Why June would suit Blair best

Tony Blair is being infuriating. At least to some advisers, the man just isn't concentrating on what matters. They are all hunched over election grids and focusing on when to announce the campaign. And the Prime Minister himself? He has put it all to one side and is frantically obsessed with the detail of the foot-and-mouth crisis - a self-taught expert on sheep movements, rendering processes and vaccination strategies. He is on a personal mission against the spreading virus.

This does not mean that Blair is uninterested in the election date. His hilarious camera ambush at the Stockholm summit put paid to that idea. Rather, he has an intelligent sense of priorities. As long as the crisis is out of control, the danger of a backlash against an electioneering government is great. He can't assume that he will win very big against the Tories until he is winning against foot-and-mouth.

"Winning", in this context, may not mean the actual numbers falling. The curve of the epidemic suggests that won't happen for a while. But it would certainly mean no more cases outside the areas already infected, completing the disposal of the vast heaps of dead animals, an end to complaints that it was taking too long to kill infected herds, and effective "firewalls" protecting other areas. Until this long list is ticked off, the Prime Minister is clear that the election question must come second. If he feels a May election is impossible, what does he do? First, he offers trauma counselling to many around him. As a senior minister told me recently, 3 May has been the preferred choice for as long as four years now: the date was pencilled into the diaries on the very day that Labour swept into Downing Street in 1997. All that planning, all those grids.

Blair's best argument is that it simply doesn't matter electorally. According to the Ministry of Agriculture, there are 64 seats where foot-and-mouth has already been detected; 14 of those were Labour gains at the last election, where a 5 per cent swing to the Tories from Labour would result in Labour losing the seat. To go ahead with a May election in the face of the foot-and-mouth outbreak would more than likely alienate enough voters in those constituencies - which include Northampton South, Medway, Stroud, The Wrekin, and Sittingbourne and Sheppey - to cause that 5 per cent swing. With 14 fewer seats for Labour, and 14 more for the Tories, 28 seats are gone, just like that. We can reasonably add another four or five constituencies that will probably have been hit by early May, which will knock another ten off Labour's majority. Yet that still leaves Labour with a more than comfortable majority of more than 100. As I argued in this column last week, many in the party would be quite happy to see the numbers come down into double, not treble, figures.

Then there's that big political opponent for Labour: voter apathy. The apparently good news for Labour is that, according to the elections guru David Butler of Oxford University, in his book The People Have Spoken (Hansard, just published), "when there is a crisis . . . more will flock to the polling station". Tracking voter turnout throughout the 20th century, he finds it highest in the elections of January 1910, October 1924 and February 1974 - each of which was called suddenly and dramatically. No one could claim that a spring election was called suddenly, but the highly charged atmosphere and sense of national crisis would, on Butler's evidence, increase turnout. This might help Labour.

The problem with history, and psephology, is that elections are dynamic. Overshadowing all this is the threat of a May national backlash against a government perceived to be cynical, removed from the light of the countryside and determined to promote its own interest first.

As I write, No 10 opinion, though seriously divided, is beginning to move towards a delay. How would that happen? First, hardly anyone supports October. The recession might have hit, and it's a horribly long pause for a political system geared up to go now. It has the whiff of Jim Callaghan's famous, and fatal, delay in autumn 1978, which helped usher in the Thatcher era six months later.

No, it would almost certainly be later on in May or early June. Some Labour people would like Blair to give no indication as to when. But if he is to take the moral high ground, the obvious way would be to announce that the local elections will be delayed a month, to allow more time to control the virus. June is then the preferred option for the general election, too.

This removes the main anti-June argument, which is that Labour might have just been battered in May local elections. It makes Blair seem magnanimous, self-confident and generous. It gives him the time he really needs to get on top of the crisis. It removes any mileage the Tories might have got from accusing him of ruthlessness or cynicism - while at the same time giving them some painful financial problems. Oh yes, and it gives people another few weeks to realise the useful effect of the Budget measures. It might even be possible for another interest-rate reduction to be squeezed in.

The main argument against June is that it would, in effect, mean a two-month election campaign. That would either bore the electorate to sleep or risk the Labour campaign leadership making a major slip-up, simply through fatigue or by taking their eye off the ball. The other arguments, about grids and party planning, should not come into it: voters do not care.

This, then, is the real balance sheet facing Blair. For what it's worth, from the outside, this tally of arguments, as related by insiders, seems to point marginally to June. But one thing is certain: the decision will be based entirely on the raw calculation of what would maximise Labour's vote. That's not being cynical; that's just politics.