Prospects for Electoral Reform in Canada

Lessons from Quebec's Near-Adoption of PR in 1984        

Henry Milner

        Throughout democratic countries, electoral reform has been under some considerable discussion in recent years, with Japan, Italy and New Zealand instituting major changes. In Canada, the 1993 federal election exposed the distortions built into its winner-take-all voting system based on legislators elected by plurality from single-member districts.
        In the election, Canadians clearly wanted to show the ruling Progressive Conservatives that they had lost confidence in them, and the party won only 16% of the popular vote. However, the workings of the voting system turned a show of non-confidence into a massacre. Rather than electing 46 of 295 members that a proportional system would have provided, the Tories elected only two. By contrast, two regionally-based parties, the Bloc Quebecois and Reform, with 13% and 19% of the popular vote respectively, elected 54 and 52 Members. The voting system also turned the victorious Liberals' 41% of the vote into a very solid majority of 177 seats.
        Despite several such results at a federal and provincial level in recent years, there is no perceptible clamor for electoral reform. And while there is much dissatisfaction with the representativeness of Canadian legislators, the link to the electoral system is seldom drawn. More typically, and especially in the 1993 campaign rhetoric, party discipline was blamed for the perceived unaccountability of the elected politicians.
        Recall has been proposed as a solution by the program of the fledgling Reform Party, which came second in the popular vote in the 1993 federal election; indeed, according to a March 1993 Gallup Pool, recall is endorsed by almost 80 percent of Canadians. The closest PR came to finding its way to the Canadian national political agenda was in a proposal by the Pepin-Robats Commission in the late 1970s calling for a mixed electoral system with 60 (of 342) MPs to be elected from party lists.
        This recommendation, like the rest of the Report, fell on deaf ears. Indeed, when the Mulroney government created the Lortie Commission on electoral reform and party financing in 1990, the mandate was narrowly interpreted from the outset, so changing the electoral system was expressly excluded as a non-starter.

Quebec: A Broken Promise to Adopt PR

        As far as the provinces are concerned, the situation has been no different -- except in Quebec, where the appearance of the Parti Quebecois (PQ) at the end of the 1960s changed the equation. A mass party committed to Quebec independence, the PQ could not ignore the injustice inherent in an electoral system that allowed it only a handful of seats in the Quebec assembly despite its winning 24 and 30 percent of the vote in the 1970 and 1973 election campaigns respectively.
        The PQ's program promised to incorporate PR into Quebec's electoral system and, in fact, the governing Liberals added a commitment to electoral reform to their program in the 1970s. When he won power in 1976, Rene Levesque set up a Ministry of State for Parliamentary and Electoral Reform, its mandate including consideration of alternate voting systems for Quebec.
        The Ministry took up this mandate in earnest at the beginning of the PQ's second term of office and, in 1982, it proposed a regionally based PR system not unlike that used in a number of European countries. Parties would present a list of candidates in each region and voters would choose the party, as well as to indicate preferences within the lists, which would determine the order of candidates elected. The seats were to be allocated to the parties by the d'Hondt divisor system.
        The proposal was favorably received by Levesque and his cabinet -- in part, perhaps, because the popular support of the PQ had been falling. Beset by a deep recession and severe conflicts with its supposed trade-union allies, it could count on barely one-third of the voters, a situation guaranteed to produce far fewer than one-third of the seats under plurality voting.
        In June 1983, the National Assembly unanimously requested Pierre F. Cote, Quebec's Chief Electoral Officer, to chair a commission to investigate reforming Quebec's electoral system. During the Commission's four weeks of public hearings, a large majority of the briefs presented called for a form of PR. In a report presented to the National Assembly in March 1984, Cote recommended a system of territorial proportional representation (PR) similar in nature to that proposed by the Ministry. The main difference was that electoral district boundaries would largely conform to MRC (county) boundaries, which meant they would be smaller in size with twenty-one regional districts with three to fourteen seats per district.

