What happens if the next British election reflects today's opinion polls and Labour is wiped out—slumping from today's 350 MPs to 200 or even fewer? The conventional wisdom is that moderates and Blairites sit for the more marginal seats, and leftist old Labour types for heartland safe seats that would survive a disaster. The result, it is claimed, is that the party would implode. The parliamentary Labour party (PLP) would become a socialist rump, with the moderate minority splitting to form a new party, or perhaps to join the Liberal Democrats.
As often with conventional wisdom, there is something in the underlying assumption. Labour's very marginal seats are disproportionately held by party loyalists. If Labour were to lose its 44 most marginal seats, for example, it would lose just one MP who backed the left-wing John McDonnell for the leadership against Gordon Brown last year. And seats held by Labour with a majority of 10 per cent or less are disproportionately held by MPs who have not cast one dissenting vote against the government since 2005.
Yet anyone hoping for a leftist takeover of the PLP will be disappointed. For one thing, the effect is not very strong. For example, of those Labour MPs who are currently committed to fighting the next election, around 62 per cent have not voted against the party whip since the last election. If Labour loses every seat up to, say, a majority of 5 per cent, then 70 per cent of those who fall are loyalists, 30 per cent rebels. So loyalists go faster than rebels, but mainly because there are more of them; any disproportionality in the rate of defeat is marginal.
Moreover, there is nothing at all in the belief that a catastrophic result will produce a rump PLP dominated by the left-wing Campaign Group. Most of these effects even out after the most marginal seats have fallen: once Labour has lost every seat with a majority of 15 per cent or less, rebels and loyalists go in roughly the same proportions.
And this sort of calculation almost always ignores the effect of retirements. With up to two years to go, we can only guess how many Labour MPs will announce that they are not going to contest the next election. So far around 40 have said they are off, but expect that number to rise, especially if the outcome continues to look bad. How enjoyable do you think Westminster was for the 165 Conservative MPs who hung on in 1997? Formally announcing retirement early also allows an MP to start openly chasing post-Commons employment opportunities, enabling them to line something up before a possible deluge of ex-MPs floods the job market.
Those who have already announced that they are going are older and disproportionately from the left. About 70 per cent of them have voted against the party whip since 2005, compared to around 40 per cent of the PLP overall. Almost one in five backed John McDonnell, compared to 8 per cent of Labour MPs as a whole. Retirements remove a sizeable chunk of the left (the same was true in 2005, when the rebels retired at about three times the rate of the loyalists): those going include Lynne Jones, Bob Marshall-Andrews, Alan Simpson and John Grogan.
Under most circumstances, the overall ideological effect of retirements and defeats roughly cancel each other out. For example, of the seats Labour now holds with a majority of up to 25 per cent, if it lost every one, the proportion of the PLP who supported McDonnell for the leadership or Jon Cruddas for the deputy leadership last year would remain almost identical to what would happen if the party retained every one of these seats.
Many details will change in the next couple of years and that's before we attempt to measure the ideological positions of any incoming Labour cohort or the effect of boundary changes. But the general pattern is clear: if the PLP shifts ideologically after a heavy election defeat, it will not be because of a huge change in its personnel. It will be because of the lessons those remaining draw from that defeat.
