In 2019 there were local elections in May, European Parliament elections in June and a general election in December. This Insight looks at how the general election result compares to the results of these other elections.
General election results are counted and declared at constituency level, while local and European Parliament election results are counted and declared at local council area level. This means general election results cannot easily be compared to local and European election results. However, academics have modelled these results to get an indication of what the results would look like at national and/or constituency level.
Local elections: Indicating a ‘hung Parliament’
Local elections took place in 248 councils in May 2019 in England and 11 councils in Northern Ireland. There were local mayoral elections in six English councils too (for more information, see Results of the 2019 local elections in England).
Professor John Curtice produced an estimate of how political parties would have fared across Britain if this had been a general election (assuming voters behaved the same) for the BBC. His ‘projected national share of the vote’ (PNS) suggested that Labour and the Conservatives would each have won 28% of the vote in Great Britain, with the Liberal Democrats taking 19%.
Sky News produced its own estimate of how the local election results would translate into a general election result. It suggested that the result indicated a House of Commons made up of 280 Conservative MPs, 268 Labour MPs and 28 Liberal Democrat MPs. The Guardian, reporting this estimate, said the likely outcome of such a Parliament would be a minority administration run by Jeremy Corbyn and supported by the SNP and Lib Dems, which would probably result in a second referendum on EU membership.
The charts below show how these estimates compare to the actual results achieved in the general election held later in the year.
European Parliament elections: Indicating a Brexit Party Government
Professor Chris Hanretty has modelled the results of the June 2019 European Parliament elections to estimate how constituencies would have voted if it had been a general election. His model shows that the Brexit Party would have won the most Westminster seats. At the time, he warned against seeing this as a predictor of the party’s performance at the next general election; in 2014 UKIP won the most votes in 44% of local authority areas, but the party won just one seat in the 2015 General Election.
The chart below on the left shows the vote shares parties won at the 2019 European Parliament and general election. The chart on the right compares the estimated number of MPs parties would have won based on their performance at the European Parliament elections, with the number they actually won at the general election.
While these charts highlight substantial differences between the June and December elections, it is worth noting that the combined vote share of the Brexit Party and Conservative Party, the two major Leave supporting parties, were fairly similar: 41% at the European Parliament elections and 46% at the general election.
Estimated constituency vote share
Another way to compare how parties did at these two elections is to look at the difference between their estimated constituency vote share at the European elections, and the vote share they won at the general election. Averaged across all constituencies, the Conservatives won 35 percentage points more in December than they did in June; Labour won 19 percentage points more. The Brexit Party won an average of 30 percentage points fewer in December than in June.
Why do people vote differently in different types of elections?
It’s fair to conclude that the results of the local and European Parliament elections held in 2019 were not a reliable indicator of the general election. This is due to a few factors.
Firstly, several months passed between the May and June and the December elections, during which politics moved on and the country changed Prime Ministers.
Secondly, turnout is substantially different at different types of elections. It is generally around 35% at local elections, it was 37.2% at the European Parliament elections, and 67% at the general election. The smaller subset of people voting in local and European Parliament elections may not be representative of the general population;
Finally, people are motivated by different reasons when they vote in different types of elections. There is evidence to suggest that local and European Parliament elections are considered by political parties and voters as ‘second-order’ elections: they are seen as less important because they do not result in the formation of a government. Governing parties tend to lose votes at these elections as people express their discontent with government policies, and because people are not worried about ‘wasting’ their vote on smaller parties. People may also have different policy preferences at different levels of governance (see for example this article on the European Parliament). That said, academics Rallings and Thrasher show that in the UK, local elections were seen as more important than European Parliament elections. This is in line with the vote shares recorded in the 2019 elections, as shown in the chart below: Labour and the Conservatives won most at the general election, least at the European Parliament elections, and were somewhere in the middle at the local elections (based on the projected national share of the vote, or PNS, described above).
Further reading
General Election 2019: Brexit, House of Commons Library.
General Election 2019: Turning votes into seats, House of Commons Library.
General Election 2019: Marginality, House of Commons Library.
About the author: Elise Uberoi is a researcher specialising in social and general statistics at the House of Commons Library.