The Politics of Voting: Reforming Canada's Electoral System

The Politics of VotingReforming Canada’s Electoral System

DEnniS Pilon

York University

Toronto, Canada 2007, 2016

Copyright © 2007, 2016 Emond Montgomery Publications limited. All rights reserved. no part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, photocopying, electronic, mechanical, recording, or otherwise, without the prior written permission of the copyright holder.

Emond Montgomery Publications limited60 Shaftesbury AvenueToronto on M4T 1A3http://www.emond.ca

Printed in Canada.

We acknowledge the financial support of the Government of Canada.

Acquisitions and development editor: Mike Thompson

Copy editor: Sarah Gleadow

Production editor: Jim lyons, WordsWorth Communications

Proofreader and indexer: Paula Pike, WordsWorth Communications

Text designer: Tara Wells, WordsWorth Communications

Cover designers: Stephen Cribbin & Simon Evers

Library and Archives Canada Cataloguing in Publication

Pilon, Dennis, 1965-The politics of voting : reforming Canada’s electoral system / Dennis Pilon.

includes index.iSBn 978-1-55239-236-2

1. Canada. Parliament—Elections. 2. Proportional representation—Canada. 3. Voting—Canada. 4. Representative government and representation—Canada. i. Title.

Jl193.P54 2007 324.6'30971 C2007-903884-0

To Doris Anderson and Vera Samuelson

v

Contents

Revised Preface (2016) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix

Chapter One

Introducing the politics of Voting

introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1The Purpose of the Book . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

Depoliticized Politics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4Putting the Politics Back in . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

The Plan of the Book . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9A Final Word About “Balance” . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

Chapter t wO

why Voting Systems Matter

introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12How important is the Voting System? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

The Academic Debate over Voting Systems . . . . . . . . . . . . . . . . . . . . . . . . 13The Mechanical and Psychological Effects of Voting Rules . . . . . . . . . . . . 14The Social-Choice Critique of Voting-System Effects . . . . . . . . . . . . . . . . 15

What is a Voting System? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18The Component Parts of a Voting System . . . . . . . . . . . . . . . . . . . . . . . . . . 18What Are the options?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19Classifying Voting Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

Chapter three

Investigating plurality

introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26How the Current System Works and its implications . . . . . . . . . . . . . . . . . . . . 28

Debating the Effects of Plurality Voting . . . . . . . . . . . . . . . . . . . . . . . . . . . 30Examining Plurality Voting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

The Plurality Approach to Representation . . . . . . . . . . . . . . . . . . . . . . . . . 33Governing Majorities, Party Competition, and individual Voter

Strategy Under Plurality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48

v i / ConTEnTS

Chapter FOur

exploring proportional representation

introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49How PR Works and its implications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50

Debating the Effects of Proportional Voting . . . . . . . . . . . . . . . . . . . . . . . . 51Examining Proportional Voting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53

The PR Approach to Representation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53Governing Majorities, Party Competition, and individual Voter

Strategy Under PR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61False Alternatives: Majority/AMS Systems and Parliamentary Reform . . . . . . 70

non-proportional Voting-System Reforms . . . . . . . . . . . . . . . . . . . . . . . . . 70Parliamentary Reform . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72

Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73

Chapter FIVe

explaining Canadian and Comparative Voting-System Choices

introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75The origins and Maintenance of Canadian Electoral institutions . . . . . . . . . . 76

Debating the origins of institutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77The History of Canadian Voting-System Reform . . . . . . . . . . . . . . . . . . . . 79

The origins and Maintenance of Comparable Western Electoral institutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83

The Modern Period of Voting-System Reform . . . . . . . . . . . . . . . . . . . . . . . . . . 86Recent Voting-System Reform in Western Countries . . . . . . . . . . . . . . . . . 87Recent Voting-System Reform initiatives in Canada . . . . . . . . . . . . . . . . . 89

Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92

Chapter SIx

the politics of Choosing

introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94Debating How to Get PR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95

let the Politicians Decide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95let the People Decide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96let the Courts Decide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104Debating Who Should Decide . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105

Where Do We need PR? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106PR and the Municipal Scene . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107PR and the Provincial Scene . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111PR and the Federal Scene . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114

Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117

ConTEnTS / v i i

Chapter Se Ven

Debating the pr alternatives: StV Versus MMp

introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119STV Versus MMP: A Voting-System Grudge Match . . . . . . . . . . . . . . . . . . . . . 120STV Versus MMP: The Terms of Debate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120

Simplicity Versus Complexity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123Pro-party Versus Anti-party . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125The Quality of local Representation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126Diverse Representation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127Proportionality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130

Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132

Chapter eIght

Debating Voting-System effects

introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134The Debate About Values . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134The Debate About Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136

Plurality’s Alleged Strengths . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137PR’s Alleged Weaknesses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145

new issues: Separating Defensible from indefensible Claims . . . . . . . . . . . . . . 154Voter Turnout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154Political Culture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155Defensible Claims . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156

Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157

Chapter nIne

the politics of Voting-System reform in Canada

introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158Why the Voting System is necessarily Political . . . . . . . . . . . . . . . . . . . . . . . . . . 159The Politics of Contemporary Canadian Voting-System Reform . . . . . . . . . . . 162

The Public . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163Parties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165Media . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169Civil Society . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172Bringing it All Together . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 176

Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 177

Further resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179

glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 181

works Cited . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203

ix

Revised Preface (2016)

Shortly after his surprising federal election victory in october 2015, Prime Minister Justin Trudeau reiterated his party’s campaign promise that the most recent federal contest would be the last to operate under Canada’s traditional “first-past-the-post” voting system. Since then, various members of the liberal government have described our voting system as “antiquated,” and have suggested that the country needs a voting method that will “make every vote count,” one that will better represent the diversity of the country. To demonstrate their resolve, the government has steadfastly resisted calls to slow the reform process or submit their eventual reform proposals to a public referendum. Meanwhile Prince Edward island’s liberal government will hold, in the fall of 2016, a second referendum on changing their voting system and new Bruns-wick’s liberal government has started talks aimed at establishing an all-party commit-tee to examine voting system reforms. ontario also recently passed legislation to allow municipalities to adopt various forms of ranked balloting.

The sudden revival of interest in voting system reform caught most pundits and academic observers off guard. Given what occurred in the years after this book was originally published in 2007, it seemed that the voting system issue was like Monty Python’s proverbial parrot: if it was not dead, it was certainly sleeping, and it seemed likely to sleep for a very long time. From an initial positive vote in favour of the single transferable vote (STV) form of proportional representation in British Columbia’s referendum in 2005, everything quickly went sideways for electoral reform supporters. Prince Edward island voted down their mixed-member proportional (MMP) voting system reform later that year, ontario voters defeated a bid for a similar system in 2007, and detailed plans to pursue the issue in Quebec and new Brunswick were eventually derailed and abandoned. Federally, Stephen Harper’s Conservative minor-ity government fulfilled the previous liberal minority government’s promise to ex-amine the issue by farming it out to an anti-reform think tank that, not surprisingly, reported against making any change. When British Columbia’s second referendum on the voting system witnessed the reform proposal again fail to meet the government’s super-majority threshold (indeed, fail to match even its own previous level of sup-port), most observers declared this most recent era of Canadian voting system reform to be well and truly over.

There were many opinions about why all these voting system reform initiatives failed. Anti-reform forces declared their efforts were decisive in discrediting the pro-posed alternative voting systems, the media concluded that Canadians just liked the status quo, a host of pundits announced that nobody really cared about the issue except wonks and special interest groups, while reformers blamed a lack of education and reform rules that rigged the process to fail. Academic analysis of the various ref-erenda suggested that a clear majority of the public did support some kind of voting

x / PREFACE

system reform, but they also found that the same public was largely unaware that the referenda were going on or what they were about. As a result, the majority “no” votes were less a rejection of the reforms than an “i don’t know what this is” conservative response from voters with low information—a fairly predictable outcome given the research on direct democracy. Still, such insight does not really get to the root of why all the reform efforts failed. To revise a famous phrase: it’s the politics, stupid!

As this book made the case in 2007, voting system reform is deeply political because it affects access to the levers of political power. our voting system usually allows one party to have exclusive control of the government, with all the benefits that such a use of power entails. A shift to a different voting system that might force parties to share power is deeply worrying to political elites and the powerful interests they represent, not to mention the more partisan members of the electorate that support different parties. Thus the debates about which voting system might be “best” for British Co-lumbia, ontario, Prince Edward island, or Canada were never really about values or the performance of different systems, but about power and how switching systems might affect who gets it. As a result, the campaigns that argued against reform were stacked with recycled party hacks—often from opposite sides of the political spec-trum—that agreed on one thing: the need to maintain voting rules that rigged the game in their favour.

The first victim of such campaigns was fact-based reasoning. opponents of reform indulged in wild speculation about the possible impacts of different voting systems, fudged the facts about other countries and their voting systems, cherry-picked evi-dence to make their case, and generally tried to confuse the debate at every turn. The media, who should have been asking tough questions and helping to guide the public discussion, instead joined the pile-on, entering the debate on the “no” side and using their formidable influence to sow confusion with misleading and frankly inaccurate coverage. Reading between the lines, the media seemed to be saying that a more pro-portional voting system would allow a little too much competition into Canadian political affairs and that voters were better off with an elite-managed politics carried out by either liberals or Conservatives. They welcomed the reform defeats and coun-selled reformers to give up.

And yet in 2016 here we are again with advanced voting reform initiatives under-way in the Maritimes and at the federal level. in light of the previous disappointing results, why would politicians return to this issue once more? Again, it’s all about the politics. interest in the Maritime provinces has resurfaced because they have a real problem with the kind of election results they typically produce. Minor shifts in voting and/or differences in vote totals often lead to a dramatic overrepresentation of the government and an under-representation of the opposition. Such wonky results make the representative system appear illegitimate, weakening the government’s ability to act and have the public accept their actions. in the earlier round of voting system re-form discussion, the concern was mostly about better representing the opposition. But now, third parties are looking more competitive, which is intensifying the critical scrutiny of election results.

PREFACE / x i

At the federal level, the sustained interest in voting system reform reflects the on-going instability in the national party system. As liberal fortunes fell post-2004, the nDP increasingly promoted the issue, and this was aided by the stark under-represen-tation of Greens throughout the second half of the decade. liberal interest in voting system reform escalated dramatically when the party fell to third place nationally in 2011, with a majority of liberal MPs voting in favour of an nDP motion for propor-tional representation in 2014. When the nDP reaffirmed its support for voting system reform after becoming the official opposition, arguing that if elected to government in 2015 it would simply introduce a form of proportional representation, the liberals matched their promise, suggesting that, if elected, they too would commit to 2015 being the last election to use the first-past-the-post system. As we know, the liberals did win, and while many expected them to go quiet on their voting system reform promise, the government has instead repeatedly pledged to go forward with some kind of change to the status quo.

So what is actually going to happen? The recent past suggests that probably nothing will happen. But as this book recounts, voting systems have been changed in the past, both here in Canada and abroad. So change is possible, if the conditions are ripe. in the Maritimes, a lot will depend on whether any major party buys into reform as a good idea, and whether the rules for change are realistic (i.e., not rigged so that the process will fail, as in the past). At the federal level, a great deal depends on how federal liberals understand their electoral and governing prospects going into the future. if they believe that everything has returned to the typical Canadian “normal”—a duopo-ly in which liberals dominate and occasionally lose power briefly to the Conserva-tives—then no change is the likely outcome. on the other hand, if the liberals are worried about ongoing future strategic losses to the Conservatives due to vote splits with other parties (the nDP, Bloc Québécois, and the Green Party), then a switch to something like the majoritarian alternative vote could be in the cards as a minor re-form. That would likely shunt minor party votes back to the liberals, but leave every-thing else in the political system basically the same. Finally, if liberal voter research is telling them that their recent victory was weak and tentative, and that the decade-long instability in the party system is not over, then some form of proportional represen-tation (or a modified less-than-proportional system) might start to look attractive. After all, half a loaf is better than none, particularly if being seen at the centre of the political spectrum means that your party will likely be a participant in most governments.

Since the liberals announced their intention of going forward with their voting system reform promise, there have been three broad responses to what is going on. one approach insists that everything about changing the voting system is constitu-tional and would, at the very least, require a referendum to sanction any change. Another approach says that, since there is no perfect or “right” voting system, the issue is simply a matter of taste, and it depends on what you prefer to produce in terms of electoral outcomes. The last approach argues that voting system reform is a means to significant and necessary democratic reform and, as such, is not really up for legislative debate or a public vote, because changing from an undemocratic system to a more

xi i / PREFACE

democratic one is really the only acceptable option. The first position is fomented by the federal Conservative party and most of the media for obvious political reasons: they desperately want to avoid any change in the voting system that would open up the political market to more competition. The second position is defended by the federal liberal party and most Canadian political scientists as a stance that allows for change, if necessary, but not necessarily any change. The last position is held by the rest of the federal parties (the nDP, Greens, and Bloc Québécois), most reform advo-cates like Fair Vote Canada, and the author of this book.

The first argument, that the voting system is constitutional and would require a referendum or constitutional process to change, is simply wrong, as well as being in-ternally and logically inconsistent. Parliament has the right to change the voting sys-tem, full stop. Various attempts to argue that different sections of the constitution limit voting system reform are either historically uninformed or specious. They amount to special pleading for maintaining a voting system that is seen to benefit the Conservative party. Attempts by Conservatives and the media to seize the democratic high ground by insisting that any change of voting system requires approval by the public in a referendum fall afoul of the glaring and obvious inconsistencies of such a position given their support of phony majority governments. Are we to believe that it is necessary and democratic that at least a majority of voters must sign off on changing the voting system but that it is okay for 39 percent of voters to win a majority of the seats and have their party dominate the legislative and policy agenda for a four-year term? Some democrats.

The second position, that voting system choice is just a matter a taste, is untenable on democratic grounds. Democratic institutions should not be chosen on the basis of “preferences,” but on how well they accomplish what we know voters are trying to accomplish in using them. Voting systems that consistently frustrate a significant number of voters in their efforts to gain representation and mirror our social and political diversity while allowing a minority of votes to commandeer the majority of legislative voting power should not be on the table. You don’t get to “prefer” non-dem-ocratic outcomes. And while it may be true that there is no single perfect voting sys-tem, there certainly are imperfect ones, like Canada’s first-past-the-post system, if furthering and enhancing the democratic experience is our goal.

The third position, that voting system reform is a matter of urgent democratic reform that should be simply implemented by the government as soon as possible, is the only one that has any serious academic support in terms of comparative evidence or a historical understanding of the development of Canadian political institutions. As set out in this book, Canada’s traditional single-member plurality voting system is a pre-democratic holdover that was not designed or intended to do anything but keep power in the hands of the people that implemented it. in terms of performance it consistently fails to represent what most individual voters say with their votes, mis-represents legislative results for parties in terms of their popular support, and tends to create phony majority governments that do not enjoy the support of a majority of Canadian voters, all of which lead to all sorts of problems. By contrast, an examination of western countries that are politically comparable to Canada, and their experiences

PREFACE / x i i i

with different forms of proportional representation, suggests that we can do better, and can create more accurate and inclusive representation in our legislatures with governments that really do reflect a majority of the Canadian electorate. Critics com-plain that PR is all about giving parties what they want, but that is just empty cam-paign rhetoric. PR is about empowering individual voters to get what they want by equalizing their voting power.

What about arguments that say that changing the voting system is something that should be sanctioned by the public through a referendum to avoid self-interested parties doing what is best for them? This is either misguided populism or Conservative party talking points. Yes the nDP, Greens, and Bloc might support PR because they think it might help them electorally, but a PR system would give them just what they have coming in terms of turning their percentage of the votes into seats—no more, no less. So even if their motives were self-interested, this is a case where their self-in-terest and the public interest coincide. Conservative arguments for the status quo, by contrast, really do have little but electoral self-interest to justify them, as is made clear at various points in this book. But the bigger point is, just submitting something to a vote doesn’t make the outcome necessarily democratic. As we have seen over the last decade or more, referendums are easily manipulated by political actors who want them to fail. They are subject to information distortion, neglect, or bias from the media and those opposed to change, and this does not typically lead to reasoned and informed public debate. Furthermore, expectations of public involvement on this question are unreasonable and out of line with how voters actually cope with issue complexity. Canadians and their voting system are like intimate strangers: despite li-aising at every election, the public knows little about its workings or effects. As a result, our political parties, as representatives of the public, must lead on this question. in the end, voters will take their cues from the party they support: if their party supports reform, they probably will too; if their party opposes it, then they will likely have problems with it as well.

This book provides readers with all the basic information they need to examine and understand voting systems and what is at stake in the debate over voting system reform in Canada. it critically investigates the many erroneous and inflated claims made about our current voting system and those used in other countries. Although it was written in 2007, most of the arguments and evidence in this book still stand in terms of the basic and typical workings and outcomes of various voting systems. The most basic claim—that voting system reform is always political, indeed that it is ne-cessarily so—has certainly been borne out by the post-2007 events. And it remains true of the present moment. The federal liberal government is making all the right noises about voting system reform, and given their stated commitments for a new voting system—making every vote count, bettering the representation of diversity in the House of Commons, changing the winner-take-all culture of political interaction on Parliament Hill—some form of proportional representation appears to be the logical choice for change. indeed, only a form of PR will make every vote count by equalizing voter power. A majority of voters supported parties that were committed to voting system reform as policy in the last election. The liberals have the power to

xiv / PREFACE

simply introduce a more democratic voting system. All that remains is to discover whether there is the political will to do it.

AcknowledgmentsThis book is dedicated to two strong, inspirational women: one of Fair Vote Canada’s founders, Doris Anderson, and my grandmother, Vera Samuelson.

Many people have contributed in various ways to this book: William Chadwick, Howard Cherniack, Henry Milner, the members of Fair Vote Canada (particularly larry Gordon and June Macdonald), Mark leier and the History Department at SFU, the faculty and graduate students at York University Political Science (particularly leo Panitch, Fred Fletcher, Geoff Kennedy, and Charles Smith), the faculty in the Depart-ment of Political Science at the University of Victoria, Bryan Palmer and Joan Sangster in Canadian Studies at Trent University, and my students at York University, the Uni-versity of Toronto, Trent University, and the University of Victoria. A number of people read the book or its outline in draft form and provided helpful comments: Graham White, Bob MacDermid, Matt James, David newberry, larry Gordon, Rich-ard Bisaillon, Joyce Green, Heather Macivor, Steve Patten, Peter Russell, and Michael Stein. For providing me with materials to aid my research, i thank Hilary Pearse, Mark Warren, and Shane Gunster. Research support in terms of funding came partly from an SSHRC internal research grant from the University of Victoria, and a research grant from the Frost Centre for Canadian Studies at Trent University. Research assistance was provided by Tim Gauthier and Dustin Dunlop. Emond Publishing worked hard on this book. Thanks to Sarah Gleadow, Jim lyons, and especially acquisitions editor Mike Thompson, who steered the project from original idea, to publication, to this new re-issue.

These acknowledgements would not be complete without thanks to my husband, Dann Hoxsey, and our children Ryann and Jessie, all of whom now know much more about voting systems than the average person.

Dennis PilonYork University July 2016

1

CHAPTER onE

introducing the Politics of Voting

introductionConsider the following election scenarios:

Scenario 1: A party that has secured considerably less than a majority of the votes has nonetheless won a majority of the seats and control of the government.

Scenario 2: The party with the most votes wins fewer seats than the second-place finisher and remains in opposition while the less popular party forms the government.

Scenario 3: The governing party gets more votes than they did in the previous elec-tion and increases its percentage of the popular vote but still loses the election.

Scenario 4: one party receives more than half a million votes but wins no seats while another wins 22 seats for a roughly similar amount of support.

Scenario 5: of the millions of votes cast in the election, fewer than half are counted toward electing anyone.

These election results seem perverse, and contrary to common-sense notions of fairness and justice. They sound like just the sort of results our media report occurring in what they call tinpot dictatorships, banana republics, or unstable authoritarian regimes. But in the cases above, the seemingly irregular results are not the product of meddling by generals or apparatchiks or authoritarian leaders. All the scenarios have actually occurred in Canadian elections. Scenario 4 can be related to results in the 2004 federal election, where 582,247 votes for the Greens gained them no seats while 474,247 votes for the

2 / t h e p o l i t i c s o f vo t i n g

liberals in the Maritime provinces alone won them 22 seats. Scenario 3 sounds like the 2006 provincial election in new Brunswick, where the governing Conservatives increased their total number of votes and their percentage of the overall vote but still won fewer seats than the rival liberals. Scenario 2 can be linked to the 1996 and 1998 provincial elections in BC and Quebec, where in both cases the most popular party lost to the second-place finisher. And Scenarios 1 and 5 describe nearly all Canadian elections. We regularly witness election results where a “majority” government is won by a party that gains less than a majority of the votes. Meanwhile millions of votes, often representing as much as half of the total ballots cast, do not contribute to the election of any representatives at all—they are effectively “wasted.”

These results run contrary to Canadian expectations about election results and in-deed defy common sense. nobody goes to vote thinking that his or her ballot won’t be counted. People expect that a party with more votes should get more seats, and that parties with roughly equal levels of support should be awarded a similar number of seats. it seems reasonable that a government that gained more support in an election should be re-elected. And terms like “majority government” should mean something—in this case, that the government actually represents a majority of the people. Yet far too often our election results repudiate such common-sense assumptions. Worse still, such results are typically accepted as normal by our political elites and media, who very rarely ask any questions or draw public attention to them.

Part of the reason that election results such as these go uncontested is that Canad-ians are largely unaware of the details of our elections and have a poor understanding of the workings of our electoral institutions. This isn’t surprising when we recognize how little effort is made by our government and media to help people understand such details. After all, most election-day coverage consists of screaming headlines about a “landslide majority” for this or that party, or a televised party leader claiming to have won a “mandate” to govern and implement his or her party’s policies. But the media and the politicians seldom point out that these “majorities” rarely rest on a real majority of the votes. one key purpose of this book is to change that. The more the public is aware of how our electoral institutions work and the anomalous results they regularly produce, the greater the chance that such results will no longer be passively accepted. The point of this volume is to focus attention on the voting system as the key factor producing these distorted results and the primary institution for reform if we want to eliminate them. in drawing attention to the centrality of the voting system we will also address why this institution and its workings are typically obscured, poorly understood, and very difficult to change.

The Purpose of the BookThere are a host of books and reports on many aspects of electoral reform in Canada. in fact, “electoral reform” is a very broad topic, including just about everything con-nected with elections: the voting system, campaign finance laws, voter registration rules, the regulations governing political parties, and much more. This book will focus more narrowly on the voting system component of electoral reform debates—specif-ically, on the voting system defined as the mechanism by which votes are translated

i n t r o d u c i n g t h e p o l i t i c s o f vo t i n g / 3

into representation in our legislatures. (For definitions of various voting systems and related terms, see the section in Chapter 2 entitled “What Are the options?” begin-ning on page 19.)

Within this subtopic, there are a few books and reports that address the voting system in Canada and make a case for reform (loenen 1997; Milner 1999, 2004a; law Commission of Canada 2004 [hereafter lCC 2004]). Yet for various reasons, these works fail to make a strong enough case for reform. in criticizing them i mean no disrespect; each makes an important contribution to the larger debate. overall, how-ever, the works are simply not critical enough of many of the traditional—but largely unsupported—characterizations of contemporary and historical Canadian electoral politics and political institutions. The works also tend to ignore the politics of voting- system reform, focusing instead on the values that Canadians might wish to see in their voting system. But as will be argued here, a successful reform process requires, at a minimum, a different approach to both these issues.

This book will take a decidedly different approach in addressing voting-system re-form in Canada. The purpose of the book is twofold. First, it argues that Canada should adopt proportional voting systems for all levels of government. Drawing from Canadian and comparative examples, it makes the case that there are no compelling democratic arguments for the retention of our single-member plurality (SMP) voting system and that the needs of our modern divided polity require some form of pro-portional representation, or “PR.” in doing so, this book challenges many conven-tional and largely unquestioned assumptions about Canadian politics and political institutions in a bid to shift the terms of debate. Second, the book argues that the way reformers and academics are presently discussing voting systems in Canada is funda-mentally flawed and actually militates against the possibility of achieving any con-crete reforms. There is a curiously apolitical quality to much of this discussion, despite the obvious capital-P political implications of what is being discussed. By contrast, the key premise of this book is that decisions over democratic institutions like voting systems are fundamentally and inextricably political. By “political,” i mean that they are linked to deep and abiding differences within the polity, differences that are ex-pressed and mobilized into different movements and political parties. At one level this might appear uncontroversial, as most people would link voting and politics. Af-ter all, most of the key decisions that people make about politics in western countries involve voting. But the point here is that the choice of the voting instrument itself is also the result of politics, and that an appreciation of this fact is crucial to academic and public discussions about voting systems and the actions taken to reform them.

These two themes—broadening the current terms of the Canadian voting-system debate and making politics central to the discussion—obviously require further elab-oration. Regarding the former, we can easily point to a host of assumptions embedded within conventional discussions about Canadian voting-system reform that are ac-tually highly contestable. For instance, most commentators on reform, pro and con, accept that local representation is terribly important in Canadian elections. Yet there is precious little evidence to support this view. We also regularly hear that Canadians prefer majority government, that the plurality system allows voters to “throw the ras-cals out,” and that PR systems are consensual, among many others assertions. But all


these alleged truisms can and will be challenged throughout this book as we explore the substance of the debate over voting systems. our other theme—that voting-system reform cannot be apolitical—is perhaps more controversial, and must be sketched out more fully now to properly set up the approach that this book will pursue.

Depoliticized PoliticsThe way we vote in Canada has become increasingly contested of late, and a host of political, legal, academic, and public forums have seriously reconsidered our commit-ment to the status quo. But these discussions have tended to leave the politics of this question out of the deliberations, preferring instead to focus on values. Two key prob-lems emerge from this decision. First, inattention to the politics of institutions—to the fact that institutions themselves are the product of political contestation and competition—prevents analysts from understanding how voting systems function in practice, how and why such rules were first chosen or have been kept in place, and how any proposal for change might fare when cast into the organized realm of politi-cal mobilization and combat. This information is crucially important if people want to critically examine the conflicting claims made about voting systems or assess the realistic possibility of their reform. Second, leaving out the politics of voting contrib-utes to a discussion of voting systems where reformers try to link abstract values to institutional designs, an approach that flattens key value distinctions between the systems and curtails a critical reappraisal of the traditional voting-system debates. By contrast, a political approach would see reformers mapping the country’s real politi-cal divisions and asking how institutions might affect them, as well as critically exam-ining the possible politics behind the claims made for all aspects of our existing institutions and potential alternative arrangements. But this latter approach remains largely untried. instead, the absence of politics in recent Canadian reform discussions is pervasive, reflecting a particular scholarly understanding of both institutions and voters. Yet, as will be argued here, this view is one that is increasingly a barrier to un-derstanding voting systems and the process of voting-system change.

The consistency with which so many academics, political operatives, and reformers choose to leave politics out of a discussion of voting-system reform clearly requires some explanation. There are both normative and analytical aspects to the decision. on the normative side, some argue that effective deliberation over some issues requires us to depoliticize the process. The argument is that deliberation must be about the value of ideas, not simply a contest of numbers. if the deliberation were conducted on the basis of conventional organized politics, it would begin and end by adding up which forces had the most votes and determining their position on the issue (Petit 2004: 52–3). in a way, this view of party self-interest foreclosing any real deliberation process does accord with the history of parliamentary discussions of voting-system reform in Canada. There are many examples of opposition parties successfully raising the issue, but typically the government, with the aid of a legislative majority, has easily voted the idea down with little real discussion (Pilon 2006; Seidle 1996).

So what is the alternative? Proponents of depoliticization argue that, instead of the competitive party system form of politicized discussion, we should look to models


like Canada’s independent boundary commissions as a concrete example of how to separate deliberation from political self-interest (Petit 2004: 53). By the 1990s most jurisdictions in Canada designed their electoral maps using non-partisan boundary commissions that are “independent” from the politicians and the legislature to varying degrees (Courtney 2001: 116). These commissions are staffed by appointed judges, political scientists, and other notables who take advice from politicians and the public but ultimately make decisions on boundaries based on their deliberations about what is in the public good.

A similar logic has informed the recent discussions about Canadian voting-system reform. Reform advocacy groups like Fair Vote Canada have argued that politicians are essentially in a conflict of interest vis-à-vis the voting system as they have a vested interest in supporting the system that will advantage their party (Fair Vote Canada 2005). in other words, they can’t be trusted to take a broader view of the public inter-est. Academics too have underlined the self-interest at the heart of politician-centred decision making on such questions (Mendelsohn and Parkin 2005: 141–2). Even poli-ticians seem to think that politics should be left out of the deliberations. in setting the terms for the creation of their respective citizens’ assemblies, both the BC and ontario provincial governments expressly excluded politicians from the pool of potential members.

Analytically, some experts say that the non-political approach is simply a reflection of the reality of today’s voters. Essentially, they argue that modern voters are not di-vided politically in important and abiding ways. instead, they claim that increasing numbers of voters are alienated from the political system and do not identify with the organized political players who dominate it. of those who continue to participate and express a preference between the candidates, the majority are now “flexible partisans” who change their party vote from election to election depending on the issues. There is certainly a considerable body of cross-national research that would appear to sup-port these observations (Pharr et al. 2000; Drummond 2002; Gidengil et al. 2004). As such voters do not appear to strongly support or oppose any party, it makes sense to orient the discussion toward the kind of values that they might wish to see embodied in their political institutions—values like local representation, majority government, accountability, representativeness, and stability. Both these normative and analytical insights have informed the recent popularity of using either expert commissions or values-driven citizens’ assemblies to deliberate over the advisability of voting-system reform.

Attempts to depoliticize the reconsideration of democratic institutions, however, have ultimately proven unsuccessful. Efforts to deny politics—understood as party or politician-centred politics—do not so much depoliticize as differently politicize. And this different politicization has had an impact in terms of the participants’ understand-ing of how politics and institutions work. it has contributed to an undifferentiated “people versus politicians” view of politics, which tends to provide an all-purpose answer to political problems amounting to “politicians don’t care” or “parties are too powerful.” This has the effect of foreclosing important lines of inquiry, like the more complex political reasons behind the origin of our democratic institutions and why they may have been kept in place or reformed at different points in history (Pilon


2006). of course, in this, the recent trend in public deliberation mirrors a longstanding practice in the study of Canadian political institutions that has also tended to assume rather than explore the reasons for our use of different institutional arrangements. For instance, many scholars assume that our present use of single-member ridings somehow reflects the public’s preference for local representation. But this ignores the fact that, historically, multi-member ridings have been used extensively in most Can-adian jurisdictions (Qualter 1970: 118–23; Siaroff 2006: 200–3).

Another impact of this different politicization is that it has tended to sever the consideration of institutional values from the concrete political realm, contributing to a mismatch between the laudable goals of the reformers and political reality in terms of organized political forces and their interests. For instance, in 2004 the BC Citizens’ Assembly (BC-CA) decided in favour of the single transferable vote (STV) form of PR, which was opposed by all three of the province’s major parties and most key groups in civil society, over the mixed-member proportional (MMP) form of PR, which was favoured by two of three major parties and many forces in civil society, including most of the public submissions to the BC-CA itself. Even in terms of the deliberation process, the values approach tends to limit critical inquiry precisely be-cause it converts what should be seen as potentially questionable attributes of different systems into morally equivalent value commitments. Thus local representation is characterized as a question of value (i.e., do we like or prefer it) rather than as some-thing that should be investigated and assessed in terms of its real functioning within the political system (i.e., does it really matter in political outcomes).

Putting the Politics Back InWe cannot really depoliticize politics—nor should we want to. if anything, the prob-lems that have led to calls for voting-system reform relate to the false narrowing of political interests and debate. Thus we need more politics, not less. This book argues that deliberation over possible voting-system reforms must be made within a context of appreciation for political differences. At all costs, we must avoid the current tenden-cy toward a populist anti-politics, which really amounts to a denial of the existence of political differences among Canadians. in populist rhetoric, political divisions are typically blamed on self-interested outsiders, the inference being that most political problems could easily be solved if it weren’t for interfering politicians and political parties (Gibbins and Stewart 2002: 87–8). Yet despite public criticism of political par-ties and conventional conflict-oriented political discourse, our parties and the tenor of our political discussions do reflect real and abiding differences about politics among Canadians.

in loose ways, Canadians can be arrayed on a continuum of left to right in terms of ideas about society, values, the role of government, and economics. Some favour more social services, some less. Some are for public broadcasting, some are not. Some approve or will tolerate gay marriage, while others oppose it. Some believe the economic system is fair, while others complain it is biased toward the already privileged—and so on. The fact that these differences may not correspond neatly to existing political parties or remain wedded to the same party choices in each election does not make


them any less real or abiding (nevitte et al. 2000: 48–57; Blais et al. 2002: 106). Some academics would argue that, while recognizing these differences, we should still en-courage a depoliticized process so that deliberation participants can pursue a result that best suits the “public interest” (Mendelsohn and Parkin 2005: 143–4, 146). But all decisions affecting public policy or institutional design help some and hurt others, which means that there is and can be no “unified” public interest. This is just another way of denying politics. As there really is no place beyond politics, we should just be as frank as possible about how politics enters into this particular discussion.

Here too there may be debate about exactly how politics is part of the discussion on voting-system reform. After all, although voting in the 2005 BC election split 46% to the right and 42% to the left, nearly 58% of the same voters endorsed the STV vot-ing system. obviously, strict party lines did not determine all of the voting. Yet there are two ways in which politics affects such choices. While it may be true that many voters do not initially connect particular institutional choices with their political in-terests, as deliberations over institutions like the voting system move forward many voters will likely seek to understand any potential changes in these terms, and perhaps seek direction from existing political parties or organizations in civil society. More to the point, even more than voters, politicized organizations within civil society (unions, business lobbies, think tanks, etc.) and political parties will quickly examine any pro-posals for change in terms of how they understand what is in their political self-interest and seek to influence their supporters on the question. indeed, this is just what appears to have occurred in BC in the period before the referendum vote (Ratner 2007).

To take seriously the politics of voting is to set aside the current focus on values as a means of judging voting systems in favour of exploring the political rationales and forces behind institutional choices, both past and present. Rather than speculating about the values that allegedly informed the initial adoption of different voting sys-tems or that have allegedly kept them in place, we need to assess the actual political reasons why such arrangements were introduced and maintained. instead of conduct-ing discussions about an ideal Canadian polity, we would be better served by investi-gating how our politics is concretely organized in the here and now, as this reality will condition the reception of any proposal that is generated by a deliberative body and ultimately affect its functioning if it is accepted. Furthermore, we need to understand the politics behind the arguments for or against change, or between supporters of different forms of PR, or about where and how to seek reform, or concerning the composition of the reform and anti-reform campaigns. Such strategic information can tell us a great deal about why reform initiatives succeed or founder. Thus understand-ing the politics of voting—that institutions are embedded in a process of political contestation—is crucial for a successful deliberative process that might contribute to change. ignoring politics means ignoring the space where the deliberative results will emerge and where they will ultimately be struggled over.

Another reason to set aside the “values” approach is that it tends to limit critical scrutiny of all aspects of the voting-system debate. By converting issues like majority government and local representation into value questions, deliberators shift from a critical exploration of what these things mean and how they function within the po-litical system to a normative assessment about whether they like or prefer them.


Whether one likes local representation may be irrelevant if it can be demonstrated that it plays no important role in politics. And if it is irrelevant to political outcomes, it should certainly not be allowed to detract from other factors that may be demon-strably central to political contestation (such as the representation of parties). Perhaps one of the most serious problems with the values approach is the way it flattens all values into moral equivalents. Thus stability and majority government are offered as potential values informing voting-system design equal to proportionality and coali-tion government. As voting-system analyst David Farrell suggests, the choice between them is “ultimately … a normative judgment call” (Farrell 2001: 12). Such an ap-proach ignores how the normative bases of these different values may be very different, with some falling far short of democratic norms.

The politics of the “norms” debate is itself very interesting. For a long time, north American political science was wholly hostile to proportional representation (Rustow 1950: 107–8). Part of this reflected American arrogance that US institutions were, by definition, how democracy should be designed; another part reflected realpolitik opposition to the kind of results that emerged from western European PR systems, especially where the communist left was strong (Pilon 2005). But by the 1980s and 1990s, particularly after the outbreak of voting-system reform in western countries, the consensus among north American political scientists had changed. The one-dimensional defence of plurality shifted to a view that essentially claimed that both PR and plurality were fine, depending on the values and circumstances of the polity where they were to be implemented. As Richard Katz—the most cited author on this position—has claimed, “There is no universally correct, most democratic electoral system” (Katz 1997: 308). This has led many analysts to claim that there is no basis for arguing that plurality is undemocratic—it is just a matter of different values. indeed, G. Bingham Powell argues that the majoritarian and proportional approaches to elec-tions are the “two great approaches of democracy” (Powell 2000: 21).

This particular “values” issue is worth exploring in more detail. in theoretical form, the majoritarian vision defends overrepresentation of the majority view in elec-tions so as to assure government stability, effective policy implementation, and clear accountability between government and the electorate. But there is an important slip-page between the theory and practice of the majoritarian vision under plurality: few legislative majority governments in the system ever get a majority of the votes cast. Even defenders of the majority view admit that, normatively, the approach is on shaky ground for just this reason (Powell 2000: 129, 237).

Given these general trends, plurality’s supporters are left arguing that a voting sys-tem that regularly produces majority governments from the largest voting minority is a value that can be defended as democratic in the design of a voting system. Basically, they seem to be arguing that if enough people like a system that distorts election out-comes by awarding the largest minority views with a majority of seats, then that is oK. A further implication of this position is that if the largest minority opinions—say, the two largest political parties—collude to keep others out, then that is oK too. While such views are regularly defended, the values undergirding such positions are difficult to defend as democratic ones. They fail in terms of effective representation of our political divisions by over- and underrepresenting different positions. They also


violate the notion of majority rule by awarding an electoral minority the majority of legislative voting power.

The politics of such value orientations should be clear: those who would highlight stability and majority government as the key values in designing electoral systems at the expense of representing our political differences favour values that limit democ-racy and political contestation. The fear of democracy embedded in such views is well established, from anti-democrats like Tocqueville to reluctant conservative democrats like Schumpeter (Femia 2001: 46–60, 83–4). indeed, in defending the majority “vision” of democracy, scholars typically produce a list of supporters with decidedly undemo-cratic pedigrees—James Madison, Walter lippman, and William Riker, among others (Powell 2000: 10–11; for critics see Mackie 2003, Roper 1989, and levin 1992). So the choices among the “values” are not morally equivalent; some values can be readily defended as democratic while others cannot. only by examining the politics behind these value choices can we recognize this significant difference.

it is only once we have put the politics back into our analysis of voting systems—in other words, recognized the divided nature of the Canadian polity and how that has affected and will affect deliberations over institutional change—that we can return to the question of values. in this book, a strong case will be made that plurality voting limits both the expression of our political differences and the degree of political com-petition open to Canadians, and thus violates key democratic values. Furthermore, attempts to justify plurality based on the alleged need for majority government and unproven fears of instability will be shown to be unconvincing and rooted in undem-ocratic values. By contrast, it will be argued that we can base our democratic institutions firmly in democratic values by adopting some form of proportional representation.

The Plan of the BookWith this brief introduction to the book and its approach, the focus now shifts to the debates over voting systems and voting-system reform in more detail. Chapter 2 ad-dresses the debates over voting-system effects, specifically why different experts think that the voting system is either inconsequential or central to political contestation and election results. The chapter then explores how different voting systems work in some detail, including an examination of their component parts. Chapter 3 reviews the claims made both for and against our present plurality system, touching on prob-lems of representation, the lack of authentic majority governments, the stifling of electoral competition, and the incentives for negative strategic voting. Chapter 4 ex-plores the potential of proportional representation in terms of representing voters, parties, and our social diversity; producing more intuitive majority governments; providing for a more competitive electoral politics; and creating opportunities for qualitatively different forms of strategic voting. Chapter 5 sets out the historical roots of Canadian and comparative voting systems, highlighting how early conversions to PR coincided with struggles for democracy itself, and sketching out the different eras of Canadian voting-system reform and the changing political and social contexts that informed them. Chapters 6 and 7 examine the “politics of choosing” in three different areas that are relevant to the campaigns for voting-system reform, specifically debates


over how to seek reform (legislative, referendum, courts), where to seek reform (civic, provincial, federal), and what kind of reform to seek (STV vs. MMP). Chapter 8 criti-cally engages the debates over voting-system effects, especially as they are typically cast in public debates over voting systems and with particular attention to the common assertions made in defence of plurality voting and against PR systems. The chapter then turns to some of the new issues that have been raised in connection with this debate, such as the potential effects on voter turnout and political culture. Finally, Chapter 9 sums up the arguments and key evidence of the book, assessing the possible paths and barriers to changing our voting system while noting potential areas of support and typical sources of opposition. Depending on the reader’s interest, the chapters may be read out of order and certain sections—as they may be more or less interesting or rel-evant to different readers—may be skipped altogether.

Methodologically, this book approaches the debate over voting systems and voting-system effects concretely and comparatively, with specific attention to western countries. Many existing texts address the possible effects of voting-system change for Canada speculatively (i.e., arguing that the adoption of PR could lead to X result or Y result, etc.). But speculating in a vacuum about what might or could happen in Canadian politics under different voting rules is a worthless exercise, as it gives us no real insights into what the impact of different institutional arrangements might realistically be. A better way to assess how voting-system change might impact Canada involves engaging what actually happens in both the Canadian and relevant comparative contexts. This means examining the details of Canadian election results and political mobilization to assess what the effects of our present voting rules are, and attending to the effects of different voting systems in other countries—specifically, countries whose political contexts can be realistically compared to that of Canada.

Though it has become increasingly popular in comparative political science to com-pare countries regardless of their differing patterns of democratization and economic development (Shugart and Wattenberg 2001), there are serious problems with such an approach (for a discussion of these problems see Pilon 2005). As such, this volume will limit its comparative examples to other western countries whose patterns of democratic and economic development are somewhat similar to those of Canada (similar at least when compared to the very different patterns in developing countries, for instance). This approach allows us to assess the range of possibilities that might realistically emerge from change. For instance, while some speculate about how the adoption of PR could lead to X or Y result, our concrete and comparative approach seeks to understand what has actually occurred in countries using different voting systems. of course, comparative experience does not exhaust the possible effects of change; something new might emerge in Canada. But examining concrete experience at least gives us something solid to assess, while speculation cannot be critically assessed at all.

A Final Word About “Balance”This book represents a “strong case” approach to addressing the debate over voting-system reform in Canada. it sets out from a particular normative position on democracy, moves on to examine how well our current voting system matches up with that position,


and then explores which systems might provide a better match. is taking a normative stand on an issue a problem for political scientists? not if the practice of modern Canadian political science is anything to go by. After all, most political scientists nor-matively favour democracy over other ways of organizing politics.

There are plenty of examples of normative positions being defended by Canadian political analysts, including those focusing on questions of institutional arrange-ments. For instance, John Courtney does not merely analyze the origins and effects of independent boundary commissions as a way of designing constituencies for Canadian elections—he actively endorses them as a better way of doing democracy (Courtney 2004: 71). in other words, he defends them in normative terms, as well as in terms of his analysis of how they work. it is important to underline that normative positions are seldom taken up without an analysis of how things work in practice. To make a strong case for PR in this book, for example, would have little persuasive effect if the normative commitment did not reflect a substantive engagement with the facts, both pro and con. Thus one can make a strong case for something and argue from a par-ticular position but still present a balanced and objective treatment of the subject matter. in fact, this is just what a number of previous volumes arguing for voting-system reform in Canada have done (irvine 1979; loenen 1997; Milner 1999, 2004a).

Still, some who focus on balance insist that all positions should be assessed in terms of their strong and weak points. The point that will be made in this book is that the commonly accepted strong points in favour of plurality do not hold if we require them to measure up to democratic norms. The focus on balance also ignores how making a strong case for an issue can sometimes help clarify what is at stake in a de-bate, or shift the terms of the debate into new and potentially more insightful areas. More seriously, a fetishization of “balance” can actually lead to an unbalanced and unfair debate because it ignores the advantages that one side may already enjoy—spe-cifically, the status quo. in the case of voting-system reform in Canada, plurality is the hegemonic position and as such enjoys the advantage of seeming “normal” and “proper” simply by virtue of its long use. Thus any fair discussion must actually spend more time making the opposing case, precisely to overcome some of the unearned advantages enjoyed by the status quo.

As noted above, this is not the first book to argue for voting-system reform in Canada. What distinguishes this book from the others, however, is its insistence on the politics of our voting-system arrangements and the need to introduce this politi-cal dimension into our present discussion of voting systems and their reform. in the end, each reader will judge the strength of the arguments put forth here and make up his or her own mind.

12

CHAPTER TWo

Why Voting Systems Matter

introductionDo voting systems matter? Do they influence the choices that voters make? Do they shape the political outcomes that emerge from their use? in debates over voting-system reform, these questions are crucial. if the answers to these questions are “no,” there is no point in changing voting rules. But if they are “yes,” the politics of voting-system choice become apparent and the need to debate which voting system should be used becomes unavoidable. Here, the debate can tend to assume an “all-or-nothing” charac-ter, with reformers blaming the voting system for all the ills of politics while defenders of our current system claim it has no real impact on the results. neither of these ex-tremes, however, can be sustained.

This chapter will make the case that the voting system does matter, that it does have effects, and that its impact on political results—though more subtle than the traditional critique—is no less important. After establishing that voting systems matter, we’ll explore the range of systems in use in western countries and review the details of how they work, with particular attention to their component parts.

How important is the Voting System?The debate over voting systems is really a debate about voting-system effects. in the Canadian context, proponents of change assume that the adoption of a different sys-tem would allow voters and parties to act differently. Thus it is not merely that different voting systems count votes differently, but that different voting-system arrangements alter the incentives that voters and parties face to do certain things: to vote for one

w h y vo t i n g s y s t e m s m at t e r / 13

party over another, to try to appeal to one group of voters over another, and so on. Defenders of plurality also highlight the potential effects of changing our voting system, painting a gloomy picture of minority or coalition government, a proliferation of parties, and a general increase in political instability. The substance of these claims will be taken up in subsequent chapters; first, we must address a more elementary question, one that the pro and con sides may overlook—that is, whether we can say that voting rules have any real impact on elections results. Academics fall across the range of debate, with some highlighting a strong impact of voting rules, others sug-gesting more mild effects, and some claiming that voting rules have no discernable effects at all.

The Academic Debate Over Voting SystemsAcademics have long debated the importance of voting systems. on one side, some felt that voting systems were largely a result of or a response to other social and political forces. lipset and Rokkan argued that different voting systems, following party systems, emerged from differing patterns of social and political cleavages in modern societies. Thus societies with few social cleavages utilized plurality or majority voting systems while societies divided by multiple social cleavages opted for proportional voting sys-tems. in this model, the voting system is an effect or result of other political forces (lipset and Rokkan 1967; Rokkan 1970). on the other side, some argued that voting systems conditioned or created the party systems. Duverger proposed that the voting system was the key causal factor in determining the nature of the party system in a given locale. Thus the use of plurality or majority voting contributed to the existence of a two-party system while the use of proportional voting allowed a multi-party system to emerge. in this model, the voting system is the cause of the political results, such as the kind of party system that is produced (Duverger 1954). While a great deal of de-bate has ensued over these points, they are not in fact mutually exclusive. it is possible both that social factors contribute to the adoption of a particular voting system and that, once adopted, the voting system conditions the nature of political competition.

Few serious scholars today hold to either extreme view—i.e., that voting systems have no effect on political results or that voting systems are the key factor affecting political results. instead, the real debate is about the relative weight of voting-system effects and whether the trade-offs produced by any given system are worthwhile. Thus some admit that Canada’s use of the single-member plurality system does exert some influence on the results of our elections, but add that it is just one factor among many and as such not terribly important. others argue that our use of the SMP system has not prevented new parties or ideas from emerging in Canada when these became suf-ficiently popular (Courtney 1999). or they suggest that SMP clearly affects who and what is represented in our legislatures, but in the end this is not very important as it has little impact on how our executive-cabinet form of government operates in Par-liament (Aucoin and Smith 1997). Some point to the high degree of turnover in the membership of the House of Commons, suggesting that this proves that our current system is already quite responsive to what voters want (lortie 1997), while others argue


pragmatically that changing political structures will be very difficult and only take re-sources away from more important challenges to the substance of politics (Jenson 1997). in all of these defences, there is a bias that political organization, mobilization, or the countervailing influence of other institutions can and ultimately will trump any of the admitted effects of the voting system.

The Mechanical and Psychological Effects of Voting RulesSuch defences ignore the ways in which voting-system effects are dynamic, shaping not just static results but the very process of contestation from which results emerge. Duverger characterized it as a combination of mechanical and psychological effects (Duverger 1954). Mechanical effects refer to the observable qualities of a voting system. Plurality, for instance, will reward voting support that is concentrated geographically over voting support that is more spread out. in Canada, this has meant that regionally concentrated parties like Social Credit and the Bloc Québécois tend to be overrepre-sented while more regionally dispersed national parties like the nDP are consistently under represented. A PR system, by contrast, will tend to respond to voting support more equally regardless of where the votes are cast.

Such effects are easy to track and quantify, but voting systems also have psycho-logical effects. For instance, because plurality systems tend to discriminate against voting support that is dispersed, some voters may decide not to support political op-tions that they fear are not popular in their geographic area. of course, the end result is a self-fulfilling prophecy. Because people fear that their choice may be unpopular in their area, they vote for something else, thus helping to make their fear a reality. These sorts of psychological effects, though more difficult to track and prove than mechani-cal effects, are no less real. Thus the fact that new parties emerge in Canada from time to time, or that a great many MPs fail to get re-elected, does not vindicate our present electoral arrangements. neither fact tells us very much about the impact of our current system. Both results could just as well be characterized as indications of the system’s failure, with the past two decades of party decline and anti-politician rhetoric serving to illustrate the public’s frustration with trying to make the current system work.

An appreciation of these mechanical and psychological effects leads to a shift in emphasis about the relative weight of the voting system in affecting political results in Canada. All the stuff of politics—culture, organization, issues, mobilization—mat-ters, but gauging how these factors interact in the context of the voting system is also crucial to understanding Canada’s political system and its results. The importance of this can be seen in a host of areas. First, the voting system affects the incentive struc-ture for both parties and voters. Both make decisions about how to act based on how effectively they think they can get their message across and achieve results through the existing institutional arrangements. Second, the voting system affects the kind and degree of diversity that is represented in our legislatures, from the diversity of party viewpoints to the social diversity embodied in the elected members. How accurately the election results reflect what the public wants, or whether parties make an effort to represent everyone in society, is crucially influenced by the choice of voting system.


Finally, the voting system is key in facilitating or hindering a clear dialogue between citizens and their representatives. Election-day results can either clarify or obfuscate what citizens want on the table, depending on how their votes are translated by the voting system. These effects will be explored in more detail in subsequent chapters.

The effects of voting systems, then, are less about the production of specific results than about affecting the nature of political competition. This cuts across both the strong and weak “effects” positions in the academic debates. For instance, the strong position held that PR would lead to a multi-party system and plurality to a two-party system, while the weak position argued that demands from society would overcome institutional constraints. The evidence over the last 50 years is that a number of PR-using countries have had party systems with just two or three competitors (Germany, Austria) while a number of plurality systems have had multi-party systems (Canada, india). This would seem to support the weak effects position. on the other hand, we can point to recent changes in voting systems that did seem to allow for an expansion of the party system, particularly in the conversion from plurality to PR, and some dy-namic changes in the social diversity of the representatives (new Zealand). This would appear to support the strong effects position. in fact, neither position captures what is going on. What we need instead is a model that combines social and institutional factors, recognizing that while a change in institutions may affect how politics works, it does not produce any specific results on its own. As Peter Mair notes, voting systems “provide at best ‘facilitating conditions,’ the impact of which will also be mediated by a variety of other institutional cultural factors” (Mair 1992: 85). in a nutshell, voting systems are like the aperture through which politics tries to flow—they are basically more or less permissive, more or less competitive. The dam affects the flow of water, but not if there isn’t any water flowing.

The Social-Choice Critique of Voting-System EffectsDespite the preceding discussion, a consensus has not been reached on the effects of voting systems. Recently, an influential body of work emanating from economics has argued that regardless of what people may think they are doing with voting, it cannot be demonstrated that voting allows them to accomplish what they set out to achieve. Social-choice theorists claim to have shown logically that no voting system can prop-erly or fairly aggregate what voters are saying with their votes (Arrow 1963; Riker 1982). if these theorists are correct, then attempts to change the voting system are pointless. in fact, in the extreme position, voting—and by extension democracy it-self—is pointless. There are a number of aspects to this argument. The basic premise is that, any time there are more than two choices available, there is no fair or accurate way to pick a winner.

if we take a case where a group of 12 diners must settle on one option as their col-lective choice for dessert, with a choice of pie, ice cream, or cake, it may turn out that no single option emerges as the consensus for the group and that no single option even gains a clear majority of the group’s preferences. What happens now? one way to resolve the issue would be to ask the diners to rank the options from their most to


least preferred. in the typical social-choice case scenario, the preferences of the group would look like this:

4 diners prefer: p > i > c

4 diners prefer: i > c > p

4 diners prefer: c > p > i

What has emerged is called a “cycling” of preferences. Depending on where we start counting, a majority prefers pie to ice cream, ice cream to cake, cake to pie, and so on. What this means is that there is no single result that appears to be decisively more popular than the others. So, which option would win? According to social choice, it would depend on who could manipulate the agenda of how the votes were cast. For instance, pie is less popular than cake but more popular than ice cream. As ice cream is more popular than cake, the pie partisans would want a vote to take place between ice cream and cake first. Then, in a showdown between pie and ice cream, pie would win. if the first vote involved pie versus cake, however, then cake would win, though it would lose in the next round to ice cream. Between the two versions of the vote, ice cream has gone from the first-eliminated loser to last-standing winner, yet the order of the voters’ preferences has remained unchanged. From this example, it is clear that who wins is rather arbitrary, depending entirely on the order in which the vote is tak-en. Social-choice analysts argue that cycling and the strategic manipulation of the voting process are in fact the norm in democratic settings—an argument which, if correct, would suggest that voting results are arbitrary and meaningless (Riker 1982).

The above debates are not just of academic interest to the mathematically minded. no less a figure than Canada’s chief electoral officer has drawn on the notion of cy-cling to argue that there is no “perfect” voting system and thus no pressing need to address voting-system reform (Kingsley 2001). Furthermore, a number of academic critics of the reform movement have chided those calling for “fair voting” by referring to what they consider definitive proof from social-choice work that no such system can be designed (Kam 1997; Flanagan 2005: 99–100). if these critics are correct, then reforming the voting system would indeed not be worth the effort.

The logical study of the implication of voting rules is an old tradition, much older than the century or so of voting-system research. it stretches back to the French Revo-lution, and since then has re-emerged with force seemingly every half-century. Con-tributions have come from the politicians of the French revolutionary period (Condorcet, Borda), the writer of Alice’s Adventures in Wonderland (lewis Carroll, writing under the name C.l. Dodgson), and more recently from nobel-winning math-ematicians (Kenneth Arrow, Amartya Sen). Considerable advances in what is typically called “voting theory” have regularly emerged over the past half-century. American political scientist Gary Cox recently lamented that the two traditions of voting study—voting theory and voting-system analysis—have rarely had much dialogue. Cox’s book Making Every Vote Count sought to redress that and, since its publication in 1997, a raft of social-choice approaches to voting systems have emerged (Cox 1997; for an earlier effort see Dummett 1984).


But Cox’s sophisticated social-choice analysis of the strategic coordination prob-lems facing voters and parties through voting systems has seldom been matched. There was a reason why voting theory typically attracted little attention from those studying real voting systems: they were talking about fundamentally different things. Voting theory deals with how voting is conducted in committees. it involves questions where only one option may be chosen from among many. Voting-systems analysts study how voters make collective decisions about representation for an entire legislative body. By its nature, such voting cannot have just one winner. in the example of the diners choosing a dessert, the committee decision requires that one option win out. in a representative election, the partisans of cake, ice cream, and pie might all see their choices reflected in the outcome. of course, if the representative system uses the single-member plurality system and only one winner can emerge from each riding, it does begin to resemble the conditions of the committee voting. But the committee condi-tions do not hold here, either. Due to the left–right dimension of real politics, voting choices among parties are not likely to be as circular as the choices between desserts and thus cycling is highly unlikely.

Recently, the social-choice approach to voting has come under a devastating critique. Critics had long pointed out that the problem of cycling was actually just a logical possibility. That preferences could cycle does not mean that they must or even will on most occasions. in fact, in a recent analysis of a host of supposed examples of cycling in the US Congress, no evidence of cycling could be discovered at all (Mackie 2003). Even as a theory of committee behaviour, social choice is wide of the mark. Few real committees operate the way that the committees imagined by voting theorists do. in real committees, cycling is rare because people deliberate, negotiate, and change their positions based on the actions and reactions of others (Buchanan 1954: 118–21). in fact, this is why so many students of Canadian legislatures have kind words for the work of legislative committees even when they are highly critical of other aspects of legislative behaviour (Docherty 2005: 165–72). Social choice and voting theory, with all their logical proofs, are simply too removed from the actual practices of democratic interaction to be of much use in assessing the concrete effects of voting systems.

Probably the strongest evidence supporting the view that the voting system really is important is the historical record of just how hard such rules have been to change and just how closely political elites guard any decisions about altering them. Despite a recent spate of reform in western countries since the 1990s, voting systems are typically very difficult to change. From the 1920s to the 1990s only one western country—France—changed its national voting system. Even the recent reforms in Japan, new Zealand, italy, and the UK involved protracted battles in which largely unwilling po-litical elites attempted to dodge the reform bullet (Pilon 2005). in Canada, too, the recent provincial discussions of voting-system reform bear the marks of a conniving political class bent on resisting reform. in each case, the party in power mouthed platitudes in favour of open discussion on the issue but then attempted to derail the process by insisting on super-majorities for any change to succeed. Clearly the politi-cians—regardless of what they might say for public consumption—believe the voting system is extremely important, or they wouldn’t work so hard to prevent any tamper-ing with the system that has put them in office.


What is a Voting System?With some sense that voting systems are important, we now turn to examining just what constitutes a voting system and exploring how voting systems work. The voting system comprises the subset of electoral rules that determine how votes cast in an election will be translated into representation. At the same time, the voting system is part of a larger electoral system, which encompasses all aspects of election adminis-tration—rules for candidacy, campaign finance regulations, voter registration, balloting rules, etc. There is often confusion on this point, as some commentators use the terms “voting system” and “electoral system” interchangeably or fail to distinguish vote-counting rules from other aspects of election law and electoral practices. Ultimately, the definitions of such terms are a matter of convention and shifting academic and public discourses. For our purposes, the “voting system” will refer exclusively to ques-tions of vote counting and representation.

The Component Parts of a Voting SystemThe rules of a voting system guide the actions of both the voters and those responsible for counting the ballots. The voting system sets out how voters should mark their ballots, how those votes should be aggregated before counting them, and how the winners will be determined. These three broad areas of concern are referred to as bal-lot design, districting, and voting formula (Rae 1971).

Ballot design involves how voters are to mark their ballots. Some ballots call for voters to mark their choice nominally, with either a check or an “X” next to their choice. others ask voters to mark their choices ordinally, ranking the candidates in terms of first, second, and third preferences, and so on.

Districting refers to the geographical area within which the votes will be gathered and counted as well as to the number of candidates to be elected. Sometimes the district is the whole country—as in israel—but typically districts are smaller, representing particular defined areas. Such districts may elect only one member, or two, or ten, or thirty, or more.

Voting formula refers to how the aggregated votes will be added up to determine a winner or winners. in some cases, the winner must simply gain a plurality of the votes, or more votes than any other individual candidate. in other cases, the winner must receive a majority of the votes cast. Finally, in still other cases the formula allots representation in proportion to the votes cast for different competitors.

As can be seen, choices must be made within each component element of the voting system. When these factors of ballot design, districting, and formula are combined in different ways they produce different voting systems. it should be noted, however, that certain design elements are incompatible with others. For instance, a plurality formula cannot be combined with a preference ballot structure, nor can a party-list PR system be instituted in a single-member district. This is where the “system” part comes in. The component parts of a voting system do not work in isolation, but rather have their effect through their systemic interaction.


Voting-SyStem ComponentS

plurality majority proportional

Districting single/multi-member single/multi-member multi-member/mixed

Ballot design nominal nominal/ordinal nominal/ordinal

Formula plurality majority proportional

What Are the Options?There are essentially three major kinds of voting systems: plurality, majority, and pro-portional, with a smaller category of semi-proportional systems that do not fit neatly into the typology.

major Voting SyStemS

Voting-system family Voting-system variants Voting-system use by country

plurality single-member plurality Canada, United States, United Kingdom

multi-member plurality Canada, United States (some local elections)

majority double ballot France

alternative vote australia (lower house)

proportional party list norway, Sweden, Finland, Denmark, Belgium, netherlands, italy

single transferable vote ireland, malta, australia (Senate)

mixed-member proportional

germany, new Zealand

Semi-proportional limited vote United States (some municipal elections)

cumulative vote United States (some school board elections)

single non-transferable vote

japan (1947–1993)

parallel japan, italy (1994–2004)


PlURAliTY VoTinG SYSTEMS

Plurality voting systems combine single- or multi-member districts with the plurality voting formula and a ballot structure calling for a nominal “X” to indicate the voter’s choice or choices. To gain election under this system, the winning candidate need only have more votes than any other individual candidate. if just two candidates are running, the winner will have more than 50% of the vote. But if three candidates are running, the winner could succeed with as little as 34% of the vote. Such results are not merely a statistical possibility. in the 2004 federal election, one candidate won his riding with just 27% of the total (Saskatoon/Humbolt). Canada and the United States use single-member plurality for most elections, though both have used multi-member plurality for national and subnational elections in the past. Multi-member plurality has fallen out of favour in recent years, though it is still used in a few municipal appli-cations such as Vancouver’s ten-member city council. in a multi-member plurality election, voters have as many votes as there are candidates to be elected, and winners are the top vote-getters. For example, in Vancouver, voters have ten votes and the elec-tion winners are the candidates with the top-ten vote totals. The UK used plurality exclusively in the past, but now uses it only for national elections and for local elec-tions in England and Wales.

MAJoRiTY VoTinG SYSTEMS

Majority voting systems combine single- or multi-member districts with a majority voting formula and a ballot structure calling for either a nominal or an ordinal mark-ing. There are basically two approaches to majority voting: the double ballot and the alternative vote. The double-ballot approach requires voters to mark an “X” on their ballot; if one candidate gains a majority in the first round of voting, they win the seat. if no candidate gains a majority of the vote, however, a second round of voting is held a few weeks later, usually involving only the two most popular candidates from the first ballot. Within the narrowed field of candidates, one of the remaining candidates usually gains a majority of the vote in the second round. in practice, double-ballot rules may vary. For instance, some countries require low vote-getters to drop off the ballot in the second round while others do not. Canadian political parties use a varia-tion of the double ballot when they elect their leaders, though in these cases the rounds of balloting may continue beyond just two if a candidate has not secured a majority.

The other approach to majority voting is the alternative vote. Here, voters are re-quired to mark a numeric preference for the candidates on their ballot. At the end of balloting, the first-preference choices for the candidates are added up; if a candidate has secured a majority of the ballots cast, she is declared the winner. However, if no candidate has gained a majority of the votes cast, then the lowest vote-getter is elimi-nated and her ballots are redistributed on the basis of the second preferences indicated on the ballots. if a candidate now has a majority, a winner can be declared; if not, the process is repeated until someone does gain a majority. Countries using these systems include France (double ballot) and Australia (alternative vote). Both of these approaches


to majority voting are typically conducted in single-member ridings, though multi-member ridings have been utilized in the past.

PRoPoRTionAl VoTinG SYSTEMS

Proportional voting systems are varied and extremely flexible. They can combine single- and multi-member districting, or simply use one or a number of multi-member districts combined with a proportional voting formula and a ballot structure allowing for nominal or ordinal choices. There are three basic forms of proportional representation, or PR: party list, mixed-member proportional (MMP), and the single transferable vote (STV).