It is difficult to imagine more favorable circumstances for electoral reform in the Canadian context than those of Quebec in the early 1980s

        The Cote report took up the Ministry's argument that not only did PR allow for the fair representation of different political organizations, but also that it promoted the election of women and visible minorities, as well as strengthening regional political institutions.
        While received favorably among intellectuals and well-informed journalists, there was no sign of any groundswell of popular support for the proposal, however. Given its solid lead in the polls at this point, the Liberal opposition refused to enter any discussion of principle over electoral reform, stating simply that is could not support an unpopular government's tinkering with the electoral system in order to hold on to its seats in the legislature. Despite this, the government tried to press on ahead. Would it be possible to get sufficient popular support to make it impossible for the Liberals to block reform? The first obstacle proved to be within the Parti Quebecois itself.
        The PQ executive had itself presented a brief to the Corte Commission. Declaring itself in favor of PR, it nonetheless sought a moderate form that would maintain single-member districts. Its proposal was for a compensatory system based on the German model -- except that, unlike Germany, where half the seats are drawn from party lists and added to constituency results to bring the overall totals to proportionality, it would allocate only a quarter of the seats for this purpose, which would give an advantage to the party winning the most of the single-member seats.
        Thus the party executive found itself in disagreement with its own government. The issue was placed before an enlarged meeting of the caucus at a retreat in August 1984. But the Cabinet failed to gain endorsement for the Plight recommendations. Even the fact that the PQ had sunk below 25 percent in the polls by this point -- which would have spelled total electoral collapse if an election were held under plurality -- was insufficient to convince the majority of the deputies.
        For its part, the cabinet rejected the mixed member PR (MMP) proposal, which might have won support from a majority of the caucus. The main reason given was that, since it was publicly identified with the party, MMP could not gain the kind of bipartisan, above-politics approval needed to win acceptance for a reform of this magnitude.
        A party-government committee composed of three cabinet ministers, the deputies and three members of the executive was created to come up with a solution. But after intensive meetings over a two-month period, no solution was achieved and, in the end, the proposal was dropped.
        The issue soon faded as the PQ split over whether to put Quebec independence on the back burner. When the next election came late in 1985, Levesque had resigned and his successor, Pierre-Marc Johnson, made a creditable showing in a losing effort under plurality, winning twenty-three of 122 seats with 38 percent of the vote. No one has raised the banner of electoral reform since.

Political Psychology Behind the Failure

        It is difficult to imagine more favorable circumstances for electoral reform in the Canadian context than those of Quebec in the early 1980s -- and yet nothing came of it. To get at the explanation, we must look at the workings of the plurality system at the level of the constituency and the perceptions and interests of the actors there.
        It is well known that single-member constituencies provide incentives for catering to narrow local interests. For example, a German study found the legislators elected from single-member districts to be far more concerned with obtaining and locating specific government projects than those elected from the list.
        But there is much more that distinguishes politics at the base when that base is made up of single-member constituencies. William Irvine's analysis of the Canadian electoral system and its weaknesses in Does Canada Need a New Electoral System? (1979) is the most thorough and insightful on this.
        Irvine's argument starts from the nature of party support under plurality, which he describes as overvaluing the behavior of the least-partisan citizens. The logic of the system is for parties to appeal to the "volatile voters," those who can make the difference between winning and losing marginal seats -- since that is what counts -- at the expense even of their own traditional supporters.
        As a result, in the long run, more and more voters become "volatile;" that is, they loosen and then lose their ties with the party. The parties that "purport to be 'everybody's instrument'" are in fact "nobody's instrument. . . . The party is robbed of its institutional usefulness to the electorate."
        Moreover, the voters' volatility means rapid turnover among legislators, making it difficult for individuals to envisage a political career through continuous service to the party and the electorate. (Irvine cites figures showing only 10 percent of defeated Canadian party candidates running in the next election.)
        There is thus little party continuity at the local level: missing are the local party activists to serve as representatives, antennae and organization-builders between elections. Lacking such persons, the party under plurality turns to "experts" and pollsters to tell it how to appeal to the voters, thus further alienating traditional supporters. Irvine concludes by endorsing PR as the system under which parties have reason to build faithful voting blocs around individuals who are in the process of advancing their political careers.
        To take this argument even further, let us consider the party whose candidate was successful in the plurality district. The party relies on that Member of Parliament (MP) as its link at the base with its supporters. But that link is an uncertain one, for the MP needs the party organization only to assure renomination. In between, she or he emphasizes a different role, that of liaison between the local population -- irrespective of political stripe --and governmental institutions. It is only during the general election that the party is the primary -- practically the only -- factor at play in the relationship between elected and elector.
        Indeed, anecdotal evidence from Quebec suggests that in by-elections, in contrast, local factors can play a more important part in determining the outcome, giving active local candidates an advantage over nationally known "prestige" candidates. In sum, the normal, everyday activities of the local MP contradict his or her partisan identity -- the very identity which was the basis of his or her election in the first place.
        This provides the context for understanding the response of the PQ's Quebec legislators to the proposals of their PQ leaders. First of all, being attuned to local elites and not to the intellectuals writing on the op-ed pages, they perceived no preoccupation with electoral reform among the electors.
        It is not impossible to overcome such apathy -- as the New Zealand case shows -- but it requires a special coordinated effort to educate the public on the often subtle effects of electoral arrangements. One factor for the success of such a campaign would be the involvement of local legislators from favorable parties.
        Had the PQ deputies actively supported electoral reform, they might have created a sufficiently positive climate for public discussion. In such a climate, the Liberal opposition might have been forced to enter a debate over the merits of the electoral system. In sum, while support from the PQ legislators would not have guaranteed passage of the reform, failure to do so ensured its being passed over.

