The New Zealand Electoral Commission has released the split-voting statistics from the 2014 general election.
This is a great service provided by the Electoral Commission, showing in each electorate (district) what percentage of voters for each party list cast their vote for that party’s candidate or any other candidate in the electorate. To make it even better for those who like analyzing voting statistics, they offer CSV files.
The NZ Herald offers a summary of key electorates.
Fruits and Votes 
The NZ Herald article implies that Labour and Green voters ought to have voted National in Epsom, so the ACT leader didn’t win the constituency. What would be the advantage of that? ACT was unlikely to win top-up seats.
LikeLike
It would have meant 0 seats for ACT instead of 1, with no effect on the National total. Just goes to show how de-coupled the tiers are under MMP.
LikeLike
If, for a threshold-clearing party like National, gaining an additional nominal-tier seat does not affect its overall seat total, then doesn’t it show how coupled the two tiers are?
LikeLike
I thought that would be your reaction… I should probably have used a different term. Perhaps ‘different purpose’ describes itbetter. What I mean is that the MMP nominal tier (rarely) has an effect on the overall allocation of seats, whereas under MMM, both votes do affect it and thus are ‘coupled’ in their meaning to the voter (though quite the opposite in the actual calculation of seats). Essentially the point I was making before, in the thread about an AV-nominal tier under MMP.
LikeLike
At the 2014 General Election 31.64% of voters split their vote compared to 30.70% in 2011, 29.63% in 2008, 28.71% in 2005, and 39.04% in 2002. Has anyone analysed the reasons for these shifting numbers?
Although perhaps the most important question is, why is it always so much higher than in any other MMP jurisdiction?
LikeLike
Except for the 2002 percentage of split voting, that looks like a steady rate, with random fluctuation. The higher percentage in 2002 was probably due to United Future being the option for some usual National voters; it was the first year that the party entered a post-electoral agreement with Labour, which was very much dominant that year. The UF has not done well in the party-list vote since then.
I am not sure of the exact contribution, but a large chunk of the split-voting is Green and New Zealand First voters on the party list preferring the local Labour candidate.
As for why the rate of split-ticket voting is so much greater in New Zealand than in Germany, I’d point to two, probably related, reasons: (1) the legacy of district representation under the FPTP era has made local representation more important to New Zealanders; (2) unlike in Germany, local MPs get more resources for their constituency service duties. So voters care who represents them at that level.
By the way, split voting has been on the rise in Germany since the 1990s. This points to another possible explanation: more partisan choices mean more split voting, other things equal.
And, yes, I am working on some papers on these issues. Stay tuned.
LikeLike