❍ Party list

The party list form of PR utilizes multi-member districts where voters typically indi-cate support for a list of candidates representing different parties. After balloting, the votes are totalled and parties receive representation in proportion to the votes cast for them. Thus if Party A gained 30% of the total vote, the top 30% of the names on that party’s list would be declared elected. in some cases party-list systems allow voters to give preference to particular names on a party list; if enough voters alter the order of the list, it can alter the order of who gets taken off the party’s list. in practice, however, such individual preferences usually have little impact on the results. Party-list PR was commonly adopted in western countries that changed their voting systems in the early 20th century, and today is associated with the Scandinavian and Benelux countries.

❍ Mixed-Member Proportional

The mixed-member proportional (MMP) form of PR utilizes both single- and multi-member districts where voters usually mark two ballots, one for an individual running in the local single-member riding, and one for their party preference from a national party list. After balloting, the results of the local contests are added up according to plurality voting rules, and then the party-ballot results are totalled. The party-ballot results are used to establish what the overall proportional results of the election should be. if Party A gains 30% of the party votes, then it is entitled to 30% of the to-tal seats when the election is completed. To determine which members of Party A will gain election, a number of steps must be taken. First, any members of Party A that have won the local single-member plurality contests are declared elected. Second, the percentage of seats these local winners represent is subtracted from the total percent-age of seats to be accorded to Party A. Finally, the rest of Party A’s representatives are then drawn from the party list. let’s say that Party A’s victories in the local ridings amount to 15% of the total seats up for election. Because Party A won 30% of the party vote, it is still entitled to another 15% of the total representation available and, as such, the first 15% of the names on their party list would be declared elected. in a sense, MMP elections look like two elections, one where parties win local ridings and another where the party-list process compensates parties whose popular vote exceeds


the percentage of seats that they won in the local contests. This system was invented in postwar Germany and was most recently introduced in new Zealand.

❍ Single Transferable Vote

The single transferable vote form of PR utilizes multi-member districts that typically range in size from three to five members. Here, voters mark their choices preferen-tially (i.e., first, second, third, etc.) across the range of individual candidates contesting the election. To win a seat, a candidate must win a quota, or a proportion of the over-all vote. For instance, in a five-member riding a winner would require roughly 20% of the total votes to secure election. After balloting, the votes are totalled on the basis of the first preferences marked; if any candidate has achieved the quota, they are declared elected. if no candidate has achieved the quota, the lowest vote-getter is dropped and his or her ballots are redistributed on the basis of the second preferences marked on them. This process is repeated until someone gains a quota and is thus elected. When someone achieves the quota, the votes they gained in excess of their quota are also re-distributed, though only for a portion of their original value. let’s say that Candidate A has achieved twice the quota. What this means it that she really only requires half of each vote cast for her to secure election—the other half of every vote can go to help someone else gain election. This is how STV produces proportional results. of course, not all examples are so mathematically simple and today computer counting is popu-lar for STV elections. This process of elimination goes on until all the positions in the multi-member riding have been filled. STV is currently used in ireland, Malta, and for various levels of government in Australia. Historically, STV was widely used in Canada, with provincial applications in urban ridings in Alberta and Manitoba and municipal applications in Winnipeg, Calgary, Vancouver, and 16 other local governments.

SEMi-PRoPoRTionAl VoTinG SYSTEMS

Systems in the last subcategory of voting systems are often referred to as “semi-proportional” because they produce results that are not fully proportional but tend to offer better minority representation than either plurality or majority voting systems. Three of these systems are actually variants of multi-member plurality voting: the limited vote, the cumulative vote, and the single non-transferable vote. The final sys-tem is sometimes referred to as mixed-member majority (to distinguish it from the proportional mixed system, MMP) or as a non-compensatory mixed voting system. let’s examine each in turn.

❍ The limited Vote

With the limited vote, electors in a multi-member constituency have fewer votes than the number of positions to be filled. For instance, in a three-member riding with lim-ited voting, voters would have two votes. What this means is that a minority that votes strategically will have a good chance of gaining representation because the “limit” on the votes prevents the largest group of voters from winning everything. To use a concrete


example, let’s propose that Party A has 14 supporters while Party B has 10 supporters. if this were a typical multi-member plurality election, voters would have as many votes as there are seats. Thus if there were three seats, voters would have three votes each—meaning that Party A would be assured of winning all the seats since each Party A candidate would have more support (14 votes each) than each Party B candi-date (10 votes each).

now, let’s rerun the election using limited-vote rules. There are still three positions to be filled, but this time voters have just two votes. if Party B strategically decides to run just one candidate, she will likely get the maximum potential Party B votes: 10. What this means is that Party A cannot run three candidates and hope to win all three seats because their 28 votes will not be enough to elect three candidates; one candi-date will end up with fewer votes than B and thus B will gain election. To many, this seems only fair, because as a body of opinion B represents 40% of the electorate to A’s 60%. The single non-transferable vote (SnTV) is a more extreme form of the limited vote because it limits the number of votes that voters may cast to a single vote. This has the effect of forcing both major and minor competitors to be very strategic in as-sessing their voting support and deciding how many candidates they should run. The limited vote was used historically in the UK and ontario and is currently used for a host of local elections in the United States. SnTV was long used in Japan, Taiwan, and South Korea, though recently some of these countries have adopted different systems.

❍ The Cumulative Vote

The cumulative vote operates in an opposite fashion to the limited vote. instead of limiting the number of votes an elector can cast, the cumulative vote allows him to cast multiple votes for the same candidate. Voters in a three-member riding would get three votes, which they could cast in any combination for a single or multiple candi-dates. They could give one vote to one candidate and two votes to another, or one vote to three different candidates, or all three votes to one candidate. As with the example above, the less-popular Party B would be strategic under cumulative voting to limit the number of candidates it runs and encourage its supporters to “cumulate” their voting support on the remaining standard-bearers. By doing so, the party would be assured of gaining some representation. The cumulative vote was used historically for school board elections in the UK and has been mandated by the US courts for use in a number of school board elections, particularly across the American South. it was also used for 100 years in illinois for state elections.

❍ The Parallel Voting System

The most recent semi-proportional voting system to emerge has been the parallel or mixed-member majority system (also known as a non-compensatory mixed system). Basically, this system looks like the proportional MMP system, with legislative represen-tation split between single-member-riding MPs and party-list MPs. The key difference between this system and MMP is that there is no compensatory relationship between the two levels of representation. What this means is that the two arenas—the local


choice and the party-list choice—run “parallel” to each other as self-contained elections, leading to results that may depart substantially from proportionality. For instance, let’s say that there are 100 seats to be elected, half in single-member ridings and half from a party-list pool. When the votes are counted, Party A has won 40 local riding seats and 40% of the party-list votes. Under MMP rules, Party A would be awarded no more seats because their 40 local riding victories already equals the 40% of the party-list total. But under the parallel system, Party A would win their 40 local seats and be awarded 40% of the party-list pool, adding 20 more seats to their total. The parallel system offers something to smaller parties in that it has a party-list pool of seats. in its larger workings, however, it tends to inflate the support for larger parties, leading to less-than-proportional results overall. This system has recently been intro-duced in Japan, Russia, and a host of other countries.

Classifying Voting SystemsGrouping voting systems together is a task that generates its own debate (Blais 1988). Some commentators prefer to group voting systems by whether they tend to create a majority government or not. Thus plurality and majority systems are both considered majoritarian voting systems, less for the results they create in particular ridings than for the overall results they create in terms of producing single-party legislative majority governments (norris 2004: 67). others distinguish voting systems by virtue of their mechanical parts. Here, the distinction is between pure and mixed systems. Pure sys-tems utilize one combination of formula, districting, and ballot structure while mixed systems utilize two. Thus single-member plurality would be a pure system, while mixed-member proportional would be mixed, as it comprises both the single-member plurality combination with a multi-member party-list proportional combination (Massicotte and Blais 1999). But the breakdown utilized in this volume is premised on the actual election results that the different systems tend to create where they are used (Pilon 2001; lCC 2004). Thus the plurality–majority–PR breakdown accords with the logic of what goes on in the constituencies where the voters are using such systems. This is where we need to begin if we are to understand how the systems work.

These descriptions of the main voting systems and their workings provide enough information to understand how such systems might be used in practice and what some of the implications of using them might be. For the purpose of debating voting-system reform, this information is more than sufficient. There are volumes that provide much more information about the technical side of voting-system design and adminis-tration (newland 1982; Farrell 2001; lCC 2004). But the details of the various formulas and divisors (Sainte-lague vs. D’Hondt vs. Hare) or the transferring mechanisms utilized in STV (pile method vs. weighted-inclusive Gregory method) are details that only experts in elections administration would be required to know; they are irrele-vant to debates concerned with choosing one voting system over another. So too are many of the technical debates over the allegedly perverse functioning of this or that voting rule. For instance, some allege that certain voting systems are “non-monotonic” or, put plainly, irrational in their effects because in some cases voters may further


their most preferred choice by actually choosing it second (Doron and Kronick 1977; Dummett 1984). This problem—elegantly demonstrated in a host of academic jour-nal articles—is, however, a red herring. As with the social-choice problem of cycling, the incidence of non-monotonicity is just a theoretical possibility; it almost never oc-curs in actual voting. in fact, in a study of STV elections in northern ireland covering a 22-year period, researchers could not find even one case of a non-monotonic result (Farrell 2001: 150). The debates over voting systems involve big questions and big results, not the splitting of hairs over debates in mathematics. We must judge the per-formance of voting systems in practice, not in theory.

ConclusionVoting systems matter. They influence the choices that voters and political parties make, and thus play a key role in shaping the political outcomes that emerge from their use. But voting-system effects are not specific or automatic. They do not pro-duce a particular kind of party system, for instance, or contribute to any necessarily immediate results at the moment they are changed. instead, voting-system rules should be seen to operate within a political and social milieu, where the mobilization of politics comes up against and is influenced by the nature of the rules affecting po-litical competition. These two factors must be combined into a single analysis if we are to avoid the all-or-nothing character of most debates on voting-system effects. With that insight, we can examine many different kinds of voting systems in an at-tempt to gauge how the stuff of Canadian politics is presently affected by our choice of voting system, and how that politics might be different under a different set of vot-ing rules.

26

CHAPTER THREE

investigating Plurality

introductionFor most Canadians, the voting system is not something they think about a great deal. They assume that the way they vote is the only way to vote. Come election day, they go to the polls, mark an “X” by their chosen candidate, and wait to see if they are on the winning side. in other words, the whole process by which votes are translated into representation is simply taken for granted. That there might be other ways to elect rep-resentatives would not occur to most people, as they have never voted in any other way.

A number of implications flow from recognizing this fact. First, most people are unaware of the problems associated with our form of voting because they lack any context by which to judge the results our process produces. Second, they fail to under-stand the mechanics of how our system generates its results. This means they cannot critically examine the claims made about our elections. it is not surprising, then, that they tend to misunderstand the nature of the results produced. There is perhaps no better example of this than the public’s concept of majority government. Polling sug-gests that nearly 50% of Canadians believe that the majority government typically produced by our system enjoys the support of a majority of the voters (Bricker and Redfern 2001). But, as pointed out earlier, this is seldom the case. no doubt the screaming headlines in the newspapers about the election of yet another “landslide majority government” can help explain this confusion. nonetheless, the widespread public ignorance about what our voting system is and how it works allows these me-dia distortions to go unchallenged.

The purpose of this chapter is to outline how our current voting system works and explore the implications of these workings for voters, parties, and our political sys-tem. in doing so, we will critically address the many claims made for the current system and the main problems identified with its use, paying specific attention to questions of representation, majority government, electoral competition, and strategic voting.

i n v e s t i g at i n g p lu r a l i t y / 27

Front page from the globe and mail, october 26, 1993, following the election of jean Chrétien’s Liberals. nowhere on this page is it reported or even suggested that the huge Liberal majority was achieved with only 41% of the votes cast. reprinted with permission from the globe and mail.

FigUre 3.1


How the Current System Works and its implicationsFor most voters, their knowledge of the voting system only extends to understanding how to correctly mark their ballot. Some also understand how the votes are counted up and a winner is produced. But this level of knowledge does not mean that most voters understand how the system works as a whole. For instance, very few voters could explain why there is usually such a variance between the popular vote for parties and the percent-age of seats won by them. Yet this is just the kind of understanding that is required if voters are to make informed judgments about the efficacy of their democratic institutions.

our voting system in Canada at all levels of government is typically called the plu-rality system. it is also regularly referred to as single-member plurality, “first past the post,” and the relative majority system. The mechanics of the system are fairly simple and straightforward, though the political implications of the system and its results are much less so. in a plurality voting system, the candidate with the most votes wins, re-gardless of the percentage of the total votes that number represents. With two candi-dates running, the winner will have secured more than 50% of the votes. But whether that majority represents 80% or 51% does not matter—all that matters is that the leading candidate has more votes than any other. For example, in the election below, Candidate A has registered just over half of the voting support compared to Candi-date B, who received almost the same level of support:

Candidate A 51% winnerCandidate B 49%

Despite the fact that voters seem fairly evenly divided on the choice, Candidate A wins 100% of the representation because 2% more of the voting population preferred her to her rival. Fifty-one percent have gained a representative that 49% did not endorse. or, put another way, 51% of the voters have a representative while 49% have no rep-resentation at all. But that is the logic of the plurality system—it is an all-or-nothing approach to representation where the winner “takes all.”

With more than two serious candidates running, the winner may not even gain a majority of the votes. in the example below, Candidate A wins with far less than a majority of the voting support because the majority of voters are split on an accept-able alternative:

Candidate A 39% winnerCandidate B 35%Candidate C 26%

Even though 61% of voters wanted someone else, Candidate A wins the seat under plurality because she has gained more votes than any other single candidate. nor is this sort of result merely a statistical possibility; in a typical Canadian election, often more than half of the ridings are won under such circumstances. in fact, the margin of victory can be shockingly low, as this result in the Saskatoon/Humbolt riding from the 2004 federal election demonstrates:


Conservative 27% winnernDP 26%liberal 25%independent 20%Green 2%

Because the Conservative in this riding was “first past the post,” he won—despite the fact that 73% of the voters wanted somebody else. This demonstrates how the “post” is not fixed. in our examples above alone, the post has shifted from 51% to 39% to 27%. Such results have led many to question how appropriate the horse-racing meta-phor is in describing the workings of this voting system (at the track, of course, the post doesn’t move). But the metaphor has less to do with the structure of the race-track (i.e., where the post is) than the relationship between the horses. The point is that, in a horse race—even though mere seconds may be all that separate the winning horse from his competitors—the winner gets all the glory. So too in plurality elec-tions, winning only requires that the leader stay in front of the pack. How far in front or how large the pack is do not matter.

These examples also demonstrate how the plurality system typically creates a gap between the overall levels of voting support for different parties in an election and the proportion of seats allotted to them. For instance, let’s assume that the candidates running in this example of a local-riding contest represent different parties:

Party A candidate 39% winnerParty B candidate 35%Party C candidate 26%

if these results were reproduced in every riding, Party A would capture 100% of the seats with just 39% of the total votes cast. of course, these sorts of results don’t typ-ically occur; few parties are stronger than all the others in every region where an election is being held. However, where a party is considerably stronger they can take all the representation, as this example of the 1986 provincial election in new Bruns-wick demonstrates:

popular Vote Seats

Liberal 60% 100%Conservative 29% 0%nDp 11% 0%

Sixty percent is a considerable amount of support, but we cannot say that the voters in the example above really intended to give all the representation to one party and nothing to the rest. This result was the product of the mechanics of the voting system. Plurality allows the strongest party in each riding to capture every seat up for election, and lacks a mechanism to reward parties with widespread strength that nonetheless


do not come first in any particular riding. This 1986 provincial result was extreme, though not unusual. over the past decade, there have been a host of lopsided results in new Brunswick and PEi. The PEi opposition ranks were composed of two mem-bers in 1989, one in 1993 and 2000, and four in 2003, despite opposition support ranging from 34 to 43% of the popular vote. Meanwhile, at the other end of the country, a similar result occurred in BC in 2001 when the liberals turned 57% of the popular vote into 97% of the legislative seats.

To gain advantage under the plurality system, however, a party needn’t be stronger than all the others in every region. Where one party is the strongest in just a majority of ridings, it can turn a minority of votes into a majority of seats. in fact, this is the typical result in plurality voting systems. it is this ability to win seats with various lev-els of support that explains why plurality vote totals—the popular vote for parties—don’t match up with the proportion of seats won by the parties. in the example below, Party A takes control of a five-member council with just 40% of the total votes cast:

riding 1 riding 2 riding 3 riding 4 riding 5 popular Vote

party a 40 party a 42 party a 38 party a 39 party a 41 40% party B 35 party B 33 party B 37 party B 46 party B 44 39% party C 25 party C 25 party C 25 party C 15 party C 15 21%

Undoubtedly, given Canadian-media practices, Party A would be declared to have “won a majority” based on this result. And they have won a majority—of seats. But obscured in the coverage and the details of the voting results is the fact that a majority of voters did not endorse Party A and did not intend to award them a majority of the seats. This result was the product of the workings of the voting system, not the voters’ intentions. The voters gave Party A 40% of the votes; the system converted that into 60% of the seats. Meanwhile, Party B also got nearly 40% of the votes but gained no bonus. And Party C has been excluded altogether, winning no seats despite gaining 20% of the votes.

Debating the Effects of Plurality VotingThe dynamics set out above have led researchers studying the effects of plurality vot-ing to underline a number of general tendencies associated with the system. First, plurality systems will typically produce single-party legislative majority governments because of the system’s tendency to overrepresent the leading vote-getter. Thus parties gaining 40–50% of the popular vote may expect to gain 60–75% of the seats, depending on how the opposition vote is divided. Second, plurality systems will tend to discour-age the formation of new parties and hinder the success of third parties because, to gain access to the system, such competitors must be able to come first in a riding against the existing major parties. Third, plurality systems will tend to reward parties with geographically concentrated support over parties with more dispersed levels of support (Blais 1991: 240–3; norris 2004: 84–8). What these tendencies amount to is a voting system regularly associated with single-party majority government and a


fairly small, stable two-party system—one where each competitor regularly alternates as the government and opposition, with perhaps one more party emerging from time to time that focuses on regional demands (Powell 2000).

Though the public is largely unaware of the workings of plurality voting and thus cannot be said to endorse the general tendencies of the system, plurality does have its defenders, particularly among politicians, media commentators, and academics. in-deed, one study found a considerable gap between the opinions of politicians and the public on the voting system, particularly among regional and potential governing parties (Black and Hicks 2006a: 21–4). on the whole, plurality’s defenders tend to approve of its general tendencies as a necessary set of limits that help democracies function. Thus for them, the fact that plurality tends to create majority governments out of minority voting support for the leading party is a good thing, as it contributes to government stability and clarifies just who is responsible for policy outcomes. This means that a government can effectively introduce policies and voters know whom to hold accountable for them. if voters are unhappy with the government, plurality’s tendency to overrepresent can work for the opposition as voters shift their support and “throw the rascals out.” Plurality’s defenders also applaud the system’s tendency to restrict party competition as they see this as an effective check on extremism and an incentive to political cooperation. As new parties face serious obstacles to break into the system, political interests are instead encouraged to work within existing “big tent” parties that can broker an effective consensus among different views (Powell 2000). it is for just this reason that John Courtney defends the plurality system, which he claims has led to the creation of “pan-Canadian” political parties capable of bro-kering the difficult compromises necessary to keep a country like Canada together (Courtney 1999). Even the geographic bias of plurality is defended as a key element of local representation within the larger political system.

needless to say, such defences of plurality have fuelled much debate. Some critics, like British Conservative lord Hailsham, argue that plurality’s bias toward stability borders on an “elective dictatorship” as majority governments often remain indiffer-ent to opposition parties and the public until just before election time (norris 1995: 73). in terms of policy, however, one might argue that plurality is not really that stable at all. As a small shift in votes may “throw the rascals out” and bring the opposition into power, a relatively minor shift in public opinion might result in a dramatic swing in policy direction. indeed, this happened repeatedly in the UK after World War ii, as minor movements in votes led to a change in government and consequent shifts in policies like regional economic development, income policies, and nationalization (Wilson 1975; Stout 1975). one result was the successive nationalization and dena-tionalization of British Steel, hardly a “stable” result for the economy (Zaharadis 1999). it is for reasons such as this that some argue that plurality leads to govern-ments that are more extreme than the voters (Powell 2000: 196, 221).

nor do critics accept that plurality’s limits on political competition aid political cooperation and help curb extremism. After all, a key criticism of plurality is that it actively militates against cooperation between parties because of its all-or-nothing dynamic. As for political extremism, there is considerable debate about the degree to which the voting system can be held responsible. Most analysts agree that the lower


threshold of election in PR systems makes it easier for new parties, including extrem-ist ones, to gain representation (Jackman and Volper 1996). But this does not explain why such parties emerge when they do, why their support rises or falls, or why sup-port for extremist parties varies across countries that use PR. Thus PR cannot be said to promote extremism, though it may allow it to emerge when it has become suffi-ciently popular (Carter 2002). When such views do emerge under PR, however, they are not overrepresented as they might be under plurality or majority voting rules. in examining the changing fortunes of the recent populist right in Europe, we can see that its electoral support has waxed and waned. indeed, as such parties have attempt-ed to influence governments, they have had to moderate their stances on a host of controversial issues (Brug and Fennema 2003; Heinisch 2003). Meanwhile, majoritar-ian systems have also witnessed the rise of electorally successful extremists, such as le Pen’s national Front in France. Even under plurality, racist politics is gaining traction, as the increasing electoral success of the British national party in local and national elections demonstrates. The party saw 20 local councillors elected in 2005 and gained nearly 200,000 votes in the general election that same year (BBC 2006).

John Courtney’s faith in the ability of plurality to foster pan-Canadian parties can also be challenged by the clear breakdown of pan-Canadianism throughout Canadian history, particularly since the 1960s. The obverse of pan-Canadianism is the regional politics that the plurality system facilitates with its overrepresentation of geographi-cally concentrated political movements. A decade before Courtney’s observations, Alan Cairns had already pointed out that exaggerated regionalism was the much more typical result of our voting system than pan-Canadianism (Cairns 1968). Since then, the regionalizing tendencies of our voting system have been only too apparent, par-ticularly in the 1980 election, which witnessed a majority liberal government gain no seats in the West despite considerable voting support, and the 1993 federal election, which translated diverse views across the country into regionalized blocs of support for one party or another (Swayze 1996).

on their own terms, the claimed benefits of the plurality system can easily be chal-lenged. However, when we shift to broader normative issues of democratic participation and decision making, plurality fails utterly. Specifically, plurality fails to adequately represent voters, parties, or our social diversity. it fails to produce authentic majority government and militates against fair and open party competition. Finally, plurality encourages a negative strategic voting that limits voters’ choices and discourages their participation in the political system.

Examining Plurality VotingSeveral basic and crucial ideas must be central to just about any discussion of democ-racy in the modern world. Contemporary democracy involves, among other things: representation; majority rule; protection of minorities; free and fair competitive elec-tions; and transparent, straightforward democratic institutions. These are really the minimum conditions for democratic legitimacy. other factors may be added to suit local conditions and values, but lacking any one of these basic conditions undermines the claims to democracy altogether. As we shall see, plurality voting compromises every


one of these essential conditions, from representation to majority rule to institutional transparency. let us take each issue in turn.

The Plurality Approach to RepresentationWe call our system a “representative democracy” because those who deliberate and make decisions on the public’s behalf are supposed to be able to claim some legitima-cy as representatives of the public. Plurality fails this most basic requirement. it does a lousy job of representing what the public says with its votes, begging the question of just how “democratic” its results can be. Specifically, plurality fails to effectively rep-resent individual voters, it fails to accurately represent support for parties, and it fails to adequately represent our social diversity.

REPRESEnTinG inDiViDUAl VoTERS

People trot out to the polls on election day with a purpose in mind: to see their votes contribute to the election of a representative that in some way reflects their views about politics. But with plurality voting, many needn’t bother. in any given riding, depending on the vote split, an average of between 45 and 65% of the votes are “wast-ed”—they don’t end up contributing to the election of anyone. in the example below, Party A voters have cast an effective ballot; their votes have been counted toward the election of a representative. But Party B and Party C voters, despite in this case com-prising a majority of the votes cast, have failed to elect a representative. As such, their votes are effectively wasted:

Party A 4,000 votes winnerParty B 3,500 votesParty C 2,500 votes

Add this up across all the ridings in the whole election and anywhere from a large mi-nority to a majority of the electorate are wasting their time casting a ballot. in fact, every Canadian federal and provincial election results in millions of wasted votes. Between 1980 and 2006 an average of 49% of the total votes cast were wasted. All parties are affected: the percentage of wasted votes ranges from 26% for the Bloc, to 43% for the liberals, to a staggering 80% for the nDP, and fully 100% for the Greens (FVC 2006). of course, if we shift from party performance to the election results as a whole, we can see that wasted votes are not just a problem for the smaller parties. in the 2004 federal election, liberal and Conservative supporters actually cast 70% of the total wasted votes in the election, amounting to nearly four million votes (Gordon 2004). This is because there are vast regions of the country where neither party can capitalize on their electoral support—think the liberals in the West and the Conservatives in Quebec and Canada’s major cities. in most cases, it matters less what you vote for than where you vote for it. in the 2004 federal election 178,507 Conservative votes in Saskatchewan elected 14 MPs while nearly twice as many Conservative votes in Quebec elected no one (FVC 2006). it is hard not to see these latter Conservative votes as anything but “wasted.”


Defenders of plurality voting often object to the notion of a “wasted vote.” The complaints take a number of forms. Some argue that in any election the number of seats is limited and it stands to reason that not all possible views in the community can be represented. The point is obviously true, though trivial. The fact that a repre-sentative system cannot, by definition, represent every possible view does not refute the “wasted vote” idea. Even though it is possible that 10,000 voters could harbour 10,000 different ideas about representation, this is not what typically occurs, either in our election example above or in the real world. instead, nearly all voters in Canadian elections cast their ballots for parties with significant voter support, but depending on where they live, their votes may or may not count on election day. other defenders of plurality claim that voting is about expressing deeply felt political opinions and that a “vote is not wasted if it expresses what the elector wants to say on the one occasion when individuals can register their political views” (Rose and McAllister 1992: 135). of course, if this were really the point of voting then there would be no need to pick a winner from among the competitors.

Another argument against the notion of wasted votes claims that there are no wast-ed votes under plurality because, as the strength or weakness of the losing candidates ultimately determines who comes out on top, even those votes that do not back the winner have played a key role in deciding the outcome. Few would deny the existence of this relationship between the winning and losing piles of votes under plurality, but the point is irrelevant to the stated concern—specifically, that many voters feel they have no positive outcome as a result of voting. other critics of the “wasted vote” thesis argue that there is little evidence that MPs do not respond to or act for constituents who did not support them electorally (Franks 1987: 65). This may be true as concerns the ombuds-role of the local representative, but it clearly can’t be true when it comes to more broad and zero-sum issues of policy—just the ones that divide the parties that voters are choosing between. Elections hardly mobilize voters on the basis of a “better ombudsperson”; they are about big questions and the direction of government policy. And here, plurality’s winner-take-all approach leaves those unsuccessful at the ballot box without representation of their views. in this way, their ballots have not contributed to the election of anyone to speak on their behalf and thus are “wasted.”

Beyond the wasted votes, this example underlines another grave problem with plurality voting: not all votes are treated equally. Why is 40% of the vote good enough to win in one riding but 50% is required in another riding? Why could 7,279 voters lay claim to a Conservative representative in Churchill River-Saskatchewan in 2004 but 20,570 Conservative voters failed in Simcoe north-ontario in the same election? one of the most basic principles of democratic voting is that votes should have equal weight in the outcome or the notion of “one person, one vote” ceases to have much meaning. But plurality does not, and frankly cannot, satisfy this criteria. Some claim this is more a problem of drawing the boundaries for the ridings than a problem with the voting system per se. They would point to gerrymandering, or the constitutional overrepresentation of certain regions, as the reason such results occur. it is true that variation in the sizes of ridings in Canada—from a 2006 average of 26,919 registered voters per district in PEi to an average of 76,400 in downtown Toronto—is a factor in our skewed results. But even if all the ridings had exactly the same number of voters,


there would still be no guarantee of voting equality under plurality because there is no set threshold for winning. if we compare ridings of similar size in the 2006 federal election, we find winners with varying levels of support—55% in Willowdale-ontario, 43% in Vandreuil-Soulanges-Quebec, 37% in Saanich-Gulf islands-BC, and so on. Depending on the competitive situation in any given riding, the winning threshold might be well above or below 50%, meaning that there can be no equality in the effect of votes in the election as a whole.

From the point of view of the individual voter, plurality looks like a lousy deal. The odds typically range from a one-in-two to a two-in-three chance that her vote will be wasted and not count toward the election of anyone. in addition, there is no way to assure that individual votes will have equal voting power in the election as a whole. For voters, the system is experienced as unpredictable, unequal, and seemingly arbi-trary in its differential treatment of individual votes—hardly the attributes one would expect of a system that is supposed to act as the key interface between an individual and her democracy.

REPRESEnTinG PARTiES

The second major representational failing of plurality is its inability to represent party support accurately, both in terms of the real levels of voting support a party enjoys and where that support is. Party results are important, as voters regularly report that party label—the names and perceived identities of the different parties—is the key factor informing their decision about how to vote (Clarke et al. 1980: 102, 216, 218–9, 222; nevitte et al. 2000: 133–4; Blais et al. 2002: 116–7). Plurality tends to distort the election outcomes for parties, overrepresenting regional parties and larger parties with geo-graphically concentrated support while underrepresenting (or not representing at all) smaller parties and parties with more diffuse support. These tendencies were particu-larly apparent in the 2000, 2004, and 2006 federal elections. Regional parties like the Bloc Québécois and parties with solid regional blocks of support like the liberals were overrepresented, while parties with more diffuse support like the nDP or the Greens were either underrepresented or not represented at all. nor are these tendencies new—nationally, they can be traced to the decisive break with a two-party system in 1921.

These problems in misrepresenting parties under plurality have a number of nega-tive effects on the political system. At the most basic level, underrepresenting some parties while overrepresenting others is simply unfair. Why should 10.5% of the votes give the Bloc 51 seats in the 2006 federal election while 17.5% of the votes for the nDP delivers only 29 seats? The inequalities become readily apparent if one divides the different parties’ vote totals by the number of seats they have won. in 2006 this would mean that every Conservative seat represented 43,339 voters while every nDP seat represented 89,296 voters. Conservative supporters would appear to have had twice the voting power of new Democrats in this election, at least in terms of turning their votes into representation. And even this result was something of an improvement. in 2004 Conservatives had almost triple the voting power of new Democrats, breaking down to 40,000 voters per Conservative compared with 111,000 per new Democrat. These sorts of inequalities have been constant in Canadian elections, and have hurt


smaller parties with more diffuse support in particular. Although this typically affected the nDP and is now affecting the Greens, it also was a factor in the underrepresenta-tion of Progressive Conservatives after 1993.

Another key problem stemming from plurality’s inability to reflect party strength accurately is the fact that election results often send the wrong message to the public about the country’s political choices. Because the seat totals for parties rather than the total votes cast for them are typically the ongoing focus for the media, it is easy for voters to get the wrong idea about just how popular different parties are and where their support is. For instance, in the 1997 federal election the Reform party gained three times as many seats as the Progressive Conservatives (60 to 20), leading most commentators to observe that the former had won the battle for right-of-centre voters. But a look at the vote totals reveals that the parties were in fact equally popular: the difference in their popular vote amounted to less than 1%. Reform just appeared more popular because their concentrated support in Alberta allowed them to win more seats than the Conservatives, whose support was more spread out across the country. Even those parties that are more successful see their representation distorted, as they tend to win seats in their strongest areas and nowhere else, making their support seem more regional than it is. Though tilted toward Alberta in terms of winning seats in 1997, Reform did have considerable voting support in ontario as well—but not enough in any particular riding to win. The effect was to make the party appear more regional than its support really was. if elections are meant to provide information about what voters think, then plurality is severely garbling the information it transmits.

The 1993 federal election may rank as the nadir of party representation under plurality in Canada. The election witnessed the dramatic decline in popularity of the

% of votes % of seats

2000 Federal Election 2004 Federal Election 2006 Federal Election

NDP New Democratic PartyG GreenO Other

0 10 20 30 40 50 60 0 10 20 30 40 50 60 0 10 20 30 40 50 60

L

BQ

CA

PC

NDP

G

O

LEGEND L LiberalCA Canadian AllianceBQ Bloc QuébécoisPC Progressive Conservative (2000),

Conservative (2004, 2006)

FigUre 3.2 CanaDian FeDeraL eLeCtion reSULtS: 2000, 2004, 2006

Source: Compiled from elections Canada Statement of Votes 2000, 2004, 2006.


governing Progressive Conservatives, a steep loss of support for the nDP, some im-provement for the liberals, and an impressive entry of two new parties: Reform and the Bloc Québécois. The results in terms of seats won appeared stark. Though the liberals gained a majority of seats, the country appeared balkanized along regional and French/anti-French lines. The election results seemed to suggest that westerners were for a largely anti-Quebec Reform party, that Quebecers were for the separatist Bloc Québécois, that ontarians were entirely for the liberals, and that only a few locales stood by the nDP or the PCs. Yet a closer look at the actual voting results supports a very different interpretation. There is no denying that Canadians voted for change in that pivotal election, but a glance at the regional popular vote for the different parties demonstrates that a majority of westerners actually supported parties with more moderate views on Quebec, that a majority of Quebecers voted for parties keen on staying in Canada, and that nearly half of ontarians voted for parties other than the liberals. Even the PCs and nDP were more popular than their seat totals suggested. For instance, 16% of the national-vote total for the PCs translated into less than 1% of the total seats in the House of Commons.

The 1993 plurality results painted a picture of a desperately divided country balkan-ized along regional and language lines—one in which each region strongly supported only one point of view, and they were ones that suggested little room for compromise or productive negotiation. But the voting reality was something different. in fact, looking at the popular votes, the country was not so regional at all, and within each region a healthy debate on the key questions could be found. The fact that the voting system could not capture this complexity arguably nearly brought the country to ruin two years later when Quebecers only narrowly voted to remain in Canada.

FigUre 3.3 1993 FeDeraL eLeCtion: regionaL DiStortionS

Source: Compiled from elections Canada Statement of Votes 1993.

% of votes

West % Ontario % Quebec %Maritimes/

NFLD %

LEGEND

0

10

20

30

40

50

60

70

80

90

100

OthersReform OthersLiberal OthersBloc OthersLiberal

% of seats


Some defenders of plurality discount the need to represent parties effectively. They claim that the voting public is increasingly turning away from parties and demanding that their representatives vote according to the wishes of local constituents rather than slavishly adhering to party discipline. in fact, plurality supporters claim that this is the great strength of their chosen system—that it maintains such a strong link between a local area and its representative. There is no denying that a strong anti-party dis-course has arisen in the last few decades, coinciding with a drop in political-party membership numbers and public involvement in party-organized political activity (Gidengil et al. 2004: 124–5; Cross 2004: 19). Many of today’s voters are “flexible par-tisans” who refuse to be loyal to a particular party (Blais et al. 2002: 116–7). instead, they view each election as an opportunity to shop for the best option and switch be-tween parties depending on what each can offer them at the given time. But accepting that there is greater volatility in the relationship between voters and parties as well as a general decrease in party loyalty does not automatically legitimate a local, non-party focus for representation.

Despite declining faith in parties, voters nonetheless continue to report party-like divisions in their political views—non-party “independents” are rarely elected. And despite being frustrated with parties, voters still report that party label is the key fac-tor determining their vote. in fact, when asked, a majority of voters agree that parties are a crucial component of the democratic process. in a 2000 survey, 69% of respond-ents agreed that “without political parties there cannot be true democracy” while just 23% disagreed and 9% had no opinion (Howe and northrup 2000: 29). on the other hand, endorsing local representation runs into a host of contradictions. These include the fact that voters’ knowledge of federal and provincial politics and local members is uneven, that most political problems are hardly local in origin or nature, and the fact that—despite the total number of people that local members represent—few people seek help from them (Howe 2003; norris 2004: 238–41). But even if such problems could be addressed, there is a larger, more serious flaw in the “local representation” arguments that cannot be overcome: just whose local views is the local member ex-pected to act on? in the case of a vote for or against a hot-button issue, does the member vote in line with the 40% who elected her or the 60% who voted against her? The parties, for all their problems, do represent real political divisions in Canadian communities—divisions that do not disappear by romanticizing local representation (nevitte et al. 2000: 48–51). Plurality, with its winner-take-all logic, cannot accom-modate the political divisions that exist within all these ridings.

Canadians undoubtedly have a love/hate relationship with parties, and plenty of evidence can be marshalled that many believe parties are failing as vehicles of repre-sentation (Howe and northrup 2000: 30). But the polling through which such evidence is gathered is often too limited in asking the public about their views. Undoubtedly, most voters would like their local politicians to listen to them—and they tell pollsters as much. But the polls do not ask those same voters how a local member is supposed to be accountable to divergent views in the local area at the same time. on the other hand, despite their negative responses about parties, voters continue to identify par-ties as the key players in politics and the key factor in their voting decisions. in other words, they vote to be represented by parties. When faced with a choice between


choosing a local candidate or their most preferred party, just 5% of survey respondents were prepared to choose the local candidate (Blais et al. 2003). The fact that plurality fails so spectacularly in accurately representing what voters say in voting for parties is a pretty damning criticism of the system’s claims about representation.

REPRESEnTinG SoCiAl DiVERSiTY

The third major representational failing of plurality is its inability to adequately rep-resent our social diversity. Canada is a diverse country made up of women and men; French, English, and Aboriginals; many faiths; people from a variety of ethnic and racial backgrounds; different sexualities; and so on. To look at our legislatures, how-ever, one might think Canada was a country primarily of rich, professional men, with a much smaller subgroup of professional women and only a smattering of diversity in other terms. Yet women make up more than 50% of Canada’s population while visible minorities comprise 15% of the total. Despite this, women have held an average of just 20% of the seats in Canadian legislatures over the past decade (Cool 2006: 2). levels of visible-minority representation have increased in terms of the total proportion of federal MPs—up to 7% in the 2004 election—but do not match the increasing pro-portion of visible-minority citizens across the country (Black and Hicks 2006b: 26–7). What this means is that the recent increases do not really represent an improvement on the underrepresentation of visible minorities at the federal level. And working people, regardless of colour and gender, are typically wholly absent from our legislatures.

What explains this mismatch between the face of the country and those who rep-resent us in the political realm? Some blame the public, claiming either that they are unwilling to vote for diversity or, conversely, that they do not really care about identity representation when it comes to elections. But serious work on these concerns has

FigUre 3.4 ComparatiVe repreSentation oF Women UnDer pLUraLity

Perc

ent

LEGEND

United Kingdom

United States

Canada

0

10

20

30

40

50

60

70

80

90

100

2005199519851975

Source: Compiled from data cited in Childs et al. 2005; trimble and arscott 2003; Cool 2006; amer 2006.


countered both suggestions. instead, researchers have found that the public does believe that it is important to reflect social differences in legislative chambers, and that voters are not unwilling to vote for diversity (Black and Hicks 2006a: 27, 31; Black and Erick-son 2006). So the problem with our unrepresentative legislatures cannot be blamed solely on the voters. in fact, the root lies elsewhere—namely, in the logic of the plurality system itself, and specifically in the incentives structure it creates for political parties.

Plurality cannot effectively represent diversity because its basic logic involves a winner-take-all competition between parties. As there can only be one winner, politi-cal parties of all stripes tend to select candidates in a conservative manner, choosing a candidate that they feel will appeal to the most people and alienate the fewest. Histor-ically in Canada, such candidates have typically been white males, with some recent exceptions wherever a non-white minority group forms a majority of the population in a particular riding. Yet this is only half the story. inside the party, the race to gain a party nomination is also winner-take-all, with all candidates organizing and signing up mem-bers to help them win the nomination. Here too the process has tended to privilege white males, particularly professional or wealthy white males as they have typically enjoyed superior resources and access to resources compared with lower-class white males, women, and visible minorities. The result of this logic is a situation where the overwhelming majority of candidates, particularly in the parties’ “winnable ridings,” end up being affluent white males (Mcivor 2003: 29–32). Thus the choices that voters can make in seeking to represent diversity are necessarily limited from the start.

Some suggest that these historic problems of diverse representation with plurality are being overcome today. Supporters of plurality point to a general trend of progress

FigUre 3.5 repreSentation oF ViSiBLe minoritieS (Vm) in CanaDa anD HoUSe oF CommonS

Perc

ent

0

2

4

6

8

10

12

14

16

18

20

2004200019971993

LEGEND Percentage of VM in populationPercentage of VM MPs

Ratio to population:

0.47% 0.56% 0.42% 0.48%

Source: adapted from Black and Hicks 2006: 27.


on diversity issues in countries using plurality, with particularly striking improve-ments in recent years in countries like the UK. if we compare women’s representation by decade, there does seem to be some improvement in plurality systems despite the logic outlined above. Do such trends effectively counter the criticisms that plurality cannot handle diversity? Assessing the evidence, the success of the British labour party in dramatically increasing the number of women in the House of Commons—from just 9% in 1994 to 18% in 1997—does seem to suggest that where parties make diverse representation a priority, they can make progress despite plurality. in this case, labour overcame the competitive pressures benefiting men in plurality nomination races by effecting centralized control over the nomination process and introducing women-only shortlists for nominations in winnable labour seats. However, labour’s quota policy was quickly struck down by the courts and further progress has failed to materialize; two subsequent elections witnessed only minor improvements in wom-en’s representation (Childs 2000). in the end, labour’s gambit had really only brought the UK up to the standards in most other plurality-using countries, which remain low and seemingly stuck around 18 to 20% of seats for women (Childs, lovenduski, and Campbell 2005: 20). What is striking in both the UK and Canada is that, after sudden increases in the numbers of women elected, both countries appear to have reached a plateau over the last few elections, with some decline in a few jurisdictions in Canada (Trimble and Arscott 2003). At this rate, some women’s groups predict that equality between women and men under plurality rules in Canada may take another 117 years (Equal Voice 2007).

The problem of women’s representation under plurality voting highlights how these rules contribute to an incentive structure for parties where they consistently fail to respond to changing public demands for representation. Even if we are seeing some slight improvement in women’s representation today, it must be underlined that par-ties under plurality have responded very, very slowly, and well behind public opinion on the issue. Demands for new kinds of representation emerge all the time, and it should be the mark of a modern representation system how well it encourages parties to respond. Calls for improvements in women’s representation first emerged in most western countries in the 1960s, fuelled by a re-emergent women’s movement and an increase in women’s participation in the economy and in higher education (Bashevkin 1993: Mcivor 1996). But while social attitudes about the appropriate role of women in society—particularly in politics—changed, the representation of women in legis-latures elected by plurality did not.

in a comparative study of women’s legislative representation among countries where levels of women’s political activism was high, the voting system was found to be the key determinant affecting their success (Caul 1999). Thus throughout the 1970s and 1980s the all-or-nothing logic of plurality made it difficult for these new demands for representation to be heard, let alone met. Despite public visibility and a strong presence in civil society, women’s groups found it difficult to influence the parties from within or without. Typically, parties would claim either that the all-or-nothing logic of plurality prevented them from running more women (as it might make them uncompetitive) or that they were powerless to control the nomination processes for candidates (as the processes were a matter for the local constituency association)


(norris 2004: 189). Given that it has taken over three decades to address these obsta-cles—and that the results have been mixed and unsatisfying at best—we can reasonably ask how plurality might encourage parties to respond to any new demands for repre-sentation. The track record suggests an answer: not very quickly, and not very well. This is hardly acceptable in a diverse and rapidly changing country like Canada, one with a raft of past differences still unresolved (Aboriginal claims; Quebec; regional alienation) and new challenges on the horizon.

Under plurality, parties fail to represent our social diversity and tend to respond poorly and very slowly to demands for new representation. Combined with its poor representation of parties and its tendency to waste votes or weight them unequally in electoral outcomes, plurality clearly appears to be a poor choice of voting system for a “representative” democracy.

Governing Majorities, Party Competition, and Individual Voter Strategy Under PluralityThe logic of the plurality voting system has effects that resonate in all corners of Can-adian political behaviour, affecting government, political parties, and individual voters. Plurality affects the nature of governing majorities, the process of party competition, and the strategic considerations facing individual voters. By its supporters, plurality is lauded for regularly producing stable majority governments, effective party compe-tition, and clear strategies for voters seeking party and governing accountability. its critics condemn it for creating phony majority governments, restricting electoral and party competition, and coercing electors into an awkward form of strategic voting. At the root of these different perspectives are some fundamentally different notions about what democracy amounts to and how elections either further or limit its po-tential. By examining the nature of majority government, electoral competition, and strategic voting under plurality, we can see how the arguments in favour of the plural-ity system are anything but democratic.

MAJoRiTY GoVERnMEnT

one of the most consistent observations about legislatures using plurality voting to elect their members is that there is a marked tendency toward electing single-party majority governments. Any scan of comparative electoral research bears this out (Blais 1991: 240–1; lijphart 1994: 98). Most Canadian governments have been major-ity governments on these terms, and nearly all governments at the provincial level fit the bill. From 1921 (effectively the beginning of multi-party competition federally in Canada) to 2006 there have been 26 federal elections resulting in 16 legislative major-ities for a single party. But the key point here is that the “majority” being referred to is a majority of the seats. Such “majorities” do not necessarily reflect a majority of the voters. in fact, just the opposite is nearly always the case—such majorities typically represent only a minority of the voters. indeed, from our 26 federal contests, only two election results (1940 and 1958) can be considered clear majority governments by reference to the popular vote (i.e., 51% or more). This is a fundamental characteristic


of plurality voting: it tends to manufacture phony majority governments by inflating the legislative support of the largest party.

When is a majority not a majority? it would appear: nearly always, under plurality voting. This seems curious, as one of the principle values that allegedly distinguish democratic societies from non-democratic ones is that democracies are bound by a notion of majority rule. of course, tolerant societies must find ways to respect mi-norities, either through constitutional safeguards or other means, but in the end “democracy” means that the various zero-sum decisions that have to be made should be made by majority rule. Yet the “majority” making the decisions in plurality voting is nearly always a minority.

Defenders of plurality’s phony majorities justify its distorted results in two ways. on the one hand, they claim the results are acceptable because today’s government will be tomorrow’s opposition, and vice versa. in other words, the results may be dis-torted in one party’s favour today, but that will change tomorrow when in another election the results are distorted in favour of the other party. on the other hand, they claim that phony majority government is a good thing in itself because it affords one party a stable tenure in government, allowing them to introduce their policies unhin-dered. And in the end, if voters do not like that government’s policies, they can “throw the rascals out” at the next election and give the opposition a turn. These notions of “alternation in power” and “stable majority government” are two of the central de-fences of the plurality system (Powell 2000). Yet as we shall see, neither withstands much scrutiny, with one failing as an accurate description of how plurality systems really work and the other for its obvious violations of basic democratic principles.

The “alternation in power” thesis rests on a host of untenable assumptions. First, it assumes a politics of equal two-party competition where voter swings can put the government out of power and the opposition in. But Canadian history is replete with

FigUre 3.6 FeDeraL eLeCtion WinnerS: popULar Vote, 1921–2006

Out of 26 federal elections since 1921:

– 16 legislative majorities– 10 legislative minorities– but only 2 clear popular vote majorities (51% or greater)

0

10

20

30

40

50

60

70

80

90

100

2006

2004

2000

1997

1993

1988

1984

1980

1979

1974

1972

1968

1965

1963

1962

1958

1957

1953

1949

1945

1940

1935

1930

1926

1925

1921

Perc

enta

ge o

f po

pula

r vo

te

Source: elections Canada; thorburn 1985: 349.


evidence that plurality does not fit this idyll. When a third party enters the scene, there may be no alternation in power despite public frustration with the government. indeed, if the public cannot agree on a single alternative, an unpopular government may be lodged in power for some time. Most of the 20th-century Canadian federal experience counters the “alternation” thesis, as the liberals dominated government—maintaining power for two-thirds of the century—despite seldom gaining a majority of the vote. The trend is even more pronounced after the advent of multi-party politics in 1921, with the liberals ruling for 59 of the 79 remaining years of the century, or three-quarters of the total. At the provincial level, there are a host of examples, includ-ing the unbroken, multi-decade rule of Social Credit and Conservatives in Alberta, Social Credit in BC, the Conservatives in ontario, the Union nationale in Quebec, the CCF in Saskatchewan, both liberals and Conservatives in different parts of the Mari-times, and so on. in some of these provincial cases, the parties in question did win a majority of the popular vote (particularly in Alberta), but in others the lack of alter-nation reflected the inability of the voters to settle on a single alternative. Thus the controversial Harris Conservatives in ontario and the 1990s nDP in BC both man-aged to win two terms, despite the existence of a consistent majority of voters who opposed their policies.

A second untenable assumption of the “alternation in power” thesis is that every-one eventually gets overrepresented so the occasional underrepresentation is oK. This ignores the fact that third parties stand little chance of ending up in power, as well as the fact that—in some cases—even second parties are structurally disadvan-taged. in BC, the left CCF/nDP has been the second party in the legislature since 1952 but has gained office only three times (1972, 1991, and 1996). in Alberta, both the postwar liberals and new Democrats have formed the official opposition but never held government. in ontario, the second-place liberals lost government in 1943 and didn’t return until 1985. of course, even when there is an alternation in government, it is seldom a case of voters “throwing the rascals out.” Under plurality, governments may get less, the same, or more support, and still be defeated, depending on how the votes break down. in 1975 the BC nDP lost power despite maintaining the level of support that had brought them to power in 1972. in 1935 the federal liberals won a majority government with the same percentage of the vote as had caused them to lose control of the government in 1930. in 2006 the governing new Brunswick Conserva-tives gained more support than in the previous election as well as more votes than the opposition liberals—and still lost the election. imputing change to the voters ignores how the state of party-system competition is much more important in whether gov-ernments remain in power or fall. in the end, it is the plurality system and its tendency to inflate the legislative support of the most popular party that ultimately puts gov-ernments in or votes them out.

Regardless of how it is created, defenders of plurality hail the system’s tendency to exaggerate the support of the leading party because this usually contributes to the election of a single-party legislative majority government—a factor they claim is cru-cial to government stability and accountability. Because a single party will control the legislative agenda, they will be able to introduce their policies unhindered by the op-position, thus assuring stability. At election time, the public will be certain about who


has been responsible for the government’s actions, thus assuring a clear line of ac-countability. The obvious implication of this view is that a government that lacks a single-party majority (e.g., either a coalition or minority government) will be unstable and unable to introduce its policies, or will be unaccountable because voters will be un-sure which party did what and the parties will shift blame onto each other. This “stable majority government” position is perhaps the key defence of plurality (Powell 2000). indeed, until recently, it was almost an article of faith among north American political scientists that “stable majority government” outweighed all other considerations in voting-system choice (Rustow 1950; Corry 1951: 276–83; Beck 1963: 63; lovink 1970: 509; for a review of key influences on the discussion see lijphart 1977: 6–14).

The “stable majority government” arguments fall down on a number of fronts. First, normatively, there can be no democratic argument for denying due representa-tion or inflating representation. To suggest that it is legitimate for a party gaining 40% of the vote to have 60% of the seats simply makes no sense by any democratic logic. Second, Canadians collectively do not typically vote for a one-party majority govern-ment in that they do not give any one party a majority of their support. if one-party government was so important to them, then surely their voting patterns for parties would reflect that. Third, there is little evidence from the comparative experiences of other western countries to support the view that where one party does not gain a leg-islative majority the government is then necessarily unstable (lijphart 1999a; Boston et al. 2003). Canada has had many examples of effective minority and coalition gov-ernment at both the provincial and federal level (Forsey 1964; Dobell 2000). if some instability can be identified in such cases, it has had less to do with the lack of a legis-lative majority than with the unstable incentive structure of plurality itself. Because plurality tends to overrepresent larger parties, there is often some pressure from po-litical elites within the minority government to go back to the polls when their party appears to be gaining public support and hope that, this time, their minority vote will be transformed into a legislative majority (Dobell 2000). of course, in voting systems where votes and seats more accurately match up, there is little incentive for minority governments to rush back to the polls. Fourth, the stable majority governments produced by Canada’s plurality system have failed miserably on questions of account-ability to voters over the past four decades, as countless voter surveys and many dra-matic upheavals in the provincial and federal party systems have demonstrated (Pammett and leduc 2003; Tanguay 2004: 270–1). Finally, the whole focus on “stable majority government” begs the question: stability for whom? An important line of criticism about Canadian democracy is actually that our system is too stable, allowing the PM-dominated executive branch to control our politics at the expense of construc-tive parliamentary deliberation (White 2005: 65–7). Besides, as minority governments in Canada have performed well, and as voters seldom give any one party a majority of their support, there seems little obvious benefit or justification in allowing phony majority governments to monopolize power.

Aside from its dubious democratic legitimacy, there are other reasons to be wor-ried about plurality’s tendency to create phony majority governments. one serious problem is how the regular election of phony majority governments obfuscates the political results of elections, sending distorted signals about the voters’ collective


judgments and intentions. This problem is compounded when we recognize that few voters clearly understand the workings of our electoral system and its implications for party representation and government formation. As a result, nearly 50% of voters confuse legislative majorities with popular-vote majorities, assuming that if a party has won a majority of the seats it has also won a majority of the votes. Yet as we have seen, this is almost never the case. legislative majorities under plurality are typically granted to the largest minority of voters, though this is not always the case. Recently in Canada, we have seen a number of cases where the voting system gave the second-place party more seats than the first-place finisher. in 1996 in BC, 1998 in Quebec, and 2006 in new Brunswick, the second largest minority of voters got more seats than the largest minority of voters—clearly a perverse result! nor are such results rare. one study of western countries over the past 25 years demonstrated that one in five plurality elections delivers these kind of perverse results (Powell 2000: 130). in these examples and in the more typical phony majority governments under plurality, the gap be-tween the actual results and the public’s understanding of them allows some political views to enjoy an inflated legitimacy while diminishing the importance of others. Aside from the ethical issues involved in obscuring the transparency of the election results, one of the long-term effects of phony majority government is to weaken the link between voters and the policy deliberation process, with implications for policy efficacy and voter mobilization.

PARTY CoMPETiTion

other serious problems associated with the phony majority governments typically produced by plurality voting include the impact on party competition and individual voter strategies. A key aspect of democratic politics in the electoral arena is that a competitive dynamic must be present to assure any accountability. if existing parties and politicians are not meeting public expectations, then the public should be able to anticipate replacing them by voting for someone else. But here, plurality places a host of barriers in the way of effective party competition, essentially privileging both status-quo parties and the more regionally based parties. one barrier involves what Duverger called a mechanical effect of plurality voting—the fairly high threshold to gain elec-tion. Plurality rules insist that a party must gain more votes than all others in a riding to win a seat, regardless of how much support it may have overall in the election. So even though the Green party won more than half a million votes in the 2006 federal election, it did not gain any representation because its votes were not concentrated geographi-cally. Yet the liberals, Bloc, Conservatives, and nDP all won various individual seats with just thousands of votes in each of the ridings. Plurality’s bias in aggregating votes in single-member ridings prevents parties with broad geographic support from get-ting into the system. nor were the Greens the only victims; both the nDP and the Progressive Conservatives were dramatically underrepresented in the 1993, 1997, and 2000 elections because their support was less regionally concentrated than that of their opponents. The difficulty in getting past the high threshold for election has psycho-logical effects, too. Given the uphill battle to gain election, potential competitors are less willing to run, and voters worry about “wasting” their vote on new and unproven


parties and their candidates. Such barriers insulate existing parties from challenge, limiting the competitive accountability process that elections are supposed to deliver and alienating frustrated voters—particularly young people—from the political system.

STRATEGiC VoTinG

These barriers to competition also intensify the strategic dimension of voting in a way that puts individual voters at a disadvantage. Voters using plurality often complain that they feel compelled to vote “strategically”—i.e., not for their first choice but for the one that appears most competitive (Blais and Bodet 2006). of course, voters are ulti-mately free to cast their ballots as they like. But if they want to be effective—if they do not want to “waste” their votes—they must vote strategically for the candidate that they can support who has the best chance of winning. Parties constantly underline this dimension at election time, particularly those that benefit from plurality’s over-representation. For example, federal liberals will often warn potential nDP voters that they must vote liberal or risk allowing the Conservatives to win (Clarkson 2005: 201, 255). in this case, following the logic, voting for the left might result in a victory for the right—just the opposite of these voters’ desires. of course, if everyone accept-ed this logic the nDP would never win a seat. Thus rather than being true in all cases, the logic undergirding strategic voting here depends crucially on an assessment of just how strong the parties are competitively in any given locale (Bowler and lanoue 1992; Blais 2002) and on changes in the character of party-system competition from election to election (Blais and Bodet 2006). Assessments of strategic voting in various Canadian elections have ranged from a low of 3 to 6% (Blais et al. 2001; Blais and nadeau 1996) to a high of 11 to 16% (Merolla and Stephenson 2006). Yet all analysts agree that some degree of strategic voting does occur.

The problem for voters is that it is very difficult for individuals to obtain the stra-tegic information to inform such a decision. They would need to have some sense both of what other voters were planning to do and of the competitive position of the various parties in the running (Rose and McAllister 1992: 126–7). of course, very, very few individuals command the resources to get such information. instead, for most individuals, their sense about which candidates are competitive typically comes from gauging the relative levels of advertising for each competitor and following the polls that report what other voters allegedly think. But these proxies for immediate infor-mation have their own drawbacks. Advertising may tell us more about who is good at fundraising than who has the support of voters, while polling tends to privilege estab-lished candidates (whose campaigns or supporters in the news media usually pay for them) over newcomers. And as both advertising and polling are seldom locally driven, their application to local circumstances may be useless and misleading. For instance, a party may be the leading party in a certain riding even though it is running third in support in national polls. Yet seldom can a local candidate, let alone a member of the public, afford to commission a poll to gauge such support (Cross 2004: 115–6). one study discovered that in the 1988 election nearly 50% of the voters guessed wrong in trying to figure out which candidate was least likely to win (Blais and Turgeon 2004).


Some argue that complaints about strategic voting are without substance because political decision making is by its nature strategic. Thus all decisions about voting necessarily involve assessments of strategy regardless of the voting system in use (Kam 1997). The point is true, though trivial. if a choice is to be made, there are few situa-tions where such a decision would not involve the weighing of options and thus a strategic assessment of the choices. But this is not what complaints about strategic voting are focusing on. Rather, they are focusing on the restricted nature of the condi-tions under which a choice must be made. Voting theorist Michael Dummett argues that voting systems should, as much as possible, offer voters a clear path by which to make the decisions they wish to make (Dummett 1984: 81). in other words, negative incentives should not be a key feature of the voting process. Voters’ strategies should ideally be based on various positive actions they might like to see—more or less of this or that policy. Perhaps they wish to see more fiscal responsibility than the federal liberals have provided, and so strategically decide to opt for the Conservatives despite concerns about the latter’s approach to Canadian unity. Clearly, there are strategic considerations in this decision. However, they differ from the fundamentally negative strategic considerations facing voters under plurality. Following the example above, these would involve continuing to vote for the liberals despite the concerns for fear that voting Conservative might allow the nDP to gain election—a result the voter does not think, rightly or wrongly, will address fiscal responsibility.

ConclusionPlurality voting does not represent what people are saying with their votes in Canada. its claimed benefits are either illusory (the strong locale–representative link) or cannot be justified in democratic terms (phony majority government; restricted party com-petition). Voting decisions in Canada are largely party decisions and are seldom made on the basis of locality or the local representative. Plurality certainly does tend to create majority governments out of minority support and unduly limit party competition, but such “benefits” are hard to square with democratic values. Meanwhile, plurality’s demonstrated weaknesses defy common-sense notions of electoral fairness. Parties large and small find themselves misrepresented; millions of votes—usually some-where near a majority of those cast—are wasted and do not count toward the election of anyone; and the diversity of Canadians in terms of demography and political opin-ion fails to register where it matters most, namely, in our allegedly representative chambers. Some minorities get overrepresented to the point of gaining a “majority” of the legislative seats and 100% of the political power, while other significant minori-ties are dramatically underrepresented or not represented at all. in every election, voters face an unenviable task of trying to get sophisticated political messages through what is essentially a medieval voting mechanism, with predictably poor results. The distorted and ineffective representation that emerges from plurality then calls into question the quality of democratic deliberation that such legislatures can produce. if our objective is to enhance the quality of Canadian democracy, then plurality can no longer be defended as a means to that end.

49

CHAPTER FoUR

Exploring Proportional Representation

introductionThe many failings of Canada’s plurality system call out for a response. Exploring the concrete alternatives to our current system is an obvious place to start. Canada could consider switching to any of the voting systems presently in use in the countries now considered to be “democratic” worldwide. Yet such a response is slightly premature. Before we can assess different models of voting and their suitability for use in Canada, we must critically examine the assumptions made by different families of voting sys-tems as these inform the basic logic underlying how the systems work. For instance, the logic of plurality is an all-or-nothing contest where the winner “takes all.” This logic is at the root of all the problems associated with the system: wasted votes, stifled party competition, inaccurate and inadequate representation of parties and social diversity, phony majority governments, and negative strategic-voting pressure for voters.

if we want to overcome the problems we’ve identified with plurality, we not only need a different voting system—we need one that operates on an entirely different logic. The response to our current problems must involve a rejection of the all-or-nothing logic of electoral competition embedded within plurality in favour of a more inclusive and democratic logic that seeks to make every vote count. This approach would alter both the kinds of results our elections produce (no more wasted votes, distorted election results, or phony majority governments) and the incentives to par-ticipate for both parties (more competition) and citizens (less negative strategic vot-ing). Such an approach also has important implications when it comes to considering concrete voting-system models, causing us to focus our attention primarily on forms of proportional representation.

This chapter will make the case that only forms of PR can make every vote count and thus really address the problems we currently face in Canada. in order to do this,


we will set out how PR systems work and discuss the implications that flow from such workings, as well as show how PR affords better representation, a more legitimate ma-jority government, a more competitive environment for parties, and a more positive space for strategic choices by voters. Additionally, we will assess other reform options, including non-proportional voting systems (e.g., majority and parallel voting systems) and changes to parliamentary practices, and demonstrate why these are insufficient.

How PR Works and its implicationsimagine it is election day in Canada sometime in the future. The results start pouring in and it becomes clear that a majority government has been elected. But in this case, the “majority” really does represent a majority of the voters. Under the country’s new PR voting system, the votes and seats finally add up in a transparent and straightfor-ward manner. The majority government is actually composed of two parties who had clearly signalled to the public during the campaign that they would be prepared to work together if they garnered enough support. Though neither party gained a majority of the vote individually, their combined support did exceed 50%. Given the PR voting sys-tem now in place, that meant they gained a majority of the seats. This result, a refreshing alternative to the distortions and obfuscations of our current plurality system, could easily become the norm in Canadian elections if we moved to proportional voting. in fact, the recent conversion to PR by new Zealand—a country shaped by a similar Brit-ish parliamentary tradition and history as Canada—demonstrates how well PR might fit into our existing institutional arrangements. Most new Zealand PR elections have worked out very much like the scenario sketched out above (Boston et al. 2003: 19).

in turning our attention to PR, we are examining the way most western countries do democracy. PR systems are widely used in the most economically developed coun-tries, with only the UK, Canada, and the United States still continuing to use plurality. Most of the countries that use PR have had extensive experience with proportional voting over the past century. Thus there is plenty of practical experience and a range of approaches to draw from in considering how we might “do” PR in Canada.

if we focus our attention on the election results, the workings of all PR systems are both straightforward and transparent. Basically, if you get the votes, you get the seats. A party that gains 20% of the vote, for instance, stands to gain about 20% of the seats, more or less. Depending on the specific institutional design of the PR system, there may be a slight or more-marked departure from pure proportionality in the election results. But on the whole, PR election results tend to match votes with seats in a transparent, reliable, and mirror-like fashion. Countless academic proportionality indexes measure the gap between votes and seats for parties under different electoral systems, with con-sistent findings: PR systems deliver proportional results (Powell 2000: 96; lijphart 1994: 160–2). Where PR systems do depart from proportionality, it has less to do with PR than with the choices of the system designers or the effects of political gerrymandering. in some systems, like those used in Germany and new Zealand, explicit minimum thresholds were introduced to limit the entry of small parties. Such rules skew the results slightly in favour of the larger parties (lijphart 1994: 22). in other systems, politicians have intervened to reduce the overall proportionality of the system in a bid to aid their

e x p lo r i n g p r o p o rt i o na l r e p r e s e n tat i o n / 51

party and hurt their competitors. in ireland, politicians from the governing parties re-duced the size of the typical multi-member constituency from five to three throughout the 1970s, thus limiting how representative the system could be. When a new govern-ing coalition increased the size of the ridings again in the 1980s, the proportionality of the system improved dramatically (Gallagher 1986: 258–9; Busteed 1990: 64–70).

