Alternative vote: an electoral system for Canada.
I agree with other contributors to this issue of Inroads that changing the federal electoral system in Canada would be a good thing. My reasons, however, are, in part, not the same as those of some observers, and, as a result, my proposal differs from some of those advanced in this issue.
First, I care not at all for the complaint that women and visible minorities find it hard to get elected under our single member plurality or first-past-the post (FPTP) system. That situation has been changing gradually and will continue to change due to larger social trends. Trying to sell electoral reform as affirmative action, at a time when affirmative action is on the way out as public policy, will hurt its chances of ever being adopted.
Second, I am not overly concerned with exact proportionality in the conversion of popular votes into seats. While I am not against proportionality, I regard it as an instrumental good, not as an end in itself. Pursuit of proportionality for its own sake is doctrinaire. The real test of an electoral system is its impact upon the larger political system in specific circumstances of time and place.
At this level, FPTP serves Canada poorly. Ever since Alan Cairns' classic article in the Canadian Journal of Political Science in 1968, it has been understood that FPTP contributes to the regionalization of the political/party system, which has now reached an extreme in Canada. Only the Liberal Party still has any claim to be a national party, and its claim is rather weak. In the 1997 election, the Liberals were badly beaten everywhere except in Ontario and its satellite communities in anglophone Quebec.
As Stephen Harper and I argue in a recent article, (1) fragmentation of the Canadian party system was inevitable. It has been growing throughout the twentieth century because it reflects deep divisions of political culture among the country's regions. The realities of Canadian society tend to generate a multi-party system, so we need an electoral system that fosters co-operation and coalition-building among parties. FPTP, in contrast, encourages parties to play a war of attrition against each other, hoping to open up a small lead in popular support that can wipe out a competitor's seats. FPTP could even be dangerous for Canada's continued existence because it induces parties to play to their regional base, thus exacerbating our ever present regional and ethnolinguistic tensions.
FPTP is so counterproductive for Canada that almost any alternative would be better, including AV, the alternative vote (Australia); the run-off or second-ballot (France); STV, the single transferable ballot (Ireland); or MMP, the compensation-seat approach (Germany, New Zealand). Any of these could be successfully adapted to Canada so as to encourage party co-operation and reduce dangerous regional antagonisms. Nevertheless, some are better suited to Canada today than others.
The compensation-seat approach has the most supporters among those who have written on Canadian electoral reform, and is defended here by Henry Milner. In the German model, half the members of the legislature are elected from territorial constituencies under FPTP and the other half are elected by a list form of proportional representation (PR), with numbers juggled to achieve overall proportionality. The ratio of FPTP to PR members does not have to be 50-50, however; Kent Weaver, elsewhere in this issue, proposes a ratio more like 90-10. This would not achieve overall proportionality in converting votes to seats, but it would allow major parties to win at least a few seats in all regions of the country, thus inducing parties to become less regional in their approach to campaigning and governing.
STV also has its supporters. (2) The main advantage of STV, as compared to MMP, is that it does not put so much power into the hands of party leaders. Under MMP, the party machine would nominate the candidates on the party list, and, even more importantly, would determine their position on the list. Being high on anyone's list would almost guarantee an electoral victory, unless you were running for a very small party; being low on anyone's list would mean you were grist for the political mill, perhaps hoping to get a higher ranking in a future election. Many voters might see MMP as involving an unacceptable increase in party regimentation, which is already unpopular in Canada. In contrast, STV allows the voters themselves to sort out the ranking of candidates sponsored by their preferred party.
MMP has other problems, too. To accommodate both territorial and PR members, the legislature is usually quite large. The German Bundestag has over 600 members. Would we agree to double our House of Commons to make room for the list members? Creating so many new MPs would be controversial, to say the least. On the other hand, there would be a huge uproar here if we cut the number of territorial MPs in half in order to keep the House at 300 members. The problem is reduced if, as Weaver proposes, the number of PR members is kept small; but so are the benefits of proportionality.
A political problem with any compensation system is that it would create a new pool of seats for which the Bloc Quebecois could not compete. Since the Bloc runs only in Quebec, it would not be eligible for compensation seats from other provinces; and since it already gets more than a proportional share of seats within Quebec, it would not get any of the compensation seats in that province. In the overheated climate of our endless national unity debate, separatists would surely attack this type of electoral reform as a manoeuvre to weaken Quebec's influence in the House of Commons.
