would also give more representation to racial minorities. Indeed, the increased representation of African Americans, Latinos, and Asian Americans would probably more than offset the

210

Objections to Proportional Representation

slight influence that one or two white supremacist officials might have in our legislatures. In addition, extending political representation to such extremist groups might encourage them to moderate their views. The political culture in most legislatures encourages compromise, not confrontation. Finally, the ability of such groups to have a political voice through the electoral system could decrease their sense of political frustration and make them less likely to employ violence or other undesirable means to attract attention to their views. In this regard, it is interesting to note that a recent study of voting systems in other countries found that the use of proportional representation has discouraged discontented ethnic groups from “engaging in extreme forms of resistance to the status quo.”48 Ultimately, the importance we attach to the possibility of PR encouraging political extremism is a matter of political judgment. It depends largely on how likely it is that such groups could really prove a threat to democracy in the United States. Is democracy so shaky in the United States as to be endangered by a few small extremist parties? This hardly seems to be the case. Moreover, as seen earlier, the history of PR in democratic countries suggests that any risks from extremist elements is exceedingly low. Looking at how PR has worked in the United States, Steven Mulroy has reached a similar conclusion: “There is no evidence that where alternative systems have been tried in the United States, there has been any marked increase in the number of candidates considered ‘radical’ or ‘extremist,’ let alone the number elected.”49 Richard Pildes and Kristen Donoghue also looked explicitly at the issue of extremist candidates in their study of the use of a semiproportional system, cumulative voting, in Chilton County, Alabama. They noted “a longstanding concern about cumulative voting is that it enables the election of fringe or extremist candidates . . . . After seven years, there is no evidence of this problem arising in Chilton County. While the potential for election of fringe candidates exists in theory, it has not occurred in practice.”50 If the risk of extremism is this minimal, it hardly makes sense to prohibit the use of PR in the United States on those grounds. Even if a few extremists were elected to office here, it would seem a small price to pay for an electoral system that has all the numerous and considerable political advantages described in earlier chapters.

PR Undermines Geographic Representation Another complaint is that PR would eliminate local district representation. SMP elections typically involve small geographic districts, each of

Objections to Proportional Representation

211

which has its own legislative representative. This system is thought to encourage a close identification between the voters and their representatives. Legislators can be trusted to understand and represent the interests of voters in their particular neighborhood, city, county, or district. Proportional representation, it is argued, weakens this relationship. It sometimes eliminates districts entirely, such as in urban areas where an entire city might be considered one district in which all candidates for city council vie for office. And even when districts are used in PR, they can cover much larger geographical regions and may be anywhere from five to ten times as large as current districts. Proportional representation systems are thus thought to make it more difficult for legislators to keep in touch with their constituencies, and voters could find it harder to identify who represents their particular area and interests. Angus Maude suggests that if PR were adopted in Great Britain, “electors would not have ‘their own MP’ [member of Parliament] to whom they can go with their problems and who will take matters up for them with the appropriate Minister or authority . . . . Ordinary citizens will have to fend for themselves in the jungles of national or local bureaucracies.”51 This criticism of proportional representation suffers from a variety of problems. First, champions of single-member districts exaggerate the close relationship that supposedly exists between local representatives and their constituents in single-member plurality systems. This relationship is often largely mythical. A study commissioned by the Senate Subcommittee on Intergovernmental Relations found that only 46 percent of a national sample could identify their congressional representatives and only 41 percent could name their representatives’ political affiliations.52 Similar studies in the singlemember districts of Great Britain show that almost half of British constituents do not know the name of their own national representative and that more than 75 percent cannot identify some legislative action taken by their representative or name something that official has done for their particular district.53 These figures hardly support the notion of a natural and intimate relationship between voters and representatives in single-member districts. This lack of connection between local representatives and their constituents is certainly not surprising, given that most people in single-member districts do not even vote for the candidate who ends up representing their district. Half of the American electorate does not vote at all. And of those who do vote, many vote for the losing candidate and so are unlikely to identify with the winning one. As a result, many constituents may not feel particularly close to “their” representative. Indeed, many might feel uncomfortable

212

Objections to Proportional Representation

approaching an official they voted against, someone who is unlikely to be sympathetic to their political concerns.54 Compare this to PR elections, which have significantly fewer abstentions and wasted votes. Under PR rules, the vast majority of voters end up electing one member in these multimember districts. Thus, almost everyone in these larger districts can find a legislator who will lend a sympathetic ear to their political concerns. In this sense, many more constituents can feel closer to one of their representatives in a multimember PR district. PR also often gives constituents the option of several different officials of the same party to whom they can relate. A Republican voter can choose whether to work with either the moderate or the more conservative Republican official elected from their district. In STV systems, these multiple party representatives have even been known to compete with each other to give the best constituency service in their district and thus increase their chances of reelection. In general, then, if one is concerned about encouraging close relationships between constituents and officials, the best option would actually be proportional representation. However, the most important flaw in this criticism of PR is that it is simply not true: PR systems usually do not prevent the election of geographical representatives. Most PR countries use regional districts for the explicit purpose of ensuring that representatives come from all geographical areas of the nation. There are a few very small nations—Israel and the Netherlands— that have eliminated districts entirely and elect all candidates from nationwide lists, but these are the exception. Thus if we were to use PR in state legislative elections, for example, every region of the state could have its own multimember district and its own officials to represent its particular geographical interests. When the single transferable vote has been used in cities, local districts have sometimes been eliminated and all the city councilors have been elected citywide. But even this does not necessarily eliminate geographical representation. In this form of PR, people cast votes for individual candidates, ranking them in their order of preference. If voters feel strongly that they want to be represented by someone from their area of town, they can rate that candidate number one on their ballot and help him or her to get elected. STV does not guarantee geographical representation, but produces it if that is what voters want.55 Finally, for those who want to absolutely guarantee local geographical representation, there is the additional-member system of proportional repre-

Objections to Proportional Representation

213

sentation. This version of PR was designed explicitly to combine small singlemember districts with a PR system. Recall that this system uses a double ballot. On the first part citizens vote for a local district representative in a plurality contest. The second part of the ballot is used for a party-list vote. Onehalf the legislature is made up of officials elected in the single-member district contests, with the other half added from the party lists in such a way as to ensure that each party’s total number of seats corresponds to its proportion of the party-list vote. Using this dual system, Americans could keep their local district representation and still enjoy the benefits of proportionality in their legislatures. In the end, then, it is simply untrue that there is a necessary tradeoff between PR and local representation. If they want, Americans can have both.

Conclusion: PR’s Advantages Far Outweigh Its Disadvantages Initially, there seems to be a considerable list of complaints against proportional representation. But after careful examination, most of those criticisms fall by the wayside. Most of them are either inaccurate, greatly exaggerated, or apply only to certain forms of PR. More important, there is little or no evidence to back up the critics’ dire predictions of the political disasters that will occur if the U.S. were to adopt proportional representation. The only evidence offered by critics of problems with PR abroad rely on highly selective and misleading examples like Italy and Israel—atypical countries that use extreme forms of PR. The record of decades of PR use in dozens of other Western democracies reveals a system that is well functioning and popular. The history of PR use in the United States also shows few if any substantial problems with this voting system. After examining the use of alternative voting systems in this country, Richard Engstrom of the University of New Orleans concluded that the “alleged [negative] consequences have yet to be documented, and what little empirical evidence exists points to their absence rather than their presence.”56 All of this is not to say that proportional representation is a perfect electoral system. Some forms of PR make it too easy for very small parties to get elected and thus complicate legislative politics. One form does not allow people to vote for individual candidates. Sometimes multiparty legislative coalitions fall apart. At times extremist parties may win some representation. Some of these problems could be minimized by choosing particular kinds of

214

Objections to Proportional Representation

PR or by adjusting such things as the threshold level. But the fact remains that no voting system is without some disadvantages. Critics of proportional representation often act as if finding a few problems with this system is enough to justify dismissing it. But to say that a voting system has some disadvantages is not to say that it is an undesirable system. One must consider whether the advantages of that system outweigh its disadvantages. In the case of proportional representation, that is clearly the case. As has been seen in this book, its benefits are numerous: it minimizes wasted votes, ensures fair representation for all political parties, eliminates gerrymandering, encourages more issue-oriented campaigns, gives voters more choices, increases the representation of minorities and women, fosters better voter turnout, encourages power sharing, enlivens political debate, and increases political equality among citizens. And as we have just seen in this chapter, the few political risks that would accompany the adoption of PR—such as a slightly higher chance of electing extremists—pale in comparison to the political rewards it would bring. In contrast, the political advantages of single-member plurality elections hardly justify the many political problems they cause. The advantages of SMP elections are much less impressive than those of PR. At best this system guarantees geographical representation, uses simple ballots, creates singleparty legislatures, makes it somewhat easier to throw a party out of power, and makes it more difficult for extremists to get elected. These benefits, however, are overshadowed by the heavy political costs that must be paid for using this electoral system: unrepresentative legislatures, numerous wasted votes, violations of majority rule, limited ballot choices, unfair obstacles to minor parties, rampant gerrymandering, uncompetitive elections, frivolous campaigns, discrimination against minority and female candidates, lower voter turnout, less power sharing, restricted political debate, and less political equality. The choice of an electoral system is an important one and it should not be made in a capricious or arbitrary manner. We should look at the arguments of both proponents and critics of each system. We should not rely on the myths associated with each system, but rather on a careful examination of the evidence. And we should not consider just one or two effects of these systems, but weigh all their various advantages and disadvantages. When we do this, it becomes clear that proportional representation elections are greatly superior to single-member plurality elections. It is not even close.

10

Versions of Electoral Reform for the United States

After it is shown that proportional representation is superior to single-member plurality voting, an important political question still remains: Which of the three general forms of PR—party list, mixed-member, or single transferable vote—would be best for the United States? Actually, the choice is more complicated than that because there are other alternatives to an SMP voting system besides these PR systems.1 Some electoral reformers, for instance, are advocating the adoption of instant runoff voting or semiproportional systems such as cumulative voting. In this chapter, we will first look at what is wrong with these other non-PR reforms and why they are not good substitutes for proportional representation. Once we get these inferior reforms out of the way, we will then turn to a discussion of the specific PR systems themselves and consider what kinds of advantages and disadvantages they have compared to one other.

The Unacceptable Alternative: Instant Runoff Voting One of the electoral reform ideas gaining increasing interest in the United States is a system called “instant run-off voting” or IRV. As of this writing, it looks likely that several states and cities will be adopting this reform in the near future. IRV is a winner-take-all voting system that is de-

216

Versions of Electoral Reform

ballot 5. Instant Runoff Voting

signed to solve some of the problems of plurality voting. In IRV, voters rank order the candidates running for office, putting a “1” next to their favorite candidate, a “2” next to their second choice, and so on. (See ballot 5.) To determine the winner, the number one preferences of all the voters are counted first. If a candidate receives more than 50 percent of the first choice votes, he or she is declared elected. If no candidate receives a majority, then the candidate with the fewest votes is eliminated. The ballots of supporters of this defeated candidate are then transferred to whichever of the remaining candidates they marked as their next preference. The votes are then recounted to see if any candidate now receives a majority of the vote. The process of eliminating the lowest candidate and transferring their votes continues until one candidate receives a majority of the continuing votes and wins the election. Assume that in the accompanying example Rosen received 45 percent of the vote, Kleinberg 44 percent, Chou 8 percent, and Royce 3 percent. Since no candidate received more than 50 percent, the lowest candidate, Royce, is dropped out and his votes transferred to their second choices. Say that they

Versions of Electoral Reform

217

all went to Kleinberg. She would now have 47 percent, but no one yet would have a majority. Chou is eliminated next. Say that 7 percent of his vote is transferred to Rosen and 1 percent to Kleinberg. Rosen would then win with 52 percent of the vote. Instant runoff voting is thus a majority voting system, not a plurality system. In essence, it is a series of runoff elections, with the candidate with the least chance to win being eliminated each time. Traditionally, runoff elections required a second election held sometime after the first one. All candidates except the top two would be eliminated thus ensuring that the final winner received a majority of the vote. Instant runoff voting eliminates the need to hold a separate second election, thus sparing all the expense and administrative aggravation involved in holding two elections. Instant runoff voting has several advantages over plurality voting. First, it ensures that the winner is supported by a majority of the voters. Second, it allows people to vote for minor party candidates without fearing that they are throwing their votes away. If their minor party candidate cannot win and is eliminated, then their vote is transferred to the major party candidate they prefer next. Because of this feature, this system also eliminates the “spoiler” problem—where a minor party candidate takes away enough votes from one major party candidate to ensure the election of the other major party candidate, who would not have won otherwise. With IRV, voters need not worry that a vote for a third-party candidate will inadvertently help to elect the candidate they like the least. So what is the problem with instant runoff voting? In fact nothing is wrong with this reform as long as it is being applied to single-office elections—such as those for mayor, governor, etc. Indeed, IRV would be very beneficial in those elections by eliminating spoilers and ensuring that the winner is supported by the majority of voters. Since proportional representation cannot be used to improve single-office elections, it makes sense to use IRV to do so. Unfortunately, however, some politicians and election reformers have proposed that IRV also be used for legislative elections, instead of PR. The idea would be to keep single-member districts, but use IRV instead of plurality voting to fill the seats in the legislatures.2 But using IRV for city councils, state legislatures, or even Congress is a bad idea—for several reasons. First, instant runoff voting is a very poor substitute for proportional representation. It offers very few political benefits compared to using PR for legislative elections. Although it has a few advantages over plurality voting, IRV is still a winner-take-all system, prone to all the other drawbacks of these sys-

218

Versions of Electoral Reform

tems. As one veteran electoral scholar, professor Wilma Rule, has observed, there are several important things that IRV does not do—but PR does. “[IRV] is a majority system which leaves out the political minority especially women and ethnic minorities, and third and other small parties.”3 Thus IRV does nothing to help solve our voting rights problems, or aid in the election of more women. Nor does it ensure fair and accurate representation for all parties, including minor parties. IRV slightly reduces but does not eliminate most of the enormous numbers of wasted votes in plurality elections. It also does not produce multiparty legislatures that truly reflect the variety of political views in the electorate. Finally, IRV eliminates none of the problems associated with redistricting, such as uncompetitive districts and partisan gerrymandering. In short, in legislative elections, IRV is not much better than plurality elections; and as a winner-take-all system, it remains grossly inferior to PR. Adopting PR would bring a number of badly needed changes to American elections and American politics—adopting IRV would not. Instant runoff voting is being promoted for legislative elections in part because it seems an easier reform battle to win. IRV is simpler to explain that PR and it is easier to get mainstream politicians to support this reform. Some major party officials see IRV as working to their advantage. In New Mexico, some Democratic lawmakers want IRV so that Green candidates can no longer act as spoilers and allow Republicans to win what have been traditionally Democratic districts. In Alaska, it is the Republicans that have been hurt by minor-party spoilers and are now interested in this reform. But the support of these politicians simply demonstrates how weak a reform IRV actually is. Major-party politicians can embrace IRV because it is relatively harmless and does not pose a threat to their dominance, or to the two-party system. It will not allow the routine election of minor-party candidates. Indeed, it will primarily help the major-party candidates because they will no longer have to worry about spoilers. Minor parties would be marginally better off under IRV than under plurality rules. They would be able to attract more votes for their candidates and this could help to increase their political legitimacy. But in the end, such parties would remain largely excluded from the governing process. They still would get neither seats nor any real power in our legislative institutions. Some IRV proponents acknowledge its inferiority to proportional representation for legislative elections, but maintain that it can serve as a “stepping stone” to PR. As Caleb Kleppner, Majority Rule Project Director for the Center for Voting and Democracy, has explained: “Although IRV is a majoritarian

Versions of Electoral Reform

219

system that would not significantly improve the representation of political minorities, it represents a substantial improvement over plurality and runoff elections, and it helps pave the way for future PR efforts.”4 It is thought that once people are used to ranking candidates in IRV, it will be easier to convert them to the single transferable vote or other forms of PR. Perhaps—but this may be wishful thinking. There is no empirical evidence that adopting the IRV aids in further electoral reforms.5 In addition, it is unclear how reformers would be able to easily shift political gears after the adoption of IRV and begin to promote a change to PR. They would have to turn from enthusiastic supporters of IRV to severe critics of that system—an inconsistency that would surely not go unnoticed by the public, the media, and policymakers. Even worse, the adoption of instant runoff voting for legislative elections could actually set back the PR movement in the U.S., or at least distract from it. Resources in the electoral reform movement are limited, so any effort for IRV must come at the expense of promoting PR. In addition, using instant runoff voting to eliminate spoilers and “fix” some of the problems of plurality elections could give politicians an excuse for not considering other reforms such as proportional representation. Indeed, in New Zealand, IRV became the preferred choice of many of those who were opposed to PR and wanted to maintain a winner-take-all voting system. For all of these reasons, it would probably be best if the advocacy of IRV were confined to the area of single-office elections. Working for “better” winner-take-all elections is only appropriate when no other alternatives are possible. It makes little sense in legislative elections, where much better alternatives—such as proportional representation—are available. Advocating IRV’s use in legislative elections thus might prove to be a strategic error for the electoral reform movement. The struggle to adopt PR—and all the major political benefits that it will bring—is difficult enough without making it harder by promoting IRV for these elections. Those interested in genuine electoral reform for legislative elections would be better off keeping their “eyes on the prize,” and continuing to press for the adoption of proportional representation elections in the United States.

The Semi-Acceptable Alternative: Semiproportional Systems Some American electoral reformers have promoted cumulative and limited voting—both semiproportional voting systems. In fact, these systems are

220

Versions of Electoral Reform

ballot 6. Cumulative Voting

now more numerous than pure PR systems in the United States, thanks largely to their implementation as settlements in voting rights suits. More than 100 local jurisdictions now use these voting systems. There is also a campaign to bring back cumulative voting to Illinois, which used this system to elect their state legislature from 1870 to 1980.6 Interestingly, cumulative voting is also used for the election of boards of directors in hundreds of corporations, where it has successfully allowed minority stockholders to elect some representatives to these corporate governing bodies. In cumulative voting, candidates run in multimember districts. Voters have as many votes as there are seats—for instance, three votes for three seats. Voters cast their votes for individual candidates and the winners are the ones with the most votes. (See ballot 6.) The distinctive characteristic of this system is that voters may “cumulate” or combine their votes on one or more candidates instead of having to cast one vote for each candidate. In other

Versions of Electoral Reform

221

ballot 7. Limited Voting

words, voters may distribute their votes among the candidates in any way they prefer—such as three votes for one candidate, one each for three candidates, two for one and one for another, and so on.7 The limited vote also utilizes multimember districts and multiple votes. The difference is that voters have fewer votes than the number of seats to be elected. And they may cast only one vote per candidate. So in a three-seat district, voters might have only two votes, which they could cast for their two favorite candidates. (See ballot 7) In the end, the candidates with the most votes win office.8 The main aim of these two systems is to produce more proportional results than winner-take-all systems, and to allow for some representation of minority political groups. In a five-seat cumulative voting district, for instance, a racial or political group with as little as 17 percent of the vote could elect one of their representatives—as long as they all cast their five votes for that one candidate. In practice, these systems have often worked to increase minority representation. In some local elections in the south, for example, cumulative voting has allowed racial minorities to win election for the first time.9 However, the problem with these systems is that while they have the potential to create proportional election results, they may not always do so. They are inconsistently proportional. Sometimes they produce proportional results, but often the results are only somewhat proportional, or even glaringly unproportional. At times the largest party might win all the seats in a

222

Versions of Electoral Reform

district and minor parties might be denied representation entirely. It is even possible for the second-place party to win the most district seats. These systems also have the potential to create manufactured majorities, where a party that gets less than a majority of the votes get a majority of the seats in the legislature. In other words, while these systems can sometimes be as fair as truly proportional systems, they can also be as unfair as plurality systems. This tendency to produce inconsistent results is why these systems are called “semiproportional.” The main flaw in these systems is that how well they work depends entirely on how many candidates the parties nominate and whether the voters cast their votes in the most strategically effective way. Consider the examples shown in tables 10.1, 10.2 and 10.3. These represent the same election taking place in a three-seat district using cumulative voting. The district is composed of 30,000 Democrats and 16,000 Republicans who have three votes apiece. The fair and proportional result would be the election of two Democrats and one Republican, which happens in table 10.1. It is important to note, however, that in this example the Republicans nominate only one candidate, Gilpin. What if the Republicans decide to nominate two candidates, as in table 10.2, in the hopes of winning more seats. Then the Republican vote could be split and they would win no seats. The Republicans could salvage a seat in this situation if most of the Republican voters could somehow

table 10.1 Proportional Results in a Cumulative Vote Election Democrats Candidate

Votes

Lakeman

35,000*

Wright

30,000*

Hoag

25,000

Republicans Candidate Votes Gilpin

48,000*

90,000 Result

2 seats

*Winning candidates. Note: 30,000 Democrats and 16,000 Republicans with three votes each.

1 seat

223

Versions of Electoral Reform

be convinced to cast all three of their votes for only one of the candidates. But that would obviously be difficult to do. Table 10.3 illustrates another possible malfunction of this system. In this case, the Democratic voters make a mistake and give too many of their votes to their favorite candidate, Lakeman, and not enough to the other Democrats. The result: a situation of minority rule where the Republicans win two out of three of the seats. So as these examples demonstrate, the fairness of the election results in cumulative voting (and limited voting as well) is not automatic at all, but is dependent on nominating and voting strategies. These are not simply hypothetical problems: in a 1991 cumulative vote election in Peoria, Illinois, two liberal black candidates ended up splitting much of the African-American vote and neither of them was elected. In short, while semiproportional systems are an improvement over singlemember plurality elections, they are unreliable and inconsistent compared to truly proportional systems such as party list PR, the mixed-member system, and the single transferable vote. In true PR systems, it makes no difference how many candidates are nominated or how the voters cast their votes—the results will be proportional. This obvious superiority is what led most European countries to embrace PR instead of semiproportional systems as the best way to remedy the problems of winner-take-all elections.10 Likewise, it makes little sense to promote these semiproportional systems in

table 10.2 Disproportional Results in a Cumulative Vote Election Democrats Candidate

Votes

Republicans Candidate Votes

Lakeman

35,000*

Gilpin

24,000

Wright

30,000*

Humphreys

24,000

Hoag

25,000*

48,000

90,000 Result

3 seats

*Winning candidates Note: 30,000 Democrats and 16,000 Republicans with three votes each.

No seats

224

Versions of Electoral Reform

table 10.3 Minority Rule in a Cumulative Vote Election Democrats Candidate

Votes

Republicans Candidate Votes

Lakeman

60,000*

Gilpin

24,000*

Wright

15,000

Humphreys

24,000*

Hoag

15,000

48,000

90,000 Result

1 seat

2 seats

*Winning candidates Note: 30,000 Democrats and 16,000 Republicans with three votes

the United States. They are better than SMP systems, but why settle for an unreliable, semiproportional system when reliable and fully proportional systems are available? Many advocates of semiproportional systems acknowledge that they have some serious drawbacks when compared to full PR systems, but insist that systems like the cumulative vote are more politically feasible—that they have fewer political obstacles and a much greater chance of being adopted in the United State than PR. They emphasize that the cumulative vote is not that different from at-large voting—a system with which many Americans are familiar and comfortable. In fact, cumulative voting has sometimes been called “modified at-large voting” to stress the similarities between the two. It is thought that it is much easier to get the support of politicians, election administrators, and citizens if the electoral change is not too drastic. But it is not clear at all that semiproportional systems are easier to “sell” than proportional systems. Cumulative and limited voting have often run into serious political opposition where they have been proposed. Moreover, if citizens are given the proper information about PR, many will find it an easily understandable and desirable voting system. This was certainly the case in New Zealand. In 1993, after several years of study and debate, New Zealand voters approved a change from their tradition single-member plu-

Versions of Electoral Reform

225

rality system to mixed-member proportional representation. Similarly, when American voters have been exposed to PR, many have responded quite positively. In the 1990s, both Cincinnati and San Francisco held referendums on PR; and although the pro-PR forces had little money for their campaigns, they were still able to attract more than 42 percent of the voters in both cities. This clearly indicates that many Americans can easily appreciate the appeal of this system. Thus, since there seems to be little real difference in the political feasibility of these two kinds of voting systems, it makes little sense to promote the inferior one.

Party List Proportional Representation Now that we can dismiss these inferior electoral reforms, we can turn to a consideration of the better options—the various forms of proportional representation. Throughout the book, we have been primarily concerned with describing the major advantages that all of these PR systems share, such as fairer representation for parties, minorities, women, and so on. But there are some differences worth noting as well. Each PR system has its unique advantages and disadvantages when compared to the others, and understanding these differences is an important element in choosing the appropriate PR system for a particular political circumstance. Let us begin this analysis by considering party list proportional representation. In party list proportional representation, parties run competing slates of candidates in large multimember districts. (To refresh your memory about how these PR systems work, you might want to refer back to the descriptions of each system in the introduction.) Party list PR is the most popular form of this voting system; it is used in more than 80 percent of countries using PR elections. Nevertheless, many electoral reformers in the United States sometimes do not give this form of PR serious consideration. Some fear that this system puts too much emphasis on parties and that Americans voters might not embrace it for that reason. But this is based on a mistaken understanding of how party list PR works. For example, some people believe that party list ballots cannot accommodate independent, non-party candidates—but this is not true. Similarly, others believe that party list PR does not allow people to vote for individual candidates at all, but only for a predetermined party list. But as was shown in the introduction, this is not necessarily true. Open party list systems allow citizens to vote for their favorite candidate, thus increasing his or

226

Versions of Electoral Reform

her chances of moving up on the list and being elected—and this form of party list voting is by far the most prevalent one among Western democracies. One major advantage of this form of PR is that it is often the most proportional. It is very accurate in giving political parties and groups the share of seats in the legislature that reflects their voting strength. Party list systems typically use districts with large numbers of seats; and in general, the more seats in a district, the easier it is to allocate them proportionately among the parties. For this reason, the single transferable vote (STV), which tends to use relatively small districts, often does not produce as proportional results as party list PR. So in areas with many parties vying for office, party list PR would best ensure that they all received their fair share of seats. However, this strength also carries with it a weakness. Because it tends to use large districts, this system may tend to weaken the traditional geographical link between elected officials and their local constituents—especially compared to STV and the mixed-member system. This tendency is taken to the extreme in a few very small countries, such as Israel, where the entire country is one district, with no local or regional districts at all. But most party list countries create some geographical representation by dividing the nation into regional districts to ensure that all areas of the country have some representation. Clearly this regional-district approach would be best for the United States as well. Clearly as well, an open-list would be better than a closed-list system. American voters prefer to vote for individual candidates, not just parties. Also, an open-list system allows factions within parties (such as African Americans within the Democratic party or fundamentalist Christians within the Republican party) to promote the election of their favorite candidates within the parties’ slates. Thus party list PR ensures not only fair representation for parties, but for other kinds of political groups as well. Party list proportional representation could be used at every level of election in the United States. In could be easily used for local city and country elections. Most likely, the entire city or county would be one large district where all the parties vied for seats—similar to the at-large plurality arrangement still found in some local areas. However, this system could not be used in areas of the country that use nonpartisan elections that require candidates to appear on the ballot with no party designation. Party list voting obviously requires party designations to work. In elections for state legislatures or the U.S. House of Representatives, it would be best to divide the state up into several regional districts to ensure geographical representation. A hundred-seat state senate, for instance, could

Versions of Electoral Reform

227

be elected from ten 10-seat districts. A fifteen-member U.S. House delegation could be elected from three 5-seat districts. Note, however, that there is no reason why the regional districts would have to be the same size. In order to avoid splitting up cities or counties, it might make sense to create some larger state senate districts with 10–15 seats, along with some smaller districts with 5–10 seats. Party list proportional representation is a very flexible and workable voting system. It has a good track record in other Western democracies. Also, it is very easy to understand compared to some of the other PR systems. For all of these reasons, it deserves serious consideration by those promoting proportional representation.