To say that 20% of the vote will typically result in 20% of the seats under PR sounds clear, but it doesn’t explain just how this result is produced. This is often where com-mentators from non-PR countries claim that everything gets very complicated. if we turn our attention to the internal mechanics of just how PR systems operate, things definitely get more complex. Unlike plurality, where the representation is accorded through a series of all-or-nothing contests, PR shares out the representation as a pro-portion of the total representation available—thus a party with 10% of the vote will be awarded 10% of the seats. The specific mechanism by which seats are awarded in-volves achieving a quota of the total vote. if the whole country is one constituency and there are 100 seats available, then the quota that must be achieved to gain election is 1%. A party that gained 1% of the total votes would have reached the quota for and would be awarded one seat. A party that received 10% of the votes would have achieved ten quotas and thus would be entitled to ten seats. of course, in these exam-ples the math is fairly straightforward—the calculations appear more complicated when there are, say, 156 seats and a party has 17.5% of the vote.

Just who fills the seats won by a party depends on the seat allocation rules in place within the given PR system. in a party-list form of PR, the parties would establish an ordered list of candidates (set in some cases by the party leader, in others by the party members) and the candidates would be elected from the list in that order. There is a great deal of variation among western PR-using countries in terms of how the order of the list is decided. However, most utilize some form of democratic procedures, either by virtue of government statute or individual party policy. Voters too have influence on the structure of the lists in many cases, though there is much debate about how substantive that influence is (Farrell 2001: 83–7).

in an STV form of PR, the voters themselves decide the order by which candidates are elected because candidates achieve the quota for election as a result of the numbered preferences marked by individual voters on their ballots (see figure 4.1). The mechan-ics of seat allocation under PR are clearly more complicated when compared with the fairly simple “most votes wins” allocation method of plurality. on the other hand, the results produced by PR allocations are much more transparent and straightforward than the distorted ones produced by plurality. When it comes to trade-offs, transpar-ent results are ultimately much more important than simple allocation methods.

Debating the Effects of Proportional VotingResearchers studying the long-term effects of PR have underlined a number of general tendencies with these systems. First, PR systems tend to produce minority or coalition governments. This occurs because the systems reflect the popular vote for parties fairly accurately and no one party typically gains a majority of the votes. of course, plurali-ty systems also seldom see a majority of voters support one party. The difference,


however, is that where plurality typically gives a bonus in seats to the largest party—thus manufacturing a legislative majority government out of a minority of votes—PR systems do not exaggerate the legislative results for the largest parties to the same de-gree. Second, PR systems are more open to multi-party competition because the threshold for election is lower and the district from which the votes are drawn is larger. Thus in PR, a new political competitor need only gain a quota of the total votes in a fairly large district rather than having to secure more votes than all others in a much smaller riding, as under plurality. Third, PR systems are typically more broadly rep-resentative, both in terms of the diversity of parties that are included in the legislature and the governing ranks and in terms of the diversity of the individual legislators themselves (Blais 1991: 243–6; norris 2004: 88–93, 187). What these three tendencies amount to is a voting system regularly associated with multi-party coalition govern-ment and a fairly stable multi-party system—one where new competitors have access to the system, where changing coalitions of parties (typically composed of a dominant party and a smaller ally or allies) alternate in government, and those elected better re-flect the social diversity of the polity (Powell 2000).

Critics of PR paint a different picture of its general tendencies, claiming that it contributes to government instability, a fragmented party system, legislative sclerosis, domineering parties, and a lack of clear accountability between government actions and the politicians responsible. These criticisms will be dealt with in detail at various points

mmp Ballot—new Zealand StV Ballot—British Columbia

FigUre 4.1 BaLLotS

Source: Chief electoral office, new Zealand; Final report of the BC Citizens’ assembly on electoral reform, December 2004, p. 4.


in the book. However, it should be underlined now that in making such arguments the critics are seldom comprehensive in their coverage of PR-using countries. instead, they focus on the countries that they feel will best support their claims—typically, italy and israel—while ignoring the bulk of western countries that use PR, such as Germany, Spain, the Benelux countries, Scandinavia, etc. (see Courtney 2004). But as we shall see, a more balanced assessment of the countries using PR undermines all these criti-cisms. in fact, a critical reading of italy and israel—allegedly the exemplars of the sins of PR—will demonstrate that even they do not fit the caricature of PR created by its critics. For instance, italy is alleged to have been very unstable, witnessing “50 govern-ments in 50 years.” But the claim is exaggerated and rests on a dubious definition of “different governments.” in fact, italy had fewer elections than Canada between 1948 and 1992, and the so-called changes in government were more akin to cabinet shuffles in the Canadian system as the new governments often had most of the same ministers as the old ones. The critics also neglect to mention that one party (the Christian Dem-ocrats) was the dominant force in every coalition for four decades and named every prime minister from 1945 to 1981 (Seton-Watson 1983: 113; Furlong 1994: 119, 147). no wonder some claim that italy’s problems actually stemmed from too much stability, not instability (Fabbrini and Gilbert 2000: 34; Radaelli and Franchino 2004: 941–2).

PR systems utilize sophisticated allocation methods to create results that appear intuitive and transparent. Though the counting process can be complex, PR voting rules deliver results that roughly mirror what people have voted for. This key differ-ence between PR and plurality has a host of implications both for the workings of the larger political system and when it comes to addressing the concerns raised about plurality in the previous chapter. Specifically, it will be demonstrated that PR provides better representation, a more effective form of majority government, a more competi-tive environment for parties, and a more positive space for voters to make strategic decisions.

Examining Proportional VotingWe have noted already that, at a minimum, modern democracy involves representa-tion; majority rule; protection of minorities; free and fair elections; and transparent, straightforward democratic institutions. There is much discussion today about modi-fications or alternatives to these aspects—a greater use of referendums or innovative community participation projects, for instance—but few would disagree that the elements set out above form the minimum conditions by which a jurisdiction can reasonably claim to be “democratic.” in the last chapter, we explored in detail how plurality managed to violate or usurp all of these conditions. now, we will demonstrate how only PR voting systems can deliver substantively on their democratic promise.

The PR Approach to Representationin a representative democracy, the quality of representation is crucial. The Swiss voting reformer Ernest naville summed it up nicely over a century ago when he argued that “[i]n a democratic government, the right of decision belongs to the majority, but the right of representation belongs to all” (Hoag and Hallett 1926). in other words, majority


decisions—the hallmark of democratic decision making—have little legitimacy if those making them do not accurately represent the voting public. only proportional voting really addresses this concern. This is because, unlike the winner-take-all ap-proaches of plurality and majority voting, the PR approach to representation is to make every vote count. This means maximizing the ability of individual voters to af-fect the outcome, representing the cumulative voting results for parties accurately, and eliminating the artificial barriers to achieving better representation of our social diversity. Because PR more accurately reflects what people say with their votes, it can lay greater claim to creating a genuinely “representative” democracy. This can be demonstrated by examining the specific ways in which PR systems better represent individual voters, political parties, and our social diversity.

REPRESEnTinG inDiViDUAl VoTERS

it only makes sense that if voters are going to go to the trouble of trooping to the polls on election day, they want to see their efforts amount to something if at all possible. if they are casting their vote for the Monster Raving looney Party or the clown who regularly appears at the local supermarket, the choice of voting system is unlikely to have much effect on their chances of electing a representative. Voting for candidates with little support is a losing strategy regardless of the voting system. But for voters who are casting their ballot for a serious political competitor—which is, after all, what most voters do in the ballot booth (99% of Canadian voters in 2006)—then the choice of voting system matters a lot. only PR tries to make every vote count toward the election of someone. Enid lakeman, the veteran UK voting-system reformer, re-ferred to it as the “power to elect” (lakeman 1982). PR systems are essentially more efficient in turning individual votes into representation because they waste fewer votes and tend to weight votes equally.

The superiority of PR from the point of view of the individual voter can be brought into stark relief by comparing it to the treatment of individual voters under plurality. Under plurality, individual voters will only have an impact on the results if they vote with the most popular choice in their local area. if their choice is not the most popular, their vote will not count toward the election of anyone—it will be, in effect, wasted. A comparison of recent elections in Canada with elections in PR-using countries like new Zealand, Germany, and Scotland shows that while our system wasted nearly 50% of the total votes (2004), the other countries’ PR systems wasted just 1% (nZ 2005), 4% (Germany 2005), and 6% (Scotland 2003), respectively. nor are such results un-usual; on average, Canadian voters wasted 49% of the total votes cast in each election from 1980 to 2006 (FVC 2006).

The gap highlighted here is typical of the different results produced by plurality and PR systems. it is the product of significant differences in institutional design. Plurality wastes more votes than PR because it decides winners on the basis of a winner-take-all formula. if a candidate wins a seat with 40% of the vote, it means that 60% of voters have come up empty handed; multiply that by all the ridings up for election and it amounts to a lot of unrepresented voters. it also wastes votes because it breaks up the total election results into a series of mini-elections in each geographic riding.


What this means is that support spread across riding boundaries cannot be calculated as a whole. These rules aid parties with concentrated geographic support and hurt those whose support is more spread out. PR, by contrast, is much more efficient with votes because it tallies the results for the election as a whole and awards representation on the basis of achieving a quota of the total votes. in this way, a voter needn’t be part of the largest group in a specific geographic area to gain a representative. What this means is that, under PR, a vote will count regardless of where it is cast. The experi-ence of Canada’s federal Green party can help make our discussion more concrete. in 2006 the party gained 664,068 votes—a considerable amount of support. in strictly proportional terms, the votes amounted to almost 5% of the total votes cast. But the party did not elect any representatives because their vote was too spread out; nowhere was it concentrated enough to come in first ahead of all other competitors. But if Canada had been using a PR system, those votes from across the country would have added up to something like 15 seats.

Because PR systems count votes regardless of where they are cast, they are not bi-ased in favour of geographically concentrated political opinions like regional parties or larger parties with solid concentrations of support. This means that individual voters can support new parties or smaller parties whose support is more spread out with less fear of “wasting” their vote. But such voters are not the only voters that have more power. Under plurality, even voters supporting the most popular parties can find themselves relatively powerless if they live in an area dominated by another party. Thus federal Conservative voters in East Vancouver or downtown Toronto, or liberal voters in most of Alberta—despite supporting very popular parties nationally—find themselves in a distinct minority where they live and never cast a vote that counts to-ward the election of a representative. PR would ensure that those voters too had “power to elect.”

PR not only gathers up all the votes and converts them into representation more efficiently than plurality, it does so more fairly because it weights each individual vote more equally. Recall that plurality voting occurs in separate geographic ridings, which may or may not have equal numbers of voters in them, and that the threshold to win a seat can vary among ridings depending on the nature of the political competition. Typically, rural ridings in Canada have far fewer voters than urban ridings, meaning that rural voters can elect a member with far less support than would be needed in an urban riding. For instance, in 2006 rural voters in Kenora-ontario gained a federal MP with just 9,937 votes while urban voters in Toronto-Centre-ontario required 30,874 votes to secure theirs. in this comparison, each rural voter appears to have three times the voting power of the urban voters. However, even if we could divide the ridings such that they had equal numbers of voters, this wouldn’t solve the problem. This is because the threshold for winning in plurality is not fixed, but rather depends crucially on the nature of political competition in a given locale. in plurality, the winning candidate may have 60% of the vote, or 50%, or 40%, or even as little as 26%. if we compare ridings with roughly similar numbers of voters, we can see that the threshold to win can vary widely, meaning that not all votes will have an equal impact on the results.

By comparison, PR accords each voter’s contribution roughly equal weight in the outcome because the threshold for election is more consistent. Thus regardless of


the competitive situation in a given geographic locale or the weakness of a competi-tive party in a certain region, individual voters are more assured that their votes will count and will have an impact similar to every other vote cast in the election. From the point of view of the individual voter, this is a much better deal than the arbitrary, lottery-like results typical of plurality.

REPRESEnTinG PARTiES

Another area of improved representation under proportional voting—in fact, the one that garners the lion’s share of attention among both academics and reformers—involves the representation of political parties. And there is good reason for this. Politi-cal parties, despite a bad rap in the press and from the public, are crucial facilitators of democratic debate and public participation. Without parties, the range of potential issues and political interests becomes hopelessly complex and overwhelming for aver-age citizens. Where parties are absent from political competition, as in many local elections, public knowledge about and involvement in politics drops precipitously. This isn’t a criticism of the public. The complexity of modern society makes an enor-mous number of demands on individuals that largely preclude them from becoming briefed on the intricacies of public policy. This is where parties help guide political choices, as each one tends to represent a different approach to political questions and the concrete policies needed to address them. These choices among parties then help frame public debate and foster turnout in elections by providing voters with different options for responding to the issues (Chandler and Siaroff 1991). And voters confirm this, both by voting largely for political parties rather than independents, and by telling pollsters that party label is the key information they use in making voting decisions (Clarke et al. 1980: 102, 216, 218–9, 222; nevitte et al. 2000: 133–4; Blais et al. 2002: 116–7). With so much riding on parties as the key interface between the public’s myriad views on issues and the need for specific policy decisions, one might expect that accurately representing parties’ voting support would be a paramount value of a democratic institution. We’ve demonstrated that this is not the case with plurality voting. But it is with PR systems.

Put simply, PR systems more accurately represent voter preferences for parties than non-PR systems do. if we examine proportionality indexes for different coun-tries, which measure the gap between the percentage of votes cast for different parties and their ultimate percentage of seats won, the evidence is clear: PR systems match votes with seats. of course, this can happen accidentally in plurality systems, or even consistently where the entry of new parties is unduly restricted for other reasons (as in the US). The point is that, under PR, this relationship of votes and seats is consistent across elections. Thus a new party that knows it has mobilized 10% of the population can likely count on getting 10% of the seats.

The implications of this proportional dynamic for representing parties are many. First, there is the ethical concern. From a “natural justice” perspective as well as from the perspective of plain-old common sense, it’s only fair that parties get what they’ve got coming from the voters. if 20% of voters want Party X, who is to say they can’t have such representation? As the system is supposed to be a “representative” democracy,


FigUre 4.2 proportionaLity inDexeS

Farrell Disproportionality index for Western Countries in the 1990s

Country Voting system Level of disproportionality

Denmark . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 1.55netherlands . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 1.58Sweden . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 2.13Belgium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 3.12germany . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—mmp 3.38Switzerland . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 3.62Finland. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 3.91norway . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 3.92new Zealand . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—mmp 4.36ireland . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—StV 5.39United States . . . . . . . . . . . . . . . . . . . . . . . . . . . plurality 5.43italy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . semi-pr 7.00australia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . majority 10.31japan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . semi-pr 10.78United Kingdom. . . . . . . . . . . . . . . . . . . . . . . . plurality 15.19Canada. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . plurality 15.49France . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . majority 21.37

Source: adapted from Farrell 2001: 157–8.

Lijphart Disproportionality index for Western Countries 1945–1996

Country Voting system Level of disproportionality

netherlands . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 1.30Denmark . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 1.83Sweden . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 2.09austria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 2.47germany . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—mmp 2.52Switzerland . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 2.53Finland. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 2.93Belgium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 3.24italy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 3.25ireland . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—StV 3.45United States . . . . . . . . . . . . . . . . . . . . . . . . . . . plurality 4.90norway . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 4.93japan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . semi-pr 5.03australia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . majority 9.26United Kingdom. . . . . . . . . . . . . . . . . . . . . . . . plurality 10.33new Zealand . . . . . . . . . . . . . . . . . . . . . . . . . . . plurality 11.11France . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . majority 11.34Canada. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . plurality 11.72

Source: adapted from Lijphart 1999a: 160, 162.


the standard for the representation should be set pretty high. in other words, repre-sentation should be as accurate as possible. Second, election results are a form of communication. They relay among voters what they are collectively thinking about the key political concerns of the day. This is another reason why a correct representa-tion of such views is so important for the health of democratic dialogue and debate. Recall that plurality voting tends to distort this communication in a host of ways, presenting different parties as more or less popular than they really are both nation-ally and regionally. For example, in the 1993 federal election, such distortion led to results that mapped the country as completely polarized on regional and ethnic lines. A PR system, on the other hand, would have reflected the real diversity that existed within all regions of Canada, thus laying the basis for some accommodation across this regional issue. With PR, parties could campaign across the country, even in areas where they had limited support because they would know that any votes they received would contribute to their overall results. no longer would certain parts of the coun-try appear one-dimensional in their political interests (e.g., Alberta, Quebec). no longer would minority views in a particular locale be orphaned by the way the votes are added up (e.g., Conservatives in Toronto, liberals in Calgary). And the resulting transparent relationship between votes and seats would allow voters to better under-stand the various political divisions in the country, and the political necessity of working with others to get things done.

REPRESEnTinG SoCiAl DiVERSiTY

in addition to our political party differences and the policy differences they represent, there are differences that transcend party identification that many think also need to be represented in our political sphere. or, to put the issue differently: the consistent absence of certain groups from our representative system raises questions about whether such absences are voluntary, a product of the public’s assessment of repre-sentational priorities, or whether they reflect structural barriers to participation from certain groups. The facts themselves are pretty stark. our legislatures are composed primarily of well-funded white men, with well-funded white women a distant second-ary pool and only a smattering of our greater multicultural and economic diversity present at all. We know the results have little to do with public attitudes, as Canadians are either keen on diverse representation or are not prepared to vote against candi-dates who represent diversity. Still, women make up just 20% of the federal Parliament, a figure that has hardly changed since 1993 despite the fact that Canadian voters have said they would be happy to vote for a female candidate. By contrast, countries with similar attitudes toward women as Canada that use PR systems have produced results that are dramatically different—and better—than our own. All evidence suggests that where demands for diverse representation emerge in PR systems, they are much more quickly translated into political representation than in our all-or-nothing system in Canada (Rule 1987; Reynolds 1999; Kentworthy and Malami 1999; norris 2004).

A fascinating example of this dynamic was recently highlighted in new Zealand when the country switched from plurality to PR voting. Prior to the switch to PR in 1996, new Zealand’s representation of women stood at 21%; other aspects of diversity


were poorly represented in the legislature as well, despite the country’s considerable Aboriginal population and regional pockets of multiculturalism. But in the first election under PR, women’s representation increased to 29% while indigenous representation rose from 7% to 13%, essentially mirroring the proportion of indigenous people within the population at large. nor were these results a fluke. Since 1996 three more PR elections have occurred; the addition of representatives reflecting the country’s multicultural makeup has only served to diversify new Zealand’s new Parliament even further. After the 2005 election, women’s representation had risen to 33%, indige-nous representation to 16.5%, and visible-minority representation to 4.2% (Vowles et al. 2006: 274; levine 2004: 657). Though attitudes about racial and gender diversity among new Zealanders are complex, there is some evidence of increasing levels of public support over the past few decades for a more diverse characterization of new Zealand’s national identity (lui and Sibley 2006). The switch to PR has helped represent that shift by offering very different incentives to parties to respond to social demands for diversity.

Basically, PR beats out plurality on diversity because it avoids plurality’s all-or-nothing logic. Recall that in single-member plurality systems, the problem is rooted in the nomination battles occurring in each separate riding. Given their social and economic advantages, men tend to come out on top. But because all PR systems use multi-member ridings to some degree, parties can avoid this all-or-nothing battle at the nomination stage. if a party is ordering a list of candidates for a multi-member riding, they can balance their list with men and women or whatever range of diversity they see fit. Thus the structure of competition under PR is an important factor in ex-plaining why parties can more easily offer a diverse range of candidates at election time (norris 2004: 188–90). of course, in noting this, it must be underlined that those seek-ing more diverse representation are not a homogeneous group. For instance, while it seems likely that a shift to PR in Canada would aid both the country’s ethnic minorities and its indigenous peoples in increasing their levels of representation for the reasons set out above, there are many other challenges to reaching this goal, and the nature of such challenges may diverge sharply among the many groups seeking diversity.

Some critics object that crediting the voting system with producing more diverse representation goes too far. instead, they argue that the countries with better repre-sentation of women also tend to be countries with more progressive attitudes about women. Thus seeing more women in a parliament is really about changing cultural or political values rather than institutional rules (Courtney 2004: 152–4). Such argu-ments, however, can’t explain why some countries with fairly positive views about the role of women in public life have nonetheless failed to elect many women to public office. And the fact that these countries all tend to be ones still using plurality only further undermines the argument. of course, cultural and political values do matter. The introduction of PR does not create diverse representation out of nothing. After all, the countries credited today with great representation of women—i.e., the Scan-dinavian countries—did not always do so well in this area. For the first 50 years of its use, PR did not lead to many women getting elected in these countries either (Bysty-dzienski 1994: 59; Matland 1993: 747–8; Sainsbury 2004: 71–3). Thus the fact that, structurally, PR systems can better accommodate diversity does not actually tell us


why they do so at different times. Here, we need to turn to how the competitive pro-cess under PR tends to make parties more responsive to changing public opinion and the mobilization of new ideas in civil society.

A concrete example will demonstrate the interactive process that fuels the poten-tial for more diverse representation in PR systems. if we look at the historical process of improving women’s representation in the Scandinavian countries, it becomes ap-parent that nothing automatic occurred to produce these countries’ now-impressive results. The process began with the rise of a Scandinavian women’s movement and the entry of large numbers of women into the paid workforce and political parties. But as this also occurred in Canada and the United States—countries still lagging comparatively on women’s representation today—clearly that was not the only factor. it also required a party that would be prepared to take up the issue. Here, a significant difference emerges: the competitive nature of the PR system gave the Scandinavian women’s movement much more leverage than plurality gave their north American counterparts.

like women elsewhere, Scandinavian women took their concerns initially to the political left, as they felt those parties would respond more readily to a shift in tradi-tional values concerning gender (Skjeie and Siim 2000: 352–4). While not entirely

LEGEND

0

10

20

30

40

50

60

70

80

90

10020051975

Swed

en

Finl

and

Denm

ark

Neth

erla

nds

Norw

ay

Spai

n

Belg

ium

Aust

ria

Germ

any

New

Zea

land

Swit

zerla

nd

Irel

and

Ital

y

Aust

ralia

Cana

da

Unit

ed K

ingd

om

Unit

ed S

tate

s

Fran

ce

Perc

ent

Plurality/Majority Proportional

FigUre 4.3 Women’S repreSentation UnDer proportionaL repreSentation anD pLUraLity/majority Voting SyStemS

Source: adapted from Kittleson 2006; international parliamentary Union 2005.


wrong in their assessment, the women were not impressed at how slowly the political left responded. But as PR systems are more open to new parties and party competition, the women had some options. in the end, Scandinavia’s female political activists either mobilized across party lines; threatened to start their own parties; and/or affected the behaviour of a key party, which had the effect of forcing other political contestants to pay attention. in norway, a cross-party women’s alliance made a breakthrough at the municipal level in 1971 by organizing women to give voting preferences to women. The result was that women’s representation in a number of major cities increased from 20 to 50% (Matland 1993: 747–8). in Sweden, female members of the leading Social Democratic Party threatened to launch their campaign for more elected women inde-pendently of their party. At the same time, the broader women’s movement influenced the small Socialist left party to move decisively on women’s representation, which had the effect of forcing the larger Social Democratic Party to act as well. And as the left re-sponded by running more female candidates, it created a “contagion effect”—other parties felt compelled to respond or risk allowing the left to be the only beneficiary of the votes of those who might vote on gender lines (Jenson and Mahon 1993: 86–91; Sainsbury 2004: 71–4). The point is that achieving diverse representation requires a social dimension—political demands and organizing—even under PR. Where coun-tries lack a women’s movement or social attitudes conducive to women’s participation in politics, PR will not do much to change things. But where such demands do exist, the more competitive environment for parties under PR tends to make them more sensitive to shifting public views, and the structure of political competition—particu-larly the existence of multi-member ridings—makes it easier for parties to respond (Matland and Studlar 1996; norris 2004: 188–90).

When it comes to representation, PR systems are demonstrably superior to winner-take-all approaches to voting like plurality or majority systems. PR systems better represent individual voters because they efficiently translate most votes into represen-tation—few votes are “wasted” in PR systems. Proportional voting also tends to weigh votes equally in determining the results. Thus, for an individual voter, PR provides a more reliable and fair “power to elect” than winner-take-all methods where the voter’s physical location makes all the difference in whether their vote counts and for how much. PR also better represents parties by more accurately translating party votes into seats, both at the national and local level. Finally, PR systems better represent the social diversity of modern democracies by both responding to changing social de-mands and structuring political competition in such a way as to allow parties to offer diverse slates of candidates at elections. The “representation” part of representative democracy cannot be sacrificed as a tradeable value for “stability” or other such non-democratic rationales. only PR can accommodate the complex representational needs of modern democracies.

Governing Majorities, Party Competition, and Individual Voter Strategy Under PRBeyond better representation, a shift to a form of proportional representation would lead to other key changes in Canada’s political system as well. For instance, the use of


proportional voting would fundamentally alter the nature of our governing majorities, competition among political parties, and the strategic incentives affecting individual voter decisions in the ballot booth. Under PR, the term “majority government” would mean that the government actually enjoyed a majority of the voting public’s support. Party competition would increase, as the barriers to new participants would be low-ered and votes would be more efficiently and fairly translated into representation. And as most votes would count toward the election of someone, voters would escape the negative strategic considerations inherent in plurality voting about “wasting” one’s vote. Each of these changes—affecting majority government, party competition, and strategic voting—would mark a profound improvement in Canada’s electoral democracy. As such, each bears closer scrutiny to set out just why PR would produce these results.

MAJoRiTY GoVERnMEnT

one of the key arguments for plurality voting and against PR hinges on the desirabil-ity of single-party legislative majority government. Most defenders of plurality are so keen on this feature that they are prepared to distort the election outcome to get it (i.e., allow a party with a minority of the vote to gain a majority of the seats). As stat-ed previously, the ethical arguments justifying such a practice are either dubious or non-existent. Plurality’s defenders instead rely primarily on pragmatic arguments—specifically, that democracy requires a stable government that can efficiently intro-duce its program, and one that voters can subsequently pronounce on at election time either by re-electing it or by “throwing the rascals out” (Courtney 2004: 137–9). This assumes that PR governments are not stable, not able to introduce their policies effi-ciently, and not accountable to voters for what they do. But there is little evidence to support these broad characterizations of either plurality or PR. While phony majority governments under plurality can “efficiently” introduce policies, it is not clear that such practices are either democratic or desirable. nor is it clear that the public can “throw the rascals out”—that will crucially depend on the nature of political party competition. on the other hand, when the evidence is examined, PR performs well on government stability, getting policies introduced, and being accountable to voters.

A good example of the efficiency and accountability problems linked with plurality is the long rule of the Thatcher Conservatives in Britain. Though Thatcher’s then radical right-wing policies never enjoyed majority support in the country, the inabil-ity of the public to settle on an alternative party throughout the 1980s meant that her Conservatives could always turn a minority of votes into a majority of seats. Similarly, during the 1990s in Canada, provincial governments of both the right (ontario Con-servatives) and left (BC nDP) introduced policies that a majority of their citizens opposed, but vote splits among the opposition parties allowed them to stay in power. To note that all these governments were eventually defeated hardly justifies plurality’s claims about efficiency and accountability. The “efficient” introduction of policies that most people do not want cannot be characterized as positive by any democratic criteria. nor were the various publics in these cases able to “throw the rascals out” in a timely manner. in most cases, the “accountability” took ten to fifteen years—hardly a


mark of efficiency. And it often came at a considerable cost. in the case of Britain, the evolution of new labour as a response to the Conservatives had the effect of dramati-cally narrowing the range of policy debate and political choices for voters (Blyth 2003).

The critical claims about the performance of PR governments are no more com-pelling. There is little evidence to support the view that PR governments are unstable, short-lived, ineffective, or unable to introduce policy. Studies of PR government per-formance in western countries comparable to Canada demonstrate that the average length of a PR government is nearly the same as that of a plurality government: 1.8 years for PR governments versus 2.5 years for plurality governments (Woldendorp et al. 2000: 79). of course, much depends on the measurement of what is considered a “government.” Many researchers consider any change in the prime minister or cabinet to constitute a change of government. Thus Mulroney’s Conservative government, which was re-elected in 1988, is counted twice in these comparative figures because near the end of its term Kim Campbell took over as the prime minister. Realistically, however, this change did not represent a change of the government in the Canadian context. A more effective measure of stability or instability would be to compare the number of elections in different countries, as this suggests the degree to which politi-cians can govern without recourse to formal public input via voting. Here, we can see that there is virtually no difference in the frequency of elections on the basis of voting-system use in western countries: between 1945 and 1998 plurality-using countries averaged 16.7 elections while PR-using countries averaged 16 elections (iDEA 2007).

in terms of economic health, quality of life, and innovative policy responses, western PR governments have some of the best performance ratings (lijphart 1999a: 263–70,

FigUre 4.4 goVernment LengtH in WeStern CoUntrieS, 1945–1998 (in DayS)

Country Voting system average governing days

italy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 331France . . . . . . . . . . . . . . . . . . . . . . . . . . . . . majority 336Finland. . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 404Belgium . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 511Denmark . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 638germany . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—mmp 660australia . . . . . . . . . . . . . . . . . . . . . . . . . . . majority 661Sweden . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 752norway . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 774new Zealand . . . . . . . . . . . . . . . . . . . . . . plurality 794netherlands . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 879ireland . . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—StV 900Canada. . . . . . . . . . . . . . . . . . . . . . . . . . . . . plurality 947United Kingdom. . . . . . . . . . . . . . . . . . . . plurality 995

Source: Compiled from Woldendorp et al. 2000: 79.


1999b: 320–3). Here too, though, much depends on how one defines vague terms like “efficiency” and “stability.” Most of the negative claims made about PR systems focus on the fact that government control is seldom in the hands of a single party. instead, PR countries enjoy minority or—more typically—coalition governments. in fact, these governments are arguably more stable and efficient than the phony majority govern-ments of plurality systems. They are more stable because they are more inclusive and responsive to different views. This avoids the policy extremes typical of plurality where Party A routs Party B’s policy legacy when elected and Party B returns the favour when they get back in power (lCC 2004: 143). Powell argues that, by contrast, PR elections tend to produce governments that are closer to the “median voter,” i.e., the position that can embrace the broadest range of views (Powell 2000: 182, 186, 193–6). not surprisingly, policy under PR develops more collaboratively—and some would argue it is ultimately more successful—precisely because there is more consensus behind it (lijphart 1999b: 325). This is a different, and more substantive, way of understanding stability and efficiency.

While critics of PR try to cast the debate as a choice between stable majority gov-ernment and unstable minority/coalition government, the choice is really between phony majority government and real majority government. Majority governments under plurality almost never represent a real majority of the voters; they are “phony” because they are the product of the workings of the system, not the desires of voters. At the very least, such governments represent a tyranny of the minority—the obverse of most basic democratic principles and common sense. But coalition government under PR represents a real majority government precisely because that government usu-ally does represent a majority of the voters (Powell 2000: 100). Such majorities have

FigUre 4.5 nUmBer oF eLeCtionS in WeStern CoUntrieS, 1945–1998

Country Voting system number of elections

germany . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—mmp 14italy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 14norway . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 14Finland. . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 15ireland . . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—StV 15France . . . . . . . . . . . . . . . . . . . . . . . . . . . . . majority 15United Kingdom. . . . . . . . . . . . . . . . . . . . plurality 15netherlands . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 16Belgium . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 17Sweden . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 17Canada. . . . . . . . . . . . . . . . . . . . . . . . . . . . . plurality 17new Zealand . . . . . . . . . . . . . . . . . . . . . . plurality 18Denmark . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 22australia . . . . . . . . . . . . . . . . . . . . . . . . . . . majority 22

Source: Compiled from iDea 2007.


proven stable and effective in responding to the complex challenges of modern gov-ernment. in fact, the diversity embedded in such majority governments (which results from the fact that they are an explicit coalition of different parties) is part of their strength. Such diversity ensures a greater mobilization of the public, a greater range of policy perspectives, and more debate within both Parliament and the press about the substance of political differences. of course, there is nothing preventing a single-par-ty majority government from coming to power under PR—all that would be required is that the public actually vote for it. Even minority governments under PR have per-formed well, proving much more stable and effective than minority governments under plurality (Strom 1990: 89).

These real majority governments under PR have many advantages over the phony majorities created by plurality. For instance, they are not just a better form of govern-ment—they are a better form of politics. The phony majority government process of plurality reproduces the winners-and-losers approach to political contestation that focuses attention on leaders and horse-race-style elections. Voters may attend to the process while the race is on but then are expected to retire from the political field when the “elected dictatorship” takes power and effects its stable control of the system. Under PR, however, the politicking continues after election day, contributing to an ongoing public dialogue about politics and a more deliberative approach to political contestation. The process of government formation after the PR election may result from a pre-election agreement among parties or a post-election set of negotiations between the winning parties (Powell 2000: 74). Either way, the media must comment on—and the public is exposed to—the process and substance of potential political cooperation and disagreement. Henry Milner argues that this dynamic encourages parties to alter their behaviour too by increasing their efforts to educate voters on the key issues (Milner 1997, 2004a). This raises the level of public awareness about the real problems of political differences and the necessary trade-offs required to gain broad consensus on issues. instead of the all-or-nothing mentality of plurality politics, PR forces people to face the reality of political pluralism—that people disagree.

The style of governing effected by real majority governments under PR is also dra-matically different. Unlike plurality, whose phony majority governments are effectively insulated from parliamentary or public pressure between elections, a coalition major-ity government under PR can ignore neither its constituent member parties nor the public. ignoring a member party of a governing coalition could affect the govern-ment’s majority, so the government listens to its partners and shows more respect for internal differences (Powell 2000: 215–20; Martin and Vanburg 2004). And when the public becomes mobilized around an issue, it may have more influence precisely be-cause there is more political diversity within the governing ranks (lupia and Strom 1995). What this means is that a publicly engaged form of competitive politics con-tinues between elections, not just at election time.

PARTY CoMPETiTion

The key factor that gives PR systems these governing and political dynamics is effec-tive political party competition. in other words, PR systems allow voters to punish and


reward party behaviour more freely, more accurately, and more quickly than plurality systems. Duverger’s observations about the mechanical and psychological effects of voting systems are relevant here—PR systems have lower thresholds for election and a more efficient way of converting votes into seats (lijphart 1994: 23; Powell 2000: 96). Thus if voters wish, a greater diversity of political opinion can be represented. And because PR operates in a more representative fashion, the psychological barriers to supporting a new or small party also disappear. All this adds up to a more competitive form of political contestation, effectively altering the incentives that influence how parties behave.

For instance, because PR systems do not inflate the support of bigger parties or block the entry of new parties as plurality tends to do, parties must work harder in PR systems to maintain their support and attract new voters or risk losing out to new players. Thus if new ideas come onto the scene in PR systems, such as an increased public interest in the environment, they will be reflected politically much more readily. While both Europe and north America have registered increasing interest in envi-ronmental issues and political parties focusing on them, only Europe has seen a signif-icant electoral breakthrough for Green parties. This is mainly because PR has not barred their entry with unduly high thresholds to gain election. or if existing parties fail to follow through on their promises while in office under PR, voters can shift their support to other parties or support a new party without fear of “wasting” their vote. Under PR, even small shifts in support can send a strong signal to a party not to take its voters for granted. And unlike plurality, where a voter unhappy with Party left usually only has Party Right as the alternative option, a PR voter can usually find an-other party with views that are at least somewhat similar to those of the party he or she is presently unhappy with. PR also allows voters more say in governing coalitions, as shifts in support among the government’s coalition partners tend to shift the weight of influence within cabinet to the parties gaining more support (Powell 2000: 173–4).

The critics of PR don’t like the fact that it is so openly competitive. instead, they much prefer more restrained party competition, ideally limited to a two-party system. Furthermore, the critics argue that PR’s open form of political competition leads to a proliferation of parties, political paralysis and instability, and an undue influence accorded to smaller parties in allegedly dictating policy and holding the system to ransom. Again, such claims are seldom backed up by a fair representation of the facts. Taking each in turn, it can be shown that none of the criticisms can be maintained.

First, the fact that PR systems may allow for greater political competition does not mean that PR automatically causes more parties to emerge. The number of parties in PR-using countries ranges from few to many. But the raw number of parties gives a misleading sense of what is going on under PR. Just as plurality systems have major and minor parties, so do PR systems. Assessing the major parties, PR systems have similar party-system dynamics to plurality—typically, two major parties or party blocs, and a few minor parties who form the coalition partners of the majors. in fact, plurality systems have proven prone to multi-party outbreaks on a consistent basis, as British and Canadian experiences can attest to. As a result, researchers analyzing the “number of parties” question do not treat every party the same. After all, that would


mean that the UK would be considered a strong multi-party system from the 1970s on, as the liberal party revived and the smaller regional parties gained a foothold in Parliament. But as the latter parties were tiny, representing few votes and seats, research-ers “count” them for less. For instance, in the 1974 election, three regional parties gained just shy of 6% of the popular vote and won 26 seats while the two largest par-ties won 596 seats between them. Though the liberals won 18% of the votes, they won only 13 seats. Realistically, despite regional parties, the UK is seen as a two-and-a-half-party system, with labour and the Conservatives as the major parties and the smaller liberals as the “half” (given their voting support, if not seat-winning potential). Factoring in the size of parties, then, research suggests very little difference between plurality and PR systems—typically, three parties under the former and three to five in the latter (norris 2004: 82–5; lijphart 1994: 104–5). PR party systems, specifically as concerns the number of parties, have also tended to remain stable over long periods of time (Mair 1991: 49–53).

Second, these party systems in Europe have generally avoided both paralysis and instability in the postwar period. Some critics try to reach back to the interwar period in Europe to condemn PR, particularly noting the experience of Weimar Germany and its chronic economic, social, and political instability (Wiseman 1997). But seri-ous scholars of both Germany and voting systems acknowledge that the country’s problems in that period were beyond what an alternate voting system could have fixed (Pulzer 1983: 92–3, 104–5). Besides, Germany has used PR since 1945 with stable and positive results, effectively debunking the simplistic “PR equals instability” for-mula (Klingemann and Wessels 2001).

Third, the allegation that small parties unduly dominate PR systems is not borne out by practical experience in western Europe. The relevance and impact of small parties in different countries will depend on many factors, including cultural norms of inclusion, where the parties sit on the ideological spectrum, their size, etc. (Muller-Rommel 1991: 5–6). normatively, the bias against small-party influence is difficult to defend. Small parties deserve to be heard as legitimate representatives of minority views, and the larger parties should listen to them because they themselves have not gained a majority of the public’s trust—they too are “minority” parties, though larger ones. And small parties only gain influence if bigger parties let them. After all, larger parties could cooperate among themselves and cut the smaller parties out. This has, in fact, occurred on many occasions in PR systems where the larger parties have formed a “grand coalition” precisely because they felt small parties were being too de-manding (e.g., Germany from 1966 to 1969). Finally, the public in PR systems has demonstrated little patience with smaller parties that make unreasonable demands, tending to punish them in subsequent elections. A famous example of this occurred recently in new Zealand, when the erratic new Zealand First party overplayed its hand in coalition negotiations and in supporting the government—their support dropped considerably in the next election (nagel 2004: 135–6).

The anti-PR arguments rooted in a fear of effective political party competition fail for many of the same reasons that anti-PR arguments against effective representation fail: they violate basic common sense about democracy. As the name of the system is “representative democracy,” it should represent people effectively and accurately. So


too is modern representative democracy about competition. For instance, the whole point of the western critique of former eastern-bloc claims to “democracy” was that there was no real competition in those systems (Pravda 1978). Setting aside more substantive or participatory notions, the basic view of democracy in our system is that voters should be able to choose their representatives in a free and fair election, and that accountability can be effected by sometimes deciding not to choose them again and instead choosing others. Anything that interferes with this basic logic only further limits even this rather modest approach to democracy. Given that we know that voters use party label as the key information to make voting decisions, by what demo-cratic criteria can we argue to limit the competitive dynamic of the system vis-à-vis party representation? The fact is that there are no “democratic” arguments for limiting party competition, only undemocratic ones. leaning on rationalizations about stabil-ity versus instability tends to obscure this. Democracy is about change, or the potential for it, and as such is inherently unstable. That is its strength, not its weakness.

STRATEGiC VoTinG

in addition to altering the incentives for parties, the more competitive electoral environ-ment under PR offers different incentives to voters. Basically, PR dramatically reduces the pressure for negative strategic voting, allowing voters greater latitude in making

FigUre 4.6 aVerage nUmBer oF partieS aCroSS WeStern CoUntrieS, 1945–1996

Country Voting system average number of parties

Switzerland . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 5.24Finland. . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 5.03italy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 4.91netherlands . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 4.65Denmark . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 4.51Belgium . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 4.32japan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . semi-pr 3.71France . . . . . . . . . . . . . . . . . . . . . . . . . . . . . majority 3.43norway . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 3.35Sweden . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—party list 3.33germany . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—mmp 2.93ireland . . . . . . . . . . . . . . . . . . . . . . . . . . . . . pr—StV 2.84United States . . . . . . . . . . . . . . . . . . . . . . . plurality 2.40Canada. . . . . . . . . . . . . . . . . . . . . . . . . . . . . plurality 2.37australia . . . . . . . . . . . . . . . . . . . . . . . . . . . majority 2.22United Kingdom. . . . . . . . . . . . . . . . . . . . plurality 2.11new Zealand . . . . . . . . . . . . . . . . . . . . . . plurality 1.96

Source: adapted from Lijphart 1999a: 76–7.


strategic decisions about who to vote for. Under plurality, voters constantly face re-strictive, multidimensional limits on the choices they may want to make because the all-or-nothing dynamic of the system punishes everyone but the most popular candi-date. Voters must try to figure out in advance who will be most popular and then decide strategically how best to act. This is negative for a host of reasons. First, voters cannot gain the necessary information to make such a decision, and what information is available is usually inadequate (i.e., national polls on party standings) or biased to-ward those who have paid to gather it. Second, what voters want to vote for under these conditions becomes less important than what they think they can get. Thus they may be voting less for something than against something else. Third, their need to act in this way is dictated by the system design, not by a genuine strategic problem (such as the absence of other supporters for their choice). in other words, a local choice is often only “un-strategic” under plurality because the voter cannot make common cause with other supporters of the same choice who live outside the locale but are voting in the same election. PR systems, by comparison, reduce and often eliminate this nega-tive strategic pressure with voting because locale is much less important, more votes are effectively translated into representation, and individual votes weigh more equally in the overall results.

Essentially, PR provides more straightforward approaches for voters to get what they want politically. Some political scientists complain that trying to eliminate stra-tegic voting is impossible—there will always be a strategic dimension to voter choice. But here, they are confusing different things. of course, wherever there is a choice to be made there will be strategic elements to consider. The question is whether the strategic dimensions are fundamentally negative or positive. negative strategic voting limits the choice-making process by design, regardless of the substance of the choices to be made. By contrast, positive strategic voting is about a voter weighing what they want to accomplish with how they think they can best get it. For instance, a voter in a PR system might shift his vote within the coalition of governing parties to help tip the policy balance in a certain direction. Here, we have a voter making a strategic de-cision about how to act to further his political goals in a positive way, in a way that links what he wants to do with what he can do. Though the evidence for this sort of voter behaviour is mixed, parties nonetheless have to respond to the more dynamic form of competition in PR systems (Strom 1997). negative strategic voting is all about the “best of the worst” or the “lesser of two evils,” where voters feel compelled to vote for something they don’t want in order to avoid getting something worse. The classic examples in the Canadian context involve voters from both left and right cast-ing their ballots for the centre candidate in a bid to forestall victory by their ideological opposite. Clearly, there are some important differences between these two kinds of strategic voting. With few justifications for retaining the negative kind, we can limit or eliminate it with PR.

The introduction of PR in Canada would do more than just better represent Canad-ians and what they say they want at the ballot box. it would also fundamentally alter the way our governments are formed and maintained in power, how parties compete, and the strategic pressures facing voters when they have to make voting decisions. PR sys-tems offer a chance to gain the benefits of real rather than phony majority government,


increased political party competition and the greater choice and accountability that would flow from it, and a more straightforward strategic situation for voters when they come to make decisions about how to vote. Such changes would mark a consider-able improvement in the Canadian democratic process and would undoubtedly affect the way our politics works for the better. Challenges would remain, of course; PR is no panacea for solving real political problems. What PR does offer us is a different and better way to take up those challenges.

False Alternatives: Majority/AMS Systems and Parliamentary ReformSome critics of our present plurality system do not agree that PR is the best alterna-tive. For some, the key problem in our current way of doing things is that a candidate can win a riding with less than a majority of the vote. They recommend switching from plurality to some form of majority voting to address this particular problem. others call for what are popularly referred to as “mixed” systems—usually, our pres-ent system with some small measure of proportionality added on (though far less than a proper proportional representation of the popular vote for the different par-ties). Still others claim that the focus on the voting system is beside the point because it will do little to change how politicians and their parties behave once they are elected. instead, they believe that the solution lies in the reform of Parliament and parliamen-tary rules and/or processes. What is important to underline here is that these proposals have very different aims and agendas that are often much more meagre and partial than those associated with PR. They are “false alternatives” to PR because they cannot address the shortcomings of plurality that any PR system can. of course, to some, they appear to be attractive stopgap measures when compared to the problems with plurality. But taking each reform in turn, it will be argued here that such approaches will fail to address the problems of plurality or match the promise of PR.

Non-proportional Voting-System ReformsA good number of people at election time, particularly the partisans of the major par-ties, notice that plurality winners fail to gain the support of a majority of voters in their district. This is what fuels a great deal of speculation and rumour mongering about vote splitting and third-party candidates. For the major parties, new competitors may split the vote and allow another party to “go up the middle” between them and the new entrant. not surprisingly, they are keen on a reform that will alleviate this problem. Various forms of majority voting address this issue by assuring that the winner in a district really does gain a majority of the vote, either through run-off vot-ing (as used in France) or by allowing voters to indicate second and third preferences on the ballot as with the alternative vote (as used in Australia). With this voting sys-tem in place, parties and voters needn’t worry about vote splitting, as less-popular candidates will be eliminated until the race is between the two strongest competitors. Tom Flanagan promoted majority voting for just this reason to aid Canada’s right


when two right-wing parties were competing against each other federally from 1993 to 2001. Flanagan’s thinking was that, if right-wing voters could determine which right-wing party had the advantage in their riding, they would probably vote for it (Flanagan 1999, 2005). As it was, particularly in ontario, the two parties were just cancelling each other out and helping the liberals win most of the seats. of course, right-wingers were not the only ones who thought majority voting was a good idea. Some on the left thought majority voting could aid the nDP against the Greens, or prevent the right from benefiting from vote splits between the liberals and the nDP. Some even argued that majority voting might help new parties because potential sup-porters wouldn’t need to worry about “throwing away their vote” on a new and un-tried entity. While all this sounds promising, the practice of majority voting ultimately doesn’t deliver.

Majority voting is really a solution for the major parties and mostly serves to en-trench their superior position in the system even further. At least under plurality, a small party can hurt a larger party and sometimes force them to acknowledge them and their issues. This is not the case under a majority system because such parties will be eliminated in secondary rounds of voting or transfers of ballots. Majority voting has the effect of funnelling support back to the major parties, requiring them to do little in the way of political horse-trading to get it. And while majority voting may as-sure that winners in single-member ridings receive something near a majority of the local support to gain election, it does nothing to address the other glaring problems with the plurality system. Majority systems, like plurality, also tend to manufacture a phony legislative majority government out of a popular-vote minority. Majority sys-tems do not proportionally reflect the voting support for different parties or improve political-party competition. in fact, they offer up the same barriers and distorted op-portunities as plurality, particularly the overrepresentation of regional parties and national parties with regional pockets of support (Farrell 2001: 60–4; Jansen 2004). Majority systems do not better reflect social diversity. As with plurality, they typically operate in single-member ridings, thus offering just one representative to represent the diverse interests and identities of a local area (Rule 1987; Reynolds 1999; Kent-worthy and Malami 1999). Basically, majority systems ease the pressure of negative strategic voting by limiting vote splitting and some wasted votes for the major parties. But they do little to increase an elector’s “power to elect” or to make politics more accountable to the voters. As such, majority voting is a subpar reform if our goal is to make every vote count, end the tyranny of phony majority government, and make parties more accountable to their voters.

other popular suggestions for voting-system reform in Canada of late, particularly among academic elites, are less-than-proportional “mixed” or additional-member systems (AMS). These should not be confused with what are fully proportional “mixed” systems, like the mixed-member proportional option recently chosen in new Zealand. non-proportional mixed systems look similar to MMP, but the ratio of single-member ridings to list seats is heavily weighted toward the former and the system does not seek to make results for political parties proportional. instead, the point of such schemes is to basically leave our system the way it is, composed mostly of single-member plurality ridings, but then add a dollop of extra representation that can be


used to address particular interests and problems. The number of extra representa-tives in the various mixed proposals varies considerably (irvine 1985: 72–88; Jansen and Siaroff 2004). one proposal called for just 10% of the total representation in the House of Commons to comprise the additional members (Weaver 1997). Such a small number of extra seats would do little to alter the typically distorted results of the main plurality election. nor could such a meagre proposal do much to better the rep-resentation of diversity, either for parties or society, or prevent the election of a phony majority government.

So what would be the point of making such change? Typically, proposals for non-proportional mixed systems eschew democratic principle and rely on pragmatic jus-tifications for their design. in the 10% mixed model noted above, the modeller sought to aid the major governing parties who are typically shut out in different regions of the country, particularly the liberals in Alberta and the Conservatives in Quebec. Regionalism had been the major factor pushing voting-system reform in the 1970s and 1980s in Canada, and most proposals in this period expressly sought to address the regional polarization of the country’s party system. But despite many reports and much money spent studying the question, the problems of regionalized politics proved an insufficient motivation for voting-system reform. The 10% mixed model defended its meagre offering by claiming that its very modesty would make it attrac-tive to politicians fearful of change; such an analysis has proven to be, alas, mistaken (Weaver 1997: 508–9). The politicians have demonstrated little interest in such tin-kering reforms. Besides, the problem of regionalism is now just one of many issues pushing a public reconsideration of Canada’s voting system. These underwhelming mixed proposals are too modest in their aims to answer the problems now raised on the question of electoral reform. The models appear to have fallen between two stools: they are far too limited to attract public attention and support, and yet are still too threatening for political elites to act on. And, as with the majority system, these pro-posals will not address the problems at the root of the plurality system—specifically, distorted representation, phony majority government, and an uncompetitive party system. The time for elite-negotiated half-measures is long past.

Parliamentary Reformof course, some feel that the focus on voting systems is misplaced. Specifically, they believe that a change in voting system will do little to alter the behaviour of parties or their members once they are elected to Parliament. instead, they argue that what we need is a change to Parliament itself. This might involve altering how committees work; strengthening the hand of the legislature over the executive (i.e., the PM and cabinet); limiting or weakening party discipline; or a host of other administrative or procedural reforms. The basic view here is that, regardless of the voting system, poli-ticians and their parties will still be central to the political process and there is little reason to believe that they will act differently. For instance, Smith and Aucoin argue that the introduction of PR would do little to alter the executive dominance of the Canadian parliamentary system, which is what a lot of commentators focus on as a key problem in our governing approach (Aucoin and Smith 1997). Such views assume


that a change in the voting system will not affect party actions or government func-tioning or an individual legislator’s behaviour.

But the gist of our argument thus far is that a change to the voting system will do precisely all these things. The end of phony majority government will lessen executive dominance because coalition or minority governments must be more mindful of governing partners and of Parliament itself. Party leaders will have to attend to what their legislative followers want and think a bit more closely because each vote will be needed to get things done. And individual legislators will have more meaningful roles in their parties, in the legislature, and in committees. Frankly, the “parliamentary re-form” line is hardly new and has precious little to show for all its efforts (Massicotte 1999). And that has a lot to do with our present electoral system: phony majority governments may mouth platitudes about reform, but nothing is forcing them to act differently. However, in a system where no party has a majority and the electoral competition for voter support is fierce, the incentives for parties to change their be-haviour are very different. ironically, as other experts on legislatures have noted, we may do more for parliamentary reform by getting PR than by settling for parliamen-tary reform instead of PR (White 2007).

Majority voting, non-proportional mixed systems, and parliamentary reforms are false alternatives to proportional representation. none gets to the root of the prob-lems embedded in plurality voting—specifically, bad representation, phony majority government, restricted political party competition, and poor incentives to participate for both voters and parties. Both majority voting and the mixed systems offer solutions to problems for some, mainly for the existing major parties. They do not answer the complaints of the Canadian public or improve upon the democratic quality, substance, and justice of our electoral process. But we require much more to really address the problems inherent in the plurality voting system. At a minimum, we need a system that effectively represents our political and social diversity; ends the tyranny of phony majority government; creates real competition among parties large and small, old and new; and does away with the negative strategic pressures affecting voters under the current system. none of the false alternatives can do this. nor can mere parliamentary reform. PR, by contrast, can.

Conclusiononly some form of PR has the potential to give Canadian voters what they vote for. Proportional voting would effectively represent what voters say with their votes in terms of party representation and open up the competitive arena to new ideas and more diverse representation, if that is what Canadians want. PR would end the dispir-iting practice of wasting nearly half the votes cast in every election and orphaning millions of voters because they happen to live in an area where their political view is not the most popular. A PR election would produce results that would send a clearer message to Canadians about what they think politically than our present system does. The more competitive environment would alter the incentives to participate for both voters and parties. While some citizens might miss the all-or-nothing aspects of plu-rality, many more will likely appreciate the stability of a system where every vote


counts. More competition might also force political parties to work harder to inform citizens about the issues and get them out to the polls.

The comparative experience with PR among western countries suggests that these expectations are realistic, while the alleged trade-off involved in the switch—that we might sacrifice stability for better representation—is not borne out by the concrete experiences of other western countries with PR. PR can deliver a government that is both effective and representative, and one that offers stability and increased competi-tion. By contrast, other proposed reforms like majority voting, non-proportional mixed systems, and parliamentary reform are false alternatives to PR because they fall far short of what is required to address the shortcomings of our present system.

75

CHAPTER FiVE

Explaining Canadian and Comparative Voting-System

Choices

introductionCanada remains one of just three major western countries still using the plurality voting system. The United States and the United Kingdom also use it, though British loyalty to the system may be faltering—over the last decade, they have introduced a host of proportional voting systems for regional and local elections. Everywhere else, people vote using other methods. A few western countries use majority voting (France, Australia) but, on the whole, most western democracies use some form of propor-tional representation. in fact, they have been using proportional methods of voting for nearly a century. Tellingly, none of the newly democratizing countries in the for-mer Soviet bloc or the global South have opted for first-past-the-post, either.

So why does Canada continue to use the plurality system? How have we gotten into a situation where we are seemingly wedded to such a poorly performing voting system? Why do most other western countries use something else? Some say that in-stitutions like voting systems mirror the broad cultures in which they are used, that the Anglo-American democracies are “adversarial” and thus favour the all-or-nothing approach of plurality, while European countries are allegedly more consensual and thus favour the more inclusive, cooperative style of PR. others suggest that these in-stitutions emerged and remain in place through the consent of the governed. But such explanations quickly falter when confronted with the actual historical record of the origins of western voting systems and their reform. in Canada as elsewhere, vot-ing systems emerged from struggles over power, not principle, while their reform was often fuelled by struggles to make these countries minimally democratic.


This chapter will explore the historical struggles over Canadian and comparative western voting systems—specifically, the origins of the Canadian system, past efforts to reform the system, and the shift to proportional voting in Europe—to highlight the social and political dynamics that have fuelled change and provide some much-needed context for our struggles today.

The origins and Maintenance of Canadian Electoral institutionsTo hear some commentators talk, one could be forgiven for thinking that our single-member plurality (SMP) system is an old and intrinsic component of the Canadian political system, steeped in tradition and ultimately rooted in our constitutional in-heritance from the United Kingdom. in addition, elaborate justifications for the system are repeatedly proffered that argue it reflects deeply rooted cultural and political val-ues, values that demonstrate Canadians’ abiding concern for local representation. Such arguments have the effect of making our use of SMP seem either inevitable or consensual, an imperative of our constitution or the “will of the people.” But nothing could be further from the historical truth.

Historical claims that SMP forms a part of our “constitutional inheritance” are weak and ultimately ring hollow because Britain itself only shifted to SMP decisively in 1885, nearly two decades after Canada was established. in fact, Britain used multi-member constituencies extensively in the 19th century and experimented with the semi-proportional limited vote in a number of urban centres (Steed 1990: 190–1). From 1918 to 1948 Britain even used the STV form of PR in a handful of multi-member constituencies representing the major universities. And Canada didn’t settle on SMP until many years after Confederation, with a few double-member ridings surviving until 1966 (Qualter 1970: 118). nearly all Canadian provinces utilized multi-member ridings at one time or another, with BC and PEi maintaining dual-member ridings into the 1990s. nor is it clear that our parliamentary “inheritance”—presumably involving responsible government—requires any particular voting scheme to be effective. in-deed, Australia quickly shelved SMP after World War i in favour of a majority system for their lower House and later introduced PR for their Senate elections, despite a similar legacy of British institutions (Farrell and McAllister 2005; Uhr 1999).

Cultural defences of SMP are also weak and suffer from a poor reading of history. Proponents typically imply that a kind of cultural consent for SMP exists simply be-cause the system has been used for a long time. Few actually explain how Canada’s cul-ture has influenced our choice of voting system. For instance, Canadian political scientist Heather Macivor claims that different voting rules reflect different political values, with SMP valuing majority government and PR valuing proportionality (Macivor 1999: 26). But she neglects to mention just whose political values these institutions represent or how they came to be enshrined. This slippage—from a description of the dynamic of voting systems to an explanation of why they are used—is common in political science accounts of voting systems and their effects. For instance, Richard Katz argues that “[i]nstitutions embody the value commitments of a polity,” though

e x p l a i n i n g c a na d i a n a n d c o m pa r at i v e vo t i n g - s y s t e m c h o i c e s / 77

a paragraph later he allows that “[n]ot only values shape institutions” (Katz 1997: 7). But voting-system choices are either the product of values or they are chosen for other reasons—they cannot be both.

Debating the Origins of Institutionsin a review of the relevant academic literature, Pippa norris suggests these “values” explanations emerge because analysts “seek to locate the normative values underlying the choice of electoral systems within coherent theories of representative democracy.” For instance, we hear how PR systems are used in “consensual democracies” because they “aim at power sharing among multiple political actors to maximize deliberation, bargaining, and compromise” (norris 2004: 67). The implication is that PR systems were chosen for these reasons. But when we examine the historical record in western Europe, just the opposite appears to be the case: PR was chosen primarily because it would exclude certain political forces from power. Put bluntly, proportionality only came to be “valued” by those with the power to act on it in World War i-era Europe, when new political forces from the left appeared to be in a position to “value” majority (Boix 1999; Pilon 2005). in Canada too, historical evidence shows that plurality’s adoption was an elite decision, with no public discussion or input (Phillips 1976; Pilon 2006). The notion that our voting system’s historical emergence was informed by a cultural bias for a close relationship between MPs and their ridings can find little to recommend it historically. Michael Steed’s conclusions about 19th-century British constituency reform can be extended to Canadian circumstances: put starkly, the shift from a predominantly multi-member plurality system to SMP was the product of party interests and little more (Steed 1990: 199). Thus the representative–constituency link allegedly so valued by Canadians is instead a historically uniformed, post hoc justification for what were primarily self-interested political decisions.

The real origins of Canadian institutions like our voting system must be sought in the pre-democratic and elitist politics that founded the country. in 1867 Canada was not a democracy, even by the broadest definition. Though the notion of “responsible government” was a key component of the Confederation deal—thus assuring that gov-ernment would be accountable to an electorate—the suffrage was so narrow that only the most economically privileged white men could vote. Even for them, elections were erratic and staged to favour the governing coalition, often marred by corruption, and compromised by a lack of secret voting (Ward 1963; Franko 1992). These arrange-ments were not accidental; Canada’s founders were not democratic in sentiment. For instance, both John A. Macdonald and his Quebec lieutenant George Cartier were keenly worried that the Canadian system might allow “democratic passions” to break out and sought numerous safeguards to prevent public influence from going too far. These included the electoral restrictions and manipulations noted above, in addition to a strong defence of an appointed Senate. in these pre-democratic times, politicians could be refreshingly frank in their aims—at least as concerned those with little po-litical influence. in the negotiations over Confederation, Macdonald proclaimed his intent was to design a system “of constitutional liberty as opposed to democracy” and erect safeguards to protect “the right of the minority.” The minority Macdonald was


concerned about were the rich, whom he helpfully noted “are always fewer in number than the poor” (Brady 1958: 71). He needn’t have worried; the narrowness of the 1867 franchise pretty much excluded everyone who might advance anti-rich sentiment.

of course, things did shift from the founders’ narrow view of public input and participation in our political system—eventually, simultaneous election dates, the secret ballot, and a broader franchise were introduced. Such changes are referred to obliquely in most texts on Canadian democracy, the product of a vague and seeming-ly inexorable process of modernization, juxtaposed against a list of the relevant dates involved (Courtney 2004; Dyck 2007: 305–6; for a better, albeit still incomplete ver-sion, see Elections Canada 2007). But the process of change itself is the story that we need to recover. How and why elites agreed to make these specific changes may tell us a great deal about how the reform of political institutions works more generally.

The reform process basically had to do with political self-interest, elite competi-tion, and pressure from below for access to the political system, particularly from the emerging industrial labour force. The fallout from the Pacific Scandal in 1872 (where donors to the Conservative party were exposed as having gained generous and unten-dered contracts to build a national railway as a result of their political connections) discredited Macdonald and his financial backers, allowing a new coalition of politi-cians and their financial backers to come to power. This new elite promptly changed a number of election rules that they felt worked against their interests—particularly the open ballot and multiple voting days—though they did not expand the suffrage a great deal (Ward 1963: 158, 167–8). The expansion of voting rights took longer and resulted primarily from agitation by organized labour, who wanted the state to legiti-mize the position of unions to represent their members and legally strike against their employers (Franko 1992).

over three decades, each of the traditional parties (liberals and Conservatives) slowly and incrementally extended the franchise to different segments of the white male working-class population when they were in power. The competition between these two parties made extending the vote attractive as a way of recruiting more support, while the incremental nature of the process allowed them to judge the kind of politics that might emerge from such actions. The parties’ fears of working-class radicalism proved unfounded, as most of the new voters supported one of the two existing elite parties. This sense of tactical advantage was also the key factor motivating the extension of the vote to women at the federal level during World War i—the pro-war Union government felt that women with husbands or brothers at the front would support their policy positions (Bashevkin 1993: 8–9).

The point is that these changes were not inevitable, evolutionary, or the product of a sudden conversion to procedural fairness and democracy on the part of either elites or the public. instead, they emerged from various pragmatic assessments by the then-existing political elites about how best to maintain or further their interests and the interests of those who supported them—particularly those who supported them fi-nancially—in the face of competition from other elites, increasing demands from the public, and unpredictable and unforeseen historical developments. Principles neither informed nor got in the way of such decisions, though in defending their choices elites often cloaked them in principle. This helped make their decisions (in reality


self-serving) appear to be in the public interest as well. Thus changes in election ad-ministration, voter registration, and the franchise were reforms calculated to benefit those enacting them. The shift back and forth from a provincial to a national fran-chise in the late 19th century is a case in point. Each change was designed to benefit the changer, with a federal franchise considered to be in the interest of the Conserva-tives and a provincial franchise thought to benefit the liberals (Qualter 1970). Whether such elites did benefit is beside the point, as sometimes the powerful mis-judge their opponents or circumstances change in unpredictable ways (as we will note below with the 1951 reform of BC’s electoral system). neither development alters the intent informing the change; the elite rationale behind the choice becomes no more principled just because some public good accidentally emerges.

