Complete the following sentence: "On 7 June the single largest bloc of the electorate will vote for . . . " Did you write "Labour"? Wrong, almost certainly. The correct answer is likely to be "no party". For the first time since (almost) all adults were given the vote in 1918, abstainers are likely to outnumber voters for the winning party. The elected government's claim to a popular mandate will look more threadbare than ever.
The arithmetic works out as follows. The polls in week one of the campaign put the parties' support at 31 per cent Conservative, 50 per cent Labour, 13 per cent Liberal Democrat. This is within a whisker of the polls at the same stage before the 1997 election. If they are exaggerating Labour's actual vote by the same margin as in 1997, Labour will end up with about 44-45 per cent of the vote.
But the polls also point to a significantly lower turnout. The numbers "certain" or "very likely" to vote are seven to ten percentage points lower than at the same point in 1997. The "won't votes" and "don't knows" (who are "won't votes" in disguise) are up by a similar amount. This suggests that turnout on polling day could dip from the 71.5 per cent of 1997 to somewhere in the region of 62-65 per cent this time - the lowest level since 1918. Adjusting for the inaccuracy and age of the register in June, an official turnout of 65 per cent amounts to 68.5 per cent of those actually eligible to vote. If 45 per cent vote Labour, Labour will have 13.5 million voters. But the abstainers will total about 13.9 million. And, in reality, the number of non-voters is much larger because, according to checks made against the 1991 census, another 3.5 million people, especially the young, the unemployed and blacks, do not register at all.
The reluctance to vote recorded in recent polls confirms the lamentable turnout in UK elections since 1997. In local elections, it has averaged 29-30 per cent, well down on the 40-50 per cent typical since the 1940s. In the London elections, it was 33 per cent; and in the 1999 European parliamentary election, an extraordinarily low 23 per cent. In the inner-city areas of Merseyside, Greater Manchester and Tyneside, fewer than one in five voted in the local or European elections. At by-elections in heartland Labour seats, it was an abysmal 25 per cent in Wigan, 24 per cent in Tottenham and 20 per cent in Leeds Central, the lowest in any by-election since the war. The more Labour the area, the lower the turnout. And the lower the turnout, the worse Labour did: hence the Conservative "win" in the Euro elections, with the support of 9 per cent of the electorate.
The left blames new Labour "modernisation". The rot started in 1997, it claims, when turnout dropped to the lowest level since 1935: Labour's traditional, working-class supporters were less than enthused by the Blairites' embrace of big business, cold-shouldering of the trade unions and flirting with the Liberal Democrats. Since 1997, goes the reasoning, the early freeze on public spending, a deteriorating health service and cuts in disability and single- parent allowances have only deepened the alienation.
In The Rise of New Labour (OUP, 2001), Anthony Heath and colleagues, analysing the British Election Study for 1997, show that turnout did indeed fall furthest among the Labour-supporting "core working class" of trade union members, council tenants and the unemployed, while it actually increased among middle-class Labour supporters. But, the Labour salariat aside, turnout fell across the board, among Conservative as much as Labour supporters. There must have been a general, national factor at work. And if old Labour disaffection was so significant, the Socialist, Green, Scottish Nationalist and Plaid Cymru candidates should have done better in the low-turnout Labour heartlands.
Why, then, did turnout fall so sharply in 1997? Not because of Conservative abstentions, as the Tory high command likes to comfort itself: there was no correlation between the decline in constituency turnout and the size of the Conservative majority in 1992. Nor because sleaze had turned the electorate off politics. According to the British Election Study, the proportion claiming to "care a good deal" which party won was just as high (76 per cent) as in the Wilson, Heath and Thatcher years.
The culprit was the polls and their unremitting message of Labour victory. Labour and Conservative supporters alike regarded the result as a foregone conclusion and felt less incentive to vote, especially in safe seats. Historically, turnout has turned more on the closeness of the contest than on the degree of party polarisation. It was low in 1983 (73 per cent), despite Thatcherism and the challenge of the SDP-Liberal Alliance, because everyone expected Labour to be trounced. It was high in 1951 (82 per cent) despite one-nation Conservatism and the wholesale absence of Liberal candidates, because people expected a close result. It was high in February 1974 (78 per cent) and 1992 (78 per cent), when polls fed media speculation about hung parliaments.
Turnout will fall to a record low on 7 June because the drumbeat of the polls is "certain Labour win". Does it matter? Not for the result. Turnout will hold up best in the Labour marginals, where the result will be decided. Firmness of support and intention to vote were weaker among intending Labour than Conservative voters earlier in the year, but have now caught up. The usual differential abstention at Labour's slight expense will not necessarily be more marked this time but, even if it is, the cost will be closer to six seats than the 60 forecast by a deliberately alarmist Millbank.
But record abstentions should matter to the self-proclaimed party of social inclusion. Across wide tracts of run-down inner cities, a large majority are detached from politics. Over the past century, first the churches, then the trade unions, now the Labour Party have lost their fragile grip on the urban poor. Britain is following the US path to a depoliticised underclass, without a place in our system of representation.
Ivor Crewe is professor of government and vice-chancellor of Essex University