Boris Johnson, Brexit and electoral blowback: the perils of a late year general election

Clifford Williamson | 01 October 2019

With all the talk of a general election in the autumn to take advantage of a ‘Boris Bounce’ and public anger at Europe, Number 10 would do well to learn the lessons of history over the dangers of an electoral contest in autumn-wintertime — especially a ‘who governs Britain’ type of campaign.

In the twentieth century there were 13 general elections fought in autumn-winter and in 11 of those the incumbent party lost seats (I have excluded the 1918 and 1931 elections as they were fought on a national ticket and therefore not comparable). The number of seats lost ranged from 246 in the 1906 election to just 2 in December 1910, but on average 61 seats were lost by the party in power at the dissolution.

Manufacturing consent

The purpose behind Number 10’s current thinking is said to be manufacturing consent for a no-deal Brexit, or alternatively to make a quick dash for votes following an exit from the EU. Yet there are some very significant examples of attempts to use an election to achieve this type of purpose: in 1973, in the two elections of 1910, and in 1923. These all backfired in various ways.

The most obvious parallel to the current situation is that faced by Edward Heath in late 1973 following the start of the second miners’ strike of his administration and the declaration of a state of emergency. The PM sought a renewed mandate (he did not have to go to the country until at least the middle of 1974) on the question of ‘Who Governs Britain’, and although the election in Feb 1974 was somewhat inconclusive, one message that came through was ‘not you, Ted’.

In January 1910, the Asquith administration sought a new mandate in the ‘people vs peers’ struggle over the House of Lords’ blocking of the 1909 Budget, that had introduced taxes on the wealthy to fund ambitious social reforms and the first state old age pensions. It should have been a simple matter of mobilizing the people against privilege. Yet it boomeranged spectacularly with the Liberals losing 123 seats and their overall majority, forcing them into relying on support of the Irish Nationalists. A second election in December that year was equally indecisive.

Like the 1910 Liberal government that sought a mandate for specific policies, the first Baldwin administration with its majority of 77 went to the polls over the issue of protectionism and the end of Free Trade known as tariff reform. This issue had been a source of division, much like Europe now, for over 20 years in the Conservative Party. The party lost 86 seats and paved the way for Ramsey MacDonald to become Prime Minister in a minority Labour government.

Now the good news

Although history is generally against incumbents, there are two examples where an autumn/winter poll was successful and these are 1959 and 1974. The most intriguing and totally against the trend was 1959, when Macmillan went to the country in October. The election was fought off the back of a series of diplomatic triumphs involving Khrushchev and Eisenhower, and more importantly in the midst of a ‘never had it so good’ economic boom, and it earned Harold Macmillan an increased majority for the Conservatives through gaining 20 seats and the sobriquet of ‘Supermac’. The October 1974 Election was the second election that year with Labour making 18 gains overall and gaining a slim working majority.

If an autumn or winter election seems too risky, Boris Johnson could delay till the summer. However, two examples, 1978 and 2007, stand out as cases where baulking at going to the country has been seen as a disaster.

During the summer of 1978 James Callaghan wrestled with the idea of an early general election. Labour’s opinion poll fortunes had improved following the economic catastrophes that had forced the government to seek an emergency loan from the IMF in 1976 and inflation had fallen from its 25%+ high in 1977. The PM however hesitated and in September at the TUC conference memorably announced in song that there would be no general election that year. The election when it did come in May 1979 on the back of the Winter of Discontent and a vote of no confidence resulted in a Labour defeat and 18 years of Conservative rule.

Nearly thirty years later Gordon Brown, who like Callaghan, had inherited the premiership rather than won a mandate at the polls, toyed with an early general election, the previous one having only been held in 2005. His decision not to has also been viewed in retrospect as a major error of judgment considering the advent of the financial crisis in 2008 and the discrediting of his management of the economy.

A final factor to consider is that there has not been an autumn-winter general election in the UK since October 1974. The conventional wisdom is that, where possible, governments prefer spring-summer elections to maximize hours of daylight and the prospect of better weather that could help turnout, though there is no empirical evidence to support this contention. Turnout has been comparatively low for all of the last 40 years of spring-summer elections in the UK, especially since 2001 when on average less than 70% (60% in 2001) of the electorate have voted in each election.  Indeed, the highest turnouts of the 20^th century were in January-February 1910 with 86.6% and in February 1950 with 84%. In both, the incumbents lost a large number of seats.

To judge from most of the twentieth century, a general election in autumn-winter is a recipe for unexpected blowback, whether due to disrupting preparations for Christmas, Seasonal Affective Disorder or the weather. Any planners pushing for one should pay heed to history’s lessons.

• Tags:
• Williamson, Clifford
• Power and politics

About the author

---

Clifford Williamson is Lecturer in History at Bath Spa University.