The History of Canadian Voting-System ReformWe can see this party self-interest dynamic at work particularly on the question of voting-system reform in Canada (this section draws on Phillips 1976, Pilon 1999, and Pilon 2006). The earliest calls for some degree of “minority representation” came from elements of the traditional political elite who were being poorly served by the existing system. For instance, when a significant element of the liberal party coalition considered breaking away in 1874 to form Canada First (a more firmly English na-tionalist party), a new voting system that would guarantee minority representation was a key priority. When this group abandoned their efforts to create a new party, they also abandoned their interest in voting-system reform. later, around the turn of the century, Quebec Conservatives tried to raise interest in PR for national elections as a means of bettering their dramatic underrepresentation from the province. But as they could convince neither their party nor Parliament, little came of the initiative. There were social reformers and more marginalized groups like organized labour that were also interested in voting-system reform early in Canada’s history, but none had much influence on those who could make a change.

When parties saw change as being in their interest, however, they could alter voting rules quite rapidly. The best example was the introduction of the semi-proportional limited vote for elections to the ontario legislature from the multi-member riding of Toronto. in the 1880s the provincial liberal government found itself shut out of any representation in the provincial capital, despite attaining a considerable amount of support. To counter this, the governing liberals introduced the limited vote in 1885 as a means of effecting minority representation for liberal voters in the capital. The plan worked: subsequent elections returned at least one liberal in the multi-member riding. However, when it appeared that the scheme might allow a labour candidate to gain election to the legislature, the government quickly reverted to the more-exclusive plurality voting in 1893. These examples demonstrate how party self-interest—not values—was responsible for voting-system choice. later voting-system changes were also responses to the same dynamic, though the context informing the need for change altered considerably.

There has been significantly more debate over, and change to, Canadian voting systems than most contemporary political analysts realize. From World War i to the


1950s the voting system and its reform was regularly a topic of debate and struggle at all levels of government and in most areas of the country. At the federal level, propos-als for voting-system reform were discussed every decade from World War i to the late 1940s. Parliamentary committees produced reports on the subject in the 1920s and 1930s and a vote held in 1923 to experiment with proportional voting in a few urban ridings gained considerable support, though not enough to pass. indeed, the failure of PR to be introduced in the 1920s was, on the surface, somewhat surprising in that two of the three parties elected to the 1921 Parliament—together comprising a ma-jority of voting members—had committed themselves in the election campaign to the introduction of proportional voting. But here too, examining party interests is informative. What had changed between 1921 and 1923 was the federal liberal party’s sense of confidence about its political future. Before the 1921 election the liberals appeared to be unravelling under the pressure of the wartime split in the party that divided English and French members, and the move by farmers and labour—usually key constituencies in the liberal coalition—to run under their own party banners following the war. After the election, however, despite the breakthrough for the farm-ers’ Progressive party and the minority governing status of the liberals, the party felt more confident about eventually recouping their losses. But the liberals would con-tinue to dangle voting reform before the voters and other parties over the next decade whenever conditions appeared to be moving to the disadvantage of their party.

Though voting reform would not occur at the federal level in this period despite much discussion, there were a host of reforms to voting systems at the municipal and provincial levels. over the interwar period, 19 towns across western Canada adopted proportional voting for civic elections while Alberta and Manitoba both adopted mixed majority/PR voting systems for provincial elections. Elite and party self-interest did not motivate all these changes, though they were key for the ones that remained in place for any length of time. During and after World War i, earnest and idealistic civic reformers succeeded in getting PR adopted in a number of smaller Canadian towns either by referendum or a vote of city council. But they could not keep these new vot-ing systems on the basis of civic-reform arguments. in most areas, the locals quickly grew tired of the complicated counting procedures required for proportional voting, the delays in getting the results, and the fact that the system did not seem to produce markedly different results from those that might have been achieved under plurality voting. Small towns in BC like nelson, new Westminster, and Mission used PR once and then sought repeal almost immediately.

But where politics was sharply divided, particularly along class lines, voting-system reforms tended to be motivated by the self-interest of traditional parties and the changes proved to be more long lasting. Thus in 1920 Manitoba’s liberal government rushed through changes to municipal and provincial voting systems, fearing that political support for organized labour was rising precipitously. in Alberta, farmers-turned-politicians formed a majority farmer government and introduced similar changes to benefit themselves and divide their opponents in the traditional parties. in each case, values took a back seat to political expediency. Both provinces gained extensive ex-perience with these majority and proportional voting systems, utilizing them from the 1920s to the 1950s.


FigUre 5.1 Voting-SyStem CHangeS in CanaDa

provincial System adoption repeal application

ontario . . . . . . . . . . . . . . . . . . . . . LV 1885 1893 multi-member toronto ridingmanitoba . . . . . . . . . . . . . . . . . . . StV 1920 1955 urban ridings aV 1924 1955 rural ridingsalberta. . . . . . . . . . . . . . . . . . . . . . StV 1924 1956 urban ridings aV 1924 1956 rural ridingsBC . . . . . . . . . . . . . . . . . . . . . . . . . . aV 1951 1953 all ridings

municipal System adoption repeal

Calgary. . . . . . . . . . . . . . . . . . . . . . StV 1916 1961 aV 1961 1973Lethbridge . . . . . . . . . . . . . . . . . . StV 1928 1929edmonton . . . . . . . . . . . . . . . . . . StV 1922 1928Vancouver. . . . . . . . . . . . . . . . . . . StV 1920 1923Victoria . . . . . . . . . . . . . . . . . . . . . StV 1920 1921South Vancouver . . . . . . . . . . . . StV 1918 1929West Vancouver . . . . . . . . . . . . . StV 1917 1930nelson . . . . . . . . . . . . . . . . . . . . . . StV 1917 1919port Coquitlam . . . . . . . . . . . . . . StV 1917 1921mission City . . . . . . . . . . . . . . . . . StV 1917 1921new Westminster. . . . . . . . . . . . StV 1917 1919regina . . . . . . . . . . . . . . . . . . . . . . StV 1920 1926moose jaw . . . . . . . . . . . . . . . . . . StV 1920 1925Saskatoon. . . . . . . . . . . . . . . . . . . StV 1920 1926north Battlefield . . . . . . . . . . . . StV 1920 1924Winnipeg . . . . . . . . . . . . . . . . . . . StV 1920 1971transcona . . . . . . . . . . . . . . . . . . . StV ca. 1941–44 1971St. james . . . . . . . . . . . . . . . . . . . StV 1922 1971St. Vital . . . . . . . . . . . . . . . . . . . . . StV ca. 1931–34 1971

aV = alternative voteLV = Limited voteStV = Single transferable vote

Source: adapted from pilon 1999.


After World War ii, interest in voting-system reform again rose in the face of po-litical uncertainty. At the federal level, some liberal and Conservative MPs worried about rising support for the socialist Co-operative Commonwealth Federation (CCF), especially in the wake of the CCF capture of provincial government in Saskatchewan in 1944 and the party’s surprisingly strong showing in ontario in 1943. Many MPs and influential business groups thought that some kind of majority voting was required at the federal level to keep the CCF out of power nationally. interest was particularly strong in the run-up to the 1945 and 1949 federal elections. But the CCF break-through never came and interest in the issue died out.

At the provincial level, voting-system reform remained a hot topic in certain lo-cales. in British Columbia, a coalition of liberals and Conservatives had ruled the province for most of the 1940s. After the 1941 election, where the CCF emerged with the most votes, the two traditional parties decided to work together rather than let the left “go up the middle” between them and win on a vote split. Though successful and popular with the voters as a governing coalition through two elections, there was constant pressure within each of the different governing parties for the coalition to end. When the arrangement finally broke down in 1951, the government introduced a majority voting system to prevent the CCF from benefiting from vote splits between liberal and Conservative voters. The plan worked—but not for the former coalition partners. Under the new system, the CCF did not win power; instead, a new right-of-centre coalition benefited and eventually crowded out the former governing parties.

The repeal of the majority and PR systems in Canada’s western provinces was sim-ilarly motivated by party interests rather than by principle or values. in BC, the new governing Social Credit party turned their minority government of 1952 into a ma-jority government in 1953. Though the majority voting system had helped secure their surprise victory, the party decided they didn’t need majority voting anymore and instead returned to plurality voting. in 1955 and 1956 Alberta and Manitoba also abandoned their hybrid majority/PR voting systems in favour of a return to plurality voting. Despite all sorts of public-spirited rationales for the decisions offered by the changers, the evidence is fairly clear that party self-interest was paramount in moti-vating the changes.

The origins and maintenance of Canada’s plurality voting system—as with most aspects of Canadian political institutions—has little to do with principle, values, or the consent of the governed. our institutions have taken, changed, or maintained their shape according to the interests of political elites, initially in pre-democratic circum-stances and later via political contestation and struggle. The instances of voting-system change are striking illustrations of how quickly alleged “traditions” can be cast aside when they are no longer seen to serve the strategic interests of those exercising power. Thus our lengthy interlude with majority and proportional voting systems at the provincial level did not reflect some conversion to consensual values of governing and then a subsequent rejection of them decades later, as one might gather from conven-tional political science volumes. instead, the overall point is that our voting systems, both past and present, are fundamentally political results and remain in place for po-litical—not principled—reasons.


The origins and Maintenance of Comparable Western Electoral institutionsin contemporary debates over voting systems in Anglo-American democracies, op-ponents of reform often reject attempts to use western European experience with proportional voting as a point of comparison, claiming that these countries are too culturally and ethnically homogenous to make for good comparisons with the more diverse societies of the new World. The implication is that PR works in Europe be-cause everybody basically gets along and there is fundamental agreement on the most potentially divisive issues of the day. Following from the political science literature on the topic, the fact that these countries use PR is further proof of their social commit-ment to consensus and inclusion as a way of doing politics because PR allegedly “aims” to respond to these values. By comparison, Anglo-American democracies are much more socially diverse and—according to this line of reasoning—PR would only underline that diversity, with negative effects for everyone. And in any event, the fact that these democracies don’t already use PR is taken as confirmation that consensus and inclusion are not values relevant to these political systems. The thrust of all these arguments is that PR-using countries use PR because they are not terribly diverse or politically divided, and as such have little problem operating by consensus with every-one included in decisions.

The problems with such arguments are many, including circular logic, bad rea-soning, and an ignorance of the depth of political division and struggle within these PR-using countries both past and present. First, to suggest that PR-using countries are consensual and then point to the fact that they use PR as proof of their consensus orientation is obviously circular reasoning. Second, to assume that the consensus pro-duced in PR-using countries flows from pre-existing cultural predilections neglects the role of the institutions themselves. Perhaps the causal arrows should be reversed? That is, it is not a consensual culture that brings about PR, but rather that PR facilitates or forces a greater degree of consensus decision making on the group. By this logic, Anglo-American democracies could be just as consensual as European democracies if they used PR instead of their all-or-nothing voting systems. But the most serious problem with the “consensus” explanation of PR use in Europe is that it cannot with-stand an encounter with the historical facts. As will be demonstrated, PR in Europe was the product of conflict, not consensus, and it has remained in use because it has proven politically the most effective means of managing those conflicts.

The origin of PR use in Europe coincided with the most significant political strug-gle of the early 20th century: the achievement of minimally democratic structures of governing (this section draws from Pilon 2002, Pilon 2005, and Colomer 2004). in the 19th century most western European countries used either plurality or majority voting systems to elect various legislative bodies. What must be underlined here, however, is that such bodies were not “democratic” by even the most minimal standards. in some cases, the franchise was so narrow that only the wealthiest males could vote. in others, the franchise might be more open but the legislature elected by it did not control the


governing executive. The end result was the same whichever path was followed—gov-ernment was not based on the consent of the masses (Beyme 2000). The reasons for this were fairly straightforward. Europe was deeply divided between traditional con-servatives and reforming liberals who often could not agree on changes to traditional forms of rule, let alone sanctioning democracy. in fact, neither typically had kind words for mass suffrage or democratic rule. Their fear was that, if given the vote, the masses (who were the majority) would use the state to further their own interests, just as conservatives and liberals had done. Then, late in the 19th century a new force arose—an organized working class that supported socialist parties, which demanded workers be included in the political system (Rueschemeyer et al. 1992; Eley 2002). over the next three decades, the balance of force between these groups (conservatives, liberals, and working-class political parties) would shift the alliances between them, affecting both the speed of democratization and the timing of the introduction of proportional voting.

The first countries to switch to PR did so as a means of forestalling the drive for democratic rule. Conservatives in Belgium (1899), Finland (1905), and Sweden (1907) all introduced PR as a way to divide the pro-reform camps and hobble both liberal efforts to gain parliamentary sovereignty and socialist efforts to achieve full-male-suffrage democracy. in the same period, German conservatives introduced PR for a host of elections (insurance boards, local government) with the same intent. And for a time, such strategies seemed to work. But the coming of war in 1914 wrenched the traditional social fabric in all European countries, breaking down and discrediting conservative forces as the war dragged on while strengthening their liberal and social-ist opponents. With the cessation of hostilities in late 1918, most countries on the continent were facing conditions not unlike a political—or actual—civil war. it was within this context that PR swept the continent. Basically, a minimal form of democ-racy could no longer be put off. Conservatives wanted to assure themselves a place in the new order and limit what they perceived as ballooning support for the left; liberals agreed, as they too were concerned about the stated economic agenda of the socialists. As both liberals and conservatives believed that the left would be very powerful under a mass suffrage—and a mass suffrage could no longer be resisted—they sought out any institutional ways that might prevent them from taking power. PR was their best hope for limiting the gains of the left. in fact, the left too was keen on PR, but for different reasons. The postwar left was deeply divided between constitutionalists committed to elections and revolutionaries inspired by the Bolsheviks in Russia. PR would allow their differences to be represented while still allowing joint action on occasion. So one could say a “consensus” for PR existed in Europe after World War i, but it was a consen-sus based not on principle or values but rather on a pragmatic assessment of political power.

The wholesale conversion of European countries to proportional voting thus co-incided with and was conditioned by the shift to democratic forms of governing. it was all about establishing a new equilibrium among conflicting social forces in historical circumstances that were rapidly changing and unpredictable. nor did the struggle end there. interwar Europe continued to witness these different groups contest democracy and its boundaries, contributing to another major war and reconstruction period


where PR emerged again for many of the same reasons. What kept PR on the table in re-establishing European democracies after World War ii and prevented attempts to repeal the system at various junctures thereafter were precisely these deep and abiding divisions within each country. The key division fell along a left–right axis, where the existence of a powerful left party with a strong state interventionist agenda usually forced all non-left parties to act together to prevent or limit the left’s actions. Under these circumstances, any interest in repealing PR quickly foundered on political reality because a change back to a majority or plurality system would threaten both to po-tentially overrepresent the left (furthering their power) and benefit one non-left party to the exclusion of the others. internal divisions and a fear of the left meant PR stayed in place. Again, this “consensus” for PR—if that is the appropriate term—was based on the reality on ongoing conflict and division within these countries, not on some cultural bias toward inclusion and agreement.

incidentally, this conflict dynamic in Europe was also the factor that moved key voting-system reforms in Canada, particularly those following the social and political upheaval after each of the world wars (this section draws from Pilon 1999 and Pilon 2006). What was different then was less the cultures of Europe and Canada than the balance of political forces competing in each space. Canada’s liberals and Conserva-tives had slowly acclimatized to a limited form of democracy, and until World War i had successfully incorporated or marginalized all other political movements. Thus “democracy” itself appeared less dangerous to them than it did to their European counterparts. nonetheless, when left-wing movements threatened to become success-fully political in Canada, their opponents too suddenly discovered the merits of pro-portional voting. As one commentator astutely noted, the rise and fall of the political left often acted as a barometer for elite interest in voting-system change (Phillips 1976: 323).

FigUre 5.2 pr ConVerSionS in WeStern eUrope

Country year full pr system adopted

Finland. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1906Sweden . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1907netherlands . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1917germany . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1918Switzerland . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1918Belgium . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1919*France . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1919*norway . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1919italy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1919Denmark . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1920*

* partial pr systems were adopted in Belgium in 1899 and Denmark in 1915. France would later repeal and re-adopt pr many times.

Source: Compiled from pilon 2005.


Winnipeg’s business community was very active in convincing the liberal provincial government to rush through changes to both municipal and provincial voting systems for the city in 1920, not long after the upheaval of the general strike, as a means to limit what they perceived would be a crushing victory for left forces in the coming elec-tions. Similar sentiments were voiced in a number of other locales across the country. in the end, Canada’s political left could not maintain its electoral threat into the 1920s and the urgent elite-level interest in voting-system reform dissipated. But where the left remained strong politically, as in Winnipeg, PR remained in use for decades.

As with Europe, in Canada this dynamic was not limited to a single historical period. it was reproduced in BC after World War ii, when a disintegrating governing coalition of liberals and Conservatives turned to voting-system reform to stem the electoral threat of the socialist CCF. Tradition was forgotten as newspapers and political elites defended the sudden reform as only “right” and “proper.” And when the threat was considered to have passed, these same authorities blithely mouthed sentiments about a return to tradition and simplicity. on the other hand, there were a few unique wrinkles to the Canadian experience. in the case of Alberta, voting-system reform was fuelled by a farmer politics that was excluded from the mainstream parties and then maintained by populist parties who successfully exploited some of the distinct feder-al/regional issues in the country. But here too, it was the dissipation of these distinct political forces that prefaced the repeal of PR, not a change in values or culture. The institutions changed because the political usefulness of the arrangements was either no longer clear to those with the power to alter them (Manitoba) or was clearly not useful to them anymore (Alberta).

The Modern Period of Voting-System ReformAnother discussion of voting-system reform began in Canada just as the last rem-nants of the former period were extinguished. Winnipeg was nominally still using a proportional voting system for city elections when the Manitoba provincial govern-ment amalgamated it into its surrounding suburbs in 1972—an occasion that allowed a complete government restructuring and return to traditional plurality voting. A year later, Calgary too removed the last remnants of its half-century of municipal voting reform when it repealed the majoritarian alternative vote in favour of plurality as well (reforms in the early 1960s had converted the city’s STV system to AV). Yet just four years later, interest in proportional voting systems on the national scene was resur-rected in response to the separatist threat embodied in the election of the Parti Québécois as a provincial government in 1976. Three years after that, the Pepin-Robarts Task Force on Canadian Unity called for a very modest and only mildly proportional reform to Canada’s voting system.

in the same year, Canadian political scientist William irvine published the first serious political monograph on the question entitled Does Canada Need a New Voting System? He answered “yes” and he was not alone—a number of other respected political analysts agreed that some kind of mild reform was in order. The nDP’s Ed Broadbent attempted to get the issue moving politically during debates over redistricting in 1978. in 1981 he was offered a chance to put a proposal on the issue to the House by


Pierre Trudeau, whose liberals were still smarting over being shut out of the West de-spite considerable voting support in the 1980 election. in various proposals, PR or a limited form of PR was touted as a potential salve for resurgent regional woes, the crisis of Canadian unity, and as a way to balance the representation of the major par-ties in all parts of the country (Seidle 1996). But ultimately, nothing happened.

Regionalism, Canadian unity, the phony balkanization of party support—these issues did not prove serious enough to motivate any governing party in Canada to consider voting-system reform. Even the sop to PR in the 1992 Charlottetown Accord, where the to-be-elected Senate would use STV, was quickly traded away by political elites before the deal came to a public vote. it was telling that these new proposals for reform betrayed little knowledge either of Canada’s past experience with alternative voting systems (or the conditions that fuelled their adoption) or of the voting-system reforms in Europe. Some appreciation of the kinds of crises that had informed insti-tutional change in the past might have proven useful.

Recent Voting-System Reform in Western CountriesAt the same time, voting-system reform was re-emerging as a serious issue in other western countries, though for very different reasons than the ones proffered in Canada. Basically, again, party self-interest was key (this section draws from Pilon 2002 and Pilon 2005). While many commentators—academic or otherwise—have highlighted the role of the parties, scandal, and shifting public opinion in pushing these reforms forward (Shugart and Wattenberg 2001; Gallagher and Mitchell 2005), most have overlooked the consistent importance of particular parties in these changes and the impact of larger economic changes in influencing why reform emerged in the period in question and not in others. The important context here was the long-term effects of the economic crisis of the 1970s and early 1980s on the western party systems, and on the political left in particular.

The 1970s had shaken the economic and political consensus forged in the postwar West, forcing a reappraisal of existing policies by those on both the right and left of the political spectrum. The “golden age” of capital/labour accommodation was draw-ing to a close. For various complex and disputed reasons, inflation was eating up the wage gains of workers while the wage demands were putting pressure on the long-term profitability of capital. Something had to give. Though the traditional left attempted to hold on to their position for a time, powerful right-wing think tanks, the new dominance of corporate media, and their own weak links to their voting constituen-cies eventually pushed most left-wing parties toward the right. in some cases the change came quickly, but in others it was contested. Voting-system reforms emerged in a number of countries out of this process of remapping the left, which involved shifting their strategic role in the party system, and a struggle to redefine their internal coalition dynamics. At the same time, voting-system reforms emerged out of struggles over issues of state economic policy that crossed the ideological spectrum from left to right (Pilon 2005).

France set the stage for these trends in 1986 when the French Socialist party engi-neered the replacement of the double-ballot majority voting system with a form of


PR. The party had been elected in 1981 with Communist party support to introduce a host of socialist measures designed to resist and roll back the new neoliberal eco-nomic policies ascendant in parts of Europe. But by 1984 the Socialists had embraced neoliberalism and were looking for new, more centrist allies to keep them in power. As the logic of the double-ballot majority system essentially forced them into a coali-tion with the Communists to their left, a change to a proportional system would give the Socialists more freedom to seek allies elsewhere without endangering their own electoral prospects. The new voting system, however, did not prevent the Socialists from losing power and the new right-wing government quickly reverted to majority voting. nonetheless, the reform had served its purpose. By the 1990s the Communists were no longer a key factor and the Socialists had other partners to work with.

Voting-system reform really took off in the 1990s, with successful changes effected in new Zealand, italy, Japan, and the United Kingdom. in all cases, the changes were fuelled by struggles within the left to reposition the dominant left party and by inter-party struggles over a national versus international focus for the economy. in each country, the consensus in favour of the traditional postwar Keynesian management of the economy gave way under pressure from the worldwide economic crisis of the 1970s and early 1980s. Some parties defended the status quo but others agitated for more radical change, specifically for a more neoliberal management of the economy.

in new Zealand, the left-wing labour party came to power in 1984 promising a fairly traditional social democratic government. But, once ensconced in power and wielding a majority in the legislature, the party turned radically to the right and in-troduced some of the most neoliberal policies in the western world. Many labour party supporters were outraged; their sense of betrayal fuelled demands for voting-system reform. A number of anomalous election results in the 1970s that had worked against labour had motivated the party to explore electoral reform when they came to power, though party leaders thought a Royal Commission would keep party activists happy but ultimately accomplish little. instead, frustration with government policies and a strong report in favour of a new voting system combined in a call for a referendum on the issue. labour repeatedly promised to hold the referendum but did not. Even-tually, the subsequent national government was forced to hold a referendum on the issue—it passed, bringing into force a proportional voting system by the mid-1990s.

in italy, the refounding and repositioning of the powerful Communist party as a more social-democratic party in the late 1980s—combined with the increasing inte-gration of italian business with Europe—broke up the traditional coalitions holding the country’s party system in place. Scandal and public frustration with italy’s politics were hardly new and cannot be credited with sparking voting-system reform. The changes in the party system were much more crucial, as they disturbed and eventually destroyed the network of support that maintained the status-quo politicians and lim-ited judicial and political challenges to their hegemony. once the traditional system was forced open, a host of populist initiatives surfaced and gained support among the fractured political class, including a public-driven referendum to alter the voting system away from its high degree of proportionality. These changes were mostly supported by the new party system and remained in place until recently, when the country re-turned to its old system.


in Japan, the story is similar. The Japanese party system had been balanced between a strong centre-right and a weaker left divided along issues of peace and state-led economic development. But the slow decline of left voting power throughout the 1970s and 1980s, particularly for the Japanese Socialist party, led to a reformulation of the political centre and an opening for new coalitions with the long-dominant centre-right. Disagreements on the right—particularly on the issue of state-led eco-nomic development versus new neoliberal methods—led to enough defections from the governing centre-right coalition in the early 1990s to push them out of power for the first time since the 1950s. A new centre coalition formed a short-lived government that accomplished little more than changing the voting system.

in the United Kingdom, voting-system reform has been effected at the subnational rather than the national level. Akin to new Zealand’s labour more than a decade before, Britain’s traditional labour party reinvented itself in the late 1990s as a neoliberal party intent on market reforms and pro-business policies. This makeover was not uncontroversial and party elites, keen to dampen opposition to their objectives, uti-lized proportional voting to divide their opponents or marginalize them completely. Thus what began as a policy fall-back position in case labour lost the 1997 national election became an opportunity for the new leadership to cement their control over the party and the country. For instance, the shift from plurality to proportional voting for UK representatives to the European Parliament allowed labour leaders to control the nomination process more effectively and root out their opponents within the party. And the introduction of proportional voting for the new regional assemblies assured that such bodies would be more divided and less able to challenge the policy dominance of Westminster.

Recent Voting-System Reform Initiatives in Canadain Canada, the voting system also came back to prominence by the end of the century. The breakup of Canada’s traditional party system in the 1990s brought renewed atten-tion to the distortions of the SMP system, highlighting its regional overrepresentation and often extreme distortions between party votes and party seats. The three-party system of the 1980s gave way to a five-party system in the 1990s and then to a four-party system after 2003. Yet there was no return to “politics as usual.” instead, the four-party system contributed to the election of minority governments in 2004 and 2006. At the same time, concerns over declining voter turnout and the apparently rising public hostility to conventional politics moved some elites to reconsider voting-system change as a potential solution (Seidle 2002; Tanguay 2004). Pushing elite interest was the rise of a new public advocacy group focused specifically on the voting system. The formation of Fair Vote Canada in 2000 gave the country its first full-time, grassroots, national advocacy organization committed to the adoption of proportional voting for all Canadian elections. Since 2000 FVC has channelled considerable energy (though rather limited financial resources) into public education, media outreach, and political party/government lobbying, with some success. By 2003 PR was consistently making the agenda of proposed solutions to Canada’s “democratic deficit.” For instance, in 2004 the law Commission of Canada released a report calling for the introduction of


a mixed-member proportional voting system for federal elections. But despite the persistence of a federal multi-party system and an increase in media interest in the is-sue, the traditional national parties have shown little desire for voting-system reform. instead, most of the debate on the question is happening at the provincial level, with active discussions in PEi, new Brunswick, Quebec, ontario, and BC.

PRoVinCiAl iniTiATiVES

Provincial interest in voting-system reform emerged for a host of reasons, some simi-lar to those that caused it to emerge at the federal level, some entirely different (see various chapters in Milner 2004a and 2004b). Similar to federal experience, public opinion about government and politicians at the provincial level had been sinking for some time along with voter turnout. A number of provincial politicians thought a consideration of voting-system reform might reverse these trends. But in a departure from federal experience, provincial election results had delivered a series of seemingly perverse results throughout the 1990s. in the Maritimes, opposition parties were often reduced to a handful of seats despite considerable support, thus limiting how effective they could be as opposition and calling the legitimacy of the system into question. Governments in PEi and new Brunswick began studying voting-system reform as a means of correcting this problem. in Quebec and BC there were elections where the party with the most votes did not “win” the election. Such counter-intuitive results moved the parties affected to promise a review of the voting system if they gained power. Meanwhile in ontario, the liberal party made a consideration of voting-system reform one of its democratic reform planks in the run-up to the 2003 provincial elec-tion. When all these governments came to power, they did in fact follow through on their promises to consider reform, though all dragged their feet on the issue for most of their mandate. Again, the rise of public advocacy groups focusing on the voting system—Fair Voting BC, Mouvement pour une démocratie nouvelle in Quebec, and Fair Vote Canada—made a difference in keeping the issue on the agenda.

Events started rolling when Quebec and BC struck their reform bodies. Quebec utlilized a combination of legislative committees and public consultations to consider changes to the voting system. As all three legislative parties had committed to voting-system reform in the election campaign of 2003, a consensus for change appeared highly likely; indeed, a proposal for a non-proportional mixed voting system emerged early on. over the next few years the government continued to pledge support for the change, though specific details about when the system would be introduced or take effect kept changing.

in BC, the liberal provincial government made good on its promise to create a public body to study the provincial voting system and perhaps make recommenda-tions for change that would be the subject of a public referendum in late 2003. The BC Citizens’ Assembly comprised 161 people chosen randomly to represent each rid-ing across the province. After a year of learning about voting systems, hearing from other citizens and experts, and deliberating among themselves, the BC-CA recom-mended that BC adopt an STV form of PR. Despite indifference from the media and a lack of adequate funds from government to promote the vote, nearly 58% of voters


in the 2005 referendum were prepared to try out the new system, despite the fact that few really understood how it worked. Unfortunately, the government had saddled the vote with a super-majority rule that required 60% + 1 to pass. Though pressured to introduce the change anyway, the government refused but offered another vote on the question in 2009 as a compromise.

in the Maritimes, PEi moved quickly from a report recommending a vote on a mixed-member proportional system to a vote in november 2005. That same year, the new Brunswick government agreed to abide by the direction of an expert panel on voting systems about how to consider reform and ontario’s liberal government es-tablished the terms for their Citizens’ Assembly (the oCA), modelled roughly on the process used in BC. in May 2007 the oCA also recommended a PR system, though in their view an MMP form of PR was more suitable for ontario circumstances than the STV model recommended by the BC-CA.

But despite all this activity, the commitment of provincial politicians to the issue has proven to be largely superficial. in each case, the government has acted either to stall the decision-making process or to rig the process against the likelihood of change. Thus the Quebec government keeps consulting on the question and putting off the decision about when the new system will be introduced or take effect (Cliche and Charbonneau 2006). in BC, the government lumbered the referendum process with a super-majority rule—a decision with no precedent in either voting-system reform or Canada’s use of referendums—and starved the public-education budget of funds. in PEi, the government kept changing the rules of the process in the weeks before the vote—including what would be considered a winning threshold—while refusing to fund public education or open enough polling stations on the voting day (McKenna 2006). in new Brunswick, the Conservative premier committed to a referendum vote on a PR system for 2008 but then proceeded to call a provincial election before allowing voters to decide on the issue, thus casting doubt on his commitment to it. in ontario, the government waited until late in its term to finally act on its commitment to create a citizens’ assembly (thus limiting public exposure to the process or ideas before a vote might be taken) in addition to following BC’s cynical lead of imposing a super-majority requirement on the result.

not surprisingly, the results from such haphazard commitments have been mixed to poor. The PR referendum failed in PEi with just 36% voting in favour. BC voters fell 2.5% short of their 60% threshold. Quebec still dithers about when and how voting system reform will happen. it is not clear whether new Brunswick’s new liberal gov-ernment will honour the Conservative government’s commitment to a PR vote in 2008. And the ontario process faces many of the same hurdles as that in BC in terms of threshold and public education. The end result of all this effort may be no voting-system change at all.

it is telling to compare the successful voting-system reforms in other countries re-cently or those in Canada’s past to the consistent pattern of failure of efforts to reform Canada’s voting system over the past four decades. Attempts to dislodge plurality based on regional issues, Canadian unity, distorted party standings, and—most recently—concerns about declining public confidence in the political system have all proven in-sufficient motivators for change. Without some party—specifically, a potentially


governing party—seeing change as being in their electoral self-interest, the chances of reform appear to diminish considerably. internationally, changes occurred as part of a process of shifting the left toward the centre of the political spectrum and a struggle over the national or international orientation of the economy, depending on how that affected the particular larger party system. in some cases, PR was part of a struggle to prevent that shift (new Zealand), while in others it was part of helping it come to pass (italy, Japan, and the UK). But Canadian circumstances are different, and the electoral left and the national/international aspects of the economy do not play the same signal role in our party system that they do in other western systems. This may prove a cru-cial stumbling block to change. or it may mean that wholly new factors might come into play, such as the mobilization of the electorate for a referendum or a court chal-lenge to the existing system based on Charter guarantees of equality.

ConclusionCanada’s first-past-the-post system has proven very resilient to reform over the past century and a half, despite some considerable efforts to change it. The public has re-mained largely unaware of its effects, taking their cues from the political parties they support, who—for the most part—have supported the system and its unrepresenta-tive workings. only recently has there been any suggestion that the public should both understand and have some say in the choice of the voting system they use to elect their representatives (Pilon 2002).

Traditionally, the use of plurality voting has been defended simply because it has long been used in Canadian circumstances or by arguing that it responds to a particu-larly adversarial Canadian political culture. But such arguments are both logically and empirically without foundation. Canadians have had little input into the decisions over voting systems and voting-system change at any level of government in our his-tory. The fact that Canadian voters have used what they were offered by our political elites cannot seriously be considered a kind of consent. Consent implies “informed consent”; in other words, choosing requires some knowledge of the choices. But evi-dence drawn from Canadian public opinion surveys on voting rules suggests a deep ignorance about the basic workings of our current system, let alone any of the other choices (Bricker and Redfern 2001). nor can Canadian voters be said to value local representation above all other concerns: few can say much about local candidates or local politics. Rather, the key factor in the voting decision is the party, and this is obvi-ously what would be paramount in representation if voters had their way. Thus the public cannot be seen as the author of our institutional choices.

instead, choices about institutional design—such as the kind of voting system we will use—have been contained within the strict purview of political parties. Because it has served their interests, we have used plurality. But when it has not, or when they have feared that it will no longer serve their interests in the face of new and threaten-ing political competition, they have proven more than willing to cast “tradition” aside in favour of arrangements that they believe will serve them. They have been wrong, as the liberal/Conservative forces in BC discovered in 1952 when their voting-system reform backfired on them, but that does not change the intent of their actions.


These historical examples are not merely of antiquarian interest; they tell us some-thing important about the nature of institutional change, something perhaps still relevant to us today. Voting-system reforms have succeeded where powerful political elites have seen them as being in their interest. Specifically, where governing parties have been keen on change, change has moved very quickly indeed. This underlines how voting systems are not the product of principle or values but rather are forged out of political contestation involving struggles for power and influence. Thus chang-ing our voting systems today may also require the harnessing of such political power, or the creation of a situation where the powerful become interested in reform.

94

CHAPTER SiX

The Politics of Choosing

introductionFinding broad agreement that there is something terribly wrong with our current single-member plurality system does not mean a consensus exists about anything else connected with voting-system reform. How should reformers seek change? By refer-endum? Through the courts? Where should PR be applied in Canada? At what level of government? So far we’ve made a general case for voting-system reform in Canada. The current system has been found wanting and some form of proportional repre-sentation has been forwarded from many quarters as the necessary alternative. But as soon as we shift our attention from the problems of the current system to the poten-tial alternatives, the discussion can often lose its focus and reform forces may sacrifice their unity in the scramble to defend their view of the “one true reform.” People can get stuck on very specific ideas about how reform should proceed, where it should be applied, and what form it should take. Some say that politicians should change the voting-system laws, while others demand that citizens make the decision in some form of referendum. Some are keen to see PR introduced in both houses of Parlia-ment, while others insist that only the House of Commons should be reformed. of course, at some point in the reform process, choices must be made. But as this chapter will demonstrate, none of the choices are as dire as some reformers seem to think, or nearly as important as the basic choice to replace plurality with PR.

Some of the key choices that emerge in the debates involving voting-system reform centre on how to pursue reform, where to seek reform, and what kind of specific PR system to choose. The question of how to pursue reform involves choices about whether to seek legislative changes to the voting system from politicians, or seek a public process culminating in a binding referendum vote to change the system, or seek a court intervention using the Charter to overturn existing laws. The question of where to seek reform involves decisions about whether PR should be applied to some or all jurisdictions—federal, provincial, municipal—and whether it should be applied

t h e p o l i t i c s o f c h o o s i n g / 95

to just one or both of the House of Commons and Senate of the federal Parliament. The question of what kind of PR system to choose involves choices between off-the-shelf models like STV and MMP or some new hybrid or less-than-fully-proportional mixed system.

This chapter will review the debates surrounding the choices about how and where to seek reform, while the next chapter will take up the debates over the “right” choice of PR. Here, we will argue that getting PR by any means is better than not getting it, and that PR should be pursued for all jurisdictions in Canada (with a caveat that the Senate is a special, more complicated case).

Debating How to Get PRThere are three methods touted for getting PR systems adopted in Canada. There’s the old-fashioned way of simply lobbying politicians and political parties to do it. More recently, a popular notion has arisen that politicians can’t and shouldn’t be trusted with the decision as they are basically in a conflict of interest: they are being asked to set the rules by which they themselves stand to benefit or not. As a result, the deliberation and decision making about voting rules should be done by the people themselves, first in a deliberative body like a citizens’ assembly, and then through a vote of all the citizens in a referendum. Finally, some claim that the constitution offers a route to reform—specifically, that various clauses of the Charter of Rights and Free-doms could be invoked to strike down the existing plurality voting system and bring about something else. Each of the proposals has its supporters and detractors, as well as its strengths and weaknesses.

Let the Politicians DecideThe traditional approach of just letting the politicians determine the voting system has its attractions. if a majority party in a parliamentary system like Canada’s favours reform, change can occur very quickly. Historically, politicians designed all the original Canadian voting systems and have been responsible for all the changes at the federal and provincial levels thus far (Phillips 1976; Pilon 2006). of course, the strength of this approach is also its weakness. We can find no examples of politicians and parties prepared to change a voting system unless it coincided with or furthered their inter-ests (Phillips 1976; Pilon 2006; Seidle 1996). Thus politicians designed the various systems with their own electoral benefit in mind. And any reforms were similarly motivated by self-interest. in some cases, as with the provincial voting-system reforms in Manitoba in 1920 or BC in 1952, the motivation was defensive—to prevent an an-ticipated competitor from the left from gaining advantage under the old plurality rules. But in Alberta, the shift to a new voting system was authored by a new political competitor, in this case the victorious farmer party which sought to entrench its dominance over the old parties and even its sometime urban allies in the labour movement. More recently, politicians have proven largely uninterested in debates over voting systems and their lack of interest has made the traditional route seem like a dead end for reform. When the issue does arise in Parliament, it is often shuffled off


to a committee and quietly killed with indifference—a very traditional response by Canadian parliamentarians.

A modification of the traditional approach involves the delegation of consider-ation over voting systems by politicians to expert panels and Royal Commissions. new Zealand got the ball rolling in the mid-1980s when an appointed Royal Com-mission recommended the country shift from plurality to PR. A series of perverse election results in the 1970s and 1980s had moved the newly elected labour govern-ment of 1984 to promise a study of the issue. Few expected it to recommend change or be noticed by the public. But as both of new Zealand’s major parties alienated the public in the 1980s and 1990s, the Royal Commission report gained a high degree of respect from the public and bolstered efforts to push the reform process along (Jack-son and McRobie 1998). other expert panels were not so successful. in the UK, the newly elected labour Party established a one-man Jenkins Commission in 1997 to examine the voting system. Jenkins’s ultimate recommendation for a modified majority system, however, offended reformers and was ignored by the government. in Canada, premiers in PEi and new Brunswick established commissions to examine the voting system and make recommendations about how to proceed. Both commissions rec-ommended a referendum to allow voters to choose between plurality and MMP—not necessarily the result the premiers had anticipated. nonetheless, the politicians had the final say. in PEi, the government sank the initiative with a lack of education, a rushed timeline, an off-election-year vote, a super-majority threshold and minimum-turnout rule, and a reduced number of voting locations. in new Brunswick, the premier promised to hold the vote but then held a general election instead. His loss in the election left the status of the referendum up in the air.

Let the People DecideFrustration with the traditional politician-oriented route has moved the bulk of Can-adian reformers to support a citizen-centred approach to voting-system reform, one where some kind of appointed body deliberates over the issue and then makes a rec-ommendation that the public can decide on in a referendum. Fair Vote Canada has repeatedly lobbied for this kind of process in all parts of Canada and for all levels of government, receiving support from a host of other public interest groups. The strat-egy is partly principled, partly pragmatic. Proponents of the public deliberation and decision model claim that this is the most ethical and democratic way to set the rules of the electoral game, as the politicians cannot be trusted to make an unbiased deci-sion about the voting system when they are so directly affected by it. instead, average citizens, because they are not seeking election, can approach the question more ob-jectively and consider what is really in the public interest. Pragmatically, the citizen approach prevents the politicians from failing to act or trying to kill the initiative in committee because it also includes a mandated vote.

The idea that the public should be engaged in issues of democratic reform has long been seen in practice at the civic level in Canada, with a host of public votes on mu-nicipal voting systems stretching back to the beginning of the 20th century (Hahn 1968). The farmer and labour populists of the interwar period also supported such


approaches for provincial and federal issues as well (laycock 1990). More recently, international experience has drawn attention to the potential of public votes to force a change in the voting system, specifically in new Zealand and italy in the 1990s (Vowles 1995; Donovan 1995). Finally, concerns about trends that indicate a declining faith and participation in the political system have increased academic interest in citizen-engagement strategies around democratic reform (Seidle 2002; Tanguay 2004).

PEoPlE DECiDinG: THE BC CiTiZEnS’ ASSEMBlY

All these influences—in addition to some peculiar elections results in British Columbia in the 1990s—contributed to the creation in 2003 of the BC Citizens’ Assembly, a group tasked with examining voting systems and making a recommendation to the public to be considered in a referendum. Throughout 2004 the BC-CA met on weekends, first to learn about voting systems and then to deliberate over which one would be best for BC. The 161 members were drawn from every riding in the province through a combination of random and self-selection. Half were women, half were men, and as a group they reflected a good portion of the province’s middle-class diversity, though visible minorities were underrepresented (no Aboriginals were chosen through the initial process—two were added later) and, needless to say, the economically challenged and those without 9-to-5 weekday jobs were not well represented (James 2007).

The BC-CA members could have decided in favour of the status quo (single-member plurality), but early on it became apparent that the group had an appetite for reform (MacDonald 2005). The education phase served to focus members on the values that they thought should be key in a representative electoral system. At an early stage, mem-bers highlighted proportional party results as a key value they would like to see in BC’s voting system. From then on, it was clear that some form of PR would be recommend-ed, though the particular kind would only emerge much later in the year. Research that tracked the CA members’ views and how they changed over the course of the pro-cess suggests that MMP was an early favourite of most participants. But in the end, the BC-CA chose to recommend the STV form of PR to voters in the referendum that would accompany the provincial election of spring 2005 (Blais et al. 2007). on elec-tion day, a majority of voters endorsed the BC-CA’s decision.

The BC-CA process and its results were both remarkable achievements. They were historic in their implications, as no government anywhere in the world had ever handed control over such a decision to a group of citizens (Pilon 2002, 2005). The tasks facing the BC-CA’s administrative and educational leaders were considerable: to organize and educate a group of people with no previous background in the issue and to produce a decision within the one-year timeframe in order to present it to the larger public before the fixed-date election scheduled for May 2005. And the commitment and performance of the BC-CA members impressed all observers of the proceedings in both the learning and deliberation phases and the public hearings. All in all, the BC-CA proved that citizens can take on complex topics and are keen to be involved, if they believe their efforts will matter. Many observers credited public confidence in the assembly members with contributing to the impressive result for the largely unknown STV-PR proposal (Cutler et al. 2007). These positive responses to the BC-CA’s work


accelerated calls for similar models of public engagement around institutional and democratic reform in other parts of the country, and even in other parts of the world.

❍ opposition to Reform

But not everyone was happy with the BC-CA or its decision. The decision came as a surprise to those who had followed the CA’s deliberations, as most knew that MMP was initially more popular with members and that 80% of public submissions had called for some form of MMP (Milner 2005). Politically, the recommendation for STV found few friends among the organized political forces in the province. Though both of the parties represented in the legislature eschewed public campaigning on the issue, claiming that they preferred to let the people decide, neither the liberals nor the nDP would endorse the change. The governing liberals, despite Premier Gordon Campbell’s support for the BC-CA process, were not keen on any change to the voting system. Many liberal members wondered why they had bothered to keep this particular cam-paign promise to publicly examine the merits of the voting system (Palmer 2005). in fact, opposition within caucus on this question had been crucial in changing the threshold for the referendum victory from a simple 50% + 1 to a super-majority rule.

By contrast, the nDP had committed officially to switching the provincial voting system to a form of PR, but they had envisaged an MMP model. STV, on the other hand, was not popular with new Democrats, whether they were elites or the rank and file. Many confused it with the 1952 and 1953 elections in BC where the province had used the majoritarian alternative vote—a system nDPers felt had “robbed” the party of victory (Ruff 2004). others in the party thought STV was not proportional enough and would not help improve the representation of women and visible minorities. of course, PR itself—despite official party policy—remained a divisive issue within the BC nDP, and some members were only too happy to condemn STV as a way of bury-ing the larger issue altogether. Curiously, even the leadership ranks of the third-place Greens, a party with no seats in the legislature, also condemned the STV proposal (though it is widely believed that many Green party activists and supporters ignored their leaders on the issue). When Green party leader Adrienne Carr managed to force her way into the televised leaders’ debate, she failed to use the opportunity to promote the PR referendum because she preferred MMP.

once the BC-CA’s decision was made in favour of STV, public opposition to the recommendation began emerging from individuals connected with most of the organ-ized political forces in the province. Though the official parties had sworn off getting involved, the emerging “no” campaign—cleverly named “Know STV”—featured former party insiders from the nDP, Social Credit, the liberals, and the Green party. At one point during the election campaign, former nDP Premier Dave Barrett and former Social Credit Premier Bill Bennett issued a joint statement condemning STV (Ward 2005). Business organizations came out against STV; many of the major unions in the province encouraged their members to vote “no”; and most of the media re-mained indifferent, neutral, or hostile to the proposal, focusing mostly on the novelty or complexity of the new system (Brown et al. 2005). A number of left-wing writers claimed that STV was a right-wing plot to entrench the right in power forever, while


a number of sources on the right argued that STV would threaten the stability of the provincial economy (Anderson 2004; Potvin 2005; Clemens and Veldhuis 2005). The BC-CA appeared to have chosen a voting system with few friends and many well-organized enemies. once their official work was done, most of the assembly members joined an alumni association that campaigned locally during the election for the change, but they had difficulty getting the sustained attention of the local media and had little influence with local politicos (Ratner 2007).

The political fallout from the BC-CA process spilled over into the federal arena when the federal nDP backtracked on their support for a CA process nationally. Ed Broadbent, admittedly “once an early fan of the BC process,” now condemned the BC-CA as a “disaster” akin to “designing a health care system without consulting doc-tors.” For Broadbent, the problem with the BC-CA model was that it did not include a proper role for political parties, which in the end contributed to a proposal that was, in his view, anti-party and unrealistic (Broadbent 2005; Massicotte 2006). others went further, accusing the BC-CA research team and its advisers of unduly influencing the outcome (Redekop 2005). leaving aside this specific accusation for a moment, a num-ber of academic commentators have offered criticisms of the CA process, wondering if non-expert citizens really can come up to speed on complex topics like voting sys-tems without being unduly influenced by the experts, or highlighting the problems in the process—particularly, a gender dynamic that might have diminished the CA’s in-terest in the question of diverse representation (these claims are assessed critically in lang 2007b).

❍ Biasing the People?

The accusations of bias are hard to substantiate. interviews with CA members do not suggest that they considered the research team biased, and an examination of the learning-phase teaching materials confirms a fairly neutral presentation of the PR options (Ratner 2007). other ways to assess bias might be to examine the research backgrounds of the research team and the expert consultants that were brought in help out on various topics. The research director and his deputy did both publish material on STV prior to taking on their positions, though one was somewhat critical of the system as used in ireland while the other thought there was some merit to its use in Australia (Carty 1981; Sharman 1986). The experts were also mixed in their commitments. David Farrell, the author of the book used to guide the CA research and discussion, did publicly state his preference for STV; however, his book is an even-handed treatment of the subject that includes both the merits and challenges of the main PR systems (Farrell 2001). Aside from a less-than-clear presentation of the possible impact of voting systems on the representation of women, the invited experts could not be portrayed as having any particular axe to grind with regard to the voting-system choices. Though Redekop argues that a dynamic expert presentation could have unduly influenced the politically unsophisticated BC-CA members, lang points out that members had access to a variety of sources of electoral system expertise and often brought their own independent research into the deliberations (Redekop 2005; lang 2007b).


if there were legitimate problems with the education phase of the BC-CA process or the choice of STV by the BC-CA members, they were problems involving poor de-cisions about how to approach the study of voting systems and a faulty analysis of BC politics—not a bias for one voting system over another. in both cases, the BC-CA suffered from a lack of politics. The BC-CA’s educational materials (accessible on the assembly’s website—see bibliographic entry) spent too much time assessing voting systems as a matter of deciding among competing values and too little on concerns related to realpolitik. The former has become a very popular way of approaching the question of voting-system reform, but it effectively depoliticizes both the consider-ation of the different systems and the ultimate choice itself. Discussion tends to focus on what is best for individual voters and citizens cast against an ideal-type model of the polity where the real and abiding divisions in the polity are absent. The adoption of this approach allowed BC-CA members to talk freely about what they might want in a polity—what they thought was normatively desirable—divorced from the need to assess whether politics really can or do function in such a way or whether such notions might realistically come to pass. A different tack would begin with an assessment of where the politics of the province really are, why they are organized in such a way, and then move on to how the voting system interacts with that reality. Such an ap-proach might have sensitized BC-CA members to how the different existing political forces might have responded to different proposals for change.

A concrete example of these problems can be traced back to the three overarching values that BC-CA members committed to early on in the process to guide their deci-sion in evaluating the voting systems: voter choice, fairness, and local representation (BC-CA 2004). of the three, only one arguably reflects the concrete world of political contestation: fairness—for political parties. local representation was pushed by an aggressive and organized northern lobby in the BC-CA, which, by some accounts, was highly influential in the final decision to choose STV (MacDonald 2005; lang 2007a). The problem here is that the available evidence suggests that people do not vote on a local/non-local or urban/non-urban basis (see Chapter 8 for more details on this argument). People vote in both rural and urban spaces by party and are divided along party lines in both. if there was a bias in the BC-CA process, it was in this privileging of local representation as the basis for choosing members. Having been chosen on the basis of locale, it is hardly surprising that a number of members thought that “local” representation was important—particularly those from rural areas (who were effec-tively overrepresented by virtue of this selection process).

Meanwhile, the factors that are demonstrably divisive and in need of representa-tion—i.e., the party differences across the province—had no representation because party representatives were banned from BC-CA membership and the BC-CA did not seek their input. indeed, as one observer noted, neither BC-CA members nor the re-search staff could seem to find anyone who would defend or explain the positive implications of party discipline (lang 2007b). This would turn out to be a serious problem, as the commitment to “voter choice” as a key value demonstrated. Those seeking greater voter choice were concerned that parties had too much power vis-à-vis citizens and were determined to make individual elected members more accountable to voters as opposed to their parties (MacDonald 2005). of course, this betrays a classic


populist rejection of the role of the political party in modern democracy. Basically, in this populist understanding of politics, politicians are arrayed against “the people,” with the effect of diminishing the important differences that really do exist among the people—the very left–right ones that parties represent. A serious discussion of party discipline would have highlighted how cohesive and uniform voting in a legislature by members of the same political party is another approach to securing politician/voter accountability; in this case, one focused on the policy differences among voters. While there is some evidence that the research staff did address the issue and try to encourage members to take a more sophisticated view, clearly not enough time was devoted to critically addressing these anti-party biases (see the education materials on the BC-CA’s website).

❍ The Political Context

of course, the BC-CA members were not unaware of the political environment of the province. But when asked what role practical politics played in their decision, mem-bers made it clear that they chose STV on the basis that it was the best option, not on the basis that it was the most likely to gain the support of political elites (Pearse 2007). Such statements demonstrate the serious flaws in the depoliticized model of deliberation characterized by the BC-CA. Members appeared to make decisions based on abstract values and their own local experiences rather than as a result of any deep engagement with the concrete political issues that divide British Columbians or a consideration of the role that organized political parties play in representing them. instead, they arguably chose the wrong issues, privileging what appears to have been a romantic localism and a populist voter-choice approach over a more realistic appraisal of the political divisions and organization of the province. Even the politics of their opportunity—that voting-system changes are exceedingly rare—seemed lost on them if we believe that their de-cision was based on what was “right” rather than what seemed “possible.” in their defence, some BC-CA members thought their decision was balancing real political interests (lang 2007a). The problem was that they were listening to the wrong people, in this case the northern members advocating local representation and the populists who were hostile to parties. And there were other factors influencing their decision, like the BC-CA chair’s decision to prevent any discussion of designing models that might increase the size of the BC legislature. Apparently, for many BC-CA members, this made the MMP option less attractive than it might have been because the “local” ridings would have had to be considerably enlarged (Ferejohn 2007; lang 2007b).

The lack of attention to politics did more to limit the success of BC-CA’s recommen-dation than anything else. First, their proposed STV model did appear to discriminate against one of the key parties. Because there would be larger multi-member ridings in urban areas and smaller multi-member ridings in rural areas, this would mean that the system would be most proportional in areas of strength for the nDP (thus allowing the liberals to gain more representation where they were weaker) and least propor-tional in areas of strength for the liberals (thus denying the nDP any advantage where they needed it). Anyone with an appreciation of realpolitik would have seen that such a proposal would not be accepted by the nDP under any circumstances.


Second, two of the three largest, best-organized parties in the province—the nDP and the Greens—were on record as supporting PR in an MMP form. By choosing STV over MMP, the BC-CA turned potential allies—who together wielded considerable organizational resources that might have been marshalled to mobilize support for the decision—into opponents, without gaining the organized support of any other group in the process. When these self-inflicted wounds were added to the government-inflicted ones—specifically, the super-majority threshold for victory, a lack of public education money to get the issue before the voters, and the indifference and/or hostility of the media—the results from the vote are all the more surprising. However, here too, ob-servers note that the wording of the referendum question probably had an important influence on the voters’ decisions. on the ballot, voters were asked if they supported STV, as recommended by the Citizens’ Assembly. Polls showed that few voters knew or understood STV, but the question directed voters to the source of the recommen-dation and many appeared to trust what had come from “citizens” (Cutler et al. 2007). needless to say, those designing the ontario CA process decided to put the wording of their question in the government’s hands rather than in the hands of their CA. Thus this particular advantage may no longer hold in future votes.

These critical comments about the BC-CA, its deliberation process, and its mem-bers’ ultimate choice of which voting system to recommend to the public should not be interpreted as a condemnation of their efforts or even a necessary rejection of their decision to recommend an STV form of PR. The BC-CA members, by their efforts, have rightly become a model of citizen engagement and a rich source of information—and indeed, inspiration—for those keen on bringing the public into deep discussions about our democratic processes and institutions. The point here is to underline that, whatever decision groups like this make, they must be decisions that grapple with

Yes

Should British Columbiachange to the

BC-STV electoral systemas recommended by the Citizens’ Assembly on

Electoral Reform?

No

FigUre 6.1 BC-St V BaLLot

Source: government of British Columbia. Copyright © province of British Columbia. all rights reserved. reprinted with permission of the province of British Columbia. www.ipp.gov.bc.ca.


concrete political circumstances. Perhaps if the BC-CA members had considered how their recommendation would be received by the province’s organized political forces, they might have been able to speak more effectively to the party concerns and might have gained more support for their proposal.

❍ Aftermath

Still, for all its problems, the BC-CA’s recommendation nearly passed but for the gov-ernment’s arbitrary super-majority threshold. As the situation after election day in BC looked ridiculous—the government with just 46% of the vote was not prepared to accept the referendum results with 57.5%—Premier Campbell eventually agreed to have another vote in 2009, though with the same super-majority rule still in place. The frustration with the citizens’ assembly process—particularly, the unwillingness of the governments in question to act in good faith and the media’s unwillingness to foster a genuinely open discussion on the question—has moved some to call for legal action to get things moving.

For a time, reformers had hoped that ontario would not follow BC’s precedent on the threshold, a hope that was fuelled by the government’s delay in clarifying their position. But in the fall of 2006 the McGuinty liberal government announced that they too would seek a super-majority verdict from the public. Meanwhile, the ontario media’s response to their citizens’ assembly was even more indifferent and/or hostile than the response in BC had been. Columnist after columnist dismissed the import-ance of the question or the need for change (Gwyn 2006; Campbell 2006; Urquhart 2007; Toronto Star Editorial Board 2007; Denley 2007). instead of using the oCA to help their readers figure out what was being considered and make an informed deci-sion, both reporters and columnists tended to short-circuit the process, often simply telling them what to think, illustrated with skewed and unrepresentative examples of different voting systems from other countries. The first meeting of the oCA was com-pletely ignored by both print and broadcast media.

The citizens’ assembly process was supposed to get around the blocked path to reform, specifically the entrenched opposition of the politicians and the media gate-keepers, but it has proven subject to their influence as well. in both BC and ontario, politicians have simply “invented” new traditions and rationales for super-majority rules, clearly in a bid to frustrate reform. Politicians in both provinces defended their manipulations by arguing that changing the voting system is a reform of great signifi-cance and as such should require 60% + 1 of the popular vote as well as a majority vote in 60% of the ridings to pass (the latter rule to allegedly protect rural influence on the question). But such arguments lack any historical or comparative precedent. Canada has seen ten voting-system reforms at the provincial level, with two each in BC and ontario, all passed into law by simple majority votes in their legislatures. The origins of all Canadian provincial and federal voting systems were the result of simple majority votes by the founders. nor can politicians argue that the voting system as a law is of “foundational” importance and as such requires a super-majority to pass. on the contrary: voting system laws in Canada are just statutes that can be passed, amended, or repealed by simple majority votes.


Comparatively, other western countries have also changed their voting systems by simple majority votes of their legislatures. Where public referendums have been held, including the most recent in italy and new Zealand, the thresholds have been 50% + 1 (Pilon 2002). nor can our politicians draw from the country’s traditions on referen-dum votes more broadly: conscription for World War ii was decided in a national referendum based on 50% + 1, as were the Quebec sovereignty referendums in 1980 and 1995 (Boyer 1992; leduc 2003). Even the nunavut referendum that proposed to establish a gender-equal legislature—surely a relevant precedent—was decided by 50% + 1 (Young 1997). Finally, the rural-voter guarantees too are bunk because, as any perusal of Canadian election statistics demonstrates, rural voters do not vote as a bloc. Thus it is false to assume that these voters have more in common with their neighbours than with potential urban political allies. in other words, the call for super-majority votes is a brand-new innovation of political elites for stated reasons that are not convincing. The only precedent these politicians can seem to find for their super-majority rules are municipal plebiscites in a number of provinces, mostly dealing with liquor issues.

Let the Courts Decideone way to unblock the political barriers to change would be to turn to the courts. An obvious route would be to challenge the super-majority rules for the referendums on the grounds that they clearly treat individual voters differentially, effectively inflating the voting power of those against change while diluting the voting impact of those seeking change. While the courts have been prepared to entertain deviations from voter equality in some circumstances where clear and compelling reasons could be marshalled in their favour—as with the need to slightly overrepresent rural areas—there appear to exist no such rationales for the current super-majority rules. Thus a case could be made that these rules violate Charter guarantees of equality in voting rights.

But some activists are not satisfied with simply taking on the super-majority rules. They think the courts can be used to strike down the single-member plurality voting system altogether. if successful, they might then seek a remedy of PR from the courts or a remedy that would force the politicians to consider alternatives (Knight 1999). one case along these lines was organized on behalf of Joan Russow, federal leader of the Green party in the 1990s. Russow argued that the bias in the plurality voting system against smaller non-regional parties constituted a violation of her rights under the Charter (Morgan 2000). More recently, political scientist Heather Macivor has argued persuasively that a strong case could be mounted against the single-member plurality system as violating section 3 guarantees of democratic rights (specifically in terms of a relative parity of voting power) and section 15(1) guarantees of equality rights (because SMP does not effectively represent social diversity). But as Macivor notes, despite the strength of the case, there is no assurance of victory by using the courts because no one can predict how the justices may take up this issue, despite some promising and seemingly relevant past decisions. And given the expense and long delays involved with court actions, the legal route is a risky strategy—especially if the government


adopts a “deep pockets” defence strategy, as appears to have been the case with the eventually abandoned Russow challenge (Macivor 2006: 12).

Debating Who Should DecideEach approach to seeking voting-system change has its supporters and opponents, and broad political differences come into play on each; differences that could threaten the cross-party coalition seeking an end to plurality. When voting-system reform be-gan to re-emerge seriously in Canada near the end of the 1990s, the opposition or lack of interest from politicians made other approaches seem attractive. The populist backlash against the political class—represented in the 1993 election results that clob-bered the PCs, hurt the nDP, and gave rise to both Reform and the Bloc—heightened interest in non-traditional methods of public decision making, such as referendums. And with the success of voting-system reforms in new Zealand and italy, both of which involved public referendums that did an end run around politician opposition to reform, the referendum appeared to be the tool that could break the logjam. Thus both the populist right and left could agree that a public approach and vote was a good and proper strategy for achieving change.

But the process and decision of the BC-CA turned some of the populist and institu-tional left away from CAs. The federal nDP now wanted a Parliament-centred approach to voting-system reform, which the liberals also backed when they had to. Mean-while, the court strategy was always in the background, apparently moving forward, but with little in the way of concrete progress to show for it. Both left and right re-form forces initially rejected the court approach, as both favoured a citizen-centred deliberation and decision-making process. Besides, judges were felt to be not that far behind politicians in favouring status-quo institutions. The consensus for a non-court approach worked out pragmatically for the broad reform coalition, as the right within the group was extremely hostile to a court strategy. now, however, the court strategy is re-emerging with more force, partly due to some surprising success in chal-lenging federal party funding rules and partly out of frustration with government manipulation and a lack of media support for the CA process. The politics of choos-ing a way forward in pursuing voting-system change could prove formidable.

After five years of heady progress in getting voting-system reform on the public and political agenda, there seem to be positives and negatives to all three of the main approaches to seeking change. First, the politicians have softened somewhat on the issue of voting-system reform. After all, as many commentators have pointed out, it was the politicians themselves who opened up the space for the voting-system com-missions in PEi and new Brunswick, the citizens’ assemblies in ontario and BC, and a mixed public/legislative exploration of the question in Quebec (Massicotte 2006). Though the politicians’ motives may have been suspect and though they have tried repeatedly to stymie the process, every opening has the potential to slide out of their control, as it nearly did with the BC referendum. Second, the citizens’ assembly pro-cess is not without challenges of its own, not the least of which is its reliance on both fair dealing from its government sponsor and a measure of goodwill from the media; neither has been forthcoming. on the other hand, traditional politician-centred


processes have fallen flat. Ed Broadbent’s parliamentary-committee approach, struck in 2005, failed to produce any results, despite the existence of a minority government dependent on nDP support (ibbitson 2005; Massicotte 2006: 11–2).

The beauty of the citizens’ assembly process is that it prevents politicians from withdrawing their support when they believe that killing the initiative is within their power. if Broadbent had sought a modified citizens’ assembly approach at the na-tional level, even a change of government wouldn’t have derailed the process. Yet the problems emerging out of government-sponsored referendums and citizens’ assembly approaches in PEi on the one hand and BC and ontario on the other, suggest that just relying on referendums and citizens’ assemblies may not be enough, either. Peo-ple forget that new Zealand’s successful referendum process was not intended to be so by the politicians. in fact, most analysts considered the politicians’ choice of a multi-stage referendum process an attempt to kill off an issue that had dogged them for over a decade (Jackson and McRobie 1998). Canada’s politicians appear to have come up with a more successful spoiling strategy. That is why some serious consider-ation of court strategies alongside these other approaches should not be rejected out of hand.

