# In Mixed-Member Proportional representation system what percentage of people vote for different party than MP's party?

In Mixed-Member Proportional representation system electors have two votes first for an MP and second for party. My question is there some figure of how many people vote for party in second vote different from the MP's party in the first?

The reason I ask this is that if most people vote for same party in their second vote as the MP's party in the first vote then why have two votes? Why not have one vote for an MP like any first past the post system then from that calculate how many votes each party got?

• That’s two questions. You should edit to limit it to one question at a time. For the first question, there definitely are statistics from Germany at least. Apr 3, 2018 at 19:32
• @chirlu, I don't need answer to second question. I just added that so that people know why I am asking.
– MrDi
Apr 3, 2018 at 19:34
• Can you link to the statistics from Germany please?
– MrDi
Apr 3, 2018 at 19:35
• For the record, there's an issue with the logic in your second paragraph, that shows up more clearly in the related question "if most people vote for party X, why have party Y?". Apr 3, 2018 at 19:58

## 2 Answers

New Zealand has used an MMP system since 1996.

In the 2017 general election, 27.33% of voters split their vote (that is, used their candidate vote for a candidate of a party different from that for which they used their party vote).

In general, splitting votes is much more common for supporters of minor parties. Of those who voted for National or Labour (the two major parties), only 14.13% and 22.94% of voters respectively split their vote. In comparison, of those who voted for the Greens, NZ First, or ACT (the three minor parties who hold seats), 66.62%, 59.06%, and 81.27% of voters split their vote.

This makes sense, as typically general electorates are won by either National or Labour (unless a major party endorses a smaller party for a particular electorate). A supporter of a minor party will typically vote for either the National or Labour candidate with their candidate vote, and vote for their preferred minor party with their party vote.

These figures is broadly in line with previously elections, with 31.64% split in 2014 and 30.70% split in 2011.

Also based on NZ experience. Note that in NZ MMP (note UK AMS systems are less proportional than this) the party vote decides the number of seats for each party in parliament. The electorate vote helps to choose which individuals get to occupy those seats, but only afffects seat numbers if there is an overhang (a party wins more electorates than its share of party votes justifies).
Vote splitting: Often vote splitting occurs where people wish to suppport a well-known local personality but not their party. It does not change much.

There is a potential scam with MMP which might be hard to implement, but which I think needs attention. Well known senior members of a party resign and stand as independents, and their original party does not stand a candidate against them. They may then win electorate seats which help their old party over and above their entitlment arising from party votes. I think it just needs some 'nod,nod, wink, wink' to get their party to go along with it. Something like this has happened in NZ, with the Epsom 'cup of tea'deal.

Why two votes? The scam above could be eliminated by having one vote per voter, being both for a candidate and for their party, as you suggest.

If there is a threshold, then it is useful to allow voters a second choice, so as to avoid perverse incentives for voters. That is another issue.

• "Well known senior members of a party resign and stand as independents, and their original party does not stand a candidate against them. They may then win electorate seats which help their old party over and above their entitlment arising from party votes." This doesn't appear to be very like the the Tea Tape scandal, which was press behaviour. The endorsement of a candidate from a existing different party, involving no resignations seems to have been open and expected rather than a scandal. The "Alba" party in Scotland's most recent elections is probably closer, but still not quite a match. Oct 7, 2021 at 9:09
• At least based on the wikipedia article; en.wikipedia.org/wiki/Tea_tape_scandal Oct 7, 2021 at 9:10