Dr Chris Stevens considers lessons from previous elections.

Predicting elections is a fool’s game. I was at an election party of academics in 1992. When we went around the room at the start of the evening, I alone predicted a Conservative majority, which I put at 20.  As the night wore on, and, as a victory for John Major edged closer, those in the room began to blame me personally for the Tory win. On the face of it my prediction of 20 was impressive. Major won a majority of 21. In reality, my prediction was completely wrong, I had anticipated that Major would win far fewer votes than he did. In fact, he won more votes than anyone had before, beating the previous high-point of 1951, when Attlee won more votes, but fewer seats than Winston Churchill. Major won so many votes, he should have had a majority matching those secured by Margaret Thatcher in 1983 and 1987.  The problem for him was anti-Tory voting in marginals constituencies.

In my last blog, I predicted the the next election would see a hung parliament with Labour as the biggest party.  This was based on a number of factors. First, Labour starts from a very low position. To win a majority, Keir Starmer needs a swing bigger than that achieved by Tony Blair in 1997. Second, given SNP dominance in Scotland, and the reduction of the number of Scottish seats after devolution, Labour will need to win in traditional Tory areas.  This it did in 1945, 1966, 1997 and 2001, but it will be much harder next time, given different political dynamics.  Third, an anti-Tory majority will not necessarily translate into a Labour majority

That blog immediately seemed wrong. Just before it went to press, we had a poll of 28,000 voters that suggested a large Labour majority. Then, we had anticipation in the press of a Conservative recovery based on Rishi Sunak’s steadying of the Conservative ship. Then, we had the prediction that a possible collapse of the SNP would benefit Labour. Then, the May 2023 local elections suggested a hung Parliament. Now, we have had three fateful bye-elections on 20 July 2023.

The message delivered by Conservative politicians on local election night was the results show that Labour cannot win an overall majority.  When politicians go into elections minimising their prospects in order to create low expectations and claim unexpected successes, they do so not just to make themselves feel good and rally their supporters, but because they know that political optimism is an electoral asset. The Conservative Party was probably pleased, therefore, that the broadcast media tended to follow the Tory message and make Labour the news story. In a similar vein, the Conservative party will be delighted that the focus of much of the broadcast media on July’s bye-elections is on Starmer’s leadership and policy choices.

A hung Parliament seemed plausible following the local elections because of the assumption that governing parties close the gap on their opponents as the next elections draws closer. So, Conservative spokespersons have been at pains to emphasise that their electoral travails are part of the normal political cycle, and that recovery will occur. Boris Johnson put down the massive bye-election defeats to the Liberal Democrats at North Shropshire in December 2021 and Tiverton and Honiton in June 2022, to mid-term blues, something that few took seriously but few bothered to refute.  Before last week’s elections, Rishi Sunak and others said that Governments did not win bye-elections. He was right up to a point.  The Conservative Party did not win any of the 22 bye-elections between 1992 and 1997, but the Labour Party won five of the sixteen bye-elections between 1997 and 2010.

Sunak will hope that by waiting until the last minute, he can benefit from an upswing.  He is likely, therefore, to ignore any advice that a general election in May to coincide with the next local elections will enable him to avoid another pre-general-election drubbing at the polls. Prime Ministers tend to hold elections after four years when things are going well, as Thatcher did in 1983 and 1987 and Blair did in 2001. It is counter-intuitive, however, to hold an election in order to minimise the scale of a probable loss, because something might always turn up for the better.  So prime ministers tend to wait for five years when things are going badly, as Callaghan did in 1979, Major did in 1992 and 1997 and Brown did in 2010. Waiting can be a disaster. When a general election was widely expected in 1978, Callaghan used the party conference to announce that he would wait to the following year. Had he gone to the polls in 1978, he might still have lost, but he would have avoided the “Winter of Discontent”, and Thatcherism might never have found its feet. The one exception to this dynamic is the 1992 election and that brings me back to my opening comments and the main point of this blog.

It is a commonplace that politicians tend to devise their election strategy by re-running the last election: the winner to replicate it, the loser to replay it.  Now, however, neither Sunak nor Starmer, nor the media are giving any thought to the 2019 election. Minds are fixed firmly on 1992. The dominant narrative of 1992 is that opinion polls put Neil Kinnock and Labour in the lead, but Major went on to win and that this was because by replacing Thatcher with Major, the Tories had refreshed their appeal. Conservative politicians hope that we are in a 1992 situation; Labour politicians fear that we might be. Both have been encouraged to do so by the media.  Ever since the high-point of the Boris Johnson premiership, the liberal/left press has warned that Sunak will draw a line under the bad times, and present his government as something new, as they believe that Major did in 1992.