So often the discussions about which path to change reformers should pursue is conducted at the level of values and principles—for example, the idea that the people should decide, not the courts or the politicians. But the problem with such approach-es is that they ignore the real-world inequalities that affect the likelihood of success for the different approaches. The people may not get to choose, or get a fair chance to choose, precisely because of the inequalities that exist between citizens and govern-ment or the inequalities of influence among those who have the ear of the press. All potential means of dislodging plurality should be pursued. in fact, it may be a combi-nation of factors that ultimately results in change. The italian referendum vote that sparked the change in that country’s voting system involved a series of successful court challenges to the political system and then, after the referendum vote, a politician-centred deliberation over the specifics of what ultimately would replace the former system (Sassoon 1995; Donovan 1995). Given the difficulty of getting voting systems changed, reformers should not be fussy about how change ultimately happens but should rather take victory however it comes.

Where Do We need PR?With three levels of government in Canada, what level should be the priority for voting-system reform? Some reformers are focused on the national level, given its key importance in the Canadian system. others have pragmatically focused on provincial reform efforts, as that is where the action seems to be. Still others would like to see a municipal campaign for voting-system reform, particularly in our larger and more diverse cities. Certainly, complaints about Canadian democracy are heard at all levels and the specific representational problems are consistent across jurisdictions. over the past decade, academic reformers have focused largely on the federal level, which offers its own unique challenges. introducing PR for federal elections would require


tailoring the proposal to the constitutional requirements for representation already embedded in the 1867 BnA Act (renamed the Constitution Act, 1867). This includes required deviations from pure proportionality, given existing constitutional guaran-tees of minimum levels of representation for certain provinces. A further debate is whether PR should be sought for both the House of Commons and the Senate, or just the House of Commons. The provincial level has its own challenges, given the popu-lation imbalance between the urban South and more rural, small-town centre and north in nearly all provinces. Finally, the municipal level in Canada often has its own traditions of politicking, including an absence of formal party names and structures that makes using PR seem to some to be either irrelevant or not worth the effort.

it will be argued here that we should seek PR for all levels of Canadian govern-ment. As politics and political divisions exist across them all, the arguments against plurality in Chapter 3 and for PR in Chapter 4 apply to each level. Furthermore, each of the challenges posed in the different arenas can be addressed and overcome.

PR and the Municipal Scenelet’s begin with what many commentators assume is the least likely avenue for reform: local government elections. Many commentators assume that municipal elections would not need a PR voting system because the electoral process is too local and per-sonalized. in this view, a city councillor is a well-known local personality with strong links to her constituents and their concerns. The whole élan of local politics, then, is more intimate and collaborative than other levels of government. PR, with its implicit assumption of enduring and serious divisions, would not be appropriate for local elections. Some analysts go so far as to claim that PR cannot be applied to the munici-pal level because local elections lack political parties—a functional requirement for a PR system in their view. And even if PR were appropriate, the design of any propor-tional system would require multi-member districts or an increase in the size of single-member districts, ultimately detracting from the local element that most analysts think is the strength of the local government system.

These criticisms are based on some enduring myths about how local politics really works in Canada, particularly in our urban centres. When we look more closely at the municipal scene, we see that many of the same enduring political divisions that char-acterize other levels of government are there too, meaning that PR is just as relevant. There are a number of positions that need to be challenged to make the case for PR at the local level—specifically, biases about the importance of locality, the absence of parties and politics, and the inability to apply PR without formal political parties.

MUniCiPAl MYTHS

Many modern Canadian municipalities govern millions of people. Most Canadians live in large urban centres—in fact, in 2001 64% of Canadians lived in just 25 census metropolitan areas—and providing them with local services is a complex and expensive job involving difficult political questions about who gets what. The idea that munici-pal politics is primarily about taking care of local concerns—i.e., tending to the local


ward and neighbourhood priorities—ignores how local governments make decisions that affect all their citizens regardless of where they live. Decisions about transporta-tion, downtown development and revitalization, urban parks and recreation, business versus residential property tax rates, and so on—all these questions tend to divide on a broad left–right continuum that mirrors those in place for provincial and federal elections. What this means is that any local factors in municipal elections are crosscut or informed by these other non-local concerns. This is backed up by the election results where, in the local wards (if they exist), political competition tends to divide between candidates associated with the left, centre, or right of the political spectrum.

The most obvious evidence that locality is not terribly important on the urban scene is the fact that many Canadian municipalities do not even accommodate local representation. For instance, Vancouver has no individual local wards—the whole city is one ten-member “at-large” ward. Most local governments in BC use at-large multi-member constituencies, as do a considerable number of medium-to-small towns across Canada (Tindal and Tindal 1995). Another blow to the “local” thesis is that the population size of ward ridings has increased in most major cities over the last decade due either to municipal amalgamations or a failure to match population increases with increases in levels of representation. The government of ontario effectively eliminated local government in Toronto in 1999 when it amalgamated Toronto and its suburbs into the former regional government boundaries, resulting in significant increases in the size of local wards (Thomlinson 2000). if locality is so important, someone forgot to tell the local and provincial politicians who authored and sup-ported these changes. on the population/representation problem, Vancouver hasn’t increased the number of city councillors since 1957, despite a dramatic increase in the voting population since then.

Another enduring myth about local politics is that parties are not a part of what goes on. This is wrong in two important ways. in some cases, it is just empirically wrong; there are party systems in place in a number of Canadian municipalities. Vancouver, Montreal, Winnipeg, and a host of other locales have been divided by competing slates for municipal office for decades, some since the 1920s (Tindal and Tindal 1995: 251–66; Smith and Stewart 1998: 27). These slates clearly represent the traditional left–right party-system continuum, with the left typically challenging business domi-nance in the city and focusing on improving or expanding city services while the right focuses on maximizing private development and minimizing public services (Culter and Matthews 2005). The slates are parties in all but name, and voters respond to them just as they do to parties at other levels of government. An examination of the voting patterns for slates in Vancouver elections reveals a high level of cohesion among voters supporting one side or the other—most give their support to all the designated candidates of their slate. in fact, whenever a high-profile councillor has left his slate to run independently, he has lost. But even where formal slates are not in use, quasi-slates—or what Smith and Stewart call “emergent parties” or “associated independents”—exist in many locales (Smith and Stewart 1998: 26).

The “lack of parties” thesis is wrong in another way, even in locales where coherent slate voting is not practised. Councillors in Toronto do not run in formal slates, but it would be absurd to suggest that political parties are not involved in city politics.


Many local candidates, particularly the successful ones, are linked to provincial and federal parties and they draw on these party links to get elected (Tindal and Tindal 1995: 267–8). An examination of Toronto city politics can easily link most candidates with the main provincial and federal parties. Though party lines may get blurred on occasion—particularly for candidates on the centre-right—they are there. After all, many provincial and federal representatives got their start at the local level; it is hard to believe that they only discovered party lines when they left the local scene. A more accurate characterization of local politics, then, would underline how it is not so much that parties are absent from municipal contests as much as that their role is regularly obscured from the voters.

if formal slates exist, the case for PR becomes clearer and essentially similar to those made at other levels of government. in a way the case may be even more press-ing, as the distortions in the results at the local level using slates can be extreme. The use of a multi-member at-large system in Vancouver has regularly produced signifi-cant distortions in the representation of different slates and dramatic shifts in levels of representation from election to election, despite broadly consistent levels of support for the two main groups over the past two decades. This is because multi-member plurality tends to exaggerate the all-or-nothing tendencies of single-member plurality. in fact, if slate voting is cohesive, it can lead and has led to one slate winning all repre-sentation on elected bodies despite the fact that their opponents were nearly as popular, as occurred in the Vancouver city council election of 1996.

one response to the problems of slate voting under multi-member plurality has been a call for a return to wards electing a single representative (Smith and Stewart 1998). Since the 1960s numerous campaigns have been launched in Vancouver with this goal in mind; the city has voted many times on the question, with mixed and un-clear results (Berger 2004: 13–7). of course, aiding this constant focus on the civic voting system has been the self-interest of one of the key political players. The left slate coalition—more typically underrepresented by the at-large system than the right—has consistently championed a ward system. But more recently some reformers have called for a proportional reform of the city’s voting system, recognizing that a return to single-member wards will not fix all the problems of representation at the city level (Tenant and West 1998a, 1998b). Unfortunately, a lack of unity on an alternative to the at-large plurality system has contributed recently to keeping it in place (Ward 1996). nonetheless, Vancouver remains a compelling case where a proportional voting sys-tem would effectively represent the different bodies of political opinion in the city.

What of cases at the municipal level where political competition is not organized into formal slates? Does this mean that PR could not be used and is irrelevant in discus-sions of institutional reform? The short answer is “no.” Critics here seem unaware that the STV form of PR does not actually require formal party labels to create propor-tional results. Because the voters in STV decide how to rank their preferred potential representatives, they can decide to what extent they think different candidates are linked together by policy and ideology. Voters in most locales have some sense of, or seek information about, where various politicians are coming from politically. Even without a party label, voting patterns in terms of the preferences made would tend toward a left/right split.


Returning as an example to Toronto, which does not have formal slates, most voters in the last decade understood that Jack layton, olivia Chow, David Miller, and many others were from the left while Case ootes, Rob Ford, and Jane Pitfield were from the right. Thus we might expect that those same voters using an STV system would pass preferences along based on which politicians they perceived were closest to each other. of course, nothing would stop voters from not doing this, just as some voters in Van-couver ignore the slates and vote on the basis of the candidates. But the trends in Vancouver suggest that most voters do look to slates to inform their decisions and it stands to reason that voters in Toronto would similarly use a left/right division to inform their choices under STV. But, as mentioned, STV allows voters to gain repre-sentation based on any division that is relevant to them. if not a left/right split, then STV could proportionally represent any other distinct bodies of opinion that exist with-in the polity. This is why STV is often used in many non-political party contexts where there are, nonetheless, abiding divisions that require representation. Contemporary examples of STV use in this vein include elections to the Anglican Church Synod and the nomination process for the Academy Awards.

As more and more of the contradictory effects of our increasingly globalized world manifest in urban centres and as the demography of our largest cities continues rap-idly to change, an effective representation of our complexity and diversity is essential. Systems like multi-member plurality and single-member plurality are blunt and inef-fective in representing even the most basic disagreements within our political system. Canada has already had considerable experience with STV use at the municipal level, in a host of designs. Winnipeg used both a ten-member at-large district with STV for municipal elections, as well as a number of smaller STV districts at different times. one proposal for reform of the Vancouver civic voting system in 1996 called for the use of a mixed-member proportional system that would combine a number of single-member wards with a top-up pool from a multi-member constituency. Depending on the town and the nature of political divisions, either approach could work effectively. in fact, such schemes are in use at the municipal level in a host of countries. Germany uses MMP and party-list PR for local government, while ireland has long used STV for municipal contests. Both Scotland and new Zealand, which use MMP for legisla-tive elections, recently adopted STV for use at the local level. Thus, in addition to our own historical experience with municipal PR, there are plenty of working examples of PR at the local level from which Canada could draw some practical lessons and inspiration.

As local political differences seem as abiding as those at other levels of govern-ment—and thus face the same problems getting represented under plurality—there seem few reasons not to pursue effective representation of them for municipal contests. of course, some common sense must be applied, and it may be that sophisticated methods of representation like PR are not appropriate or feasible in the many very small towns that dot the country. indeed, an indiscriminate approach to “modernizing” elections in western Canada after World War i contributed to the repeal of PR in many of the smaller towns that had adopted it with zeal; the administrative effort involved was simply too much for some locales to bear, particularly where local elections did not seem that divisive (Pilon 2006). Still, the decision is one that should be relevant


to the politics of the locale. in the case of cities in major and secondary urban regions, there should be less debate about whether PR is or isn’t appropriate, as these locales possess undeniable political divisions and the resources to handle increased complex-ity in voting and democratic administration. What is less clear is where the political will for change would come from. As at other levels of government, the greatest barriers to change are the politicians elected by the current system.

PR and the Provincial SceneThe lack of progress with voting-system reform on the federal scene, despite some discussion of it at various times from the 1970s through the 1990s, has shifted some attention to the greater likelihood of a provincial breakthrough. Some argue that this is the standard pattern of policy innovation in Canada. They suggest that most new ideas manifest at the provincial level before being adopted federally. Regardless of whether this is true, the provincial realm has become the de facto arena of voting-system re-form in Canada because provincial politicians in Quebec, PEi, British Columbia, new Brunswick, and ontario have been prepared to open up the discussion. However, this doesn’t mean that reform is necessarily more likely at the provincial level. in each case, the politicians have only unwillingly allowed some discussion of the issue, usu-ally as a response to a series of perverse election results or in the hopes that some very modest reform might emerge.

But as demands have grown out of their public consultation processes for more thoroughgoing reform, governments have placed more and more barriers in the way of change (Massicotte 2006). Certainly, the super-majority rules that privilege those in favour of plurality or that give more influence over the result to rural areas are the obvi-ous forms of manipulation. less obvious are the more subtle ways of influencing the results, like starving the process of proper public education on the issue (BC), limit-ing access to the polls (PEi), or simply stranding the process in endless committees/consultations (Quebec). Yet none of this takes away from the urgent need for propor-tional voting to be adopted for all provincial and territorial elections—it just moderates some of the more unrealistic enthusiasm. The barriers in the way of provincial reform are many, including the challenges of a severe rural/urban split in population densities, disagreement among reform forces about what form of PR to seek, and the enduring hostility of most politicians and the mainstream media to any change from plurality.

Despite the governments’ public discourse about modernizing elections and dem-ocracy, the voting system emerged as a government-sponsored issue in each province for primarily political reasons. in both Quebec and BC, a government had been re-elected in the 1990s with less voting support than its main competitor. When the losers then won the next contest, each had to contend with supporters and members of the general public who thought the voting system should be examined. Also propelling the issue forward was academic analysis that suggested that both these parties were structurally disadvantaged vis-à-vis their main electoral competitor. The argument was that both the BC and Quebec liberal parties had to win a higher proportion of the vote to gain the same level of representation as their nDP and PQ opponents, re-spectively (Massicotte 2006: 8).


in the Maritimes, the political conditions were a bit different. in the case of both PEi and new Brunswick, the conventional plurality system was regularly delivering very lopsided results—the opposition often had just one or two seats in the House, thus calling into question the legitimacy of the representative system. Premiers in both provinces established commissions to look into the problem, hoping they would recommend rather mild and unobtrusive reforms (Milner 2004b). in ontario, the liberal party committed itself to a BC-style citizens’ assembly and vote while in op-position. The party seemed to be hedging its bets. Historically, the ontario liberals have spent most of their time in opposition despite being the number-two party in the province for most of the 20th century. Even if they gained power in the coming election, getting re-elected had always been a problem for the liberals (Pilon 2004). once in power, each government underestimated the level of interest-group organiz-ation on the issue and the emerging consensus among academic experts that some kind of change was both practical and desirable, both of which helped tip the recom-mendations from the public process toward more serious change. now, governments faced trying to stymie the process without appearing to block it—a very difficult task given the quasi-public nature of the deliberations.

THE RURAl/URBAn DiViDE

one of the key issues introduced into the discussions was the rural/urban divide, which also tended to align with the various governments’ bases of support and the clientelistic networks that might stand to lose something from any change in the voting system. Even though in none of the provinces do voters make the rural/urban split a key factor in voting—in fact, in all cases voting is split among the major competitive parties to a greater or less extent—this rural/urban dimension came to dominate dis-cussions of provincial voting-system reform. Again and again, self-appointed local spokespeople invoked an unproblematic identity of “northern” or “rural people” who were somehow arrayed against their urban opponents (lang 2007b). The only prob-lem was that these rural and northern locales were united on very little politically. Thus instead of focusing on how a proportional voting system would accurately rep-resent the actual political divisions across the province (which are based on party identification), debates focused heavily on how a shift to PR would mean larger rural ridings and a lower level of rural representation in provincial houses (MacDonald 2005). The “rural vs. urban” bias in the discourse around reform had the effect of shifting the discussion away from the key issues that had driven consideration of the reform in the first place—that is, poor representation of what voters clearly do vote for: parties.

of course, the imbalance in all provinces between the concentrated southern urban population centres and the less-populated northern and rural areas is a real challenge for any voting-system designer. This very challenge was one reason that both Alberta and Manitoba pragmatically settled on mixed systems in their earlier periods of voting-system reform, ones that combined multi-member proportional voting in urban areas with single-member majority voting in rural areas (Phillips 1976). But here again we let a parochial approach to examining the problem limit our thinking. other countries


also face great swathes of underpopulated geography contrasted with smaller, highly populated areas. Sweden, norway, and Finland all contain highly urban and populat-ed cities at their countries’ southern tips while vast expanses of land stretch north with much smaller, dispersed populations. Their response has been to decentralize a great deal of decision making to the local level, as much as is realistically possible (Thomas 1985: 219). The same provincial governments that are acting on voting-system reform now also make the key decisions about local government. if they were sincere about the problems of underpopulated areas, they could incorporate a restructuring of local government responsibilities and financing into their process of voting-system reform. Unfortunately, the rural representation issue is, in reality, a red herring designed to mobilize opposition to changing the voting system. While these governments have mouthed platitudes about maintaining rural influence, their policies have belied such sentiments (for instance, see Beers 2005). or, to put it differently, those people in rural areas that support the government and its policies—supporters not much different from the government’s urban supporters—have maintained influence. Those op-posed, both rural and urban, share a similar lack of influence.

The real problem in getting voting-system change at the provincial level mirrors that at every other level of government in Canada, and that is the lack of determined political party champions. in BC, Quebec, and ontario, various parties had commit-ted to considering voting-system reform while in opposition, partly in response to pressure from supporters and partly as a kind of insurance policy if they lost again. But victory altered their incentives to support change. They had been, at best, hostage converts to the issue, moved by a fear that structurally they might be disadvantaged by the existing rules of the game. once elected, the losing spell appeared to be broken. nor have strong alternative supporters of change emerged. Though on paper the BC nDP supports a proportional voting system, the party remains deeply divided about changing the voting system, making them an unreliable supporter (Ruff 2004). The ontario nDP is more committed to change, but as the third party in the ontario system its influence is much less significant. The PQ is similarly ambiguous on pro-portional voting, supporting it when pressed—particularly when in opposition—but stalling or ignoring it once in office (Massicotte 2006). only the Greens remain con-stant in supporting PR, but they are still too weak at the provincial level to exert much influence.

Prospects for opening up a discussion about voting-system reform in the other re-maining provinces do not look good. The nDP in Saskatchewan and Manitoba—two provinces where the party regularly comes to power under plurality rules—are, not surprisingly, opposed to any consideration of reform. in Alberta, the dominance of the provincial Conservative party means that it too has little interest in allowing the question a public hearing. in nova Scotia, it is unclear whether a fundamental shift in the party system is occurring. What was once a traditional two-party system of Con-servatives and liberals has expanded due to a considerable increase in support for the nDP. For a number of elections now the three parties have gained roughly equal sup-port, producing successive minority governments. At this point, none of the parties wants to change the rules, as it is unclear who is moving up or down in popularity. Both the liberals and the nDP hope that the other party will decline and that they


will then benefit from an eventual assumption of majority governing power. For their part, the Conservatives too are unwilling to countenance change as they hope to translate their minority governing status into a majority sometime in the future. Some discussion of the issue appears to be emerging in newfoundland, but no dis-cernable public demand has arisen. of the territories, only the Yukon has studied the question but decided against pursuing it (McKinnon 2005).

Some might feel this “party self-interest” explanation of these recent reform initia-tives at the provincial level is too narrow an interpretation, as it ignores the possibility that some of the politicians—particularly Gordon Campbell and Dalton McGuinty—may have allowed the CA process to go forward out of a genuine commitment to a public consideration of reform. it may be that these politicians supported these initia-tives. As we cannot look into their minds, we will probably never know if their public statements did indeed match their personal commitments. But their personal “prin-ciples” are not the only factors guiding government policy on such issues. Despite the considerable power of party leaders and government premiers, they do answer to and must retain the support of their caucus, extra-parliamentary party, and the influential financial interests that tend to back them. in the end, if the actions of their govern-ments on this issue are anything to go by, the principles of individuals like Campbell or McGuinty appear to have been trumped or vetoed by other apparently more im-portant interests.

Despite a promising burst of activity focused on voting-system reform at the prov-incial level in the last half-decade, nothing concrete has been accomplished since the earliest rumblings in 2002. Two commissions, two CAs, and one mixed legislative/public approach have yet to produce a single successful voting-system reform. it is entirely conceivable that this interesting window of opportunity could draw to a close with only discussion to show for it.

PR and the Federal SceneMost academic work on voting-system reform in Canada focuses attention on the federal level. This is not surprising given the regional and language issues of the 1970s, the constitutional battles of the 1980s and 1990s, and the party-system meltdown and realignment from the 1990s to the present. The federal scene in particular seems to present itself as needing what voting reformers are offering: a way to represent our political and social diversity in a more sophisticated way. But voting-system reform at the federal level is arguably the most complex of all, necessarily involving an ac-commodation with various constitutional imperatives and raising the divisive ques-tion of what to do with the Senate. Some even argue that PR cannot be applied federally in Canada, as the constitution requires the use of single-member plurality. This last argument, as we shall see, is wholly uncompelling but the other issues do create challenges for voting-system reformers. Again, there are solutions that can ac-commodate these limits. on the other hand, there is some debate on just how these traditional limits would apply in a new voting system—something the courts may have to decide. in any event, any case for using PR for the federal House of Commons must begin with a response to these challenges and objections.


THE HoUSE oF CoMMonS

one of the more novel and surprising objections to introducing PR for federal elec-tions recently has been the argument that such a move would be unconstitutional. Rob leone argues that the “spirit of SMP is embodied in the preamble of the consti-tution which states that Canada’s government is based on that of Britain.” He then goes on to argue that a number of court cases involving Canada’s electoral system also infer that only single-member plurality is constitutionally appropriate, specifically citing Justice Mclachlin’s claim that our voting system forms part of the conventions we inher-ited from Britain. leone also notes that section 40 of the constitution makes reference indirectly to the voting system (leone 2006-07: 57–8). This new constitutional objection is surprising because the historical evidence against it is so overwhelming. SMP was not the norm in the UK until 1885—more than a decade after the 1867 BnA Act was writ-ten. Multi-member districts remained in use in both Canada and the UK well into the 20th century (Steed 1990). Britain even used non-plurality voting systems for a host of ridings between 1867 and 1885 and between 1918 and 1948 (Hart 1992). As for section 40 of the constitution, the wording says nothing about which voting formula is to be used to determine winners, though it does set out a schedule for single- and dual-member riding use in the original four provinces (irvine 1985: 103). However, section 40 is not really constitutional in the sense that it is subject to the constitution-al amending formula: its direction is modified in the first instance by the phrase “[u]ntil the Parliament of Canada otherwise provides … .” This effectively means that decisions about districting and district magnitude, the focus of section 40, are a mat-ter that Parliament can decide for itself. The voting system in Canada is just a law like any other—one that can be changed by a simple majority vote. Attempts to anchor SMP in the constitution lack any historical or legal basis and as such are unconvincing.

our constitution does not prescribe or proscribe the use of any particular voting system, but it does contain provisions that may affect the design of whichever system is chosen. There are a number of clauses that might have the effect of diminishing the overall proportionality that could be achieved with any PR system design. Both the Senate and grandfather clauses in the constitution effectively allow less-populated re-gions of the country to be overrepresented. The Senate clause ensures that sparsely populated PEi, which proportionately would be entitled to just one MP, gets at least as many MPs as it has Senators (four). The grandfather clause states that the less-populated provinces will never fall below the level of representation they achieved in 1985. The effect of these two clauses depends on the size of the overall House. if the House were increased in size, their effect would be limited or eliminated. on the other hand, even without an increase in the size of the House, these departures from strict proportionality in according seats to regions many not, in the end, have a decisive impact on the results.

Another constitutional provision that may affect the design of any PR proposal is contained in sections 51(1) and 52, which call for a proportional division of the seats in the House of Commons by province. This rule may affect the design of a mixed-member proportional system by limiting the application of compensatory seats to results within each province. Allocating compensation seats at the national level, as is


done in many European countries, creates more proportional results overall. But using provincial boundaries would not greatly diminish the proportionality of the results in the larger provinces. With over 100 seats in ontario, for instance, the province could use half its federal ridings as single-member districts and then use the other half as compensation seats—with fairly good proportional results. The challenge comes with the smaller provinces and territories, which have far fewer seats to work with. For instance, with just four seats in PEi, proportional results would be highly unlikely. it is possible that the courts would re-evaluate what the constitution means in the context of a different voting system, perhaps recasting the proportionality rule to include only the single-member district seats. or, they might allow national alloca-tions to have some influence on the allocation of seats within provincial boundaries as a “reasonable” departure from provincial proportionality (irvine 1985: 104–6). However, only a court test case or challenge would clarify this. in the end, though, even with these limits, a PR system would still deliver far superior results to our pres-ent plurality system.

THE SEnATE

Still attending to the federal level, some academics and reformers also think we should demand PR for a reformed and elected Senate (laycock 1994; Carty 1997). Unlike our provincial legislatures, Canada’s federal Parliament is a bicameral institution, a fact often forgotten or unnoticed by most people because our appointed upper House is largely marginalized in our democratic discourse by the elected—and therefore more legitimate—lower House of Commons. The appointed Senate has been the subject of much criticism throughout Canadian history, with proposals for reform that range from electing Senators to abolishing the upper House altogether (lutzig 1994). The issue is a divisive one. Some, particularly on the right of the political spectrum, feel that modern democracies must be bicameral so as to create a system of checks and balances on government. These reformers call for some kind of elected Senate. But others, ranging from provincial advocates to the centre-left, argue that the Senate is redundant and unnecessary. not surprisingly, these reformers support abolition. in the Canadian context, a third group includes those who wish to just ignore the Senate, as it is relatively powerless in its current form. Their view is strengthened by the reality that Senate reform is a constitutional question, which means it is very difficult to do anything about. Still, proposals for a PR-elected Senate have emerged from time to time, as part of the Charlottetown Accord negotiations (ca. 1990–92) and, most re-cently, in a rather provocative, innovative, and perhaps questionably constitutional set of reform proposals from the Conservative minority government of Stephen Harper. Should PR for the Senate be added to the campaign to change the voting system for the House of Commons? it is not a straightforward question.

The problem with addressing PR for the Senate is that one must first address what the role of the Senate is or should be. For those opposed to an upper House, electing the Senate will only legitimize it and give its deliberations more authority. Some might wonder how anyone could be opposed to converting an appointed chamber into an elected House. Basically, it amounts to differing philosophies of government.


in the US model, checks and balances exist to slow government and legislating down, as each branch of the government can exercise a veto—at least for a time—over legis-lation. Thus the House of Representatives can reject bills from the Senate and vice versa, the president can veto bills approved by Congress, and so on. in Canada’s par-liamentary system, the “efficient secret” (as one scholar called it) is that things can get done because the executive is linked to the legislative branch and, with a majority of votes in the House of Commons, can pass any bill it likes. Some think this system gives too much power to government and want an elected Senate to slow the government down. others say that the way to fix the problem is not to introduce a veto body but to alter the nature of governing in the legislative branch, and the way to do that is to introduce PR for the lower House. in the bicameral “solution,” checks and balances mean that deadlock can result and little may be done on difficult issues. But the PR solution helps create broader coalitions to support government actions, both by bet-ter representing political diversity and by preventing the monopolization of power by a minority (as is common under plurality). Thus PR limits the power of the executive by forcing it to work more collaboratively without preventing things from getting done. The bicameral approach is defensive and reactive; the PR approach is proactive and collaborative.

There is an element of realpolitik involved in these discussions of Senate reform. if any government succeeds in moving an elected Senate forward, the question then be-comes: what should PR supporters do? if they don’t like the Senate, will they like it any more or less elected by plurality rather than PR? obviously, if it comes to that, even those opposed to upper houses will have to think long and hard about what to do. Many will undoubtedly cut their losses and agitate for a PR voting system for the Senate. But in the interim, pressure must be brought to bear on the government for avoiding the more obvious locale for reform—its own legislative chamber, the House of Commons. normatively, there is something highly suspect about a government advocating change for somebody else while scrupulously avoiding discussion of its own chamber’s voting system. But, as this book has consistently argued, normative concerns are seldom the driving force of institutional change. it’s all about politics.

Canadians are divided about politics at every level of government. They don’t agree about the policies that should be introduced or the role that government should play in society. These differences form the basis of our political community. They must be represented accurately for effective deliberation to occur and for genuine compromise to emerge, if possible. An effective representation of differing views is also the foundation upon which the legitimacy of the decision-making process rests, regardless of legislative outcomes. Thus pursuing PR for all levels of government makes sense.

ConclusionThe road to reform is filled with many potholes and any one may pitch the reform forces off the path to their goal if they are not careful. in some cases, the choices we face divide reformers along the underlying political divisions that they have stepped outside of in order to address voting-system reform. Conservatives tend to have


strong views on the illegitimacy of the courts when it comes to democratic institu-tions, while the centre-left is more amenable to court challenges to get past political blockages. The right typically feels strongly about an elected Senate where the left ei-ther dismisses it as a priority or seeks to abolish it. in other cases, reformers simply disagree about priorities in terms of which level of government is most in need of re-form, or have developed strong opinions in favour of one form of PR over another.

This chapter has tried to address these challenges with some pragmatic advice and concrete analysis to counter what are often flights of principled fancy—if only re-formers could shape the terrain upon which they struggle to make change! But the reality is that reform forces must seize whatever opportunities arise, given the power inequalities that exist between those that can make change and those that want it. They must forward their agenda wherever an opening emerges, at whatever level of government. And they must avoid overheated commitments to particular pet voting systems. The key distinction that divides the concrete struggle to change voting sys-tems is between proportional and non-proportional voting systems. That is, and must remain, the key focus of change. All other details can be addressed more readily after a change to some proportional voting system has occurred. But within the confines of our present system, the nuances of what reformers want will remain ignored.

119

CHAPTER SEVEn

Debating the PR Alternatives: STV Versus MMP

introductionProportional representation is a generic term for any proportional voting system. Thus supporting PR means supporting a general principle—that is, the idea that there should be roughly proportional results for the parties competing in our elec-tions—but it doesn’t tell us very much about just how such results might be produced. The general principle of PR is great for getting people to think critically about our existing un-proportional plurality voting system. The next step, however, usually in-volves exploring how proportional representation would work in practice. And here is where all hell can break loose. it’s funny how quickly a group united in their con-demnation of plurality can disintegrate into warring factions for STV, MMP, or some modified, untested invention of an amateur voting-system engineer.

There are essentially three kinds of PR. The party-list approach is the most pro-portional and, in many ways, the easiest to grasp. A party gets 40% of the vote and the first 40% of the names on their party list are declared elected. But despite its simplici-ty, party-list PR was never a favourite among citizen-driven PR movements in the UK, Canada, and other English-speaking countries. in fact, in nearly all applications in Europe it was introduced by elites with no public input, in an environment where party representation was considered most important and local representation was felt to be irrelevant. Having said that, all these countries were geographically small—un-like Canada. not surprisingly, for geographic reasons if nothing else, no one is calling for party-list PR in Canada today. The debate is really between the other two main forms of PR: the single transferable vote (STV) and mixed-member proportional (MMP). This chapter will review the debate over the “right” version of PR and argue that STV and MMP could work equally well in Canada, providing they are fully proportional.


STV Versus MMP: A Voting-System Grudge Matchin most cases in life, choice is good. But experience in voting-system reform move-ments suggests that even two choices may be too many. Basically, it is not uncommon for voting-system reform advocates to develop very strong opinions in favour of one form of PR over another. in some cases, these views reflect a fetishization of one or more aspects of the voting system. STV supporters claim their choice will allegedly empower the people over the parties and is thus superior. MMP fans hold that their system maintains the single-member riding—the allegedly crucial link between con-stituents and a local representative—and thus is better. in other cases, the choices reflect political objectives that fall on and across the left–right divide. For instance, there generally seems to be a centre-right bias for STV and a centre-left bias for MMP (though each system has some support on the other side of the political spectrum as well). The divisions over which form of PR is best can become so bitter that some-times partisans would rather maintain plurality if they cannot win support for their particular system. in the 2005 BC referendum on STV, the bulk of the “no to STV” campaign argued that its supporters were actually not opposed to PR per se, but sim-ply to the STV version of it (Anderson 2005). Some STV supporters returned the compliment, suggesting they would never support an MMP proposal.

Fair Vote Canada (FVC), the leading PR advocacy group in Canada, stands as an excep-tion to this trend. it supports both STV and MMP, though negotiating the balance has proven tricky. in 2004, partly amid fallout over the BC-CA’s decision to endorse STV, FVC managed to lose both its president (Doris Anderson, who denounced STV) and its vice-president (Bruce Hallsor, who supported it) (Doris Anderson 2005; YES for BC-STV 2007). Clearly, debates over which is the better PR system can get out of hand, threaten-ing the unity of reform forces and thus the campaign to get rid of plurality altogether.

Debates over which form of PR is better or worse suffer from many problems. These include overstatement, institutional determinism, romanticism, reductionism, misplaced emphasis, and many more. Probably the most glaring omission from such debates is politics itself. in the same way that defenders of plurality ignore institutional factors that influence politics, the various PR-system partisans ignore how politics can affect the results that emerge from institutions. in the end, their arguments are just the inverse of their opponents’ indifference to institutions—they appear to be ar-guing that the wrong choice of rules will create an insurmountable barrier to certain kinds of politics, a position that is indifferent to political innovation. A critical review of the arguments for and against STV and MMP will demonstrate the weakness of these positions and the futility of much of the debate. Reformers have got to put the “politics” back into their analysis if they want to grasp how institutions and human agency interact. More to the point, a sense of politics is crucial to the reform project they have undertaken and to its eventual success or failure.

STV Versus MMP: The Terms of DebateDivisive battles over particular proportional voting systems are nothing new. Most historic movements for voting-system reform made little progress until they sorted

d e b at i n g t h e p r a lt e r n at i v e s : s t v v e r s u s m m p / 121

out which version of PR they were prepared to champion. in the early 20th century Anglo-American countries settled on STV as the only kind of PR they would campaign for (Hoag and Hallett 1926). A host of factors influenced their thinking, including a populist anti-politics, anti-party sentiment; a concern to give voters influence on the individual candidates elected; and the idea that STV would encourage political coop-eration over competition. Some reformers, particularly in the British context, did highlight the ability of STV to represent serious differences within political parties as well. it wasn’t just coincidence that many of the early champions of STV in the UK found themselves in a minority position within their parties on various issues like irish home rule, British imperialism, and questions of suffrage extension (Hart 1992). All in all, the stated objectives of these historic reformers—less party discipline, more minority representation, and greater cross-party cooperation—do not sound that much different from the objectives of present-day STV partisans.

Today’s STV supporters claim that the system is superior to all other forms of PR because it is voter-centred, because elected members have more freedom from party rule, and because it retains an element of local representation. it is voter-centred be-cause the voters mark their ballot preferentially, ultimately determining just which members of a party get elected. Since voters get to number their preferences across the candidates on the ballot, they decide which members of a party they like more than others, or they can ignore party lines and rank candidates on any basis they like. STV supporters defend their claim that elected members have more freedom from party rule under STV on the basis that, since party members gain election by getting per-sonally endorsed by the public along with their party, they have more moral authority to act for their constituents in any dispute with their party. Besides, representatives can argue that if they do not attend to their locality, they may lose their personal support in the next election. Finally, the local basis of STV is defended in that the multi-member ridings are smaller than in other PR systems and can thus be adapted to local geogra-phy and accommodate dispersed populations. These claims have resonated with all sorts of contemporary anti-party populists, both left and right, who are suspicious of PR systems that give parties the power to decide which candidates will be elected and who wish to see a decline in or end to the high levels of party discipline in Canadian politics. Supporters on the right have particularly warmed to STV’s promise to make local members more accountable and re-assert locality and local interest against the dominance of parties, though some on the left also like this idea.

STV has elicited a storm of criticism in the Canadian context over the last decade. Critics complain that STV is too complicated, both in that it expects too much from voters (i.e., they must number their choices across a range of candidates) and in that it lacks transparency when it comes to adding up the votes and producing winners. Certainly the media theme in the 2005 BC STV referendum was that STV was both confusing and complicated. others complain that STV takes away the single-member riding of our current system and creates multi-member ridings that are too large geographically to be considered genuinely local.

Many of the points that STV supporters see as positives are converted into nega-tives by its critics, particularly the system’s alleged potential effects on party discipline. Some critics worry that STV creates too much competition between members of the


same party, thus potentially weakening the efficacy of parties in our political system. Canadian women’s groups have also complained that STV is a bad choice for women because its multi-member riding structure is too small to create the positive incen-tives that researchers tell us are necessary for bettering diverse representation. STV multi-member ridings typically elect just three to five members, while experts claim that seven-member ridings are a minimum to improve women’s representation. Fur-thermore, women’s advocates claim that international experience with STV has not provided compelling results in terms of boosting representation for women (Doris Anderson 2005; Equal Voice 2005). Additionally, some argue that STV ridings would be too big geographically for women to campaign in due to the costs of covering such large areas. But one of the key complaints lodged against STV by those who support PR is that the system simply is not—and cannot be—proportional enough given its structural bias in favour of small multi-member ridings.

For most of the 20th century STV was the preferred choice of would-be voting-system designers in English-speaking countries. it still is in the United States, where a candidate-based discourse and weak party discipline make any other form of PR appear unintelligible and seemingly irrelevant (Amy 1993). But elsewhere a shift oc-curred in the 1970s as the discussion of voting-system reform moved away from citi-zen reformers and into the hands of academics and politicians. Experts and professional politicos understood the problems of conventional politics differently than populist reformers did, and were less moved by anti-elitist and anti-party senti-ment. Their concerns were to address the shortcomings in the political system in terms of accomplishing what parties and elites thought was needed.

in Britain, the Hansard Society broke with tradition in the 1970s and recommended a modified MMP system, primarily to better represent the parties proportionally and regionally (Hart 1992). in Canada, the Pepin-Robarts Task Force on Canadian Unity recommended a less-than-proportional MMP system in 1979 in the face of regional discontent and the possible secession by Quebec. As academic- and politician-authored studies piled up in the 1980s and 1990s in Canada, the balance swung firmly into the MMP camp (Seidle 1996). William irvine dismissed STV as divisive and complicated in his 1979 monograph Does Canada Need a New Electoral System? He recommended MMP instead (irvine 1979). A series of MMP proposals have been offered up since then. By the time the BC-CA got started, few academics expected any serious consid-eration of anything but MMP (Milner 2004b; Blais, Carty, and Fournier 2007).

MMP supporters stress the pragmatic nature of their choice. They make a virtue of its continuity with our existing single-member-based system while highlighting its capacity for complete party proportionality. For these reasons, some claim that MMP represents the “best of both worlds” between SMP and PR (Shugart and Wattenberg 2001). There is a strong sense of realpolitik in these arguments. Proponents want to be seen as offering something to keep both the voters and parties happy (they think that voters want something familiar and believe that parties want a system that is perceived as party friendly). MMP supporters also maintain that their system is easier to explain and simpler to understand than other forms of PR. Activists keen on diverse represen-tation claim that MMP is much better than STV in increasing the numbers of women and visible minorities that get elected. Finally, where serious parties have taken on


commitments to voting-system reform in Canada over the past few decades—the nDP, the Greens, the Parti Québécois—they have almost uniformly endorsed MMP.

When the BC-CA chose an STV model over MMP as the voting system recommen-dation they would make to the people of the province, many voting-system activists and academics were surprised (Milner 2005; Massicotte 2006). But the activists and academics had underestimated the degree of criticism aimed at today’s parties be-yond elite opinion and academic studies. The BC-CA claimed their decision to reject MMP was based in part on concerns about undue party control under that system, particularly over the party list of candidates. They were concerned about who would choose this list and who the list members might be accountable to. The general feeling was that MMP was too party-centric, a problem that BC-CA members had already identified with the current system (MacDonald 2005; lang 2007b).

Thus the debate between STV and MMP forces really boils down to their views of parties and what they believe the problems with parties are. STV supporters tend to see the existence of parties, and their behaviour, as a key contemporary political prob-lem that voting-system reform should address. MMP supporters see the problem as being not so much with parties per se as with the fact that the current plurality system does not represent parties effectively. STV advocates hope to affect party behaviour by creating more competition within parties, while MMP advocates hope to affect party behaviour by increasing the competition between parties. in STV, voters directly dis-cipline the politicians—with only indirect effects on their parties. in MMP, voters directly affect the standings of the parties—with only indirect affects on the politi-cians. Partisans for each system tend to be convinced that their highlighted problem is the most important, that it is in fact the key hinge upon which the success of the entire reform enterprise rests. no wonder they battle in a “take-no-prisoners,” fight-to-the-finish style.

The intense debate over the claimed necessary effects of the different forms of PR represents a fetishization of certain aspects of both systems. The partisans have taken what are at best general tendencies of the systems and converted them into causal laws about their effects on politics and representation. STV does offer voters greater latitude in making choices, but parties remain key players in STV systems, exercising considerable control over the range of candidates nominated and ultimately elected. MMP systems are by definition supposed to be fully proportional, but they can also be designed in ways so as be less than proportional in their results—as the recent Quebec “MMP” model demonstrates. it is time to scale back the claims made for both systems to more modest proportions. Much of what passes as informed discussion of proportional voting systems is both bad analysis and ultimately bad strategy from the point of view of securing reform. let’s review the claims and counterclaims to sort out what is defensible from what is not.

Simplicity Versus Complexityone of the key arguments in the debate over PR systems involves complexity. Put simply, some people think that “simple” is better. Here MMP supporters claim the high ground, arguing that STV is too complex for voters to use, for returning officers


to count, and for the general public to understand in terms of just how the votes are converted into representation. By contrast, MMP is allegedly more straightforward because it involves plurality voting (a system already known to the public) for one group of seats and party-list PR (the most mathematically intuitive form of PR) for the other group. of course, the MMP forces must tread carefully playing the simplicity card; at least in terms of voting and adding up the ballots, plurality trumps both MMP and STV in a duel over simplicity.

The simplicity arguments are dubious on a number of grounds. Why is simplicity so important for electoral systems but not other arenas of life? Most people in western countries have to learn to operate cars, personal computers, and audio-visual remote controls, all of which have become increasingly complex in the past two decades. Most adults have to use bank machines for financial transactions, and the interfaces on such systems are more complex than the requirements of voting under STV. Com-plexity seems fine—indeed, increasingly mandatory—in most areas of life, so the arguments against it in voting seem arbitrary and unconvincing. of course, com-pared with plurality voting and counting, all other voting systems are complex. But then plurality also produces substandard election results, so one must assess the trade-offs. is simplicity in voting and counting the ballots important enough to ac-cept distorted election results? Put this way, simplicity doesn’t sound so impressive.

The simplicity arguments also fail important empirical tests. Farrell compares rates of ballot spoilage in the 1990s across a broad range of countries and voting sys-tems and finds that Canada (SMP), ireland (STV), and Germany (MMP) had nearly the same level of spoiled ballots—between 1.3 and 1.5% of the total votes cast (Farrell 2001: 202). nor is this trend unusual. The fact is that irish voters have used STV for nearly a century, and their levels of spoiled ballots do not suggest that voters there find the system too complex (McKee 1983; Farrell 2001). Voters in Manitoba and Al-berta—literally millions of Canadians—used STV for three-and-a-half decades, and most had no trouble marking their ballots. For instance, ballot spoilage in Winnipeg ranged from 0.9 to 1.8% of the total votes. in Calgary and Edmonton, where ballot spoilage was higher, scholars argue that the basic workings of STV were less the cul-prit than particular rules about accepting or rejecting ballots (Jansen 1998: 87–8; Jansen 2004: 658–60). it is hard to believe that Canadians would be less able than their irish peers in taking up a voting system like STV. And it is particularly uncon-vincing that people today—more educated and complexity-savvy than any prior gen-eration—could not cope with STV at least as well as Canadians voting with it circa World War i, the Depression, and World War ii. As for counting the ballots, again, historical and comparative evidence suggests it can be and is done with few problems. Today one can even count an STV election with a computer (utilizing paper ballots to maintain a paper trail), thus getting instant results.

The argument that MMP is more simple and straightforward than STV is also un-compelling. As activists in PEi discovered during their referendum in 2005, anything that it not familiar to political elites and media gatekeepers can be made to sound confusing and strange to the public (McKenna 2006). Though MMP retains Canada’s familiar single-member ridings, the party-list part of the voting system and its rela-tionship to the single-member districts can sound confusing to the uninitiated; the


three stages to the vote-counting process required to allocate all the seats in the elec-tion can appear particularly complex. A visit to the Elections new Zealand website and a look at their breakdown of the allocation of seats for the 2002 national elections under MMP should disabuse even the most partisan supporter of the notion that the system is “simple” (Elections new Zealand 2007). of course, most aspects of our cur-rent plurality system are confusing to voters and media elites. For instance, few can explain why the popular vote for parties and their percentage of seats won typically diverges wildly under our system. Polling suggests that many voters confuse legislative majorities with popular-vote majorities, thus granting more support to winners than they really deserve (Bricker and Redfern 2001). Thus all voting systems are complex in various ways. The real issue with complexity in any voting system is the amount of public education undertaken to raise public awareness about the voting instrument. Where effective education strategies are adopted, there are few problems with the complexity of any voting system.

Pro-party Versus Anti-partyif complexity arguments are unconvincing, what of declarations that this or that PR voting system is either pro- or anti-party? STV supporters claim that their choice is a voter-centred system, while opponents claim STV is divisive for parties, creating com-petition among candidates of the same party and preventing representatives from taking a broader view of the issues than their local balliwick. The claims on all sides suffer from serious distortion and overstatement. STV does allow voters to express a more sophisticated set of preferences across the candidates up for election. For those who study elections, STV ballots offer a mine of rich information about how voters see the party system and the relationship of parties to each other. And surveys report that voters like making preferences (Farrell and McAllister 2003). But the fact that voters can make a range of choices under STV does not mean that this system is any more “voter-centred” or “anti-party” than any other.

The reality of modern politics is that parties are key to the political system—no voting system is going to change that. Australian use of STV for its upper legislative chamber from 1949 to 1983 (changes thereafter made the system more like party-list PR) did not prevent the Senate from being dominated by just a few parties with strong party discipline (Hughes 2000; Stone 1998). Malta’s use of STV from 1921 to the pres-ent has witnessed the dominance of a two-party system over the entire era (Proctor 1980; lane 1995). The irish also have a strong party system where typically just a few parties dominate (Busteed 1990). Even Canadian experience with STV provincially was characterized by fairly conventional party-like behaviour, despite the election of populist and coalition governments (Phillips 1976; Jansen 1998). nor is there much evidence to support the view that STV leads to party fragmentation because candi-dates of the same party compete against each other in a multi-member riding. in fact, candidate defeat in ireland is much more the product of swings in party support than competition between candidates of the same party. in Malta, where party competi-tion is stable and restricted to two parties, representatives do tend to be defeated by members of their own party, but this reflects the impact of a series of special rules


that motivate parties to run many more candidates than there are seats in the multi-member ridings and that have little to do with the STV system specifically (Gallagher 2000: 96–9).

Thus STV does not prevent strong parties from being at the centre of competitive politics. After all, voters ultimately choose from the candidates nominated by their party—the party still controls the range of offerings. Marsh argues that voters in ire-land do pay attention to individual candidates, but that this works within a system that is very much dominated by parties. He adds that irish election campaigns are defined by parties and that strong party discipline is characteristic of the irish parlia-mentary system (Marsh 2003), an assessment confirmed by other experts in irish elections and politics (Gallagher 2000: 109–10). Voters could vote for independent candidates—and such candidates might stand a better chance of getting elected with STV rather than in our system—but typically they do not (Carty 1981; Bowler and Farrell 1991). in practice, a small number of independents do get elected in ireland but most tend to be former party representatives disgruntled with their party leader-ship. Because parties mobilize resources, policy positions, and people, voters tend to take their cues from the parties they trust in STV systems, just as in other systems. This view means that STV should not be seen as being any more or less friendly to parties than any other voting system.

Conversely, MMP should not be seen as more party-centric than STV. The idea that an MMP system gives undue influence to party bosses because they will control the order of the party lists is a view that ignores the considerable influence of party hierarchies on candidate nominations in all systems. it also ignores any engagement with the actual practices of MMP-using countries where various methods of ordering the lists are in use, including a state-mandated democratic approach in new Zealand (all parties must order their list democratically through a vote of party members) and a party-centred democratic one in Germany (where competition between parties has pushed more democratic nominating processes).

The Quality of Local RepresentationAnother area of dispute between partisans of different PR systems is the quality of local representation offered by each. STV supporters claim that their small multi-member ridings maintain a degree of local representation while providing other benefits, such as increased proportionality for parties. in fact, they argue that their approach is su-perior to the single-member riding approach because individual voters are more likely to find a representative they agree with in a multi-member riding. MMP supporters disagree, arguing that STV’s multi-member ridings do not really constitute local rep-resentation and that the maintenance of the single-member riding is most crucial for maintaining effective local representation. STV partisans counter by noting that the single-member ridings under MMP must be much larger than typical plurality ridings so as to carve out some seats for the party list elections, and that this compromises the quality of single-member-district representation. They also raise objections about the “two classes” of MPs created by MMP, as some elected members will come from


local ridings while others will be elected from a party list. They complain that the party-list members will not be chosen by the electorate and will have less to do than the constituency representatives.

Both approaches suffer from an overvaluation of the efficacy and importance of local representation. The bottom line is that, while local representation is much talked about by politicians and elites (and even voters themselves often put their concerns in what is essentially a hegemonic discourse—what Peter Mair calls the “ideology of local representation”) local representation is not really that important in modern politics (Mair 1992). As a recent study noted, even in electoral systems that allegedly encourage a “personal vote” so that local candidates pay attention to local concerns, there is little evidence that local factors have much independent impact on voter choice (Morgen-stern and Swindle 2005). Put bluntly: people vote “party,” not “locality.” Most have difficulty naming the politicians with the highest media profile (leaders, cabinet minis-ters), let alone the local MP, and very few are prepared to vote for a local candidate who is not also the candidate of their favoured party (Gidengil et al. 2004; Blais et al. 2003). in new Zealand and Scotland, much of the “two classes of MPs” controversy was ac-tually fuelled by the politicians themselves, either to gain advantage over their political adversaries or to discredit the MMP system (lundberg 2005). in light of these facts, it seems absurd to design a voting system around an issue that is clearly not decisive in terms of election results. of course, if you can have a degree of local representation, there might be compelling arguments to do so (Docherty 2005). But present discus-sion assumes the pre-eminence of this issue, despite much evidence to the contrary.

Diverse RepresentationMMP proponents become greatly impassioned when the discussion turns to the question of diverse representation. Here, they often state categorically that STV has not and cannot accomplish the goals of those seeking greater equity in terms of rep-resenting our social diversity. There are a number of strands to this argument. on the one hand, these critics argue that countries that have used STV have performed poor-ly in terms of electing women and visible minorities. At a broad sweep, it is hard to disagree. Malta, ireland, and Australia have had some of the lowest levels of women’s representation across western industrialized countries (lane 1995; White 2006; Rydon 1994). on the other hand, they argue that the problem is structural, not merely cultural. Specifically, they suggest that because STV has smaller multi-member constituencies than other PR systems and an effectively “open list” approach to ranking who gets elected, women and visible minorities stand little chance of gaining election. Small districts are a problem because they intensify political pressure to get near the top of the list. if three parties are likely to elect members from a five-member riding, only those members with the top rankings for each party stand to get elected; according to the critics, they will likely be men. The voter influence in STV is also seen as a prob-lem. Critics maintain it will prevent parties from seeking to give women and visible minorities an advantage by placing them higher on their lists because the voters will rank men over women. For all these reasons, a number of women’s organizations in


Canada have argued that STV would be worse than plurality (Equal Voice 2005). Certainly this thinking was behind the concerted women’s movement opposition to the adoption of STV in BC in 2005.

The “diverse representation” arguments against STV represent some of the worst examples of deterministic thinking about the impact of institutions. Political scientists have long held that PR systems tend to be more open to calls for diverse representa-tion than plurality and majority systems (Rule 1987; Matland 1993; Matland and Studlar 1996; norris 2004). The reasons have to do with the combined effect of the voting formula and multi-member districting. The multi-member districts allow parties to craft a balanced ticket that reflects different demands for representation. By contrast, this is hard to do with single-member districts because, with only one winner, there is often a battle to gain the nomination and the winner is typically the candidate with the best resources—usually a man. With multi-member districts, the opportuni-ties for diverse representation will also depend on the competitive situation of the parties—more parties means fewer representatives elected from the lists by each one. This is why so many analysts insist that the longer the list, the better the chance of electing more diversity (Matland 2005).

in a large multi-member riding, a number of parties might each elect more than a few members from their list, thus increasing the chances that one or more of those elected would be women or visible minorities. Activists use such directives to con-demn PR systems that use smaller districts. But what they forget is that the voting formula is perhaps even more important in the dynamic that moves calls for diverse representation forward. it is the proportional formula that facilitates greater party competition and, with more competition, those seeking diverse representation need only influence one party to possibly create a “contagion effect.” Where parties make diverse representation an issue, and the level of proportionality is such that party in-novation on issues can impact other parties, progress can be made on this question even in PR systems with small ridings. nor does the research suggest that the ability to indicate preferences necessarily works against women; in a number of cases, women have used such mechanisms to increase women’s representation (Matland 1993). in other words: politics still matters in gaining different forms of representation, even in a PR system. And this means that STV can facilitate better representation of women, where the political will exists (Gallagher 1986; White 2006).

STV critics have been led to their conclusions by inattention to the methodological limits of their expert sources, an unwillingness to accord cultural factors due weight in influencing levels of diverse representation, and by a misreading of the election re-sults themselves. The research into the relationship between women’s representation and voting systems has produced a complex account of the factors that contribute to just how diverse the election results will be in any given locale. But all researchers agree that women’s representation is generally best in party-list PR systems and weak-est in plurality and majority voting systems (norris 2004). in examining the specific factors influencing this result, researchers demonstrate that the highest levels of di-verse representation have historically tended to coincide with PR systems that have the largest multi-member ridings. This is one reason MMP supporters condemn STV, where the average riding size is typically just three to five members.


But these research results are presented with many more qualifications than STV’s critics let on. What we have here is a correlation between different factors, not an argu-ment that the existence of one factor causes the other. What researchers are saying is: “this is the relationship we have discovered thus far, that with X-kind of multi-member ridings under PR we tend to find Y-level of women’s representation.” it is not saying that the existence of X causes Y. This is an important distinction because STV critics do just that—confuse the correlation of different factors with the idea that one of the factors causes the other. But as one academic who uses this method cautions, “correlation does not mean causation, no matter its strength or statistical significance.” instead, she notes that although better women’s representation and PR is strongly correlated, “the precise reasons for this pattern remain a matter of speculation.” Thus the method is good at spotting patterns but weak at explaining what gives the patterns shape (norris 2004: 23). What this means is that while such research can highlight past pat-terns, it cannot necessarily anticipate new developments. For instance, taking this method back to 1950—when few women were elected anywhere—researchers would have suggested that there was no relationship between the voting system and women’s representation because there was no measured correlation. in other words, the method is like a snapshot of a particular moment but it does not tell us why certain patterns have emerged, continue to exist, or how they might change. For that we need a broad-er understanding of political change, one that takes seriously the nuances of different political contexts and the agency of political actors.

STV critics also tend to underplay another important aspect in the study of voting systems and diverse representation: the impact of culture. one of the reasons that STV fares poorly in cross-national voting-system comparisons is that so few countries use it for national elections. With such a small sample—just three countries—it is hard to know how STV might work in a number of diverse settings. As it stands, all three countries have strong cultural factors that have militated against progress on diverse representation. All three have witnessed strong Catholic influence on their political systems, with the effect of limiting the mobilization of a modern women’s movement. in Malta and ireland, the Catholic Church has been a dominant institu-tion, both socially and politically (lane 1995; Girvin 1996). in Australia, working-class Catholic voters had strong links to the labor party, thus blocking one typical avenue of feminist influence (Coleman 1988; Brett 2002). What is telling is that as the institutional and party influence of Catholicism has waned considerably in the last three decades, noticeable improvements in women’s representation have emerged in all three countries (White 2006; Curtin 2006).

As sketched out above, improved women’s representation has emerged historically from a number of factors working together, including a mobilized women’s move-ment in civil society, pressure on the party system to take up the issue, the presence of an electorally popular left or post-material party, and a voting system porous enough to allow for keen inter-party competition and to make parties ready to respond com-petitively to what other parties are doing. But if one of these factors is missing, little progress on women’s representation is likely, regardless of the institutional structure.

nor is the experience of STV nearly as dire as the critics allege. A more careful ex-amination of the election records demonstrates some compelling evidence that using


STV did make a difference in levels of women’s representation in Australian elections between 1949 and 1983 (after 1984 changes to the STV ballot essentially converted the system into a form of party-list PR). Comparing the STV-elected Senate with the majority-elected lower House, women’s representation in the former averaged 8.6%—well above the lowly 0.6% average in the latter (Pilon 2007). At the local level, Kathleen Barber discovered that STV did improve the representation of people of colour in ohio from the 1930s to the 1950s (Barber 1995). The use of STV for community school board elections in new York City between 1970 and 2000 also led to highly rep-resentative results in terms of the different neighborhoods’ ethnic and racial diversity (Weaver and Baum 1992; Sinha 1993; Richie 1997). Even irish national elections dem-onstrate a slow and steady improvement in women’s representation—despite Catholic influence and a weak women’s movement—suggesting that even marginal improve-ments in the status of women could be more readily absorbed politically by STV than, say, Canada’s plurality system over the same period (White 2006; Pilon 2001).

The evidence against institutional determinism extends beyond STV. Even more modest approaches to minority representation (semi-proportional systems like the limited vote and cumulative vote) have proven effective in producing results in terms of representing diversity in various local elections in the US (Brockington et al. 1998; Bowler et al. 2003). nor is MMP’s record on diverse representation entirely stellar. Germany was one of the last western European countries using PR to move forward on women’s representation, and until the 1990s many experts considered PR to be as “problematic” as STV for women and those keen to further diverse representation (Kolinksy 1991; Rule 1987). Thus an assessment of the possibilities under different institutional ar-rangements must balance institutional factors with social and political ones.

ProportionalityEven if all of the issues above could be dealt with in a civil manner, for some combat-ants the debate between STV and MMP would remain intractable. For them, the key

FigUre 7.1 aVerage perCentage oF Women eLeCteD UnDer St V, By DeCaDe

Country jurisdiction 1970s 1980s 1990s 2000s

australia Senate 6.9 18.0* n/a n/aaustralian Capital territory State legislature n/a 23.5 23.6 35.3ireland national legislature 3.2 7.2 12.0 13.2ireland european parliament 13.0 10.0 30.0 38.0malta national legislature 4.0 3.0 5.5 9.3northern ireland assembly 5.1 3.8 13.3 16.7tasmania State legislature 1.0 7.1 25.5 24.0

* Ballot changes after 1983 converted the StV system to a de facto party-list pr system.

Source: Compiled from White 2006; Curtin 1999; tasmanian parliament 2007; Centre for advancement of Women in parliament 2007; Lane 1995; elections aCt 2007; maltadata 2007; Whyte 2007; australian parliament 2007.


0

5

10

15

20

25

1983

1981

1980

1978

1977

1975

1974

1972

1971

1969

1968

1966

1965

1963

1962

1961

1959

1958

1956

1955

1954

1953

1951

1950

1949

Perc

ent

LEGEND Senate (PR-STV)Lower House (majority)

FigUre 7.2 Comparing Women’S repreSentation in aUStraLia, 1949–1983

Source: parliamentary Library, parliament of australia.

FigUre 7.3 Comparing Women’S repreSentation: ireLanD anD CanaDa

Perc

ent

LEGEND Canada Ireland

0

5

10

15

20

25

2006

2004

2002

2000

1997

1993

1992

1988

1987

1984

1982

1981

1979

1977

1974

1973

1972

1969

1968

Source: Lovenduski et al. 1993; Bickerton and gagnon 1994; Kittleson 2006; Cool 2006.


issue concerns levels of proportionality. MMP supporters claim that STV cannot, by design, achieve full proportionality. This is because the small multi-member ridings have a built-in threshold; in a five-member riding, the threshold to win a seat would be approximately one-fifth of the total votes, or 20% (though in practice some voters express only one preference, so the threshold in this kind of riding usually ranges from 17 to 18%). This is much higher than new Zealand’s MMP threshold of just 4% for parties to gain a party-list seat. on the face of it, STV would appear to be a much less proportional voting system than MMP. The accusation is sort of true and sort of not true. A 17 to 20% threshold is clearly very high compared to other PR systems. There is no denying that for a political force whose support is sparsely but evenly spread across the country, such a system would act as a barrier to gaining representa-tion. But in practice, few political players are so completely dispersed. Usually, any political operative has some geographic base, meaning that they can gain election under STV despite what appears at first glance to be an insurmountable threshold.

A case in point would be the irish Green party. With national voting support in the range of just 1.2 to 3.8% over the past two decades, the party would seem an unlikely candidate to gain election via the STV form of PR. But given their regional concentra-tion of support in a few cities, they managed to gain representation in the 1989, 1992, 1997, and 2002 parliamentary elections. While the Green party in ireland would be unlikely to gain election under a plurality system, given multi-member ridings and a proportional voting formula under STV they could gather enough support to get elected, despite a formal threshold of about 17 to 18%. And they are not an exception. ireland’s traditional third party, labour, has regularly converted an average of 10% of the national vote into roughly 10% of the seats in the legislature. So too have the centre-right Progressive Democrats regularly turned 5% of the national vote into 5% of the parliamentary seats.

Again, in practice, election results in ireland have mostly proven fairly propor-tional for all parties, large and small, with a slight bias toward larger parties and against smaller parties that exists in all PR systems. The only exceptions have been those periods where politicians reduced the average riding size to three members. However, when they were raised back to five members, proportionality returned (Gal-lagher 1986). nor is this kind of manipulation restricted to STV. An MMP system can be similarly manipulated if the number of list seats is reduced, if the party-list alloca-tion is broken up into small regions, or if the list and district seat allocations are not compensatory in nature. And, in the end, if people are concerned about the threshold in STV, there is no reason that some kind of compensatory feature could not be added to open the system to non-geographically concentrated political opinion (Pilon 2001; also discussed in Gallagher 1986).

ConclusionThe debate over the “perfect” PR system wastes a lot of time and valuable energy pitting people who should be allies against each other. The whole enterprise is misdirected and poorly informed. Much of it is simply bad analysis of voting systems and their possible effects. Participants tend to ignore how all voting systems can be characterized


as complicated and how all systems can be manipulated to produce suboptimal results. These partisans are often guilty of a deterministic reading of how voting systems work, one that overstates the impact and effect of mechanical details while underestimating the dynamic interplay of structural arrangements with social forces (e.g., women’s representation). This leads to overstatement about voting-system effects on parties, local and diverse representation, and individual politicians. The view here is one of a technical problem, to be ameliorated by the application of the correct instrument. The wrong instrument then must lead to failure—this is why the PR partisans are so dogged about their pet system.

All of this misplaces the proper emphasis in voting-system debates, which involves the broad representational differences between plurality and PR—the key factor that crucially affects the level of political competition and the nature of government forma-tion in each system. Get this division right and all the other problems can be addressed, either through political action (now freed up considerably with more effective com-petition and an end to phony majority government) or through further reform. And this last point bears repeating: there is nothing preventing further reform from taking place after a PR system is introduced. in fact, the rules of many PR systems have been tweaked or altered after the initial adoption of the system, most of them in a more proportional direction (lijphart 1994: 52–6). But it is bad strategy to divide the PR forces on such questions before plurality has been displaced. Reform is much harder within plurality than within a PR system. And the BC referendum on STV demon-strates that divisions within the PR camp over which system is best only work to the benefit of those seeking to keep plurality.

134

CHAPTER EiGHT

Debating Voting-System Effects

introductionA lot of things are claimed about voting systems, most of them debatable. The debates involve why countries use the systems they do, how the systems work, what kinds of results they typically produce, and what might happen to a country—say Canada—if the voting system were changed. While most of these issues have already been ad-dressed in this book in one form or another, it is useful to bring them together in one chapter to compare their contrasting logic and evidence, particularly as these comprise the “greatest hits” of the voting-system debates. The debates tend to take two forms, one focused on values and the other focused on performance. Academics have fostered the values discussion (characterized as a choice between adversarial or consensual systems, or majoritarian or proportional models), while a much broader group—in-cluding academics, politicians, pundits, activists, citizens, etc.—have commented on voting-system performance in various ways. This chapter will review the classic debate points about voting systems, particularly questions of stability, local representation, and government accountability, to critically assess their validity. it will also take up some of the newer issues attached to voting-system reform campaigns, such as their poten-tial impact on voter turnout and styles of political engagement, in an effort to separate defendable from indefensible positions.

