A study reported in the mainstream media this week suggested that New Zealand First voters would have preferred that Winston Peters had gone with the National Party after the 2017 General Election. There has been much wailing and regret since the 2017 election, and the composition of the Sixth Labour Government is responsible for a great proportion of it. Numbers man Dan McGlashan, author of Understanding New Zealand, has the stats.
At the top level, the statistics do suggest that a slim majority of New Zealand First voters preferred National after the last General Election. The correlation between voting National in 2017 and voting New Zealand First in 2017 was 0.04, whereas the correlation between voting Labour in 2017 and voting New Zealand First in 2017 was -0.15.
These are very weak correlations – neither of them are considered statistically significant. The National one is positive and the Labour one negative, which does indeed tell us that the overlap between New Zealand First voters and National voters was larger in 2017 than the overlap between New Zealand First voters and Labour voters.
This does make Peters’s decision to go with Labour instead of National somewhat surprising. One explanation for it may be that Peters was judging his voters based on what they were in 2014.
In the 2014 election, the demographics of the New Zealand First voters were different. The correlation between voting New Zealand First in 2014 and voting National in 2014 was -0.34, and between voting New Zealand First in 2014 and voting Labour in 2014 it was 0.11. This correlation with National voters is statistically significant, which means the two groups are significantly different to each other.
So although it might be true that a majority of New Zealand First voters in 2017 would have preferred that Peters went with National, a majority of New Zealand First voters in 2014 would have preferred that Peters went with Labour, had he come to hold the balance of power then.
The reason for the change is the considerable number of Maori voters who switched from New Zealand First to Labour between 2014 and 2017. In 2014, the correlation between voting New Zealand First and being Maori was strong, at 0.66. New Zealand First lost the confidence of many of these voters during the next three years, and by 2017 the correlation between voting New Zealand First and being Maori had fallen to 0.38.
Because the correlation with being Maori and voting Labour is also strong (0.42 in 2014 and 0.58 in 2017), it can be seen that the shared Maori connection may have been enough to tilt New Zealand First’s loyalties towards the Labour Party.
A second point is that New Zealand First are nationalists, and concomitantly have a high proportion of people born in New Zealand among their voters. The correlation between being born in New Zealand and voting New Zealand First in 2014 was 0.69, and in 2017 0.54.
This high proportion of New Zealand-born voters makes New Zealand First very different to National. The low-tax, low-solidarity model of the National Party appeals strongly to those born overseas, and this is reflected in their voters.
The correlation between being born in New Zealand and voting National in 2017 was -0.41, which reveals the depth of globalist sentiments among National voters. The correlation between being born in New Zealand and voting Labour was 0.22 in 2017, on the border of statistical significance, but much closer to New Zealand First than to National.
New Zealand First, therefore, shares two very strong qualities with Labour that they do not share with National – a high proportion of Maori support and a high proportion of New Zealand-born support. These qualities may have been instrumental in making Peters’s decision.
So although it may be true that New Zealand First voters in 2017 would have preferred Bill English as Prime Minister, there are solid strategic reasons for Peters to have made the choice he did (whether he came to regret it afterwards must be the subject of a different analysis).
Understanding New Zealand, by Dan McGlashan and published by VJM Publishing, is the comprehensive guide to the demographics and voting patterns of the New Zealand people. It is available on TradeMe (for Kiwis) and on Amazon (for international readers).