1 Brief to Canadian House of Commons Special Committee on Electoral Reform
Dr. Dennis Pilon Associate Professor York University
July 23, 2016
Relevant research
My research focuses on the practical workings of, and concrete results produced through, the use of different voting systems across western industrialized countries, with particular attention to the political processes accompanying their introduction and reform. My 2007 book, The Politics of Voting: Reforming Canada’s Electoral Systems, is the only single-authored academic work on the topic focusing on Canada, while my 2013 book, Wrestling with Democracy: Voting Systems as Politics in the Twentieth Century West, compares 18 western countries over 150 years examining every case of national voting system reform. Additionally, I have numerous book chapters and journal articles that examine both Canada’s historical experiences with voting system reform and the use of alternative voting systems as well as the more recent reform efforts across five Canadian provinces. More broadly my work examines the historical struggle to either further or resist the democratization of our political institutions. Recommendation The only option that will advance the five principles set out in the government’s motion creating this committee, namely, effectiveness and legitimacy, engagement, accessibility and inclusiveness, integrity, and local representation, as well as fulfill the government’s pledges to make every vote count and better represent the diversity of Canada, is some form of proportional representation (PR). After consultation, the government should decide which version of PR would best suit Canada and simply introduce this reform in time for its use in the next election. No referendum is warranted or desirable. No constitutional issues represent any serious barrier to proceeding in this way.
Rationale
Since the government reiterated its campaign commitment to make 2015 the last election under our present voting system we have seen three broad ways of talking about the process of voting system reform emerge. One view says that the voting system has a quasi-constitutional status and that it should not should not be altered without either invoking the constitutional change process or, at the very least, have any alternative proposal subjected to a national referendum. Another view suggests that voting systems are simply a matter of values and outcomes and depending on which values one holds and which outcomes one prefers, any voting system is as good as any other – voting system choice here is characterized as a matter of taste.
Finally, the last view insists that voting system reform is intrinsically and inescapably 2 linked to democratic reform – to the extent that an alternative voting system can be shown to produce demonstrably superior democratic outcomes i.e. outcomes that better facilitate what voters are trying to accomplish by voting and better match what voters say with their votes, there is no real debate about change, such a system should simply be introduced. The view that federal voting system is quasi-constitutional and as such requires a constitutional process to change it, or a national referendum to approve any alternative, is wholly without merit, lacking any compelling evidence to support its claims or logical coherency in making its arguments. The choice of, and process for, reforming the federal voting system are neither constitutional issues. The constitution only affects the voting system in that section 51 of the Constitution Act 1867 says that provinces must have seats allotted to them in a manner proportionate to their population, subject to other constitutional strictures like the Senate and grandfather clauses. This means that as a long as any chosen voting system abides by these rules, there is no constitutional barrier to using it. Parliament’s right to choose its own voting system was confirmed by sections 40 and 41 of the original British North America Act of 1867, which basically stated that matters pertaining to the conduct of the elections were matters for Parliament to decide. While the electoral details of these sections have clearly been superceded by other acts and thus are ‘spent’, the over-arching intent of the sections that accord Parliament the power the make election law is clearly still in force. Others have tried to argue that the recent Supreme Court of Canada decision in the Senate reference case means that the provinces may now have an interest in the federal election law (Pal 2016), or that section 51 of the Constitution Act 1867 provides a set number of districts for the provinces and so can only be changed to accommodate a new voting system by changing the constitution (Bowden 2016), or that the use of referenda to change voting systems in Canada has now become a convention (Dutil 2016). These are weak and poorly supported arguments. On the first issue, the Senate reference case addressed a situation where provincial interests in the institution were well established historically and spelled out in the constitution, which is not the case for the voting system.
On the second concern, section 51 does not constitutionally entrench a specific number of districts but rather uses a formula to establish the number of seats a province should have, without stating how they should be organized into districts. The section and its amendments over the years have scrupulously avoided mentioning districting, and it should be remembered that a few dual member ridings were used federally from 1867 to 1966, which had no impact on the constitutionally determined number of seats allotted to the provinces in which they were used. Finally, on the last concern, attempts to argue that the recent use of referenda for voting system change decisions in Canadian provinces means that the practice has become a kind of convention for voting system change at the federal level is simply ridiculous. If following a policy more than three times made it a convention there would be very little room for government to legislate. Conventions do not emerge in this manner. If history matters, then the Canadian ‘convention’ for voting system reform is for governments to simply legislate new voting systems into existence. Historically, Canada has had ten successful voting system reforms at the provincial level, none of which involved referenda.
More importantly from the point of view of seeing conventions as normatively important practices, research shows that the reasons these various governments 3 turned to referenda recently in Canada and abroad was not because such governments were committed to voter participation in the process as a public good but that they thought a referendum represented the best way to defeat the reforms while not appearing to publicly oppose them. Setting constitutional issues aside, attempts to argue that the government must submit any reform proposal to a referendum – indeed, that it would be undemocratic for the government to act without such public involvement – are specious and logically inconsistent.