The Debate About Valuesin their 2002 discussion paper, the law Commission of Canada stated that “[c]hoosing a voting system is about choosing between different values” (lCC 2002: 29). it has become commonplace in most political science texts and public processes to begin a discussion of voting systems with a consideration of values. Typically, we are encouraged

d e b at i n g vo t i n g - s y s t e m e f f e c t s / 135

to reflect on the values that we think should underlie a democratic voting system, and helpful prompts are usually provided in the form of values we should consider—values like the quality of representation, government performance and accountability, local representation, and so on (lCC 2002; BC-CA 2004; Watling 2006).

Most political scientists go further and claim that the kind of voting system in use in a given country is itself a reflection of larger social and political values, the implica-tion being that the system was adopted to facilitate the expression of such values (Katz 1997; lijphart 1999a). Specifically, they claim that SMP countries favour or embody an adversarial political culture while PR-using countries reflect a more consensual approach to politics. in the values paradigm, the debate over voting systems is both a discussion about what we value and would like to see in a representative system and an assessment of what we think the larger cultural values are that animate politics in our particular locale. Thus reform involves a compromise between individual and collective values vis-à-vis politics. All this sounds very balanced, but the privileging of values as the key determinant in voting-system choice is usually asserted rather than defended. The fact that we might want an approach to voting-system reform that is driven by values does not mean that values have driven or will drive the process of changing actual voting systems. if values are not the root causes of change, we may be misplacing our efforts and wasting our time.

The currently dominant “values” approach to examining and appraising voting systems is deeply problematic for at least three reasons. First, when political scientists and public processes begin with values, they may simply reproduce a dominant and largely unquestioned view in north American society that all political institutions and political results have their origins in public values and choices. This has a host of im-plications, not the least of which is foreclosing important lines of research that may prove crucial to the process of voting-system reform. if all political results are the un-problematic product of public wants, then there is no need to examine just how various arrangements have come about or why various efforts at change have succeeded or failed. Such ready-made, all-purpose answers prevent the public and experts alike from applying insights from past experience to the present.

This relates to the second problem with the values approach, which is that it is fac-tually incorrect, both historically and in contemporary settings. Historical work on voting-system reform demonstrates that political institutions in modern societies were influenced by political interests, not values, and formed largely without input from citizens. in the case of Britain, and by extension in the cases of the Anglo-American colonies it influenced, the voting system was not so much the product of a grand design as the result of an ongoing process of pragmatic tinkering and accommodation to shifting political interests (Hart 1992; Pilon 2005). in the case of western Europe, the countries adopting PR could hardly be characterized as being driven by “consensual” values in designing their voting systems. in fact, most were riven with such serious political divisions that potential civil war and revolution were actually the key impe-tuses to voting-system reform. in Germany, Switzerland, Sweden, and Belgium (among others), the end of World War i was accompanied by street demonstrations and an increase in radical political demands (Pilon 2002, 2005). So there is little support for sweeping claims that adversarial versus consensual social values somehow led to the


adoption of different voting systems in the past. in fact, the causal arrows might run the other way. if such societies appear either adversarial or consensual today, that may have less to do with values than with the long-term effects of the voting system push-ing politics toward either conflict or compromise. Meanwhile, contemporary survey work demonstrates that voting systems have remained largely unquestioned not be-cause they enjoy support, but because most people remain unaware of them or their workings. indeed, analysis of voters’ knowledge of Canada’s plurality system has dem-onstrated that many do not grasp the basic workings and implications of the system (Bricker and Redfern 2001). Voters can hardly be credited with authoring a system that most do not understand.

Finally, the values approach limits critical inquiry into key tenets of the voting-system debate by converting analytical questions into value considerations (“do we like local representation?” rather than “is local representation demonstrably important to the political system?”) and by flattening all values relativistically (thus “diverse represen-tation” and “majority government” are equally valid “values” in choosing a democratic voting system). A more critical approach would argue that an analysis of the real workings of our institutions and their political implications should precede any consideration of the values we think we may want in a voting system. Furthermore, a critical stance would argue that not all values are equal when it comes to evaluating the implications of different voting systems; democratic values should be given priority. in fact, some “values”—like getting a majority government from a minority of the votes—may not be normatively defensible at all by democratic standards.

The currently fashionable values approach is a barrier to change. it obscures the political contestation that has been central to historical and contemporary episodes of voting-system change while misdirecting public and elite attention toward factors with no demonstrable influence on the process. Pragmatically, such an approach will contribute to the failure of reform initiatives and a disillusionment of the participants, particularly the public.

The Debate About PerformanceThe bulk of detailed debate over voting systems typically concerns their alleged perfor-mance. Great claims are made about the effects that will flow from one model, while dire assessments are offered about another. We have reviewed the deficiencies of the plurality system and the positive aspects of PR in previous chapters. now let us reverse the approach, critically assessing the alleged positive aspects of plurality and the com-mon complaints made against PR systems. in doing so, we will be responding to the discussion as it is commonly framed in Canada. in a nutshell, the plurality system—while perhaps less representative—is defended for its stability, for its tendency to cre-ate majority governments, for linking voters to a particular representative, and for allegedly giving voters the final say on government formation. PR, by contrast, is said to create political instability, constant minority government, and a lack of accountability, both in terms of a local representative and government formation. in this section, we’ll explore these competing claims about plurality and PR.


Plurality’s Alleged StrengthsThere are four key pillars in the defence of plurality voting: simplicity, a tendency to produce stable majority government, accountability of a local member to a geographic constituency, and the ability of voters to hold the government to account and potentially “throw the rascals out.” These four factors are typically highlighted in public debates over voting systems as compelling and convincing arguments in favour of plurality voting. let’s take each in turn.

SiMPliCiTY

Plurality has long been defended for its simplicity and transparency for voters and election officers alike. Voters need only examine the ballot, find the name they wish to support, and mark an “X” in the space provided. At the conclusion of voting day, election officers need only empty the ballot box, divide the ballots by the markings made upon them, and declare the most popular candidate the winner. Such voting instructions are easy for voting officials to communicate and fairly readily understood by voters, regardless of their level of education or political sophistication. Adding up the results with such a system can also be quickly mastered and executed, even by novice electoral officers. This is generally what is meant by the “simplicity” of the plurality voting system and, under-stood in these terms, there can be little doubt that the claims are true. Barring any good reasons to count votes another way, simplicity in voting process and administration seems like a good thing. Furthermore, proponents argue that such simplicity is key to voter confidence in the system as the administration of the vote is very transparent—basically anyone can follow the logic of the ballot markings and the ballot-counting process.

But there are two flaws in the “simplicity” defences of plurality voting. on the one hand, the characterization of plurality’s simplicity is one-sided, attending only to the acts involved with voter administration. The simplicity of a voting system and the trans-parency of its results should not be limited to voting and ballot counting but should extend to the results produced by the system as a whole. And here, plurality is anything but simple and transparent. in fact, nearly half of voters do not understand that a vic-torious “majority” government seldom represents an actual majority of voters (Bricker and Redfern 2001). Few voters can explain why there is nearly always a gap between the percentage of votes cast for parties and the percentage of seats they win. Thus in terms of the election results, rather than in terms of the election administration, plu-rality is one of the least transparent systems.

on the other hand, even if one values simplicity in voting and vote counting, to what extent should that commitment be adhered to in the face of competing and perhaps more compelling needs and values? To suggest the need for simplicity must trump everything else is not a compelling argument for a number of reasons. First, there is little support for the view that voters cannot handle more complex voting systems. in fact, most voters in western countries already do (Farrell 2001: 202–3). Second, as the voting results are a kind of political communication, the simplicity and transparency of those results is arguably much more important than having simple vote-counting methods. Finally, few voters attend to the workings of the plurality


system, despite its simplicity, so it is false to claim that this simplicity is key to its le-gitimacy and to the accepted legitimacy of its results. in practice, the public’s sense of the legitimacy of our institutions has more to do with the behaviour of political actors and the media. if political elites are concerned about the system and its results, that serves as a cue for public concerns as well. Thus it follows that how political elites re-spond to any alternative method of voting will also prove more important than whether voters can follow all aspects of the vote count.

in the end, the simplicity arguments are often disingenuous. our political elites have proven only too willing to embrace complexity when it has served their interests, whether we are discussing voting systems or a host of other topics. As political scientist Jean-Pierre Derriennic has noted, the same politicians who claim to be able to explain the complexity of free-trade agreements to voters nonetheless throw up their hands at the prospect of working with alternative voting systems—despite the fact the poten-tial results from the latter are far less uncertain than those of the former (Derriennic 2005). And any serious political operative understands that it is the role of organized political forces—be they parties or interest groups—to signal to the public that they should be concerned about a particular complex political institution from time to time. That leaves appeals to simplicity operating on the level of fear and ignorance, as elites attempt to exploit what people don’t know in order to keep in place what seems fam-iliar, for reasons other than those publicly stated.

STABiliTY AnD MAJoRiTY GoVERnMEnT

if the simplicity arguments amount to an insincere appeal from people who know better, a great deal more effort goes into the defence of plurality as the voting system that best ensures stability and the likelihood of majority government. Basically, those who argue for plurality on the basis of stability and majority government believe that stability in governing is crucial to the well-being of a country, and that the best way to secure that is through the election of a single-party majority government. Though the “majority” is often based on less than a majority of the popular vote, proponents nonetheless argue that this is an acceptable trade-off given the benefits that ultimately accrue to the governed.

in this scenario, a party can develop a coherent program and, if they win a majority of seats, can be assured of having the legislative ability to see the program introduced in its entirety. Such a government can also focus more on the long-term interests of the country because their legislative majority allows them to introduce changes that, while temporarily unpopular, might prove to be in the nation’s interests at some time in the future. By implication, the lack of single-party majority government would mean a lack of coherence in policy, confusion about what the government is doing or can accomplish, a watering down of their agenda, and an inability to administer strong and unpopular medicine to the people for their own good. in the end, proponents ar-gue that voters themselves approve of such majority government arguments because, according to the polls, Canadians express a preference for majority government over minority government. The need for “majority government” is probably the most com-mon defence of plurality.


if we turn to the evidence, there is no doubt that plurality does tend to result in legislative majority governments where it is used. The United Kingdom produced 21 majority governments from 26 elections over the course of the 20th century. in Canada, too, plurality has produced many majority governments, though with less frequency than plurality’s defenders might admit. From the breakdown of Canada’s traditional 19th-century two-party system in 1921 until 2006, 10 out of 26 elections have resulted in a minority government, despite constant adherence to plurality voting. interest-ingly, this debate is irrelevant in one of the last major countries still using plurality, the United States, because their Congressional system does not utilize the parliamen-tary notion of responsible government. in fact, the behaviour of the US Congress often looks to be the opposite of the claims made for plurality regimes; in that system, there is no guarantee that either the executive or legislative branch can ensure passage of legislation, even with a nominal majority in the different chambers.

of course, the fact that plurality voting tends to produce legislative majority gov-ernments does not necessarily prove that such governments are in fact more stable than minority governments or coalition governments. A lot depends on how we define stability. Some would argue that the policy swings typical under plurality—that Party X comes to power and removes Party Y’s policies only to see the opposite occur when Party Y returns to office—are highly unstable. nor are the alleged benefits from plurali-ty’s “stability” well documented. A clear example of this can be seen in the 1990s at the provincial level in ontario and BC. ontario lurched from liberal to nDP to Conser-vative government in just three elections, with wide swings on policy accompanying each change in office. in BC, the old-right Socreds were displaced by the left nDP, who were then defeated by the new-right liberals. Any gains produced by the dramatic shifts in policy under each “majority” government seem dubious and difficult to quantify. on the other hand, many of the alleged problems with minority government under plurality have less to do with the fact that the governments are minorities than with the incentive structure of the voting system itself. Because plurality exaggerates the support of the biggest vote-getter, it creates an incentive for a minority government to go back to the polls (Dobell 2000). Minority governments in PR systems do not face the same incentives and as such do not “destabilize” as readily as under plurality.

The biggest problem with the “majority government” arguments that are utilized to defend plurality voting is that, unwittingly or otherwise, they rest on fundamentally undemocratic values. For instance, the obsession with stability—despite all evidence to the contrary about the stability of coalition and minority governments in western societies—suggests either a lack of faith in democratic deliberation or an unwilling-ness to abide by genuine majority decision making. no one wishes for “unstable” government, but in a democracy instability is factored into the design because people are supposed to be allowed to disagree. To sacrifice all other claims of democratic le-gitimacy to the need for stability is not healthy for a democratic society. There are other models of government that make stability the overarching principle (dictator-ship, fascism, etc.), but it is not democracy and nor should it be.

The other defences of majority government are just variations on this undemo-cratic riff, whether we are talking about the paternalism of the view that says that governments must be empowered to act against the wishes of their populations “for


their own good” or about the anti-majoritarism behind the view that popular minorities who get a majority of seats via the voting system should be able to introduce policies that a majority of people don’t want. Such views violate one of the most basic assump-tions of democratic theory—that is, that majority rule is the defining characteristic of the decision rule in democracy. The most telling problem with the majority govern-ment argument is that there has never been anything stopping a majority of people from voting for the same party, either now or in the past. The fact that they seldom do is actually a crucial piece of information in a democratic society, one that we shouldn’t allow our so-called democratic institutions to obscure.

The attempt to shore up support for plurality’s inflated majority governments by suggesting that the public prefers majority to minority government is also suspect on at least two counts. First, it is not clear that people are saying they approve of such re-sults with the answers they provide in opinion surveys. in some cases, the wording of the questions in the polling that bolsters this view is unclear or unclear in its implica-tions. For instance, recall that most voters are confused about the results produced by plurality voting, with nearly 50% mistaking legislative majorities for popular-vote majorities (Bricker and Redfern 2001). For such voters, the question “do you support majority government” is absurd because, in their view, our majority governments do result from majority support. The obverse implications of such a question would be “do you think that a majority vote should result in a minority government?”—clearly an absurd outcome.

on the other hand, the question—even if properly understood—is stated unclearly because it focuses on wants or desires rather than on anticipated results. i may like the idea of majority government—for my party—but such a view is divorced from the fact of whether it is likely that my party will gain a majority. it is telling that when the ques-tion is rephrased away from desires and toward outcomes, the results of polling are very different. For instance, when asked whether the voting system should award seats to parties in proportion to the votes they receive—which is just another way of de-scribing PR—64% of respondents said “yes” (Bricker and Redfern 2001). This would appear to counter glib attempts to marshal public opinion behind phony majority governments. of course, even if a majority of voters thought that distorted election outcomes were oK, that would not necessarily be a strong argument in favour of such governments. While most democratic theory accepts majority rule as a way to make decisions about specific policies, process, and representation issues cannot be sub-jected to the same decision-making approach. if they were, 51% of the voters could decide to disenfranchise the other 49%—hardly a democratic outcome.

loCAl REPRESEnTATion

Aside from majority government, another common defence of plurality voting is that it maintains a strong local link in terms of representation. Many participants in the voting-system debate, both pro and con, accept the view that local representation is a key part of Canadian politics and is much valued by individual Canadian voters. indeed, one reformer claimed that any new voting system would need to maintain single-member ridings because Canadians “would be unlikely to accept an electoral system


that deprived them of having a single MP to represent them” (Milner 1999: 41). The local representative is defended as a crucial link between voters and the political sys-tem, one that allows individuals to seek help and the redress of grievances that involve government while also allowing voters to hold someone—in this case, the individual local politician—accountable for what is going on in the provincial or federal capital. There are a host of academics and politicians who underline these two key local func-tions, arguing that the local link serves to provide incentives for politicians to maintain a presence in the communities they claim to represent. Furthermore, they worry that any change to the voting system that would eliminate this local factor, or even just di-minish it, would endanger the quality of Canadian democracy (assessed critically in Pearse 2005).

The broad acceptance of local representation as a key component of Canadian democracy is surprising, as much of the current “debate” represents a triumph of assertion over argument. Proponents of local representation insist it is crucial to Canadian politics but seldom bother to mount a serious case to defend it. Even where serious academic work does focus on the local, it tends to highlight local effects in po-litical campaigning (i.e., spending money or organizing locally) rather than argue that locality and specifically local concerns are the basis upon which voters make their choice on election day (Carty and Eagles 2005). When we turn to the evidence, the case for the importance of the local member in our political system quickly crumbles.

Whether we are talking about representation or accountability in our political sys-tem, the local realm doesn’t really matter that much. The representation that a local member can give voters amounts to a rather mediocre ombuds-service—help with getting a passport or getting the attention of a government bureaucrat. on the other hand, if local voters want to talk policy they are going to run into a problem unless their MP’s political views line up with theirs. This means that many local people—usually between 40 and 60% of them—can forget having any influence over the legis-lative voting decisions of their local politician. nor is the alleged accountability of local representation all that substantive. The ideal would have it that voters elect a member and then re-elect or defeat that member based on their assessment of that member’s behaviour. But whether a member gets elected or re-elected is more crucially depen-dent on the competitive party situation in her riding than on shifting voter opinion. For instance, a candidate in a competitive three-cornered contest might win with 35% of the vote. in the next election, that candidate may get more support but still lose if support for the third party falls. in this scenario, the losing candidate’s loss may repre-sent less the voters’ verdict about her individual performance than the arbitrary results of the voting system’s workings. More generally, Canada has a particularly high turnover in terms of our elected members, which academics credit to swings in party support rather than voter judgments of local politicians (Docherty 1997: 51–7).

Why do so many uncritically accept the importance of the local in Canadian politics? Peter Mair argues that the seemingly unquestioned support for the position amounts to an “ideology of local representation,” one that results from the complacency of aca-demics who have passively accepted its allegedly functional importance (one that they fail, however, to investigate) and from its vigorous defence by self-interested politicians (Mair 1992). Yet even the most cursory investigation of the evidence reveals that the


local has proven unimportant to both Canadian politicians and voters. Politicians can hardly be credited with a serious interest in the local. For most of Canada’s history, multi-member ridings were used for a number of provincial and federal elections, and they continue to be used for municipal contests in some locales. The move to single-member ridings was motivated by competitive political party pressures and key court rulings, not some desire to connect local voters to politics.

nor has the alleged concern for local representation prevented politicians from constantly increasing the size of local ridings in nearly all provincial and federal juris-dictions. For instance, today’s federal ridings contain, on average, five times as many people as they did in 1867 and twice as many as in 1953 (Pilon 2001). And these trends are similar for provincial jurisdictions as well (Pilon and Docherty 2006). Even the much-vaunted ombuds-role is hardly traditional; it wasn’t until the 1970s that politi-cians gave themselves the resources to act as a kind of local representative by funding the opening of local constituency offices, allocating permanent offices in provincial and federal legislatures, and aiding travel costs so that members could regularly visit their local areas (Docherty 1997). The romantic notions of the local member collide with a history of decisions by politicians that make plain just how unimportant local representation has been to them in the grand scheme of Canadian politics.

Assessing Canadian voters’ views about local representation is a bit more compli-cated. Because of the “ideology of local representation,” it is not surprising to hear voters claim that they think having a local member is good or that they would like their member to listen to local voters more (Anderson and Goodyear-Grant 2005). But if we are to understand to what degree voters make voting decisions on the basis of local representation, we must be more critical of these claims. There is a host of other evidence that challenges these survey results. For instance, countless academic studies have underlined that voters use party as the key factor in making their decision about which candidate to support. Some studies have supported this conclusion by demonstrating that there is little bonus for representatives who build up a local profile compared with those who do not; the overarching factor affecting their re-election

Popu

lation

ave

rage

, in

tho

usan

ds

0

10

20

30

40

50

60

70

80

90

100

199719741953192519001867

FigUre 8.1 inCreaSeS in aVerage CanaDian FeDeraL riDing SiZeS, 1867–1997

Source: Compiled from elections Canada historical data.


will be the more general swing for or against their party (Crewe 1985). There is some debate in Canada about the magnitude of “personal voting,” or the bonus that an in-dividual MP can gain by his or her personal profile. However, the amount is small, and linked less to local factors than to cabinet position and national media attention (irvine 1982; Docherty 1997, 2005).

other studies have asked voters to rank the factors that influence their voting deci-sion, and party comes out on top. one recent study did find that 40% of voters claimed that the individual candidate did factor into their voting decision. However, the much more crucial finding was that only 4% of respondents were prepared to rank the individual candidate higher in their vote choice if it meant deviating from their pre-ferred party choice. in other words, 96% of voters are making their choice primarily on the basis of party identification (Blais et al. 2003). As Marsh notes in the context of irish voter surveys: “it seems likely that although many voters may vote a party ticket they will rationalise this to themselves in terms of candidate qualities” (Marsh 2007: 22). Another way we can assess the importance of local representation to voters is to get a sense of what people know about local members and gauge how many ac-tually seek their help. in both cases, the figures are unimpressive. People can more readily identify parties and party leaders than regular politicians (with the exception of rural areas, where the name of a local politician is typically better known). And fewer than one in five survey respondents report contacting their local member (Gidengil et al. 2004; norris 2004: 240–3).

Probably the most damning evidence against the “local representation” argument is simply that the voting results themselves demonstrate that people do not vote on the basis of locality. if the local were important, we might expect to see politicians elected on the basis of local identities rather than party identifications. on the other hand, if the policy differences between different parties were more important to voters, we might expect local votes to be divided among parties, and the winning local member to represent one of them. of course, we know that the latter is the case in Canada at the provincial and federal level. in each locale, voters divide their support among the different parties—there is no homogenous local interest to be represented. The debate over local representation is important because those who make claims for it use it to defend our current plurality system or limit the range of options we may consider for reform. now, if we are going to reform our voting system and some people like the idea of local representation, then such advocates should be free to make their case. But as with so many other aspects of the voting-system-reform debate, normative appeals should be brought into dialogue with the actual practice of Canadian politics. if trade-offs have to be made in the design of a new voting system, then they should be based more on those factors that are demonstrably important to Canadian voters—like party representation—and less on those for which the evidence is slight at best.

GoVERnMEnT ACCoUnTABiliTY

The final key argument used to defend plurality is that the system assures that the gov-ernment will remain accountable to voters because it allows them to “throw the rascals out.” Basically, this argument holds that under plurality it is the voters who are ulti-mately in control of the formation of government and, by implication, the defeat of


government. This is contrasted to the alleged experience of PR systems, where the voters elect parties but it is negotiations between the parties that decide who will form the government. Critics claim that this process can lead to situations where a significant group of former government supporters change their vote but the governing parties remain in place regardless by negotiating support from new coalition members. This would be less likely in plurality precisely because of the system’s tendency to exagger-ate the impact of vote swings and overrepresent the most popular parties. For instance, if enough voters from Party A switched to Party B in a plurality election, then Party A would probably lose its inflated majority of seats while Party B would now enjoy an inflated majority. Because the shift in votes would lead to defeat for the government and the election of an alternative government, proponents argue that this means the voters are in control and that they are making government accountable to them.

The “government accountability” defence of plurality is the weakest of the four pillars. it rests largely on an idealized notion of party competition where two parties get alternating chances to offer the same things to voters. But this ignores both the fact that different parties do not tend to offer the same things to the public and the fact that the voters themselves are not united about what political results they want. Thus it is not clear how and to whom the mechanism is effecting accountability. For account-ability to make sense in political terms, there has to be some relationship between what people want, what government does, and how people respond to government action or inaction, as the case may be.

But who is gaining the accountability when a government changes under plurality? Recall that most plurality governments are elected by a minority of voters. logically, it makes no sense to argue that the change represents accountability for those who sup-ported the government, as the new government probably owes its election to a different minority of voters. And the new government’s minority of voters have not enjoyed any accountability from the change either—they did not support the previous government in first place. it is possible that there are some voters in the middle of the spectrum who switch back and forth between the major parties and that they could be said to enjoy the accountability of the change, though such voters would appear to have such broad political tastes that it becomes too difficult to judge what “accountability” amounts to in practical terms (if you are prepared to vote left and right, on what basis do you make distinctions about government performance?).

The fundamental problem with the “government accountability” defence is that it ignores politics. To argue that voters can gain accountability from government by replacing “party left” with “party right” is to ignore how real people in real political situations understand the notion of accountability. it is obviously ridiculous to argue that left-wing voters will be making “their” government accountable to them by re-placing it with a right-wing alternative. of course, leaving aside political divisions, the logic of the government accountability argument is simply not convincing. We are supposed to believe that accountability is satisfied under plurality because a small shift in votes can take a majority government away from a minority of voters and give it to a different minority of voters. in practice, even this distorted logic may not be what really happens, as governments can fall or rise based on changes in party competition rather than changes in support for the government.


oTHER DEFEnCES oF PlURAliTY

The four major arguments we’ve reviewed do not exhaust the possible defences of our current system. There are even weaker arguments offered to defend the continued use of plurality in Canada. For instance, there are many essentially pragmatic arguments that suggest that the fact that Canada is a relatively successful country means that the “plurality system has served Canada well” and that, with such a track record of success, we shouldn’t consider changing it (Toronto Star Editorial Board 2007). Setting aside the fact that such arguments could easily be used to justify authoritarian rule—and indeed were used to justify fascist governments in italy, Spain, Portugal, and a host of authoritarian regimes in latin America—the success of Canada is very much in the eye of the beholder. We are and have always been a politically divided country with strong disagreements about what success means, how to measure it, and how to spot it when it rides into town. The “success” argument is really a non-argument, as it avoids engaging the substantive issues raised by both proponents and critics of the status quo. it is designed to try to dismiss the discussion altogether.

Another weak argument claims that even if we allow that our current system is not ideal, we do not know what change might bring and as such we should do nothing. John Courtney argues that Canada has a host of problems that might be exacerbated by adopting PR. He claims that, as we don’t know what might result from change, we should just stick with the status quo (Courtney 2004). The problem with the “we don’t know about change” argument is that it ignores the fact that we don’t know about non-change, either. in other words, it is possible that the results of our current system could change and lead to disastrous results in the future, because the future is unknow-able with or without change. of course, some might argue that our current system has already led to some pretty terrifying results—in particular, the overrepresentation of a number of regional interests that threaten the unity of the country.

The traditional defences of the plurality system in Canada—simplicity, majority gov-ernment, local representation, and government accountability—prove to be less than compelling when some time is spent examining their assumptions and the evidence that might be brought to bear on the case. in each example, we have shown the claims to be variously one-sided (simplicity), undemocratic (majority government), unsustainable (local representation), or simply incoherent and illogical (government accountability).

PR’s Alleged WeaknessesHardly a newspaper story is written that refers to PR in Canada that does not make reference to at least two of the four standard complaints offered by the critics of the system. Thus we tend to hear about Germany’s “complex proportional voting scheme” or the chronic instability associated with PR in italy or israel. or reporters will refer to the overweening power of the political parties in PR systems or the indecisiveness of the election results. Here, we will critically review the allegations that PR is too complex and opaque in its results, that it fosters instability and minority government, that it engen-ders party dynamics that are too stable and party dominant, and that it leads to a lack of accountability at both the local level and in terms of government formation.


CoMPlEXiTY

As with claims that plurality voting is simple, arguments suggesting that PR is complex are true in some ways but equally untrue in others. The point is that the complexity debate as regards PR is often one-sided. if we are referring to the process of balloting and vote counting, there is little doubt that PR systems are more complex than the simple methods of plurality. The complexity of the vote-marking procedures can range from something as simple as plurality—just marking an “X” next to a party name—to more involved markings requiring preferences to be made both among parties and among the candidates within or across parties. Vote counting under PR systems is undeniably complex, as it typically involves aggregating votes across a larger constituency than in plurality and then allocating representation on the basis of the achievement of quotas of the total votes rather than just a plurality of the votes.

But the complexity of a voting system must not be assessed simply on the basis of ballot marking and vote counting. The system’s complexity must also be assessed on the basis of how clear the results of the election appear to be, given how people have voted in the election. And here, PR systems are much clearer in the results they pro-duce than plurality voting systems. Whereas the relationship between votes and results under plurality is often distorted and the reasons for such distortions are opaque and hard to explain, the results in a PR election typically make intuitive sense to voters because the percentage of votes for each party tends to closely match the percentage of seats won by each party; that 20% of the votes will produce 20% of the seats in a PR system (give or take a few percent) is a much more clear and transparent result than 40% of the votes turning into 60% of the seats, as can occur under plurality.

in a sense, the simplicity argument offers a trade-off. People must decide what is more important—a simple way of adding up the votes but distorted and unclear elec-tion results, or a more complex way of adding up the votes that produces a fairly straightforward election result. in this, the choice should be clear: transparent election results should not be sacrificed to simple vote-counting methods just because the methods are simple. of course, there is something disingenuous about the simplicity defences of plurality. There are many aspects of our electoral system that are complex and largely beyond the general working knowledge of most voters, including how districts are allocated, how party funding is organized, and how voter registration is administered. But plurality’s defenders do not insist that simplicity should rule in all these areas. nor do they seem to have a problem with complexity in all other aspects of our modern society. The demand for “simplicity” in vote counting appears to be just special pleading, not a matter of principle or performance that might require us to sacrifice other important goals for our voting system. Thus it hardly amounts to a compelling reason to reject alternative voting systems with other attractive qualities.

inSTABiliTY

if there is one criticism of PR that towers over all others—appearing in nearly all critical accounts of the system—it is that proportional voting systems lead to instability, either of the government or of politics more generally. Specifically, discussion on this point has focused an inordinate amount of attention on the experiences of just two countries: italy


and israel. PR as used in both is blamed for creating political instability, unresponsive legislatures, and overpowerful parties, particularly minor ones. We typically hear how italy has suffered with “50 governments in 50 years” or that israeli governments are held hostage to tiny extremist parties that exercise influence beyond their support in the community. The general impression is that little can be accomplished in such legislative settings, leading to public frustration with government. The “two i’s” critique of PR remains a common reference in contemporary debates about voting-system reform in Canada and as such warrants a detailed examination. Here, we will critically assess the arguments about italy and israel’s experiences with PR, focusing particular attention on the logic of the arguments and the methods of comparison before moving on to a more general discussion of PR and instability in other countries using the system.

❍ Just How Unstable Are italy and israel?

our first line of inquiry in assessing PR use in italy and israel must be to establish the veracity of the claims that these countries have been unstable. There are a number of ways we can convert these broad claims into readily testable propositions. To assess their stability or instability, we can look at the length of term in office for a govern-ment, the frequency of elections, and changes in the structure of the party system. let’s take each in turn.

one standard test of stability/instability involves measuring the length of govern-ment, or how long a particular administration can stay in office. As noted earlier, the average length of government in a PR system is about 1.8 years, compared with 2.5 years for plurality (Woldendorp et al. 2000: 79). The gap between the two systems is less than a year—hardly a remarkable difference. But here, italian experience appears to depart from other European PR-using countries. Scholars report that between 1948 and 1994 italy had 48 different governments, with an average length of office of just ten months (Seton-Watson 1983; Furlong 1994). However, what is referred to as a “change of government” in the italian setting is actually more akin to a cabinet shuffle here in Canada. if we examine how long key ministers held their positions without break, we get a better idea of the length of government. For instance, prime ministers in italy in this period held office for an average of 27 months, or two-and-a-quarter years. An-other key post, the minister of the interior, averaged 29 months without change. in fact, most ministers holding office between elections retained their positions through these successive “governments.”

Another way to measure “instability” is by the frequency of elections. The more elections called (beyond a regular cycle of three or four years), suggests intractable political problems, perhaps just the instability that PR critics complain about. But between 1948 and 1994 italy and israel had only 12 and 13 elections, respectively, or one about every four years. By contrast, Britain and Canada had 13 and 15 elections in the same period. At the electoral level, then, it appears the neither italy nor israel had call to go to the polls any more than Anglo-American democracies.

instability can also be gauged by examining the structure of the party system over time. Dramatic changes in party strength may signal serious political instability. But curiously, both italy and israel have been marked by long periods where one party dominated the governing process, even though these ruling parties often lacked outright


majorities. For instance, despite the fact that italy saw more changes in government than Canada in the postwar period, it should be remembered that one party in italy—the Christian Democrats (DC)—named all the prime ministers from 1944 to 1981 and remained the pivotal and largest party in both government and Parliament until 1992 (Furlong 1994). israel has seen three different dominant coalitions in its history, with the first lasting from 1949 to 1977, the second from 1977 to 1984, and the last from 1984 to 1990. in all three periods, the prime minister and all senior ministers came from the largest parties; minor parties were able to exact little by way of concessions in terms of cabinet posts. only in the 1990s has this changed somewhat with the rise of more orthodox religious parties (whereas before religious parties were less ortho-dox) and the decline of the labour party (Diskin and Diskin 1995; Rahat and Hazan 2005: 338–40).

Most of the criticisms directed at PR involve claims about its influence on parties and party behaviour. PR is alleged to fuel the creation of many new parties, to lack in-centives to make legislators respond to constituents, and to allow small parties to dominate the political arena. Unfortunately for the critics, the historical practice with PR in italy and israel does not support their claims. Both countries were divided po-litically at the time they chose PR, and subsequent elections merely reflected these divisions. However, over the postwar period the number of parties remained relatively constant in both countries. From 1950 to 1990 there was little change in either party system (Diskin and Diskin 1995; Furlong 1994).

The complaints about unresponsive MPs in PR systems may have some merit, but the question of whether MPs anywhere else are terribly responsive must be raised. in other words, it may not be PR that is to blame, but other factors that are common in non-PR systems as well—for instance, the increasing power of media or the reliance of political parties on corporate or public funding. Criticisms that focus on how PR cre-ates MPs that are unaccountable to voters often romanticize how accountable MPs are under SMP rules. Additionally, in the italian case, one factor that contributed to a weakened party discipline much more than the use of PR was the practice of secret balloting by MPs in the House, which effectively prevented party leaders from enforc-ing party discipline (Bull and Rhodes 1997).

Finally, the question of small-party influence—specifically, the idea that these parties exercise an unfair amount of influence—is also not supported by the historical record. Certainly, negotiations had to be conducted for major parties to get minor-party sup-port, but there were always alternatives. When the italian DC grew tired of the demands of the coalition partners to the right in the early 1960s, they struck a new deal with the socialist party to their left—an arrangement that held for over a decade (Furlong 1994). By the same token, when both of israel’s major parties felt that small-party demands were becoming too onerous in 1984, they formed a grand coalition between themselves and left the others out entirely (Diskin and Diskin 1995). nor were italy and israel unique in this regard; other PR-using countries, such as Germany and Austria, had similar experiences.

The “instability” arguments in the cases of italy and israel find little support when we examine factors such as the length of government rule, the number of elections, or the structure of the party system. of course, all this isn’t to say that there is nothing


unique about italy and israel—clearly, there is. The point is rather to demonstrate that the condemnation of the voting system as the author of these countries’ woes is poorly supported by evidence or effective arguments. Such an approach tends to suffer from overstatement, overgeneralization, and a lack of attention to the real political and historical contexts of the two locales. The critics seem to infer that if italy and israel had just adopted plurality instead of PR, they would not have struggled with the politi-cal issues that they have. But this ignores the historical contexts under which PR came into use in both countries. in italy, the adoption of PR reflected the extreme polariza-tion between the Catholic right and the communist left, with the former unprepared to sanction a plurality victory for their adversaries (Pilon 2005). in israel, the adoption of PR was essential in forging national unity in a country whose existence was forcibly disputed both from within and without (Diskin and Diskin 1995; Rahat and Hazan 2005). As these conditions have largely remained in force ever since, one could argue that it is the long-term existence of essentially warlike conditions in and around the country that has had much greater influence on the hyper-inclusive approach to gov-ernment than PR, as nearly all countries under threat of war tend to maximize internal unity through coalition or national-unity governments.

❍ Questions of Method

Beyond the empirical question of whether italy and israel are unstable as polities, there are larger methodological issues with using these examples to make the case against PR that must be addressed. The issues involve sample selection and the limits of comparison. For instance, even if there were some basis for the complaints levelled against the use of PR in italy and israel, that still wouldn’t explain the frequency with which these two examples appear in anti-PR arguments. As nearly all European coun-tries use some form of PR, it is not clear why only these two countries are regularly marshalled as evidence by the critics. What we have here is what political scientists would call a “sampling” problem, akin to someone trying to use the wrecking yard as the pool of cars to argue that cars in general don’t work. of course, the first thing any fair-minded person would say is that the cars in the wrecking yard do not reflect how cars generally work. if we want to draw conclusions about how cars work in general and we don’t have the time or resources to test every single kind of car that exists in the world, then we have to put together a representative sample.

The same is true if we want to test propositions about the workings of different voting systems. Thus those that invoke only italy and israel to discredit PR are guilty of rigging their samples. or, put another way, they are selectively choosing evidence to fit their conclusions. A representative sample on the question of PR and its effects would have to include more countries than just these two. of course, as soon we do that, it quickly becomes apparent that the anti-PR generalizations are even less persua-sive. ironically, unlike many of the claims made against PR, most of the generalizations about the effects of single-member plurality systems—phony majorities, misrepresen-tation, regional balkanization—do hold up across different countries.

The other methodological problem noted above highlights the limits of the com-parative approach: not everything can be compared unproblematically to everything


else. For instance, when people raise the examples of italy and israel, the implication is usually that adoption of PR in Canada would produce similar political conditions. Aside from reiterating the point made above—that PR did not in fact create the “problems” it is typically accused of creating—we must also discredit this type of comparison. Canada is a very different place both socially and politically than israel or italy. A change of voting system is not going to introduce israeli-style intractable religious and ethnic differences or automatically produce an italian-style large Com-munist party. And the point must be stressed that voting systems do not of themselves create social reality or, indeed, any specific political results. instead, voting systems differ in the relative degree of openness they may or may not possess to political com-petition. if we do want to make comparisons, we need to establish ones with countries that are somewhat similar to Canada—or seriously limit what we can draw from the process of comparison altogether. in many ways, Germany would be much more com-parable to Canada, with its federal system, its large Catholic minority, and its level of economic development; or new Zealand, with its legacy of British immigration and institutions, considerable Aboriginal population, and some degree of ethnic diversity. Tellingly, neither has been accused of italy- or israel-like problems.

❍ Broader Comparisons of PR and Stability

if we turn to a more representative sample of PR-using countries—specifically, those that would make for an effective comparison with Canadian circumstances, such as western European countries—the accusations connecting PR with instability gain little support. PR governments tend to last nearly as long as plurality governments: 1.8 years on average for PR as compared with 2.5 years for plurality (Woldendorp et al. 2000: 79). PR-using countries have not had recourse to return to the polls with more frequency than plurality-using countries; between 1945 and 1998 PR countries averaged 16 elec-tions, compared with an average of 16.7 elections for plurality countries (iDEA 2007). And western PR countries have not been unstable in terms of their economic develop-ment. in fact, in terms of economic growth and quality-of-life measures, western PR governments have reported consistently high performance ratings (lijphart 1999a: 263–70, 1999b: 320–3).

The ritual invocation of italy and israel by those opposed to PR—and their claims that PR has contributed to political instability, unresponsive legislatures, and over-powerful parties—has little empirical backing. Upon examination, the arguments critical of the experience of PR in these countries are weak and lack a proper compara-tive context. in fact, the whole approach to singling out italy and israel while ignoring the experiences of other countries using PR casts doubt on the integrity of such research projects, demonstrating at the very least a faulty and biased sampling procedure. From this false empirical base, the critics compound their error with problems of emphasis, tending toward overgeneralized and overstated comments about the specific political effects of voting systems. At the root of this work is often a rather simple notion that democratic institutions come into being by “choice,” free from the influence of specific historical contexts and power struggles. of course, when we turn to a truly representa-tive sample of PR-using countries, the argument that “PR equals instability” appears


even weaker. instability, understood either in terms of government tenure or the larger party system, is not a problem for PR systems in western countries.

PARTY DoMinAnCE

Another key criticism of PR actually inverts the “instability” thesis, claiming instead that the problem involves too much stability. in these complaints, PR creates sclerosis in the political system, leading to coalition governments or certain parties that cannot be effectively dislodged from government by elections, either because there are parties that are “un-coalitionable” (like the Communists in italy) or because certain parties are centrally located in the political spectrum and thus end up as everyone’s preferred junior partner (like the FDP in Germany). A related point argues that parties in PR systems are generally too strong and centralized. Concretely, these criticisms have emerged within the countries using PR, with debate in the netherlands about the tendency toward stasis in the political system emerging from the 1960s on, and concerns in Germany about the overweening power and influence of the major state parties on the right and left. By inference, the critics suggest that plurality is less likely to become blocked and that parties under plurality are less dominant.

There is certainly evidence of long periods of government stability in a number of PR systems. in Sweden, the Social Democrats have dominated government since the 1930s. The CDU has been the dominant party in Germany for most of the postwar period. The DC was a part of every italian government from 1946 to 1994—and many other examples could be found. But then, one can find long periods of government stability in plurality systems as well. The Conservatives have dominated postwar Brit-ain. The liberals ruled for most of the 20th century in Canada. And so on. if we shift our attention to the subnational arena in Canada, we can find even more examples of stability under plurality—particularly in Alberta, BC, and ontario. Then again, we can also find party and coalition alternation in PR systems. So the argument seems weak when we subject it to a comparative analysis. Both across plurality and PR systems and among those countries using PR, stability and change can be found all over.

A more serious problem with the “stability” argument is that it underestimates the more subtle kind of influence voters can have under PR beyond defeating a govern-ment or driving a party out of office. Because PR systems tend to produce coalition or minority government, governments tend to be more responsive to shifts in voter support. if a centre-left or centre-right coalition government witnesses a shift in voting support among their coalition members, such a shift tends to signal in which direction policy should move. Thus voters may not need to defeat the government or remove a party from office to affect a government’s actions. of course, strategically placed parties may find themselves the subject of much attention, but they too can find themselves on the outside of a coalition if their partners think things have become too “stable.” in fact, this has occurred on a number of occasions in Germany, where a grand coalition of left and right has eliminated the need to work with the centrist FDP. Meanwhile, the concerns raised in the netherlands about government stasis led to considerable debate and some reform initiatives, but in the end those making the case in the 1960s and 1970s did not convince enough people that their concerns warranted changing the


system in a substantial way. The issue has re-emerged recently, though discussions are focused on a shift to a different form of PR rather than a move away from PR (Andeweg 2006; Dutch Civic Forum 2006).

The idea that parties are more dominating in PR systems is also poorly articulated. Parties are dominant in all systems, as they are a fact of political life. The only difference is that plurality systems have a strong romantic characterization of local accountability that tends to distort how powerful parties actually are in all aspects of plurality voting. Yet few analysts would doubt the power of parties in plurality systems like the US or Canada. in the German case, the concerns about party power were also rooted in issues not directly connected with the voting system. German parties utilized state-sanctioned foundations to fund their efforts, and these foundations received a con-siderable amount of their money directly from the state. Concerns were raised that this relationship might lead to an unbalanced situation between voters and parties, with the latter able to maintain themselves despite declining support from voters (Scarrow 2004, 2006). What is interesting is that this sort of trend has only extended beyond Germany in recent years. in fact, in Canada, new state financing laws have in-creased the amount of state support for political parties and raised many of the same concerns articulated in the German case. Thus party dominance is not solely the product of the voting system or an issue that should be linked only to PR systems.

loCAl ACCoUnTABiliTY/GoVERnMEnT ACCoUnTABiliTY

The questions of party dominance and stability re-emerge in a slightly different way when critics of PR turn to the question of accountability, specifically in terms of local representation and government formation. on the one hand, critics argue that PR systems lack effective local representation, thus leading to a political system where voters cannot control which politicians are elected. Either the version of PR lacks a local representative altogether, as in the party-list form, or it offers at best a diluted degree of locality in either the multi-member STV form or the necessarily enlarged single-member riding version of MMP. on the other hand, they argue that PR lacks account-ability as concerns government formation because it is the parties—not the voters—who will decide how the government is formed. This will occur because, as no party will typically secure a majority of the votes, the voting process will not decisively determine the outcome of the election, unlike in the typical case under plurality.

The “accountability” arguments as applied to PR suffer from the same incoher-ence and inconsistency as those offered for plurality. As local accountability is largely a myth in plurality systems, it seems unreasonable to expect a higher standard from PR systems. in fact, some PR systems are refreshingly frank about how unimportant this factor is. in Sweden, voters do not expect national legislators to be bargain-style ombuds-people; they have an effective ombuds-service for that or they have decentral-ized decision making to the local level of government, where appropriate. it would appear that a lot of the academic discussion that criticizes PR systems on this question is still in the grip of a sort of ideological commitment to local representation, despite the considerable evidence to suggest that local representation doesn’t really matter. The commitment has led many reform-oriented academics to applaud the MMP form


of PR as the “best of both worlds” because it retains single-member ridings (Shugart and Wattenberg 2001). Though here it is telling that, when we turn to the concrete aspects of the analysis, the “local” may not be very local at all. For instance, American ridings are anything but local. The average size of a US congressional house district is 500,000 people—five times the average size of a Canadian federal riding (Kromkowski and Kromkowski 1991). This makes a mockery of “local” representation. Meanwhile, the multi-member ridings that are seen as less than ideal in Europe often have fewer people in them than one single-member riding in the US. The “local accountability” issue is often raised, but seldom demonstrated to be an effective or important part of any western political system.

Criticisms of the accountability of government under PR are even less compelling than the criticisms of local accountability. in this case, critics claim that because the question of government formation is seldom clear from the voting results, the decision about forming the government is decided by the parties rather than the voters. This is contrasted to the experience under plurality, where the voters are alleged to have been in control of the decision precisely because the voting results usually do present an unambiguous result about who should get to form the government. The key problem with this argument is that the logic is faulty. The voters do not “choose” the government under plurality—they are only given an opportunity to mark their ballot for a local candidate of a particular party. if their local candidate’s party ends up with a majority of seats, then that party gets to be the government. Thus there is a step in between the voter’s choice and government formation, just as there is in PR. The fact that plurality will typically award a minority of voters with a majority government is the result of its systemic features, not of the voters’ “choices.”

on the other hand, if plurality really is being credited with giving voters control, then it must be recognized that it is an arbitrary and capricious form of control; the kind where 40% may give some voters the ability to form the government—or not. in 1930, 45% was not enough to give liberal voters “control” of the federal government, though five years later it suddenly was. in 1972, 38.5% only gave the liberals a minority government, though in 1997 the same amount was good enough for a majority govern-ment. And many more examples could be cited. To speak about “control” or “choosing” implies a degree of consistency in actions and reactions that plurality results fail to provide. nor is it really compelling to talk about “voters” (i.e., as a whole) choosing governments under plurality when the results typically show that only a minority could be credited with the successful choice. Surely, to speak of the “voters” collectively choos-ing would at least imply a majority getting to choose? Thus for a host of reasons, the “plurality government formation” arguments are largely incoherent and illogical.

By contrast, a case can be made that, in practice, the majority of the public can be said to “choose” who forms the government in a host of PR countries because poten-tial coalition arrangements are often signalled during the election itself (Powell 2000). Thus when they cast their votes, voters already know who their parties are likely to ally with to form a majority government. in these cases, PR actually performs better on accountability in terms of government formation. Some argue that, despite these agreements, there are many cases in PR countries where a party gains support but is turfed from government or, conversely, loses support but remains in government. one


study suggested that this occurred in 25% percent of the cases, compared to just 5% of the cases in plurality systems (Katz 1997: 166–9). The problem with such studies is that they do not track the long-term responses of voters to what appear on the surface to be counter-intuitive coalition-making results. in new Zealand, voters responded to a counter-intuitive coalition deal made in the 1996 election by severely punishing the re-sponsible party in the subsequent election (nagel 2004: 130). or it may be that the co-alition deals worked out in PR systems after elections did ultimately meet with the approval of a majority of voters in those countries. There is also evidence that plurality systems can deliver counter-intuitive results; indeed, in one study of western countries it was discovered that the systems let the “loser win” nearly 20% of the time (Powell 2000: 130).

PR has been accused of complexity, instability, too much stability and party domi-nance, and a lack of accountability in terms of local representation and government formation. But a critical scrutiny of the arguments and the facts used to support such accusations has found them wanting. PR systems create more transparent results; pro-vide stable government and party competition; allow for nuanced shifts in government on policy and composition; and perform no worse than plurality on local representa-tion, and arguably much better on accountability in terms of government formation.

new issues: Separating Defensible from indefensible Claimsover the past two decades, a number of new issues have entered into the debate over voting systems—in particular, concerns over declining voter turnout in elections and what is felt to be the increasingly ugly tone of political campaigns. Some reformers have highlighted how a shift to proportional voting might affect both of these concerns, perhaps increasing voter turnout and fostering a more consensual, less confrontational approach to politics. These two themes alone have generated considerable excitement among reformers and more detached academics. But the claims may exceed what can be demonstrated with the available evidence about the potential impact of a change of voting system. it may be that such forecasts are too optimistic, which could condi-tion people to expect too much from reform and be disappointed with results that fall short. in the campaign for voting-system reform, advocates need to distinguish be-tween defensible and indefensible claims. Here, we’ll explore the claims made about voter turnout and political culture under PR, as well as more general claims about the potential impact of PR on the political system.

Voter Turnoutit is hardly surprising that voter turnout has become linked with voting-system reform in many people’s minds. Simply put, PR countries would appear to enjoy higher turnout in their elections than countries using plurality voting. Studies differ in their assess-ments of the magnitude of the effect: some find a modest 3 to 5% difference between PR and plurality countries while others claim up to 11% more turnout in PR systems


on average. The typical bonus awarded to PR systems over plurality tends to run to about 7 to 8% (Blais and Aarts 2006). For many, then, it stands to reason that a shift to PR voting should lead to an increase in voter turnout in Canada (CRiC 2001). But such a conclusion does not necessarily follow from the facts. new Zealand recently changed its voting system from plurality to a form of PR but witnessed voter turnout levels fall slightly rather than improve. And PR countries too have witnessed a steady decline in voter turnout over the past three decades, despite maintaining consistently higher levels than plurality countries. The factors affecting voter turnout are much broader than just the voting system, including the organization of the party system, changing patterns of voter mobilization, and the shift to a largely media-centric form of campaigning as the chief means of conveying political messages to voters (Gray and Caul 2000; Geys 2006).

What all this means is that the impact of the voting system on levels of voter turn-out is actually hard to isolate. it may be that the effect of a shift to PR in Canada would increase voter turnout, but it is also possible that such an effect could be masked by other factors driving voter turnout even lower. This appears to have been the case re-cently in new Zealand, where PR did rally new voters and previously less-motivated voters but could not completely counteract a long-term trend of declining turnout informed by changes in how parties mobilized or did not mobilize voters (Vowles 2002). The decline in personal contact between parties and voters has been highlighted as a key factor influencing the less-motivated members of the public in particular to vote (Gerber and Green 2000; Franklin 2004). it stands to reason that increased open-ings in the political system might encourage those who are presently discouraged or marginalized to participate, and there is some evidence to support this both in the Canadian context and abroad (Pammett and leduc 2003; Galatas 2004; Geys 2006). But on the other hand, there is evidence that some voters sit out where elections do not appear to be decisive or the choices are too varied (Brockington 2004).

of course, we might decide that it is better to trade participation by “horse race” voters for that of more policy-oriented voters—but we should understand that we might not end up ahead in terms of total numbers in the end. For all these reasons, those keen on voting-system reform must be cautious about promises of increased voter turnout re-sulting from a shift to PR. The evidence to support such conclusions is mixed at best.

Political CultureAnother popular refrain among reformers is that the move to PR will contribute to a shift in the tenor of politics itself. in some versions of this pitch, advocates utilize insights from the academic Arend lijphart, who argues that PR systems are character-ized by an inclusive and consensual approach to politics (lijphart 1977, 1999a). in others, proponents work out their own version of the consensus position, arguing that PR forces parties to find points of agreement ultimately because they will have to work together in a coalition. The “coalition effect” will also be thought to lessen the vicious-ness of the parties’ attacks on one another precisely because they may have to work together at some point (Milner 1999). Scholars typically qualify such statements with admonitions that conflicts and disputes will still arise under PR, but some of the more


excitable activists on the question tend to cast the “PR equals consensus” line a bit lit-erally. While there is something to what these positions set out, proponents can too easily paint a portrait of consensual politics where conflict, partisan loyalty, and strong opinion all but disappear. Such a portrait does not match the political practice where proportional voting is used. instead, political partisanship and divisions remain at the forefront of political discourse. in fact, some of the divisions—particularly on either side of the left or right division—may be sharpened. it is important to underline that the change to politics that PR can accomplish is not a populist denial of politics or a cultural shift in the way politicians interact so much as an alteration of the incentives for parties to work together. PR creates a situation where a grandstanding kind of op-position will get little in the way of results. it also creates incentives for coalition building, though such efforts do not suppress politics as much as create a space to ne-gotiate differences into an acceptable compromise.

The problem with the “end of conflict” position of some reform advocates is that it can raise public expectations to an unreasonable level, and those disappointed ex-pectations can fuel a backlash not just against politicians but against the new system itself. This is what nearly occurred in new Zealand when reformers oversold the “end of politics” argument to frustrated voters in the run-up to the voting-system change. However, when the new voting system produced a diverse Parliament that proceeded to grandstand over the formation of the first government and then over subsequent policy decisions, many voters felt cheated and called for repeal. over the longer term, however, as politicians came to realize that the old tricks would not necessarily work in the new system, voters came to appreciate what their new voting system could offer and set aside their unrealistic expectations (Karp 2002). But to avoid such a crisis, re-formers should be careful in making promises about how PR will alter political cul-tures. Canada is a politically divided society and PR won’t make the differences within it go away. The point of PR is precisely to better respect those differences, not cover them up, and offer a better way to represent and broker among them.

Defensible Claimsif better voter turnout and a consensual culture of politics are not readily defensible positions in making the case for PR, there are some concrete claims that one can make that can be defended, particularly as concerns the accuracy of election results, an in-creased level of competition among parties, and a qualitative change in the strategic pressures facing voters in making their voting choice. First, as set out in Chapter 4, PR will lead to more accurate results for voters in terms of their political party choices, providing the new system is fully proportional. Second, there is no denying that a move to PR will increase the competitiveness of the party system, both by better representing existing parties where they are weak (thus giving them an incentive to expand their support) and by lowering the barriers to new competitors (thus forcing existing parties to be more accountable or risk losing support to others). Finally, it can also be de-fended that a shift to PR will alter the strategic dimension of political competition from the voters’ point of view by reducing negative strategic-voting pressures (voting for the “best of the worst”) and increasing their positive strategic-voting power (choosing


among parties to further their policy goals). The latter will occur because under PR voters will probably have a range of parties to support rather than the more narrow “party left and party right” opportunity provided by plurality.

PR advocates must be careful in the claims they make about the possible outcomes of voting-system reform. Claims that PR will increase voter turnout or replace conflict-oriented party competition with a new “politics of consensus” are not well supported by strong evidence. it is possible—and arguments sound compelling—that PR could be expected to increase Canadian voter turnout, but the comparative evidence does not necessarily “prove” that this will be the case. in fact, some important counter-evidence can be marshalled. on the other hand, claims that focus on the more competitive results of elections held under PR can be firmly grounded in evidence and thus constitute strong and defensible claims, particularly those that focus on improved accuracy in election results, increased levels of party competition, and an altered environment of strategic voting.

ConclusionA good deal of what passes for debate over voting systems is speculation, misinforma-tion, overstatement, and outright assertion. When we turn to the facts, the story of the performance of these different voting systems turns out to be very different. in the Canadian context, the debate over voting systems has included a debate about values and a debate over comparative voting-system performance. The “values” debate is more recent and largely driven by well-meaning but ultimately misguided practitioners of civic engagement-style encounters between experts and the public in various contexts—typically, settings where a new voting system is being considered. But a values approach is misleading and poorly suited to examining voting-system reform. it encourages an uncritical view about the origins of electoral institutions and the reasons why such arrangements have been kept in place (thus limiting our understanding of just how difficult change will be), and it limits a critical inquiry into all aspects of the voting system by turning key components—for instance, local representation—into values.

The more far-reaching debate over voting systems in Canada has concerned ques-tions of voting-system performance. Here, we encounter many long-standing but poorly substantiated claims about the workings of our plurality system and those of PR in other countries. These include old chestnuts like the typical casting of the stability and alleged accountability of plurality against the instability and lack of accountability associated with PR. Such generalizations, however, do not withstand serious scrutiny. But there are also new debates concerning the potential effects of voting systems, spe-cifically their potential impact on voter turnout and political culture. While there is some encouraging evidence to support such views, there is also counter-evidence that suggests that such claims may be overstated. in the end, what is not debated is that plurality is simple to count and tends to produce majority governments, while PR more accurately represents voters and allows for more political competition. The real debate should be about what such results mean for a democratic society.

158

CHAPTER ninE

The Politics of Voting-System Reform in Canada

introductionit’s time to pull together the various arguments in this book and apply them to the present condition of voting-system reform in Canada. in many ways, it seems like the best of times: never have so many different voting-system reform initiatives been pursued at the provincial level in Canada as in the last decade. But in other ways, it seems like the worst of times: two initiatives have failed to produce results (BC and PEi), others appear stalled (Quebec) or in limbo (new Brunswick), and another seems to be getting short shrift from the media and opinion leaders (ontario). it is entirely possible that none of these reform initiatives will bear fruit. The way forward seems unclear. So far, reform-oriented academics and activists have focused their efforts on debating the values and performance of different voting systems, with middling to poor success. it’s time to alter that strategy.

Part of the problem has been rooted in a less-than-critical approach to many prob-lematic assertions about our political system. This book has sought to shift the terms of the debate by challenging the myths that have arisen—specifically, the allegedly key role of local representation in our political system, the pressing need for stability in government above all other values, and the so-called accountability that our present system it is supposed to deliver to voters. Additionally, the book has challenged many of the myths about the allegedly negative performance of PR systems. But more to the point, this book has demonstrated that questions of voting-system performance have not been central in decisions about choosing plurality or PR, either in the past or in contemporary settings, despite what public debate on the issue might seem to suggest. instead, voting systems have been chosen primarily for political reasons, in order to serve the interests of mobilized political forces. Even in the most recent and more

t h e p o l i t i c s o f vo t i n g - s y s t e m r e f o r m i n c a na da / 159

publicly deliberative reform processes, the success or failure of the reform initiatives has depended crucially on what organized political forces have done to aid or hinder making change happen.

The central argument of this book is that the institutions of voting are themselves political. Thus the reform of a voting system is a fundamentally political decision, in-volving a necessary appreciation of the political divisions at work in a given locale and how they are mobilized into concrete political action. The divisions are manifest in the mobilized political power we recognize as parties, and what parties want or don’t want is an important factor in efforts at voting-system reform. But political divisions are not simply the creation of parties. Unlike our presently dominant populist dis-course, which seems to posit a world where folks could all just get along if it weren’t for meddling politicians, there are real political divisions that cut across Canadian civil society—divisions that are reflected, albeit imperfectly, in our political parties. Those seeking voting-system reform must come to grips with both these aspects of politics if they want to increase their chances of success.

This chapter will sum up the insights from this book that support the view that our electoral institutions are—and indeed, must be—political, and that changing them requires an appreciation of this politics and its dynamics. We will then apply that ap-proach to understanding the politics of securing voting-system reform at the present time in Canada.

Why the Voting System is necessarily PoliticalThis book began with a criticism of efforts to depoliticize discussions of voting-system reform. i argued that attempts to go “beyond politics” invariably end up somewhere that is just differently political, not genuinely un-political. The push to depoliticize the discussions has two forces behind it—one academic, the other public—and both are deeply problematic. The academic approach assumes that institutions should be designed outside of the sphere of political contestation and that such a process will produce institutions defined by values, not self-interest. The public approach lumps all politicians together as the problem and refashions the notion of political struggle away from party competition and toward a battle between “the politicians” and “the people.” Both approaches appear to assume that there is someplace beyond politics where better or more fair decisions can be made. But both are mistaken. ignoring politics means that when decisions emerge from these allegedly depoliticized places, they will be cast back into the conventional politicized world and will succeed or fail as a result of their ability to survive in such a contested environment. And nor will these results simply reflect the actions of self-interested politicians, as politicians and parties and the decisions they make do bear some relationship to actual divisions among people within the polity. Given this reality, we would be better off to recognize the politics going on here from the outset and, when considering voting-system change, make strategic decisions that address and incorporate them.

The historical record of voting-system change itself recommends such an approach. As recounted at various points in this book, voting systems throughout western countries


were the product of political self-interest, not of values or value-driven assessments of voting-system performance. Though the rhetoric of voting-system discussion almost always takes on a value- or performance-driven discourse, these are post hoc rational-izations informed by considerations of power rather than by idealistic values. now, some readers might object that this line of analysis veers dangerously close to the populist anti-party explanations criticized above. After all, populists claim that politicians are “in it for themselves,” and i am arguing that voting systems have been designed by parties to serve their self-interests. But there are important differences dividing the two explanations. let’s explore them in more detail.

The populists use a discourse of “the people versus politicians/parties,” which has the effect of denying the fact that there do exist real political differences among people. in a sense, such critics seem to be saying: “Everyone i know thinks the same, so these political problems must be the creation of politicians.” it is easy to see how such views could become popular. The hallmark of modern societies, as opposed to traditional ones, is the fluidity of group membership. Within various constraints, people can as-sociate or not with different people, allowing those who would prefer to mix only with others who share their ideas and opinions to do so. it is not a big step to move from thinking that the consensus in your group is, in fact, the consensus in the larger polity, especially when your self-selection process means that you seldom encounter anyone who disagrees with you. Thus such populists typically call for an end to parties or party discipline, or for more local representation, or even for direct democracy like referendums—all with the confidence that such changes will allow the consensus they have come to believe exists to finally emerge.

The “populist” approach to understanding politicians is very different from one that focuses on party self-interest as the animating force in institutional choice. The former assumes an undifferentiated political class whose members have more in com-mon with each other than with the public they claim to represent. Thus institutional choices are about protecting politicians, as a group, against the interests of the public, also as a group. The latter approach, however, assumes that political self-interest differs among parties and that these differences do bear some links to real and abiding dif-ferences among the people in civil society. Furthermore, this self-interest cannot just be asserted; it must be explored concretely in order to see both what parties are trying to accomplish in taking various positions as well as how such positions may relate to what people in civil society may or may not want from politics. let me be clear: i am not arguing that parties are a transparent reflection of public wants, as obviously par-ties have considerable power to choose and shape what issues to take up from the public. But it is equally problematic to argue that parties and their policies have no links to those from whom they seek to gain votes.

Voting-system reform is an interesting case in point because, in some cases, it has emerged out of elite decisions within the party regarding strategy while, in others, it has been forced onto the agenda by party members and supporters. This is why the positions of different parties have changed throughout Canadian history and why they have even differed across the country in the same period. Thus both federal liberals and Conservatives have supported voting-system reforms at various times, depending on what they saw as being in their self-interest. The federal nDP has both opposed


and supported a switch to PR at different historical moments and, today, different branches of the provincial nDP seem to favour or oppose reform depending on their likelihood of forming a provincial administration.

Complaints about political-party behaviour notwithstanding, the populist take on politics goes further, in the end amounting to a denial of real political differences among Canadians. But the populist case against politics is both empirically wrong and normatively suspect. let me deal with each of these issues in turn. There are numerous facts that seem to call the populist non-political idyll into question. Canadian election and survey data demonstrate that we are divided in our political opinions. Across the country—indeed, in every riding—Canadians give their support to parties across the political spectrum. Survey work demonstrates that these voting patterns are not just random, but reflect fairly clear ideological differences among us concerning the key questions of politics, from social ones involving abortion and sexual orientation to economic ones relating to the market and government regulation. Though Canad-ians may not be well informed on the nuances of policy, they do have a sense of where the different parties stand generally on issues that are important to them and they choose accordingly (nevitte et al. 2000: 48–51). it is because most evidence concludes that party label is the key indicator informing individual voting decisions that this book makes the case that we need a voting system that can effectively translate those decisions into representation (Clarke et al. 1980: 102, 216, 218–9, 222; nevitte et al. 2000: 133–4; Blais et al. 2002: 116–7). if “party” is what people are voting for, then our political institutions should reflect that as accurately as possible, and that means switching to some form of PR. Anything that interferes with registering what voters are saying they want with their votes—like most of the so-called positive features of our present plurality system—must be rejected in this analysis as undemocratic.