STV is also open to objections. A perennial problem is the complexity of the voting process, in which the voter has to choose and rank candidates for perhaps five to ten seats in a multi-member constituency. Depending on the number of parties in the race, this could mean looking at as many as 50 names. Canadian geography is also a problem. Outside our major metropolitan areas, Canada is a very large, thinly settled country. The multi-member rural ridings demanded by STV would range from large to huge, which would make adoption politically difficult.
Finally, given Canada's underlying propensity towards a multi-party system, both STV and MMP would make majority government difficult to achieve. Since the end of World War II, there have been only two elections, 1958 and 1984, in which the winning party received 50 percent or more of the popular vote. It may be a vice rather than a virtue, but FPTP does tend to turn electoral pluralities into representational majorities, and Canadians are used to majority government. Any system that makes majority government unattainable is going to face practical difficulties getting adopted. Even if advocates of electoral reform can persuade the voters to renounce majority government, can they persuade the Liberal Party, which has benefited more than any other party from this aspect of FPTP? And, can electoral reform ever be accomplished without support from the Liberals?
Because both MMP and STV face practical difficulties in getting adopted, it is worth considering other alternatives for electoral reform. The French-style run-off is worth a look, but it would be easily attacked on grounds of the expense of conducting the second ballot. Let me, therefore, focus on the Australian-style alternative vote (AV), which has three big advantages.
First, it would involve a relatively simple transition. The House of Commons would not have to be enlarged, and all constituencies would remain single-member and territorial. There would be no problem of huge rural ridings, party domination of list rankings, or an impossibly complicated ballot. The only difference from FPTP is that voters would be allowed to rank candidates for one seat in order of preference rather than merely to choose their single most preferred candidate. In the past, the expense and time involved in counting and recounting the ballots was a valid objection, but this need no longer be a practical problem in the computer age.
Second, AV has Canadian roots. It was used in Alberta provincial elections from 1926 through 1955, in Manitoba from 1927 through 1957, and in British Columbia in 1952 and 1953. (3) Three provincial parties have recently used AV as part of a direct selection process for party leader; (4) and the Reform Party of Canada, the Alberta Progressive Conservatives, and perhaps other parties are now using AV, instead of the more time-consuming run-off, to nominate local candidates.
Third, as was done in Alberta and Manitoba, AV can easily be combined with STV. Rural ridings can use AV while densely populated, metropolitan ridings can use STV. Voting becomes more complicated in the cities, but the fundamental idea of the preferential ballot remains the same. This sort of compromise would allow a proportional element to be incorporated into the overall system without enlarging the House of Commons.
There are, of course, some disadvantages to this proposal. First, AV does not produce proportionality. It does not purport to be a form of PR; and the record in Australia, Alberta, Manitoba and B.C. clearly shows that proportionality does not result in practice. Most single-member contests are won by the first-ballot leader; (5) and even when the transfers are significant enough to determine who will form the government, as in B.C. in 1952, they do not produce overall proportionality between popular votes and seats. However, to the extent that STV is adopted for urban seats, a degree of proportionality can be built into the system without producing unwiedly multi-member rural ridings. There is thus the possibility of a real-world compromise between those who place high value on proportionality and those who appreciate AV for other reasons.
A weakness of preferential systems like AV and STV is that they can sometimes eliminate a "Condorcet winner," a candidate who would beat every other candidate in a series of head-to-head, two-person contests, but this doesn't seem to happen very often in practice. After making a detailed study of the use of the run-off method (which in this respect is similar to AV) in choosing Canadian political party leaders, I could not find any cases in which this happened. In any event, even if AV does occasionally eliminate a Condorcet winner at the constituency level, it is a defect that pales against those of alternative electoral systems. (6) FPTP distorts electoral results, including eliminating Condorcet winners, more readily than AV. And PR systems can produce political configurations in which small parties, because they become necessary for Cabinet formation, enjoy far more influence than their popular support warrants (e.g. the role of the Free Democrats in Germany).
What positive effects might AV (with or without STV) have upon Canadian politics? It can facilitate coalition building between political parties without forcing them to merge. AV has played this role in Australia for three quarters of a century, allowing the National (previously Country) and Liberal parties to maintain a coalition that can compete with Labour in winning elections and governing.
