75 more votes in the right places – that’s all the Conservatives would have needed to govern alone

If just 75 people in the right constituencies had voted differently in the 2017 General Election, Theresa May would now have a working majority in the Commons. Chris Terry (Electoral Reform Society) examines some of the more extraordinary anomalies thrown up by the first-past-the-post system this year. In North East Fife the SNP beat the Lib Dems by just two votes, but under the ‘winner takes all’ system all the Lib Dem votes in the constituency were wasted ones. Similarly, Labour lost out because it piled up votes in areas where it was already certain to win. He argues for an electoral system in which every vote carries equal weight.

Photo: duncan c via a CC-BY-NC 2.0 licence

Britain’s first-past-the-post (FPTP) electoral system (in)famously results in marginal and safe seats where many votes are considered ‘wasted’ – but the 2017 General Election has shown just what random results that can throw up. The existence of such seats is a feature of a FPTP electoral system, not a bug. FPTP is meant to function so that there are two sets of safe seats for each of the main parties – and as the pendulum swings from one party to another, the marginal seats switch from one party to another. The party winning the majority of seats then forms the government.

But this results in major democratic issues. It means that, for instance, election battles become concentrated on a smaller number of voters. These areas and voters are targeted relentlessly and often become the battleground of the election. In turn, these swing seats see a huge amount of tactical voting based on who might beat the ‘greater of two evils’. We estimate that six and a half million people ‘held their nose’ on June 8. With one in five voters trying to second-guess each other, we are left with a ‘lottery election’ where casting a ballot is like casting a die.

Partly as a result, this election saw the second highest electoral volatility (the movement of votes between parties) since 1931. People are switching sides and shopping around at astonishing levels, as they try and game the system. That’s a context which makes it hard for parties to ‘target’ accurately. This election, they got their targeting wrong – and hence the ‘true battlegrounds’ were not where they thought they were.

Attention has been lavished on surprising gains, but less noticed has been the consolidation of some safe seats into ultra-safe seats – places such as Liverpool Walton, where the Labour majority is 77%, Birmingham Ladywood with a Labour majority of 69.5% or East Ham where it is 70.5%. Indeed, Labour now holds 34/35 of the safest seats, while the Conservatives hold none. This creates an inefficiency in the Labour vote. By piling up votes in safe seats, it is harder for the party to win a majority nationally. To win elections, it is best to win individual seats by small rather than large margins, and to lose by large ones: to do the opposite is inefficient, as it makes it more likely that a party will lose an election even if they win the most votes.

Although supermajorities of the size of Labour’s are less common for the Conservatives, some sizeable majorities still exist. Christchurch has a Conservative majority of 49.7%, South Holland and the Deepings 49.5% and Hampshire North East 48.2%.

While in many ways it may be satisfying for campaigners to achieve gigantic majorities in their home areas, once a party has achieved first place in a constituency under FPTP those extra votes do nothing to help secure more seats in the House of Commons. They are, put simply, wasted. Meanwhile, at the opposite end we have ultra-marginal seats. Eleven seats were won by fewer than 100 votes. The most marginal seat after the 2017 election is North East Fife, where the SNP beat the Lib Dems by a mere two votes. If one SNP voter had gone to the Lib Dems the seat would have tied, resulting in the drawing of lots. If two had gone to the Lib Dems then they would have won the seat – and yet the SNP will represent all of the voters in that seat until the next election.

At the same time, the Conservatives could have held Kensington if just 12 voters had changed their minds – that’s 100% of local power hinging on a dozen votes.

These bizarre binaries are much reduced under multi-member PR systems. Indeed, if 75 voters in the right seats had voted differently the Conservatives would have a working majority, meaning a deal with the DUP would have been unnecessary. Just 75 votes switching hands would have given the Conservatives a working majority, 533 votes switching hands would have given the Conservatives an absolute majority. 645 voters could have deprived the Conservatives of a majority with the DUP, depriving them of a deal.

These figures demonstrate the relatively tiny number of people who actively change election results in the UK. The importance of marginal campaigns is such that Professor Justin Fisher has said that the idea of a national campaign may, effectively, be dead, as national resources are spent almost entirely on supporting constituency level campaigns. In addition to this issue with the national campaign, an analysis of the local, constituency-based campaigns at the 2010 election was carried out by the Electoral Reform Society in 2013, finding that the difference in expenditure was 22 times as high in the constituency with the highest campaign spend per vote as it was in the lowest.

This practice means that people and areas can, rightly or wrongly, feel ignored – as many in Scotland felt before 2014. Two things hold: votes should count everywhere and hold equal value, and people should be represented equally regardless of where they happen to live. The case for these arguments is now stronger than ever.

This post represents the views of the author and not those of Democratic Audit.

Chris Terry is a Research Officer at the Electoral Reform Society.