Just as European and American leaders prepare for a summit on the global financial crisis, Canada, the first of the G7 countries to go to the polls since the crisis began, votes today (Australian time) in its third general election in just over four years.

In 2004, the Liberal government lost its parliamentary majority but carried on as a minority government with the support of the left-wing New Democratic Party. That support was withdrawn the following year, and in December 2005 the Liberals were defeated in a vote of confidence.

The ensuing election returned the opposition Conservatives as the largest party, but well short of an overall majority. (Adam Carr as usual has the figures.) Conservative leader Stephen Harper formed a minority government, which has carried on somewhat unsteadily with the support of the pro-independence Quebec Bloc. Last month, Harper seized the opportunity for a dissolution in the hope of winning a majority.

That now looks unlikely, although the Liberals have also failed to make up the ground they were hoping for. The most recent opinion polls (conveniently summarised at Wikipedia) show both major parties down slightly from their 2006 vote, to the benefit of the New Democrats and the Greens.

With single-member electorates and first-past-the-post voting, however, votes translate very imperfectly into seats. The biggest losers from this are the Greens; they look set to maybe double the 4.5% they received in 2006, but because their vote is thinly spread they will struggle to win a seat. The Quebec Bloc, on the other hand, with a similar level of support, will again win a bundle of seats, because its vote is heavily concentrated in one part of the country.

That means that to win a majority, the Conservatives need not 50% but probably less than 40% — hardly an impossible target.

William Bowe, blogging for Crikey at the weekend, considered the electoral effects of the financial crisis and contrasted two hypotheses: one, attributed to Peter Brent, that incumbency and length of service were the key variables; the other, from Adam Carr, that voters were distinguishing between right and left-wing governments, and punishing the former.

So far, as Bowe concludes, the evidence inclines towards Carr’s hypothesis. But in Canada, even if voters clearly opt for a centre-left government, their perverse electoral system looks like denying them that choice.