Referenda are blunt instruments, best suited to narrow policy issues that the public stands a reasonable chance to become informed about (e.g. municipal recreation issues). The use of referenda in Canada for voting system reform since 2005 demonstrates the many weaknesses of going this route. Academic study of these processes has highlighted government manipulation of the process, media indifference and/or bias in the public deliberation over the issues, and an inability of the public to cope with the complexity of the issue. Indeed, academic survey research during and after both the BC and Ontario referenda demonstrated most voters were unaware a referendum was going on, let alone what it was about. In Ontario, researchers discovered that a majority responded positively to the proposal for a voting system that would retain a local member and increase proportionality of the party outcomes, they just didn’t know that was what the referendum was offering them – clearly a perverse outcome (Cutler and Fournier 2007; Carty, Cutler and Fournier 2009).
Repeatedly through these referenda we have seen that neither the public nor the media were able to become informed in a way that would facilitate their participation in deliberating over the details of this issue. Cynics might argue that this is precisely why anti-reform forces are championing a referendum – because they feel confident that such a process would guarantee defeat. But what research shows is that such a defeat would not be the product of a reasoned rejection of the merits of the proposed alternative voting system by an informed public, but rather a frustrated rejection of the whole process of consultation by a public with low information on the issue, a very typical coping mechanism used by voters facing too much issue complexity. The referenda advocates can also be challenged for their inconsistent interest in and application of a majority rule. They insist that changing the voting system must be approved by at least a majority of Canadian voters to be legitimate but Canadian governments and all the decisions they make are defended as legitimate even though they seldom have the support of a majority of Canadians.
Why is a majority required in one instance but not the other? If getting majorities for policy decisions is the hallmark of democratic legitimacy, and that is the gist of the referenda advocates’ argument, then surely all governing decisions should be subjected to such strictures. And the best way to assure that is to shift to a voting system that would require legislative majority governments to actually represent a majority of voting Canadians. The selective application of the majority principle by referenda advocates to the question of voting system reform alone is logically inconsistent and appears, frankly, hypocritical. The view that choosing amongst voting systems involves decisions about competing values and preferred outcomes suggests that the choice of a voting system is primarily a matter of taste. If you like having a local member and prefer there to be legislative majority governments in office, then first-past-past meets your needs. If you think that every vote should count and that a party’s legislative representation should mirror its popular support, then a form of PR is closer to your values and preferred outcomes. The problem with this approach is that it does not distinguish amongst the various possible values and outcomes – and it should. Our democratic institutions should not be designed on some relativistic set of values but should be judged against how well they facilitate what Canadians are trying to do when they participate in elections. Here we can use what we know from the academic study of voters and the reasons informing their vote choices to focus our discussion of institutional design. For instance, we know from a considerable body of research that party choice is the key factor influencing how people vote, not the specific attributes of local candidates. Elections outcomes themselves confirm this – local candidates devoid of party affiliation are seldom elected in Canadian national elections (for a longer discussion of the research on voter choice and the relevant academic sources, see Appendix I).
As such, it makes sense that our electoral institution facilitate Canadians getting the party representation that they want.
At the same time, values that do not meet democratic standards should be ruled out from the start – in other words, choosing undemocratic options should not be an option. For instance, I may prefer there to be a single party legislative majority government, but if a majority of my fellow Canadians are not prepared to settle on one party then what I am really saying is that I want to be able impose a legislative majority on them against their expressed democratic wishes, a dubious democratic outcome. Many of the values associated with our existing firstpast-the-post are patently undemocratic, namely that it: (1) tends to produce single party legislative majority governments supported by only a minority of the voting population, (2) restricts party competition,
(3) inaccurately translates voting support into party representation,
(4) often leaves over the half of the votes cast not contributing the election of anyone, and
(5) limits the representation of social diversity.
First-past-the-post fails to adequately satisfy any of the government’s five criteria for an effective voting system. Allowing people to ‘choose’ it would make a mockery of the reform process and break the government’s campaign promise. Those that support the values approach are fond of saying that there is no perfect voting system, but that doesn’t mean there aren’t some that are highly imperfect ones like first-past-the-post, especially from a democratic point of view.
The last view presents voting system reform as a matter of urgent democratic necessity, based on observable facts about what Canadians are trying to do by voting in elections. If an alternative voting system can be shown to produce demonstrably better democratic results than our present voting system by more accurately reflecting what voters say with their votes in terms of legislative representation, then government should simply introduce it. To do so would be to continue the more than century-long process of attempting to reform Canada’s political institutions to be more substantively democratic. At Confederation, the first past the post voting system was part of an elite governing system that was explicitly not democratic. Over the following decades various reforms – the secret ballot, the extension of the franchise, the codification of voter registration regulations, etc. – were introduced, often for reasons of short-term partisan gain. Despite this, the long-term effect of such changes was to further enhance the democratic substance of the electoral process.
Yet the voting system has mostly eluded reform, for reasons that have everything to do with partisan self-interest. Because the voting system is the key aperture through which voting demands must pass, our two major national parties have seen fit to keep the opening as narrow as possible so that only they could squeeze through. The public, on the other hand, confront the voting system as an intimate stranger: despite meeting up at every election, most have little knowledge about its workings or are even aware that any alternatives to it exist.