Single Transferable Vote The single transferable vote—sometimes also called “choice voting” or “preference voting”—has a unique advantage over other forms of proportional representation: It already has a history of successful use in the United States. Two dozen cities used STV during the early and mid-twentieth century, and Cambridge, Massachusetts continues to use it today. (See appendix C.) It is also used to elect the Australian Senate and the national legislature in the Republic of Ireland. In the single transferable vote, candidates run in multimember districts and voters rank order them—giving a “1” to their first choice, a “2” to their second choice, and so on. All the first choice votes are counted and any candidate that reaches the threshold needed to win office is elected. If there are any seats remaining, a transfer process takes place to determine who will fill those positions. For example, the lowest candidate is dropped out, and his or her ballots are transferred to the voters’ second choices to help those candidates win—very much like the transfer process that takes place in instant runoff voting. The purpose of this transfer process is to minimize the number of wasted votes and maximize the number of voters who actually elect a candidate of their choice. And not incidentally, STV also produces proportional representation for parties. Although this voting system is easy to use, the complexity of the vote transfer process sometimes makes it difficult to explain to people—especially compared to the other forms of PR. The single transferable vote can be used on all levels of government. It has already been used successfully on the local level in the United States to

228

Versions of Electoral Reform

elect city councils, school boards, county boards, etc. It many cases, the entire city was considered one district, but there is no reason why cities cannot be divided up into several multimember districts. States could also be divided into many regional multimember districts to use STV elections for state legislatures or the U.S. House. Again, the districts could be of different sizes to take into account local geographical and political boundaries. One of the appeals of this system is that it is purely candidate oriented. There is no party list voting, as in the other proportional representation systems. Voters cast their votes for individual candidates only. Voters also get a great deal of flexibility in how they cast their vote. They need not stick to one party. A Democrat, for instance, may rank a Democratic candidate first, but then give her second choice to a female Republican candidate she particularly likes. STV is also the only form of PR that can be used in nonpartisan elections. Candidates can be simply listed on the ballot with no party designation, which makes this form of PR the system of choice for local areas that want to retain their nonpartisan approach to elections. The single transferable vote gives voters a choice between different members in each party, which has its advantages and disadvantages. On the one hand, this allows voters to choose the candidate that represents the faction of the party with which they identify. So a moderate Republican could choose to support a moderate Republican candidate. On the other hand, this situation can exacerbate internal tensions in parties and fuel destructive rivalries between candidates of the same party. We already see some of that behavior now in party primaries, where the candidates in the same party spend more time criticizing each other than they do their opponents. In STV, this same kind of internal party bickering can also extend into the general election. So while the choice between different party candidates is good for voters, it may not be good for party cohesion. STV also has the advantage of allowing voters to choose how they will be represented. Single-member district voting assumes that most people want to be represented geographically. But STV allows voters to be represented in any way they want. For example, a feminist may join with others who share her views and help to elect a female candidate. Fundamentalist Christians can similarly band together to support a candidate with their moral and religious values. And if voters believe geographical representation is important, they can vote for someone from their local area. Thus in STV, voters can organize themselves into “voluntary constituencies” and be represented in the ways they think are most important. (A similar effect can be achieved in a

Versions of Electoral Reform

229

completely open party list system, where voters are also given the opportunity to choose the individual candidates of their choice.) The single transferable vote also has the tendency to encourage more civil campaigning. Candidates are discouraged from throwing mud at each other by the fact that they hope to benefit from the second choices of voters. If they engage in vicious personal attacks on their opponents, they risk alienating these possible supporters. Finally, STV can offer the opportunity to eliminate primaries. In a city council election, for instance, all the candidates could run in the general election. The ones with the least voter support would be eliminated first in the counting process, just as they would have been if a separate primary were held. Thus STV avoids the considerable expense, administrative burden, and low turnout that characterize primary elections. There are then a number of unique advantages to the single transferable vote. The fact that it is a purely candidate-centered system and that it has a history of successful use in the United States makes it a particularly attractive in the American context.

Mixed-Member PR Mixed-member proportional representation (MMP) is probably the “hottest” form of proportional representation. Versions of MMP have been recently adopted in New Zealand, Scotland, and Wales. It claims to be the “best of both worlds”—incorporating the best elements of single-member district voting and the best of PR. In this system, half the members of the legislature are elected in single-member districts and the other half are elected from a party list ballot and added on to the district members to ensure that all parties have their appropriate share of seats in the legislature. (Again, see the introduction for a more detail description of this system.) Unlike party list proportional representation, MMP does ensure local geographical representation. The use of relatively small single-member districts means that all local areas will have their own representative to further their interests in the legislature. And yet this system also ensures proportional representation, so that all parties and political groups receive their fair share of legislative representation as well. Another advantage of MMP is that its double vote arrangement allows voters to better express their political views. For example, voters may split

230

Versions of Electoral Reform

their votes between two different parties as a way of expressing support for a particular coalition. Supporters of the Greens could cast their party list vote for their own party, and then cast a vote in the single-member district contest for the major party candidate they liked the best—say a Democrat. In this way, they would effectively express their support for a coalition of Greens and Democrats. Some critics of the mixed-member form of PR worry about the two different classes of legislators that it creates: some who are beholden to specific geographical constituencies, and others who mainly represent the parties. It is thought that the legislators from the party list portion of the ballot will not have close ties to constituents and will not have to spend as much time on constituent service. However, this does not seem like a particularly serious problem. After all, many Americans cities have a dual electoral system, with some city councilors elected from districts and some elected at-large, and there are rarely any complaints from voters about this arrangement. There is sometimes a concern that keeping single-member districts could bring some problems typical of SMP systems—such as gerrymandering. And while gerrymandering could certainly take place in this system, it could not give a party more than its fair share of seats—as it can do in SMP systems. The PR portion of this system guarantees that all parties will receive their appropriate number of seats in the legislature. For this reason, gerrymandering is less of a concern in this system than it is in SMP elections. It is also possible that a single-member district election among a variety of candidates could result in the election of a local representative who did not get a majority of the vote. But that would be true only if plurality voting rules were used. If a runoff system, such as the instant runoff vote, were used this problem would be effectively avoided. Use of IRV for district contests should thus be seriously considered by those thinking of using MMP elections. There are some restrictions on the use of mixed-member systems. They would be obviously inappropriate for jurisdictions that use nonpartisan elections—since it requires a party list vote. Also, MMP does not work well where there are relatively few seats at stake. It would be a poor choice, for instance, for a small city council of seven members. If four were elected in districts, that would only leave three seats to be used to ensure proportional representation of the parties in the council—a number that is probably not sufficient for ensuring accurate representation. Similarly, if a state delegation to the U.S. House of Representatives were a small one, say five seats, this would also be a poor choice for MMP; other forms of PR, such as the single

Versions of Electoral Reform

231

transferable vote, would be more suitable. MMP would work quite well in states with large House delegations, cities with large councils, and of course state legislative elections. It is no accident that mixed-member PR is growing in popularity. It can be seen as a reasonable compromise between single-member systems and proportional representation, and thus has a unique and powerful political appeal. Advocates of SMP get to keep their district representatives and proponents of PR get to have truly representative legislatures. Since it keeps part of the old winner-take-all system, this form of PR is not seen as such a radical change, and thus it may appear less threatening that other forms of PR. For this reason, reformers should take a very careful look at this version of proportional representation.

Another Consideration: District Size No matter what form proportional representation takes in the United States, one of the most important electoral design decisions concerns the size of the election districts—that is, how many seats are elected in each district. This is usually called district magnitude, and it is a choice that has several important political ramifications. First, if district magnitude is too low— say three or four seats—gerrymandering becomes a possibility. It is sometimes tempting to create as many small multimember districts as possible to give each local area their own representatives. But the smaller the districts, the greater the possibility that lines can be drawn to deny some group their fair share of representation. In a three-seat district, for instance, any party with less than 25 percent of the vote will be unable to elect anyone. This kind of gerrymandering has sometimes been a problem in Ireland, where many districts have only three or four seats. Political scientists have usually recommended that PR districts have at least five seats to minimize the possibility of gerrymandering.11 District magnitude also plays a critical role in determining the size of the party system and the number of parties that win seats in the legislature. The more seats contested, the lower the natural threshold for gaining office and the larger the number of minor parties that are likely to win seats. 12 This relationship is illustrated in table 10.4. The table shows how the number of parties elected tends to increase as the district magnitude gets larger. Note that in a five-seat district, only two parties elect candidates, while in the sixty-

232

Versions of Electoral Reform

seat district, all four parties win seats, including the two small parties. In Israel the district magnitude is very high—120—with all the seats in the Knesset elected in one countrywide district. This leads to the propagation of many very small parties—sometimes more than a dozen in the Knesset. As we saw in chapter 9, some believe the multiplication of too many small parties can lead to political problems. So if we want to keep the number of parties at a more reasonable level, relatively small district magnitudes—such as five- to nine-seat districts in an STV system—should be considered. On the other hand, large multiseat districts do offer some advantages. Primarily, they produce a much more proportional distribution of seats among parties. As a rule, the more seats in a district, the easier it is to match the proportion of seats with the proportion of the vote. This is also illustrated in table 10.4. The numbers in parentheses are the percentages of seats won by each party, which can be compared with the percentage of the vote won by each party. In a five-seat district, the percentage of seats only roughly approximates the percentage of votes. Although party A receives 48 percent of the vote, it receives 60 percent of the seats. In the sixty-seat district, the correspondence of the percentage of seats to the percentage of votes becomes quite close—party A gets 48.3 percent of the seats for its 48 percent of the vote. Because of the greater accuracy some PR countries use very large districts to ensure precise proportionality in seat allocation. Germany, for in-

table 10.4 Relationship of Number of Legislative Parties to District Magnitude in PR

Party

Percentage of Votes

Seats won in a 5-Seat District

Seats won in a 15-seat District

Seats won in a 60-seat District

A

48%

3 (60%)

7 (46.7%)

29 (48.3%)

B

44%

2 (40%)

7 (46.7%)

26 (43.3%)

C

5%

0 (0%)

1 (6.6%)

3 (5.0%)

D

3%

0 (0%)

0 (0%)

2 (3.5%)

Number of parties winning seats

2

3

4

Note: Results reflect party list system using the largest remainder formula to distribute seats.

Versions of Electoral Reform

233

stance, determines the total allocation of party seats on the national level. If we were to use the mixed-member system of PR, we might use similarly large districts—with the entire city, state, or nation serving as the party-list district. However, in such situations we would want to set an arbitrary legal threshold—a percentage of the vote a party must receive in order to win any seats in the legislature. Such a threshold would serve to discourage the emergence of too many small parties. What exactly should that threshold be? The one percent threshold used in Israel is clearly too low. A more reasonable level is probably 3 to 5 percent, as is the case in many PR countries today. This would allow for some additional minor parties, but avoid the excessive fragmentation of the Israeli system.

Which PR System Is Best for the United States? Clearly proponents of proportional representation should consider carefully which PR system they propose for the elections they wish to reform. Each form of PR has attractive features. Party list PR is the most proportional form and has long record of successful use in many Western democracies. The mixed-member system lays claim to blending the best of both worlds— single-member districts and PR. And the single transferable vote is our homegrown version of PR and it is a purely candidate based system. In some cases, local considerations could tip the scales in favor of one system or another. If an area uses nonpartisan elections, then the single transferable vote would be the obvious choice. Also, if a city council or school board is very small, a system like mixed-member PR would be inappropriate. However, in many situations, most forms of PR could be made to work. Indeed, in the end, a good case can be made for all of these proportional representation systems and it may make little sense to obsess too much about their differences. In fact, the primary danger facing the PR movement is not that it might opt for the “wrong” version, but that it could waste valuable time and energy squabbling over which system is best. This kind of quarreling has plagued many other electoral reform efforts in the United States. For example, disagreements among competing camps of reformers often have hobbled attempts to reform the Electoral College system. While they have all agreed on the inequities and irrationalities of the Electoral College, they have often strongly disagreed over which particular system should replace it.13 Some have advocated replacing the college with a popular vote for the

234

Versions of Electoral Reform

presidency, and others have offered less radical plans to simply modify the college to remove some of its more disagreeable qualities. These splits have often prevented reformers from presenting a united front and are one reason why past reform efforts have failed. The PR movement must make a concerted effort to steer clear of the same fate. At some point, naturally, people and organizations interested in promoting proportional representation will have to coalesce around a common plan for the jurisdiction under consideration. Such a decision will necessarily involve some analysis and discussion. But if those deliberations are to be fruitful, the participants must be able to stand back and keep the bigger political picture in mind: that party list PR, the single transferable vote, and the mixed-member system are all vastly superior to single-member plurality elections and to weaker reforms such as instant runoff voting and cumulative voting. All of these PR systems would minimize wasted votes, give minor parties fair representation in our legislatures, broaden the range of political debate, increase the numbers of women and racial minority officeholders, increase voter participation, encourage power sharing, and make our government more responsive and democratic. Equally important, all of these systems have been used with success in other countries. In this sense, we could not go wrong choosing any of these systems of proportional representation for use in the United States.

11

The Political Prospects for PR in the United States

In the end, the case for proportional representation is strong. The political benefits of PR are many and significant, and the political risks are minimal. And as demonstrated in the previous chapter, proportional representation could take several practical forms in the United States—all with proven track records in other countries. But the superiority and viability of proportional representation do not ensure that it will be adopted in the United States. Electoral reforms are not enacted simply because they are good ideas. Choices about electoral systems are not made in a debating society in which the best arguments win but in a political system in which the most powerful political interests prevail. Thus any attempt to adopt PR must be understood in the context of the American political system, in which selfinterested groups compete for power and advantage. These political realities mean that the effort to adopt PR will involve an intense and perhaps prolonged political struggle. Some powerful individuals and groups benefit from our plurality election rules, and they will fight hard to maintain those rules. This chapter will discuss the prospects for adopting proportional representation in the United States. It will examine the main obstacles that stand in the way of proportional representation—which include public ignorance, constitutional barriers, and the opposition of political elites. We will also look at several ways that these obstacles might be overcome.

236

The Political Prospects for PR

The Need for Public Education Undoubtedly one of the most serious obstacles faced by the proportional representation movement is public ignorance about electoral systems, particularly alternative electoral systems. Many Americans still have little awareness of proportional representation. They do not understand how PR works or what advantages it has over our current election system. This will have to change if the proportional representation movement is to begin to develop widespread public support. For this reason, public education has to be the number one priority of the PR movement today. For such an education effort to be effective, it must not only make more Americans aware of PR and how it works, it must convey the following points as well. •

•

•

•

Electoral systems matter. Most Americans do not think about electoral systems. But they should. The choice of an electoral system is not a trivial one; it has wide-ranging political implications. These systems not only determine who gets elected but also how power is shared and who participates in policymaking. Our current single-member-plurality voting system causes numerous political problems. Many Americans have an “if it ain’t broke, don’t fix it” attitude. They need to see that our current voting system is broken—that is, it has created a number of serious political problems, including lack of choice at the polls, wasted votes, unrepresentative legislatures, gerrymandering, and so on. And unless it becomes clear to the public that many our election problems are systemic—caused by a flawed electoral system—they cannot be expected to support the need for systemic solutions such as the adoption of proportional representation. There are alternatives to SMP. For most Americans, SMP is the electoral system they grew up with and the only one they know. So when they become discouraged about our elections, they tend to become disenchanted with elections in general. They need to be shown that the problem is not elections in general, but our SMP electoral system in particular. They need to become aware that there are other electoral systems that work much better and that transform voting into a more meaningful and powerful political act. PR has advantages for all Americans. Sometimes PR is seen as being primarily of advantage to racial minorities or supporters of

The Political Prospects for PR

•

237

third parties. But in fact a more just and democratic electoral system works to the advantage of all Americans. It ensures that major and minor parties, liberals and conservatives, and whites and minorities will all get their fair share of seats and political power. Most voters want more choices at the polls, more competitive elections, fairer representation, and more responsive policymakers—and they need to appreciate how PR can help to achieve those shared political goals. PR has worked well where it has been used. The superiority of proportional representation voting is not merely a matter of speculation. Its advantages have been proven in practice in many other Western democracies. Promoters of PR must make it clear that this is a well-tested reform—that it has had a long and successful track record in other countries and that few of them have experienced significant problems with this voting system.

Fortunately, recent years have seen some significant strides being made in educating Americans about electoral systems and proportional representation. In the early 1990s, when the PR movement was in its infancy, it is safe to say that almost all Americans were ignorant of the political effects of SMP elections and largely unaware of alternatives such as proportional representation. Today, that situation has changed and an increasing number of Americans know about PR and its political advantages. Much of the increase in public awareness about PR can be largely attributed to the efforts of Robert Richie and his colleagues at the Center for Voting and Democracy (CVD) in Washington D.C. The Center is the pre-eminent organization educating policymakers and the American public about PR and other alternative voting systems. Like the Electoral Reform Society in Great Britain and the Proportional Representation Society of Australia, CVD has played a key role in increasing public awareness of this issue. It has developed a substantial literature on electoral reform and offers a web site that provides comprehensive information about these alternative systems. It has also been successful in its media outreach program, placing numerous opinion pieces in newspapers and magazines and encouraging reporters and political columnists to write about PR and electoral system reform. Articles about electoral reform have appeared in a wide range of major media outlets, including Time, The New Republic, The New York Times, The Atlantic, The Boston Globe, The New Yorker, The Christian Science Monitor, The Chicago Tribune, The Washington Post, and USA Today.

238

The Political Prospects for PR

An equally important, though less obvious, development in the area of public education has been the emergence of local groups promoting proportional representation. Several state-based groups—such as Californians for Proportional Representation, Fairvote Wisconsin, and Fairvote Minnesota—have begun to organize around the issue of electoral reform. These groups have been working to educate local politicians, activists, and citizens about proportional representation. Their efforts have included letters to the editor and op-ed pieces submitted to local newspapers, discussions about electoral reform with local and state officials, and talks given to local groups, such as the League of Women Voters, Chamber of Commerce, labor unions, Sierra Club, and so on. Some PR groups have also encouraged the creation of local and state electoral study commissions. In Illinois, for instance, Illinois Citizens for Proportional Representation and the Midwest Democracy Center were instrumental in setting up a bipartisan commission to study the possibility of reinstating cumulative voting for state legislative elections—a semiproportional system used in Illinois from 1870 to 1980. The commission’s final report supporting the return of cumulative voting proved to be a very useful tool for educating the press and the public about this alternative electoral system.1 The efforts of these local and state organizations, along with those of the Center for Voting and Democracy, have gone a long way toward familiarizing many Americans with the problems associated with our current electoral system and the benefits of proportional representation. Clearly, however, this education effort must continue if the movement is to grow.

Legal and Constitutional Obstacles Several legal and constitutional barriers can interfere with the adoption of proportional representation. On the national level, for instance, it would be difficult to adopt PR for the U.S. Senate because the Constitution stipulates that senators be elected one at a time in each state. Obviously PR cannot be used for such single-office elections. One can imagine, however, several different ways that Senate elections could be modified to accommodate PR. For example, a mixed-member PR system could be used. In that approach, each of the states would elect one senator, and these fifty officials would make up half of the Senate. Fifty more senators would be elected by party list PR, from regional lists, to make up the other half of the Senate. But

The Political Prospects for PR

239

such drastic changes would require a constitutional amendment. Unfortunately, amending the Constitution is a very arduous process, requiring the agreement of two-thirds of Congress and three-quarters of the states. This would be an extremely difficult political hurdle to surmount. Because of this, most PR reformers have suggested leaving the Senate as it is, and adopting PR for the House of Representatives. Using proportional representation for the U.S. House would require no constitutional amendments. The Constitution does not require singlemember district elections for this body. Article I, Section 4, of the Constitution clearly gives to the states and to Congress the power to decide how House seats will be elected: “The Times, Places and Manner of holding Elections for Senators and Representatives, shall be prescribed in each State by the legislature thereof; but the Congress may at any time by Law make or alter such Regulations, except as to the Places of Chusing Senators.” The only legal barrier to the use of PR for House elections is congressional legislation passed in 1967 that requires the use of single-member districts. Interestingly, the original intent of these bills was not to prevent the use of PR, but to stop the use of plurality at-large elections. In the 1960s, some southern states were trying to switch to at-large elections to discourage the election of African Americans to the House, and Congress wished to prevent this political maneuver.2 In recent years, several pieces of legislation have been introduced in Congress to repeal the 1967 law and allow states to use alternative electoral systems like proportional representation. For example, in the 106th Congress, Representative Cynthia McKinney (D, GA) introduced the Voters Choice Act (HR 1189) and Representative Mel Watt (D, NC) introduced the States’ Choice of Voting Systems Act (HR 1173). While such bills have garnered more sponsors each time they have been introduced, they currently stand little chance of passage. It is clear, however, that passing such legislation is considerably easier than trying to amend the Constitution. The legal and constitutional situations on the state level are more varied. Some state constitutions, such as those in Connecticut, Kansas, and Michigan, require single-member plurality elections for their state legislatures. Fortunately, the constitutions of many others, including Mississippi, North Carolina, and Oregon, are silent as to the method of electing these bodies, and so proportional systems could be adopted without constitutional amendment. However, even where constitutional amendments are required, the process is often much easier in the states than it is on the federal level. Seventeen states

240

The Political Prospects for PR

require only a simple majority of the legislature to change the constitution. And all but one state require only a majority of the state’s voters to approve such measures. More important, sixteen states allow citizens to bypass the legislature through the initiative process to put a constitutional amendment directly on the ballot. In most states, petitions for initiatives require the signatures of only 5 to 10 percent of the electorate and can be eventually passed by a majority vote.3 The relative ease in changing state constitutions is evidenced by the fact that literally thousands of changes to state constitutions have been made in the U.S. since the 1970s.4 Thus changing state constitutions to allow for PR elections is much more feasible than changing the federal constitution. Legal obstacles are even fewer on the local level. Adopting proportional representation on the city level, for instance, usually only requires changing a city’s charter, which can often be accomplished by a referendum or initiative. Undoubtedly, this is one reason why when PR has been adopted in the United States, it has almost always been on the city level. Nevertheless, legal and constitutional obstacles, especially those on the state and federal levels, cannot be ignored by proponents of proportional representation and must be addressed in any reform effort.

Political Opposition Attempts to bring proportional representation to the United States have also been encountering some political opposition. Some Americans do not like the way PR would change the political landscape. And not every American believes that all political groups deserve their fair share of representation in our political system. For example, the tendency of PR to increase the representation of racial and ethnic minorities has aroused opposition in some political quarters. Some who have opposed the use of majority-minority districts to give minorities fairer representation, such as Abigail Thernstrom of the Manhattan Institute, have already come out publicly against adopting proportional representation. Also, it is significant that racial issues played an important role in the repeal of PR in Cincinnati in the 1950s. To mobilize whites in opposition to PR, repeal advocates intentionally exploited growing tensions over the civil rights movement and the emerging political power of blacks on the city council. Today, explicitly racist opposition to PR would be less likely to be voiced publicly. Nevertheless, race could easily become

The Political Prospects for PR

241

an issue in any political battle over PR. Race and racial antagonisms remain important and volatile dimensions of American politics, and some politicians have become experts at exploiting racial tensions in subtle but effective ways. Some opponents to PR might find ways to introduce the race issue into the debate, particularly in areas with large populations of racial minorities. Ideological antagonisms have also fueled some opposition to proportional representation. PR would expand the political spectrum in the United States—allowing parties, politicians, and ideologies now on the fringes of American politics to win elections and occupy some seats in our city, state, and federal legislatures. When PR was previously used in American cities, this development created some strong opposition. The case of the Communists elected in the STV elections in New York City during the 1930s and 1940s is the best-known example. Today some Americans are alarmed by what they see as the radical ideologies of the Libertarians or the Greens; and clearly some would oppose PR simply to discourage such parties and politicians. Fortunately, those who oppose PR on racial or ideological grounds are likely to be in the minority. The majority of Americans believe in racial fairness and also want to see other parties challenging the Republicans and Democrats—and thus are likely to line up on the side of proportional representation. Commentators on the political left have speculated that some business interests could also align themselves against proportional representation. As Anthony Thigpenn has argued, “The most formidable opponent is the business community that currently controls [elected officials] and the rest of our politics.”5 Corporations, through their large campaign contributions and well-financed lobbying efforts, do have considerable influence in both major parties; and they may not want to see new parties gain power in legislatures. Minor parties, such as the Greens, which favor greater controls on corporate activity, would naturally be considered a threat to those interests. Given the political power of business interests, any opposition from this quarter would have to be taken seriously. It is not at all clear, however, that business would automatically oppose PR. Some might even support it, as has been the case in Great Britain. But reform would not necessarily be blocked even if there were substantial business opposition to PR. It did not block it in New Zealand. Business leaders and corporations were a major source of money spent in the effort to oppose the PR referendum in New Zealand in 1993, and yet strong grassroots support for PR led it to victory at the polls.

242

The Political Prospects for PR

Opposition of Political Elites By far the most serious political obstacle to adopting proportional representation in the United States is the opposition of political elites, particularly elected officials of the two major parties. Even though the electoral reform movement is relatively new in the United States, it is already clear that elected officials are some of the most vocal opponents to this kind of political change. This is to be expected. Most Democratic and Republican officials are unlikely to enthusiastically embrace a change that would allow minorparty candidates to take some of their seats and that would probably prevent either party from ruling on its own ever again. So reformers face a dilemma: the politicians who are in the best position to enact electoral reform are those who are elected by and benefit from the current system. This political dilemma is nothing new—it has been faced by virtually every effort to reform elections in the United States, from women’s suffrage to term limits. Historically, politicians in the two major parties have used whatever political weapons were at their disposal to maintain their political monopoly and to block the intrusion of minor parties into their political domain. They have been particularly adept at using election rules to discourage third parties and their candidates. State election rules have often been deliberately constructed to hamper third-party efforts by preventing their automatic inclusion on the ballot and by setting up onerous petition requirements.6 Unquestionably, however, the single-member plurality system has always been the main bastion against the rise of minor parties. As we saw in chapter 4, its rules make it extremely difficult for minor-party candidates to win office. This barrier to minor-party power is one of the best defenses the two major parties have against challenges to their dominance, and for this reason alone they will oppose alternative voting systems such as proportional representation. To appreciate how far the two parties will go to oppose proportional representation and to prevent any growth in minor-party influence, consider the role they played in the effort to repeal PR in New York City in 1947. One of the strangest features of the repeal campaign was significant Republican leadership support of the effort to reinstate single-member plurality elections. Under proportional representation, the Republicans had actually been able to increase their political representation in the New York City Council. Before PR the Democrats completely dominated city government. In 1935 the Democrats elected 95.3 percent of the council with only 66.5 percent of the votes.7 In PR elections in 1937 the Democrats were able to

The Political Prospects for PR

243

win only 50 percent of the seats, with the Republicans winning 11.5 percent and the rest of the seats going to minor parties. During the 1947 repeal attempt, a majority of the Republican council members sought to save their seats and opposed the elimination of PR—clearly out of self-interest. But most Republican party leaders worked vigorously for the repeal campaign and contributed to its victory.8 As a result of the return to SMP elections, Republican representation went from three seats to one seat in the first post-PR election, and since then the Republicans have sometimes failed to elect even one member to the city council. Clearly, the Republicans felt so strongly about preserving the two-party system that they were willing to lose seats on the council to see its return—a revealing example of how adamant the dominant parties can be in their opposition to sharing power with minor parties. As in New York, many other successful efforts to repeal proportional representation in U.S. cities were led by local leaders of the formerly dominant parties. Some electoral scholars argue that it was this highly organized opposition from political elites previously privileged by SMP arrangements—and not any faults of PR—that was a key factor in many repeal campaigns. As the political scientist Bernard Grofman concluded: “The repudiation of PR by all but one of the American cities which used it has often been alleged to demonstrate its unsuitability for use in the United States. Actually, as far as we are aware, in those cities where the [single transferable vote] was used it worked well, and its ‘undesirable’ consequences were mostly in the eyes of the previously impregnably entrenched majority beholders.”9

Overcoming the Obstacles Some of the political obstacles to proportional representation—particularly the opposition of Republican and Democratic officials—seem very formidable at first glance. But they can be overcome. Indeed, such obstacles have been overcome in the past by other movements to make elections more democratic in the United States. If the opposition of elected officials were enough to prevent reform, women would not have the vote, racial minorities would still be excluded from the polls, and we would never have had term limits or reforms in the campaign financing process. History has shown us that it is possible to change election systems for the better, in spite of the initial opposition of political elites.

244

The Political Prospects for PR

Overcoming this kind of opposition requires building a widely supported reform movement, for which PR advocates have already been laying the political groundwork. They have been building bridges to other political organizations and reform movements. Environmentalists, public interest groups, third parties, civil rights activists, campaign finance reformers, feminists, good government organizations, term limits supporters, and others have expressed strong support for PR and have the potential to form an effective political coalition around this issue. Let us now turn to a consideration of what this PR movement can do to most effectively promote this reform and to overcome the political obstacles discussed earlier.

