Ben Davies Blog Research Vita

Later this year, New Zealanders will vote in a referendum on whether to legalise voluntary euthanasia under the conditions specified in the End of Life Choice Bill (hereafter “the Bill”). Members of Parliament (MPs) read (i.e., debated) the Bill three times, each time holding a conscience vote on whether to progress the Bill towards becoming legislation. The table below presents the percentage and fraction of MPs who voted in favour of the Bill, separated by political party and reading.1

Party First reading Second reading Third reading
Act 100% (1/1) 100% (1/1) 100% (1/1)
Green 100% (8/8) 100% (8/8) 100% (8/8)
Independent 100% (1/1) 100% (1/1) 100% (1/1)
Labour 80% (37/46) 72% (33/46) 72% (33/46)
National 36% (20/55) 33% (18/55) 29% (16/55)
NZ First 100% (9/9) 100% (9/9) 100% (9/9)

Most MPs in the coalition government voted in favour, including all MPs from the Green Party and NZ First. In the Bill’s final reading, 72% of Labour MPs followed party leader Jacinda Ardern’s vote in favour, while 71% of National MPs followed party leader Simon Bridges’ vote to oppose. Overall, about a third of Labour and National MPs voted against their party lines.2

New Zealand uses a mixed member proportional electoral system: voters submit votes for a political party and for a representative of their local constituency. Consequently, some “list” MPs enter parliament because they are ranked highly within a party that received many votes rather than because they were the preferred candidate among their local constituents. The table below shows that Labour and National list MPs were more likely to vote along party lines than non-list MPs in the Bill’s third reading.

Party List MP adherence Non-list MP adherence
Labour 88% (15/17) 62% (18/29)
National 80% (12/15) 68% (27/40)

The difference in list and non-list MPs’ adherence to party lines has at least two explanations. First, non-list MPs have non-party reasons to be in government—namely, to serve their local constituents—and so may accept weaker idealogical matches than list MPs when self-selecting into party affiliations. This weaker matching would reduce the idealogical polarisation and inertia among non-list MPs relative to list MPs. Indeed, all of the MPs who changed their votes between the Bill’s first and third readings were non-list MPs.

Second, list MPs have stronger incentives to signal loyalty to their party because they cannot rely on support from local constituents to get elected. If list MPs consistently oppose their leaders then they may be demoted within their parties and, consequently, become less likely to re-enter parliament at the next election. Thus, to the extent that MPs want to maximise their chances of re-election, list MPs may be more willing than non-list MPs to ignore their conscience and vote along party lines.

It would be interesting to separate the idealogical sorting and signalling motives that drive greater adherence among list MPs. One strategy could be to track individual MPs across votes and governments, and analyse whether their propensity to vote along party lines is greater when they are list MPs than when they are non-list MPs. However, I can’t find any up-to-date vote data online and don’t particularly want to create them by trawling through decades worth of Hansard documents.3 Perhaps one of my readers is up for the challenge?


  1. The data used in this post are available here.
  2. This overlap in preferences among Labour and National MPs reflects the idealogical overlap between the two parties at the centre of the political spectrum.
  3. There was an online database of conscience votes among New Zealand MPs, but the database was shut down in late 2019 and hadn’t been updated since 2012.