Now we know why the Prime Minister felt compelled to call the election a year early: so he could take the summer off. "I believe it is important that we take the summer to prepare for the launch of the new government in September," he explained to reporters Sunday. See, if you held the election in the fall, that would be too late. September would already be over.

The early election is not much of an issue in itself. Such use of the prime ministerial prerogative may serve as a symbol of the arrogance of the present government, or better as an example of the grotesque concentration of power in executive hands -- the one being intimately linked to the other. But it does not threaten the Liberal grip on power. Indeed, it's not clear anything can.

This will disturb many people, most of all the media, for whom the prospect of 36 days filled with headlines like "Liberals still in front" is a nightmare too horrible to contemplate. Press coverage in recent days has been peppered with articles trying to make the case that the Liberals, notwithstanding opinion polls showing them 25 to 30 points ahead of their nearest rivals, are in some sort of trouble. The speculation has been helped along by confessional quotes from unnamed party insiders, just the kind of highly trained people you expect to blurt out their innermost fears three days before an election.

One theory, known in the literature as the poll-sitters' paradox, predicts the Liberals will lose precisely because they are so far ahead. In almost every federal election in the last forty years, it is recalled, the front-runner on the day the writ was dropped endured a sharp decline in suport as the campaign wore on. Projected majorities shrank to minorities, as in 1965 and 1972, or in some cases to outright defeat: the last election being the most dramatic example.

There are two problems with this theory. One, in every previous campaign, there was a credible government-in-waiting, a national party with a reasonable prospect of taking office in the event the other stumbled. No such party presents itself this time. Two, the Liberals have already suffered a fairly calamitous drop in popularity: from the sweltering 57 per cent registered in last summer's Angus Reid poll to the more temperate 42 per cent recorded at the latest reading. That it now seems to have stabilized suggests most of the froth has been blown off.

A second theory holds that the Liberals' national poll numbers are deceiving: there are in fact four elections, with a different opposition party showing strength in each region. The Tories are said to be the threat in Atlantic Canada; the Bloc maintains its hold on fortress Quebec; the West is Reform and NDP territory; while in Ontario all four national parties could take seats.

So although the Liberals have a big lead in the overall standings, they could be nipped to death in these regional dogfights.

It's a persuasive analysis, marred only by the fact that the Liberals are ahead in every region: far, far ahead, in some cases. In Atlantic Canada, according to Reid, they are 17 points in front; in Ontario, 21 points; even in B.C. and Alberta, supposed Reform strongholds, the Liberals have a modest lead.

As for Quebec, even if the Tory resurgence is not so pronounced as that astonishing CBC poll makes out, it now seems evident that Jean Charest is at least as much a threat to the Bloc, for whom Gilles Duceppe is fast proving a deadweight, as he is to the Liberals. If the Tories split the nationalist vote with the Bloc, as they have split the right-wing vote with Reform elsewhere, they could end up doing in Quebec what they have done in the rest of the country: elect more Liberals.

Then there are those academic studies purporting to show precise statistical correlations between this or that economic event and electoral defeat for incumbents. Two professors at the University of Toronto have found, for example, that whenever growth in real personal disposable income per capita -- dollars in your pocket, for short -- falls below 2 per cent in the year before an election, the voters took it as their cue to boot out the governing party. As incomes have in fact been falling for more than six years, that would seem to bode ill for the Grits. Another study finds the same correlation with rising taxes.

Again, nice, but it runs into some more fundamental electoral arithmetic: four opposition parties, two limited by region, two by ideology, ranged against the democratic world's most successful and adaptable governing party. The Liberals will have to fall a lot further to make this election interesting.