What Really Happened On May 1?

At the heart of the 1997 election result is a conundrum. The defending government was presiding over steady growth, low inflation, falling unemployment, a buoyant housing market, cheap mortgages and falling income tax rates - yet it lost power; and not only was it ejected from office, it secured a lower share of the vote than any governing party this century. How come?

The BBC's exit poll, conducted by NOP, helps to explain what happened. A total of 2,356 voters throughout Britain completed questionnaires outside polling stations on election day, immediately after voting. The main findings that emerge are:

• The economy did matter: memories of the recession of the early nineties proved more important than the subsequent recovery in shaping attitudes.

• Labour managed to neutralise the negative factors that lost it the 1992 election.

• The Conservative Party was seen as narrow and divided, while Labour was seen as united and good for all classes.

• Tony Blair was seen as stronger than John Major (although not more trustworthy).

• Most people thought Labour policies would make things better; a substantial minority thought Conservative policies would make things "a lot worse".

Table 1 shows the details

Women swung more to Labour than men; the result was to close the gender gap. For the first time since 1945, the votes of women alone were sufficient to give Labour an overall majority. (In 1964, 1966 and October 1974 - the other post-war elections at which Labour secured an overall majority - men voted Labour in sufficient numbers to ensure an overall majority; had only women voted, Labour would either have lost the election or, at best, been able to form a minority government.)

Young voters swung to Labour to a remarkable extent. In 1992, 18- 29-year-olds preferred the Conservatives over Labour by 41%-38%. In 1997, Labour voters outnumbered Tories by 56%-22% - a swing of 18.5%. The swing to Labour was progressively lower in each succeeding age group: 15% among 30- 44-year-olds, 11.5% among 45- 64-year-olds and only 1% among men and women over 65. Indeed, at 34%, Labour's support among the over 65s was actually slightly less than in 1992.

Never has the age profile of the two main parties differed so greatly. Only 35% of Conservative voters were under 45; 65% were 45 or older. The figures for the other two main parties were: Labour - 51% under 45, 49% 45 and over; Liberal Democrat - 44% under 45, 56% 45 and over.

Analysed by housing tenure, the swing to Labour was greatest among mortgage payers, and smallest among council tenants. Detailed analysis by the BBC on election night of the pattern of constituency results shows that the seats with unusually high numbers of semi-detached homes swung particularly heavily to Labour.

It seems that the housing slump of the early nineties, leading to years of negative equity, caused a particularly strong rejection of the Tories - for mortgage-paying semi-detached home owners were among the most numerous victims of the housing slump: for example being unable to sell their home in order to move to be nearer a new job.

Memories of the housing slump provide one example of how the economy did matter in this election - and why the Conservatives did not benefit as much as they must have hoped from the subsequent recovery. Table 2 provides a further example.

NOP asked voters about the record of the past five years. Only one in four said their standard of living was higher now, compared with 1992. Many more, 38%, said their standard of living had got worse. However, when voters made a similar five-year comparison, but about the state of the economy, the figures were rather different: more voters thought the economy was now stronger (35%) rather than weaker (31%).

Two reasons may be adduced for the difference between these two answers. First, some people feel that they have been left behind during the recovery. Indeed, for them, the Conservative slogan, "Britain is booming", may have been counterproductive for it induced some voters to react: "Maybe it is, but I am not". Of those voters who said the economy was stronger, but their families standard of living had deteriorated, more than eight out of ten voted Labour or Liberal Democrat.

Second, it may be that "the economy" is mainly about money, whereas "standard of living" embraces not just the purse or wallet, but a wider sense of the quality of life, including the quality of the health and education services, public transport and so on. Although these were not tested directly in the exit poll, other surveys have shown that large majorities of people feel that the public services have deteriorated since 1992.

One further reason why the economy did not save the Conservatives is that Labour managed to get rid of its previous reputation for economic incompetence, especially over taxes. In 1992, Labour was seen as likely to raise taxes - not as a deliberate act of policy in order to improve public services, but as a consequence of its perceived inability to run the economy properly.

In 1997's exit poll, NOP asked what they expected to happen to taxes (a) if Labour won, and (b) if the Tories won. There was little difference between the results: 61% thought a Labour government would increase taxes, 9% thought it would reduce them; 30% thought taxes would remain the same. The figures for the Conservatives were: taxes to rise - 59%, fall 5%, stay same 36%.

More generally, as Table 3 shows, Labour matched the Tories in 1997 when voters were asked which party they trusted most to take the right decisions about the economy - a far cry from 1992, when the Tories led Labour on this issue by 53%-33%. On income tax, the shift has been even greater. In 1992 the Tories led Labour by 22 points as the party most trusted to take the right decisions; in 1997 Labour led the Tories by 8 points.

It should also be noted that the Liberal Democrats scored well on both tax (20%) and education (26%), relative to their national vote (17%). The key policy that the party stressed during the campaign was that it would increase the standard rate of income tax by 1p in the £, and spend the money on education. These figures confirm the findings of campaign polls: this was a popular policy, which may have contributed to the fact that the Liberal Democrats were the only party to gain support significantly during the course of the election campaign.

The Liberal Democrats were also widely trusted as the best party to deal with "sleaze", scoring more (29%) than the Tories (23%). These figures suggest that sleaze did help drag the Conservative vote down, and that the Liberal Democrats' attempt to portray themselves as a fresher, cleaner, more positive party did them some good.

If Labour edged ahead of the Tories on the economy, it opened up a far larger gap on the wider "mood" question of how things would change in Britain under the alternative futures of a Labour and Conservative government. As many as 57% thought that "overall, Labour policies would make things in Britain better"; just 33% thought Labour policies would make things worse. The figures for the Tories were: better 31%, worse 49%. And most of the 49% who thought Tory policies would make things worse, thought they would make things "a lot worse".

In 1992, far more people were frightened of a Labour victory than a Tory victory; by 1997 the fear factor harmed the Tories far more than Labour. Labour was seen as the party of hope - albeit, for the most part, modest hopes: most of these who thought things would get better under Labour thought that things would only be "a little better".

Labour also benefited from the comparison of the alternative candidates to be Prime Minister. In 1992, John Major outscored Neil Kinnock as "best Prime Minister" by 15 points. In 1997, Tony Blair outscored John Major by 14 points - see Table 4.

One of Blair's advantages was that he was overwhelmingly regarded as a strong leader (by 77%, compared with just 23% who thought him weak), whereas most people thought Major was not (strong 35%, weak 65%). However, when NOP tested another personal attribute, trust, there was little to choose between the rival leaders. Just over half the electorate thought both men could be trusted, just under half thought they could not. Margaret Thatcher's reputation was famously summed up in one headline as "We respect her but we don't trust her"; Major's reputation seems to be the opposite: most voters like him but don't respect him.

Finally, the Tories suffered in the 1997 election by being seen as a narrow, divided party. As Table 5 shows, a majority of more than two-to-one thought the Conservatives good for one class, rather than good for all classes; but a similar majority thought Labour good for all classes, not just one. Worse even than that for the Tories, as many as 84% saw the Tories as divided - whereas 66% saw Labour as united. One element of the damning verdict that the electorate delivered to the Tories and their leader on election day was: "nice man, shame about the party".