The Route to a Hung Parliament?
Greg Cook, Lexington
One of the features of Britain’s so-called First Past the Post electoral system is that parties can win majorities of seats with far less than 50% of the vote, sometimes as in 2005 and 2015 with well below 40% of the vote. However, the size of majority a party can win depends on a number of factors and not just the margin of its lead in votes. So in 1992 the Conservatives won a majority of just 20 despite a seven percentage point advantage, whereas in 2005 Labour won a majority of 64 on a lead of just three points.
In 2019 the consensus is that a Conservative lead of less than five points would make a hung parliament at least a possible, if not a probable, outcome despite the fact that in theory they require a swing of less than one per cent to win a majority. The average of current polls gives them a lead of 10% which would equate to a swing from Labour of roughly 4%.
If every seat in the country had the same swing this is how votes would translate into numbers of MPs:
|Con lead of||Equates to Lab to Con Swing of||Number of Con gains from Lab||Implied number of seats in total||Con Overall Majority|
While the Labour-Conservative swing is overwhelmingly the most important measure, the swing to other parties is what counts in seats where they are in first or second place. Currently polls are showing net swings from Conservatives to the Liberal Democrats and Scottish National Party, which may mean the Tories would lose some seats to them but too few for it to upset the overall pattern.
It is the potential distortions to the pattern of uniform swing and its relationship with seat output that make a hung parliament possible, of which there are many sources:
- Unequal Electorates
The current constituencies are now the most out of date there have ever been, drawn up on electorates taken in the year 2000. Two boundary reviews since then have not been implemented leaving constituency size varying between only 50,000 or so (mainly in Wales) to over 90,000. Unlike 1992 however, when the largest constituency electorates were mainly in safe Conservative areas and this was one of the main reasons the Tories won only a small majority, the recent trends in electorate growth and in particular the huge expansion in many inner city seats mean this will probably have little partisan effect.
- Incumbency Effects
There is no doubt that new MPs defending their seats for the first time on average do better than other candidates, and where a previous MP of another party lost the last time that effect may be magnified (double-incumbency). The scale is not quantifiable and will vary according to the type of seat it is (usually greater in rural seats and “Celtic fringes” of Cornwall, Wales and Scotland). Most MRP models do not take account of this because their individual constituency samples are too small. Many of Labour’s most marginal seats were gained last time and this factor could make a real difference, especially if the national swing is less than 3%.
Current polls are showing that the SNP have increased their support from 2017 mainly at Labour’s expense. The Tories would lose one or two seats on that swing but the pattern of anti-nationalist tactical voting and incumbency in Scotland will offset that.
The two big variables relate to the politics of the election and the extent to which Brexit patterns, assisted by organised tactical voting, reshape party support and in so doing render uniform swing if not redundant then incapacitated:
- Conservative Remain Seats
At the start of the campaign the expectation was that the Liberal Democrats would cut a swathe through the super-affluent seats which contain most of the pro-Remain ideologues, much as they had done in the European Election. Most but not all of the seats concerned are currently Conservative-held and others, notably Kensington, are seats the Tories would win on national swing; so this effect would definitely have impinged on their potential to win an overall majority. A number of constituency polls and to a lesser extent the YouGov MRP analysis have confirmed this is a real phenomenon. However none of them has actually shown the Lib Dems ahead in any of the constituencies concerned and all were taken at a time when they were stronger in the national polls. Even allowing for additional tactical voting it looks a tall order for them to gain many of these seats. In others their effect may be to assist the Tories by syphoning off Labour votes in key Labour-held marginals (Battersea, Croydon Central, Enfield Southgate) or others they are hoping to gain from the Tories.
Against expectations the Tories’ biggest problems appear to be outside London with constituency polls in Esher & Walton, South Cambridgeshire and Wokingham, all of which have Tory majorities of 15,000-25,000, showing them ahead by just single figures. If these are real there are several other seats in the high-tech Remainer clusters of the A1, M3 and M4 corridors where no polls have been done that may be equally competitive. Again though, the Lib Dems would actually have to win some of them for this to be an impediment to a Tory majority.
- Red Wall
Indeed, the scenario described above as it stands might actually lead to the opposite distortion, one which enhanced the Tory majority. If the Tory vote is really under the stress that it seems in the South East region, but they are losing few seats, then there must be other parts of the country where they are performing better and may gain more. The YouGov MRP accordingly showed a number of seats where they estimate the pro-Tory swing to be well above 5%. The stereotype of the seats concerned is that they are white ex-coalfield or “post-industrial” cities and towns but really they are much more diverse. For example there is a cluster of four Labour-held marginals in North East Wales, one of the remaining high-skilled manufacturing heartlands, another in the West Midlands conurbation, which is demographically very different and includes some BAME communities. What unites most of them is that they lack much of a professional middle class.
If the Tories sustain their national lead at its current level then there are many more seats of this kind which they can gain on an above average swing than they might lose on regionalised counter-swing.
What Might Change?
If the Tories maintain a lead of six points or more then it is very likely that will bring them a majority of some scale. It needs a change in the national position for any of these variables to become significant. One source of that might be a complete collapse of the Lib Dems which enabled Labour to push up towards 40% of the vote and replicate the 2017 result. Even then, it does seem that the Lib Dems have re-established themselves as a credible option, which can only limit Labour’s scope to add to its vote and actually gain marginal seats.
The Brexit politics of this election seem almost certain to create some extraordinary results somewhere and the Liberal Democrats remain the most likely means by which the result may be distorted. It requires them to establish momentum over the final week of the campaign such that they are able to maximise their reach, including to vulnerable Tory voters in the seats where those extraordinary results are achievable. For now, though, the Tories remain in pole position.