Wooing Disgruntled Major Parties Despite the example of New York City mentioned earlier, it would be a mistake for reformers to write off the major parties as potential supporters of proportional representation. The opposition of the Democratic and Republican parties may not be as automatic nor as monolithic as it first appears. Certainly the national party organizations will be opposed to PR, but this may not be uniformly true of the parties on the local and state levels. Consider a state like Massachusetts, where one party—the Democrats—effectively monopolizes political life. Republicans have nothing like a fair level of representation: the state delegation to the U.S. House of Representatives is entirely Democratic. There are similar situations in other areas of the country. In Idaho, for instance, it is the Republicans who are overrepresented and dominate political life. About 30 percent of the state votes Democratic, but in 2000 only 8 percent of the seats in the state Senate went to Democrats. In such situations, the party that is suffering from the injustices of the single-member plurality systems—the underrepresentation, the gerrymandering, etc.—has a strong interest in proportional representation, and could prove to be a valuable political ally to local PR movements. Major parties who have suffered from the unfairness of the SMP system have become proponents of PR in other Western countries. In Great Britain, for instance, the Labour party was out of power throughout most of the 1980s and 1990s, and began to explore the PR option. Within the party, there arose the group promoting proportional representation called the Labour Campaign for Electoral Reform—the largest single-issue membership organization working in the party. And eventually the leader of the

The Political Prospects for PR

245

party, Tony Blair, ran for office on the pledge of allowing a public referendum on the electoral system. (Unfortunately, after the Labour party won control of the parliament, many party members and Blair himself began back-pedaling from the promise of a referendum.) In New Zealand, the public referendum that eventually brought PR to that country in 1993 came about largely because of support from various major parties. In the 1987 election campaign, the Labour prime minister promised to allow a public referendum on PR; but as with Tony Blair, he retreated from this pledge once elected. In the next election, the rival National party sought to exploit this broken promise by vowing to allow such a referendum if they came to power—a position they found themselves unable to back away from once they won parliament in 1990. As these examples illustrate, major-party opposition to PR is not always assured and reformers need to cultivate the support of these powerful organizations whenever possible. Finally, it should also be kept in mind that even if major party officials oppose proportional representation, many of the grassroots members of these parties will find PR in their interest. Democratic and Republican voters in districts and states where they are routinely outvoted and under- or unrepresented would benefit greatly from a PR system that allows both parties to win their fair share of seats. The vast majority of wasted votes cast in elections in the United States are not cast by third-party supporters, but by frustrated members of the two major parties. Thus PR reformers could find a very sympathetic ear among major party voters who are currently suffering from the political injustices of our single-member plurality system. For instance, it is not surprising that in Seattle, a city long dominated by Democrats, some Republican party supporters have expressed interest in switching to proportional representation.

Legal Pressure While most elected officials cannot usually be expected to voluntarily promote electoral system reform, it is sometime possible to pressure them into reluctant support. This has been the case in the area of voting rights. The Voting Rights Act and its amendments have given electoral reformers the opportunity to challenge in federal court plurality electoral systems that discriminate against minority voters. The courts have been particularly willing to overturn plurality at-large systems that have made it impossible for racial and ethnic minorities to elect any officials of their own. Some legal

246

The Political Prospects for PR

scholars, such as Steven J. Mulroy, have argued that the federal courts should not automatically assume that single-member districts are the only appropriate alternative to at-large systems, and he has urged lawyers and judges to pursue proportional systems as a remedy in voting rights cases.10 Most federal judges still seem hesitant to embrace alternative electoral systems, but some voting rights attorneys, such as Edward Still, have been successful at using legal challenges under the Voting Rights Act to pressure local officials into accepting these systems. City and county officials are often averse to spending the considerable amount of money necessary to defend their at-large system in federal court, and thus are receptive to out-ofcourt settlement offers. In several cases in Alabama, Texas, and elsewhere, voting rights attorneys have insisted that semiproportional systems—cumulative and limited voting—be substituted for at-large elections, and local officials have (often reluctantly) accepted these systems to avoid further costly litigation. These more proportional systems now are being used in more than 100 local jurisdictions in the United States and constitute the most successful effort to date to promote the adoption of alternative electoral systems. Currently, however, this kind of legal pressure on officials can have only a limited impact because it can be applied only where the representation of racial and ethnic minorities is at issue. Some legal scholars have hoped that a broader legal challenge to plurality elections, based on constitutional grounds, could be used to force the adoption of more proportional systems on a wide scale in the United States. They have argued that the constitutional principle of one-person, one-vote, based on the equal protection of the laws clause of the Fourteenth Amendment, could reasonably be interpreted to require the use of proportional representation in American elections. John R. Low-Beer puts forward one version of this argument in his Yale Law Journal article entitled “The Constitutional Imperative of Proportional Representation.”11 He argues that the consistent misrepresentation that takes place in single-member plurality elections—particularly the misrepresentation caused by gerrymandering— violates the U.S. Supreme Court’s stated principle of “fair and equal representation for all citizens.” He demonstrates that it is inconsistent for the justices to assert that the dilution of the strength of citizens’ votes through malapportionment is unconstitutional, while maintaining that the more severe vote dilution that routinely takes place through advertent and inadvertent gerrymandering is not.12 The only effective way to ensure fair and equal representation, Low-Beer argues, is to adopt proportional representation. PR is

The Political Prospects for PR

247

the simplest and most elegant way to ensure both minority representation and majority rule. He concludes that “the time has come to take seriously the argument that the Constitution requires admitting the principle of proportionality into the electoral systems of state and local government.”13 However, given the current ideological bent of the Supreme Court, it seems unlikely in the near future that the Court will be open to such arguments in favor of proportional representation. Indeed, some of the Court’s latest decisions about election systems suggest it could be hostile to PR. For example, in 1997 the Court ruled in Timmons v Twin Cities that the states had the right to ban the use of “fusion” ballots in elections. Fusion is the practice of allowing candidates to be listed on the ballot several times under different parties’ labels. For instance, a candidate might be listed separately as both a Green candidate and as a Democratic candidate, and voters could choose under which party label they would cast their vote for that candidate. This ballot system is largely seen as a way to help the election efforts of minor parties. In the Timmons case, Minnesota had passed legislation outlawing this kind of ballot, and the New Party challenged this law. The Supreme Court ruled in favor of the state, arguing that the states were justified in maintaining electoral laws—such as a ban on fusion—that favor the two major parties over minor parties. Writing for the majority, Chief Justice Rehnquist arrived at a remarkably political conclusion: “The States’ interest permits them to enact reasonable election regulations that may, in practice, favor the traditional two party system and that temper the destabilizing effects of party splintering and excessive factionalism. The Constitution permits the Minnesota Legislature to decide that political stability is best served through a healthy two party system.”14 Clearly Rehnquist had accepted the argument that multiparty systems create political instability—one the myths about proportional representation that was critiqued in an earlier chapter. The Timmons ruling suggests that the Court would also support a state’s right to ban proportional representation. In this regard, it is significant that one Justice’s opinion in this case explicitly mentioned PR and alluded to its alleged tendency to create instability: “Systems of proportional representation may tend toward . . . fragile coalitions that diminish legislative effectiveness.”15 These kinds of misinformed judicial opinions about PR do not bode well for the chances of using the federal courts to force states to adopt this voting system. For this reason, successful legal challenges to winner-take-all electoral systems are now likely to remain confined to those pursued under the Voting Rights Act.

248

The Political Prospects for PR

Capitalizing on Electoral System Failures A number of scholars studying electoral system change in other countries have noted that the chances of reform increase dramatically when electoral systems experience “systemic failures.”16 Systemic failures occur when the electoral system dramatically malfunctions and creates an obviously problematic election outcome—an outcome that provokes public discontent. Such systemic failures focus attention of the problems of the electoral system, increase public pressure for change, and make it difficult for politicians to ignore the issue. If such an electoral crisis were to occur in the United States, this would clearly serve to energize the electoral reform movement here. A prime example of this kind of systemic failure occurred in New Zealand and eventually resulted in the abandonment of SMP and the adoption of PR. Several election incidents occurred that served to focus public attention of the political deficiencies of their single-member plurality system. First, in 1978 and again in 1981, national elections resulted in the party that came in second place in the polls winning a majority of seats in the parliament. Such obviously unfair and undemocratic results created a growing call to consider alternatives to the SMP system. Then, after the 1984 election, the new government embarked on a radical change in economic policy—one that drastically deregulated the economy and cut back on social services. None of these policy initiatives had been revealed during the campaign, and many of them were highly controversial. Moreover, the party that initiated these changes had received only 43 percent of the national vote. There was a widespread feeling that the election had produced a government that was unaccountable and unresponsive to a majority of the citizens. The growing sense of political crisis and public discontent is what eventually fueled the successful effort to adopt proportional representation elections. Many citizens believed that the coalition governments that would come with PR would be more representative and more accountable. Persistent systemic failures in plurality voting in Great Britain have also increased interest in alternative systems there. There the problem has been the consistent and blatant misrepresentation of political parties. The centrist third party, the Liberal Democrats, has routinely been denied anything close to its fair share of seats in parliament. In the 2001 parliamentary elections, for instance, that party won 19 percent of the votes, but was relegated only 8 percent of the seats. Equally disturbing to many British voters are the numerous manufactured majorities that SMP elections have produced. In-

The Political Prospects for PR

249

credibly, no ruling political party in that country has won a majority of the vote in national elections since World War II. For almost twenty years, the Conservative party ruled Britain while never receiving more than 44 percent of the vote. In recent years, the Labour party has taken their place, but again it has not received a mandate from a majority of the voters. Some citizens there find it disturbing that parties of both the left and the right have been able to impose their policies on a nation where the majority of voters actually cast their ballots against those ruling parties. These numerous and obvious incidents of electoral injustice are one reason why the PR movement is much further along in Great Britain than it is in the United States. Unfortunately, we have not had this same kind of dramatic malfunctioning of our single-member plurality system on the national level. There was a clear systemic failure in the 2000 election, but it occurred in the wrong part of the electoral system. The Electoral College system managed to elect the candidate who lost the popular vote—an obviously undemocratic result that triggered a great deal of criticism of this system and calls for it to be abandoned in favor of the direct popular election of the president. But this failure in the Electoral College did little to help the campaign against SMP legislative elections. What is really needed is an obviously unfair result in a congressional election—such as the persistent manufactured majorities that have occurred in Great Britain. However, manufactured majorities have been less common in the United States. And when they do occur, they often go largely unnoticed because the media never bother to compare the parties’ share of seats to their proportion of the vote in national or state elections. We have actually had some manufactured majorities in the House of Representatives in recent years—the Republicans won a majority of the seats in that body in 1996, 1998, and 2000, even though they received only 48.9, 48.9 and 47.9 percent of the nationwide vote in those years. But these results went unreported in the press, and these manufactured majorities were not as outrageous as those in Great Britain where the winning party has often received less than 43 percent of the vote. The relative lack of severely distorted results in national elections in the United States is largely attributable to the absence of successful minor parties in the United States. In other SMP countries, it has been the presence of minor parties that attract substantial numbers of voters—such as the Liberal Democrats in Great Britain—that has caused much of the distortion in party representation. Political scientists have attributed the lack of viable minor parties in the U.S. to a number of factors, including unusually difficult ballot

250

The Political Prospects for PR

access requirements and the ability of dissenters to run as candidates in major party primaries instead of joining a minor party.17 If, however, a minor party were to grow to a substantial size in the United States, this could create the kind of clearly distorted representation that would give the PR movement an important boost. Imagine, for example, that the Libertarian party in the United States garnered 20 to 30 percent of the nationwide vote but received few or no congressional seats. The unfairness of such an outcome would be obvious even to those who did not share that party’s political views. Furthermore, if the Libertarians were to take that large a portion of the vote away from the Democrats and Republicans, this would likely create the kind of clear manufactured majorities that have plagued Great Britain. But while such a third-party scenario is possible, it does not seem likely to develop in the near future. However, while there has been a lack of systemic failures and outrageous election results on the national level, we can find many examples of these on the state level and local level. Every election for the U.S. House of Representatives produces some state delegations that are clearly unrepresentative of the level of party support found in those states. As mentioned earlier, Massachusetts’ delegation contains no Republicans even though this party usually attracts 30 percent or more of the vote in these elections. This is clearly an unfair and undemocratic result—a good example of the gross distortions possible in SMP elections. Similarly outrageous electoral results also occur on the city level. Cities that use at-large plurality elections are particularly apt to produce unfair and unrepresentative results. Unfortunately, as with the national elections, the press usually does not even calculate or publicize the percentage of the statewide or citywide vote won by each party, so most citizens are unaware of how unrepresentative the results of SMP state elections can be. Reformers need to highlight these kinds of systemic failures and draw public attention to them. Unfairness of racial representation may be another electoral injustice that can be taken advantage of by PR reformers. Clear examples of unfair representation of racial and ethnic minorities continue to occur on the state and local levels. And the situation for minorities may get worse in the near future as well. As shown in chapter 6, the Supreme Court has been looking skeptically at the majority-minority districts that have allowed minorities to win more seats in Congress, state legislatures, and city councils in recent years. If a substantial number of these majority-minority districts were to disappear and along with them the minority representatives, this could prompt

The Political Prospects for PR

251

a severe political reaction on the part of minority communities. This lack of fair representation would be an important and politically useful example of the serious problems created by our continued reliance on single-member plurality elections.

Building a Movement Based on Public Discontent While serious election failures would certainly be useful to the proportional representation movement in the United States, there are also other ways to build public support for this movement. PR activists could follow the political path blazed by other election reform movements such as term limits and campaign finance reform. These movements have demonstrated that many Americans can become galvanized around electoral reforms that promise to reinvigorate American democracy. Both of these reforms seemed in their initial stages to be political “long shots,” and they faced many of the same political obstacles—such as hostile politicians—that PR faces today. Yet both have achieved some success by using referenda to pass reform legislation on the state and local level. These changes were accomplished not so much by capitalizing on a specific election fiasco, but by tapping into the general level of political dissatisfaction felt by many Americans. Even when there is not a specific election crisis in the United States, there remains a serious underlying crisis of public confidence in the political system. Polls consistently reveal that three-quarters of Americans distrust the federal government.18 They also show that considerably more than half of Americans (62 percent) do not think that public officials care what they think, 42 percent believe that they have no say in what government does, and 64 percent think that government is “pretty much run by a few big interests looking out for themselves,” rather than for the benefit of all citizens.19 Americans are also very dissatisfied with their party system. Less than a quarter of them (23 percent) believe the two-party system “works fairly well,” and a clear majority wants to see other parties challenge the Democrats and Republicans.20 Our abysmal level of voter turnout—dropping to a mere 33 percent in the 1998 congressional elections—is yet another indication of the depth of political alienation felt by the public. The term limits and campaign finance reform movements have been able to build on this general sense of political dissatisfaction, and the proportional representation movement has begun to do the same. Indeed,

252

The Political Prospects for PR

many of the people already being drawn to PR—including environmentalists, public interest lawyers, civil rights activists, labor organizers, feminists, third-party supporters, and community organizers—are clearly discontented with the workings of American politics. They feel that the current American political system has not been responsive to large segments of the public, and that basic political reforms are needed. They see PR, along with other issues such as campaign finance reform, as part of an effort to make the political system more open and more democratic. They are not so much interested in remedying technical faults in the electoral system; rather they want to improve access to power for themselves and other Americans traditionally left out of the political process. But if the proportional representation movement is to continue to grow, it must appeal not just to political activists but to average Americans as well. To accomplish this, reformers must tap into the political frustrations felt by most voters. They must show these voters that many of their frustrations are related to a flawed SMP electoral system. Citizens need to realize, for example, that they have little choice at the polls because we have a single-member plurality electoral system that actively discriminates against third parties. And the reason they feel that voting is a waste of time is that much of the time their votes are actually wasted. Americans must also begin to recognize that one reason they feel that government does not represent them is because many of them are literally denied representation in our current SMP system. If this strategy of mobilizing political discontent is successful, then the PR movement will be able to follow in the footsteps of term limits and campaign finance reform and garner the public support necessary to win some significant political victories. Most Americans want our democracy to work better. They want more parties, more say in government, and policies that better reflect the public will. Interest in PR will grow if it is made clear that electoral reform is not simply about new vote counting procedures, but also about revitalizing American democracy. PR must be seen as a part of an effort to empower more citizens and to create a political system that will be more responsive to the needs of all Americans. A historical perspective is helpful here. A citizen revolt against politics as usual prompted many of the early-twentieth-century experiments with PR in American cities. Average citizens were fed up with the waste and corruption that were typical of the political machines that ruled many large American cities at that time. They were ready to seriously consider changes in the

The Political Prospects for PR

253

structures of elections and city government in an effort to regain public control over these institutions. PR was usually voted in with a package of other structural changes, including nonpartisan elections and the city manager form of government. Citizens’ frustrations with entrenched politicians and dysfunctional political institutions set the stage for these early municipal reforms, and this same scenario is certainly possible today.

One Reform Strategy: Local Experimentation As the PR movement grows it must face a decision about where best to focus its energies for reform. Clearly the most rapid way to spread the use of proportional representation elections in the United States would be to promote its adoption for state and federal elections. But electoral system change at this level is also the most difficult politically. At the state level, for instance, the pro-PR movement would have to be large and well-funded in order for its message to be reach voters state-wide. Much work would also have to be put into developing support from a variety of statewide political organizations that would form an effective political coalition for electoral change. In addition, the higher the political level, the higher the political stakes, and the more intense the likely opposition to PR. For all of these reasons, the most practical path toward electoral reform in the short run may be a bottom-up approach—one that begins with electoral experiments with proportional representation on the local level. This local-level approach is, of course, how the initial wave of proportional representation began in the United States and it resulted in two dozen cities adopting this alternative voting system. The major advantage of this approach is that the adoption of PR by city councils, school boards, or county commissions could be accomplished relatively easily. No large national or statewide movement would be necessary. A small but wellorganized band of local reformers could build a citywide political coalition that would have a good chance of winning. In many areas, there would be no need to change state legislation or constitutions to adopt PR on the local level, and reform could be accomplished with a referendum to change the city or county charter. Also, organized opposition would be less intense on the local level. While local party leaders and elected officials might oppose such a change, the opposition of organized interest groups is less likely. Of course the accompanying disadvantage is that it would also be relatively easy

254

The Political Prospects for PR

to repeal PR on the local level—the fate that befell many of the early reform efforts. However the relative lack of political obstacles still makes the local level the most likely arena for beginning efforts to make PR a reality in the United States. If local jurisdictions began to adopt proportional representation, it could become contagious and encourage more widespread adoption of this system. Once PR could be seen at work, it would become demystified for more and more Americans. The advantages would become more obvious and concrete, and the disadvantages could be seen as minimal in practice. Voters in an SMP town dominated by a one-party government could look at a nearby PR town and see a much more diverse and representative government. It would become increasingly clear that large numbers of votes need not be wasted as with single-member plurality arrangements and that nearly all citizens can come away from elections with someone representing their views in government. As mentioned earlier, one main obstacle to PR in the United States is the lack of public awareness of this system and its advantages. Once this system can be appreciated at first hand, public support is likely to grow quickly. Some local experimentation with alternative voting systems is already taking place. As noted earlier, voting rights litigation has ushered in the use of semiproportional systems in many dozens of local jurisdictions—mainly in the South. Indications are that these systems are working relatively well, and thus it is possible that they could become contagious, with citizens demanding that these better voting systems be used in other elections as well. Several attempts have also been made to adopt PR on the city level. In the 1990s, referendums were put on ballots in Cincinnati and San Francisco to use the single transferable vote for city council elections. In both cases, reformers were able to put together coalitions of local groups to promote this change. In San Francisco, for instance, the PR referendum received support from Latino and Black organizations, the local National Organization for Women chapter, the Green party, and the United Farm Workers and other labor unions. Significantly, some leaders in the dominant Democratic party supported this change. There was also some significant support on the conservative side of the aisle, since San Francisco Republicans are an underrepresented minority there and were naturally interested in winning their fair share of seats. Unfortunately, the PR movements in these two cities had only a few months, and little money, to educate citizens about this new kind of voting

The Political Prospects for PR

255

system. Although the final vote was close, neither referendum won. However both garnered an impressive level of citizen support for such a new political idea. In San Francisco, the PR referendum won 44 percent of the citywide vote and in Cincinnati, 45 percent of the vote. These close calls indicate that many Americans are willing to seriously consider PR on the local level and that this remains a fertile political ground for PR organizers. It may be only a matter of time until some cities adopt PR for their local elections in the United States.

Reasons for Optimism: PR Reform in a Larger Historical Context While the proportional representation movement is still relatively young in the United States, there are reasons to be optimistic about its prospects in the long run. First, the case for PR rests solidly on political principles and values that traditionally have been central to American democracy. If Americans value political equality, fair elections, a wide range of choices at the polls, representative legislatures, and lively political debate—and they do— then clearly proportional representation will have an intrinsic and powerful appeal in this country. Proportional representation is the electoral system that best embodies these basic principles of democracy—principles that are a vital part of American political culture. In other words, the democratic political instincts of most Americans will incline them to be supportive of PR. The public response to PR referendums that took place in Cincinnati and San Francisco demonstrates that when Americans finally hear about PR many of them quickly come to appreciate its advantages and support its adoption. This bodes well for the PR movement in the United States. PR advocates can also take heart from the fact that American elections have gone through a large number of reforms throughout our history. American democracy is, and has been, very much a work in progress. The founding fathers, as insightful as they were, did not invent the perfect democracy. And so during the life of our country, many political movements have struggled to make our political system more responsible and accountable to the public. Many of those movements—such as women’s suffrage—have promoted changes in elections. It can sometimes appear that election rules are static and immune from change; and this may actually be true in the short run. But in the long run, the opposite has been true. Elections and the rules sur-

256

The Political Prospects for PR

rounding them have been slowly but constantly evolving in the United States—usually in a more democratic direction. We now elect U.S. Senators directly, instead of their being chosen by state legislatures. The right to vote has been extended to women and minorities. We no longer have poll taxes or literacy tests. Primaries are now utilized in presidential elections. Many states have adopted the initiative and referendum to allow citizens to vote directly on policies. Many obstacles to voter registration have been removed. Restrictions on campaign contributions have been put in place. Term limits and the public financing of elections have also been adopted in several states. Even our voting systems have not remained static. For the first 150 years of the Republic, at-large elections—the most unfair form of plurality voting—were common throughout the country. As recently as the mid 1950s, more than half of the state legislative districts in the then 48 states were multimember at-large districts, and only nine states chose all their representatives in single-member districts. Writing about state elections in 1955, the political scientist Maurice Klain observed: “It is the single-member district that stands outside of history, an alien to the past, a newfangled and upstart thing.”21 The last fifty years have seen a dramatic decrease in the use of atlarge elections in states and many cities as well. So even within the plurality framework, the electoral system has been changing. In short, the American election system is an ever evolving one—one that has often been modified in response to political pressures. In this context of constant change, the adoption of proportional representation in the United States seems less of a remote possibility. Our ideas about democratic elections have changed over time as well. At this point, it may seem unlikely that most Americans will ever come to embrace the notion that all citizens should have the right to representation. But it should be remembered that most Americans once had trouble embracing another basic democratic concept: that everyone should have the right to vote. As Alexander Keyssar explains in his book, The Right to Vote: The Contested History of Democracy in the United States, throughout most of our history large numbers of Americans resisted the idea of universal suffrage.22 He found that virtually all the founding fathers were supportive of restrictions on the right to vote—restrictions based either in wealth, gender, or race. And well into the twentieth century, many Americans opposed the efforts of women and minorities to win the vote. But that eventually changed, and virtually all of us now take the principle of universal suffrage for granted. If Americans can come to see the right to vote as an essential element of dem-

The Political Prospects for PR

257

ocratic elections, then it is not at all far-fetched to think that they could also come to accept and promote the right to representation as the next logical step in the democratization of American elections. Finally, proponents of proportional representation in the United States can also take inspiration from the trend of electoral reform on the international level. During the last hundred years the worldwide trend in electoral systems has been away from plurality and majority systems and toward proportional representation. With hardly an exception, democratic countries have been replacing these unrepresentative plurality systems with various forms of proportional representation—not vice versa. We are now only one of a very few developed nations that continue to use plurality arrangements, and signs of a movement toward PR in all these other countries are clear. In addition, emerging democracies around the world have also followed the trend toward proportional representation. In Eastern Europe and the former Soviet Union, very few of these fledgling democracies have chosen singlemember plurality elections, with the vast majority opting for some form of proportional or semiproportional voting system.23 Proportional representation has also been the system of choice for many new democracies in Africa, including South Africa. Particularly interesting for those in the United States has been the progress of the proportional representation movement in Great Britain. That country is the traditional home of single-member plurality elections, and nowhere has this system been more entrenched. And yet there are many hopeful signs that Britain is moving increasingly toward PR. Polls routinely show considerable public support for PR. And as mentioned earlier, when Prime Minister Tony Blair was elected in 1997, he pledged a public referendum on electoral reform. He also appointed an Independent Commission on the British Voting System, which produced a report recommending a change to a more proportional electoral system.24 Blair eventually backed away from supporting those recommendations and his own promise of a referendum, but progress toward PR has continued on other fronts. Britain now uses PR for electing its members to the European Parliament. More significant has been the adoption of mixed-member proportional representation elections for the newly created parliaments in Wales and Scotland. These were important victories for the PR movement in Great Britain. If that country eventually abandons SMP elections for a more proportional system for its national elections, this could have significant implications for the United States. It will be increasingly difficult for the United States to continue to

258

The Political Prospects for PR

cling to winner-take-all elections if the ancestral home of this system has rejected it for proportional representation. Given these national and international trends, the temptation is to say that the tide of history is on the side of proportional representation. And in a sense it is. But such notions should not be allowed to give the impression that this change will be easy or inevitable. The door to electoral reform is not standing wide open, simply waiting for us to walk through. But neither is it closed and bolted. It remains slightly ajar, on somewhat rusty hinges, awaiting the time when a movement of dedicated reformers and ordinary citizens will put their collective weight against it and force their way through. When they do, the United States will have taken another important step forward in its long journey toward becoming a truly democratic society.

Appendix A Seat Allocation Formulas in Proportional Representation

In party list proportional representation, the seats in the legislature are allocated in proportion to the party vote. A formula must be used to accomplish this allocation, and several of them are available. Three of the most common are described here. Tables are included to illustrate the operation of each formula and how the various formulas would allocate the seats, given the same distribution of votes among four parties. These formulas vary somewhat in how proportional their results are and whether they slightly favor certain sizes of parties over others.

Largest Remainder Formula This is one of the simplest formulas. The first step is to calculate a quota. One commonly used quota is the “Hare quota,” and it is determined by taking the total number of valid votes in a district and dividing it by the total number of seats at stake in the district. In table A.1, voters have cast 99 votes, which are divided by nine seats to produce a Hare quota of 11. Each party’s vote is then divided by the quota, and the party receives a seat for each whole number produced. Thus party A received 36 votes, divided by 11, which yields three seats. As in most cases some votes are left over—three votes for party A. The same process is followed for allocating seats to the other parties.

260

Seat Allocation Formulas

table a.1 Seat Allocation Using Largest Remainder Method

Remaining Votes

Second Allocation of Seats

Final Seat Total

Parties

Votes

First Allocation Of Seats

A

36

3

3

0

3

B

24

2

2

0

2

C

22

2

0

0

2

D

17

1

6

1

2

Note: 99 voters, 9 seats, quota of 11 votes.

If there are any seats left over after this process, they are allocated to the parties with the largest number of remaining votes. In this example, one seat is left to be allocated, and it is award to party D because it has the largest remainder. One variation on this system uses a different quota, called the “Droop quota.” This divides the total votes by the total number of seats plus one. This slightly reduces the proportionality of the outcome.

D’Hondt Formula This system is named after the Belgian lawyer Victor d’Hondt, although the same formula was actually used first by Thomas Jefferson as a way to allocate congressional seats among the states. It is sometimes also called the “method of highest averages.” This system allocates seats by using what are called “divisors.” Each party’s vote is divided by a series of these divisors. In this formula the divisors are 1, 2, 3 ,4, etc. These divisions produce numbers called averages. Seats are allocated successively to the parties with the highest averages. Each time a party is allocated a seat, its vote is then divided by the next divisor to produce a new average for that party. In the example in Table A.2 , party A has the highest average in the first round with 36 votes

261

Seat Allocation Formulas

table a.2 Seat Allocation Using d’Hondt Formula

Parties

Votes/1

Votes/2

Votes/3

Votes/4

Final Seat Total

A

36(1)

18(4)

12(6)

9(9)

4

B

24(2)

12(7)

8

2

C

22(3)

11(8)

7.33

2

D

17(5)

8.5

1

Note: 99 voters, 9 seats. Number in parentheses indicates order in which seats are allocated.

and so receives the first seat. Its votes are then divided by two to produce a new average of 18. Now party B has the highest average with 24 votes. It receives the next seat and its average is divided by two and reduced to twelve. This process continues until all the seats are allocated. The d’Hondt method is known for favoring somewhat the largest party at the expense of the smaller parties. In the example in Table A.2, it allocates four seats to party A, while the largest remainder system allocated only three seats to party A. Because of this tendency, other divisor formulas have been invented to try to minimize this large party bias. One of them, the SainteLagüe formula is described next.

