Tag Archives: tactical voting

Tactical Voting in the European Parliamentary Elections 2019

Summary

Any remain voter in the Eastern Region wishing to maximise the number of pro-European MEPs needs to vote tactically, but it is difficult to predict how to do this.

Using the latest poll[1] :

  • No party with a vote share below 9% can elect an MEP. On the latest opinion poll, that excludes Conservatives, Green and ChangeUK
  • Five of the Region’s seven seats are “safe” (3 Brexit, 1 Labour and 1 LibDem).
  • The parties are evenly balanced for the remaining two seats. The best chance of maximising the remain seats is to vote tactically for ChangeUK, who would win one seat if they raise their vote share by a third.
  • Greens would need a 50% increase in vote share to win one seat
  • LIbDems would need to double their vote share to win a second seat

It would be unwise to make a decision based on these figures, since a lot can change in the 18 days before Election Day. However, postal voting will begin this week.

The challenge

For remainers, voting choices in the European Parliament elections is not simple. We are electing MEPs, potentially to represent us for the next five years, but because the result will be taken as a signal of the scale of support for stopping Brexit (in one form or another), there is a case for voting tactically for a remain party likely to win, rather than the party whose manifesto one supports.

The issue is urgent, because ballot packs are already going out to postal voters, who make up 17% of the electorate in the East. Many people will have voted before the manifestos are published,.

The tactical voting challenge

Tactical voting could increase the number of remain MEPs elected on 23rd May but doing so will not be simple.

Because the election has been called at such short notice, the remain parties had no time to agree an alliance or pact to concentrate votes. Ballot papers are no printed, so no party can withdraw, and in all regions remain voters will have to choose between LibDem, Green, Change, and perhaps Labour (if they finally declare themselves for a referendum).

By contrast, the Brexit Party appears to be concentrating all Brexit votes, and is certainly stealing votes from Conservatives and UKIP.

Tactical voting is also difficult because of the difficulty of communicating with sufficient remain voters, and the complexity of the d’Hondt voting system, which is difficult to explain to voters, and difficult to predict (see below).

EU elections v Westminster elections

Here is an outline of how European elections differ from Westminster ones.

UK Parliamentary elections EU Parliamentary elections
Vote for candidate Vote for a party
Small constituencies Large regions
1 MP per constituency 7 MEPs in the Eastern Region[2]
The candidate with the most votes wins, regardless of how many votes they get Candidates get seats according to the d’Hondt system (see below).
  Each party lists its candidates. When they win a seat the first on the list is appointed
The final result reflects the concentration of party support in particular constituencies, making most seats “safe”, and allowing a government to be elected by as little as 35% of the votes cast The final result is very roughly proportional to votes cast, but it tends to favour large parties

The d’Hondt System

The d’Hondt voting system[3] is designed to ensure that the number of MEPs elected roughly matches the number of votes cast.

MEPs are selected in a series of rounds, at the end of each another MEP is chosen, as follows:

After the election, all the parties are listed in descending order of number of votes cast.

In Round 1 the party with the largest number of votes gets the first seat. Their original vote number is then divided by the number of seats achieved (i.e. 1) plus 1 = 2, and the list is resorted.

In round 2, the party with the highest number of votes on the resorted list gets the second seat. Their original Brexit number is divided on the same basis and the list is resorted again.

This continues until all seats are filled. When a party gets a second seat, their original total is divided by 2+1 = 3 and so on.

What this means for the Eastern Region

In the Eastern Region, which has 7 seats, this would be the result, using the results of the YouGov poll of 23-26 April 2019 (shown in the Round 1 column). The winner in each round is highlighted.

Percentage of
votes
Round 1 Round 2 Round 3 Round 4 Round 5 Round 6 Round 7 % of
seats
Brexit 36 18 18 12 9 9 9 57
Labour 19 19 9.5 9.5 9.5 9.5 4.75 28
LibDem 10 10 10 10 10 5 5 14
UKIP 9 9 9 9 9 9 9 0
Conservative 8 8 8 8 8 8 8 0
Change UK 7 7 7 7 7 7 7 0
Green 6 6 6 6 6 6 6 0

The result is 3 Brexit, 2 Labour, 1 LibDem. The seventh seat goes to either Brexit or UKIP

It is obvious that the result is not directly proportional. The two parties with the highest original vote get a disproportionate share of the seats.

It is also clear that with this distribution of votes, the first five seats are relatively “safe”. If voters do what they told pollsters they would, Brexit is almost certain to win 3 seats and Labour and LibDems one each. 

After that, however, relatively small shifts in voting can make a difference. The table below shows how large an increase in votes would be needed for each party to secure one extra seat.

Initial polling % Change required (points) % change in vote share
Brexit 36 +2 6
UKIP 9 +1 11
Conservative 8 +2 25
Change UK 7 +2 28
Green 6 +3 50
Labour 19 +11 58
LibDem 10 +9 90

The parties least likely to increase their number of seats are LibDems, Labour, and Green.  Those most likely to increase their number are Brexit and UKIP, especially if they vote tactically.

On these figures, the most likely strategy to secure an additional remain seat would be for more remain voters to vote for ChangeUK, who might then win the seventh seat, displacing UKIP.

Modelling tactical voting

The following table shows the unpredictability of tactical voting. It tests what would happen if all those who plan to vote for a particular party switch to another for tactical reasons. None of these is, of course, realistic, but they show how tactical voting has unpredictable results.

On current polling Greens vote LibDem ChUK vote LibDem All Green vote Change ChUK and Green vote LibDem Change, Green, & LibDem, vote Lab
Brexit 4 4 4 3 3 3
Lab 2 1 2 2 1 4
LibDem 1 1 1 1 3  
UKIP   1        
Con            
ChUK       1    
Green            
Remain (inc Labour[4]) 3 2 3 4 4 4
Leave 4 5 4 3 3 3

On these polls (and a lot could change in three weeks).

LibDems are very likely to win one seat, but would need to double their vote share to get a second

ChangeUK and Greens each have an outside chance of winning one seat, if they can double their vote share. But since they are likely to take votes from each other, they may both fail.

Labour has one safe seat. The second is vulnerable to Labour remainers switching to a more pro-remain party.

Polling evidence

Individual opinion polls can be unreliable, so the results from several polls provide a better guide. The remainvoter website (https://www.remainvoter.com/ ) analyses the results of all polling and their likely implications for the election by region[5]. The table below shows the results to date for the Eastern Region.

They confirm that the Brexit vote is consolidating around Brexit, but that the remain vote is split between three parties.

  UKIP Brex Con Lab ChUK LibDem Green Remain total
2014 EU Electn . 3   3 1       0
Hanbury 8 Apl 1 1 2 1       0
YouGov 11 Apl 2 1 1 2       0
Opinium 12 Apl 1 1 2 2 1     1
ComRes 16 Apl   2 2 1 1 1   2
ORG  17 Apl   1 3 2   1   1
YouGov 17 Apl   2 1 2 1   1 2
YouGov 27 Apl   4   2   1   1

Stephen McNair – Norfolk for Europe 6 May 2019



[1]  YouGov 27 April

[2] The number of MEPs for each region depends on the population of the region – from 3 in the North East to 10 in the South East

[3] Named after the Dutch mathematician who designed it

[4] Assumes that the Labour manifesto supports a referendum.

[5] The website does not indicate where their regional figures come from, since most pollsters do not report at regional level, and their number are often too small to be reliable. However, applying national figures to a region is an unreliable guide, given different voting patterns.