Institutional Reasons for Legislators' Inertia

        Let us look more carefully at the legislators' lack of sympathy. Their failure to support electoral reform cannot have been simply a matter of partisan political advantage, for the deputies understood -- and had the simple mathematics explained to them by proponents of the reform if they did not -- what the numbers meant for the party.
        At 20-25 percent popularity in a two-party contest under plurality, every PQ seat was in jeopardy. Yet only a small minority of deputies favored adopting the proposed proportional system of representation. Why? The answer lies in part in inertia, in the legislators' desire to maintain existing structures, since these were the structures that got them elected in the first place, structures with which they were familiar and comfortable, structures almost certain to guarantee them renomination.
        This inertia is what made them more open to the party executive's compensatory formula which allowed the retention of -- albeit larger -- single-member constituencies. But while true insofar as it goes, this explanation is insufficient. For once the compensatory formula was ruled out, the deputies still refused to accept the territorial PR system.
        Clearly, they managed to combine their preference for the existing structures with some expectation that they could be reelected under them at close to the same likelihood of their being elected under territorial PR which they knew would guarantee their party 20 to 25 percent of seats.
        Based on discussion with many of these legislators at the time, I came to understand their reasoning. Though they could not deny the general effect of the low standing in the polls on their party's fortunes, they each believed themselves to be sufficiently immune to these effects. "Moi, je suis correcte dans mon comte," they said.

What seem to be rational choices are in fact short- sighted, both in terms of the narrow interests of the individual or party, and the wider public interest in increasing informed participation in the political process.

        The reasoning expressed their particular vantage point. Their own standing in the district, as they saw it, was solid. After all, they were dedicated and hard-working, always available to their constituents for local functions and the like. They were well-known, constantly in the local public eye; the local opposition was virtually absent. Their reasoning was reinforced by frequent encounters with constituents who never expressed anything but satisfaction.
        This distorted reasoning parallels that taking place in Parliament due to the unrepresentative nature of parliamentary representation under plurality. Just as all the signals the governing party receives from its parliamentary environment mislead it into thinking that it has the mandate of the majority of the population, so the member elected by plurality in a single-member district sees a world in which there is nothing but support for his or her actions.
        Because of this, what seem to be rational choices are in fact short-sighted, both in terms of the narrow interests of the individual or party, and the wider public interest in increasing informed participation in the political process.
        By contrast, a deputy elected along with candidates of other parties on a proportional basis in a multi-member regional district is far less prone to lose sight of the fact that it is party that links her or him to the electorate, just as a governing party elected under PR -- lacking an automatic majority either in the legislature or in the population -- is far less prone to lose sight of the fact that majority support is never automatic and must be built anew for each important legislative initiative.
        The irony is, then, that in the rare case -- as in Quebec between 1982 and 1984 -- when a governing party under plurality overcomes its disincentive to engage in electoral reform, the distorted perception at the district level under plurality can be expected to come into play to abort the process. If this proved true of legislators elected under the banner of the PQ, a party as close to a mass movement as any that have governed recently in western democracies, it is surely even more true of those elected from more traditional parties.