But just demonstrating that people do vote for parties is not the same as demon-strating that they want to. if we are talking about what voters want and about reforms to make that happen, some might argue that we should expand the discussion to in-clude the possibility of reforming or eliminating parties. After all, there is arguably much more attention focused today on the problems of party representation than on the electoral system, particularly among populists. The political party as an institution and the practice of party discipline are often characterized by populists as illegiti-mate—essentially, as interlopers in the political system coming between voters and their democracy. Survey work to determine the public’s view of parties has produced mixed results, with the public both criticizing and endorsing the role of political parties in our political system (Howe and northrup 2000). Yet even if we just focus on the negative polls, the problem with such results is that they register what people think as individuals divorced from the need to address just how our political institutions must also reconcile conflicting individual desires. Such polls seldom force respondents to think through the implications of their anti-party views—issues like how an individual local representative would represent the divided local views of her area (particularly on zero-sum decisions) or why pooling votes on the basis of geographic locale makes more sense than pooling them on some other criteria (such as party affiliation). So despite what some surveys might suggest about a public antipathy toward parties, we cannot act on such recommendations because—at least in their present form—they


are fundamentally un-political. They try to assume away what is demonstrably real in Canadian society: that we are divided about politics.

The anti-politics rhetoric is mirrored in a host of other “reforms” often touted by populists. Parliamentary reform, referendums, more power to individual MPs, less party discipline—these are just variations on the “denial of politics” thesis. instead of posing “solutions” that deny politics, we need solutions that recognize and work with our political divisions. of course, the solution proffered in this volume is proportional representation. Taking politics seriously means recognizing that Canadians are divided politically and that we must do something to reflect this fact institutionally as best we can. Recognizing political differences is arguably the first step in the democratic trinity of representation, deliberation, and decision taking. We have already demonstrated that, empirically, party representation is what Canadians vote for in elections, and that, normatively, this appears to be the only practical available vehicle to represent the real, collective political divisions that exist across Canadian civil society. The basic argument of this book—backed up by insights from Canadian and comparative experience—is that PR can best reflect our political differences as mobilized by parties through (1) bet-ter representation and (2) better accountability as a result of increased competition. indeed, public complaints about parties focus less on their illegitimacy than the fact that the parties are too similar and voters lack choice (Howe and northrup 2000).

of course, not all of our presently organized political forces may accept the need for this “solution” and, indeed, many of their supporters may share their view. While publicly mouthing platitudes about the need for stability, majority government, and the time-honoured link between the MP and her constituents, those in power will quietly note among themselves how the current system already works in their favour. This is another aspect of the “politics” of voting-system reform. Those who currently benefit from our institutional arrangements will likely use all the power they can muster to preserve them, even if that means twisting logic, distorting facts, and attempting to rig the debate in their favour. Yet between a “politics of party self-interest” and the less sharply defined political differences among the general population, there is considerable room to push an agenda of democratic reform. The gap identified by populists—that parties are im-perfect reflections of their supporters—is also an opportunity, as it creates some space to marshal public support on the issue and perhaps force the parties to respond. indeed, this was the case in new Zealand, where the public’s frustration with the actions of their parties kept the voting-system reform issue alive and ultimately opened it up to a public referendum. How such a situation might emerge in Canada is our next concern.

The Politics of Contemporary Canadian Voting-System Reformover the past decade in Canada, voting-system reform has become more publicly vis-ible than at any point since the 1920s. Many more people now have a vague idea what “proportional representation” is or a sense that our present plurality system is not the only way to count up votes. But many Canadians have no opinion on the issue, and this is crucially related to the public’s ignorance about the workings of our voting system


more generally, never mind the alternatives. While it may be unrealistic to expect the majority of Canadians to become electoral engineers, many more might become more interested in the topic if it were effectively linked to issues that they already care about: getting specific policies, the representation of their party choice, and their frustration with conventional “politics as usual.” Some players in our political system—political parties, media, interest groups, influential people in civil society, etc.—already do have opinions on voting-system reform, and their views range from positive to negative. The question is: in the event that voting-system reform becomes a serious issue, how might all these different groups interact to either further or limit the possibility of re-form? As all groups are typically a coalition of different interests, often with some contradictory tensions, there are opportunities to further voting-system reform within them all. Successful reform will involve attention both to the key players that will fur-ther change—the public, political parties, media, and interest groups—and to the impact of, and responses to, unpredictable events and opportunities.

The PublicPublic demand for change can be a powerful factor in moving a political issue forward, and voting-system reform is no exception. Traditionally, decisions about voting rules have been made by political elites with little input from the public—most of whom would have been unaware that a decision was being made or, if they did know what was happening, would have understood very little of the details of what was being de-cided. But if the public is engaged in the debate over voting systems, the room that elites have to kill the issue in committee or negotiate some other arrangement in pri-vate is reduced. A motivated public can keep the issue on the agenda, making things difficult for political elites who refuse to act. This was certainly the case with the recent reforms in new Zealand and italy, where the public’s attention and interest in the issue eventually forced the politicians to take action.

So what does the Canadian public think about voting-system reform? Public views about the voting system have changed over the past 20 years. in 1990 a survey asked Canadians if they found it acceptable or unacceptable that our voting system allows a party to win a majority of the seats without winning a majority of the votes; at the time, 39% said it was unacceptable. By 2000 the percentage of respondents opposed had risen to 49%. When the word “fair” was substituted for “acceptable” in the latter survey, the percentage opposed increased to 54% (Howe and northrup 2000). More recent polls have seen the figure rise above 60%. Polling support for our present system is even lower than these figures might imply. in 1990 only 27% of those asked were prepared to agree that our present voting system and its results were acceptable, and by 2000 this figure had slipped to just 23%. The remainder of the respondents—34% in 1990 and 29% in 2000—declared that they either had no opinion on the issue or re-fused to answer the question (Howe and northrup 2000). This latter group represents both a problem and an opportunity. if a referendum were to be held on voting-system reform, the voting of this group of “undecideds” could be crucial.

The fact that so many respondents in these polls did not take a position may speak to the low levels of knowledge among the public about the workings of our electoral


institutions. obviously, public education and public outreach by reform forces could make a difference here. But the challenges of public education on complex topics like voting systems should not be underestimated. in new Zealand, a significant commit-ment to public education on the issue from government meant that a considerable amount of material was produced in both print and television formats. By the time the final vote between plurality and PR was held, public knowledge of voting systems was fairly high (nagel 1994). But in BC, where the government committed very little in the way of resources to informing the public about voting systems in general or about the specific proposal from the BC-CA, public knowledge remained low. Very few British Columbians reported understanding STV, and large numbers had little knowledge about either the BC-CA or the referendum. in the end, most people seemed to make their decision based on their assessment of the BC-CA as average citizens with no partisan axe to grind (Cutler et al. 2007).

Did a lack of public education and low levels of citizen knowledge hurt the BC campaign for reform? After all, nearly 58% of the voters endorsed the change to STV, despite the fact that very few of them understood what STV was or how it would alter the political system. As high as the voting results were, they were lower than polls had suggested support would be. in 2000 the regional results on opposition to plurality were higher in the West, with 63% finding it unacceptable in BC. A poll during the campaign put support for a new voting system at over 60%. of course, many factors might explain the gap between the polling and voting numbers. Yet research on the BC referendum campaign claimed to discover a trend of increasing support for STV the more the public understood it, a conclusion that mirrors other research on public attitudes and voting systems (Cutler et al. 2007; Farrell and Gallagher 1999). Such findings seem to suggest that, with a greater commitment to public education, the referendum just might have passed its super-majority threshold.

A variety of polls suggest that Canadians are interested in electoral reform, though few put the issue at the top of their agenda. in a list of four political priorities—health-care, education, tax reform, and reforming Canada’s voting system—just 6% of re-spondents rated voting-system reform as their top issue (Bricker and Redfern 2001). This might reflect the public’s sense of the importance of voting-system reform, or it might reflect their sense of how likely it is that the issue will amount to anything. in any event, polls are just indications of potential support or opposition. What comes to pass in an actual vote on the question, like BC’s 2005 referendum, also depends on mobiliz-ing the public to participate. Here, normative appeals can work. Research on public opinion and voting-system reform suggests that linking reform initiatives to broad notions of fairness and justice can motivate the public both to vote for change and to continue supporting the change after it has been introduced. in new Zealand, voters who supported the shift to PR because they saw it as fair were more likely to continue to support it, even when the system later came under criticism. Even voters who are strongly partisan may depart from their party’s position on the question if they agree with the broad values that undergird the proposal (Banducci and Karp 1999).

of course, when we speak of “the public,” we must recognize important distinctions among the members of this amorphous group. Some are more active in community organizations than others. While the general public may fill their time with work and


friends (largely unaware of any larger community they do not have direct, personal links with), there is a more active—though not necessarily political—public that is engaged in community activities at church, school, or the community centre. These members of the public can act as community leaders and give direction to their less-engaged friends and neighbours about arising matters of policy. This is why the seemingly sparsely attended public meeting can have ripple effects beyond the small number of people that may turn out. These sorts of meetings are just the kind of forum to con-nect voting-system reform to the better-known, conventional issues of politics. And research suggests that it doesn’t take much to get people motivated about voting-system reform once one has gained their attention and set aside some time to focus on the issue. A number of studies have highlighted how the public’s low level of knowledge concerning electoral institutions can lead to strong responses at a normative level when they discover some of our present system’s more peculiar features; in many cases, their discoveries fuel support for change. “That’s not fair” is a familiar refrain heard at these sessions when the workings of plurality voting are set out (Farrell and Gallagher 1999).

in Canada, the public remains a largely untapped source of strength for the voting-system reform forces. Much evidence suggests that when the public gets up to speed on voting systems, they don’t like what they see in plurality voting. Even the recent criticisms in new Zealand about the performance of MMP did not lead voters to call for a return to their traditional plurality system, despite what some polls have claimed. instead, more sophisticated survey work has discovered that most sought to keep the system, reform it, or replace it with a different form of PR (Vowles et al. 2000: 8, 39). The challenge remains how to get the public’s attention and create the space and time to explore both how our current system works and what proportional alternatives might do. As will be discussed below, the opportunities in the past have usually in-volved anomalous election results, gaffes by politicians, and unpredictable breakout events that suddenly put the voting systems under the public gaze.

PartiesA great deal of attention where voting-system reform is concerned focuses on political parties—and rightly so. it is parties that ultimately pass the laws that create or change voting systems. And if a party that exercises control over a majority of votes in a legis-lature wants a new voting system, they will probably get it. This has led many scholars to underline the difficulty—indeed, the near impossibility—of achieving voting-system reform. Basically, they point out that a party with the power to change the voting sys-tem is probably unlikely to do so because, after all, the existing system got them into power in the first place. Meanwhile, a party that would benefit from voting-system reform typically never exercises the power to secure the change. it seems like a kind of catch-22. As Canadian constitutional expert Eugene Forsey once put it, the problem with getting voting-system reforms introduced is that “[t]he parties that could, won’t. The party that would, can’t.” Thus Forsey concluded, “discussion of proportional rep-resentation is just worthless. it won’t come, won’t happen” (Forsey 1985: 196). This opinion was hardly challenged in political science for most of the postwar period until


the surprising voting-system changes in a number of western countries in the 1990s forced academics to revise their views. it would appear the voting systems can be changed, despite party preferences for the status quo. no one is arguing that change is easy to come by, but the idea is not dismissed out of hand as it once was. Parties still play a central role in the debates over voting-system reform, but there have been some successful challenges to their hegemony on this issue.

it should surprise no one that parties approach the question of institutional design from a perspective of how it will affect them electorally. Parties that stand to gain a majority government with less than a majority of the vote, as the largest parties often do under plurality, are unlikely to want to change those rules. Meanwhile, parties shut out of power or Parliament altogether may be less inclined to support the status quo. Thus the self-interest of parties tends to diverge depending on whether a party is large or small, or small and nationally focused versus small and regionally focused, and so on. Yet even these conditions can change.

in Canada, the Conservatives and liberals have—not surprisingly—been the most strident defenders of plurality at the federal level. Perhaps more surprising has been the strong support for plurality from the third-place nDP. Typically in plurality-using countries, the third party is the keenest lobby for a change to the voting system. The nDP’s seemingly anomalous position can be explained by a number of factors. First, the party was influenced by British ideas about voting systems and majority govern-ments. in the UK, minority labour governments in the 1920s and 1930s could find little coalition support for their policies. This moved labour to stand by the plurality system in the hopes that, one day, they too would benefit from its overrepresenting tendencies and be able to introduce their agenda. As the CCF (the nDP’s predecessor) began to win office at the provincial level, the party hewed more firmly to the British line. Second, provincial influence in the party—particularly from those provinces where the party formed government—vetoed consideration of the issue even at the federal level. Finally, the federal party long felt it would eventually replace the liberals as the major second party in Canada’s political system and as such would benefit from the existing rules (Pilon 1999).

Yet recently, the nDP’s position has shifted. The party’s disastrous term as a prov-incial government in ontario—combined with a dramatic decline in support at the federal level in the 1990s and its continuing failure to gain any representation in Quebec—appears finally to have convinced the party’s grassroots supporters and lead-ership that its “replace the liberals” strategy is not going to work. in addition, the party has been struggling with internal divisions over what strategic direction to pur-sue, with some members calling for a move to the left and others for a move to the right. At a number of policy conventions around 2000 a commitment to electoral reform was one of the few issues to unite members (Pilon 2004). The federal party is now clearly for voting-system change, while various provincial branches of the party either offer strong (ontario) to lukewarm (BC) support or remain silent on the issue (Sas-katchewan, Manitoba, and nova Scotia).

Changing party positions on issues like voting-system reform can reflect both in-ternal and external dynamics. A dramatic case of external effects was the party-system meltdown that occurred at the federal level in 1993. Though this alone did not lead to


leadership advocacy of voting-system reform from any of the elected parties, it did ramp up pressure within the parties to address the issue. in addition, the existence of a broad multi-party system at the federal scene led to election results that highlighted some of the more curious effects of plurality voting, particularly for parties on the right. Reform and Progressive Conservative party members started to make voting-system reform an issue at conventions, with the former making an end to plurality party policy. The leadership of both parties downplayed these discussions, but as long as the vote on the right was split between two parties the issue could not be completely ignored. This experience on the right tells us something important about responses to this issue—specifically, that party members (and indeed party voters) are much less strategic in their commitments than their leaders are. What this means is that party loyalties do not necessarily determine how people may respond to appeals rooted in normative concerns like “fairness.” in new Zealand, those calling for a return to plural-ity voting are mostly national party supporters who think that the old voting system benefited their party. Yet not all national party supporters agree with their party’s position. Some think PR is fairer, even if it doesn’t privilege their party in the way that the previous plurality system did.

Some might take the above insight as a confirmation of the wisdom of the values-based approach to seeking voting-system reform. That would be a mistake. of course, we should seek change by appealing to broadly accepted democratic norms like fair-ness in terms of elections results and equality of voting power among individuals. But such value-laden appeals should be anchored in a recognition of the existence of po-litical differences: as a society, we are divided on questions of politics and agreement on procedural “values” cannot overcome or replace this. nor should we lose sight of the practical, on-the-ground aspects of politics: regardless of what values we may seek or want, political parties and their interests will come into play and party efforts will affect the outcome of our deliberations. As noted above, while some party supporters will take up a question of voting-system reform as something separate and not neces-sarily connected with their partisan commitments, others will look to their party for guidance in making a decision on such a question. Again, experience in new Zealand suggests that the question of voting-system reform did break down along party lines, with national party voters largely opposing the change while supporters of other par-ties were largely in favour (Vowles 1995). The Canadian right is also an instructive case in point. The merging of Canada’s Canadian Alliance and Progressive Conservatives into a new Conservative party before the 2004 federal election obviously altered the strategic considerations and competitive pressures facing the Canadian right, and this development undoubtedly will have important implications for their responses to calls for voting-system reform. Thus we might expect both the new right party (no longer facing another right-of-centre competitor) and right-wing voters (no longer risking splitting their vote) to be less interested in a new voting system. ignoring how the parties might direct their supporters on the question could contribute to the failure of the reform initiative. indeed, this may have been what happened in BC in 2005.

Some argue that the influence of parties is a double-edged sword. Parties can try to influence their supporters on the question, but their attempts to sway public opin-ion can backfire; if their attempts appear blatantly self-interested and self-serving, they


may ultimately have the effect of creating more support for change. in an era of strong public criticism of parties, such a scenario seems compelling, and in the recent BC case it may have motivated both major parties to avoid public pronouncements on the issue. But parties can influence public responses to voting systems in a number of ways. First, we must remember that parties represent mobilized political influence. The voters may attend to an issue when it is in the headlines, but parties have the re-sources to pursue an issue at greater length. For this reason, their concerns will remain on the agenda long after public interest in the question has faded. Second, parties can influence how voters perceive a new voting system, as they are often a key source of information about the way it works or how it is working out. Voters look to parties to give them direction when concerns are raised about how institutions work, as they cannot become expert in all the potential institutional details of government and elec-tions. Parties will also provide the public with a framework within which to understand and interpret the workings and results of various aspects of the voting system.

A good example of this influence would be the party-fuelled “controversy” over the constituency versus the party-list representatives in MMP PR systems. Some critics allege that the fact that some MPs in this system have constituency obligations while others do not has led to serious problems. But this “problem,” often taken at face value by many academics, is largely the work of the MPs themselves—specifically, those representing parties opposed to MMP (lundberg 2006). As we know, local representa-tion via single-member ridings is dramatically overrated in plurality systems, and mostly promoted/defended by the politicians (who have an obvious vested interest in portraying their role in such a romantic light). in these MMP systems, the dual-member problem plays on this myth not because the politicians are so concerned about local representation but because they are trying to mobilize public opinion against the sys-tem or against rival politicians.

The “dual-member controversy” example highlights the potential dangers of install-ing a new voting system over the heads of the politicians. Given their key position in the political system—that they are essentially the full-time observers of and participants in what goes on—parties and politicians are well placed to define the meaning of what emerges from the political process. Thus politicians may try to connect broader political problems with the workings of institutions. Just as reformers may have linked various systemic results from the political system to the use of a particular voting system, so too can politicians do the same; and, as they maintain their organizations (while reformers shut theirs down after they have achieved their goals), the parties can try to chip away at the arrangements they don’t like over the long haul. The power of the permanent political class over reformers was dramatically revealed recently in italy, when the government reverted to the country’s historically highly proportional voting system from a mixed system without public input or sanction, despite the extensive public pro-cess that had contributed to the reform of the traditional system (Massetti 2006).

The political parties are going to influence the debate over voting-system reform as well as the public’s reception of any new system and their assessment of its workings over time. All this doesn’t mean that reformers shouldn’t seek voting-system change that runs against the wishes of dominant parties. But nor should they ignore the role that parties may play in facilitating, deflecting, or subverting their efforts.


MediaThere was a time when voters would attend a public meeting to hear what prospective candidates for office had to say about the issues or politicians or their representatives would knock on doors across their riding to speak directly with potential supporters. Such practices created direct links between the political system and the people it is al-leged to serve. But over the past three decades, direct contact between the public and politicians has declined dramatically. Today, print and broadcast media are the key sources of information about politics for the public. indeed, the media often tout themselves as a quasi-public service, acting in the interests of an absent public by scrutinizing legislative activity, the decisions of politicians, and then reporting their discoveries to citizens. Thus for all intents and purposes, the media are the key inter-face between the public and the politicians. To the extent that we understand what politicians are doing, it is through the representations of their efforts that appear in and are given shape by the media. This means that the debate over voting-system re-form will also be carried out largely in the media, and that a good deal of what the public comes to know about it—what it is or will accomplish, good or bad—will be influenced by the media. There are many reasons for reformers to be concerned about this media role.

on the plus side, recent discussions of voting-system reform in Canada’s media have produced some high-profile support. Canada’s self-professed national newspaper, the Globe and Mail, came out in favour of a modest form of PR in an editorial in 2003 and made a more extensive case for their approach in a series of editorials in 2005. When the ontario Citizens’ Assembly model of MMP was released in 2007, the paper lauded their efforts—though it stopped short of endorsing the result. Globe and Mail columnist John ibbitson has repeatedly called for PR (particularly at the national lev-el), and a smattering of columnists across the country have signalled their support for a more proportional voting system, including Andrew Coyne (National Post), Paul Willcocks (Victoria Times-Colonist), Rick Salutin (Globe and Mail), and Michael Smyth (Vancouver Province). Unfortunately, these voices are a distinct minority.

When it comes to most reporters, columnists, and editorial writers in Canada, to say they do not favour voting-system reform is an understatement. They do not merely oppose it—they appear repulsed by it. Their reasons for opposing a change in the voting system vary, from a fear of “fringe” parties to government instability to a host of more idiosyncratic reasons. But one thing is consistent: apart from a few well-informed critics like Jeffrey Simpson (Globe and Mail), the media opponents of voting-system change are very poorly informed about how voting systems work and the kinds of re-sults they produce. This creates a situation where reformers and academics seem to spend most of their time just correcting the inaccuracies and distortions that appear daily in the press. The reasons for such a lopsided media response to voting-system reform are unclear. After all, the media often present themselves as being above parti-sanship and as the champion of the “little guy” against the powerful interests. Yet on this issue, they appear to be cozying up to the same establishment interests that they claim to be holding to account with their reporting. if nothing else, one might have expected a better balance of pro and con among the media, especially as the public


appears to be more evenly split on their allegiance to either the present system or some alternative.

Examining the discourse of the media on this question provides little insight into why their responses are so unbalanced. As Andrew Coyne asked in an exasperated response to a particularly one-sided editorial rejection of PR in the National Post: “What is it about proportional representation … that turns otherwise sensible indi-viduals into raving loonies?” (Coyne 2007). There seems to be both a populist rejection of reform and an anti-populist rejection of reform. Responses to the ontario Citizens’ Assembly are quite telling on this. Some reporters focus on the cost of the citizens’ assembly process or rail against an increase in the number of politicians, thus attempt-ing to appeal to public antipathy to government spending and politicians in general (Blizzard 2007; Denley 2007). others attack the whole notion that citizens should be involved in such a decision, thus rejecting what they see as a populist and largely empty exercise in citizen engagement (Urquart 2007; Campbell 2007). Additionally, such reporting is peppered with speculation about the inevitably dire results that would be produced by a proportional voting system, focusing mostly on the possible explo-sion of extremist fringe parties and political instability. The elitist and undemocratic assumptions undergirding such analyses are disturbing enough. But their seeming unwillingness to engage with facts, or their selective approach in the use of facts, re-flects poorly on these particular journalists. The “parties” question is a case in point, as most serious work on PR and party systems demonstrates that the introduction of PR into solidified democracies does not lead to excessive fragmentation of the party system (Reeve and Ware 1992: 9–10; lijphart 1999a; Farrell 2001: 161–4).

it should be noted that in trying to assess media responses to this issue, just focus-ing on what journalists and columnists do is a bit one-sided. The problem might lie with the gap between the media’s own claims about their public role in democratic deliberation and a more critical assessment of what the media are doing and why. Most serious media analysts argue that the media are a business whose top priority is mak-ing money, not serving the public (nesbitt-larking 2001; Taras 2001). Media do not seek to speak to all the public, but instead focus on those whose consumption patterns and capacities fit with the needs of the advertisers they seek. it may be that the media think the group they are speaking to are largely cynical and wholly self-interested, and thus will respond positively to populist anti-government reporting (“government is a waste of money”) or elitist anti-democratic columns (“the masses are not smart like you and me”). But here, they may be making a big mistake. in BC, the pattern of media response was similar to what has emerged in ontario. Though the Vancouver Sun and Province endorsed STV editorially, most of their columnists and reporters condemned it along with the rest of the BC media. Yet a majority of the voting public clearly re-jected the dominant media line, as nearly 58% voted in favour of change. obviously, the media’s message did not resonate with all of their readers and viewers.

of course, even if the media were prepared to grant equal time and fair treatment to both sides in the voting-system debate, there would still be some challenges in try-ing to use the media as a space for public deliberation on the question. Beyond the specific biases of this or that reporter or columnist, there are more structural biases at work in the ways that media portray issues. Those who study media note that they


tend to “frame” stories and issues in particular ways (nesbitt-larking 2001; Taras 2001). The makings of a “good” news story tend to include conflict, simplicity, and a focus on an individual or individuals. Such a bias in story selection means that issues that do not fit this mould tend to be overlooked. Another frame that is typically used in media reporting includes what is seen as “unusual.” in covering the STV debate in BC, for instance, the media constantly focused on the complexity of the STV count or the “strangeness” of the system compared with plurality voting (Brown et al. 2005). These frames do not promote an open deliberation about the pros and cons of the public policy at issue—in this case, comparing voting systems. Another difficulty with using the media for deliberation is the problem of concision, or the fact that there is often only enough time in media to confirm what people already think is true rather than challenge what may be conventionally accepted. These time constraints in media are extreme. on television, sound bites on news are typically five to ten seconds long. Even in print, where there has typically existed a greater opportunity to explore issues, the trend over the past two decades has been toward shorter and less substantive re-porting, fewer stories, and more ad copy. All of these tendencies mean that media is a problematic space within which to take up serious and complex issues.

Still, the media is where people get their political information and, if nothing else, an appearance on television or an interview in a newspaper can spark public interest in a topic, particularly if the audience is directed elsewhere to gain more detailed and nuanced information. Working with the media means understanding the media’s particular constraints. newspapers offer more space to set out a case for reform, par-ticularly in an op-ed on the opinion page. Reporter interviews are trickier, as one never knows what material a reporter may take from what is inevitably a much longer in-terview. The trade-off with newspapers is the lack of back and forth that might clarify a point, though with the rise of online website capacity for most major newspapers there is some opportunity for a kind discussion among writers and readers. Radio—and talk shows in particular—allows for a more interactive approach to taking up an issue, though the bombast of talk-show hosts and some of their listeners may not al-low for a very nuanced discussion. Broadcast media requires a more focused message from participants than a print interview. one must get to the point, and never lose sight of the point that is being made because this medium is unforgiving to those who ramble. Television is both the best and the worst of media. it is the worst because it typically has the least depth; the visual side of TV tends to overwhelm the verbal as-pect, meaning that the issues that have greater direct visual impact get more time and seem more persuasive than ones that rely on verbal argument and explanation. But television is also the best because it is where most people get their news and political analysis. A little positive TV exposure can go a long way in furthering a cause, while negative TV coverage can really hurt it.

Despite this critical appraisal of the media and its practices, voting-system reform-ers could learn something about how to effectively communicate their message to a mass audience from our conventional players in print, radio, and television. Too often, those keen on voting-system reform do not critically assess the audience they are try-ing to reach. Most seem to think that selling the public on a new voting system means that they need to explain in detail the structure and workings of our present system


as well the minutiae of the proportional alternatives. But they would do better to think about the broad results that people want to see from our political process than about the details of voting rules. Selling voting-system reform is lot like selling a car: you focus on performance, not on what is under the hood. Ultimately, as long as the car delivers on the promises, consumers will let the experts worry about why and how it works.

The media have been and will remain a central player in the voting-system reform debate. Though it would appear that many high-profile media types have already come out against changing the voting system, there are still many journalists and edi-torial writers with genuinely open minds on the issue. The present attitudes of some media personalities and the structure of the media itself present a host of challenges to reform forces, but in this such forces have little choice. Reform forces must seek creative ways to engage the media and work around some of the very real limitations noted above. if nothing else, sometimes an issue breaks out of these media-imposed frames, either because it links up with an unpredictable event (a gaffe by a politician, for instance) or because it gains a strong public response.

Civil SocietyBetween the public and the state is a vast terrain, often referred to by political scientists as “civil society.” There are permanent and more fluid elements of this space. Parties and the media are technically part of civil society, but so are local chambers of commerce, charities and community organizations, and less-structured single-issue campaigns and social movements. Civil society is where people organize to influence each other and the state. long before the media or political parties take up an issue like voting-system reform, people have been active in civil society trying to drum up interest and support. Some of this work is aimed at the general public and some of it is aimed at other or-ganized forces in civil society. A group like Fair Vote Canada sends representatives to public meetings where political issues may be raised, organizes its own public meet-ings to draw out interested members of the public, and arranges meetings with other organized groups whose members might be interested in voting-system reform. Thus before an issue breaks out as the topic of political debate in the media or between parties, there already exists a network of supporters across civil society who can respond when the issue finally becomes more broadly public. of course, opponents of change can also try to draw on their links within civil society to bolster their opposition. We have seen both tendencies at work in the recent initiatives for voting-system reform.

in new Zealand, a powerful business lobby became a key force in civil society that tried block change, using its significant financial resources to advertise its opposition to proportional voting (Vowles 1995). in Canada, there is some evidence of business opposition to voting-system reform as well. A 2007 Compas poll of business leaders and CEos found considerable opposition to changing the voting system (Compas 2007). During the BC STV referendum, the Vancouver Board of Trade urged its mem-bers to oppose the new system; the business-oriented Fraser institute, Frontier Centre for Public Policy, and the Atlantic institute for Market Studies have all published stud-ies or policy positions that warn PR would threaten economic stability and growth. But business and those on the right of the political spectrum are not united on this


issue; the Canadian Taxpayers Federation and many former Reform and Canadian Alliance members support more proportional voting, particularly in the STV form. So too are there divisions on the left, again particularly as concerns STV. in 2005 most of the major unions in BC came out against STV while the leftish Vancouver weekly entertainment paper, the Georgia Straight, urged progressive voters to reject the new voting system because they claimed it would discriminate against the poor (Georgia Straight 2005). That the paper’s analysis misconstrued and misapplied a number of its academic sources was ultimately less important than the fact that it pushed a num-ber of buttons for its left-of-centre readership.

in the battle for public opinion, what the truth really is matters less than what people believe the truth to be. in the 2005 STV referendum, there were forces in civil society from both the right and the left that used whatever rhetorical manoeuvres they thought would get voters to cast their ballots against change. often, their positions were con-tradictory. For instance, some would argue simultaneously that the problem with STV was that it would lead to a proliferation of parties and coalition government and that the problem with STV was that it wasn’t going to be proportional enough and wouldn’t help smaller parties. Clearly, the two positions are inconsistent. But here, the critics—unlike the proponents of change—were less interested in putting forward an internally consistent position. They knew that their audience was typically poorly informed about voting systems and were merely trying to throw as many potentially objection-able issues onto the table as they could in the hope that audience members would find at least one of them appealing. By contrast, supporters of change (who also ranged across left and right) often earnestly tried to respond to calls for an explanation of the weighted-inclusive Gregory method of ballot allocation in STV or an assessment of the value of the Droop quota over the Hare quota—not realizing that such requests were designed to make the proposed reform appear weird, confusing, and unrealistic.

it is curious that discussions of voting-system reform often neglect how the debate itself is usually highly politicized, particularly in the public realm. To read various accounts of reform, one might think that the whole process was simply an earnest encounter among different principled positions. But this would be to take the debate at face value. The real reasons why different groups endorse or oppose an issue and those given for public consumption may diverge sharply. For instance, during the hearings over the province’s referendum bill in January 2007, the ontario minister responsible for democratic reform stuck to a well-crafted statement on the province’s super-majority rule—despite the fact that she could marshal no compelling historical, legal, or normative arguments to support it. The statement was obviously a cover for the government’s real reasons for imposing the rule, which had everything to do with political self-interest. in the 1923 debates over voting-system reform in the federal House of Commons, speaker after speaker made eloquent appeals on both sides of the question, all rooted in principle and persuasion, but in the end the measure failed because it was not in the political self-interest of King’s liberal party (Pilon 2006). little has changed in the intervening years—although participants will all claim to be dealing with the substance of the issues at hand, in reality their positions are often determined by their understanding of how to best pursue their political interests. As a result, the politics of debating voting systems must be underlined.


Debate over voting systems is politicized because the voting system deals directly with how different collective groups may gain power over the state and its resources. The voting system is ultimately the lever of political power, the aperture through which political aspirants must squeeze. Those with the power, or who anticipate gaining the power at some point, almost always support keeping the aperture closed to the point where only they can gain access. needless to say, those farther from power usually en-dorse opening the aperture up. of course, particularly in the present, naked political self-interest is not well received in public settings in debates on the issue. Thus political interests must find other reasons to oppose or support changes offered up for public discussion.

Another group of participants in the debates within civil society are various “experts,” mostly drawn (not surprisingly) from the ranks of political science and professional political punditry. Most offer modest assessments of what change might produce, buttressed by a host of qualifications related to the difficulties involved in making such predictions. But not all are so modest in their predictions—some come out swinging. Bill Teileman, an “expert” in political communication due largely to his experience in the BC Premier’s office and as a political columnist and reporter, declared confidently that STV would not aid the election of women in BC because it had failed to do so in ireland and Malta. While Teileman cites a number of academic experts on women’s representation, he fails to mention their focus on the cultural factors contributing to the poor representation of women in both cases (Teileman 2005). His handling of the other possible effects of STV is equally problematic, whether he is addressing the chanc-es of independent candidates, how the voting system might affect campaigning, or the electoral prospects and/or impact of smaller parties. in each case, he simply cites ex-amples that support his view and ignores those that do not. Thus Malta’s experience with STV “proves” that third parties and independents cannot get elected under the system, despite the fact that readily available evidence from ireland suggests just the op-posite. These self-styled experts may play fast and loose with the facts, substituting confident rhetoric for a substantive engagement with the issue. needless to say, the public needs to assess what such “experts” say rather critically.

Some academic participants do not fare much better. University of Toronto politi-cal scientist nelson Wiseman criticizes reformers for speculating about what might result from a PR system in Canada, but then proceeds to do some speculating of his own. Specifically, Wiseman suggests that PR might lead to right-wing minority gov-ernments and that small parties would have disproportionate influence in coalition governments. in the end, Wiseman throws everything and the kitchen sink at PR, in-cluding dredging up its alleged responsibility for the rise of the nazis in Germany (Wiseman 1997). At one level, it might be understandable that academics would respond critically to some of the more exuberant and unqualified claims of PR activists—par-ticularly those that suggest reform would immediately improve voter turnout and end partisan bickering. But they hardly model a better kind of analysis with such scattershot and unbalanced responses. nor do the academics necessarily get the facts right. Wise-man’s confusion about what kind of voting system BC used for its provincial elections in 1952 and 1953 is a case in point. He says it was a form of PR, but in reality it was


the majoritarian alternative vote (Wiseman 1997). Another example is University of Calgary political philosopher Barry Cooper’s confusion about the voting formula utilized under STV in a piece written for the Fraser institute in which he argues against PR. Cooper suggests that candidates must receive a majority of the vote to be elected under STV, when in fact they only need to gain a quota (Cooper 2005). These might seem like minor mistakes, but if the experts do not understand the workings of the different systems they are criticizing, it calls into question the claims they make about their possible effects.

one of the more surprising developments to emerge from the recent voting-system debates in BC was the public response to the Citizens’ Assembly and its members. While most members of the BC-CA did not play high-profile roles in the public de-bates, they did appear at events in their communities and were interviewed by the local press about their experiences and why they made the decision they did. What came through to the public was the participants’ sincerity and their lack of any political agenda other than to do what they thought was best. in the end, an academic study credited the positive public reaction to the BC-CA and its members as one of the driv-ing forces behind the strong support for the “yes” side of the referendum—despite the fact that few members of the public really understood the proposal (Cutler et al. 2007). no doubt this factor explains why so many opponents of voting-system change in ontario have begun their campaigns by attacking the credibility of the ontario Citi-zens’ Assembly (Macleod 2007). By calling into question the legitimacy of the oCA process and role played by its members, the critics hope to derail what turned out to be a key benefit to the “yes” side in BC’s referendum in 2005.

Civil society is ultimately the source of both support and opposition for changes to our voting system. The preparatory work down here by the advocates of reform will make an important contribution to any campaign for change. But opponents of reform will also draw on their resources and networks in civil society in an attempt to prevent any change from occurring. in the process, the tone of the debate will undoubtedly shift from principles and factual analysis to politically motivated rhetoric and distortion. Being aware of the politics of the debate over voting systems does not mean that advocates of change must or should adopt the same tactics. But assessing what is really going on in a public debate—particularly in the media—can be crucial in figuring out how to re-spond effectively. in debating with opponents who are unscrupulously fudging facts and mixing up different positions, advocates must stick to their message and point out the inconsistencies in their opponents’ positions. in a “war of position,” the point is to get others off track and lead them to debating the issues that make your points look good; this is why opponents of STV endlessly tried to get debaters to deal with the technical aspects of Droop quota and the weighted-inclusive Gregory allocation meth-od used in STV. it is highly likely that they couldn’t have cared less about these technical aspects, but simply knew that public discussions dominated by such issues would turn people off change. instead of being sidetracked, reform advocates must “stay on mes-sage” and call their opponents on their manipulative tactics. Ultimately, reformers must be prepared for unscrupulous behaviour precisely because the struggle over the voting system is so political and connected with questions of power.


Bringing It All TogetherSo far, we have reviewed the many possible and probable participants in the contem-porary voting-system reform debate. The public, political parties, media, and various organized groups and powerful forces in civil society will all ultimately become in-volved, especially when a decision must be made in a public vote. But all this doesn’t tell us just how voting-system reform might emerge as an issue where no vote is pres-ently scheduled (such as at the federal level and in most of the provinces) or what it will take to push the vote over the now-conventional 60% super-majority requirement in any of the referendums that may come up. Despite the claims of “predictive” political science, no one can say definitively which factors are decisive in getting voting-system reform on the agenda or which ones will aid securing a change from one system to another. We can, however, draw broad lessons from past experience—ones that might help us anticipate opportunities and openings in the present.

Some political scientists have tried to predict the circumstances that would lead to voting-system reform in Canada. in 1979 William irvine suggested four scenarios that he thought might contribute to a change in voting system at the national level: a sus-tained period of minority government, the nDP holding the balance of power, an erosion of liberal support in Quebec, and/or an English/French polarization in the House of Commons (irvine 1979: 72). Since then, many of irvine’s conditions have arguably come to pass—but voting-system reform has not. not surprisingly, others have been more cagey about their predictions. louis Massicotte has offered three sce-narios that he thinks could conceivably lead to voting-system reform. He suggests it could come via the parties, through a referendum, or as the result of a Charter challenge. But ultimately, he seems pessimistic about the likelihood of any of the strategies suc-ceeding—with the exception of the party route where a minority liberal government might act under strong pressure from the nDP (Massicotte 2005: 90–3). Comparative scholars have developed models to predict when countries will change their voting systems, with some focusing on elite reactions to the rise of class politics while others highlight the interaction of “extreme” institutional designs with contingent political events. But such approaches have generally ignored or brushed over Canada and Can-adian circumstances, both in terms of explaining past Canadian voting-system reforms or anticipating the recent re-emergence of the issue in this country (Boix 1999; Shugart 2001).

The problem of predicting voting-system reform is really very basic. Human affairs are contingent and unpredictable because human beings have agency and can learn from past experience. Even when we can spot a pattern—as we can with the wave of simultaneous voting-system reforms across western countries around World War i—there is no guaranteeing that the conditions that contributed to those reforms will produce the same results in any other period. Politicians have long been painfully aware of this contingent nature of politics when they complain about the impact of “events.” Though they may prepare for a speech, organize the troops for electoral battle, and pay for expert advice in the form of polling, no politicians or party can predict the flow of complex events that pop up from day to day. And it is in these events that our opportunities for voting-system reform reside as well. This is where political scientists


have often been too conservative in their assessments of the potential for change. When we rationally calculate where the interests of the different parties lie on this question, the sum is always negative when we add up the parties that actually have the capacity to make change happen. But this rationality is too fixed: it lacks an apprecia-tion of events, and in particular of how the impact of unpredictable events can alter what appears rational even to the most staid and conventional power broker.

in Canada, the “events” that have recently moved the discussion of voting-system reform along include the meltdown of the national party system in 1993; the breakup of the right coalition embodied in the former Progressive Conservative party; the series of perverse election results in BC, Quebec, and the Maritimes throughout the 1990s; the election of the nDP in ontario with just 38% of the popular vote; the persistence of a multi-party system at the federal level for more than a decade; and so on. others might add even more events. Behind these events are obviously broader changes that have contributed to the “sudden” rupture with the political status quo. in the case of the breakdown of the federal party system in the 1990s, obviously the actions of the Mulroney Conservatives on a series of constitutional and governing questions over the previous decade contributed to their ultimate electoral Waterloo in 1993. The point is, all these cumulative effects finally came to a head in an “event”: the 1993 election. other events are more accidental and random. The verbal slip-up by the labour prime minister in new Zealand during a televised debate turned what was supposed to be a dodge on the question of electoral reform into a promise to hold a referendum on the issue. But after he made the promise on national television, it was hard to publicly disavow the commitment.

Probably the most unpredictable aspect of the recent voting-system reform debate in Canada is the potential long-term impact of drawing public attention to the details of our democratic institutions. As noted above, the public generally knows little about how our democracy works. But there are promising signs that, when the people do grasp how our present institutions work and the kind of results they typically produce, they don’t like what they see. There really is no predicting how Canadians might respond when enough are equipped with a working knowledge of their dysfunctional demo-cratic institutions. This is why public education and television access are such key priorities for voting-system reform initiatives. if there is a chance that this issue will break out as a priority for the public, it is much more likely where a public-education campaign—particularly one on television—is in full swing.

Though voting-system reform cannot be predicted, there are a number of steps that can be taken to further its possibility. Reformers must seek to extend their reach into civil society, strengthening links to community groups, activist organizations, political parties of all stripes, and the media. With such relationships in place, reformers will be in a position to take advantage of unpredictable opportunities as they emerge.

ConclusionVoting-system reform is political. in one sense, this is so because we are a people in a particular place—Canada—with divided views about “the good life”; the proper way to govern; and the values that should inform, shape, and limit public choices. There


is no depoliticized space where we can go to make such decisions, much to the chagrin of academic and populist advocates of a non-partisan or depoliticized process. Given this unavoidably political world in which we live, normatively we must find a way to represent our differences as the necessary first step in a democratic trinity of represen-tation, deliberation, and—finally—decision taking. This book has argued that such a normative understanding of political reality points us necessarily toward proportional forms of voting. Arguments against PR often amount to a denial of our politically divided reality (an empirically untenable position) or attempt to use undemocratic rationales (a normatively untenable position).

in another sense, voting-system reform is political because it touches on questions of power—specifically, the power exercised by conventionally organized political forces in civil society. The kind of voting system in place will crucially affect the kind of political competition that exists in any given polity, the accuracy of the representa-tion achieved, and the nature of the governing process that results. Because the voting system is the gateway to power and influence for those who aspire to exercise decision-making control over the state, attempts to alter it will spark resistance from those accustomed to exercising that power. Yet the conventional political forces are them-selves coalitions of interests, and these coalitions can be affected by unpredictable events and organizing within civil society.

Here lies one of the great challenges at the heart of voting-system reform initiatives: to promote a universalistic vision of democracy that respects our political plural-ism—one that rises above partisan appeals and direct party self-interest—even as re-formers gauge the political realities on the ground, specifically the realities informed by the realpolitik of party-centred voter mobilization and electoral politics. Between the self-interest of the organized political parties and the public’s less-formal political commitment to party voting, there is much opportunity to promote a democratic agenda. These opportunities exist in the public’s lack of knowledge about our present electoral arrangements, in the gap between party members and leaders on questions of democratic accountability, in the myriad spaces of civil society more informally, and in the “public service” discourse (if not practice) of our print and broadcast media. no one can predict when the breakthrough moment might occur for the issue of voting-system reform. But if the preparatory work has been done in the many spaces outlined here, a surprisingly minor event can sometimes blow the whole thing wide open.

179

Further Resources

Websites

Voting-System Reform Advocacy

nATionAl

Fair Vote Canada . . . . . . . . . . . . . . . . . . . . . . . . fairvotecanada.orgCenter for Voting and Democracy (USA) . . . . fairvote.orgElectoral Reform Society (UK) . . . . . . . . . . . . . electoral-reform.org.uk

PRoVinCiAl

Quebec: Mouvement pour une démocratie nouvelle . . . . . . . . . . . . . . . . . . . . democratie-nouvelle.qc.ca

ontario: Fair Vote ontario . . . . . . . . . . . . . . . . fairvotecanada.org/ontarioAlberta: Fair Vote Alberta . . . . . . . . . . . . . . . . . fairvotecanada.org/AlbertaBritish Columbia: Fair Vote BC. . . . . . . . . . . . . fairvotingbc.com

ElectionsACE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . aceproject.orgiDEA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . idea.intElections British Columbia . . . . . . . . . . . . . . . . elections.bc.caElections Canada . . . . . . . . . . . . . . . . . . . . . . . . elections.caElections ontario . . . . . . . . . . . . . . . . . . . . . . . . electionsontario.on.caElections Quebec . . . . . . . . . . . . . . . . . . . . . . . . electionsquebec.qc.ca


Canadian Voting-System ReformBritish Columbia Citizens’ Assembly . . . . . . . . citizensassembly.bc.caontario Citizens’ Assembly . . . . . . . . . . . . . . . . citizensassembly.gov.on.caQuebec Ministry of Democratic

institutions . . . . . . . . . . . . . . . . . . . . . . . . . . . . institutions-democratiques .gouv.qc.canew Brunswick Commission on

legislative Democracy. . . . . . . . . . . . . . . . . . . gnb.ca/0100Prince Edward island Electoral

Reform Commission . . . . . . . . . . . . . . . . . . . . gov.pe.ca/electoralreformKnow STV. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . knowstv.caYES for BC STV . . . . . . . . . . . . . . . . . . . . . . . . . stvforbc.com

Related Advocacy GroupsDemocracy Watch (democratic

accountability) . . . . . . . . . . . . . . . . . . . . . . . . . dwatch.caEqual Voice (women’s representation) . . . . . . . equalvoice.caFireweed Democracy Project

(democratic reform) . . . . . . . . . . . . . . . . . . . . fireweeddemocracyproject.ca

Key Academic JournalsAmerican Review of Canadian StudiesCanadian Journal of Political ScienceCanadian Parliamentary ReviewElectoral StudiesEuropean Journal of Political Research Parliamentary AffairsJournal of Canadian StudiesParty Politics

Journalism/Think Tanks with an interest in Voting-System ReformCanadian Dimension . . . . . . . . . . . . . . . . . . . . . canadiandimension.comFraser institute . . . . . . . . . . . . . . . . . . . . . . . . . . fraserinstitute.caFrontier Centre . . . . . . . . . . . . . . . . . . . . . . . . . . fcpp.orgGlobe and Mail . . . . . . . . . . . . . . . . . . . . . . . . . . theglobeandmail.cominstitute for Research on Public Policy. . . . . . . irpp.orgMaclean’s . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . macleans.caNational Post . . . . . . . . . . . . . . . . . . . . . . . . . . . . canada.com/nationalpostRabble . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . rabble.caThe Tyee . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . thetyee.ca

181

Glossary

Additional-member system (AMS)often utilized as another term for either mixed-member proportional or parallel vot-ing systems, though it is less precise because it does not indicate whether the overall results produced by the system referred to will be proportional or not. AMS systems may also differ from MMP or parallel systems in that some proposals eschew the party list and select additional members from the “best losers” in the constituency contests.

Alternative vote (AV)A majority voting system where voters rank the candidates numerically on the ballot (i.e., 1, 2, 3, and so on) in order of preference. To gain election, a winner must gain a majority of the votes cast. The ballot counting is con-ducted in stages. if at the end of the initial count a candidate has gained a majority of the votes cast, she is declared elected; if no candidate has secured a majority, then the lowest vote-getter is eliminated and her votes are redistributed at their full value to the sec-ond preferences marked on the ballots. This process continues until a candidate gains a majority. Though now used primarily in single-member ridings, AV has been used in multi-member ridings as well.

At-large votingThe colloquial term for multi-member plu-rality, though technically any voting formula could be applied “at large.” in the north American context, “at large” is used to describe municipal councils where the members are elected from across the city as a whole and by the voters in the city as a whole, rather than from single-member districts, or “wards” (as they are known at the municipal level).

Ballot structureRefers to the manner in which voters mark their preference(s) on the ballot, which can take either a nominal form (such as an “X”) or an ordinal form (such as “1, 2, 3,” and so on).

Citizens’ assembly (CA)Recently utilized in British Columbia and ontario, CAs have become a popular way for politicians to avoid accusations of self-interest in designing electoral systems. CAs are typical-ly composed of somewhat randomly selected individuals who, with expert help, review a matter of public policy and make recommen-dations to government or the public either in the form of a report or a decision that can be put to a public vote in a referendum.


CleavageA social issue or interest that may be mobil-ized into the political realm by actors in civil society, such as the media or political parties. Traditional cleavages that have been mobilized into Canadian politics include class, region, language, ethnic identity, and religion. once mobilized, these now-political cleavages be-come a factor in elections and political activity, placing constraints on or opening new op-portunities to political elites.

Contagion effectThe general idea that once one political party gains success by responding to an issue—say, bettering the legislative representation of women—other parties will be forced to re-spond or risk allowing the original innovative party to be the only one to gain advantage. This effect is more dramatic in electoral sys-tems where party competition is more open (e.g., proportional ones).

Cumulative vote (CV)Considered by many analysts to be a semi-proportional voting system, it is technically a variant of plurality voting. With CV, voters can pool a number of votes on a single candidate if they wish (for a more detailed explanation, see page 23 in Chapter 2). This allows candi-dates who appeal to a minority of voters to secure some representation.

Cycling of preferencesA term drawn from the “social choice” school of theorizing competitive elections. in this approach, whenever a choice must be made, and more than two serious options are on offer, there is a risk that no choice will emerge as the clear winner. instead, it is possible that a “cycling of preferences” will occur—i.e., that successive majorities will be found for each choice, depending on which choice is elimi-nated first (for a more detailed explanation, see page 16 in Chapter 2). Though much written about, preference cycling is merely a possible outcome of voting, one for which the evidence is slight in actual elections.

Democratic deficitA generic expression meant to draw attention to the substantive shortcomings of conven-tional western democracies—in particular, the lack of meaningful space for public participa-tion in democratic decision making and the weak accountability between government and the governed. The term has become especially popular with those addressing the declining levels of voter turnout.

Double ballotA majority voting system that utilizes a two-stage balloting process. in stage one, voters cast a ballot for their choice. if one candidate secures a majority, she is declared elected; if no candidate gains a majority, then another round of voting is held. in some cases (e.g., French presidential elections), the field of candidates in the second round is limited to the top two vote-getters, while in others (e.g., French legislative elections) all the candidates may run again if they wish. in practice, when the results of the first ballot are known, can-didates and their parties negotiate agreements with other candidates and their parties for each to stand down in certain ridings and continue running in others. This system is also known as the second ballot.

Flexible/stable partisansStable partisans stick with one party from election to election, while flexible partisans shift their support based on an assessment of a party’s past performance and/or what the party is promising for the future. over the past three decades, a majority of Canadians have shifted from being stable to flexible partisans.

Limited vote (LV)Considered by many analysts to be a semi-proportional voting system, it is technically a variant of plurality voting. With lV, voters have fewer votes than there are positions to be filled. For instance, in the case of a three-member riding, voters would have just two votes (for a more detailed explanation, see page 22 in Chapter 2). By limiting the number

g lo s s a ry / 183

of votes that an elector can cast, the system prevents the largest and most coherent voting bloc from winning all the seats. This allows candidates who appeal to a minority of voters to secure some representation.

MajoritarianSometimes, plurality and majority voting sys-tems are referred to collectively as majoritarian voting systems because, it is believed, both tend to exaggerate the legislative support of the most popular parties, contributing to single-party majority governments. in prac-tice, results may vary. For instance, French elections do not produce single-party majority governments, despite long use of the majori-tarian double ballot.

MajorityColloquially understood as “50% plus one.” Majority also refers to majority voting sys-tems, which include the alternative vote and the double ballot.

Majority governmentin parliamentary systems, a majority govern-ment is one that commands a majority of votes in the legislature and can thus be assured of passing its legislation. in majoritarian systems, such legislative majorities are typically held by a single party and are often premised on a minority of the popular vote. in proportional systems, by contrast, majority governments are typically composed of a coalition of dif-ferent parties that also represent a majority of the popular vote.

MandateThe view that a party that gains a legislative majority through an election has been autho-rized by the voters to introduce the policies that it promoted during the election. This idea of what elections sanction political actors to do with power is controversial in any setting, but comes under particular criticism in majori-tarian systems because the party that wins a legislative majority and forms the government almost always represents only a minority of

the voters. By contrast, there are cases in pro-portional systems where a coalition of parties makes a pre-election pact to govern together if elected and subsequently gains a majority of the votes between them. Arguably, the latter case has more claim to a mandate from the voters, if the mandate theory can be sustained at all.

Mechanical/psychological effectsCoined by political scientist Maurice Duverger, these terms describe the ways in which differ-ent voting rules influence election results and the behaviour of voters and political actors. “Mechanical effects” refer to the structural ef-fects of voting rules on political competition—specifically, how those rules either limit or facilitate party competition. “Psychological effects” refer to how the perception of these mechanical effects by parties and voters then influences their behaviour. For instance, plu-rality voting mechanically limits party com-petition. This contributes to the psychological effects that political elites are wary of forming new parties, and that voters are less likely to vote for such parties for fear of “wasting” their vote.

Median voterThe notion of the median voter draws insight from economic theory, which suggests that the most influential market actor is the consumer whose preferences are the least fixed, precisely because companies will court her to increase their market share. in politics, the equivalent actor is the “median” voter, who is in the cen-tre of the political spectrum and who is thus more likely to switch his support at election time between right and left. Voters who are solidly right or left (like consumers who are loyal to their brand) are essentially “captive” to their respective political parties; the latter can only gain further advantage (more votes, or more “market share”) by expanding their appeal to median voters. ironically, this com-petitive bid to gain the allegiance of the un-committed median voter leads to less choice, as both left and right parties converge on the centre of the political spectrum.


Minority governmentin parliamentary systems, a minority govern-ment is one that does not command a majority of votes in the legislature and thus cannot be assured of passing its legislation. Both ma-joritarian and proportional voting systems have produced minority governments, but the conditions under which they exist are very different in each system. in majoritarian sys-tems, minority governments tend to be unstable and short lived, as governing parties and/or opposition parties may precipitate an election when they perceive that a slight shift in popular support may allow them to secure a legislative majority government (though one still typical-ly based on a minority of votes). in PR systems, by contrast, minority governments tend to be much more stable and last longer, as it is much less likely that an election will produce an in-flated result for the leading vote-getter or turn minority popular support for one party into a legislative majority for that party.

Mixed-member proportional (MMP)Developed in West Germany after World War ii, this hybrid form of PR combines single-member districts elected by plurality with a compensatory party list that is utilized to ensure that the overall election results are propor-tional for the parties. Voters have two votes, one for the local plurality contest and another for their party choice, with the latter totals forming the basis of the parties’ claims for their proportional share of the total seats (for a more detailed explanation, see page 21 in Chapter 2).

Multi-member ridingA district where more than one member will be elected to represent the area.

Parallel voting systemThis hybrid semi-proportional voting system typically combines single-member ridings elected by plurality with a party-list form of PR. Voters have two votes, one for the local plurality contest and another for their party choice. However, unlike MMP, there is no necessary relationship between the results

produced by the two votes. What this means is that there are really two elections running “parallel” to one another. As a result, this sys-tem tends to overrepresent the larger parties, though the existence of a party-list compon-ent offers some measure of representation to smaller parties.

Party disciplineThis tradition in parliamentary forms of government operates where the legislative members of a political party agree to vote to-gether on issues in order to maximize their influence. Critics hold that party discipline is a negative feature of the political system, as it forces individual representatives to vote in certain ways—sometimes against the wishes of the voters in the locality they represent. But this ignores the fact that nearly all successful political candidates agree to run for office as representatives of a party, and, as voters make their voting decision primarily on the basis of party label, their election is due more to party factors (like their party’s policies on issues) than any local consensus on issues.

Party-list PRA form of PR where voters cast their ballot for a party rather than individual candidates and the results are typically highly propor-tional. Some party-list PR systems do allow voters some influence in choosing the partic-ular members to be elected.

PluralityThe voting formula where the top vote-getter in a pool of candidates is declared the winner, regardless of the proportion of the overall vote their total represents.

Populist/populismA popular description of anti-politician and anti-party sentiment as it translates into politi-cal activity. While populism often promotes a more active citizenry and condemns elite-driven political processes, it often does so at the expense of an appreciation of genuine political differences and the need for a con-crete means to organize and represent those

g lo s s a ry / 185

differences. in some forms, populism amounts to a denial of politics and the need for politi-cal deliberation and compromise altogether.

Preferential ballotAny ballot that allows voters to mark an ordi-nal preference (e.g., “1, 2, 3,” etc.) over the range of candidates. Examples include the majori-tarian AV and the proportional STV.

Proportional representation (PR)A generic term for any voting system that roughly matches the proportion of legislative representation achieved by different political parties with their proportion of the popular vote. Examples include party-list PR, MMP, and STV.

Quotain proportional voting systems, successful candidates gain election by securing a quota of the total votes cast. in a purely proportional party-list system with 100 seats to be elected, the quota would be 1% of the total votes. Thus a party gaining 10% of the total votes would have achieved 10 quotas and would thus elect 10 members (for a more detailed explanation, see page 51 in Chapter 4).

Responsible governmentin parliamentary forms of government—par-ticularly those with British roots—the notion of responsible government refers to the ne-cessity of a government to be sustained in the legislatures through votes of confidence. if a government cannot survive such a vote, it must either resign and be replaced or call an election.

Riding/constituency/districtThese terms are interchangeable and refer to the geographic space within which a member or members will be elected to represent the voters in that locale.

Semi-proportional voting systemsA disparate group of voting systems that can ensure a degree of minority representation. Examples include CV, lV, Parallel systems, and SnTV.

Single-member ridingA district where only one member will be elected to represent the area.

Single non-transferable vote (SNTV)This semi-proportional voting system is essen-tially a more extreme version of the limited vote. in SnTV, voters have one vote in a multi-member riding. Strategically, SnTV offers voters who support minority parties a chance to gain some representation; however, it still overrepresents larger parties (for a more detailed explanation, see page 23 in Chapter 2).

Single transferable vote (STV)A form of PR where voters mark their prefer-ences (i.e., 1, 2, 3, and so on) over the range of candidates in a multi-member riding and can-didates who obtain the quota are declared elected. The ballot counting proceeds by stag-es, beginning with an initial count of the first choices marked on all the ballots. if in this first stage a candidate exceeds the quota, she is declared elected and any surplus votes above the quota are redistributed to other candi-dates on the basis of the second preferences marked on her ballots. This process continues until no more candidates meet the quota. Then, the candidate with the fewest first choic-es is eliminated and his votes are redistributed on the basis of the second choices marked on his ballots. This process continues until some-one obtains the quota. These two aspects of the vote-counting process—distributing surpluses or eliminating low vote-getters—continue until all the available positions are filled (for a more detailed explanation, see page 22 in Chapter 2).

Social choiceAn approach to the study of voting that draws from economic theory. Social-choice theorists are interested in the logical processes involved in making decisions. While nuanced in its theoretical exploration of the possible com-plexities of voting, social choice has proven less useful in explaining political competition and what goes on in real elections.


Strategic votingColloquially, this term refers to the practice of voting for the candidate considered to be the most likely winner rather than for the candi-date that the voter most prefers. While all collective decision making arguably involves some strategic considerations, the concern with “strategic voting” is that voters face a negative incentive in making their choice (“don’t waste your vote”) that is shaped large-ly by the design of the voting system, not by a genuine strategic dilemma (i.e., the competing wants of the voter). Critics feel that strategic-voting incentives are particularly strong in majoritarian voting systems.

ThresholdThe threshold for election refers to the mini-mum amount of support that a political force must gain to win their first seat. in a purely proportional system with 100 seats, the thresh-old would be 1% of the votes cast. in some cases, explicit thresholds of exclusion exist to limit the election of political forces with low levels of support. Thus Germany has a 5% threshold that any aspiring party must sur-pass to gain a seat from the party-list portion of the voting system. in most PR systems the threshold is fixed, but in majoritarian systems they shift depending on the number of com-petitive parties in a district. Majoritarian thresholds are generally much higher than those in PR systems.

Voting formulaThe rule that sets out how votes should be added up to determine the winner or winners in an election. The basic formulas include plurality, majority, and proportional.

Voting system/electoral systemThese terms gain their meaning largely by conventional usage and are often used inter-changeably to refer to many aspects of elec-tions and electoral process. in this volume, “voting system” is used to designate the rules that determine how votes are translated into representation in an election. By contrast, “electoral system” is used to refer to the broader set of election rules (including such things as campaign finance, voter registration methods, and so on), of which the voting sys-tem is only one part.

Wasted voteA colloquial term used to designate a validly cast ballot that nonetheless does not end up contributing to the election of a representa-tive. The term is applied to majoritarian voting systems, whose “all-or-nothing” decision rule for winning a seat typically means that a ma-jority or a considerable minority of the votes are not cast for the winner. This is contrasted to proportional voting systems, where only a marginal percentage of votes do not contrib-ute to the election of a representative.

187

Works Cited

Amer, Mildred l. 2006. “Women in the United States Congress: 1917–2006.” CRS Report for Congress. Washington, DC: Congres-sional Research Service, library of Congress.

Amy, Douglas. 1993. Real Choices, New Voices: The Case for Proportional Representation in the United States. new York: Columbia University Press.

Anderson, Cameron, and Elizabeth Goodyear-Grant. 2005. “Conceptions of Political Representation in Canada: An Explanation of Public opinion.” Canadian Journal of Political Science 38, no. 4 (December): 1029–58.

Anderson, Doris. 2005. “opinion: STV is Just Plain Dumb.” The Globe and Mail, May 6.

Anderson, Mitch. 2004. “B.C. Vote Reform Could lock in Right Wing.” The Tyee, August 10.

Anderson, Mitch. 2005. “let’s Keep Vote Reform Alive.” The Tyee, May 26.

Andeweg, Rudy B. 2006. “Towards a Stronger Parliament? Electoral Engineering of Executive–legislative Relations.” Acta Politica 41: 232–248.

Arrow, Kenneth. 1963. Social Choice and Individual Values. new York: Wiley.

Aucoin, Peter, and Jennifer Smith. 1997. “Proportional Representation: Misrepre-senting Equality.” Policy Options 18, no. 9 (november): 30–3.

Australian Parliament. 2007. “The number of Women in Parliament.” Parliamentary library. http://www.aph.gov.au/library/parl/hist/noswomen.htm (accessed January 25, 2007).

Banducci, Susan A., and Jeffrey A. Karp. 1999. “Perceptions of Fairness and Support for Proportional Representation.” Political Behavior 21, no. 3: 217–38.

Barber, Kathleen. 1995. Proportional Represen-tation and Election Reform in Ohio. Columbus: ohio State University Press.

Bashevkin, Sylvia. 1993. Toeing the Lines: Women and Party Politics in English Canada. Toronto: oxford University Press.

BBC. 2006. “Profile: British national Party.” BBC news. http://news.bbc.co.uk/2/hi/uk _news/politics/3896747.stm (accessed March 3, 2007).

BC-CA (British Columbia Citizens’ Assembly). 2004. “A Preliminary Statement to the People of British Columbia.” April.

Beck, J. Murray. 1963. “The Democratic Process at Work in Canadian Elections.” in The Political Process in Canada, edited by J.H. Aitchison, 36–63. Toronto: University of Toronto Press.

Beers, David, ed. 2005. Liberalized. Vancouver: new Star.

Berger, Thomas. 2004. “A City of neighbour-hoods: Report of the 2004 Vancouver


Electoral Reform Commission.” Vancouver Electoral Reform Commission.

Beyme, Klaus von. 2000. Parliamentary Democracy: Democratization, Destabilization, Reconsolidation, 1789–1999. Houndsmill, UK: Palgrave.

Bickerton, James P., and Alain G. Gagnon, eds. 1994. Canadian Politics, 2nd Edition. Peterborough, on: Broadview Press.

