At the time this article was written Lisa Young was  Assistant Professor in the Department of Political Science at the University of Calgary. This article is based on a presentation to the Canadian Study of Parliament Group on April 25, 1998.

There are several reasons for wanting to see more representative legislative bodies in Canada. At the symbolic level, a legislative body that reflects the diversity of Canadian society affirms that society’s inclusiveness and egalitarian nature. When legislative bodies fail this test, the legitimacy of the government they support suffers in the eyes of excluded groups. Moreover, some Canadians believe that a more representative legislature or House of Commons would enact different kinds of legislation – legislation that was sensitive to the interests of women or minority groups. This article looks at some of the arguments in favour of a new electoral system.

There is little question that, in demographic terms, Canadian legislatures are unrepresentative. Our parliament and legislatures still over-represent professional white men and under-represent virtually every other segment of the population. Women comprise less than 30% of the members in any of Canada’s provincial legislatures, and only 20.6% of the current House of Commons. Although reliable statistics are more difficult to come by, a cursory glance at the membership of the House of Commons or almost any provincial legislature suggests that the diversity of Canadian society – in terms of ethnicity, ability and sexual orientation – is not reflected in our governing bodies. These patterns of under-representation are not random. They have persisted over an extended period; even more significant, they offer a rough reflection of the distribution of power, wealth, prestige and authority in contemporary Canadian society.

Several of the major Canadian parties have tried to devise answers to this question, adopting measures ranging from special training programs and recruitment practices aimed at women, to the NDP’s affirmative action plans, to the Liberal party’s appointment of women candidates. To varying degrees, these programs have been successful in increasing the number of women candidates, and may have had some effect in encouraging candidacies from other under-represented groups. The success of these efforts has been limited, however, by their awkward fit with the Canadian electoral system.

My point of departure is a simple observation: electoral rules are never neutral.

In Canada, we tend to see our territorially-based single member system as both neutral in effect and somehow natural. Upon closer examination, we can see that it is neither. By grouping citizens according to their geographic location, our electoral system privileges territorial identity. It tells us that we are, above all else, members of territorially-based communities.

When we talk about representation in Canada, we talk about it in primarily territorial or regional terms. It would be considered completely unacceptable if a province with half the country’s population had only a quarter of the seats in the House of Commons. The notion that an Albertan could adequately represent the views of a Quebecker, or vice versa, would be considered laughable. Yet we do not see an urgent need to reconsider an electoral system in which women (over half the population) hold only a quarter of the seats in the House of Commons, and we argue that non-Aboriginals can represent the interests of Aboriginals. The primacy of territory in our discussions of representation is both a reflection of the significant regional cleavages in Canadian politics, and serves to reinforce them (as Alan Cairns pointed out 30 years ago).¹

This emphasis on territorial representation has the effect of muting non-territorial claims for representation, effectively rendering them secondary. The institutional manifestation of a territorially-defined conception of representation – our single member plurality electoral system – also places practical barriers in the way of non-traditional candidates or, more precisely, non-traditional candidates who lack a geographically concentrated constituency. The most notable group in this regard is women who, unlike ethnic groups or gays and lesbians, tend not to live in geographically-concentrated groupings.

If we accept that more diverse legislatures are a desirable goal, we are left with a question: how do we get there from here?

The most effective means of increasing the diversity of Canadian legislatures is to reform the electoral system.

Single member systems make nomination contests “winner takes all” events. By selecting a white, male professional candidate to stand as their party’s sole candidate in a single-member election, the members of a constituency association can claim that they have simply chosen the “best” candidate for the job. If the majority of constituency associations independently choose similar candidates, it appears coincidental. Clearly, there is no concerted plan to exclude women, racial minorities, or others. Moreover, the logic of a single-member system requires that the most appealing candidate be selected. Under such circumstances, deviation from the norm of the professional white male candidate is noticeable.

This can, of course, be contrasted with other electoral systems. When one party organization must select more than one candidate, it has a clear opportunity to achieve a degree of diversity. Moreover, failure to achieve some sort of diversity under such circumstances may spark negative comment and limit the party’s appeal to some voters.

This raises the question, then, of how to design an electoral system that would be more likely to produce legislatures that reflect the diversity of contemporary society. The most direct (if not necessarily the most advisable) route would be to adopt an electoral system that guaranteed the desired outcome. We have seen two proposals along these lines in Canada in the past decade. The first is the Royal Commission on Electoral Reform and Party Financing’s proposal for Aboriginal Electoral Districts, and the second is the Nunavut Implementation Commission’s proposal for dual member constituencies with gender parity. The former has never received serious political consideration, and the latter was defeated in a referendum in the spring of 1997. I have argued elsewhere that the Nunavut proposal is justifiable under Canadian law and within the context of democratic theory.²

I believe that electoral systems that facilitate, rather than require, diverse outcomes are preferable. There are several reasons for this.

First, such arrangements are more flexible, and allow different aspects of political identity to be represented. For instance, the Catholic/Protestant cleavage that was so salient that it required political representation in the nineteenth and early twentieth century has faded.(Prince Edward Island has recently abandoned the dual-member system that had been used informally to ensure equal representation for Protestants and Catholics in the provincial legislature). Even 15 years ago, sexual orientation would have seemed an implausible element of political identity, but today it is highly salient. In short, the salience of various political cleavages waxes and wanes. The more flexible an electoral system, the more responsive it can be to emerging political identities.

Second, such arrangements allow parties, as essentially private organizations, to determine their own internal practices surrounding questions of representation. Under some version of a PR system, the NDP would probably adopt a system of quotas for inclusion of various groups on its list and the Reform Party probably would not. This kind of variation in internal practice is entirely appropriate in a democratic system, and would offer voters the opportunity to choose among parties that stood for different conceptions of representation.

Third, in terms of political realities, systems with guaranteed outcomes are politically unviable. Territorial conceptions of representation are deeply embedded in the Canadian public’s notion of how politics should be organized, and any deviation from this to ensure representation for other elements of political identity smacks of “undemocratic” affirmative action.

Virtually any other electoral system would facilitate the election of more representative legislative bodies than our current system does.

Whether dual member, multi-member, single transferable vote or proportional representation, a system in which a party nominates more than one candidate in an electoral district will remove the “winner takes all” dynamic discussed above and allow for some measure of ticket balancing.

Adopting a different electoral system does not guarantee representative outcomes, but would merely facilitate them. Several cross-national studies have noted women are better represented in the national legislatures of countries that employ PR electoral systems. In countries like Germany that use a mixed system, women are more likely to be elected from a party list than from a geographic constituency. These patterns do not, however, constitute guarantees. Where PR systems have elected women in significant numbers, one finds a history of women working within and outside political parties to reflect the diversity of society on their electoral lists. In the absence of such internal and external pressure, the electoral system will have little impact on the composition of the legislature. In short, for those who want to see more representative legislatures in Canada, a new electoral system is only the beginning of the process.³

Notes

1. Alan Cairns, “The Electoral System and the Party System in Canada” Canadian Journal of Political Science 1 (1968).

2. Lisa Young, “Gender Equal Legislatures: Evaluating the Proposed Nunavut Electoral System” Canadian Public Policy, October 1997.

3. For a more detailed discussions of this argument, see Lisa Young, Electoral Systems and Representative Legislatures: Consideration of Alternative Electoral Systems, (Ottawa: Canadian Advisory Council on the Status of Women, 1994).