In British Columbia, AV was introduced with the express purpose of allowing Liberals and Conservatives to vote for each other as second choices and thereby keep the CCF out of power. In the event, W.A.C. Bennett's new Social Credit movement displaced both the Liberals and the Conservatives--an ironic outcome, but still a victory of sorts for AV because a viable political force on the right emerged to confront a viable party on the left. In Ireland, STV has facilitated co-operation between Fine Gael and Labour in presenting a governmental alternative to Fianna Fail.
Admittedly, such coalition building is not automatic, as shown by the case of Alberta. In that province, Social Credit easily won all elections from 1935 through 1955, despite the use of AV in the country-side and STV in the cities. The only time the other parties came close to creating a viable coalition was with the Independents of 1940, and that movement made use of joint nominations rather than co-ordinated preferential voting on election day. (7) AV is not magic. Nonetheless, the historical record suggests that both it and STV can facilitate coalition building if parties are seriously interested in it.
This is of particular importance today, given Canada's de facto one-party-plus system in which the Liberals are the closest approximation to a national party and the other parties all have much narrower bases--sufficient to survive, but not to form a government. At the moment, the obvious candidates for co-operation would be the Progressive Conservatives and the Reform Party, particularly in Ontario. If voters on the right could rank these two parties first and second on an alternative ballot, they might be able to win some seats in that province. The number might not be large at first because polling data suggest that many of those who now vote for either party actually have the Liberals as their second choice. (In a 1997 Calgary poll, only 48 percent of Reform voters and 30 percent of PC voters named the other party as their second choice.) However, if an attempt at electoral co-operation could be made, it would affect the positioning of both parties and, over time, perhaps make each more palatable to the partisans of the other.
Of course, it would be a mistake to choose an electoral system because it might work to the advantage of certain parties at a particular moment in time. The British Columbia Liberals who brought in AV for the 1952 election learned that lesson! In practice, Reform and the PCs, particularly under Preston Manning and Jean Charest, might never be able to bring themselves to a level of co-operation where they could take advantage of AV. But even if they did not, other parties, perhaps parties not yet born, might prove more co-operative, and hence more successful.
In the Canadian context, AV is a modest proposal. Its effect on politics would be subtle and perhaps not felt for some time. Many proponents of electoral reform will want a more dramatic change. But if the practical problems associated with PR make MMP and STV politically unsaleable, those who want to achieve something rather than nothing should take another look at the alternative ballot.
(1) . Tom Flanagan and Stephen Harper, "Conservative Politics in Canada: Past, Present, and Future," in William D. Gairdner, ed., After Liberalism, Toronto: Stoddart, 1998.
(2) . For example, see Nick Loenen, Citizenship and Democracy: A Case for Proportional Representation, Toronto: Dundurn, 1997.
(3) . See David Elkins, "Politics Makes Strange Bedfellows: The B.C. Party System in the 1952 and 1953 Provincial Elections," BC Studies 30 (1976) pp. 3-26; and Harold Jansen, "The Single Transferable Vote in Alberta and Manitoba," University of Alberta: Ph.D. Dissertation in Political Science.
(4) . William Cross, "Direct Election of Provincial Party Leaders in Canada, 1985-1995: The End of the Leadership Convention?" in Canadian Journal of Political Science 29, 1997, p. 302.
(5) . See R.M. Punnett, "The Alternative Vote with the Optional Use of Preferences: Some Irish Lessons for Britain and Australia," Journal of Commonwealth and Comparative Politics 25, pp. 26-43; and J.F.H. Wright, "An Electoral Basis for Responsible Government: the Australian Experience," in Choosing an Electoral System: Issues and Alternatives, Arend Lijphart and Bernard Grofman, eds., New York: Praeger, 1984, pp. 127-134.
(6) . Tom Flanagan, Game Theory and Canadian Politics, Toronto: University of Toronto Press, forthcoming.
(7) . Alvin Finkel, The Social Credit Phenomenon in Alberta, Toronto: University of Toronto Press, 1989.
|Printer friendly Cite/link Email Feedback|
|Publication:||Inroads: A Journal of Opinion|
|Date:||Jan 1, 1998|
|Previous Article:||Electoral reform is not as simple as it looks.|
|Next Article:||How to renew Canadian democracy: proportional representation for the Commons, a Senate elected from single-member constituencies, and political...|