Black, Jerome, and lynda Erickson. 2006. “Ethno-Racial origins of Candidate and Electoral Performance: Evidence from Canada.” Party Politics 12, no. 4: 541–61.

Black, Jerome, and Bruce Hicks. 2006a. “Strengthening Canadian Democracy: The Views of Parliamentary Candidates.” IRPP Policy Matters 7, no. 2 (March): 1–44.

Black, Jerome, and Bruce Hicks. 2006b. “Visible Minority Candidates in the 2004 Election.” Canadian Parliamentary Review 28, no. 2 (Summer): 26–31.

Blais, André. 1988. “The Classification of Electoral Systems.” European Journal of Political Research 16: 99–110.

Blais, André. 1991. “The Debate over Electoral Systems.” International Political Science Review 12, no. 3 (July): 239–60.

Blais, André. 2002. “Why is There So little Strategic Voting in Canadian Plurality Rule Elections?” Political Studies 50: 445–54.

Blais, André, and Kees Aarts. 2006. “Electoral Systems and Turnout.” Acta Politica 41: 180–96.

Blais, André, and Marc André Bodet. 2006. “How Do Voters Form Expectations About the Parties’ Chances of Winning the Election?” Social Science Quarterly 87, no. 3 (September): 477–93.

Blais, André, and Richard nadeau. 1996. “Measuring Strategic Voting: A Two-Step Procedure.” Electoral Studies 15, no. 1: 39–52.

Blais, André, and Mathieu Turgeon. 2004. “How Good Are Voters at Sorting out the Weakest Candidate in Their Constituency?” Electoral Studies 23: 455–61.

Blais, André, R. Kenneth Carty, and Patrick Fournier. 2007. “Do Citizens’ Assemblies Make Reasoned Choices?” in Designing Deliberative Democracy: The British Columbia Citizens’ Assembly, edited by M. Warren and H. Pearse. Cambridge: Cambridge University Press, forthcoming.

Blais, André, Elizabeth Gidengil, Agneiszka Dobrzynska, neil nevitte, and Richard nadeau. 2003. “Does the local Candidate Matter? Candidate Effects in the Canadian Election of 2000.” Canadian Journal of Political Science 36, no. 3 (July–August): 657–64.

Blais, André, Elisabeth Gidengil, Richard nadeau, and neil nevitte. 2002. Anatomy of a Liberal Victory: Making Sense of the Vote in the 2000 Canadian Election. Peterborough, on: Broadview.

Blais, André, Richard nadeau, Elisabeth Gidengil, and neil nevitte. 2001. “Measur-ing Strategic Voting in Multiparty Plurality Elections.” Electoral Studies 20: 343–52.

Blizzard, Christina. 2007. “Voting Against new Election Rules.” Toronto Sun, April 8.

Blyth, Mark. 2003. “Globalization and the limits of Democratic Choice. Social Democracy and the Rise of Political Cartelization.” International Politics and Society 3: 60–82.

Boix, Carlos. 1999. “Setting the Rules of the Game: The Choice of Electoral Systems in Advanced Democracies.” American Political Science Review 93, no. 3: 609–24.

Boston, Jonathan, Stephen Church, and Tim Bale. 2003. “The impact of Proportional Representation on Government Effective-ness: The new Zealand Experience.” Australian Journal of Public Administration 62, no. 4 (December): 7–22.

Bowler, Shaun, and David M. Farrell. 1991. “Voter Behaviour Under STV-PR: Solving the Puzzle of the irish Party System.” Political Behavior 13, no. 4: 305–20.

Bowler, Shaun, and David J. lanoue. 1992. “Strategic and Protest Voting for Third Parties: The Case of the Canadian nDP.”

wo r k s c i t e d / 189

The Western Political Quarterly 45, no. 2 (June): 485–99.

Bowler, Shaun, Todd Donovan, and Dave Brockington. 2003. Electoral Reform and Minority Representation: Local Experiments with Alternative Elections. Washington, DC: Center for Voting and Democracy.

Boyer, Patrick. 1992. Direct Democracy in Canada. Toronto: Dundurn Press.

Brady, Alexander. 1958. Democracy in the Dominions. Toronto: University of Toronto Press.

Bricker, Darrell, and Martin Redfern. 2001. “Canadian Perspectives on the Voting System.” Policy Options 22, no. 6 (July–August): 22–4.

Broadbent, Ed. 2005. 38th Parliament, 1st Session, Standing Committee on Procedure and House Affairs. February 8.

Brockington, David. 2004. “The Paradox of Proportional Representation: The Effect of Party Systems and Coalitions on individuals’ Electoral Participation.” Political Studies 52: 469–90.

Brockington, David, Todd Donovan, Shaun Bowler, and Robert Brischetto. 1998. “Minority Representation Under Cumula-tive and limited Voting.” The Journal of Politics 60, no. 4 (november): 1108–25.

Brett, Judith. 2002. “Class, Religion and the Foundation of the Australian Party System: A Revisionist interpretation.” Australian Journal of Political Science 37, no. 1: 39–56.

Brown, Bethany, ilia Horsburgh, and Emily Klem. 2005. “Content Analysis of the Framing of the BC-STV Coverage in the Media.” Unpublished undergraduate paper, Simon Fraser Communications Depart-ment, August 2: 56 pp.

Brug, Wouter van, and Meindert Fennema. 2003. “Protest or Mainstream? How the European Anti-immigrant Parties Devel-oped into Two Separate Groups by 1991.” European Journal of Political Research 42: 55–76.

Buchanan, James M. 1954. “Social Choice, Democracy, and Free Markets.” Journal of Political Economy 62, no. 2 (April): 114–23.

Bull, Martin, and Martin Rhodes. 1997. “Between Crisis and Transition: italian Politics in the 1990s.” West European Politics 20, no. 1 (January): 1–13.

Busteed, M. A. 1990. Voting Behaviour in the Republic of Ireland. oxford: Clarendon Press.

Bystydzienski, Jill M. 1994. “norway: Achieving World-Record Women’s Representation in Government.” in Electoral Systems in Comparative Perspective: Their Impact on Women and Minorities, edited by Wilma Rule and Joseph F. Zimmerman, 55–64. Westport, CT: Greenwood Press.

Cairns, Alan C. 1968. “The Electoral System and the Party System in Canada, 1921–1965.” Canadian Journal of Political Science 1, no. 1 (March): 55–80.

Campbell, Murray. 2006. “Election Shows That First-Past-the-Post Delivers Queen’s Park.” The Globe and Mail, January 26.

Campbell, Murray. 2007. “Proportional Representation: We need to Know McGuinty’s Stand on Election.” The Globe and Mail, April 17, A10.

Carter, Elizabeth. 2002. “Proportional Representation and the Fortunes of Right-Wing Extremist Parties.” West European Politics 25, no. 3 (July): 125–46.

Carty, R. K. 1981. Party and Parish Pump: Electoral Politics in Ireland. Waterloo, on: Wilfrid laurier University Press.

Carty, R. K. 1997. “Electoral Reform Should Begin with the Senate.” Policy Options 18, no. 9 (november): 39–41.

Carty, R. K., and M. Eagles. 2005. Politics Is Local: National Politics at the Grassroots. Toronto: oxford University Press.

Caul, Miki. 1999. “Women’s Representation in Parliament: The Role of Political Parties.” Party Politics 5, no. 1: 79–98.

Centre for the Advancement of Women in Parliament (ireland). 2007. “Elections.”


Centre for the Advancement of Women in Parliament (ireland). http://www.qub.ac .uk/cawp/election.html (accessed May 25, 2007).

Centre for Research and information on Canada (CRiC). 2001. “Voter Participation in Canada: is Democracy in Crisis?” The CRIC Papers, 3 (october): 1-45.

Chandler, William M., and Alan Siaroff. 1991. “Parties and Party Government in Advanced Democracies.” in Canadian Political Parties: Leaders, Candidates and Organiza-tion, edited by H. Bakvis, 191–264. Toronto: Dundurn Press.

Childs, Sarah. 2000. “The new labour Women MPs in the 1997 British Parliament: issues of Recruitment and Representation.” Women’s History Review 9, no. 1: 55–73.

Childs, Sarah, Joni lovenduski, and Rosie Campbell. 2005. Women at the Top 2005: Changing Numbers, Changing Politics? london: Hansard Society.

Clarke, Harold D., Jane Jenson, lawrence leduc, and Jon H. Pammett. 1980. Political Choice in Canada, Abridged Edition. Toronto: McGraw-Hill Ryerson.

Clarkson, Stephen. 2005. The Big Red Machine: How the Liberal Party Dominates Canada. Vancouver: UBC Press.

Clemens, Jason, and neils Veldhuis. 2005. “The Costs of Electoral Reform.” Fraser Forum (February): 13–8.

Cliche, Paul, and Jean-Pierre Charbonneau. 2006. “Reform of the Voting System: How to Put an End to This Saga.” Le Devoir, December 19.

Coleman, Karen. 1988. “The Politics of Abortion in Australia: Freedom, Church and State.” Feminist Review 29 (Summer): 75–97.

Colomer, Josep. 2004. The Handbook of Electoral System Choice. Houndsmill, UK: Palgrave.

Compas. 2007. “Reform of Democratic institutions: institution Most needing

Repair—The Senate and not the Electoral System, Media, or Parties; Most important Reform Goal—Honesty, Efficiency, lower Taxes and not More Public Participation.” BDo Dunwoody Weekly CEo/Business leader Poll, January 19.

Cool, Julie. 2006. “Women in Parliament.” ottawa: library of Parliament.

Cooper, Barry. 2005. “The Historical Experience with Proportional Voting Systems in Canada.” Fraser Forum (February): 5–7.

Corry, J. A. 1951. Democratic Government and Politics. Toronto: University of Toronto Press.

Cox, Gary. 1997. Making Votes Count: Strategic Coordination in the World’s Electoral Systems. Cambridge: Cambridge University Press.

Courtney, John. 1999. “Electoral Reform and Canada’s Parties.” in Making Every Vote Count: Reasssessing Canada’s Electoral System, edited by H. Milner, 91–100. Peterborough, on: Broadview Press.

Courtney, John. 2001. Commissioned Ridings: Designing Canada’s Electoral Districts. Montreal and Kingston, on: McGill-Queen’s University Press.

Courtney, John. 2004. Elections. Vancouver: UBC Press.

Coyne, Andrew. 2007. “The national Post is Wrong on PR.” National Post, April 18.

Crewe, ivor. 1985. “MPs and Their Constitu-ents in Britain: How Strong Are the links?” in Representatives of the People? Parliamen-tarians and Constituents in Western Democracies, edited by V. Bogdanor, 44–65. Aldershot, UK: Gower.

Cross, William. 2004. Political Parties. Vancouver: UBC Press.

Curtin, Jennifer. 1999. “The 1998 Tasmanian Election: Women and Proportional Representation.” Australian Parliamentary Library Research Note 5. Canberra: Parliamentary library, Parliament of Australia.


Curtin, Jennifer. 2006. “Advancing Women’s interests in Formal Politics: The Politics of Presence and Proportional Representation in the Antipodes.” in The Politics of Women’s Interests: New Comparative Perspectives, edited by l. Chappell and l. Hill, 93–110. london: Routledge.

Cutler, Fred, and J. Scott Matthews. 2005. “The Challenge of Municipal Voting: Vancouver 2002.” Canadian Journal of Political Science 38, no. 2 (June): 359–82.

Cutler, Fred, and Richard Johnston, with R. Kenneth Carty, André Blais, and Patrick Fournier. 2007. “Deliberation, informa-tion, and Trust: The BC Citizens’ Assembly as Agenda Setter.” in Designing Deliberative Democracy: The British Columbia Citizens’ Assembly, edited by M. Warren and H. Pearse. Cambridge: Cambridge University Press, forthcoming.

Denley, Randall. 2007. “Don’t Scrap Election System: First Past the Post Has its Advantages.” The Ottawa Citizen, April 3.

Derriennic, Jean-Pierre. 2005. “Three Dimensions of Justice for Evaluating Electoral Systems.” in Strengthening Canadian Democracy, edited by P. Howe, R. Johnston, and A. Blais, 107–14. Montreal: iRPP.

Diskin, Hanna, and Abraham Diskin. 1995. “The Politics of Electoral Reform in israel.” International Political Science Review 16, no. 1: 31–450

Dobell, Peter. 2000. “What Could Canadians Expect from Minority Government?” IRPP Policy Matters 1, no. 6 (november): 1–20.

Docherty, David. 1997. Mr. Smith Goes to Ottawa: Life in the House of Commons. Vancouver: UBC Press.

Docherty, David. 2005. Legislatures. Vancou-ver: UBC Press.

Donovan, Mark. 1995. “The Politics of Electoral Reform in italy.” International Political Science Review 16, no. 1: 47–64.

Doron, Gideon, and Richard Kronick. 1977. “Single Transferable Vote: An Example of a

Perverse Social Choice Function.” American Journal of Political Science 22, no. 2 (May): 303–11.

Drummond, Andrew. 2002. “Electoral Volatility and Party Decline in Western Democracies: 1970–1995.” Series paper 02/02. Center for the Study of Democracy, University of California (irvine). http://repositories.cdlib.org/csd/02-02/ (accessed September 30, 2005).

Dummett, Michael. 1984. Voting Procedures. oxford: Clarendon Press.

Dutch Electoral System Civic Forum. 2006. “The Electoral System Civic Forum’s Recommendations.” Dutch Electoral System Civic Forum. http://www .nederlandpraatmee.nl/nieuws.php ?idnode=9&idCnt=574#n574 (accessed March 20, 2007).

Duverger, Maurice. 1954. Political Parties. Reprint, new York: John Wiley, 1963.

Dyck, Rand. 2007. Canadian Politics: Critical Approaches, 5th Edition. Toronto: Thomson nelson.

Elections ACT. 2007. “2004 ACT legislative Assembly election.” ACT Government House. http://www.elections.act.gov.au/Result04.html#liST (accessed February 20, 2007).

Elections Canada. 2007. A History of the Vote in Canada, 2nd Edition. ottawa: office of the Chief Electoral officer.

Elections new Zealand. 2007. “Sainte-laguë Allocation Formula.” Elections new Zealand. http://www.elections.org.nz/mmp/sainte _lague.html (accessed April 1, 2007).

Eley, Geoff. 2002. Forging Democracy. oxford: oxford University Press.

Equal Voice. 2005. “letter to Hon. Marie Bountrogianni, Minister of intergovern-mental Affairs, Minister Responsible for Democratic Renewal.” Equal Voice. Submitted october 20. http://www .equalvoice.ca/news_marie_102005.htm (accessed March 3, 2007).


Equal Voice. 2007. “The Facts Ma’am: Women in Federal and Provincial legislatures and Courts 2005.” Equal Voice. http://www .equalvoice.ca/facts_legislature.htm (accessed March 3, 2007).

Fabbrini, Sergio, and Mark F. Gilbert. 2000. “When Cartels Fail: The Role of the Political Class in the italian Democratic Transition.” Government and Opposition 35, no. 1: 27–48.

Fair Vote Canada. 2005. Statement of Purpose. Toronto: Fair Vote Canada: 2 pp.

Fair Vote Canada. 2006. Dubious Democracy: Report on Federal Elections in Canada from 1980–2006. Toronto: Fair Vote Canada.

Farrell, David. 2001. Electoral Systems: A Comparative Introduction. Houndsmill, UK: Palgrave.

Farrell, David, and Michael Gallagher. 1999. “British Voters and the Criteria for Evaluating Voting Systems.” British Journal of Politics and International Relations 1, no. 3 (october): 293–316.

Farrell, David, and ian McAllister. 2003. “Voter Satisfaction and Electoral Systems: Does Preferential Voting in Candidate-Centred Systems Make a Difference?” Paper presented to the Australasian Political Studies Association Conference, University of Tasmania, Hobart, September 29 to october 1.

Farrell, David, and ian McAllister. 2005. “1902 and the origins of Preferential Electoral Systems in Australia.” Australian Journal of Politics and History 51, no. 2: 155–67.

Femia, Joseph V. 2001. Against the Masses: Varieties of Anti-Democratic Thought Since the French Revolution. oxford: oxford University Press.

Ferejohn, John. 2007. “The Citizens’ Assembly Model.” in Designing Deliberative Democracy: The British Columbia Citizens’ Assembly, edited by M. Warren and H. Pearse. Cambridge: Cambridge University Press, forthcoming.

Flanagan, Tom. 1999. “The Alternative Vote: An Electoral System for Canada.” in Making Every Vote Count: Reassessing Canada’s Electoral System, edited by H. Milner, 85–90. Peterborough, on: Broadview Press.

Flanagan, Tom. 2005. “The Alternative Vote.” in Strengthening Canadian Democracy, edited by P. Howe, R. Johnston, and A. Blais, 99–106. Montreal: iRPP.

Forsey, Eugene. [1964] 1989. “The Problem of Minority Government in Canada.” in The Canadian Political Tradition: Basic Readings, edited by R.S. Blair and J.T. Mcleod, 314–23. Scarborough, on: nelson.

Forsey, Eugene. 1985. “Parliamentary Reform is More than Mechanics.” in The Canadian House of Commons: Essays in Honour of Norman Ward, edited by J. Courtney, 195–200. Calgary: University of Calgary Press.

Franklin, Mark n. 2004. Voter Turnout and the Dynamics of Electoral Competition in Established Democracies Since 1945. Cambridge: Cambridge University Press.

Franko, Roman. 1992. “Towards liberal Democracy in ontario: The Franchise and Policy-Making 1868–1888.” PhD disserta-tion, Queen’s University.

Franks, C.E.S. 1987. The Parliament of Canada. Toronto: University of Toronto Press.

Furlong, Paul. 1994. Modern Italy: Representa-tion and Reform. london: Routledge.

Galatas, Steven E. 2004. “Electing the First Parliament: Party Competition and Voter Participation in Scotland.” Party Politics 10, no. 2: 213–33.

Gallagher, Michael. 1986. “The Political Consequences of the Electoral Systems in the Republic of ireland.” Electoral Studies 5, no. 3: 253–75.

Gallagher, Michael. 2000. “The (Relatively) Victorious incumbents Under PR-STV.” in Elections in Australia, Ireland and Malta Under the Single Transferable Vote, edited by S. Bowler and B. Grofman, 81–113. Ann Arbor: University of Michigan Press.


Gallagher, Michael, and Paul Mitchell, eds. 2005. The Politics of Electoral Systems. oxford: oxford University Press.

Georgia Straight. 2005. “Editorial: STV Cure Worsens Disease.” Georgia Straight, May 12–19, 15.

Gerber, Alan S., and Donald P. Green. 2000. “The Effects of Canvassing, Telephone Calls and Direct Mail on Voter Turnout: A Field Experiment.” American Political Science Review 94, no. 3 (September): 653–63.

Geys, Benny. 2006. “Explaining Voter Turnout: A Review of Aggregate-level Research.” Electoral Studies 25: 637–63.

Gibbins, Roger, and Jennifer Stewart. 2002. “Wither or Whither the Populist impulse in Western Canada.” in Political Parties, Representation, and Electoral Democracy in Canada, edited by W. Cross, 132–44. Toronto: oxford University Press.

Gidengil, Elizabeth, André Blais, neil nevitte, and Richard nadeau. 2004. Citizens. Vancouver: UBC Press.

Girvin, Brian. 1996. “Church, State and the irish Constitution: The Secularisation of irish Politics?” Parliamentary Affairs 49, no. 4 (october): 599–615.

Gordon, larry. 2004. “Canada’s Electoral Victims: They’re not Who You Think They Are.” Toronto: Fair Vote Canada: 2 pp.

Gray, Mark, and Miki Caul. 2000. “Declining Voter Turnout in Advanced industrial Democracies, 1950 to 1997: The Effects of Declining Group Mobilization.” Compara-tive Political Studies 33, no. 9 (november): 1091–1122.

Gwyn, Richard. 2006. “Voting System Gives Us What We Want.” Toronto Star, January 20.

Hahn, Harlan. 1968. “Voting in Canadian Communities: A Taxonomy of Referendum issues.” Canadian Journal of Political Science 1, no. 4 (December): 462–9.

Hart, Jenifer. 1992. Proportional Representa-tion: Critics of the British Electoral System

1820–1945. oxford: oxford University Press.

Heinisch, Reinhard. 2003. “Success in opposition—Failure in Government: Explaining the Performance of Right-Wing Populist Parties in Public office.” West European Politics 26, no. 3 (July): 91–130.

Hoag, Clarence, and George Hallett. 1926. Proportional Representation. new York: Macmillan.

Howe, Paul. 2003. “Where Have All the Voters Gone?” Inroads 12 (Winter/Spring): 74–82.

Howe, Paul, and David northrup. 2000. “Strengthening Canadian Democracy: The Views of Canadians.” IRPP Policy Matters 1, no. 5: 1–104.

Hughes, Colin. 2000. “STV in Australia.” in Elections in Australia, Ireland and Malta Under the Single Transferable Vote, edited by S. Bowler and B. Grofman, 155–77. Ann Arbor: University of Michigan Press.

ibbitson, John. 2005. “Stonewalling on Electoral Reform.” The Globe and Mail, June 17.

iDEA. 2007. “Parliamentary Elections.” iDEA. http://www.idea.int/vt/parl.cfm (accessed March 25, 2007).

international Parliamentary Union. 2005. “Women in national Parliaments.” international Parliamentary Union. http://www.ipu.org/wmn-e/world.htm (accessed January 15, 2005).

irvine, William. 1979. Does Canada Need a New Electoral System? Kingston: institute of inter governmental Relations, Queen’s University.

irvine, William. 1982. “Does the Candidate Make a Difference? The Macro-Politics and Micro-Politics of Getting Elected.” Canadian Journal of Political Science 15, no. 4 (December): 755–82.

irvine, William. 1985. “A Review and Evaluation of Electoral System Reform Proposals.” in Institutional Reforms for Representative Government, edited by


P. Aucoin, 71–110. Toronto: University of Toronto Press.

Jackman, Robert W., and Karin Volpert. 1996. “Conditions Favouring Parties of the Extreme Right in Western Europe.” British Journal of Political Science 26, no. 4 (october): 501–21.

Jackson, Keith, and Alan McRobie. 1998. New Zealand Adopts Proportional Representation: Accident? Design? Evolution? Aldershot, UK: Ashgate.

James, Michael Rabinder. 2007. “Descriptive Representation in the BC Citizens’ Assembly.” in Designing Deliberative Democracy: The British Columbia Citizens’ Assembly, edited by M. Warren and H. Pearse. Cambridge: Cambridge University Press, forthcoming.

Jansen, Harold. 1998. “The Single Transferable Vote in Alberta and Manitoba.” PhD dissertation, University of Alberta.

Jansen, Harold. 2004. “The Political Conse-quences of the Alternative Vote: lessons from Western Canada.” Canadian Journal of Political Science 37, no. 3: 647–69.

Jansen, Harold, and Alan Siaroff. 2004. “Regionalism and Party Systems: Evaluat-ing Proposals to Reform Canada’s Electoral System.” in Steps Toward Making Every Vote Count: Electoral System Reform in Canada and Its Provinces, edited by H. Milner, 43–63. Peterborough, on: Broadview Press.

Jenson, Jane. 1997. “out of Proportion.” Canadian Forum 75, no. 857 (March): 27–9.

Jenson, Jane, and Rianne Mahon. 1993. “Representing Solidarity: Class, Gender and the Crisis in Social-Democratic Sweden.” New Left Review 197 (January–February): 76–100.

Kam, Christopher. 1997. “PR: A Political Shibboleth.” Policy Options 18, no. 9 (november): 29–30.

Karp, Jeffrey. 2002. “Members of Parliament and Representation.” in Proportional Representation on Trial: The 1999 New Zealand General Election and the Fate of

MMP, edited by J. Vowles, P. Aimer, J. Karp, R. Miller, and A. Sullivan, 130–46. Auckland: Auckland University Press.

Katz, Richard. 1997. Democracy and Elections. oxford: oxford University Press.

Kenworthy, lane, and Melissa Malami. 1999. “Gender inequality in Political Representa-tion: A Worldwide Comparative Analysis.” Social Forces 78, no. 1: 235–68.

Kingsley, Jean Pierre. 2001. “Speech to Fair Vote Canada Annual General Meeting, March 31, 2001.” Elections Canada on-line/Media: Statements and Speeches. http://www.elections.ca/content.asp?section=med&document=mar3101a&dir=spe&lang=e &textonly=false (accessed March 13, 2007).

Kittilson, Miki Caul. 2006. Challenging Parties, Changing Parliaments: Women and Elected Office in Contemporary Western Europe. Columbus: ohio State University Press.

Klingemann, Hans-Deiter, and Bernhard Wessels. 2001. “The Political Consequences of Germany’s Mixed-Member System: Personalization of the Grass Roots?” in Mixed-Member Electoral Systems: The Best of Both Worlds? edited by M. Shugart and M. Wattenberg, 279–96. oxford: oxford University Press.

Knight, Trevor. 1999. “Unconstitutional Democracy? A Charter Challenge to Canada’s Electoral System.” University of Toronto Faculty of Law Review 57, no. 1 (Winter): 1–41.

Know STV website. knowstv.ca (accessed April 5, 2007).

Kolinksy, Eva. 1991. “Political Participation and Parliamentary Careers: Women’s Quotas in West Germany.” West European Politics 14, no. 1: 56–72.

Kromkowski, Charles, and John Kromkowski. 1991. “Why 435? A Question of Political Arithmetic.” Polity 24, no. 1 (Fall): 129–45.

lakeman, Enid. 1982. Power to Elect. london: Heinemann.

lane, John C. 1995. “The Election of Women Under Proportional Representation: The


Case of Malta.” Democratization 2, no. 2 (Summer): 140–57.

lang, Amy. 2007a. “Agenda Setting in Deliberative Forums: Expert influence and Citizen Autonomy.” in Designing Delibera-tive Democracy, edited by M. Warren and H. Pearse. Cambridge: Cambridge University Press, forthcoming.

lang, Amy. 2007b. “But is it for Real? The British Columbia Citizens’ Assembly as a Model of Politics of State-Sponsored Citizen Empowerment?” Politics and Society 35, no. 1 (March): 35–69.

law Commission of Canada (lCC). 2002. Renewing Democracy: Debating Electoral Reform in Canada—Discussion Paper. ottawa: lCC.

law Commission of Canada (lCC). 2004. Voting Counts: Electoral Reform for Canada. ottawa: lCC.

laycock, David. 1990. Populism and Democratic Thought in the Canadian Prairies, 1910–1945. Toronto: University of Toronto Press.

laycock, David. 1994. “Reforming Canadian Democracy? institutions and ideology in the Reform Party Project.” Canadian Journal of Political Science 27, no. 2 (June): 213–47.

leduc, lawrence. 2003. The Politics of Direct Democracy. Peterborough, on: Broadview Press.

leone, Rob. 2006–07. “letter to the Editor: Keep Democracy out of the Court.” Canadian Parliamentary Review 29, no. 4 (Winter): 57–8.

levin, Michael. 1992. The Spectre of Democracy: The Rise of Modern Democracy as Seen by Its Critics. new York: new York University Press.

levine, Stephen. 2004. “Parliamentary Democracy in new Zealand.” Parliamentary Affairs 57, no. 3: 646–65.

lijphart, Arend. 1994. Electoral Systems and Party Systems: A Study of Twenty-Seven Democracies, 1945–1990. oxford: oxford University Press.

lijphart, Arend. 1997. Democracy in Plural Societies. new Haven, CT: Yale University Press.

lijphart, Arend. 1999a. Patterns of Democracy. new Haven, CT: Yale University Press.

lijphart, Arend. 1999b. “Australian Democracy: Modifying Majoritarianism?” Australian Journal of Political Science 34, no. 3: 313–26.

lipset, S.M., and S. Rokkan. 1967. “Cleavage Structures, Party Systems and Voter Alignments: An introduction.” in Party Systems and Voter Alignments, edited by S.M. lipset and S. Rokkan, 1–64. new York: Free Press.

liu, James H., and Chris G. Sibley. 2006. “Differential Effects of Social Anchoring and Attitude Certainty in Determining Support or opposition to (Bi)cultural Diversity in new Zealand.” Papers on Social Representations 15: 1.1–1.15.

loenen, nick. 1997. Citizenship and Democ-racy: A Case for Proportional Representation. Toronto: Dundurn Press.

lortie, Pierre. 1997. “A Minimalist Electoral Reform Agenda.” Policy Options 18, no. 9 (november): 22–5.

lovenduski, Joni, and Pippa norris, eds. 1993. Gender and Party Politics. london: Sage.

lovink, J. 1970. “on Analysing the impact of the Electoral System on the Party System in Canada.” Canadian Journal of Political Science 3, no. 4 (December): 497–516.

lundberg, Thomas Carl. 2005. “Second-Class Representatives? Mixed-Member Propor-tional Representation in Britain.” Parliamen-tary Affairs 59, no. 1 (november): 60–77.

lupia, Arthur, and Kaare Strom. 1995. “Coalition Termination and the Strategic Timing of Parliamentary Elections.” American Political Science Review 89, no. 3: 648–65.

lusztig, Michael. 1994. “Constitutional Paralysis: Why Canadian Constitutional initiatives Are Doomed to Fail.” Canadian Journal of Political Science 27, no. 4 (December): 747–71.

MacDonald, Jack. 2005. Randomocracy. Victoria: FCG Publications.


Macivor, Heather. 1996. Women and Politics in Canada. Peterborough, on: Broadview Press.

Macivor, Heather. 1999. “A Brief introduction to Electoral Reform.” in Making Every Vote Count: Reassessing Canada’s Electoral System, edited by H. Milner, 19–36. Peterborough, on: Broadview Press.

Macivor, Heather. 2003. “Women and the Canadian Electoral System.” in Women and Electoral Politics in Canada, edited by M. Tremblay and l. Trimble, 22–36. Toronto: oxford University Press.

Macivor, Heather. 2006. “The Case for a Charter Challenge to Canada’s Electoral System.” Paper presented to the 2006 Annual Meeting of Fair Vote Canada, ottawa, April 21.

Mackie, Gerry. 2003. Democracy Defended. Cambridge: Cambridge University Press.

Macleod, Peter. 2007. “learning to love ontario’s Citizens’ Assembly.” The Globe and Mail, January 23, online comment.

Mair, Peter. 1991. “The Electoral Universe of Small Parties in Western Europe.” in Small Parties in Western Europe, edited by F. Muller-Rommel and G. Pridham, 41–70. london: Sage Publications.

Mair, Peter. 1992. “The Question of Electoral Reform.” New Left Review 194 (July/August): 75–97.

Maltadata. 2007. “Women’s Share of Candidacies and legislative Seats.” Maltadata. http://www.maltadata.com/wom-runs.htm (accessed March 10, 2007).

Marsh, Michael. 2003. “Party Wrapped and Candidate Centred: new Evidence on Electoral Behaviour Under STV.” Paper presented at the annual meeting of the American Political Science Association, Philadelphia Marriott Hotel, Philadelphia, PA, August 27.

Marsh, Michael. 2007. “Candidates or Parties? objects of Electoral Choice in ireland.” Party Politics 13, no. 4: 501–28.

Martin, lanny W., and Georg Vanberg. 2004. “Policing the Bargain: Coalition Govern-

ment and Parliamentary Scrutiny.” American Journal of Political Science 48, no. 1: 13–27.

Massetti, Emanuele. 2006. “Electoral Reform in italy: From PR to Mixed System and (Almost) Back Again.” Representation 42, no. 3: 262–9.

Massicotte, louis. 1999. “Parliament in the 1990s.” in Canadian Politics, 3rd Edition, edited by J. Bickerton and A. Gagnon, 164–84. Peterborough, on: Broadview Press.

Massicotte, louis. 2005. “Changing the Canadian Electoral System.” in Strengthen-ing Canadian Democracy, edited by P. Howe, R. Johnston, and A. Blais, 65–98. Montreal: iRPP.

Massicotte, louis. 2006. “Electoral Reform in Canada and its Provinces.” Paper presented at the Plurality and Multi-round Elections Conference, Montreal, June 17–18.

Massicotte, louis, and André Blais. 1999. “Mixed Electoral Systems: A Conceptual and Empirical Survey.” Electoral Studies 18: 341–66.

Matland, Richard E. 1993. “institutional Variables Affecting Female Representation in national legislatures: The Case of norway.” The Journal of Politics 55, no. 3: 737–55.

Matland, Richard E. 2005. “Enhancing Women’s Political Participation: legislative Recruitment and Electoral Systems.” in Women in Parliament: Beyond Numbers, Revised Edition, edited by J. Ballington and A. Karam, 93–111. Stockholm: iDEA.

Matland, Richard E., and Donley T. Studlar. 1996. “The Contagion of Women Candidates in Single-Member District and Proportional Representation Electoral Systems: Canada and norway.” The Journal of Politics 58, no. 3: 707–33.

McKee, Paul. 1983. “ireland.” in Democracy and Elections: Electoral Systems and their Political Consequences, edited by V. Bogdanor and D. Butler,167–89. Cam-bridge: Cambridge University Press.


McKenna, Peter. 2006. “opting out of Electoral Reform: Why PEi Chose the Status Quo.” Policy Options 27, no. 5 (June): 58–61.

McKinnon, Ken. 2005. “Final Report: Electoral Reform.” Government of Yukon report. February 1: 23 pp.

Mendelsohn, Matthew, and Andrew Parkin. 2005. “Getting from Here to There: A Process for Electoral Reform in Canada.” in Strengthening Canadian Democracy, edited by P. Howe, R. Johnston, and A. Blais, 137–48. Montreal: iRPP.

Merolla, Jennifer l., and laura B. Stephenson. 2007. “Strategic Voting in Canada: A Cross Time Analysis.” Electoral Studies 26, no. 2: 235–46.

Milner, Henry. 1997. “Electoral Systems, integrated institutions and Turnout in local and national Elections: Canada in Comparative Perspective.” Canadian Journal of Political Science 30, no. 1: 89–106.

Milner, Henry, ed. 1999. Making Every Vote Count: Reassessing Canada’s Electoral System. Peterborough, on: Broadview Press.

Milner, Henry. 1999. “The Case for Propor-tional Representation in Canada.” in Making Every Vote Count: Reassessing Canada’s Electoral System, edited by H. Milner, 37–50. Peterborough, on: Broadview Press.

Milner, Henry. 2002. Civic Literacy: How Informed Citizens Make Democracy Work. Hanover, nH: University Press of new England.

Milner, Henry. 2004a. Steps Toward Making Every Vote Count: Electoral System Reform in Canada and Its Provinces. Peterborough, on: Broadview Press.

Milner, Henry. 2004b. “First Past the Post? Progress Report on Electoral Reform initiatives in Canadian Provinces.” IRPP Policy Matters 5, no. 9 (September): 1–36.

Milner, Henry. 2005. “Electoral Reform and Deliberative Democracy in British Colum-bia.” National Civic Review (Spring): 3–8.

Morgan, Ed. 2000. “Green Party Factum.” Test Case Centre, University of Toronto law Faculty. http://www.law-lib.utoronto.ca/testcase/#1 (accessed April 7, 2007).

Morgenstern, Scott, and Stephen Swindle. 2005. “Are Politics local? An Analysis of Voting Patterns in 23 Democracies.” Comparative Political Studies 38, no. 2 (March): 143–70.

Muller-Rommel, Ferdinand. 1991. “Small Parties in Comparative Perspective: The State of the Art.” in Small Parties in Western Europe, edited by F. Muller-Rommel and G. Pridham, 1–22. london: Sage Publications.

nagel, Jack H. 1994. “What Political Scientists Can learn from the 1993 Electoral Reform in new Zealand.” PS: Political Science and Politics 27, no. 3 (September): 525–9.

nagel, Jack H. 2004. “Stormy Passage to Safe Harbour? Proportional Representation in new Zealand.” in Steps Toward Making Every Vote Count: Electoral System Reform in Canada and Its Provinces, edited by H. Milner, 119–44. Peterborough, on: Broadview Press.

nesbitt-larking, Paul. 2001. Politics, Society and the Media: Canadian Perspectives. Peterborough, on: Broadview Press.

nevitte, neil, André Blais, Elizabeth Gidengil, and Richard nadeau. 2000. Unsteady State: The 1997 Canadian Federal Election. Toronto: oxford University Press.

newland, Robert A. 1982. Comparative Electoral Systems. london: Arthur McDougall Fund.

norris, Pippa. 1995. “The Politics of Electoral Reform in Britain.” International Political Science Review 16, no. 1: 65–78.

norris, Pippa. 2004. Electoral Engineering: Voting Rules and Political Behavior. Cambridge: Cambridge University Press.

Palmer, Vaughn. 2005. “other liberals not as High as Campbell on STV.” Vancouver Sun, May 11, A3.


Pammett, John, and larry leduc. 2003. “Explaining the Turnout Decline in Canadian Federal Elections: A new Survey of non-voters.” ottawa: Elections Canada: 74 pp.

Pearse, Hilary. 2005 “Geographic Representa-tion and Electoral Reform.” Canadian Parliamentary Review, Autumn, 28–32.

Pearse, Hilary. 2007. “institutional Design and Citizen Deliberation.” in Designing Deliberative Democracy: The British Columbia Citizens’ Assembly, edited by M. Warren and H. Pearse. Cambridge. Cambridge University Press, forthcoming.

Petit, Philip. 2004. “Depoliticizing Democ-racy.” Ratio Juris 17, no. 1: 52–65.

Pharr, Susan, Robert Putnam, and Russell Dalton. 2000. “Trouble in the Advanced Democracies: A Quarter Century of Declining Confidence.” Journal of Democracy 11, no. 2 (April): 5–25.

Phillips, Harry. 1976. “Challenges to the Voting System in Canada, 1874–1974.” PhD dissertation, University of Western ontario.

Pilon, Dennis. 1999. “The History of Voting System Reform in Canada.” in Making Every Vote Count: Reassessing Canada’s Electoral System, edited by H. Milner, 111–22. Peterborough, on: Broadview Press.

Pilon, Dennis. 2001. Canada’s Democratic Deficit: Is Proportional Representation the Answer? Toronto: CSJ Foundation for Research and Education.

Pilon, Dennis. 2002. Renewing Canadian Democracy: Citizen Engagement in Voting System Reform. Phase One: Lessons from Around the World. ottawa: law Commis-sion of Canada.

Pilon, Dennis. 2004. “The Uncertain Path of Democratic Renewal in ontario.” in Steps Toward Making Every Vote Count: Electoral System Reform in Canada and Its Provinces, edited by H. Milner, 249–66. Peterborough, on: Broadview Press.

Pilon, Dennis. 2005. “Why Voting Systems Change: Electoral Reform in Western

industrialized Countries.” PhD disserta-tion, York University.

Pilon, Dennis. 2006. “Explaining Voting System Reform in Canada: 1874 to 1960.” Journal of Canadian Studies 40, no. 3 (Autumn): 135–61.

Pilon, Dennis. 2007. “BC-STV and the Representation of Women: Popular Debate, Academic insights, and the Evidence from Comparative Experience—A Research note.” Paper presented to the BC Political Studies Association, Capilano College, north Vancouver, BC, May 4–5.

Pilon, Dennis, and David Docherty. 2006. “local Members and the Size of Canadian legislatures.” Paper presented to the Canadian Political Science Association, York University, Toronto, June 1–3.

Potvin, Kevin. 2005. “A Voting Scheme Made for the Rural Right.” The Tyee, March 3.

Powell, G. Bingham. 2000. Elections as Instruments of Democracy: Majoritarian and Proportional Visions. new Haven, CT: Yale University Press.

Pravda, Alex. 1978. “Elections in Communist Party States.” in Elections Without Choice, edited by G. Hermet, R. Rose, and A. Rouquie, 169–95. new York: John Wiley.

Proctor, J. H. 1980. “The Acceptance of Proportional Representation in Malta.” Parliamentary Affairs 33, no. 3 (Summer): 308–21.

Pulzer, Peter. 1983. “Germany.” in Democracy and Elections, edited by V. Bogdanor and D. Butler, 84–109. new York: Cambridge University Press.

Qualter, T. H. 1970. The Election Process in Canada. Toronto: McGraw-Hill.

Radaelli, Claudio M., and Fabio Franchino. 2004. “Analysing Political Change in italy.” Journal of European Public Policy 11, no. 6 (December): 941–53.

Rae, Douglas. 1971. The Political Consequences of Electoral Laws. new Haven, CT: Yale University Press.


Rahat, Gideon, and Reuven Y. Hazan. 2005. “israel: The Politics of an Extreme Electoral System.” in The Politics of Electoral Systems, edited by M. Gallagher and P. Mitchell, 333–73. oxford: oxford University Press.

Ratner, R. S. 2007. “Communicative Rational-ity in the Citizens’ Assembly and Referen-dum Process.” in Designing Deliberative Democracy: The British Columbia Citizens’ Assembly, edited by M. Warren and H. Pearse. Cambridge: Cambridge University Press, forthcoming.

Redekop, John H. 2005. “Commentary: An Assessment of the Citizen’s Assembly on Electoral Reform.” BC Studies 147 (Autumn): 89–102.

Reeve, Andrew, and Alan Ware. 1992. Electoral Systems: A Comparative and Theoretical Introduction. london: Routledge.

Reynolds, Andrew. 1999. “Women in the legislatures and Executives of the World: Knocking at the Highest Glass Ceiling.” World Politics 1, no. 4: 547–72.

Richie, Robert. 1997. “improving new York City’s Community School Board Elec-tions.” Testimony to the Citywide Community School Board Elections Committee on December 2. Center for Voting and Democracy Report. http://fairvote.org (accessed June 1, 2000).

Riker, William. 1982. Liberalism Against Populism: A Confrontation between the Theory of Democracy and the Theory of Social Choice. San Francisco: W.H. Freeman.

Rokkan, Stein. 1970. Citizens, Elections, Parties: Approaches to the Comparative Study of the Processes of Development. new York: David McKay.

Roper, Jon. 1989. Democracy and Its Critics: Anglo-American Democratic Thought in the Nineteenth Century. london: Unwin Hyman.

Roper, Juliet, Christina Holtz-Bacha, and Gianpietro Mazzoleni. 2004. The Politics of Representation: Election Campaigning and Proportional Representation. new York: lang.

Rose, Richard, and ian McAllister. 1992. “Expressive Versus instrumental Voting.” in Electoral Politics, edited by D. Kavanagh, 114–40. oxford: Clarendon Press.

Rueschemeyer, D., E. Huber, and J.D. Stephens. 1992. Capitalist Development and Democracy. Urbana, il: University of Chicago Press.

Ruff, norman J. 2004. “Electoral Reform and Deliberative Democracy: The British Columbia Citizens’ Assembly.” in Steps Toward Making Every Vote Count: Electoral Reform in Canada and Its Provinces, edited by H. Milner, 235–48. Peterborough, on: Broadview Press.

Rule, Wilma. 1987. “Electoral Systems, Contextual Factors and Women’s opportunity for Election to Parliament in Twenty-Three Democracies.” The Western Political Quarterly 40, no. 3: 477–98.

Rustow, Dankwart A. 1950. “Some observations on Proportional Representation.” Journal of Politics 12, no. 1 (February): 107–27.

Rydon, Joan. 1994. “Representation of Women and Ethnic Minorities in the Parliaments of Australia and new Zealand.” in Electoral Systems in Comparative Perspective: Their Impact on Women and Minorities, edited by W. Rule and J. Zimmerman, 223–32. Westport, CT: Greenwood Press.

Sainsbury, Diane. 2004. “Women’s Political Representation in Sweden: Discursive Politics and institutional Presence.” Scandinavian Political Studies 27, no. 1: 65–87.

Sassoon, Donald. 1995. “Tangentopoli or the Democratization of Corruption: Consider-ations on the End of italy’s First Republic.” Journal of Modern Italian Studies 1, no. 1 (Fall): 124–43.

Scarrow, Susan. 2004. “Explaining Political Finance Reforms: Competition and Context.” Party Politics 10, no. 6: 653–75.

Scarrow, Susan. 2006. “Party Subsidies and the Freezing of Party Competition: Do Cartel Mechanisms Work?” West European Politics 29, no. 4 (September): 614–39.


Seidle, leslie. 1996. “The Canadian Electoral System and Proposals for Reform.” in Canadian Parties in Transition, 2nd Edition, edited by A. Brian Tanguay and Alain-G. Gagnon, 282–306. Toronto: nelson Canada.

Seidle, leslie. 2002. “Electoral System Reform in Canada: objectives, Advocacy and implications for Governance.” CPRN Discussion Paper F |28. Canadian Policy Research networks. http://www.cprn.com/doc.cfm?doc=162 (accessed March 3, 2007).

Seton-Watson, Christopher. 1983. “italy.” in Democracy and Elections: Electoral Systems and Their Political Consequences, edited by V. Bogdanor and D. Butler, 110–21. Cambridge: Cambridge University Press.

Sharman, Campbell. 1986. “The Senate, Small Parties and the Balance of Power.” Politics 21, no. 2: 20–3.

Shugart, Matthew. 2001. “Extreme Electoral Systems and the Appeal of the Mixed-Member Alternative.” in Mixed-Member Electoral Systems: The Best of Both Worlds? edited by M. Shugart and M. Wattenberg, 25–54. oxford: oxford University Press.

Shugart, Matthew, and Martin Wattenberg. 2001. Mixed-Member Electoral Systems: The Best of Both Worlds? oxford: oxford University Press.

Siaroff, Alan. 2000. “Women’s Representation in legislatures and Cabinets in industrial Democracies.” International Political Science Review 21, no. 2: 197–215.

Siaroff, Alan. 2006. “Provincial Political Data Since 1900.” in Provinces: Canadian Provincial Politics, edited by C. Dunn, 175–214. Peterborough, on: Broadview Press.

Sinha, Tito. 1993. “1993 new York City School Board Elections: Effect of Preference Voting on Asian American Candidates.” Center for Voting and Democracy. http://fairvote.org/reports/1993/sinha.html (accessed April 2, 2007).

Skjeie, Hege, and Birte Siim. 2000. “Scandina-vian Feminist Debates on Citizenship.” International Political Science Review 21, no. 4: 345–60.

Smith, Patrick J., and Kennedy Stewart. 1998. “Making local Accountability Work in British Columbia, Report 2: Reforming Municipal Electoral Accountability.” Report for the BC Ministry of Municipal Affairs and Housing, June: 55 pp.

Steed, Michael. 1990. “The Constituency.” in Representation and Electoral Systems: Canadian Perspectives, edited by J. Paul Johnston and Harvey E. Pais, 186–200. Scarborough, on: Prentice-Hall Canada.

Stone, Bruce. 1998. “Small Parties and the Senate Revisited: The Consequences of the Enlargement of the Senate in 1984.” Australian Journal of Political Science 33, no. 2: 211–8.

Stout, D.K. 1975. “incomes Policy and the Costs of the Adversary System.” in Adversary Politics and Electoral Reform, edited by S.E. Finer, 117–42. london: Anthony Wigram.

Strom, Kaare. 1990. Minority Government and Majority Rule. Cambridge: Cambridge University Press.

Strom, Kaare. 1997. “Democracy, Accountabil-ity, and Coalition Bargaining.” European Journal of Political Research 31: 47–62.

Swayze, Mikael Antony. 1996. “Continuity and Change in the 1993 Canadian General Election.” Canadian Journal of Political Science 29, no. 3 (September): 555–66.

Tanguay, Brian. 2004. “Reforming Representa-tive Democracy: Taming Canada’s Democratic Deficit.” in Canadian Politics, 4th Edition, edited by J. Bickerton and A. Gagnon, 263–86. Peterborough, on: Broadview Press.

Taras, David. 2001. Power and Betrayal in the Canadian Media. Peterborough, on: Broadview Press.

Tasmanian Parliament library. 2007. “Women Members of the Tasmanian Parliament.” Parliament of Tasmania. http://www .parliament.tas.gov.au/tpl/DataSheets/MP _Women_Terms.htm (accessed March 26, 2007).


Teileman, Bill. 2005. “no: STV is not Truly Proportional and is less Democratic Than FPTP.” CCPA Monitor, April.

Tennant, Paul, and Julian West. 1998. “Half of Vancouver’s Voters Go Without Represen-tation.” Vancouver Sun, September 19, A23.

Tennant, Paul, and Julian West. 1998. “Ward System is no Better Than At-large Democ-racy.” Vancouver Sun, September 23, A17.

Thomas, Alastair. 1985. “Members of Parliament and Access to Politics in Scandinavia.” in Representatives of the People? Parliamentarians and Constituents in Western Democracies, edited by V. Bogdanor, 199–223. Aldershot, UK: Gower.

Thomlinson, neil. 2000. “When Right is Wrong: Municipal Governance and Downloading in Toronto.” in Restructuring and Resistance: Canadian Public Policy in an Age of Global Capitalism, edited by M. Burke, C. Mooers, and J. Shields, 226–61. Halifax: Fernwood.

Thorburn, Hugh, ed. 1985. Party Politics in Canada, 5th Edition. Scarborough, on: Prentice-Hall.

Tindal, C. Richard, and Susan nobes Tindal. 1995. Local Government in Canada, 4th Edition. Toronto: McGraw-Hill Ryerson.

Toronto Star Editorial Board. 2007. “Bad Electoral Medicine.” Toronto Star, February 23.

Trimble, linda, and Jane Arscott. 2003. Still Counting: Women in Politics Across Canada. Peterborough, on: Broadview Press.

Uhr, John. 1999. “The Senate and Proportional Representation: Public Policy Justifications of Minority Representation.” Australian national University Public Policy Program Discussion Paper no. 69, September.

Urquhart, ian. 2007. “Radical Voting Proposal Gains Steam ontario Citizens’ Assembly Endorses ‘Mixed Member Proportional’ Plan.” Toronto Star, February 22.

Vowles, Jack. 1995. “The Politics of Electoral Reform in new Zealand.” International Political Science Review 16, no. 1: 95–115.

Vowles, Jack. 2002. “offsetting the PR Effect: Party Mobilization and Turnout Decline, 1996–99.” Party Politics 8, no. 5: 587–605.

Vowles, Jack, Susan A. Banducci, and Jeffrey A. Karp. 2006. “Forecasting and Evaluating the Consequences of Electoral Change in new Zealand.” Acta Politica 41: 267–84.

Vowles, Jack, Ann Sullivan, Raymond Miller, Peter Aimer, Susan Banducci, and Jeffery Karp. 2000. “Electoral System opinion and the Evolution of MMP: A Report to the Electoral Commission from the new Zealand Election Study Team.” July 28: 56 pp.

Watling, Judy. 2006. “Values-Driven Electoral Reform—Conference Summary.” Canadian Policy Research networks. http://www .cprn.com/doc.cfm?doc=1589&l= (accessed May 15, 2007).

Ward, Doug. 1996. “Vote Results on Ward System ‘Pure idiocy’.” Vancouver Sun, november 18, B7.

Ward, Doug. 2005. “Ex-Premiers Barrett, Bennett Say no to STV.” Vancouver Sun, May 6, B1.

Ward, norman. 1963. The Canadian House of Commons: Representation, 2nd Edition. Toronto: University of Toronto Press.

Warren, Mark E., and Hilary Pearse, eds. Designing Deliberative Democracy: The British Columbia Citizens’ Assembly. Cambridge: Cambridge University Press, forthcoming.

Weaver, leon, and Judith Baum. 1992. “Proportional Representation on new York City School Boards.” in United States Electoral Systems: Their Impact on Women and Minorities, edited by Wilma Rule and Joseph F. Zimmerman, 197–205. Westport, CT: Greenwood Press.

Weaver, R. Kent. 1997. “improving Represen-tation in the House of Commons.” Canadian Journal of Political Science 30, no. 3 (September): 473–512.

White, Graham. 2005. Cabinets and First Ministers. Vancouver: UBC Press.


White, Graham. 2007. Communication with author.

White, Timothy J. 2006. “Why So Few Women in Da_il E_ireann? The Effects of the Single Transferable Vote Election System.” New Hibernia Review/Iris Eireannach Nua 10, no. 4 (Winter): 71–83.

Whyte, nicholas. 2007. “northern ireland Elections; Assembly; By Year 1973–2003.” northern ireland Social and Political Archive. http://www.ark.ac.uk/elections/ (accessed May 12, 2007).

Wilson, T. 1975. “The Economic Costs of the Adversary System.” in Adversary Politics and Electoral Reform, edited by S.E. Finer, 99–116. london: Anthony Wigram.

Wiseman, nelson. 1997. “Skeptical Reflections on Proportional Representation.” Policy Options 18, no. 9 (november): 15–8.

Woldendorp, Jaap, Hans Kenan, and ian Budge. 2000. Party Government in 48 Democracies (1945–1998): Composition, Duration, Personnel. Boston: Kluwer Academic Publishers.

YES for BC STV. 2007. “list of Supporters.” stvforbc.com (accessed May 23, 2007).

Young, lisa. 1997. “Gender Equal legislatures: Evaluating the Proposed nunavut Electoral System.” Canadian Public Policy 23, no. 3 (September): 306–15.

Zahariadis, nikolaos. 1999. “The Rise and Fall of British State ownership: Political Pressure or Economic Reality?” Compara-tive Politics 31, no. 4 (July): 445–63.

203

Aadditional-member system (AMS),

71, 181

alternation in power theory, 43– 4

alternative vote (AV), 181

Atlantic institute for Market Studies, 172

at-large voting, 108, 110, 181

Australia, 20, 70, 75, 76, 125, 130

Austria, 15

Bballot structure, 18, 52, 181

BC Citizens’ Assembly, 6, 90–1, 97–103, 164, 175

BC STV referendum, 97–103, 121, 123, 128, 164, 172–3, 175

Belgium, 84, 135

Bloc Québécois, 14, 33, 35, 37, 46

BnA Act, 107, 115

CCanada First party, 79

Canadian Taxpayers Federation, 173

Charlottetown Accord, 87, 116

citizens’ assembly approach to voting reform, 96 –104, 106, 181

civil society, 172–5

cleavage, 182

coalition effect, 155

complexity, 146

consensual democracy, 77

Conservative party (Alberta), 44, 113

Conservative party (federal), 33, 34, 35–7, 46, 55, 63, 78, 82, 116, 166, 167

Conservative party (new Brunswick), 44

Conservative party (nova Scotia), 113– 4

Conservative party (ontario), 44, 62

contagion effect, 182

index


Co-operative Commonwealth Federation (CCF), 82, 166

cumulative vote (CV), 23, 182

cycling of preferences, 16, 182

D“deep pockets” defence strategy, 105

democratic deficit, 89, 182

democratic passions, 77

districting, 18

double ballot, 182

Droop quota, 173, 175

Eelected dictatorship, 65

election result scenarios, 1–2

electoral reform, see also voting-system reform

inclusions, 2

voting system, and, 2

FFair Vote Canada, 5, 89, 90, 96, 120

Fair Voting BC, 90

federal government, PR relevancy, 114–7

Finland, 84, 113

first past the post, 28, 29, 92

flexible partisans, 5, 182

France, 17, 20, 70, 75, 87–8

Fraser institute, 172, 175

Frontier Centre for Public Policy, 172

GGermany, 15, 22, 50, 53, 54, 67, 84, 124,

130, 135, 151, 152

government accountability

alleged plurality strength, 143– 4

alleged PR weakness, 153– 4

Green party, 33, 35, 36, 46, 55, 71, 98, 102, 104, 113, 123

Gregory allocation method, 173, 175

HHare quota, 173

House of Commons reform, 115– 6

Iindia, 15

ireland, 22, 124, 125, 126, 132, 174

israel, 147–51

italy, 17, 53, 88, 92, 97, 105, 106, 147–51

JJapan, 17, 23, 24, 88, 89, 92

Llaw Commission of Canada, 89, 134

less-than-proportional mixed system, 71

liberal party (Alberta), 44

liberal party (BC), 90, 98, 101, 111

liberal party (federal), 27, 32, 33, 35, 37, 46, 55, 71, 78, 82, 97, 105, 153, 166, 173

liberal party (Manitoba), 80

liberal party (new Brunswick), 44

liberal party (nova Scotia), 113

liberal party (ontario), 103, 112

liberal party (Quebec), 111, 176

limited vote (lV), 22–3, 182

local government, 107–11

i n d e x / 205

local representation

alleged plurality strength, 140–3

alleged PR weakness, 152–3

ideology of, 141–2

importance of, 126 –7

Mmajoritarian voting system, 183

majority, 183

majority government

defined, 183

plurality voting, and, 42– 6, 64, 138– 40

proportional representation, and, 62–5

real versus phony, 64–5

majority voting systems, 20

Malta, 22, 125, 174

mandate, 183

mechanical effects

defined, 183

plurality voting, 46

media, 169–72

median voter, 64, 183

minority government, 184

minority representation, 77, 79

mixed-member proportional (MMP), 6, 21, 71, 91, 168, 184

Mouvement pour une démocratie nouvelle, 90

multi-member plurality

arguments for and against, 120–2

diverse representation, 127–30

local representation, quality of, 126 –7

pro-party versus anti-party, 125– 6

proportionality, 130–2

simplicity versus complexity, 123–5

BC referendum, and, 102

defined, 20

new Zealand ballot, 52

multi-member riding, 184

municipalities, PR relevancy, 107–11

Nnatural justice perspective, 56

new Democratic Party (BC), 44, 62, 98, 101, 102

new Democratic Party (federal), 14, 35, 36, 37, 46, 71, 86, 99, 105, 123, 166, 176

new Democratic Party (Manitoba), 113

new Democratic Party (nova Scotia), 113

new Democratic Party (ontario), 113, 177

new Democratic Party (Quebec), 111

new Democratic Party (Saskatchewan), 113

new York City, 130

new Zealand, 17, 22, 50, 54, 58–9, 67, 71, 88, 89, 92, 96, 105, 106, 110, 125, 127, 132, 150, 154, 155, 156, 162, 163, 164, 165, 167, 172, 177

non-proportional mixed system, 71, 72

northern ireland, 25

norway, 61, 113

Oone person, one vote, 34

ontario Citizens’ Assembly, 91, 102, 103, 169, 170, 175


PPacific Scandal, 78

parallel voting system, 23, 184

parliamentary reform, 72–3

Parti Québécois, 86, 111, 113, 123

party discipline, 184

party list PR, 21, 119, 184

Pepin-Robarts Task Force on Canadian Unity, 86, 122

plurality voting systems

alleged strengths, 137

government accountability, 143– 4

local representation, 140–3

majority government, 138– 40

simplicity, 137–8

stability, 138– 40

Canadian mechanics, 28–30

effects of, 30–2

defined, 184

majority government under, 42– 6

mechanics, 20

party competition, 46

representation, approach to, 33

individual voters, 33–5

parties, 35–9

social diversity, 39– 42

strategic voting, 47–8

political culture, 155– 6

populism, 184

power to elect, 55

preferential ballot, 185

proportional representation (PR)

alleged weaknesses, 145

complexity, 146

instability, 146 –51

lack of government accountability, 153– 4

lack of local representation, 152–3

party dominance, 151–2

alternatives to, 70

non-proportional voting-system reforms, 70–2

parliamentary reform, 72–3

approaches to achieving

Charter challenge, 104, 176

citizen-centred, 96 –104

debate on, 105– 6

politicians, design by, 95– 6

conversion to, western countries, 85

defensible claims, 156 –7

effects of, 51–3

majority government under, 50, 62–5

mechanics, 21

implications of, 50–1

mixed-member, 21

municipalities, and, 107–11

party competition, 65–8

party list, 21, 119

political culture, and, 155– 6

provinces, and, 111– 4

representation, approach to, 53– 4

individual voters, 54– 6

parties, 56 –8, 125–7

social diversity, 58– 61, 127–30

single transferable vote

BC referendum, and, 97–103, 121, 123

i n d e x / 207

defined, 22

versus MMP, 119–33

strategic voting, 68–70

voter turnout, and, 154–5

proportionality indexes, 57

provinces, PR relevancy, 111– 4

Qquota, 51, 185

Rreferendum, 104, 106, 176

representative democracy

plurality, and, 33

proportional representation, and, 53– 4

responsible government, 77, 185

riding, 185

Russia, 24

SScotland, 54, 110, 127

semi-proportional voting systems

cumulative vote, 23

defined, 185

limited vote, 22–3

mechanics, 22

parallel voting, 23– 4

Senate reform, 116 –7

simplicity

alleged plurality strength, 137–8

alleged PR weakness, 146

STV and MMP compared, 123–5

single-member plurality (SMP), 13, 20, 76, 94, 104, 115, 148, 149

single-member riding, 185

single non-transferable vote (SnTV), 23, 185

single transferable vote (STV)

arguments for and against, 120–2

diverse representation, 127–30

local representation, quality of, 126 –7

pro-party versus anti-party, 125– 6

proportionality, 130–2

simplicity versus complexity, 123–5

BC ballot, 52, 90, 97–103, 121

Charlottetown Accord, 87

defined, 6, 22, 185

municipal use in Canada, 22

provincial use in Canada, 22

seat allocation, 51

social-choice theory, 15–7, 25, 185

Social Credit party, 14, 44, 82, 98

South Korea, 23

Spain, 53

stable majority government

alleged plurality strength, 138– 40

alleged PR weakness, 146 –51

italy and israel, 147–51

position on, 45

strategic voting

defined, 186

plurality voting system, and, 47–8

proportional representation, and, 68–70

super-majority threshold rule, 103– 4, 111, 173, 176

Sweden, 61, 84, 113, 135, 151

Switzerland, 135


TTaiwan, 23

threshold, 186

UUnion nationale, 44

United Kingdom, 17, 20, 23, 31, 32, 41, 62, 66 –7, 75, 76, 88, 89, 92, 121, 122, 139, 147

United States, 17, 20, 23, 75, 117, 122, 130, 139

Vvalues, voting systems and, 8–9, 134– 6

Vancouver Board of Trade, 172

visible minorities, 39

voter turnout, 154–5

voting formula, 18, 186

voting system

academic debate re effects of, 12– 4

additional-member system (AMS), 71, 181

Canada, origins of, 76 –82

classification of, 24–5

component parts, 18–9

defined, 2–3, 18, 186

importance of, 12–7

length of governments, 63

less-than-proportional mixed system, 71

majority, 20

mechanical and psychological effects, 14–5, 66, 183

non-proportional mixed system, 71, 72

number of elections, 64

options, 19–24

parties, average number of, 68

performance, debate about, 136

defensible claims, 156 –7

plurality’s alleged strengths, 137– 45

PR’s alleged weaknesses, 145–54

plurality, see plurality voting systems

political culture, effect on, 155– 6

proportional, see proportional representation (PR)

proportionality indexes, 57

semi-proportional, see semi-proportional voting systems

simplicity, importance of, 124–5

social-choice critique of effects, 15–7

values, and, 8–9, 134– 6

voter turnout, effect on, 154–5

western democracies, origins of, 83– 6

voting-system reform, see also electoral reform

Alberta, 80, 82, 86, 95, 113

approaches to achieving, 95

courts, 104

debate over, 105– 6

people, 95–104

politician-centred, 95

BnA Act, and, 115– 6

British Columbia, 86, 90, 91, 95, 97–103, 111, 113, 158

Canada

civil society, and, 172–5

history of, 79–82

i n d e x / 209

media, and, 169–72

political parties, and, 165–8

public, and, 163–5

depoliticization, 4– 6

false alternatives to proportional representation, 70–3

federal level

House of Commons, 115– 6

need for, 114–7

Senate, 116 –7

Manitoba, 80, 82, 86, 95, 113

municipal level, need for, 107–11

new Brunswick, 90, 91, 96, 111, 112, 158

nova Scotia, 113

one true reform, 94

ontario, 79, 91, 111, 112, 158

party self-interest, and, 114

politicization, 6 –9, 159– 62

Prince Edward island, 90, 91, 96, 111, 112, 113, 124, 158

provincial level, need for, 111– 4

Quebec, 90, 91, 111, 113, 158

recent initiatives, Canada, 89

provincial, 90–2

recent reforms, western countries, 87–9

Saskatchewan, 113

Vancouver, 110

Winnipeg, 110

voting theory, 16 –7

Wwasted votes, 33, 34, 46, 47, 54, 61, 186

winnable ridings, 40

women

representation under plurality, 39, 41–2

representation under PR, 58– 61, 122, 128–31, 174

The Politics of Voting: Reforming Canada's Electoral System

Documents

Transcript of The Politics of Voting: Reforming Canada's Electoral System