A fix for Canada's voting system
By Ero Talvila
In growing numbers, influential Canadians are calling for electoral reform. Our first-past-the-post (FTPT) voting system excessively favours both strong and regionally based parties at the expense of minor parties and those whose voting strength is more evenly distributed across the country.

The most egregious example of this in Canada was probably the 1993 federal election in which the Bloc Quebecois took 54 seats with 14 per cent of the vote while the Tories took only two seats with 16 per cent of the vote.

The alternative to FPTP is proportional representation (PR), used almost everywhere else except the USA and the U.K. Discussion of the many variations of PR and of the different ways of counting ballots is beyond the scope of this article.

The Canadian Reform schemes of which I am aware have all recommended some combination of FPTP and PR. The 1979 Task Force on Canadian Unity proposed retention of the existing single-member constituencies, with the addition of 60 seats by PR. The latter were to be allotted among the political parties according to their share of the vote, either countrywide or province-by-province. On a rough guess that there were about 280 seats in the House of Commons then, that plan would have raised the total number of seats to about 340 (there are currently 301), comprising about 80 per cent FPTP seats and 20 per cent PR seats. No action was taken. Other proposals would raise the PR fraction as high as 50 per cent, the same as in Germany and some other countries.

Our proposal would set the FPTP/PR ratio to 3:1, reduce the seat total, and introduce a weighted voting system.

Since the number of seats increases after each decennial census, and since we probably have enough MPs already, the best time to reduce their numbers would be when a new system is introduced.

We have arbitrarily decided on 280 seats, which yields 210 FPTP seats -- with constituencies almost 50 per cent larger than at present -- and 70 PR seats.

In most countries using PR, voters are presented with each party's list of candidates; selection proceeds after the election, working down from the top of the list until the party has the number of members its vote total has earned.

In our system, each party leader would be free to fill out his PR quota with whomever he wants, be they failed candidates or non-candidates, according to the party's needs and in light of the election results. Selection could be on the basis of quality, expertise, gender, ethnicity, region, or something else, in order to achieve the optimum caliber and balance of representation.

Because they would have less constituency business than the FPTP members would have to deal with, the PR members would have more time to spend on policy matters and committees.

Our revised system of balloting would have voters mark their first choice, worth two points, and a second choice, worth one point. That shouldn't be too complicated. Point totals would determine candidates' standings.

What changes might we expect this new system to bring about?

Let's begin with the 3:1 FPTP/PR split. Applying this system, after the fact, to the November 2000 federal election, this is how the seat totals would have been different: The Liberals would have had 162 seats (instead of the 174 they did get); Alliance, 68 (not 66); Bloc Quebecois, 36 (not 37); the PCs, 18 (not 12); and the NDP, 15 (instead of 2). As we can see, the two smallest parties would have gained considerably at the expense of the Liberals -- who would still have maintained a comfortable majority. None of the six minor parties that contested the election would have won a single seat. The Alliance Party would only have gained two seats, but with 19 of their seats coming by PR, it could have augmented its two Ontario seats with many more in central and eastern Canada. This would have given the party a more national character, helping to reduce western alienation.

If the election had actually been fought under these proposed new rules, the results would have been different again. With every vote counting towards PR seats, there would have been less incentive to vote strategically -- that is, to pass over a preferred candidate in favour of another who might have a better chance of denying a seat to a less desirable candidate.

Weighted voting, as proposed herein, would probably result in even more votes for small parties.

Taken together, these two proposed amendments to the electoral system would strengthen the chances for survival as official parties of the currently borderline PCs and NDP, possibly even electing a member or two of the Greens -- although they would have to sit as Independents. Minor parties would have an incentive to try harder.

The proposed system is neither complicated nor difficult to implement. The biggest disruption would be the redrawing of constituency boundaries. The seat distribution would be significantly more representative of the popular will without being so radical as to be rejected out of hand by the party in power.

We think this system compares favourably with others that have been proposed for Canada.

As a related reform to the electoral laws, we propose forbidding, during an election campaign, the publication of all polls to determine voter preferences. By making the outcome less obvious, the possibliity of strategic voting would be further reduced and voters might be more interested in turning out to vote.
Ero Talvila writes from Toronto.