Sainte-Lagüe Formula This is another system that uses divisors. It uses the same basic approach as the d’Hondt method, allocating seats to the parties with the highest averages. But it uses a different series of divisors. Instead of 1, 2, 3 etc., SainteLagüe uses the odd numbers 1, 3, 5, etc. This produces a more completely proportional allocation of seats. Note that the results of the Sainte-Lagüe formula shown in Table A.3 are the same as that for the largest remainder system. Both are considered more proportional than the d’Hondt method.

262

Seat Allocation Formulas

table a.3 Seat Allocation Using Sainte-Lagüe Formula

Parties

Votes/1

Votes/3

Votes/5

Votes/7

Final Seat Total

A

36(1)

12(5)

7.2(8)

5.14

3

B

24(2)

8(6)

4.8

2

C

22(3)

7.33(7)

4.4

2

D

17(4)

5.67(9)

2

Note: 99 voters, 9 seats. Number in parentheses indicates order in which seats are allocated.

Other Systems The preceding systems are only three of the most common ones, and several others have been devised. The modified Sainte-Lagüe formula uses the divisors 1.4, 3, 5 etc., which makes it somewhat more difficult for the smallest parties to win seats and thus favors slightly the mid-size parties. The Imperiali formula uses the divisors 2, 3, 4, etc., and this tends to favor the larger parties.

Appendix B Vote Counting and Ballot Transfer in the Single Transferable Vote

The process of counting the votes and determining the winner in the single transferable vote system of PR is a bit complicated, so it deserves some detailed explanation. The accompanying diagram shows how the basic process works. First, a threshold is established—the minimum number of first-choice votes a candidate must receive to get elected. The threshold usually consists of the total number of voters divided by the number of seats to be filled plus one. Thus in a nine-seat district, a candidate would need just over 1/9⫹1 or one-tenth of the vote, to be elected on the first count. Next, all the first-choice votes are counted and all the candidates that reach the threshold are declared elected. If a winning candidate has excess votes over the threshold, then those surplus votes are transferred to the voters’ second choices. There are several ways to transfer these surplus votes. The simplest approach would be to take the last ballots that come in over the threshold and transfer them to the next preferred candidates. Thus if the threshold were 100 votes and candidate A received 200 votes, the last 100 votes would be transferred. But this approach is faulty because the next preference for the last 100 voters may not be representative of all the ballots for that candidate. To avoid this problem another mode of transfer is often used: all 200 ballots are transferred at a fraction of their value. The number of surplus votes (100) is divided by the total number of ballots (200) to produce

264

The Single Transferable Vote

figure b.1 Diagram of STV Vote Counting Procedures

the fraction of one half. Then all the second-choice votes for each candidate are added up and multiplied by one half to determine the actual number of ballots transferred to each. Thus if candidate B received 120 second preference votes and candidate C received 80, these vote totals would be multiplied by one half to get 60 and 40 ballots respectively—equaling the total of 100 surplus ballots to be transferred. This method ensures that all surplus ballots are transferred in proper proportion to the other candidates.

265

The Single Transferable Vote

table b.1 Counting of a Single Transferable Vote Election 1st Count 2nd Count

3rd Count

4th Count

5th Count

Transfer of Gomez’s Number votes and Of Votes results.

Transfer of Royce’s votes and results.

Transfer of Dains’ votes and results.

Transfer of Campbell’s votes and results.

—

Candidates

400

+300 700

700

700

Martha Dains* (Rep.)

2,300

2,300

+500 2,800

2,501

2,501

Hilde Bauer (Reform)

2,000

2,000

2,000

+200 2,200

+100 2,300

Robert Gomez* (Dem.)

2,900

2,501

2,501

2,501

2,501

Susan Rosen* (Ind.)

1,800

+99 1,899

+100 1,999

+99 2,098

+600 2,698

Edward Royce (Rep.)

600

600

—

—

—

Douglas Campbell (Dem.)

*Winning Candidates.

Once the surplus votes are transferred, then the votes are counted again to see if any candidate has reached the threshold. If no one does, then the candidate with the least number of votes is eliminated and his or her votes are transferred to the candidate ranked next on the ballot. Then the votes are recounted again to see if any now has reached the threshold. This process continues until all the seats are filled. It is helpful to see how this process works in a specific example. Consider, for example, table B.1, which shows how the vote counting and transfer process would work in a hypothetical election where six candidates are vying for three seats. The threshold in this case is 2,501 votes. After the first count, the

266

The Single Transferable Vote

Democrat Gomez has 2,900 voters and he is declared elected. But Gomez actually has 399 more votes than he needs to win. These votes are considered wasted if they stay with Gomez, so they are transferred to the second choices on the ballot. In the second count, we see the effect of this transfer. The other Democratic candidate, Campbell, gets 300 of those second choice votes, and the independent candidate, Rosen, gets the other 99. The vote totals are now recalculated to see if anyone is now over the threshold. No one is, so the next transfer takes place. The candidate with the least chance to win is eliminated and his or her votes are transferred to the voters’ second choices. This candidate is Royce, the Republican, and 500 of his votes are transferred to the other Republican candidate, Dains; and the other 100 votes are given to Rosen. Again the votes are recounted to see if anyone has reached the threshold. Dains has reached it with 2,800 votes and so she is declared elected. Once again her excess votes are redistributed to their second choices—200 to Bauer, and 99 to Rosen. But still no one has reached the threshold, so again the lowest candidate is eliminated and those votes transferred. That candidate is Campbell, the Democrat, and 100 of his votes go to Bauer, and 600 go to Rosen. This puts Rosen, the independent candidate, over the threshold with 2,698 votes, and she is the last one elected.

Appendix C The Forgotten History of Proportional Representation in the United States1

The United States has had a long tradition of single-member district, winner-take-all elections. So it is hardly surprising then that few Americans are aware of our history of experimentation with proportional representation (PR) elections. During the first half of the twentieth century, two dozen American cities used for a time the single transferable vote (STV)—a form of proportional representation that is sometimes called “choice voting” today. The story of how proportional representation came to be adopted and eventually abandoned provides some useful insights for those interested in proportional representation and in the politics surrounding electoral reform.

The Historical Roots of Proportional Representation The political roots of proportional representation in the United States originated in the Progressive Movement of the early twentieth century. Besides such issues as child labor laws, anti-monopoly legislation, and women’s suffrage, Progressives were also interested in government reform. Many were particularly concerned about the corruption in urban governments. Large cities often were dominated by party “machines,” of which Tammany Hall in New York City was the most infamous. Bribery, kickbacks, favoritism, and

268

The Forgotten History

voting fraud were rampant in these cities. The Progressives wanted to clean up these cities and blunt the power of the party bosses. Their urban reform program included such things as the nonpartisan ballot and replacing elected mayors with appointed city managers. Some Progressives also added proportional representation to this reform agenda. They argued that winner-take-all, single-member-district elections served to reinforce the power of urban political machines. It was not unusual for machines to win almost all the seats on city councils, based on only 50–60 percent of the vote. PR was seen was a way to break these one-party monopolies and to allow for the fair representation of a variety of political parties. The Proportional Representation League of the United States was also instrumental in promoting the use of PR. Founded in 1893, the league soon followed the lead of English electoral reform groups and endorsed the single transferable vote as the most preferable version of PR. In 1912 the league decided that its most realistic goal would be to promote the adoption of PR on the city level. Proportional representation received another important boost in 1914 when the National Municipal League, a leading proponent of urban reform, included PR elections in its model city charter. Soon afterward, in 1915, Ashtabula, Ohio became the first American city to adopt PR elections. Before long, Boulder, Kalamazoo, Sacramento, and West Hartford followed suit. In the mid-1920s, the first large urban areas, Cleveland and Cincinnati, adopted PR elections, and two other Ohio cities, Toledo and Hamilton, soon joined them. The greatest victory of PR advocates came in 1936 when voters in New York City approved the adoption of PR elections by a large margin. Interest in PR jumped dramatically as a result, with it eventually being adopted in eleven other cities, including seven in Massachusetts. In all, twodozen American cities joined the PR camp.

Effects on Representation What political effects did proportional representation have on the cities that adopted it? In particular, did PR fulfill the political promises of its proponents to reduce corruption, ensure fair representation, and increase voter participation? Or did it confirm the fears of PR critics who predicted confused voters, lower turnout, and increased political divisiveness? Scholars have begun to shed some light on these questions. The most extensive research to date has been produced by Kathleen Barber and several

The Forgotten History

269

colleagues. Their study, Proportional Representation and Electoral Reform in Ohio, systematically analyzed the political effects of PR in five Ohio cities. In many cases their findings were also confirmed by results in other PR cities. For example, Barber found that STV produced fairer and more proportional representation of political parties. In particular, it eliminated the tendency of winner-take-all systems to exaggerate the seats given to the largest party and to underrepresent the smaller parties. In the election before the adoption of PR in Cincinnati, the Republicans won only 55 percent of the vote, but received 97 percent of the seats on the council. In the first PR election, the results were much more proportional, with the Republicans winning 33.3 percent of the seats based on 27.8 percent of the vote, and the rival Charter party winning 66.7 percent of the seats on 63.8 percent of the vote. Similarly, in the last pre-PR election in New York City, the Democrats won 95.3 percent of the seats on the Board of Alderman with only 66.5 percent of the vote. During the use of PR, the Democrats still had a majority of the seats, but it was a much smaller one that reflected more accurately their strength in the electorate. In 1941, proportional representation gave the Democrats 65.5 percent of the seats on 64 percent of the vote. Moreover, it also produced representation for the Republicans and three smaller parties in proportion to their voting strength. Similar results occurred in the other PR cities, demonstrating that this system greatly improved the accuracy of partisan representation. Proportional representation also encouraged fairer racial and ethnic representation. It produced the first Irish Catholics elected in Ashtabula, and the first Polish Americans elected in Toledo. In Cincinnati, Hamilton, and Toledo, African Americans had never been able to win city office until the coming of PR. Significantly, after these cities abandoned PR, African Americans again found it almost impossible to get elected.

Effects on Political Machines At times, proportional representation helped undermine the power of political machines and party bosses. In several cities, such as Cincinnati, the machines lost their majorities and their grip on power. After the transition to PR, Cincinnati went from a city with one of the worst reputations for corruption to one that won praise for the integrity and professionalism of its city government. Interestingly, even in cities where the dominant party retained its ma-

270

The Forgotten History

jority, PR sometimes helped to curb the power of the party bosses. It did so by allowing the election of independent Democratic and Republican candidates—candidates nominated by petition and not beholden to party bosses. PR proponents were correct, then, in predicting that this candidate-centered system would take power away from party leaders and give more of it to voters.

Effects on Wasted Votes Proponents of proportional representation also believed it would minimize wasted votes. They argued that the ballot transfer process would ensure that most people would cast effective votes—votes that actually elected someone to office. The evidence supports this claim. In Cincinnati, the number of effective votes improved dramatically, rising from an average of 56.2 percent in the three pre-PR elections to an average of 90 percent for the 16 PR elections. Similar effects were found in other PR cities. In Cleveland, the number of effective votes increased from an average of 48.3 percent in the pre-PR period to an average of 79.6 percent during the PR period. And in New York City, the number of effective votes grew from an average of 60.6 percent to 79.2 percent with PR.

Effects on the Party System How did proportional representation affect the size of the party systems in these cities? Did it subvert the traditional American two-party system, as some critics feared it would? Not always. In some cities, PR produced a stable two-party system. In Cincinnati, the PR elections were contested between the Republicans and the Charter Committee, with no minor-party candidates winning representation. Indeed, only once in all of the PR elections in the five cities in Ohio did a minor-party candidate win office—a Socialist in Ashtabula in 1915. The situation was different in New York City—an intensely cosmopolitan area with a variety of political cultures. PR nurtured a vigorous multiparty system, where at any one time the Democrats and Republicans were joined on the city council by three smaller parties, including the American Labor party, the Fusion party, and the Communist party. In general, however, PR did not seem to automatically favor a multiparty system over a two-party sys-

The Forgotten History

271

tem, but instead it tended to produce a party system that reflected whatever degree of political diversity already existed in particular cities.

Effects on Voter Turnout Proponents of proportional representation had predicted higher voter participation, reasoning that having fewer wasted votes and more choices at the polls would give citizens more incentive to vote. Opponents had forecast a drop in turnout, with voters discouraged by complicated ballots and incomprehensible vote-counting procedures. In reality, however, PR seemed to have little effect on voter turnout. Barber and her colleagues looked at turnout rates before, during, and after the use of PR in five Ohio cities and found little correlation between voting system and the degree of voter participation. She concluded that “the emergence and disappearance of local issues and candidates appear to have had more to do with the act of voting than did the form of the ballot.”2 The scattered evidence from other PR cities seems to confirm the importance of local factors overriding the effects of voting systems, with some cities seeing increased turnout with the adoption of PR and others seeing a decline.

Effects on Political Stability Another common concern of PR critics was that it would increase political conflict and divisiveness. They worried that it would encourage so-called “bloc voting” along ethnic, racial, religious, and class lines, and that the resulting city councils would be paralyzed by conflict. In practice, PR often did result in substantial bloc voting. But as defenders observed at the time, so too did winner-take-all elections. As noted earlier, PR also tended to produce city councils that were more demographically and politically diverse. But there is no evidence that this increased political pluralism had any detrimental impact on the workings of these city councils. In the five Ohio PR cities, Barber and her colleagues found “no systematic evidence of greater dissension on PR elected councils, compared the councils elected by other means . . . . Indeed, striking decreases in conflict were found after PR/STV was implemented in Hamilton and Toledo.”3 This lack of increased conflict may have resulted from the ballot transfer process in choice voting, which

272

The Forgotten History

may have encouraged politicians to be more civil to each other so as not to risk alienating potential supporters.

The Overall Effects of PR On the whole, from the available evidence, proportional representation seemed to have had a beneficial effect on the cities that adopted it. It clearly produced more representative government and, where voters wanted it, a more diverse party system. Large increases in the number of effective votes were also enjoyed in these cities. It may not have resulted in the substantial increases in voter turnout that proponents predicted, but neither did it produce the increases in voter alienation that critics feared. And finally, even though PR city councils were often more diverse politically, this did not seem to impair their political efficiency or effectiveness.

The Abandonment of PR If proportional representation amassed such a generally favorable record, why was it eventually rejected by all but one U.S. city, Cambridge, Massachusetts? The answer to this question is complex, with a number of factors playing a role in the abandonment of PR. Sometimes the reasons were primarily local. In a few cities dissatisfaction grew over other elements of the reform charters, such as the city manager, and when the reform charter was thrown out, PR went with it. However, there were several common factors at work in many of the cities that abandoned proportional representation. For instance, this system universally came under attack from the politicians and parties who lost power and privileges. In Michigan and California, the dominant political parties mounted legal challenges and the courts in these states ruled that PR violated their constitutions. A more common attack was the effort to repeal PR by popular referendum. The referendum was a two-edged sword for PR— initially making it easier to adopt this reform, but also making it easier for opponents to challenge it. In Cleveland, well-financed opponents sponsored five repeal referendums in the first ten years of PR, with the final one succeeding. Similarly, PR opponents in Hamilton finally won their repeal effort after four failed referendums in 12 years.

The Forgotten History

273

Another common factor contributing to the demise of proportional representation was the inability of supporters to defend it effectively. By 1932, the PR League was losing steam. It was unable to finance its separate existence and had to merge with the National Municipal League. In some cities, the progressive political coalition that supported PR gradually disintegrated. Important reform leaders lost interest over the years, moved to the suburbs, or died. Two exceptions to this trend were Cincinnati and Cambridge, both of which had active and well-supported organizations dedicated to defending PR. In Cincinnati the Charter Committee aggressively defended proportional representation and it survived there for more than thirty years, despite repeated challenges. The Cambridge Civic Association has also proved to be a energetic and capable defender of PR and has defeated every repeal effort to date. Another factor working against defenders of proportional representation in many cities was the controversial nature of minority representation. Many Americans in the early twentieth century were hostile to political and racial minorities—the very groups aided by PR. Opponents of PR were not above fanning the flames of prejudice in their efforts to get rid of this reform. In particular, critics often played upon two of the most basic fears of white, middle-class Americans: communists and African Americans. In Cincinnati, race was the dominant theme in the successful 1957 repeal effort. The single transferable vote had allowed African Americans to be elected for the first time, with two blacks being elected to the city council in the 1950s. The nation was also seeing the first stirrings of the Civil Rights movement and racial tensions were running high. PR opponents shrewdly decided to make race an explicit factor in their repeal campaign. They warned whites that PR was helping to increase black power in the city and asked them whether they wanted a “Negro mayor.” Their appeal to white anxieties succeeded, with whites supporting repeal by a two to one margin. In New York City, fear of communism proved the undoing of proportional representation. Although one or two Communists had served on the PR-elected city council since 1941, it was not until the coming of the Cold War that Democratic party leaders were able to effectively exploit this issue. As historian Robert Kolesar discovered, the Democrats made every effort in their repeal campaign to link PR with Soviet Communism, describing the single transferable vote as “the political importation from the Kremlin,” “the first beachhead of Communist infiltration in this country,” and “an unAmerican practice which has helped the cause of communism and does not

274

The Forgotten History

belong in the American way of life.”4 This “red scare” campaign resulted in the repeal of PR by an overwhelming margin. Just as the adoption of the single transferable vote in New York City prompted other cities to consider this reform, its well-publicized defeat there also encouraged repeal efforts in other PR cities. PR was abandoned in neighboring Long Beach and Yonkers in 1947 and 1948. Repeal campaigns also won in Boulder (1947), Toledo (1949), and Wheeling (1951). The PR movement never recovered from these defeats; and although supporters remained optimistic, the 1950s saw the repeal of PR in one city after another. By 1962, only Cambridge, Massachusetts retained this system. While the repeal of proportional representation in these American cities is taken by opponents as evidence that this electoral system failed, it is more accurate to conclude that this system was rejected because it worked too well. PR worked well in throwing party bosses out of government—bosses who never relented in their attempts to regain power—and it worked well in promoting the representation of racial, ethnic, and ideological minorities that were previously shut out by the winner-take-all system. Thus it was the very political successes of this system that set the stage for a political backlash that was effectively exploited by its opponents and eventually led to the its demise in most of these cities.

Notes

Preface 1. Hans-Dieter Klingemann, “Mapping Political Support in the 1990s: A Global Analysis,” in Pippa Norris, ed., Critical Citizens: Global Support for Democratic Government (Oxford: Oxford University Press, 1999). 2. G. Bingham Powell, Elections as Instruments of Democracy: Majoritarian and Proportional Visions (New Haven: Yale University Press, 2000).

Introduction 1. Mary Georghiou and Martin Linton, Arguments for Electoral Reform (Guildford, U.K: Labour Campaign for Electoral Reform, 1991), Photocopy, p.4.

2. Royal Commission on the Electoral System, Report of the Royal Commission

3.

4. 5. 6. 7.

on the Electoral System: Towards a Better Democracy (Wellington, New Zealand: V. R. Ward, Government Printer, 1986). The Report of the Independent Commission on the Voting System (“The Jenkins Report”), http://www.official-documents.co.uk/document/cm40/4090/contents .htm (September 12, 2001). Henry Milner (ed.), Making Every Vote Count: Reassessing Canada’s Electoral System (Peterborough, Ontario: Broadview Press, 1999), p. 47. See, for example, Milner, (ed.) Making Every Vote Count. In the 107th Congress, those bills were HR 1173 and HR 1190. Hendrik Hertzberg, “Taxation Without Representation,” The Irwin Mann Memorial Lecture on Proportional Representation, New York University, April 15, 1997.

276

Introduction

8. Arend Lijphart, Patterns of Democracy (New Haven: Yale University Press, 1999).

9. Lani Guinier, “Forward,” in Robert Richie and Steven Hill, Reflecting All of Us: A Case for Proportional Representation (Boston: Beacon Press, 1999), p. xiii.

10. The National Elections Studies (1995–1998), The NES Guide to Public Opinion and Electoral Behavior, http://www.umich.edu/~nes/mesguide/nesguide. htm (August 1, 2001). 11. Milner, p. 16. 12. See for example, Andrew Reynolds, and Ben Reilly, The International IDEA Handbook of Electoral Design (Stockholm: International Institute for Democracy and Electoral Assistance, 1997); and Douglas J. Amy, Behind the Ballot Box: A Citizen’s Guide to Voting Systems (Westport, CT: Praeger Publishing, 2000). 13. The order of the list can be determined in several ways, such as by vote at a party convention or in something like a party primary.

Chapter 1 1. Robert Newland, Only Half a Democracy (London: The Electoral Reform Society of Great Britain and Ireland, 1982), p. 7.

2. Quoted in Robert Richie and Steven Hill, Reflecting All of Us (Boston: Beacon Press, 1999), p. 3.

3. Wasted votes in PR systems are primarily made up of votes that are cast for very

4.

5. 6.

7.

8. 9.

small parties that do not win enough votes to surpass the minimum threshold needed to win any seats in the legislature. Roughly speaking the larger the districts, the fewer wasted votes. Thus where the entire country is one district, such as Israel and the Netherlands, wasted votes are most minimized. Center for Voting and Democracy, Dubious Democracy, http://www.igc .org/cvd/reports/1999/usa.htm, (May 15, 2001). Kathleen Barber, “Proportional Representation and Boss Rule: The Case of Cleveland.” Paper presented at the 1988 annual meeting of the American Political Science Association, Washington, D.C., September 1–4, 1988, pp. 42–43. Leon Weaver, “Two Cheers for Proportional Representation in Cambridge: A Preliminary Report on Research in Progress.” Paper presented at the annual meeting of the American Political Science Association, Denver, September 2–5, 1982, p. 8. Moot, John R. “Proportional Representation Gives All a Voice.” Cincinnati Post, October 9, 1991, op-ed page. Susan Banducci, Todd Donovan, and Jeffrey A. Karp, “Proportional Representation and Attitudes about Politics: Results from New Zealand,” Electoral Studies 18 (1999): 551.

Chapter 1. Fair Representation for All

277

10. Zimmerman, Joseph F., “PR in Ireland: A Firm Tradition,” National Civic Review 72, no. 11 (December, 1983): 600.

11. Hannah Pitkin, The Concept of Representation (Berkeley: University of California Press, 1967), p. 60.

12. Ibid. p. 61. As Pitkin points out, political philosophers have developed other

13.

14. 15.

16. 17.

18.

19. 20. 21.

22. 23. 24. 25.

theories of representation that do not adhere to the portrait standard. For a description of rival theories, see Pitkin’s book. For a description and discussion of those studies, see Arend Lijphart, “The Field of Electoral Systems Research: A Critical Study.” Electoral Studies 4, no. 1 (1985): 11–12. Some argue that the primary system also handicaps third parties because it attracts political dissidents to run for office in the major parties. It was once thought that the “cube law” best expressed the tendency to give more seats to the largest party—that if the votes are divided between parties in a ratio of A:B, the seats will be divided in the ratio A3:B3. Now this ratio is thought to be somewhat less. See R. J. Johnston, “Seats, Votes, Redistricting, and the Allocation of Power in Electoral Systems,” in Lijphart and Grofman, eds., Choosing an Electoral System (New York: Praeger, 1984). Almost all of the countries in Powell’s study are multiparty systems. In a twoparty system, the second largest party is often substantially underrepresented. Center for Voting and Democracy, Seats to Votes Distortions, http:// www.igc.org/cvd/reports/1999/seats_to_votes.htm (May 3, 2001). This figure includes the large distortions that inevitably occur in small states with only one representative. They are in essence one large single-member district—which simply demonstrates the inherent distortions built into these voting systems. Jack Vowles, Electoral System Opinion and the Evolution of MMP: A Report to the Electoral Commission (University of Waikato, New Zealand: New Zealand Election Study, 2000), p. 11. Arend Lijphart, Patterns of Democracy (New Haven: Yale University Press, 1999), p. 166. Matthew Shugart, “The Jenkins Paradox: A Complex System, Yet Only A Timid Step Towards PR,” Representation 36, no. 2, (Summer 1999): 144. Mark Rush, and Richard Engstrom, Fair and Effective Representation? Debating Electoral Reform and Minority Rights (New York: Rowan and Littlefield, 2001), p. 115. Lijphart, Patterns of Democracy, p. 166. Powell, Elections, p. 251. Wisconsin’s fourth district and California’s 36th district. Other electoral systems, besides PR, can also solve this problem. One of them is instant runoff voting, which would ensure that a candidate won with a majority of the vote. However, this voting system has many disadvantages com-

278

26. 27.

28.

29.

30.

31. 32.

Chapter 1. Fair Representation for All

pared to proportional representation. See the discussion of this system in chapter 10. Similar incidents of second-place parties winning a plurality of seats have also occurred several times in Australia and France. Arend Lijphart, “First-Past-the-Post, PR, Michael Pinto-Duschinsky, and the Empirical Evidence,” Representation 36, no. 2 (Summer 1999): 134; and Powell, Elections as Instruments of Democracy, p. 144. Gallup Poll Analysis, Americans Support Proposal to Eliminate Electoral College System, http://www.gallup.com/poll/releases/pr010105.asp (August 6, 2001). Richard M. Scammon, “The Election and the Future of British Electoral Reform” in Howard R. Penniman, ed., Britain at the Polls: The Parliamentary Elections of 1974 (Washington, D.C.: American Enterprise Institute, 1975), 163–76. Royal Commission on the Electoral System. Report of the Royal Commission on the Electoral System: Towards a Better Democracy. Wellington, New Zealand: V. R. Ward, Government Printer, 1986. Powell, Elections, p. 124. Robert Luce, Legislative Principles (Boston: Houghton Mifflin, 1930), p. 393.

Chapter 2 1. Center for Voting and Democracy, “Revealing Quotes about Redistricting,” http://www.igc.org/cvd/op_eds/revealquoteredist.htm (October 2001).

2. Congressional Quarterly, Jigsaw Politics: Shaping the House after the 1990 Census (Washington, D.C.: Congressional Quarterly Press, 1990), p. 17.

3. U.S. Census Bureau, Statistical Abstract of the United States 2000 (Washington, D.C.: U.S. Printing Office, 2001), p. 286.

4. There is another possible explanation for this phenomenon. It may be that

5. 6. 7. 8. 9.

some normally Democratic voters are crossing over to vote for Republican senatorial candidates. But the crossover vote would have to be very large—almost 20 percent—which does not seem particularly likely. Alan Murray, “Redistricting: Gov. Gerry’s Monuments.” Congressional Quarterly Weekly Report, May 9, 1991, p. 1588. Alan Greenblat, “The Mapmaking Mess,” Governing Magazine, January 2001, p. 20. John H. Fund, “Beware the Gerrymander, My Son,” National Review, April 7, 1989, p.34. Center for Voting and Democracy, “Revealing Quotes.” R. G. Ratcliffe, “Delay: District Remapping will Aid GOP Goals,” Houston Chronicle, April 6, 2000, p. 33.

Chapter 2. Solving the Redistricting Problem

279

10. Center for Voting and Democracy, “Revealing Quotes.” 11. Center for Voting and Democracy, “Revealing Quotes.” 12. John Wildermuth, “State Redistricting Pie Will be Hard to Cut,” San Francisco Chronicle, April 4, 2001, p. A15.

13. Center for Voting and Democracy, “Revealing Quotes.” 14. John Wildermuth, “Lawmakers Use Creative License in Redistricting: Oddly

15.

16. 17. 18. 19. 20. 21. 22.

23.

24. 25. 26.

27.

28. 29.

Shaped Congressional Maps Would Largely Benefit Incumbents,” San Francisco Chronicle, September 2, 2001, p. A6. Cited in Samuel Issacharoff, Pamela S. Karlan, and Richard H. Pildes, The Law of Democracy: Legal Structure of the Political Process (Westbury, NY: The Foundation Press, 1998), p. 719. Center for Voting and Democracy, “Dubious Democracy.” http://www.igc.org /cvd/reports/1999/index.html (September 1, 2001). Center for Voting and Democracy, “Dubious Democracy.” Center for Voting and Democracy, “Monopoly Politics, ” http://www.fairvote. org/reports/monopoly/overview.html (September 1, 2001). Center for Voting and Democracy, “Revealing Quotes.” Ballot Access News, Vol. 16, no. 9 (December, 2000): 2. Center for Voting and Democracy, “Revealing Quotes.” David Wells, “Against Affirmative Gerrymandering.” In Representation and Redistricting Issues, edited by Grofman, Lijphart, McKay, Scarrow (Lexington, Mass.: Lexington Books, 1982), pp. 85–86. For a description of the legal battles over partisan gerrymandering, see Bernard Grofman, Toward Fair and Effective Representation: Political Gerrymandering and the Courts (New York: Agathon, 1990). Louis Jacobson and Chris Cillizza, “Taking Redistricting Out of the Hands of Lawmakers,” National Journal, March 10, 2001, p. 728. B. Adams, “A Model State Reapportionment Process.” Harvard Journal on Legislation 14 (1977): 825–904. Peter J. Taylor and Graham Gudgin, “The Myth of Non-Partisan Cartography: A Study of Electoral Biases in the English Boundary Commission’s Redistribution for 1955–1970” Urban Studies 13 (1976): 13–25. R. J. Johnston, “Seats, Votes, Redistricting, and the Allocation of Power in Electoral Systems,” in Lijphart and Grofman, eds., Choosing an Electoral System, (New York: Praeger, 1984), p. 469. Robert G. Dixon, Democratic Representation: Reapportionment in Law and Politics (London: Oxford University Press, 1968), p. 462. Arend Lijphart and Bernard Grofman, eds, Choosing an Electoral System: Issues and Alternatives (New York: Praeger, 1984), p. 7. When multimember districts get very small, as in the 3–4 seat districts used in Ireland, then some gerrymandering does become possible. Also, gerrymandering is possible in the

280

Chapter 2. Solving the Redistricting Problem

single-member districts used as part of the mixed-member form of proportional representation used in Germany and other countries. But it does not create nearly the same kind of problems as it does here in the United States. For while such gerrymandering can create safe districts for particular parties, it cannot distort the representation that parties get in these PR legislatures. This is because the distribution of legislative seats in this system is ultimately determined by the proportion of parties’ votes on the party list portion of the ballot. This ensures fair representation for all parties. 30. Nelson Polsby, “Declaration of Nelson W. Polsby in Badham v. Eu.” PS (Summer 1985): 570.

