Is history repeating itself?

Abstract

There is much that is remarkably similar about the circumstances in which the Labour Party in the UK finds itself now, and where it stood on the brink of its landslide success in 1997. The party has seemingly made itself more electable in the eyes of the electorate following a tack back towards the ideological centre of British politics. Then, in an imitation of what happened after Black Wednesday in September 1992, it has been propelled into a large polling lead after the incumbent government's reputation for economic competence was severely damaged as a result of losing the confidence of the financial markets. Although the next general election could still be nearly two years away, it would seem as though history could well repeat itself.

To illustrate how the pattern of party support has changed, we use here the data collected by the 1996 BSA survey, conducted a year before Labour's landslide victory the following year, and that obtained by the most recent survey undertaken towards the end of 2021.

Peter Pulzer famously wrote in the 1960s that ‘class is the basis of British politics, all else is embellishment and detail’.6 Thirty years later, the picture was no longer that simple. Even so, as the lower half of table 1 shows, support for the Conservatives was still rather higher among those in managerial and more junior non-manual occupations than it was among those in semi-routine and routine – that is, working-class – manual occupations. The opposite was true of Labour who, despite wanting to improve its support among middle-class voters, in fact enjoyed the support of a little over half of those in working-class jobs.

Now, after a general election in 2019 in which Labour lost a number of working-class ‘Red Wall’ seats (and with the party at the time of our survey still much less popular overall than it was in 1996), the Conservatives are only marginally more popular among those in managerial and professional jobs than among those employed in working-class jobs. Equally, Labour is only a little more popular among the former than the latter. So far at least, Labour's attempts under Sir Keir Starmer to reverse the especially heavy loss of support the party has suffered at recent elections among ‘traditional’ working-class voters appears have borne only limited fruit. In fact, the only group whose pattern of support remains as distinctive now as it was in 1996 is the self-employed who remain more inclined than any other group to support the Conservatives.

The point emerges clearly from table 2. In this table we use the BSA's value dimensions to divide the public in both 1996 and 2021 into: (i) the one-third most left-wing, the one-third most right-wing and the one-third in the centre; and (ii) the one-third most liberal, the one-third most authoritarian and the one-third in between. For each group it shows the level of both Conservative and Labour support. As we might anticipate, although the relationship is far from perfect, those on the left who are more concerned about inequality are more likely to back Labour, while those on the right are more inclined to back the Conservatives. Moreover, in the case of the Conservatives this divide matters as much now as it did in 1996, although the difference in Labour support between now and then appears rather more marked among those on the left (a point to which we will return below).

Liberals and authoritarians have also long tended to differ somewhat in their political preferences.7 In 1996, the Conservatives were a little more popular among authoritarians, Labour among liberals. But the difference was not as big as that between those on the left and those on the right. Now, in contrast, the party politics of liberals and authoritarians are markedly further apart. Indeed, the difference between them is not far short of the gap between left and right. Whereas in the 1990s, electoral politics in Britain was largely a one-dimensional battle between left and right, now it appears to be a more complex two-dimensional contest in which the distinction between a more liberal and a more authoritarian perspective matters too.

It might, though, be thought that the two dimensions are but two sides of the same coin, that is, that many of those on the left tend to be liberal and those on the right tend to be authoritarian. This is not the case. In 1996, people's position on one dimension was wholly unrelated to their stance on another, and even now, the link is only a small positive one.8 Appealing to voters’ values is no longer simply a question of adopting an appropriate stance on the question of inequality, but also one of having a position on issues of social cohesion and diversity. Irrespective of where voters stand on the left–right spectrum, support for Labour is now much higher among liberals than it is among authoritarians.

At the same time, Labour also finds itself now facing greater competition for the support of both those on the left and social liberals. Both the Greens and the Scottish National Party (SNP) are much more popular overall now than they were in 1996 – and their support is greatest among those on the left and among social liberals. As many as 18 per cent of those on the left and 20 per cent of social liberals now support the Greens or a nationalist party, compared with an equivalent figure of just 7 per cent of those on the right and among authoritarians. On either dimension, any decision to tack towards the centre now needs to balance the votes that might be gained against the trade-off of support that might potentially be lost on the left or among liberals, who may well feel they do have somewhere else they can go.

Of course, it might be thought that things have changed in the year or so since the most recent BSA survey was conducted. After all, at the end of October 2021, just as data collection for BSA 2021 was coming to an end, the Conservatives were ahead of Labour in the polls, much as they had been for most of the time since the 2019 general election. A year on, in contrast, after the fall and rise of two Conservative prime ministers in the wake of ‘partygate’ and financial turbulence on the markets, Labour was 25 points ahead. Surely this must mean that the pattern of party support has changed too?

In practice, however, this does not appear to be the case. On average, between October 2021 and October 2022, support for Labour (in terms of vote intentions) rose among those aged 65 and over by 13 points, from 22 per cent to 35 per cent. But it increased almost as much – by 12 points (from 50 per cent to 62 per cent) – among those aged 18–24.12 And while Labour support increased a little more among primarily manual ‘C2DE’ voters (by 19 points) than among primarily white-collar ‘ABC1’ voters (by 14 points), at 53 per cent the average level of support for the party among ‘C2DE’ voters was still little different from the 50 per cent registered among those in ‘ABC1’ occupations.13 Despite the dramatic rise in its support, Labour looks no more like a party that is especially successful at appealing to working-class voters than before, while the party does remain especially reliant on the backing of younger voters.

The polls do not carry the questions designed to measure people's values that appear each year on the BSA survey. However, given that how people voted in the 2016 EU referendum was strongly related to where people stood on the liberal–authoritarian dimension (and was unrelated to whether they were on the left or the right), we can use the relationship between how people voted in the EU referendum and their current vote intention as a proxy for the relationship between people's position on the liberal–authoritarian dimension and party support. This suggests that Labour continues to be more attractive to socially liberal Remainers than authoritarian Leavers. While, on average, support for Labour among those who voted Leave rose by 15 points (from 20 per cent to 35 per cent) between October 2021 and October 2022, it also rose by 15 points (from 45 per cent to 60 per cent) among those who voted Remain.