We are not, however, in a position analogous to 1992.  The Conservative Party may have been behind in the opinion polls but it maintained a reputation for economic competence, which Labour still lacked, and this was indeed reflected in opinion polls. As I argued in the academic journal, Contemporary British History, in 2002, the statecraft strategy used effectively by the Conservative party in the 1980s, which utilised the narrative of the “Winter of Discontent” to mobilise bias against Labour, had not yet played out. It is not out-of the-question that Thatcher would have won in 1992 had she been able to cling on to the leadership. Bye-election evidence showed that there was clearly little enthusiasm for a Kinnock government, a point made repeatedly by Professor Anthony King, who was playing the media role that John Curtis now occupies.  Bye-elections, he argued, turn the potential of opinion polls into electoral reality.  Prior to 1992, discontent with the Tory Government saw large bye-election swings to the Liberal Democrats, but, where the Labour Party was the main contender, the latter was only succeeding in creeping over the finishing line.

The game changer was Norman Lamont’s chaotic handling of the ERM crisis in September 1992 which shattered the Conservative Party’s reputation for economic competence.  The political and economic re-positioning undertaken by John Smith, as Kinnock’s successor in 1992, and by Gordon Brown, as his shadow chancellor, enabled Labour to take advantage of this.  The bye-election in Barking in June 1994 produced a swing of 20.5% to Labour; the Conservatives went from second to third place and their candidate, Theresa May, won fewer than 2,000 votes, a swing against her of 23.5%. Tony Blair, who became leader a month later, may have iced the Labour cake, but it seems likely that the Party would have won under Smith in 1997, had he lived. The damage done by Lamont to the Tory Party was so great, and opened so many fissures in its ranks, that the competent economic management of Ken Clarke, his successor as chancellor, was largely unnoticed. The credible claim that Clarke had provided Blair with the financial headroom to fund optimistic, expansionist policies without tax increases in 1997 went unheeded.

So what does this tell us about the next election? First, this is not 1992.  Labour’s local election successes and its victory in Selby and Ainsty are evidence of positive support for Labour of the sort that did not occur until after the 1992 election.  There will also be no easy recovery from the Kwasi  Kwarteng budget which may damage the Tories more than the ERM fiasco.  Labour critics of Starmer might also reflect on how well he has contextualised Sunak’s premiership as an intrinsic part of Conservative governance since 2010 rather than a post-Johnson renewal. Second, despite this, an effective Labour majority still appears a way off, given the swing required. One irony is that the financial incompetence of the Truss interlude has forced Starmer on to the economic back foot. One pathway for Labour would be an SNP collapse in Scotland, but despite unionist optimism there is still no evidence of that and a comfortable Labour win in the forthcoming bye-election in Rutherglen and Hamilton West might still be a false dawn. Another pathway was highlighted by John Curtis on Channel Four news last week.  Anti-Tory voting was a major factor in 1997.  In that year, the Labour and Liberal Democrat parties were only retaining 6-7% of their vote if they were not best placed to beat the Conservative party.  In last week’s three bye-elections, the Liberals polled only 3.3% in Selby and Ainsty and 1.7% in Uxbridge and Ruislip, while Labour polled only 2.6% in Somerton and Frome.  This may squeeze the Conservative Party further.

There is one last point worth making.  During the bye-elections Conservative spokespeople said repeatedly and emphatically that their bye-election losses were due to their own supporters abstaining and they found no enthusiasm for Labour on the doorstep. Well they would say that wouldn’t they! And it’s hard to argue against, without knowing exactly who stayed away. One politics academic, whom I will not name, went surprisingly further on Channel Four News, stating that the small increase in the Labour vote in Selby proved that there were few switching from Tory to Labour.  Andrew Rawnsley made a similar point in the Observer on 23 July, when he wrote that ‘there is still not that much enthusiasm for Sir Keir’s party. In Selby, the Tory vote collapsed by more than 21,000. The Labour vote only rose by less than 3,000.’ Those who share that view may need to reflect that in the Dudley West Bye-election in December 1994, which was taken to herald the beginning of the Blair ascendancy, there was a collapse of the Conservative vote by 21,234 and a swing to the Labour Party of 28.1%, but the Labour vote actually went down by 540. Blair must have been really unpopular!

Photo by Recal Media: https://www.pexels.com/photo/yellow-concrete-cathedral-under-sky-372038/