Chapter 3 1. Bowler, Shaun, Todd Donovan, and David Brockington, Electoral Reform and

2. 3.

4. 5.

6. 7. 8. 9. 10.

11. 12. 13. 14.

Minority Representation: Local Experiments with Alternative Elections (Columbus: Ohio State University Press, forthcoming). Richard A. Joslyn, “Liberal Campaign Rhetoric in 1984,” in Vermeer, ed., Campaigns in the News (New York: Greenwood, 1987), p. 45. Paul Bradford Raymond, “Shaping the News: An Analysis of House Candidates’ Campaign Communications” in Vermeer, ed., Campaigns in the News (New York: Greenwood, 1987), 13–30. Ibid., pp. 15–16. See for example, Benjamin I. Page, Choices and Echoes in Presidential Elections: Rational Man and Electoral Democracy (Chicago: University of Chicago Press, 1978), chapter 6. Ibid., p. 153. Joslyn, p. 47. Martin P. Wattenberg, The Rise of Candidate-Centered Politics (Cambridge: Harvard University Press, 1991), chapter 5. Bill Press, “Bush Breaks Promise to Protect the Environment,” Eastside Journal, Monday, April 2, 2001, p. A6. Simmons Jr., Robert 0. “Why Some Constituencies Are Better Informed Than Most about Positions of House Incumbents” in Vermeer, Campaigns in the News (New York: Greenwood, 1987), p. 51. Quoted in In These Times, March 1–14, 1989, p.13. Michael Kelly, “Though Advisors Differ, Clinton’s in Tune with All,” New York Times, September 13, 1992, p. Al. Quoted in Larry J. Sabato, The Rise of Political Consultants (New York: Basic Books, 1981), p. 144. Harold Chase, Robert Holt, and John Turner, American Government in Comparative Perspective (New York: Franklin Watts, 1980), p.126.

Chapter 3. Improving Election Campaigns

281

15. As Carey and Shugart have shown, several factors can affect the extent to which

16. 17. 18. 19. 20. 21.

22. 23.

24.

25. 26. 27. 28. 29.

30.

candidates encourage a personal vote or a party vote, such as whether a country has a parliamentary system, or whether the party controls the choice of candidates. But clearly electoral systems play a key role determining the incentives for candidates to cultivate a personal vote. John M. Carey and Matthew Soberg Shugart, “Incentives to Cultivate a Personal Vote: A Rank Ordering of Electoral Formulas,” Electoral Studies 14, no. 4: 417–439. Karen Siune, “The Political Role of the Mass Media in Scandinavia” Legislative Studies Quarterly 12, no. 3 (August 1987): 404. Paul Bradford Raymond, “Shaping the News,” p. 24. Richard S Katz, A Theory of Parties and Electoral Systems (Baltimore: The Johns Hopkins University Press, 1980), p.27. John B. Judis, “With Dukakis a Lock, Dems Eye Platform.” In These Times, June 22–July 5, 1988, p.3. Douglas Harbrecht, and Paula Dwyer. “Congress: It Doesn’t Work. Let’s Fix It.” Business Week, April 16, 1990, p. 56. Arend Lijphart, “Testimony before the California State Legislature.” California State Senate Committee on Constitutional Amendments, Sacramento, CA. March 8, 1995. Brad Rourke, Final Report of the Project on Campaign Conduct 1998 (Camden, ME: Institute for Global Ethics, 1999). Stephen Ansolabehere and Shanto Iyengar. Going Negative: How Political Advertising Shrinks and Polarizes the Electorate (New York: Free Press, 1996). p. 9. Kim Fridkin Kahn and Patrick J. Kenney, “Do Negative Campaigns Mobilize or Suppress Turnout? Clarifying the Relationship between Negativity and Participation,” American Political Science Review 93 no. 4 (December 1999): 887. Not all scholars agree on these effects. See for example the article by Wattenberg and Brians in that same issue of the APSR. Ansolabehere and Iyengar, p. 9. Quoted in Roger Lowe, “Negative Campaigns Harm System,” The Columbus Dispatch, Sunday, June 11, 2000. Lani Guinier, Lift Every Voice (New York: Simon and Schuster, 1998), pp. 254–55. George H. Hallett, Proportional Representation: The Key to Democracy (New York: National Municipal League, 1940), pp. 72–73. Leon Weaver, “The Rise, Decline, and Resurrection of Proportional Representation in Local Government in the United States,” in Grofman and Lijphart, eds., Electoral Laws and Their Political Consequences, p. 150 (New York: Agathon, 1986). Hallett, Proportional Representation, p. 73.

282

Chapter 3. Improving Election Campaigns

31. Lani Guinier in Robert Richie and Steven Hill, Reflecting All of Us (Boston: Beacon Books, 2000), p. xiii.

Chapter 4 1. U.S. Census Bureau, Statistical Abstract of the United States 2000 (Washington, DC: U.S. Printing Office, 2000), p. 289.

2. John Bibby, and Sandy Maisel, Two Parties—or More? (Boulder, Colo: Westview Press, 1998), p. 76.

3. The Vanishing Voter Project, “Many Americans Want Third-Party Choice,

4. 5. 6. 7. 8. 9. 10.

11.

12.

13. 14. 15.

No Matter Who the Nominees Are,” December 21, 1999, http://www .vanishingvoter.org/releases/12–21–99.shtml (September 12, 2001); and Bibby and Maisel, p. 75. Alan Stone, and Richard Barke, Governing the American Republic: Economics, Law, and Politics (New York: St. Martin’s, 1985), p. 121. Theodore Lowi, and Benjamin Ginsberg, American Government: Freedom and Power (New York: Norton, 1990), p. 498. Harold Chase, Robert Holt, and John Turner, American Government in Comparative Perspective (New York: Franklin Watts, 1980), p. 115. Anthony Downs, An Economic Theory of Democracy (New York: Harper, 1957), pp. 114–41. Gary W. Cox, Making Votes Count (Cambridge: Cambridge University Press, 1997), p. 230. Ballot Access News 16, no. 9 (December 5, 2000): 2. David Mayhew, “Congressional Representation: Theory and Practice in Drawing District Lines,” in Nelson Polsby, ed., Reapportionment in the 1970s (Berkeley,: University of California Press, 1971), pp. 277–81. Steven J. Rosenstone, Roy L. Behr, and Edward H. Lazarus, Third Parties in America: Citizen Response to Major Party Failure (Princeton: Princeton University Press, 1984), pp. 33–37. For a discussion of this and other ballot access obstacles to minor parties, see Richard Winger, “Institutional Obstacles to a Multiparty System,” in Paul S. Herrnson and John C. Green, eds., Multiparty Politics in America (New York: Rowan and Littlefield, 1997). Maurice Duverger, Political Parties: Their Organization and Activity in the Modern State (London: Methuen, 1954). See, for example, Stephen L Fisher, “The Wasted Vote Thesis.” Comparative Politics 5 (January 1973): 293–99. Howard L. Reiter, Parties and Elections in Corporate America (New York: St. Martin’s, 1987), pp. 47–48.

Chapter 5. Electing More Women

283

16. Robert Harmel, “The Impact of New Parties on Party Systems: Lessons for

17. 18. 19.

20. 21. 22. 23. 24.

American from European Multiparty Systems, “ in Paul S. Herrnson and John C. Green, eds., Multiparty Politics in America (New York: Rowan and Littlefield, 1997), p. 45. G. Bingham Powell, Elections as Instruments of Democracy: Majoritarian and Proportional Visions (New Haven: Yale University Press, 2000), p. 29. Arend Lijphart, “The Political Consequences of Electoral Laws, 1945–1985,” American Political Science Review 84, no. 2 (June 1990): 488–90. Arthur Miller and Ola Listhaug, “Political Parties and Confidence in Government: A Comparison of Norway, Sweden, and the United States,” British Journal of Political Science 29, 1990, p. 383. Ibid. Michael Lind, “A Radical Plan to Change American Politics,” Atlantic Monthly (August 1992): 77. G. Bingham Powell, Elections as Instruments of Democracy: Majoritarian and Proportional Visions (New Haven: Yale University Press, 2000). Lani Guinier, Lift Every Voice (New York: Simon and Schuster, 1998), p. 254. Hugh Bone, American Politics and the Party System, 3d ed. (New York: McGraw-Hill, 1965), p. 141.

Chapter 5 1. Inter-Parliamentary Union, “Women in National Parliaments,” http://www.ipu .org/wmn-e/classif.htm (August 1, 2001).

2. Quoted in R. Darcy, Susan Welch, and Janet Clark, Women, Elections, and Representation ( New York: Longman, 1987), pp. 18–19.

3. Wilma Rule and Joseph Zimmerman, “Women and Minorities in Parliaments

4. 5.

6. 7. 8. 9. 10.

and Legislatures,” The Encyclopedia of Democracy (Washington, D.C.: Congressional Quarterly Press, 1997). Lisa Grunwald, “If Women Ran America,” Life, June 1992, pp. 36–46. John M. Carey, Richard G. Niemi, and Lynda W. Powell, “Are Women State Legislators Different,” in Sue Thomas and Clyde Wilcox, eds., Women and Elective Office (Oxford: Oxford University Press, 1998), p. 93. Thomas and Wilcox, p. 93. Studies cited in Darcy, Welch, and Clark, p. 154. Sue Thomas, “The Impact of Women on State Legislative Policies, Journal of Politics 53, no. 4 (1991). Andrew Reynolds, “Women in the Legislatures and Executives of the World: Knocking at the Highest Glass Ceiling,” World Politics 51, no. 4 (July 1999): 549. Darcy, Welch, and Clark, pp. 72–77.

284

Chapter 5. Electing More Women

11. Barbara Burrell, “Campaign Finance: Women’s Experience in the Modern

12. 13. 14. 15. 16. 17. 18.

19. 20.

21.

22. 23. 24.

25. 26.

27.

28.

Era,” in Sue Thomas, Clyde Wilcox, and Susan Morrissey, eds., Women and Elective Office (Oxford: Oxford University Press, 1998), p. 27. Burrell, p. 34. Darcy, Welch, and Clark, pp. 77–81. Burrell, pp. 26–29. Elizabeth Adell Cook, “Voter Reaction to Women Candidates,” in Thomas, Wilcox, Morrissey, eds., Women and Elective Office, p. 71. Cook, p. 59. Darcy, Welch, and Clark, pp. 77–81 R. Darcy, “Electoral Barriers to Women,” in Wilma Rule and Joseph Zimmerman, eds., United States Electoral Systems: Their Impact on Women and Minorities (Westport, Conn.: Praeger Publishing, 1992), p. 224. Francis G. Castles, “Female Legislative Representation and the Electoral System,” Politics 1, no. 2 (November 1981): 22. Wilma Rule, “Electoral Systems, Contextual Factors and Women’s Opportunity for Election to Parliament in Twenty-Three Democracies,” Western Political Quarterly 40, no. 3 (September 1987): 487. Richard Matland, “Enhancing Women’s Political Participation: Legislative Recruitment and Electoral Systems,” in Azza Karam, ed., Women in Parliament: Beyond the Numbers (Stockholm: International IDEA, 1998), p. 77. Matland, p. 76. Matland, p. 79. Kathleen Barber, “Proportional Representation and Boss Rule: The Case of Cleveland.” Paper presented at the 1988 annual meeting of the American Political Science Association, Washington, D.C., September 1–4, 1988, pp. 31–32. Rule, “Electoral Systems,” pp. 478–79. While in STV, lower thresholds (the minimum portion of the vote needed to elect a representative from a party) work to the advantage of women, Matland suggests that very low thresholds in party list systems may actually hurt female candidates. A very low threshold will enable a many “mini-parties” to win one or two seats each. More often than not, a party’s main leaders are male, so if only one or two leaders are elected for small parties, they will tend to be males. Matland, p. 81. Some scholars have noted that in societies where there is a substantial bias against female candidates, systems such as open list voting actually hurt their chances of being elected. If parties have more of a commitment to women’s representation than the public, closed list systems may be better. See Reynolds, p. 555–56. Lisa Bhungalia, “NOW Advocates for More Diverse Representation in U.S. Government,” National NOW Times, Winter 2001.

Chapter 6. Fair Representation for Minorities

285

29. Quoted in Lani Guinier, “What We Must Overcome,” The American Prospect vol. 12 no. 5 (March 12, 2001–March 26, 2001).

Chapter 6 1. Milton Morris, “Black Electoral Participation and the Distribution of Public

2. 3. 4. 5.

6.

7. 8.

9.

10. 11. 12. 13.

Benefits,” in Davidson, ed., Minority Vote Dilution (Washington D.C.: Howard University Press, 1984), p. 283. Peter Eisinger, Black Employment in City Government, 1973–1980 (Washington, D.C.: Joint Center for Political Studies, 1983). R. Hayman, “Affirmative Action and Electoral Reform,” Yale Law Review 990 (1981): 1813–14. Bernard Grofman and Chandler Davidson, eds. Controversies in Minority Voting (Washington, D.C.: Brookings Institution, 1992), p. 314. Lawrence Bobo and Franklin D. Gilliam, Jr., “Race, Sociopolitical Participation and Black Empowerment,” American Political Science Review 84, no. 2 (June 1990): 387. Susan A. Banducci, Todd Donovan, Jeffrey A. Karp, “Minority Representation, Empowerment, and Participation in New Zealand and the United States.” Paper delivered at the Annual Conference of the Western Political Science Association, Seattle, WA. March 25–27, 1999. p. 15. National Roster of Black Elected Officials (Washington: Joint Center for Political Studies, 1981). CNN, “Number of Black Elected Officials Shows Modest Increase,” March 22, 2001, http://www.cnn.com/virtual/editions/europe/2000/roof/change.pop/frameset .exclude.html (August 1, 2001). Edward Still, “Alternatives to Single Member Districts,” in Minority Vote Dilution, edited by Davidson (Washington D.C.: Howard University Press, 1984), p. 247. Samuel Hays, “The Politics of Reform in Municipal Government in the Progressive Era,” Pacific Northwest Quarterly 55 (October 1964): 166–82. C. Vann Woodward, The Strange Career of Jim Crow (New York: Oxford University Press, 1966). Chandler Davidson, “Minority Vote Dilution: An Overview,” in Davidson, ed., Minority Vote Dilution (Washington, D.C.: Howard University Press, 1984), p. 11. See for example, Albert Karnig and Susan Welch, “Electoral Structure and Black Representation on City Councils,” Social Science Quarterly 63 (March 1982): 99–114, and Richard Engstrom and Michael D. McDonald, “The Effect of At-Large Versus District Elections on Racial Representation in U.S. Municipalities.” In Grofman and Lijphart, eds., Electoral Laws and Their Political Consequences (New York: Agathon, 1986), pp. 203–25.

286

Chapter 6. Fair Representation for Minorities

14. W. Berry and T. Dye, “The Discriminatory Effects of At-Large Elections, Florida State Law Review 85: 113–21.

15. 42 USC § 1973(a) (1994). 16. Thornburg v. Gingles, 478 U.S. 30 (1986). 17. Frank Parker, “Racial Gerrymandering and Legislative Reapportionment.” In

18.

19. 20. 21. 22. 23.

24. 25. 26.

27.

28. 29. 30.

31. 32.

Davidson, ed., Minority Vote Dilution (Washington, D.C.: Howard University Press, 1984), pp. 85–117, at 85. Seamon v. Upham, 536 F. Supp. 93’ (E.D. Tex. 1982) (three-judge court), rev’d on other grounds sub nom. Upham v. Seamon, 456 U.S. 37, 71 L.Ed.2d 725 (1982), on remand, 536 F. Supp. 1030 (E.D. Tex 1982). Cited in Parker, “Racial Gerrymandering,” p.97. Parker, “Racial Gerrymandering,” p. 108. Shaw v. Reno 509 U.S. 630 (1993). Ibid., at 648. Including Miller v. Johnson 1995, Shaw v. Hunt 1996, and Abrams v. Johnson 1997. David Bositis, Redistricting and Representation: The Creation of MajorityMinority Districts and The Evolving Party System in the South (Washington, D.C.: Joint Center for Political and Economic Studies, 1995), p. 28. Miller v. Johnson, 115 S.Ct. at 2498. Shaw v. Reno, 509 U.S. at 675. David Bositis, “The Future of Majority-Minority Districts and Black and Hispanic Legislative Representation,” in David Bositis, ed., Redistricting and Minority Representation (Washington, D.C.: Joint Center for Political and Economic Studies, 1998), p. 12. Lisa Handley and Bernard Grofman, “The Impact of the Voting Rights Act on Minority Representation: Black Officeholding in Southern State Legislatures and Congressional Delegations” in Chandler Davidson and Bernard Grofman, eds., Quiet Revolution in the South (Princeton, NJ: Princeton University Press, 1994), p. 335. Bernard Grofman, “Shaw v.Reno and the Future of Voting Rights,” PS: Political Science and Politics, (March 1995): 33–34. See, for example, Abigail Thernstrom, Whose Votes Count? (Cambridge: Harvard University Press, 1987). Bernard Grofman, “For Single-Member Districts Random Is Not Equal,” in Grofman, et al., eds., Representation and Redistricting Issues (Lexington, Mass.: Lexington Books, 1982). Arend Lijphart, “Constitutional Choices for New Democracies.” Journal of Democracy 2, no. 1 (Winter 1991): 75. Andrew Reynolds, “Majority and Power-Sharing Government,” in Markus Crepas, et al., eds., Democracy and Institutions: The Life Work of Arend Lijphart (Ann Arbor: University of Michigan Press, 2000), pp. 186, 178.

Chapter 6. Fair Representation for Minorities

287

33. Lijphart, “Constitutional Choices,” p. 81. 34. And even where it is possible to draw such minority districts on the state level,

35.

36. 37.

38.

39. 40. 41.

42. 43. 44. 45. 46.

47.

48. 49.

it can often require the creation of bizarre, many-sided districts that snake through a region in an attempt to capture pockets of minority voters—the very kind of district that the Supreme Court criticized in Shaw v. Reno. Delbert Taebel, “Minority Representation on City Councils: The Impact of Structure on Blacks and Hispanics.” Social Science Quarterly 59, no.1 (June 1978): 151. Hayman, “Affirmative Action and Electoral Reform.” Yale Law Journal 90 (1981): 1811–12. See Edward Still, “Voluntary Constituencies: Modified At-Large Voting as a Remedy for Minority Vote Dilution in Judicial Elections,” Yale Law and Policy Journal (1992). See, for example, Bernard Grofman and Lisa Handley, “Estimating the Impact of Voting-Rights-Related Districting on Democratic Strength in the U.S. House of Representatives,” in Bernard Grofman, ed., Race and Redistricting in the 1990s (New York: Agathon Press, 1998), pp. 51–65. Lani Guinier, Tyranny of the Majority (New York: The Free Press, 1994), pp. 99–100. This is also a property of open party list systems, and mixed-member PR that uses open list voting. For a more extended discussion of how the use of “voluntary constituencies” enhances the practice of democracy, see Thomas Christiano, The Rule of the Many: Fundamental Issues in Democratic Theory (Boulder, Colo.: Westview Press, 1996.) pp. 227–30. The description of this case comes largely from Ralph A. Straetz, PR Politics in Cincinnati (New York: New York University Press, 1958). Straetz, p. 111. Straetz, p. 119. Straetz, p. xi. Material on Chilton County was adapted from Edward Still, “Cumulative and Limited Voting.” Paper delivered at Conference on Representation, Reapportionment, and Minority Empowerment, Pomona College, Claremont, Calif., March 29, 1990. For more on the cumulative vote and other forms of semiproportional representation, see Douglas J. Amy, Behind the Ballot Box: A Citizen’s Guide to Voting Systems (New York: Praeger Publishing, 2000). Richard Pildes and Kristen Donoghue, “Cumulative Voting in the United States,” University of Chicago Legal Forum (November 1995). Shaun Bowler, Todd Donovan, and David Brockington, Electoral Reform and Minority Representation: Local Experiments with Alternative Elections (Columbus: Ohio State University Press, forthcoming).

288

Chapter 6. Fair Representation for Minorities

50. 51. 52. 53. 54. 55.

Bowler, et. al., Electoral Reform. Bowler, et. al., Electoral Reform, p. 1. Bowler, et. al., Electoral Reform. Guinier, Lift Every Voice (New York: Simon and Schuster, 1998), p. 256. National Public Radio, “All Things Considered,” July 14, 1990. Arend Lijphart, “Constitutional Choices for New Democracies,” Journal of Democracy 2, no. 1 (Winter 1991): 75. 56. Guinier, Lift Every Voice, p. 262.

Chapter 7 1. 2. 3. 4.

5. 6. 7. 8. 9. 10. 11. 12. 13.

14. 15.

Quoted in the Washington Spectator, August 1, 1990, p. 1. Hans Elau, “The Politics of Happiness,” Antioch Review 16 (1955). Charles Krauthammer, “In Praise of Low Turnout,” Time, May 21, 1990, p. 88. Jan Leighley and Jonathan Nagler, “Socioeconomic Class Bias in Turnout: Evidence from Aggregate Data.” Paper presented at the 2000 Annual Meeting of the American Political Science Association, Washington, D.C., August 31–September 3, 2000. George Will, “In Defense of Nonvoting,” Newsweek, October 10, 1983, p. 16. G. Bingham Powell, “American Voter Turnout in Comparative Perspective,” American Political Science Review 80, no. 1, (March 1986): 37. William J. Crotty, American Parties in Decline. (Boston: Little, Brown, 1984), p. 23. Crotty, p. 8. Ruy Teixeira, Why Americans Don’t Vote: Turnout Decline in the United States, 1960–1984. (New York: Greenwood, 1987). Arend Lijphart, “Unequal Participation: Democracies Unresolved Dilemma,” American Political Science Review 91, no. 1 (March 1997). Lijphart, “Unequal Participation,” p. 4. Walter Dean Burnham, “The Turnout Problem,” in Reichley, ed. Elections American Style (Washington, D.C.: Brookings Institution, 1987), p. 99. It should be noted that some scholars argue that nonvoters and voters have similar political views and so higher turnout would not translate into different government policies. But other scholars believe that increased turnout could change the political complexion of government in the United States. See Lijphart, “Unequal Participation,” p. 4. Frances Fox Piven and Richard A. Cloward, Why Americans Still Don’t Vote: And Why Politicians Want It That Way (Boston: Beacon Press, 2000). Quoted in Jack Doppelt, and Ellen Shearer, Nonvoters: America’s No Shows (Thousands Oaks: Sage Publications, 1999), p. 12.

Chapter 7. Encouraging Voter Turnout

289

16. Kay Lawson, “The Case for a Multiparty System,” in Paul S. Herrnson and

17. 18.

19. 20. 21. 22. 23. 24.

25. 26.

27.

28. 29. 30. 31.

32.

John C. Green, eds., Multiparty Politics in America (Lanham, Maryland: Rowen and Littlefield, 1997). See, for example, Sidney Verba and Norman Nie, Participation in America: Political Democracy and Social Equality (New York: Harper and Row, 1972). Mark N. Franklin, “Electoral Participation,” in Lawrence LeDuc, Richard Niemi, and Pippa Norris, eds. Comparing Democracies: Elections and Voting in Global Perspective (Thousand Oaks, Calif.: Sage Publishing, 1996), p. 220. Powell, “American Voter Turnout,” pp. 18–20. Cited in Franklin, p. 229. Franklin, pp. 226–27. Lijphart, “Unequal Participation,” p. 7. Andre Blais and R. K. Carty, “Does Proportional Representation Foster Voter Turnout?” European Journal of Political Research 18 (1990): 167–81. Shaun Bowler, Todd Donovan, David Brockington, Electoral Reform and Minority Representation: Local Experiments with Alternative Elections (Columbus: Ohio State University Press, forthcoming). Center for Voting and Democracy, “Revealing Quotes about Redistricting,” http://www.igc.org/cvd/op_eds/revealquoteredist.htm (October 2000). Jeffery A. Karp and Susan A Banducci, “Political Efficacy and Participation in Eighteen Democracies: How Electoral Rules Shape Political Behavior.” Paper presented at the XVIII World Congress of the International Political Science Association, Quebec City, Canada, August 1–5, 2000, p. 17. See Gregory Caldeira and Samuel Patterson, “Contextual Influences on Participation in U.S. Legislative Elections,” Legislative Studies Quarterly 7, no. 3 (August 1982): 359–81; and D. T. Denver and H.T.G. Hands, “Marginality and Turnout in British Elections,” British Journal of Political Science 4, no. 1 (January 1974): 17–35. Ballot Access News 16, no. 9 (December 2000): 2. Hendrik Hertzberg, “Let’s Get Representative,” New Republic, June 28, 1987, pp. 16–17. Powell, “American Voter Turnout,” p. 21. Some scholars disagree with this argument. They suggest that more political parties can create less decisive election outcomes and this uncertainty could depress voter turnout. Jackman, for instance, believes that the multipartism in PR decreases turnout somewhat, although in the end, PR has a positive effect on turnout. See Robert W. Jackman, “Political Institutions and Voter Turnout in the Industrial Democracies,” American Political Science Review 81, no. 2 (June 1987): 408. John F. Zipp, “Perceived Representativeness and Voting: An Assessment of the Impact of ‘Choices’ and ‘Echoes,’ ” American Political Science Review 79 (March 1985): 58, 50.

290

Chapter 7. Encouraging Voter Turnout

33. Jack Vowles, et al., “Electoral System Opinion and the Evolution of MMP: A

34. 35. 36. 37. 38. 39. 40. 41.

42. 43. 44. 45. 46. 47.

Report to the Electoral Commission, (University of Waikato, New Zealand: Zealand Election Study, 2000), p. 37. Cited in Vowles, p. 37. Anthony Downs, An Economic Theory of Democracy (New York: Harper, 1957), pp. 260–76. Gary Cox, Making Votes Count (Cambridge: Cambridge University Press, 1997), p. 230. Karp and Banducci, p. 1. Burnham, “The Turnout Problem,” pp.102–103. Franklin, p. 232. Thomas Byrne Edsall, The New Politics of Inequality (New York: Norton, 1984), p. 182. Kim Quaile Hill and Jan E. Leighley, “The Policy Consequences of Class Bias in State Electorates,” American Journal of Political Science, 36, no.2 (May 1992): 352. Piven and Cloward, p. 11. Piven and Cloward. Downs, Economic Theory, p. 299. Quoted in Lani Guinier, Lift Every Voice, (New York: Simon and Schuster, 1998) p. 253. Edsall, pp. 23–66. John Zipp and Joel Smith, “A Structural Analysis of Class Voting,” Social Forces 60, no. 3 (March 1982): 738–59.

Chapter 8 1. Kathleen L. Barber, The Right to Representation: Proportional Election System for the Twenty-First Century (Columbus: Ohio University Press, 2000), p. xiii.

2. Barber, p. 157. 3. For a full description of all of the characteristics of these two systems, see Arend

4. 5. 6. 7. 8.

Lijphart, Patterns of Democracy (New Haven: Yale University Press, 1999), pp. 3–4. Note that while the U.S. has many attributes of a majoritarian system, it also does have a few that are typical of the consensus model, such as a bicameral national legislature. Lijphart, Patterns of Democracy, p. 303. Lijphart, pp. 1–2. G. Bingham Powell, Elections as Instruments of Democracy: Majoritarian and Proportional Visions (New Haven: Yale University Press, 2000). Powell, p. 6. Powell, p. 34.

