Lewis Baston is author of Reggie: The Life of Reginald Maudling and several books about British general elections. He is a consultant on politics, elections and constituencies.

The Parliament elected in 2010 is coming to the end of its life; there are now fewer than six months to go. What does history tell us about what we can expect to change between now and polling day?

The classic pattern of public opinion in a parliament – Government honeymoon, mid-term discontent, Government recovery in the run-up to a general election – happens less often than one might believe. The 1992-97 government had 57 months of mid-term blues, while the 1997-2001 government had two weeks in September 2000. The Labour government of 2001-05 had a very mild mid-term dip, but then did not have much of a pre-election recovery either. The parliaments of 1974-79, 1979-83 and 2005-10 each saw several wild swings of opinion. Changes in opinion since 2010 have been interestingly gradual.


Six months out Election (GB) Swing
Con Lab Lib/ Dem Con Lab Lib/ Dem
1964 39 51 10 43 45 11 5% to Con (Govt)
1970 48 41   8 46 44 8 3% to Lab (Govt)
1973/74 34 42 23 39 38 20 4% to Con (Govt)
1978/79 46 45   5 45 38 15 3% to Con (Opp)
1983 43 35 21 44 28 26 5% to Con (Govt)
1987 39 38 21 43 32 23 5% to Con (Govt)
1991/92 40 41 15 43 35 18 4% to Con (govt)
1996/97 30 51 14 31 44 17 4% to Con (Oovt)
2001 32 49 14 33 42 19 4% to Con (Opp)
2004/05 32 38 21 33 36 23 2% to Con (Opp)
2009/10 40 28 18 37 30 24 3% to Lab (Govt)
2014/15 32 34   8


Several interesting trends emerge from the numbers. Labour consistently end up winning fewer votes in the general election than the polls would have suggested six months in advance. In some elections, events over the last six months of the parliament have gone against Labour, most notably in 1978/79. In others, polling error was partly responsible, and the voting intention figures did not reflect reality, most notably in 1992 but also probably in 1997.

In some there is regression towards the mean; the chances are that a party exceeding 45 per cent is being flattered by sampling variations and the inclusion of a number of weak supporters who will be dislodged during an election campaign. There seems little danger of this in 2015. The timetable matters: past Prime Ministers from Eden to Blair have banked their advantage at around the four-year mark. The uncertainties of a fifth year were to be avoided if possible. But this time, Parliament will go to five years anyway.[1]

Governments gain more often than oppositions: if my Conservative-supporting readers want some comfort, there are no cases of a Labour opposition gaining ground over the final six months. Most of the best Conservative showings, though, are from the 1979-92 period when the party was most dominant in campaign skills and funding, and polling was inaccurate.

But for the pessimists (and UKIP supporters) among you, people clearly do not abandon their flirtations with ‘minor’ parties and go back to the big two when the election draws near. In the past, the beneficiary has been the Liberals or Liberal Democrats, but even the most optimistic among them does not expect to repeat the pattern this time. The Lib Dems have never (at least in polling history; one suspects that Rosebery’s ratings would probably have been in negative territory in 1895) fought an election with an unpopular leader before. Nick Clegg is about to break this pattern in fine style. UKIP, the SNP and the Greens stand to inherit the protest vote, which will probably not subside much. It could even snowball if the electorate – which is ever more weakly aligned to the parties – feels in vindictive mood towards politics as usual.

But what about the economy? A curious pattern in polling history is the strong link between elections and spikes in economic optimism. Past Conservative governments have seemed able to manufacture big bounces in economic optimism just in time for an election. The only Labour government to have managed this was Blair’s in 2001 – perhaps significantly, the election at which Labour enjoyed its highest support from the press and at which Alistair Campbell was at the height of his powers.

Chart: Net economic optimism (per cent) in the months either side of general elections under the Conservatives



The optimal time for the Conservatives to have held an election might have been autumn 2013; the change in economic optimism from net -30 in March to net +23 in September in that year was a very rapid turnaround. Since summer 2013 economic expectations have remained positive, although the balance has become less positive since the peak in summer 2014. Given the huge efforts that George Osborne and his colleagues, and the media, have poured into talking up the economy in 2013-14, the risk is that the effort cannot be sustained, and the narrative will shift towards emphasising problems and dark clouds on the horizon.

It may be noted that the outgoing Major and Brown governments in 1997 and 2010 were both voted out by electorates who were optimistic about the economy. Perhaps blood-curdling admissions of economic failure are the best – though high risk – electoral strategy, scaring the electors into keeping hold of nurse for fear of something worse, as they did in 1992. In 1964, Reggie Maudling (one assumes jokingly) asked his Cabinet colleagues whether they thought a pre-election sterling crisis would be politically helpful or not.

We have always, really, been guessing about the relationship between good and bad economic news (both real and perceived) on the psychology of the electorate. But the environment in 2014-15 introduces even more ‘known unknowns’.

The two main parties are starting with the smallest combined share of the vote in history (66 per cent, compared to over 80 per cent in most elections up to 2001). The shape of the political landscape has also changed. As late as 1997, the two big parties had core votes and competed for floating votes in the same centre pool; the Lib Dems also inhabited the centre. But the ‘centre’ is now a tired metaphor that fails to capture the complexity of political competition. It is far from certain that the UKIP and SNP bandwagons, and the rather smaller Green one, will grind to a halt before polling day. These are strange times – strange enough for me to venture a Gramsci quotation in a ConservativeHome article:

“The crisis consists precisely in the fact that the old is dying and the new cannot be born; in this interregnum a great variety of morbid symptoms appear.”

Further reading

The Nuffield series of studies of General Elections are always worth picking up: one might be particularly interested in David Butler and Dennis Kavanagh The British General Election of February 1974 (Macmillan, 1974) at the moment.

The Lib Dem commentator Mark Pack has done sterling work in compiling as complete a spreadsheet of voting intention polls as is humanly possible, and I owe him thanks.

[1] See my piece in Progress for more about this.

4 comments for: Lewis Baston: Six months out, what does the past tell us about how much change to expect before polling day?

Leave a Reply

You must be logged in to post a comment.