Public and political indifference to the voting system has been giving way over the past two decades, culminating in this most recent federal election result where parties representing a clear majority of Canadian voters campaigned on a promise to change Canada’s voting system if elected.
But what should replace our present system? The choice should offer Canadians the clearest path to getting what they want out of voting. We know that Canadians participate in national elections using party label to inform their vote choice. Yet whether Canadians get the representation they prefer depends less on the popularity of their party choice than the geography of where other like-minded voters are located. In other words, our present voting system discriminates in favour of geographically proximate groups of voters and against geographically dispersed groups of voters (this is dealt with in more detail in Appendix I).
The geographic bias in our voting system cannot be defended in terms of voter preferences because we know people are not making their vote choice on the basis of purely local issues or personalities. Thus voting system reform should address this representational deficit, which is rooted in how our present system inhibits voter equality and limits party competition. Regardless of where people live, their vote should contribute to the election of their preferred party choice, if there are enough like-minded voters in the country to make common cause with. First-past-the-post cannot do that and neither can majority voting systems like the alternative vote or the French double ballot (for a more detailed treatment of why majority voting systems will not meet the government’s stated criteria for reform, see Appendix II).
Only some form of PR can equalize individual voting power. And it should be noted that unequal voting power affects the supporters of all parties, depending on where they live. Thus proportional voting is not about benefiting this or that party but about empowering individual voters, regardless of what they wish to vote for.
At this point, critics of reform often begin rehearsing a litany of crises that they allege will flow from the adoption of a proportional voting system, including but not limited to: governing instability, party fragmentation, undue influence from fringe parties, the election of extremists, voter confusion, a lack of governing accountability, and so on.
Space and time considerations limit a fulsome response but, suffice to say, such claims typically suffer from rather serious logical, empirical and methodological shortcomings (these issues are dealt with in more detail in Appendix III).
First, they often include examples from countries whose political culture and history are not really comparable to Canada (e.g. Israel and Italy).
Second, critics are often selective in their use of examples, essentially cherry-picking examples to make their case.
Third, a great deal of critique operates at the level of sheer speculation where opponents creatively dream up terrible consequences that could occur if PR were introduced. There is a proper way to do comparative research on voting systems, but it requires one to come up with comparisons that are appropriate i.e. with countries that share a broadly similar process of political and economic development as Canada, and the researcher must be systematic in examining the phenomenon rather than just seeking data that confirms what they want to believe.
The first two views of our present voting system reform process presented here are seriously flawed and uninformed. Only the third view can be supported by evidence of what we know voters are trying to do by voting. Voting system reform understood in this way gives the government a clear rationale and moral imperative to act in the best of interests of Canadians to improve their democracy by introducing a new voting system.
Conclusion
Canada’s voting system is an unreformed holdover from a pre-democratic era, kept in place because it served the partisan interests of those with the power to keep it there. But now that both public and political party attention has focused on its consistently undemocratic outcomes, it is time to abolish it in favour of something that will more effectively and accurately translate voter intentions into representation, and do so in a way that meets the criteria the government has set out for reform.
Only some form of proportional representation can address the concerns that have given rise to this reform initiative. Only PR can make every vote count, equalize individual Canadians’ voting power, accurately represent the party choices Canadians make with their votes, better the representation of diversity, and assure that governments must reach out to create a broad coalition of support to pass legislation. Such a change neither requires nor warrants a national referendum to pass. Indeed, voter equality is a first order principle that should not be put to a referendum. The reform of the voting system is also not constitutional and could easily be introduced by government as simple legislation. Finally, the choice of a voting system is not simply a matter of taste but a test of the state’s willingness to recast its institutions to further the democratic will. The choice should be based on how best to realize and reflect what Canadians communicate through the electoral process.
7 Appendices
Appendix I: Excerpt from Dennis Pilon, Brief for Quebec Voting System Case, December 2008.
Appendix II: Dennis Pilon, “Electoral Reform: Here’s Your Evidence Mr. Trudeau,” Inroads, June 2015. Appendix III: Excerpt from Dennis Pilon, The Politics of Voting: Reforming Canada’s Electoral System, (Toronto: Emond Montgomery, 2007), chapter 8, “Debating Voting System Effects.” Citations James Bowden, “Time has already run out on electoral reform,” Inside Policy: The Magazine of the Macdonald-Laurier Institute, June 8, 2016. R. Kenneth Carty, Fred Cutler, and Patrick Fournier, “Who killed BC-STV?” The Tyee, July 8, 2009. Fred Cutler and Patrick Fournier, “Why Ontarians said no to MMP,” Globe and Mail, October 25, 2007. Patrice Dutil, “Without a referendum, electoral reform is unconstitutional,” Globe and Mail, June 23, 2016. Michael Pal, “Why Canada’s top court must weigh in on electoral reform,” Globe and Mail, January 15, 2016.
Comments