Chapter 8. PR: A Different and Better Kind of Democracy

291

9. See G. Bingham Powell, “Westminster Model versus PR: Normative and Em-

10.

11. 12. 13. 14. 15. 16.

17. 18. 19. 20

21. 22.

23. 24. 25. 26. 27. 28.

pirical Assessments,” and Arend Lijphart, “First-Past-the-Post, PR and the Empirical Evidence,” both in Representation 36, no. 2 (Summer 1999): 127–132, 133–136. Jack Vowles, Electoral System Opinion and the Evolution of MMP: A Report to the Electoral Commission (University of Waikato, New Zealand: New Zealand Election Study, 2000) p. 14. Lani Guinier, The Tyranny of the Majority: Fundamental Fairness in Representative Democracy (New York: Free Press, 1994). Guinier, Tyranny, pp. 2–3. Guinier, Tyranny, p. 10. Guinier, Tyranny, p. 6. Lani Guinier, “Second Proms and Second Primaries: The Limits of Majority Rule,” Boston Review, September/October 1992, pp. 32–34. Lauren Bain, “What’s Wrong with Proportional Representation,” The Laissez Faire City Times 3, no. 27 (July 5, 1999): 4. http://zolatimes.com/V3.27 /proportional_rep. Html (August 20, 2001). Lani Guinier, Lift Every Voice (New York: Simon and Schuster, 1998), p. 254 Powell, p. 150. Powell, p. 256, footnote 9. Timothy Smeeding, Lee Rainwater, and Gary Burtless, “United States Poverty in a Cross-National Context: Luxembourg Income Study Working Paper No. 244” (Syracuse, New York: Maxwell School of Citizenship and Public Affairs, Syracuse University, 2000), pp. 15–16. Lijphart, Patterns, pp. 294–95. Woojin Moon, “The Impact of Left-Center Coalitions and Electoral Systems on Welfare Development in 17 Democracies, 1970–1990,” paper presented at the Annual Meetings of the Midwest Political Science Association, Chicago, IL, April 19–22, 2001. See, for example, James Bohman and William Rehg, eds., Deliberative Democracy: Essays on Reason and Politics (Cambridge, MA: MIT Press, 1997). Thomas Christiano, The Rule of the Many: Fundamental Issues in Democratic Theory (Boulder, CO: Westview Press, 1996.) p. 261. Amy Guttman and Dennis Thompson, Democracy and Disagreement (Cambridge, MA: Harvard University Press, 1996) p. 135. Christiano, p. 265. Lijphart, Patterns of Democracy, p. 286. Christopher J. Anderson and Christine A. Guillory, “Political Institutions and Satisfaction with Democracy: A Cross-National Analysis of Consensus and Majoritarian Systems,” American Political Science Review 91, no. 1 (March 1997): 68.

292

Chapter 8. PR: A Different and Better Kind of Democracy

29. Powell, Elections as Instruments of Democracy, p. 160. 30. Powell, p. 18. 31. Powell, p. 254. Chapter 9 1. Peter J. Taylor, “The Case for Proportional Tenure: A Defense of the British

2.

3.

4.

5. 6.

7. 8.

9.

10. 11.

12.

Electoral System,” in Arend Lijphart and Bernard Grofman, eds., Choosing an Electoral System (New York: Praeger, 1984) p. 53. Bernard Grofman, “Nonmajoritarian Election Systems in the United States.” Department of Political Science, University of California at Irvine, 1985. Photocopy, p. 10. Royal Commission on the Electoral System, Report of the Royal Commission on the Electoral System: Towards a Better Democracy (Wellington, New Zealand: V. R. Ward, Government Printer, 1986), p. 56. Leon Weaver, “Two Cheers for Proportional Representation in Cambridge: A Preliminary Report on Research in Progress.” Paper presented at the annual meeting of the American Political Science Association, Denver, September 2–5, 1982, p. 5. Richard Engstrom and Robert Brischetto, “Is Cumulative Voting too Complex? Evidence from Exit Polls,” Stetson Law Review 27 (Winter 1998): 813–34. Mark Rush and Richard Engstrom, Fair and Effective Representation? Debating Electoral Reform and Minority Rights (New York: Rowan and Littlefield, 2001), p. 47. David Farrell, Electoral Systems: A Comparative Introduction (New York: Palgrave, 2001), p. 204. Charles T. Canady, Statement Concerning H.R. 1173, State’s Choice of Voting Systems Act, before the Subcommittee on the Constitution, Committee on the Judiciary, U.S. House of Representatives, Washington, D.C., September 23, 1999, p. 2. John Bibby, “In Defense of the Two Party System,” in Paul S Herrnson and John C. Green, Multiparty Politics in America (New York: Rowan and Littlefield 1997), pp. 79–80. Jack Quinn, Donald J. Simon, and Jonathan B. Sallet, “Redrawing Political Maps: An America of Groups?” Washington Post, March 24, 1991, p. C1. The National Commission on Federal Election Reform, Hearing 1—Panel 4: The Federal Election System: Historical Perspectives, Transcript, March 26, 2001, http://www.reformelections.org/data/transcripts/h2/hearing2_p4.php?d ⫽ April⫹12%2C⫹2001 (September 10, 2001). Matthew Soberg Shugart, “The Jenkins Paradox: A Complex System, Yet Only a Timid Step Towards PR,” Representation 36, no. 2 (Summer 1999): 143.

Chapter 9. Objections to Proportional Representation

293

13. David M. Farrell, Comparing Electoral Systems (New York: Prentice Hall, 1997), p. 155.

14. See, for example, Arend Lijphart, “Constitutional Choices for New Democra-

15.

16. 17. 18. 19. 20. 21.

22. 23. 24. 25. 26. 27.

cies,“ Journal of Democracy 2, no. 1 (Winter 1991): 73; and Matthew Shugart, “Extreme Electoral Systems and the Appeal of the Mixed-Member Alternative,” in Matthew Shugart and Martin Wattenberg eds., Mixed-Member Electoral Systems: The Best of Both Worlds? (New York: Oxford University Press, 2001), pp. 46–48. For example, Leon Weaver’s study of the single transferable vote form of PR as it is used in Cambridge, Massachusetts, found little evidence for the “common assertion of STV opponents that the system leads to constant bickering because of the diversity of views represented.” See Weaver, “Two Cheers for Proportional Representation.” Kathleen L. Barber, The Right to Representation: Proportional Election System for the Twenty-First Century (Columbus: Ohio University Press, 2000), p. 89. Farrell, Electoral Systems, 204–205. Arend Lijphart, “Democracies: Forms, Performance, and Constitutional Engineering,” European Journal of Political Research 25 (1994): 8. S. E. Finer, Adversary Politics and Electoral Reform (London: Anthony Wigram, 1975). Michael Pinto-Duschinsky, “Britain’s Removal Van Democracy,” The Times, September 25, 1997. Michael Pinto-Duschinsky, “Send the Rascals Packing: Defects in Proportional Representation and the Virtues of the Westminster Model” Representation 36, no. 2 (Summer 1999): 119. Arend Lijphart, “First-Past-the-Post, PR, Michael Pinto-Duschinsky, and the Empirical Evidence.” Representation 36, No.2 (Summer 1999): 134. Lijphart, “First-Past-the-Post,” p. 134. G. Bingham Powell, Elections as Instruments of Democracy: Majoritarian and Proportional Visions (New Haven: Yale University Press, 2000), p. 72. Ibid. Arend Lijphart, Patterns of Democracy (New Haven: Yale University Press, 1999), p. 166. Even when SMP elections do produce a single-party government supported by a majority of the voters, the majority has often been a very slim one—only a little more than 50 percent of the vote. These bare-majority parties then rule a near-majority party that often has garnered more than 40 percent of the voters. Such situations can create a great deal of political discontent. In contrast, multiparty systems are much more likely to produce coalition governments that represent a larger majority of the voters. Belgium, Finland, and Switzerland, for example, have routinely been ruled by large coalitions of parties, often represent-

294

28.

29.

30.

31.

32. 33. 34.

35. 36. 37. 38. 39. 40.

41. 42. 43. 44. 45.

Chapter 9. Objections to Proportional Representation

ing almost two-thirds of the electorate. These large coalitions that make up the ruling majorities in many PR countries are much more representative than the bare majorities typical of SMP elections, and the policies they produce are apt to have a much broader base of support among the public. John Bibby, “In Defense of the Two Party System,” in Paul S Herrnson and John C. Green, Multiparty Politics in America (New York: Rowan and Littlefield 1997), p. 77. Abigail Thernstrom, Prepared Statement Concerning H.R. 1173, State’s Choice of Voting Systems Act, before the Subcommittee on the Constitution, Committee on the Judiciary, U.S. House of Representatives, Washington, D.C., September 23, 1999, p. 3. Andrew Busch, Prepared Statement Concerning H.R. 1173, State’s Choice of Voting Systems Act, before the Subcommittee on the Constitution, Committee on the Judiciary, U.S. House of Representatives, Washington, D.C., September 23, 1999, pp. 5–6. Rein Taagepera and Matthew Soberg Shugart, Seats and Votes: The Effects and Determinants of Electoral Systems (New Haven: Yale University Press, 1989), p. 63. New York Times, November 4, 1987. Arend Lijphart, “Constitutional Choices for New Democracies,” Journal of Democracy 2, no. 1 (Winter 1991): 81. Jack Vowles, Electoral System Opinion and the Evolution of MMP: A Report to the Electoral Commission (University of Waikato, New Zealand: New Zealand Election Study, 2000), p. 15. Gottlieb, Martin. “The ‘Golden Age’ of the City Council.” New York Times, August 11, 1991, p. E6. Powell, Elections as an Instrument of Democracy. Royal Commission, Electoral System, p. 49. Ferdinand A. Hermens, “From Tyranny to Democracy: Fragile or Strong?” Freedom at Issue (May-June 1990): 7. Ferdinand A. Hermens,. Democracy or Anarchy? A Study of Proportional Representation (South Bend, Ind.: Notre Dame, 1941). All quotes are from Belle Zeller and Hugh Bone, “The Repeal of PR in New York City: Tens Years in Retrospect,” American Political Science Review 42 (December 1948): 1128, 1129. New York Times, October 27, 1947. Zeller and Bone, p. 1133. Vernon Bogdanor, What Is Proportional Representation? (Oxford: Martin Robertson, 1984), p. 123. Ibid. p. 124. Jack Vowles, Electoral System Opinion, p. 16.

Chapter 10. Versions of Electoral Reform

295

46. Simon W. Gerson, Pete: The Story of Peter V. Cacchione, New York’s First Communist Councilman (New York: International Publishers, 1977), p. 109.

47. Ibid. p. 145. 48. Frank S. Cohen, “Proportional versus Majoritarian Ethnic Conflict Manage-

49.

50.

51. 52.

53. 54.

55.

56.

ment in Democracies,” Comparative Political Studies 30, no. 5 (October 1997): 626. Steven Mulroy, “The Way Out: A Legal Standard for Imposing Alternative Electoral Systems as Voting Rights Remedies,” Harvard Civil Rights/Civil Liberties Law Review 33 (Summer, 1998): 354, note 110. Quoted in Samuel Issacharoff, Pamela S. Karlan, and Richard H. Pildes, The Law of Democracy: Legal Structure of the Political Process (Westbury, NY: The Foundation Press, 1998), pp. 737–38. Angus Maude. Why Electoral Change? (London: Conservative Political Centre, 1982), p. 27. U.S. Congress. Senate. Committee on Government Operations. Sub-committee on Intergovernmental Relations. Confidence and Concern: Citizens’ Views of American Government (Washington, D.C.: U.S. Government Printing Office, 1973), pp. 72–77. Vernon Bogdanor, What Is Proportional Representation? (Oxford: Martin Robertson, 1984), pp. 104–105. Moreover, as Dennis Anderson has pointed out, studies have shown that members of groups that are underrepresented in legislatures are less likely to write to officials about their concerns. Dennis Anderson, “Alternative Electoral Systems: An Answer to our Governing Crisis,” in Paul E. Scheele, ed., ”We Get What We Vote For . . . Or Do We?” (Westport, Conn.: Praeger, 2000), p. 262. More importantly, these PR systems allow voters to choose whether they want to be represented geographically or on some other basis. A Libertarian, for instance, might decide that it is more important to be represented by a Libertarian on the city council, than by someone from his or her neighborhood. With PR, voters are able to choose to be represented whatever way they think is appropriate, be that geographically or otherwise. Rush and Engstrom, p. 58.

Chapter 10 1. There are in fact many different possible voting alternatives to SMP, including such things as approval voting, condorcet voting, etc. I will be addressing here only the alternatives being most commonly discussed today. For more information on the workings of some of the lesser-used systems see Douglas J. Amy, Behind the Ballot Box: A Citizen’s Guide to Voting Systems (Westport, Conn.: Praeger Publishing, 2000), appendix C.

296

Chapter 10. Versions of Electoral Reform

2. See, for example, the discussion of instant runoff voting in Ted Halstead and

3. 4. 5.

6.

7.

8. 9. 10. 11. 12.

13.

Michael Lind, The Radical Center: The Future of American Politics (New York: Doubleday Press, 2001), pp. 112–13. Wilma Rule, “Letter to the Editor,” Voice for Democracy: Newsletter of Californians for Proportional Representation (May–June 2000). Caleb Kleppner, “Letter to the Editor,” Voice for Democracy: Newsletter of Californians for Proportional Representation (September–October 2000). IRV supporters sometimes cite Australia as an example of this system serving as a stepping-stone to PR. In the early twentieth century, the Australian Senate first adopted IRV for its elections and eventually moved on to adopt the single transferable vote. But it took 30 years for PR to be finally adopted, hardly a good example of IRV facilitating the transition to PR. In addition, the Australian lower house also adopted IRV, and they have never changed to PR. See Institute of Government and Public Affairs, Illinois Assembly on Political Representation and Alternative Electoral Systems: Final Report and Background Papers (Urbana: Institute of Government and Public Affairs, University of Illinois, 2001). For more information on cumulative voting, see Douglas J. Amy, Behind the Ballot Box: A Citizen’s Guide to Voting Systems (Westport, CT: Praeger Publishing, 2000), pp. 115–125. For more information on limited voting, see Douglas J. Amy, Behind the Ballot Box, pp. 125–31. Richard Pildes and Kristen Donoghue. “Cumulative Voting in the United States.” University of Chicago Legal Forum (November 1995). See Andrew McLaren Carstairs, A Short History of Electoral Systems in Western Europe (London: George Allen and Unwin, 1980) p 33. Arend Lijphart and Bernard Grofman, eds, Choosing an Electoral System: Issues and Alternatives (New York: Praeger, 1984), p. 7. For a more detailed treatment of the relationship between district magnitude and number of legislative parties, see Rein Taagepera and Matthew Soberg Shugart, Seats and Votes: The Effects and Determinants of Electoral Systems (New Haven: Yale University Press, 1989), chapter 11. Lawrence D. Longley and Alan Braun, The Politics of Electoral College Reform (New Haven: Yale University Press, 1975), pp. 42–43.

Chapter 11 1. Institute of Government and Public Affairs, Illinois Assembly on Political Representation and Alternative Electoral Systems: Final Report and Background Papers (Urbana: Institute of Government and Public Affairs, University of Illinois, 2001).

Chapter 11.The Political Prospects for PR

297

2. Kathleen L. Barber, The Right to Representation: Proportional Election System

3.

4. 5. 6.

7. 8.

9. 10.

11. 12. 13. 14. 15. 16.

17. 18. 19.

20.

for the Twenty-First Century (Columbus: Ohio State University Press, 2000), p. 128. Janice C. May, “State Constitutions and Constitutional Revision: 1988–89 and the 1980s,” in Book of the States, Vol. 28 (Lexington, Ky: Council of State Governments, 1990–91), pp. 42, 44. May, “State Constitutions and Constitutional Revision,” p. 22. Robert Richie and Steven Hill, Reflecting All of Us: A Case for Proportional Representation (Boston: Beacon Press, 1999), p., 70. Richard Winger, “Institutional Obstacles to a Multiparty System” in Paul S. Herrnson and John C. Green, eds., Multiparty Politics in America (New York: Rowan and Littlefield, 1997). Ruth K. Scott and Ronald J. Hrebenar, Parties in Crisis (New York: Wiley, 1984), pp. 84–85. Belle Zeller and Hugh Bone, “The Repeal of PR in New York City: Tens Years in Retrospect,” American Political Science Review 42 (December 1948): 1128–29, 1132. Bernard Grofman, “Alternatives to Single-Member Plurality Districts: Legal and Empirical Issues,” Policy Studies Journal, special issue 3 (1980–81): p. 889. Steven J. Mulroy, “The Way Out: A Legal Standard for Imposing Alternative Electoral Systems as Voting Rights Remedies,” Harvard Civil Rights/Civil Liberties Law Review 33 (Summer, 1998): 380. John R. Low-Beer, “The Constitutional Imperative for Proportional Representation” Yale Law Journal 94, no. I (November 1984): 163–88. Low-Beer, p. 181. Low-Beer, p. 188. Timmons v Twin Cities Area New Party. 1996. 137 L. Ed. 2nd 603. Timmons, p. 611. Matthew Soberg Shugart and Martin P. Wattenberg, “Conclusion: Are MixedMember Systems the Best of Both Worlds?” in Shugart and Wattenberg, eds., Mixed-Member Electoral Systems: The Best of Both Worlds? (Oxford: Oxford University Press, 2001), pp. 572–78. Arend Lijphart, Patterns of Democracy (New Haven: Yale University Press, 1999) p. 164. Laura Myers, “Most Americans Distrust Government, Poll Finds,” The Detroit News, August 1, 1995. The National Elections Studies (1995–1998), The NES Guide to Public Opinion and Electoral Behavior, http://www.umich.edu/~nes/mesguide/nesguide. htm (August 1, 2001). The Vanishing Voter Project, Many Americans Want Third-Party Choice, No Matter Who the Nominees Are, December 21, 1999, http://www.vanishingvoter

298

21.

22. 23.

24.

Chapter 11.The Political Prospects for PR

.org/releases/12–21–99.shtml (September 12, 2001); and John Bibby and Sandy Maisel (1998) Two Parties—or More? Boulder, CO: Westview Press, p. 76. Maurice Klain, “A New Look at the Constituencies: The Need for a Recount and Reappraisal,” American Political Science Review (December 1955): 1105–19. Alexander Keyssar, The Right to Vote: The Contested History of Democracy in the United States (New York: Basic Books, 2000). Interestingly, the semiproportional system of choice in these countries has been parallel voting, a variation of the mixed-member proportional representation. For a description of parallel voting, see Douglas J. Amy, Behind the Ballot Box (Westport, Conn.: Praeger Publishing, 2000) pp. 132–37. The Jenkins Report can be found one-line at http://www.electoral-reform .org.uk/sep/publications/briefings/avplus.htm.

Appendix C 1. An earlier version of Appendix C was previously published in Representation: The Journal of Representative Democracy 34, no. 1 (Winter 1996/97). It is reprinted here with the kind permission of the McDougall Trust, a United Kingdom charity that specializes in electoral studies. 2. Kathleen L. Barber, Proportional Representation and Election Reform in Ohio (Columbus: Ohio State University Press, 1995), p. 295. 3. Barber, Proportional Representation, p. 305. 4. Robert J. Kolesar, “Communism, Race, and the Defeat of Proportional Representation in Cold War America” (University Heights, Ohio: History Department, John Carroll University, 1996), pp. 4–5.

Selected Bibliography

Amy, Douglas J. Behind the Ballot Box: A Citizen’s Guide to Voting Systems. Westport, Conn.: Praeger Publishing, 2000. Anderson, Christopher J. and Christine A. Guillory. “Political Institutions and Satisfaction with Democracy: A Cross-National Analysis of Consensus and Majoritarian Systems,” American Political Science Review 91, no. 1 (March 1997): 66–81. Anderson, Dennis. “Alternative Electoral Systems: An Answer to our Governing Crisis.” In Paul E. Scheele, ed., ”We Get What We Vote For . . . Or Do We?” Westport, Conn.: Praeger, 2000. Bain, Lauren. “What’s Wrong with Proportional Representation.” The Laissez Faire City Times, 3, no. 27 (July 5, 1999). http://zolatimes.com/V3.27/proportional _rep.html (October 1, 2001). Banducci, Susan, Todd Donovan, and Jeffrey A. Karp. “Proportional Representation and Attitudes about Politics: Results from New Zealand.” Electoral Studies 18 (1999): 533–55. Barber, Kathleen L. The Right to Representation: Proportional Election System for the Twenty-First Century. Columbus: Ohio State University Press, 2000. Blais, Andre and R. K. Carty. “Does Proportional Representation Foster Voter Turnout?” European Journal of Political Research 18, (1990): 167–81. Bogdanor, Vernon . What Is Proportional Representation? Oxford: Martin Robertson, 1984. Bowler, Shaun, Todd Donovan, and David Brockington, Electoral Reform and Minority Representation: Local Experiments with Alternative Elections. Columbus: Ohio State University Press, forthcoming. Carstairs, Andrew McLaren. A Short History of Electoral Systems in Western Europe. London: George Allen and Unwin, 1980.

300

Selected Bibliography

Castles, Francis G. “Female Legislative Representation and the Electoral System.” Politics 1, no. 2 (November 1981). Center for Voting and Democracy. Seats to Votes Distortions. http://www.igc.org/cvd/ reports/1999/seats_to_votes.htm (May 3, 2001). ———. Dubious Democracy. http://www.igc.org/cvd/reports/1999/index.html (September1, 2001). ———. Monopoly Politics. http://www.fairvote.org/reports/monopoly/overview.html (September1, 2001). Christiano, Thomas. The Rule of the Many: Fundamental Issues in Democratic Theory. Boulder, Colo.: Westview Press, 1996. Cohen, Frank S. “Proportional versus Majoritarian Ethnic Conflict Management in Democracies.” Comparative Political Studies 30, No. 5 (October 1997): 626. Engstrom, Richard and Robert Brischetto. “Is Cumulative Voting too Complex? Evidence from Exit Polls.” Stetson Law Review 27 (Winter 1998): 813–34. Farrell, David M. Comparing Electoral Systems. New York: Prentice Hall, 1997. Farrell, David. Electoral Systems: A Comparative Introduction. New York: Palgrave, 2001. Franklin, Mark N. “Electoral Participation.” In Lawrence LeDuc, Richard Niemi, and Pippa Norris, eds., Comparing Democracies: Elections and Voting in Global Perspective. Thousand Oaks, Calif.: Sage Publishing, 1996. Gerson, Simon W. Pete: The Story of Peter V Cacchione, New York’s First Communist Councilman. New York: International Publishers, 1977. Gottlieb, Martin. “The ‘Golden Age’ of the City Council.” New York Times, August 11, 1999, E6. Guinier, Lani. “Second Proms and Second Primaries: The Limits of Majority Rule.” Boston Review (September/October 1992): 32–34. ———. The Tyranny of the Majority: Fundamental Fairness in Representative Democracy. New York: Free Press, 1994. ———. Lift Every Voice. New York: Simon and Schuster, 1998. ———. “What We Must Overcome.” The American Prospect 12, no. 5 (March 12, 2001–March 26, 2001). Hallett, George H. Proportional Representation: The Key to Democracy. New York: National Municipal League, 1940. Hill, Steven. Fixing Elections: The Failure of America’s Winner-Take-All Politics. New York: Routledge Press, 2002. Institute of Government and Public Affairs. Illinois Assembly on Political Representation and Alternative Electoral Systems: Final Report and Background Papers. Urbana: Institute of Government and Public Affairs, University of Illinois, 2001. Inter-Parliamentary Union. Women in National Parliaments. http://www.ipu.org /wmn-e/classif.htm (August, 2001).

Selected Bibliography

301

Issacharoff, Samuel, Pamela S. Karlan, and Richard H. Pildes. The Law of Democracy: Legal Structure of the Political Process. Westbury, NY: The Foundation Press, 1998. Katz, Richard S. A Theory of Parties and Electoral Systems. Baltimore: The Johns Hopkins University Press, 1980. Keyssar, Alexander. The Right to Vote: The Contested History of Democracy in the United States. New York: Basic Books, 2000. Lawson, Kay. “The Case for a Multiparty System.” In Paul S. Herrnson and John C. Green, eds., Multiparty Politics in America. Lanham, Maryland: Rowen and Littlefield, 1997. Lijphart, Arend. “Constitutional Choices for New Democracies.” Journal of Democracy 2, no. 1 (Winter1991): 72–84. ———. “Democracies: Forms, Performance, and Constitutional Engineering.” European Journal of Political Research 25 (1994). ———. “First-Past-the-Post, PR, Michael Pinto-Duschinsky, and the Empirical Evidence.” Representation 36, No.2 (Summer 1999): 133–36. ———. Patterns of Democracy. New Haven: Yale University Press, 1999. ———. “The Political Consequences of Electoral Laws, 1945–1985.” American Political Science Review 84, no. 2 (June 1990): 481–6. ———. “Unequal Participation: Democracies Unresolved Dilemma.” American Political Science Review 91, no. 1 (March 1997). Lijphart, Arend and Bernard Grofman, eds. Choosing an Electoral System: Issues and Alternatives. New York: Praeger, 1984. Lind, Michael. “A Radical Plan to Change American Politics.” Atlantic Monthly (August 1992). Low-Beer, John R. “The Constitutional Imperative for Proportional Representation.” Yale Law Journal 94, no. 1 (November 1984): 163–88. Matland, Richard. “Enhancing Women’s Political Participation: Legislative Recruitment and Electoral Systems.” In Azza Karam, ed., Women in Parliament: Beyond the Numbers. Stockholm: International IDEA, 1998. Milner, Henry (ed.). Making Every Vote Count: Reassessing Canada’s Electoral System. Peterborough, Ontario: Broadview Press, 1999. Mulroy, Steven. “The Way Out: A Legal Standard for Imposing Alternative Electoral Systems as Voting Rights Remedies.” Harvard Civil Rights/Civil Liberties Law Review 33 (Summer, 1998): 334–80. Pildes, Richard and Kristen Donoghue. “Cumulative Voting in the United States.” University of Chicago Legal Forum (November 1995). Pinto-Duschinsky. Michael. “Send the Rascals Packing: Defects in Proportional Representation and the Virtues of the Westminster Model.” Representation 36, no. 2 (Summer 1999). Piven, Frances Fox and Richard A. Cloward. Why Americans Still Don’t Vote: And Why Politicians Want It That Way. Boston: Beacon Press, 2000.

302

Selected Bibliography

Powell, G. Bingham. Elections as Instruments of Democracy: Majoritarian and Proportional Visions. New Haven: Yale University Press, 2000. Rae, Douglas W. The Political Consequences of Electoral Laws. New Haven: Yale University Press, 1971. The Report of the Independent Commission on the Voting System (“The Jenkins Report”). http://www.official-documents.co.uk/document/cm40/4090/contents.htm (September 12, 2001). Reynolds, Andrew. “Women in the Legislatures and Executives of the World: Knocking at the Highest Glass Ceiling.” World Politics 51, no.4 (July 1999). ———. “Majority and Power-Sharing Government.” In Markus Crepas, et al., eds., Democracy and Institutions: The Life Work of Arend Lijphart. Ann Arbor: University of Michigan Press, 2000. Reynolds, Andrew and Ben Reilly, The International IDEA Handbook of Electoral Design. Stockholm: International Institute for Democracy and Electoral Assistance, 1997. Richie, Robert and Steven Hill. Reflecting All of Us: A Case for Proportional Representation. Boston: Beacon Press, 1999. ———. Whose Votes Count. Boston: Beacon Press, 2001. Royal Commission on the Electoral System. Report of the Royal Commission on the Electoral System: Towards a Better Democracy. Wellington, New Zealand: V. R. Ward, Government Printer, 1986. Rule, Wilma. “Electoral Systems, Contextual Factors and Women’s Opportunity for Election to Parliament in Twenty-Three Democracies.” Western Political Quarterly 40, no. 3 (September 1987). Rule, Wilma and Joseph Zimmerman. “Women and Minorities in Parliaments and Legislatures.” The Encyclopedia of Democracy. Washington, D.C.: Congressional Quarterly Press, 1997. Rush, Mark and Richard Engstrom. Fair and Effective Representation? Debating Electoral Reform and Minority Rights. New York: Rowan and Littlefield, 2001. Shugart, Matthew Soberg. “Extreme Electoral Systems and the Appeal of the MixedMember Alternative.” In Shugart and Wattenberg, eds., Mixed-Member Electoral Systems (listed below), pp. 25–54. ———. “The Jenkins Paradox: A Complex System, Yet Only A Timid Step Towards PR.” Representation 36, no. 2 (Summer 1999): 143–147. Shugart, Matthew Soberg and Martin P. Wattenberg, “Conclusion: Are MixedMember Systems the Best of Both Worlds?” In Shugart and Wattenberg, eds., Mixed-Member Electoral Systems (listed below), pp. 571–96. Shugart, Matthew Soberg and Martin P. Wattenberg, eds., Mixed-Member Electoral Systems: The Best of Both Worlds? Oxford: Oxford University Press, 2001. Still, Edward. “Alternatives to Single Member Districts.” In Chandler Davidson, ed., Minority Vote Dilution. Washington D.C.: Howard University Press, 1984.

