Since the mid-1970s governments in the UK and Australia (the ‘Westminster systems’) have on average won three full terms of office, fifteen years in the UK, and nine years in Australia. Incumbent governments appear strongly favoured to win re-election once, twice, or even three times; beyond this, the likelihood of defeat rises sharply. Only one of the seven (eventually) defeated governments below won a fifth election (Hawke-Keating), and only one has been in office for more than fifteen years (Thatcher-Major).
UK: Mean (1979-2024; assuming Conservative loss in 2024) 15 years or 3 full terms
CONSERVATIVE (Thatcher-Major): 1979, 1983, 1987, 1992 (18 years).
LABOUR (Blair-Brown): 1997, 2001, 2005 (13 years).
CONSERVATIVE (Current government, Cameron-May-Johnson-Truss-Sunak): 2010*, 2015, 2017**, 2019 (14 years).
*2010-2015 was a period of coalition government with the Liberal Democrats.
**2017-2019 was a period of minority government.
AUSTRALIA: Mean (1975-2022) 9.4 years or 3 full terms
LIBERAL (Fraser): 1975; 1977; 1980 (8 years).
LABOR (Hawke-Keating): 1983; 1984; 1987; 1990; 1993 (13 years).
LIBERAL (Howard): 1996; 1998; 2001; 2004 (11 years).
LABOR (Rudd-Gillard-Rudd): 2007; 2010* (6 years).
LIBERAL (Abbott-Turnbull-Morrison): 2013; 2016*; 2019 (9 years).
LABOR (Current government, Albanese): 2022+ (2 years+).
*2010-2013 was a period of minority government.
*Nov-Dec 2017 was a short period of minority government.
Explaining why the length of time in office was a contributing factor in these defeats helps inform us about the prospects for the current Conservative government in the 2024 UK election. Of course, each of these eleven UK elections and eighteen Australian elections has its own specific election narrative, a story developed or fleshed out in the mainstream media and by election commentators, pundits and bloggers, about why this particular election was won and lost. The problem with these explanations is they tend to treat these elections as isolated from each other, and assume that the election campaign itself is somehow decisive. They struggle to explain the consistency with which governments since the mid-1970s have managed to cling on to power.
What I’m suggesting is that election campaigns are not usually decisive in election results. Rather, elections have a default result caused by incumbency. All of the eight governments above were re-elected for a second term; in seven of the eight cases, a third term; and in four cases – 50% of the time – a fourth term. In several of these winning elections, governments went to the electorate whilst being behind, or well behind, in opinion polling. Stand out re-election wins for governments would feature Bob Hawke’s 1990 win on the back of Green Party preferences; Paul Keating’s 1993 ‘victory for the true believers’; John Howard’s 1998 GST re-election bid which the government won whilst losing the popular vote; and the UK re-election victories for Major in 1992, Cameron in 2015 and Johnson in 2019. And while in the aftermath of isolated election campaigns the narrative tends to focus on the losing side’s deficiencies (from Oh, Jeremy Corbyn! to Woah, Jeremy Corbyn!), these narratives seem problematic in the big picture – no matter how much soul-searching they inspire in opposition benches.
So, while pundits may point to Michael Foot’s Labour Manifesto of 1983 as ‘the longest suicide note in history’, this is unlikely to have really been the driving force behind Thatcher’s 1983 Conservative re-election victory. If William Hague in 2001 had staked out a more right wing policy position – or alternatively a more centrist one – it’s a bit of stretch to think he’d have then won a landslide, instead of being on the receiving end of one in Blair’s successful 2005 re-election. Blaming John Howard’s policy positions, personality or campaigning skills after the Liberal party lost to Bob Hawke in Labor’s 1987 re-election suffers from the problem that Howard went on to win four terms between 1996 and 2007. It helps to think of electorates as preferring, as a rule, the current government to the current opposition no matter of what political stripes. Better the Devil You Know, in effect: A government can point to its record, what oppositions offer is uncertain in practice, no matter what its substantive manifesto content may be. So while individual elections will produce narrative accounts specific to that election, such proximate factors may matter a lot less than more long-term considerations.
