What would happen at GE2019 if it is more like the 2019 European election than the last general election?

Previously, I looked at what might happen at a future general election using the 2017 general election results as a base and a YouGov poll to redistribute votes to see how a new general election might pan out. This poll gave the Tories a 9 point lead over Labour in second place, and accordingly in such a scenario the Tories would win a majority despite a historically low vote share, albeit with a smaller majority if the Lib Dems, Greens and Plaid Cymru were to form a pro-remain electoral pact.

But what if the distribution of electoral support has moved so much since the 2017 election that it doesn’t make sense to use it as baseline? I thought it might be interesting to see what would happen if the 2019 European election was used as a baseline instead.

Historically, European elections have been poor at predicting future general elections. Ukip won the popular vote at the 2014 European elections when they received 26.6% of the vote, but at the general election of 2015 they received only 12.6% of the vote and won only one seat. However, the gains made by the Liberal Democrats at this year’s European and locals might be an indicator that they are well placed to make gains at a Westminster election, too.

Today, Survation released a new general election poll giving the Conservatives a vote share of 28%, four points ahead of Labour on 24%, with the Lib Dems on 21% and the Brexit Party on 15%. I used Chris Hanretty’s estimates for the 2019 European elections at a Westminster constituency and adjusted these figures by the figures given by Survation for voting intention split by how respondents voted at the European election. The graphic at the top of the post shows the winner in each constituency at the 2017 general election, the 2019 European election, and a general election projection.

Using the European elections and the Survation poll leads to a projection of 344 seats for the Conservatives (up 27 from 2017), 147 for Labour (down 115), 81 for the Lib Dems (up 69), 54 for the SNP (up 19), and six for Plaid Cymru (up 2).

Whilst this would give the Conservatives a majority in the House of Commons, it is significantly smaller than the 400 or so suggested by using the YouGov poll and the 2017 general election as a base. There a number of reasons why this might be the case.

Firstly, the Lib Dem surge in affluent Remain areas in the southeast of England, as evidenced by the European and local elections, could put the party within range of victory in traditional Tory heartland seats such as Maidenhead and Witney (the seats of the last and last-but-one prime ministers). Secondly, the fragmented and unpredictable nature of first-past-the-post elections in four and five-party races means that predictions are difficult and small movements in the polls can lead to significant movements in terms of projected seats. Having a 9 point lead is a very different proposition to having a 4 point lead.

However, all these calculations are based on there being no “Unite to Remain” pact between the Liberal Democrats, the Greens and Plaid Cymru. The map below shows the same projection as above, except that the votes of those three parties have been merged into one unity candidate.

In this scenario, a pact between the Liberal Democrats, the Greens and Plaid Cymru would be sufficient to deny the Conservatives a majority.

It is inherently difficult to predict which way a general election held this year might unfold due to the unpredictability of multi-party elections in a first-past-the-post electoral system. However, if the Tories are to hold on to 10 Downing Street then they will hope that their anti-Brexit opponents remain divided whilst they protect their hardline Brexit flank against the Brexit Party. Minor movements in public sentiment could have a very large impact.

The model data can be found here.