Selected Bibliography

303

———. “Voluntary Constituencies: Modified At-Large Voting as a Remedy for Minority Vote Dilution in Judicial Elections.” Yale Law and Policy Journal (1992). Ralph A. Straetz. PR Politics in Cincinnati. New York: New York University Press, 1958. Taagepera, Rein and Matthew Soberg Shugart. Seats and Votes: The Effects and Determinants of Electoral Systems. New Haven: Yale University Press, 1989. Vowles, Jack. Electoral System Opinion and the Evolution of MMP: A Report to the Electoral Commission. University of Waikato, New Zealand: New Zealand Election Study, 2000. Weaver, Leon. “The Rise, Decline, and Resurrection of Proportional Representation in Local Government in the United States.” In Grofman and Lijphart, eds., Electoral Laws and Their Political Consequences New York: Agathon, 1986, pp. 139–53. Winger, Richard. “Institutional Obstacles to a Multiparty System.” In Paul S. Herrnson and John C. Green, eds., Multiparty Politics in America. New York: Rowan and Littlefield, 1997. Zeller, Belle and Hugh Bone, “The Repeal of PR in New York City: Tens Years in Retrospect,” American Political Science Review 42 (December 1948): 1127–48. Zimmerman, Joseph F., ”PR in Ireland: A Firm Tradition,” National Civic Review 72, no. 11 (December, 1983).

This page intentionally left blank

Additional Resources on the World Wide Web

PR Library: Readings in Proportional Representation A source of information on proportional representation voting systems— including beginning readings, in-depth articles by scholars and activists, an extensive bibliography, and a guide to related Web sites. Web address: http://www.mtholyoke.edu/acad/polit/damy/prlib.htm

The Center for Voting and Democracy A nonprofit educational organization located in Washington D.C. The Center is the leading organization in the United States dedicated to educating the public about alternative election systems. Web address: http://www.fairvote.org

Californians for Proportional Representation A good example of the state groups now organizing around the issue of proportional representation. Much good information about PR and how to promote it at the local level. Web address: http://www.fairvoteca.org/

306

Resources on the World Wide Web

Electoral Reform Society Founded in England in 1884, the ERS is the longest continuous PR organization. This site contains particularly good information about the single transferable vote—the Society’s preferred form of PR. They also have an excellent analysis of the report issued by the Jenkins Commission that recommends a new voting system for British parliamentary elections. Web address: http://www.electoral-reform.org.uk/

Fair Vote Canada This is the main organization promoting PR in Canada. Its goal is to provide concerned citizens, community leaders and activists with an organizational framework to build a nationwide grassroots campaign for voting system reform in Canada. Web address: http://www.fairvotecanada.org/

The International Institute for Democracy and Electoral Assistance Located in Sweden, this organization strives to promote sustainable democracy worldwide. It has a particular interest in improving electoral systems. It has particularly good information about women’s representation and voter turnout worldwide. Web address: http://www.int-idea.se/

Index

Note to the Index The following abbreviations or briefer wordings may be used in the index: ballot samples b. figure(s) f. or ff. note(s) n. or nn. proprortional representation proportional system(s) or PR tables t. plurality/plurality-majority systems including the single-member district plurality system widely used in the United States plurality system(s) Abrams, Robert, 83 accountability of representatives, 154–155; greater with proportional systems, 177–179. See also legislative representatives Adams, John, 32 additional member system. See mixedmember proportional representation (MMP) affirmative gerrymandering. See majority-minority districts African-American representation, 8; court-mandated cumulative voting to secure, 147–148, 148, 210, 246;

proportion elected to Congress, 126; proportion of elected to state offices, 127t.; race-based or affirmative gerrymandering and (see also majorityminority districts), 135–137, 141, 142; “safe seats” for, 142; the single transferable vote and, 144–146. See also racial and ethnic minority representation African National Congress, South Africa, 3, 119, 203 Alabama, cumulative voting in, 147–148, 210, 246; redistricting battles in, 58

308 alienation of voters, 12, 75, 84. See also voter turnout alternative vote, see instant runoff voting Amarillo, Texas, cumulative voting in school board elections, 149 American Communist Party, 180 American Independent Party, 98 American-style elections. See singlemember plurality (SMP) elections America. See United States Amy, Douglas J., A Citizen’s Guide to Voting Systems, 295n.1 Anderson, Christopher J. and Christine A. Guillory, 183–184 Anderson, John, 97–98 Ansolabehere, S. and S. Iyengar, 84 anticommunism, associated with the repeal of PR in New York City, 206, 208, 241, 273–274 antidemocratic election outcomes: disproportional representation, 33–36, 34t., 35t., 40, 42, 96; losing parties winning, 36, 43–46, 249; mythologized as fair representation, 28–29, 42; uncontested seats, 60, 61–62, 94. See also democracy; electoral failure; manufactured majorities; minority rule anti-PR objections. See proportional representation (PR)—objections to apolitical gerrymandering, 64–66 approval voting, 295n.1 Arkansas, electoral failure in 2000 Congressional election, 45 Ashtabula, Ohio: first U.S. city to adopt PR, 268; Socialist candidate elected in, 270 Asian American representation, 8, 140–141. See also racial and ethnic minorities

Index at-large elections: history of, 130–131; racial and ethnic minorities underrepresented in, 129–131, 245–256 — modified-at-large elections. See cumulative voting (CV) Australia: civic support for PR system in, 237; single transferable vote in, 21 Austria, two-party system with PR system in, 2, 100 Balkanization, fear that PR will result in, 188, 200–203 ballot-access restrictions, 108, 282n.12 ballots: German mixed-member, 20; number of invalid compared across voting systems, 189, 190 ballot samples: for cumulative voting (CV), 220b.; for instant runoff voting (IRV), 216b.; for limited voting, 221b.; for mixed-member PR, 21b.; for party list PR (closed list), 18b.; for party list PR (open list), 19b.; for single transferable vote (STV), 23b. Banducci, Susan, Todd Donovan, and Jeffrey A. Karp, 31, 128 Barber, Kathleen L., on proportional representation in Ohio’s history, 30, 120, 168–169, 268–272 Belgium: legislative power-sharing in, 172, 293–294n.27; PR system used in, 2 Bentsen, Ken, 59 Berry, Theodore, 145–146 Beyle, Thad, 58 Bibby, John, 192, 200 bicameral political structure, U.S., 193, 290n.3 Blair, Tony, 4, 245, 257 Blais, Andre and R. K. Carty, 157 Bobo, L. and F. Gilliam, 128 Bogdanor, Vernon, 294n.43

Index Bosma, Charles, 56 Boulder, Co., single transferable vote in, 21 Bowler, Shaun, Todd Donovan, and David Brockington, 69, 148–149; on voter turnout in U.S. cities, 157 Braun, Carol Mosely, 181 Britain. See Great Britain, U.K. Brother Rice High School, 173–174 Burnham, Walter Dean, 154, 162 Burrell, Barbara, 112–113 Bush, George, Sr., 76 Bush, George W., 73; and the 2000 Presidential election, 1, 44, 97 Bush v. Vera, 135 Cacchione, Pete, 208 Caddell, Patrick, 77 Cain, Bruce, 58 California: gerrymandering in, 56, 57, 60; electoral failure in 1994 Congressional election, 45 Californians for Proportional Representation: 5, 238; Web site, 305. See also Sacramento; San Francisco Cambridge, Massachusetts: conflict management in, 202–203, 293n.15; a contemporary U.S. city with proportional representation, 272–273, 274; use of the single-transferable vote in, 21, 22, 31, 86–87, 189, 195, 227 campaigns. See political campaigns Canada: electoral failures in under the plurality system, 2, 4–5; the New Democratic Party in, 165; pro-PR support in, 4–5 Canady, Charles T., 191 candidate-centered proportional representation systems, 191, 228, 233 candidates for public office: competing in proportional systems, 138–139; in-

309 dependent candidates, 98, 139, 144, 225; lukewarm support for, 6, 7, 10, 96; public funding for, 108; running unopposed, 60, 61–62, 94; whether candid or principled, 81–82; women candidates, 112–124. See also spoiler-candidate effect; legislative representatives candidate vs. party voting, 187, 191, 281n.15 Castles, Francis G., 114–115 Center for Voting and Democracy (CVD): 5, 36, 60–61, 218, 237; Web site, 305 Chamber of Commerce, 238, 273 Chilton County, Alabama, cumulative voting in, 147–148, 210 choice voting. See single transferable vote (STV) Christian Coalition, 92 Christiano, Thomas, 180–182 Cincinnati, recent referendum on PR in, 225, 254–255 —history of proportional representation in (1925–1957), 5, 144–146, 268; the Charter Committee, 145, 273; racial fears involved in repeal of PR in, 240–241, 273; use of the single transferable vote, 21, 143–147; cities: historic use of proportional systems by (see also by city), 253; recent referendums on PR in, 225, 254–255; voter turnout in, 157. See also Cambridge, Massachusetts; Cincinnati; Cleveland; New York City; San Francisco citizen satisfaction with political processes, 183–184 class bias: impacts on policymaking, 163–164; in voter turnout, 153,

310 class bias: (continued) 154–156. See also low income citizens Cleveland, history of proportional representation in, 5, 87, 120, 268; repeal of PR in, 272; use of single transferable vote in, 21, 30; women better represented with PR in, 120 Cleveland Plain Dealer, 87 Clinton, William Jefferson (Bill), 76 closed party list voting, 191. See also party list proportional representation coalitions in two-party systems, 193–194 coalition governments in multiparty systems: broad participation in policymaking with, 170–171; criticized as lacking valid mandates, 198–200; governing efficacy of, 192, 196, 207, 237; incremental changes in, 176–177, 196–197; involved in power sharing and dispersed power, 171–172, 176–179, 184, 196; objection that they are unstable, 187, 192–196; policymaking efficacy of, 192, 196; power sharing and dispersed power in, 171–172, 176–179, 184, 196; preelection knowledge of potential, 199, 204; proportional representation encouraging formation of, 98–101, 107–108. See also consensus/PR vision of democracy; multiparty systems; proportional representation (PR) Common Cause, 64 Communist Party, NYC: representation of under PR and anticommunism associated with the repeal of PR in, 206, 208, 241, 273–274 competitiveness of elections: incumbency factor reducing, 6, 67, 198; voter turnout encouraged by, 159–161

Index Condorcet vote-counting system, 295n.1 conflict management: in Cambridge, Massachusetts with PR, 202–203, 293n.15; in multiparty systems, 201–202; and the single transferable vote, 271–272. See also consensus/ PR vision of democracy Congress, U.S.: campaigns for, 70, 71, 80; voter turnout for elections to, 27–28 — House of Representatives: electoral failure in the 2000 election to, 36, 45–46, 56, 94; hypothetical example of party list PR election of, 226–227; low turnout for elections to, 155; manufactured majorities in, 43, 45, 47–48, 249; proportional representation is possible under the U.S. Constitution, 239; proportion of seats uncontested, 61–62; underrepresentation of women in, 109 — Senate: negative campaigning in races for, 83; political diversity affecting debate in, 181; underrepresentation of women in, 109 consensus/PR vision of democracy, 169, 170–171, 201; deliberative democracy and, 182; inclusive policymaking and, 177–179; political dialogue and, 179–180; political diversity and, 181–182; as win-win politics, 174–176, 184. See also conflict management; multiparty systems; proportional representation (PR) Conservative party, U.K., overrepresentation of, 37–38, 249 constituency service. See legislative representatives Constitution, U.S.: concern for minority rights in, 173; constitutional arguments for PR, 246–247; the Consti-

Index tutional Convention, 32; presenting no obstacle to PR elections of the House of Representatives. See also court rulings; Supreme Court, U.S. continuity of policy. See governmental stability corporate board elections, 220 corporate interests in electoral systems, 241 court rulings: on at-large elections, 131; mandating cumulative voting, 147–148, 210, 246; on redistricting, 58–59, 132, 134, 135. See also majority-minority districts; Supreme Court, U.S. — cases: Bush v. Vera, 135; Easley v. Cromartie, 134; Shaw v. Reno, 134, 136; Thornburg v. Gingles, 131; Timmons v. Twin Cities, 247 cracking, form of gerrymandering, 54–55, 55ff. Cube law, 277n.15 cumulative voting (CV), 220b., 220–221; in Alabama, 147–148, 148, 210, 246; corporate boards elected by, 220; current use of in the United States, 147–148, 190, 220, 221; disproportional representation possible with, 221–222, 223t.; manufactured majorities possible with, 222, 223, 224t.; proportional representation possible with, 221, 222t.; used in Texas, 190; used in Illinois historically, 220, 223 CV. See cumulative voting (CV) cynicism of voters, 12, 75, 84 Cypress, PR system used in, 2 Czechoslovakia, PR system used in, 3 D’Amato, Alphonse, 83 Darcy, R., Susan Welch and Janet Clark, 112–113

311 debate. See dialogue and debate DeLay, Thomas (Tom), 57 democracy: consensus/PR vs. majoritarian vision of, 169–172, 182t., 182; invested in the electoral system (see also electoral systems), 32, 40–41, 72, 168–169. See also antidemocratic election outcomes; consensus/PR vision of democracy; majoritarian vision of democracy — democratic governments: European democracies, 18, 19, 195, 223, 233; new democracies in Eastern Europe, 3, 18. See also by nation — democratic values: electoral reform and, 167–169, 175, 255–258; principle of majority rule (see majority rule); principle of representation (see political parties; representation); universal suffrage idea (see also voter turnout), 26, 256–257 Democratic Farm-Labor Party, U.S., 96, 106 Democratic Party, U.S.: cases of overrepresentation or underrepresentation of, 36, 244; gerrymandering by, 54–56, 59; ignoring progressive constituency, 92; minor parties and, 100, 148; potential support for proportional representation in, 43, 245; Republican platform and policies compared with, 90–92; seat allocations under the plurality system, 48, 49t., 50; splinter groups in, 103; voter identification with, 89, 200 — hypothetical examples featuring: gerrymandering, 55f.2.1.; political status with cumulative voting, 222t., 223t., 224t.; political status with single transferable vote, 228, 266; political status under PR

312 Democratic Party (continued) systems, 66–67, 99–100, 103–104, 180, 198 Denmark: electoral system of; legislative power-sharing in, 172; PR system used in, 2 developing countries and proportional representation, 3. See also Malta; South Africa d’Hondt seat allocation formula, 260–261, 261t.A dialogue and debate: campaign rhetoric, 70–72; candor of candidates, 73, 81–82; clarity and specificity of party platforms, 161–162; enlivened and clarified with PR elections, 161–162, 179–180, 214; hidden information, 73; issue-orientation discouraged in plurality/plurality-majority elections, 6, 7, 8, 69–71, 74–77, 164, 180–181; policy-related concerns of constituencies, 73, 107, 111–112, 125, 252; multiparty systems encouraging, 105–107, 201; voter turnout related to level of, 164. See also political campaigns discourse. See dialogue and debate dispersed power, 171 disproportional representation: discouraged in proportional systems, 40; gerrymandering and, 53–54; plurality system produces, 33–36, 34t., 35t., 40, 42, 96 district magnitude: in Germany, 232–233; gerrymandering and, 279–280n.29; in Ireland, 121–122, 270–280n.29; in Israel, 2, 232; large multimember districts ensuring broad representation, 66, 138, 226, 228; minor-party inclusion thresholds affected by, 231–233,

Index 232t.; women’s representation and, 122 district representation. See geographical representation; redistricting Dixon, Robert, 66 Downs, Anthony, 161, 164 Droop quota, 260 dual electoral system. See mixedmember proportional representation (MMP) Duke, David, 201, 209 Duverger, Maurice, 95–96 Easley v. Cromartie, 134 Eastern Europe, new democracies and trend toward PR in, 3, 18 efficacy. See governing efficacy Eisenhower, Dwight D., 145 elections, 32; complexity of ballots and administration of, 187, 188–191; expense of runoffs, 217; incumbency factor in (see also incumbency), 113, 120–121. See also electoral systems; political campaigns Electoral College: distortions resulting from, 44; failure of in the 2000 Presidential election, 1–2, 44, 249; public opinion on, 44; reform or abolition of, 44, 233–234; representation of women on electoral slates, 119–120 electoral failure: in Canada, 2, 4–5; in Great Britain in the 1974 Parliamentary elections, 44; in New Zealand prior to electoral reform, 45, 198; in the U.S., 45, 249–251; in U.S. House of Representatives elections, 36, 45–46; in the 2000 U.S. Presidential election, 1–2, 44, 249. See also minority rule; and by state electoral reform, 12–13, 63–66, 108, 150, 237; Canadian political parties

Index supporting, 4–5; democratic values and, 167–169, 175, 255–258; difficulty of under the plurality system; 63–66; instant runoff voting as an alternative for single-office elections, 215–219; legislative proposals in U.S. for, 239; national referendum enacting in New Zealand, 3–4, 224–225, 245; opposition of political elites to, 242–243, 272; plurality system discouraging, 95–96; pro-PR movement in Great Britain, 4, 44–45, 244–245, 249; pro-PR movement in the U.S., 4, 5–6, 241, 251–253, 273; prospects in the U.S., 215 (Ch. 10), 243–244; tradition working against, 95; Web sites on, 305–306. See also court rulings; PR prospects in the U.S.; public opinion Electoral Reform Society, Web site, 306 Electoral Reform Society of Great Britain, 237 electoral systems: comparative studies of, 153, 159, 161, 276n.12; democracy and, 40–41, 72, 168–169; impact on women’s representation (see also women’s representation), 124; party systems and, 95–96; plurality winners, 42; political dialogue and, 179–180. See also plurality/pluralitymajority systems; proportional representation (PR) electoral threshold. See threshold of exclusion Emily’s List, 123 Engstrom, Richard, 190, 213 environmental policy, 73. See also Green parties Eulau, Hans, 153–153 European democracies, PR systems predominant in, 18, 19, 223, 233

313 exaggerated majorities: producing misleading policy mandates, 196; versus earned majorities, 37. See also manufactured majorities extremist candidates and fringe parties, 188, 201, 205–210, 213 factionalism, fear that PR will lead to, 188, 200–203 Fairfield, Alabama, 130 Fair Vote Canada, Web site, 306 Fairvote Minnesota, 5, 238 Fairvote Wisconsin, 5, 238 Farmer-Labor Party, U.S., 96, 106 Farrell, David, 190, 195 feminist issues, 110–112. See also gender; women’s representation Feminist Majority, 124 feminist political efforts, 92, 99, 119, 228; organizational support for proportional representation, 122–124; third party approach, 123–124 Finer, S.E., 197 Finland, PR system used in, 2, 19 First-past-the-post (FTTP) elections. See Single-member plurality (SMP) elections Florida, electoral failure in 1998 Congressional election, 45 formulas for seat allocation. See seat allocation formulas in PR systems France, 278n.26 Franklin, Mark N., 156, 157, 163 fringe parties or groups, 188, 201, 205–210, 213, 241 Frost, Martin, 61 fundamentalist Christians, 92, 99, 104, 228 Fund for the Feminist Majority, 123 fusion ballots, 247

314 Gallagher, Michael, 33 Gans, Curtis, 155, 164–165 gender: the gender gap in voting, 111; gender and legislative preferences, 111–112. See also women’s representation geographical representation, 287n.34; ensured in single-member plurality system, 210–213, 214; oddly shaped districts and, 134, 135; possible with some forms of PR, 212, 213, 295n.55; as undermined by some forms of PR, 188, 210–213. See also district magnitude Germany: district magnitude in, 232–233; legislative power-sharing in, 172; mixed-member PR system used in, 2, 20; overstability of cited, 197; representation of women in, 110, 117–118. See also mixedmember proportional representation Gerry, Elbridge, 53 gerrymandering: bipartisan arrangements, 59–61, 94; bipartisan commissions overseeing, 64; cracking, 54–56, 55ff.; disproportional representation and, 53–54; district magnitude and, 279–280n.29; endemic in single-member plurality system, 9, 12, 14, 52 (Ch.2); examples of, 54–56, 55ff.; minimized in proportional systems, 65–67, 121, 139; neutral or apolitical districting, 64–66; packing, 54–56, 55ff., 132; partisan gerrymandering, 53–54, 56–57, 64, 246; racism and, 132–133; reduced with mixedmember proportional representation, 230; “safe seats” secured by, 59–63, 69, 121, 142, 158; Supreme Court’s silence on partisan gerry-

Index mandering, 64, 134, 135; unintentional gerrymandering, 65–66 Gingrich, Newt, 164 Gore, Al, 1, 42, 44, 97 Gorton, Slade, 193 governing coalitions: broad participation in multiparty systems (see also coalition governments in multiparty systems), 170–171; occurring in twoparty systems, 193–194 governing efficacy: of multiparty coalitions, 192, 196, 207, 237; not guaranteed with plurality systems, 197 government stability: and electoral systems, 153, 194, 272; the illusion of, 130; the issue of overstability, 197; the objection that multiparty systems are unstable, 187, 192–196; political swings occurring in plurality systems, 196–197; voter turnout and, 153 Grassroots activism for proportional representation (PR), 5–6, 241, 245 Great Britain (U.K.): electoral reform studied in, 4, 187, 257; manufactured majorities in, 37–38, 109, 249; minority rule in, 44–45, 248–249; plurality system used in, 2; potential impact of PR on political system, 41; pro-PR movement in, 4, 44–45, 244–245, 249; public support for PR system used in, 4, 44–45; representative-constituency relationship in, 211; unintentional gerrymandering in, 65; Welsh and Scottish parties in, 96. See also Canada Greece, PR system used in, 2 Green parties, 93 Green Party, Germany, 78 Green Party, U.S., 43, 89, 107, 254; endorsing proportional representation,

Index 5, 254; hypothetical status of with mixed-member PR, 230 gridlock, 192 Grofman, Bernard, 126, 137, 243 Guinier, Lani, 5, 12, 85, 87; on majority-minority districting, 142–143; on minority representation, 172–176; on pluralistic perspectives, 105–106; on “taking turns,” 175–176 Hallett, George, Jr., 86, 87 Handley, L. and B. Grofman, 136 Hansard Society Commission on Electoral Reform, 187 Hare quota, 259–260 Hasidic Jews, 141 Henry, Patrick, 53 Hertzberg, Hendrik, 9, 159 Hill, Steven and Leighley, 163 Hispanic representation, 8, 126, 132, 140–141, 148–149. See also Racial and ethnic minorities history of proportional representation in the U.S. See proportional representation in U.S. history Hitler, Adolf, 207 House of Representatives. See under Congress, U.S. Idaho, 104; Democrats underrepresented in, 244 Illinois: cumulative voting used historically in, 220, 223; effort to bring back CV, 220 Illinois Citizens for Proportional Representation, 6, 238 Imperiali seat allocation formula, 262 incumbency: as a factor affecting representation of women, 113, 120–121; gerrymandering and (“safe seats”), 59–61, 69; incumbents running for

315 reelection, 75; lack of competitiveness and, 6, 67, 198; term limits and, 121 independent candidates, 98, 139, 144, 225 Independent Commission on the British Voting System, 4, 257 Indiana; electoral failure in the (1982 state elections), 45; gerrymandering in, 56 instant runoff voting (IRV): applicable to single-office elections, 217, 219; inadequate for legislative elections, 217–219, 296n.5; majority ensured without runoff elections, 215–217, 216b., 218, 219; political minorities underrepresented with, 218; spoilercandidate effect eliminated by, 217, 277–278n.25; support for in several states, 218 institutional approach, explanation for voter turnout, 156–157, 290n.3 Institute of Governmental Studies, 58 interparty cooperation. See governing coalitions intraparty rivalries, 228 invalid ballots, compared across voting systems, 189, 190 Iowa, politically neutral redistricting in, 64 Ireland: magnitude gerrymandering in, 279–280n.29; PR (single transferable vote) used in, 2, 21; underrepresentation of women in, 121–122 IRV. See instant runoff voting (IRV) Israel: district magnitude in, 2, 232; extreme form of PR system used in, 194–195; few wasted votes in, 30; Knesset, 232; low electoral threshold in, 2, 195

316 issue-orientation of campaigns. See dialogue and debate Italy: atypical form or PR system used in, 194; unstable governments in, 192–193 Jackson, Jesse, 5 Johnson, R. J., 65 Joslyn, Richard, 70, 72 Justice Department, U.S., 59, 132, 133, 137 Kahn, K. F. and Kenney, P. J., 85 Karp, J. and S. Banducci, 159, 161 Katz, Richard, 80 Kentucky, gerrymandering in, 56–57 Keyssar, Alexander, 256 Kimberling, William, 204 Klain, Maurice, 256 Kleppner, Caleb, 218–219 Klingemann, Has-Dieter, 183 Knesset, Israel, 232 Kolesar, Robert, 273 Krauthammer, Charles, 153 Kyoto Treaty, 73 labor unions, 252, 254; as part of a proPR coalition, 238, 254 Labour Campaign for Electoral Reform, 244–245 Labour Party, U.K., 4, 244–245 Labour Party, New Zealand, 45 large multimember districts, 66, 138, 226, 228. See also proportional representation (PR) largest remainder seat allocation formula, 259–260, 260t.A Latino representation, 8, 132, 140–141, 148–149; in the U.S. Senate, 126. See also Racial and ethnic minorities League of Women Voters, 5, 238

Index legal benchmarks. See court rulings; Supreme Court, U.S. legal strategies for proportional representation, 245–247. See also Voting Rights Act of 1965 legislative representatives: accountability of, 154–155, 177–179; constituency-relationship with, 211; policy preferences of compared with constituency-preferences, 8, 40–41, 46, 105, 184–185; policy preferences of varying with gender, 111–112; post-election citizen evaluation of, 72–73 legislatures: gender gaps in legislative preferences, 111–112; legislative gridlock, 192; misleading mandates possible with plurality elections (see also manufactured majorities), 196, 198; power-sharing in, 172, 181; responsiveness of to elections, 47–48, 62–63. See also seat allocation legitimacy, political, 111–112, 155 Liberal Democrats, U.K., 38–39, 248 Liberal Party, New York, 101 Libertarian Party, U.S., 89, 93, 107, 208, 250 Lijphart, Arend: his typology of democracies, 169–171; on low voter turnout, 154; on majority turnovers, 196–197; on minority representation, 138, 139–140; on proportional representation, 12, 39, 44, 157, 195–196; limited vote, 221b., 221 Linder, John, 61 Lind, Michael, 104 litigation. See court rulings lobbying, 12 local elections. See state and local elections Locker, Jesse, 145

Index Louisiana, electoral failure in 1994 Congressional elections, 45 Low-Beer, John R., “The Constitutional Imperative of Proportional Representation,” 246–247 low-income citizens: better represented in proportional systems, 178–179; impact on policymaking, 177–179; underrepresented in two-party system, 164–165 low voter turnout. See under voter turnout Luxembourg, PR system used in, 2 McDaniel, Mark, 52 McGovern, George, 103 McHugh, Matthew, 84 McKinney, Cynthia, 239 Madonna, Terry, 57 majoritarian vision of democracy, 169–170, 173. See also two-party system; winner-take-all systems majority-minority districts, 14, 135–137, 240, 250–251, 287n.34, 133–134; advantages of proportional representation over, 140–143; assumptions about, 135–136; housing segregation and, 141; race-based creation of, 14, 135–137, 240, 250–251, 287n.34; safe seats in black majority districts, 142. See also racial and ethnic minorities majority preferential voting. See instant runoff voting (IRV) majority rule, principle of, 36, 46, 217, 218; and the 2000 Presidential election, 1–2, 44, 249; violations of, 38, 45–46, 50–51. See also democracy; representation Malta, single transferable vote in, 2, 21 management of conflict. See conflict management

317 mandates: coalition governments criticized as lacking valid, 198–200; plurality elections can result in misleading, 196 manufactured majorities, 36–39, 39t., 42, 46; in Canada, 4; discouraged by proportional representation, 40–41; exaggerated majorities, 37; in Great Britain, 37–38, 109, 249; misleading mandates resulting from, 196; the New York 1992 U.S. Senate race and, 83; underreporting of, 42; in the U.S. Congress, 43, 45, 47–48, 249 Massachusetts: history of proportional representation in, 268; Republicans underrepresented in, 27, 244, 261. See also Cambridge, Massachusetts Matland, Richard, 116 Maude, Angus, 211 media: campaign media use, 8, 70; coverage of proportional representation by, 237; underreporting of manufactured majorities by, 42 Mexican American representation, 132, 141 Michigan, electoral failure in 1994 elections, 45 Midwest Democracy Center, 5, 238 Miller, Arthur and Ola Listhaug, 102–103 Miller, Carol, 43 Mill, John Stuart, 27, 110 Milner, Henry, 13 Minnesota, Farmer-Labor Party in the history of, 96 minority rule, 36, 43–46, 249; in Canada, 4; in Great Britain, 44–45, 244–245; plurality elections may result in, 42, 48, 49t., 50; in New Zealand, 45, 198; periods of in the