A Ray of Hope

        Is there any hope, then, of instituting PR in Canada? Before responding negatively, we should remind ourselves that anyone asked to assess similar prospects in New Zealand fifteen years ago would have dismissed them as remote at best. New Zealand had over the years developed the "purest" form of plurality voting, with two highly-disciplined parties alternating periods of majority government. How are we to explain the choice just taken there? Ostensibly, there is much to compare the political situation in New Zealand with that in Canada in the last 1980s. Jack Nagel has described New Zealanders' frustrations quite similar to those of Canadians: New Zealanders had become "embittered and disillusioned with 'elective dictatorship'. . . [which] imposed radically disruptive policies without first winning broad-based consent. . . . [They were] repelled by the petty, stridently partisan quality of Parliamentary debate."
        Unlike in New Zealand, though, which experienced the special circumstances of am influential Royal Commission report and political opportunism in promising a referendum, Canadian discontent was not channeled toward the electoral system except in the support of the radical, non-parliamentary -- indeed, American -- idea of recall. Still, in responses to questions posed by pollsters in 1991, Canadians indicated dissatisfaction with the distortion built into the electoral system.
        For example, only 42 percent of those with an opinion found it acceptable that, under the existing electoral system, a party can form a majority government without winning a majority of the votes. (Quebec was lowest among Canadian regions with only 33 percent finding minority rule acceptable.) Yet a Gallup poll reported in the same La Presse poll found 64 percent of respondents to be pleased with the fact that the 1993 election resulted in a majority government.
        This may explain the failure of the losing parties, the conservatives and, especially, the New Democratic Party (NDP), to raise the banner of PR despite the evident unfairness of the result. In the case of the NDP which, when compared to its popular vote, has been under-represented in the number of seats it has been able to win in every federal election in which it has taken part -- and which has raised the matter of electoral reform in the past -- it may be that its nationalist ideology and resulting fear of falling prey to American domination makes it reluctant to support an institutional reform certain to result in minority government.
        Two things, at least, are clear from all this. First, if PR is to come anywhere in Canada, it will be in Quebec because of the recent currency the idea has had there. Second, the reform, if it is to have any hope of making it through the legislative process, will have to be based on the German compensatory model, since it is the only proportional system which allows people to have their own representative.
        This was what made Mixed Member Proportional acceptable to a majority of New Zealanders, and why the majority of PQ deputies were willing to support a compensatory system but not a regional list-based system. From the point of view of optimizing informed participation taken here, MMP is inferior to the list-based system, but it is certainly an improvement over plurality.
        Yet what are the concrete prospects? As this is written, the PQ is leading in the polls a few months before the next Quebec election [editors' note: the PQ won a majority of seats with 44% of the vote]. In the first chapter of its program, that party commits itself to institute a system of compensatory PR elections once achieving power.

        But the matter is deeper than the words of party programs. Quebec had gone farthest in North America in the direction of "concertation;" that is, of bringing business, labor, and government to work together toward common objectives. And it is well-documented that there is a close link between such "corporatist" institutional arrangements and the prevalence of mechanisms of consensual democracy such as PR.
        In any case, with the sovereignty question again on the front burner in Quebec, electoral reform is hardly making headlines. Clearly, and assuming a PQ victory, nothing will happen before the 1995 referendum on sovereignty. Still, it is quite conceivable -- especially in the eventuality of Quebec becoming a sovereign state and thus coming to politically as well as socio-culturally resemble the small corporatist societies of Europe -- that electoral reform will return to the agenda of Quebec politics.

        Henry Milner is a professor of economics and political science at Vanier College in St.-Laurent, Quebec. This piece is excerpted from a longer article that appeared in the American Review of Canadian Studies, Spring 1994.

Table of Contents
Chapter Seven