318 minority rule (continued) U.S. Congress, 36, 43–46; possible with cumulative voting, 223; in the Presidential election of 2000, 44, 249. See also manufactured majorities; spoiler-candidate effect minor parties: discouraged in plurality systems, 11–12, 33, 95, 250; discouraged in the U.S., 10, 39, 89, 98, 106–108, 242, 249–250; electoral threshold of exclusion of in PR, 201, 284n.26; emergence of new parties, 165; extent of support for in U.S., 89, 97–98, 106–108; fusion ballots helping candidates from, 247; voicing new political ideas, 105–107. See also third parties in the U.S.; threshold of exclusion; and by individual party Mississippi, racist gerrymandering in, 132, 133 mixed-member proportional representation (MMP), 20, 21b., 22t., 230–231, 233; dual-voting provides broad choice, 229–230; geographical representation ensured under, 212–213, 229; gerrymandering less of a concern with, 230; women’s representation benefiting in, 117–118, 122 mixed-member voting. See mixed-member proportional representation (MMP) MMP. See mixed-member proportional representation (MMP) modified-at-large voting. See cumulative voting (CV) Moses, Robert, 206 Mulroy, Steven, 210 multiculturalism, 125. See also racial and ethnic minorities multimember districts, large, 66, 138, 226, 228. See also district magnitude

Index multiparty systems, 88 (Ch.4); conflict management under, 201–202; emergence of new parties in, 165; encouraging dialogue and debate, 105–107, 179–180; fear that extremism is encouraged by, 188, 201, 205–210, 213; greater representation in, 104–105; issues and partisan platforms well defined in, 79–81; methods of limiting small parties in (see threshold of exclusion); new ideas encouraged in, 105–107; political campaigning in, 85–86; political debate widened by (see dialogue and debate); voter choice increased in, 101–103. See also coalition governments in multiparty systems; minor parties; proportional representation (PR) Nader, Ralph, 10, 42–43 National Association for the Advancement of Colored People (NAACP), 145 National Conference of State Legislatures, 64 National Municipal League, 268, 273 National Organization for Women, 124, 254 National Party, New Zealand, 45 National Women’s Political Caucus, 123 Natural Law Party, U.S., 89 Nazi Party, 205–207 negative campaigning: example of in U.S. Senate race, 83; problem of, 83–84, 85; U.S. public opinion on, 68, 82–83 negotiation democracy. See consensus/PR vision of democracy Netherlands: PR system used in, 2; stability of, 198

Index neutral redistricting, 64–66 New Democratic Party, Canada, 165 New Mexico: House of Representatives elections in, 43, 45–46; minority rule in, 43; support for instant runoff voting in, 218 New Party, U.S., 89, 247 New York City: elections of Communists under PR and anticommunism associated with the repeal of PR in, 206, 208, 241, 273–274; multiparty system with PR system used in the 1930s and 1940s, 100–101, 242, 270–271; racial and ethnic minority representation issues in, 141; redistricting process in, 54–56; Tammany Hall political machine in, 206, 267–268, 269–270 New York State: gerrymandering in, 55–56; history of proportional representation in, 5; manufactured majority in 1992 U.S. Senate race in, 83; use of single transferable vote, 21, 206 New York Times, 37, 60 New Zealand: minority rule under plurality system, 45, 198, 248; mixedmember proportional representation chosen in, 2, 117, 224, 229; popular support for PR in, 3–4, 241, 245; PR system leading to more positive voter attitudes, 31; rejecting the plurality system in national referendum, 3–4, 224–225, 245; representation of women increased with adoption of PR, 117–118 New Zealand Royal Commission on the Electoral System, 189, 205 North Carolina, 134; redistricting battles in, 58 Northern Ireland, political tensions under plurality system, 202

319 Norway: multiparty system in, 91, 93; PR system used in, 2; representation of women in, 110 Norwegian Labor Party (Norway), 118–119 O’Connor, Sandra Day, 134 Ohio, effects of proportional representation when used in, 195, 268–272. See also Cincinnati; Cleveland; Toledo one-party rule, 94–95 open party list voting, 19b., 19, 122, 191, 225–226. See also party list proportional representation overstability, 197 packing, 54–56, 55ff., 132. See also gerrymandering parallel voting, 198n.23 parliamentary systems, 193. See also Great Britain partisan gerrymandering, 53–54, 56–57, 64; eliminated in proportional systems, 65–67, 139; Supreme Court’s silence on, 64, 134, 135 party-centered versus candidate-centered elections, 187, 191, 281n.15 party coalitions, in two-party system, 193–194. See also coalition governments in multiparty systems party list proportional representation: closed list version of, 18b., 18–19; ensuring broad representation, 118–119, 121–123, 225, 230; open list version of, 19b., 19, 122, 191, 225–226; party mobilization encouraging voter turnout, 160, 165; track record with democratic governments, 233; women’s representation and, 118–119, 121, 123

320 party mobilization, voter turnout encouraged by, 160, 165 party platforms, clarity and specificity of, 161–162; of Democrats and Republicans compared, 90–92 party primaries. See primary system party representation. See minor parties; representation party systems, 194; voting rules and, 95–96. See also multiparty system; two-party system Pelosi, Nancy, 62, 158 Pennsylvania: electoral failure in 1998 Congressional election, 45; redistricting battles in, 57 Perot, H. Ross, 10, 76 Pildes, Richard and Kristen Donoghue, 148, 210 Pinto-Duschinsky, Michael, 197 Piven, Frances Fox and Richard A. Cloward, 155, 164 plurality/plurality-majority systems, 6, 7, 8, 33, 69–71, 74–77; lack of competitiveness in (see also candidates for public office; dialogue and debate), 6, 67, 198; media reliance in, 8, 70; negative campaigning in, 68, 82–85; plurality winners in, 42; primary system in, 229, 277n.14; public opinion on, 68; as winner-take-all systems, 2, 15, 33. See also at-large elections; single-member plurality (SMP) elections; winner-take-all systems policy continuity. See government stability policy information. See also dialogue and debate policymaking: based on misleading or exaggerated mandates, 196; broad participation in with proportional representation, 170–171, 176, 181;

Index responsiveness to elections, 47–48, 62–63 policy preferences: agreement between representatives and voters, 8, 40–41, 46, 105, 184–185, 199, 205; gender and legislative preferences, 111–112; issue-related concerns of voters, 73, 107, 111–112, 125. See also majority rule political campaigns, (Ch.14); campaign finance, 12, 108, 112–113, 121, 130, 252–253; campaign rhetoric, 70–72; candor over issues in (see also dialogue and debate), 81–82; focusing on swing voters, 7, 92, 204; hiding important policy information, 73; issues de-emphasized with, 6, 7, 8, 69–71, 74–77, 164, 180–181 political satisfaction of citizens, 183–184 political cooperation. See conflict management; consensus/PR vision of democracy political discourse. See dialogue and debate political elites, opposed to PR, 242–243, 272 political equality, 154–155, 181–182 political gridlock, 192 political issues. See dialogue and debate political legitimacy. See legitimacy political machines, 269–270 political moderation, 214 political monopoly, 61, 95; one-party rule, 94–95; uncontested seats, 60, 61–62, 94 political oligopoly, 94–95, 99 political parties. See minor parties; multiparty systems; two-party system in the U.S.; and by party political and religious minorities: religious minorities, 138; representation

Index of, 128, 148–149, 218, 221; single transferable vote encouraging representation of, 138, 273; winner-takeall-systems discouraging, 172–175. See also minor parties; third parties in the U.S. political representation. See representation political stability. See stability politicians. See candidates for public office; legislative representatives polyethnicity, 125. See also racial and ethnic minorities poor citizens, see low income citizens popular opinion. See public opinion Populist Party, U.S., 98, 106 Portugal, PR system used in, 2 postal voting, 157 Powell, G. Bingham, 34–35, 44, 105; on majority coalitions, 171, 176–177, 199; on policy preferences of representatives vs. constituencies, 40–41, 184–185, 199; on voter turnout democracies, 153 power sharing, 171–172, 184 PR. See proportional representation (PR) preference vote. See single transferable vote (STV) Presidential elections: failure of Electoral College system in 2000 election, 44, 249; minority rule resulting from, 73, 76, 103, 161; spoilercandidate effect in, 42–43, 97. See also Electoral College; single-office elections presidential systems, 100, 193 press. See media primary system, 229, 277n.14 PR Library Web site, 305 Progressive movement, U.S. (early 20th century), 130–131, 267–268

321 Progressive Party, U.S., 98, 106; candidates elected in Wisconsin in the 1930s, 96 Proportional Representation League of the U.S., 268 proportional representation (PR), (Ch.8), 213–215, 236; hypothetical examples of status of major U.S. parties with, 66–67, 99–100, 103–104, 180, 198; international trends toward, 25, 203, 258; overview of, 16–17, 17ff., 24, 276n.12; racial and ethnic minorities better represented with, 137–143, 146–147, 209; studies of PR systems, 12, 39, 40; types of (see mixed-member proportional representation; party list proportional representation; single transferable vote); women better represented with, 114–118, 115t., 116f. See also consensus/PR vision of democracy —advantages of: debate enlivened and clarified, 105–107, 161–162, 179–180, 201, 214; fair representation ensured, 29–31, 34–35, 40–41, 50–51, 214; gerrymandering eliminated or minimized, 65–67, 121, 139; issue-oriented and positive campaigns encouraged, 78–82, 85–86, 162, 214; political equality increased, 154–155, 177–179, 181–182; political minorities better represented (see also multiparty systems), 91, 93, 165, 174–175, 214; track record of governing efficacy with, 192, 196, 207, 237; voter turnout increased, 102, 157–158, 160–161, 166; wasted votes minimized, 24, 30, 159, 214, 276n.3 —objections to, 15, 186 (Ch. 9), 194; ballots and election administration

322 proportional representation (continued) are too complex, 187, 188–191; fear that balkanization or factionalism may result, 188, 200–203; fear that coalitions are less accountable to voters, 187, 198–200; fear of extremism and fringe parties, 188, 201, 205–210, 213, 241; geographical representation is undermined, 188, 210–213; minor parties may wield too much power, 188, 203–205; multiple parties may lead to instability, 187, 192–196; the party vs. the candidate is emphasized, 187, 191 Proportional Representation Society of Australia, 237 proportional representation in U.S. history, 5, 253, 267–274 (App.C); broad party representation with, 206, 270–271; first adopted in Ashtabula, Ohio, 268; in Cincinnati, 5, 21, 143–147; in Cleveland, 5, 21, 30, 87, 120; curbing power of political machines, 269–270; effects on political stability, 271–272; effects on representation, 268–269; minimizing wasted votes, 270; in New York City, 206, 208, 241, 273–274; Progressive movement initiating, 267–268; repeal of, 272–274. See also by city PR prospects in the U.S., 255–258; constitutional obstacles regarding Senate elections, 238–239; election to the House of Representatives permitted under the Constitution, 238–239; form of proportional representation that may be appropriate, 233–234; grassroots and organizational support for, 122–124, 245, 254; local experimentation and, 253–255; legal strategies, 245–247;

Index media coverage of, 237; political opposition to, 240–243; pro-PR movement, 4, 5–6, 241, 251–253; public education important to, 236–238, 254–255; referendums in two major cities in the 1990s, 225, 254–255; state and local constitutional issues, 238–240; Web sites on, 305–306. See also electoral reform public funding of candidates, 108 public interest or broad-based policymaking, 170–171, 177–179 public opinion: on abolishing the Electoral College, 44; on ballot complexity, 190; citizen satisfaction with the political process (crossnational), 183–184; on dissatisfaction with politics in the U.S., 251–252; on negative campaigning in the U.S., 68, 82–83; on support for proportional representation in Great Britain, 4, 44–45; on support for PR system in New Zealand, 38; on third parties in the U.S., 89 quotas, for female candidates, 119, 123 race-based districting. See majorityminority districts race-neutral districting, 136–137 racial barriers: to political representation, 126, 128–131; to registration and voting, 125, 129 racial and ethnic minorities: as part of the pro-PR coalition, 145–146, 254; polyethnicity, 125 racial and ethnic minority representation, 139–140, 245–246; cumulative voting court-mandated to secure, 147–148, 148, 210, 246; increase in with the single transferable vote,

Index 138, 146, 202, 240; increase in with proportional representation, 137–143, 146–147, 209; underrepresented in at-large elections (court ruling), 245–246; underrepresented in the single district plurality system, 14, 125 (Ch. 6), 126–128, 131–133, 139, 250. See also African-American representation; Asian American representation; Hispanic representation; minority representation racial polarization: fears that PR will result in, 200–201; in voting, 136–137 racism: in gerrymandering, 132–133; historical opposition to proportional representation and, 240–241, 273. See also racial barriers Raymond, Paul Bradford, 71, 80 Reagan, Ronald, 73, 164 reapportionment. See redistricting redistricting, 52 (Ch. 2); bipartisan arrangements, 59–61; bipartisan commissions handling, 64, 65; cost and time spent in political battles over, 57–59; politically neutral, 64–66; process of in New York City, 54–56; redistricting software, 58; studies of, 65, 66. See also gerrymandering Reform Party, Canada, 4–5 Reform Party, U.S., 89, 107 regional representation. See geographic representation Rehnquist, William, 247 religious minorities, 138. See also Hasidic Jews; political and religious minorities representation, 25 (Ch.1), 27, 28–29; disproportionality of in plurality systems, 33–36, 34t., 35t., 40, 42, 96; fair representation ensured with proportional representation, 29–31,

323 34–35, 40–41, 50–51, 149, 214; greater with multiparty systems, 104–105; political equality, 154–155, 181–182. See also legislative representatives; political and religious minorities; racial and ethnic minorities; underrepresentation; women Republican Party, U.S.: cases of overrepresentation of, 36, 57, 244; cases of underrepresentation of, 27, 36, 244, 261; gerrymandering by, 56–57, 58; opposition to proportional representation, 242–243; Democratic Party platform and policies compared with, 90–92; potential support for proportional representation, 245; the “Republican Revolution,” 196; right-wing of, 92, 99; seat allocations under the plurality system, 48, 49t., 50; voter identification with, 89 —hypothetical examples featuring: of gerrymandering, 55f.2.2.; of political status with single transferable vote, 228, 266; of political status under PR systems, 42, 66–67, 99–100, 103–104, 180, 198 retrospective voting, 72 Reynolds, Andrew, 139–140 Richie, Robert, 237 rigged elections, 52 Right to Life Party, 97, 102 right to vote. See universal suffrage right-wing parties, 102 Robinson, Albert, 56 Rokkan. Stein, 138 Rule, Wilma, 115–116, 218 ruling coalitions. See coalition governments in multiparty systems runoff elections, 33; expense of, 217; low voter turnout in. See also instant runoff voting (IRV)

324 Sacramento, history of PR elections in, 268 “safe seats,” 59–61, 69; gerrymandering and, 59–63, 69, 142, 158. See also gerrymandering Sainte-Lagüe seat allocation formula, 261, 263t.A sample ballots. See ballot samples Sampson, Frank, 206 San Francisco, referendum on proportional representation in, 225, 254–255 Scandinavia, 207. See also Finland; Norway; Sweden Schumer, Charles, 83 Scotland: mixed-member PR system used in, 229; rejecting plurality elections 3–4 seat allocation formulas in PR systems, 24, 188–189, 191, 259–262 (App. A), 277n.15; d’Hondt, 260–261, 261t.A; Imperiali, 262; largest remainder, 259–260, 260t.A; Sainte-Lagüe, 261, 263t.A. See also vote counting and ballot transfer in single transferable the vote (STV) seat allocation: in the plurality system: capriciousness of, 48, 49t., 50, 277n.17; rigged elections and, 52. See also gerrymandering Seattle, interest in proportional representation, 245 semiproportional systems, 219, 221–222, 246. See also cumulative voting (CV); limited vote Senate, U.S. See under Congress, U.S. sexism, 113–114. See also feminist issues Shaw v. Reno, 134, 136 Shugart, Matthew S., 194 Sierra Club, 238

Index Simmons, Robert O. Jr., 75 sincere voting, 199–200 single-member plurality (SMP) elections, 13–14, 172, 227–228, 236; an aberrant voting system (view), 256; democratic governments turning away from, 25, 203, 256, 258; the dominant electoral system in the U.S., 2, 25, 293–294n.27; less discriminatory than at-large elections, 246; minor parties discouraged in, 11–12, 33, 95; the myth of representation and, 28–29, 42; nations using (see Canada; Great Britain; United States); public discontent with in countries using, 3–6, 251–252; underrepresentation of racial and ethnic minorities in (see also racial and ethnic minorities), 131–133, 139; two-party system encouraged by, 6; underrepresentation of women in (see also women’s representation), 8 —advantages of: geographical representation ensured in, 210–213, 214; majority turnovers more likely, 197–198, 214; political mandates ensured by, 198–200; political moderation encouraged by, 214; simplicity of, 214; single-party legislatures resulting, 214; stable governments encouraged by, 153 —disadvantages of: campaigns neglect the issues, 6–8, 69–71, 74–77, 162, 164, 180–181; capriciousness of seat allocation (see also manufactured majorities; minority rule), 48, 49t., 50; excessive policy swings may result, 196, 198; gerrymandering endemic, 52 (Ch.2), 132–133; legislatures are unrepresentative (see also legislatures; underrepresentation),

Index 46, 50–51, 214; less powersharing/more zero-sum politics, 172–174, 175, 184, 202; limited ballot choices, 88, 92–93, 94, 160, 252; low voter turnout (see also voter turnout), 6–7, 164–165; misleading mandates possible with, 196–197, 198; minor parties excluded from the political process (see also minority representation; minor parties), 11–12, 33, 95; negative campaigning encouraged in (see negative campaigning); political inequality in (see also class bias), 153, 164–165, 214; uncompetitive elections (see also incumbency), 6, 59–61, 67, 69, 198; wasted votes inevitable with (see also wasted votes), 7–8, 10, 26–27 single-office elections, instant runoff voting (IRV) and, 215–217, 218, 219 single-party majorities, 94–95 single transferable vote (STV), 20, 21b., 22–24; African-American representation and, 144–146; as a candidatecentered form of proportional representation, 191, 228, 233; can eliminate primaries, 229; countries currently using, 21; criticized as too complex, 188; currently used in Cambridge, Massachusetts, 21, 22, 31, 86–87, 120, 189, 195, 227; Droop quota in, 260; geographical representation ensured with, 212, 226, 228; history of use in U.S. cities, 21, 30, 143–147, 206; hypothetical examples of status of major parties with, 228, 266; intraparty rivalry possible with, 228; minority representation possible with, 138; negative campaigning discouraged by, 86–87; threshold of exclusion in,

325 263, 264f.B, 284n.26; voluntary constituency formation possible with, 143, 228–229; vote counting and ballot transfer in, 263–266 (App. B), 264f.B, 265t.B; wasted votes eliminated in, 159; women’s representation greater with, 120, 121–122 social class bias in voter turnout, 153, 154–155, 156 Social Democrats, U.K., 38 Socialist Left Party, Norway, 118–119 Socialist Party, U.S., 98, 106 social welfare policies, effect of electoral systems on, 177–179 South Africa: proportional representation system in, 3, 18, 203; rejection of American-style elections, 203 Spain; PR system used in, 2 spoiled ballots, number compared across voting systems, 189, 190 spoiler-candidate effect: discouraging third-party voting, 96–98; inherent to plurality elections 8, 43; instant runoff voting (IRV) eliminates, 217; in the 1997 New Mexico Congressional race, 43; PR voting eliminates, 43. See also wasted votes stability of government. See government stability state and local elections: African-American representation in, 127t; constitutional issues re: proportional representation in, 238–240; cumulative vote used in, 147–148, 190, 220, 221; electoral failures in, 250; experimenting with PR at the local level, 253–255 States’ Choice Voting Systems Act (HR 1173), 239 Stevens, J. P. (Supreme Court Justice), 135

326 Storey, Tim, 64 strategic versus sincere voting, 199–200. See also spoiler-candidate effect Stretz, Ralph, 144, 146 Stuart, Gilbert, 53 STV. See single transferable vote (STV) Supreme Court, U.S.: on fair representation, 246; potential hostility of current court to proportional representation, 247; rulings on race-based gerrymandering, 14, 131, 134, 135, 137, 140; silence on partisan gerrymandering, 64, 134, 135. See also Constitution, U.S. Sweden: proportional representation system in; PR system used in, 2; representation of women in, 110, 114 Swedish Social Democratic Party, 119 “sweetheart” gerrymander, 59–60, 94 Swing voters, appeal to, 7, 92 Switzerland, PR system used in, 2 Taagepera, Rein and Matthew Soberg, 202 Teixeira, Ruy, 154 term limits, 121, 251–52 Texas: Amarillo school board 2000 election, 149; cumulative voting in, 190, 246; electoral failure in the 2000 Congressional election, 45, 56; gerrymandering and redistricting battles in, 56–57; racist and race-based gerrymandering in, 132, 135 Thatcher, Margaret, 37–38 Thernstrom, Abigail, 200, 240 Thigpen, Anthony, 241 third parties in the U.S.: ballot access restrictions affecting, 108, 249; comprising a broad underrepresented political spectrum, 89, 92, 98, 99–100, 106–108; electoral system discourag-

Index ing, 10, 39, 89, 98, 106–108, 242; spoiler-candidate effect and, 96–98; third-party strategies, 123–124. See also minor parties; and by individual party Thomas, Clarence, 143 Thomas, Sue, 111 Thornburg v. Gingles, 131 threshold of exclusion: district magnitude and, 231–233, 232t.; number of minor parties and, 201, 284n.26; in single transferable vote (STV), 263, 264f.B, 284n.26 Timmons v. Twin Cities, 247 Toledo, Ohio, effect of single transferable vote in, 21, 268, 271 turnout. See voter turnout two-party system, 14, 88 (Ch. 4) , 102–103; candidates as the lesser-oftwo-evils in (see also candidates for public office), 6, 7, 10, 96; coalitions and factions in, 103, 193–194, 250; elites of the major parties in support of, 242–243, 272; encouraged by plurality elections, 6; limited choices in, 88, 92–93, 94, 160; low voter turnout associated with, 160, 164–165; moderation encouraged by, 214; one-party rule encouraged by, 94–95; over-similarity of parties with, 90–92; plurality system encouraging, 10–12, 28–29, 95–97; possible with proportional systems, 100; public dissatisfaction with, 251–252; reasons for persistence of, 100, 108, 242–243; Supreme Court support for, 247; voter apathy or alienation and, 12, 89, 200; voter identification with major parties low, 89, 200. See also Democratic Party, U.S. one-party rule; Republi-

Index can Party, U.S.; third parties in the U.S. two-vote system. See mixed-member proportional representation uncontested seats: in the House of Representatives, 60, 61–62, 94; in state races, 60 underrepresentation: of minor parties, 11–12, 33, 95; plurality elections leading to, 14, 25, 32–36; of racial and ethnic minorities, 14, 125 (Ch. 6), 126–129; of voters (see also wasted votes), 26, 295n.54; of women (see also women’s representation), 8, 14, 109 (Ch. 5), 114 United Farm Workers, 254 United Kingdom. See Great Britain, U.K. United Nations Fourth World Conference of Women, 124 United States (U.S.): bicameral political structure in, 193, 290n.3; limited voting choices in (see also spoilercandidate effect; wasted votes), 88, 92–93, 94–95; low voter turnout (see also voter turnout), 9, 151, 155, 157, 160; minority rule in, 36, 43–46; negative campaigns in (see negative campaigning); political spectrum and third parties in, 89, 92, 98, 99–101, 106–108; as a presidential system, 100; public dissatisfaction with politics in, 251–252; representative-constituency relationship in (see also legislative representatives), 211; two-party system in (see also two-party system), 90–92, 102–103; underrepresentation of average voter in (see also wasted votes), 33; underrepresentation of minorities in (see

327 also minority representation; minor parties), 10, 39, 242; underrepresentation of racial and ethnic minorities in, 14, 125 (Ch.6), 125, 126–128; underrepresentation of women, 109–110. See also Congress, U.S.; Constitution, U.S.; democratic values; Supreme Court, U.S. United States electoral system, 1–2, 9; manufactured majorities and, 39t., 42; myths and misleading media accounts of, 28–29; single-member district plurality elections dominant (see also single-member plurality elections), 2, 25, 293–294n.27. See also Electoral College; electoral failure in the U.S. United States history of proportional representation. See proportional representation in U.S. history universal suffrage, 26, 256–257 U.S. Taxpayers Party, 102 Virginia, electoral failure in 1998 Congressional election, 45 voluntary constituencies, 143, 228–229 vote counting and ballot transfer in single transferable the vote (STV), 263–266 (App.B), 264t., 265t. voter apathy, 27; low identification with major parties in U.S., 89; wasted votes contributing to, 27. See also voter turnout voter disillusionment or cynicism, 12, 75, 84 voter effectiveness. See strategic versus sincere voting; wasted votes voter turnout, 153–155; accountability of representatives to voters and, 154–155; causes of low turnout, 151 (Ch.7), 155–157, 158, 162–163;

328 voter turnout (continued) class bias in, 153, 154–155, 156; compared across nations and electoral systems, 151–152, 152t.; competitive elections encouraging, 159–161; electoral systems affecting, 151–152, 152t., 156–157, 164; national stability and, 153; in other democracies, 151–152, 153t., 164; party mobilization encouraging, 160, 165; proportional representation encourages, 102, 157–158, 160–161, 166 —low voter turnout, 6–7, 9; causes of, 151 (Ch.7), 155–157, 158, 162–163; information-access costs related to (see also dialogue and debate; political campaigns), 164; optimistic view of, 152–153; plurality elections associated with, 6–7, 164–165; in the U.S., 9, 151, 155, 157, 160; wasted votes affecting, 158–159. See also wasted votes Voters Choice Act (HR 1189), 239 voting, comprehension and ease of use in PR, 188–190, 214 Voting Rights Act of 1965, 131, 137, 239, 245, 247 voting rights. See racial and ethnic minority representation, universal suffrage voting systems. See electoral systems voting system reform. See electoral reform Vowles, Jack, 172, 204, 207 Wales: mixed-member PR system used in, 229; rejecting plurality elections 3–4 Washington Citizens for Proportional Representation, 5

Index Washington State, electoral failure in Congressional election, 45, 47–48 wasted votes: increase in with institution of the plurality system in Cleveland, 30; inevitable with plurality elections 7–8, 10, 26–27; low voter turnout and, 27, 158–159; proportional systems minimize, 30, 276n.4; single transferable vote eliminating, 24. See also spoilercandidate effect Watt, Mel, 239 Weaver, Leon, 86–87, 293t.15 Web sites on electoral reform, 305–306 Weimer Republic, 207–207 Wells, David, 62 Western democracies, 18, 19, 195, 223, 233 West Germany, mixed-member PR system originally used in, 20 White, Bryan, 136 Will, George, 153 Williamsburg, Brooklyn, 141 Wilson. James, 32 winner-take-all systems, 2, 15, 33; instant runoffs in, 215–217, 216b., 218, 219; majoritarian view of democracy and, 171; minority representation suffering in, 172, 223; underrepresentation of women in, 116; as zero-sum politics, 172–174, 175, 184, 202. See also at-large elections; instant runoff voting (IRV); single member plurality (SMP) elections Wisconsin: Progressive Party in, 96; electoral failure in 2000 Congressional election, 45 women’s representation: compared across electoral systems, 115t.; costs of underrepresentation of, 110–112; district magnitude in PR systems

Index and, 121–122; election of women in Cleveland with single transferable vote, 120; in European democracies, 110, 114, 117–118; incumbency factor affecting, 113, 120–121; mixed-member proportional representation benefiting, 117–118, 122; party nominations of women, 114, 118–119; party slates and Electoral College slates inclusive of women, 119–120; plurality system resulting in women’s underrepresentation, 8, 14, 109 (Ch. 5), 114; proportional representation increasing, 114–118, 115t., 116f., 122;

329 proportion of women in U.S. public offices and, 109–110, 120; quotas used to increase women’s representation, 119, 123; reasons for underrepresentation of women, 112–114; the threshold of exclusion in PR systems and, 284n.26 Women’s Campaign Fund, 123 Women’s Environment and Development Organization, 124 Zeitlin, June, 124 Zeller, Belle and Hugh Bone, 206 Zimmerman, Joseph F., 31 Zipp, John, 161