The latest NZ political poll was the fortnightly Roy Morgan Research poll released on Tuesday, July 27. The poll shows some fairly large movements in support relative to the previous poll, but none of them appear to be statistically significant. National and Labour are both polling at roughly pre-budget levels within the margin of error.

As usual, the two graphs below summarise the polling averages for the party vote after the new poll. The horizontal axes represent the date, starting 60 days before the 2008 NZ General Election, and finishing on the present day. The solid lines with grey error bands show the moving averages of the party vote for each party, and circles show individual polls with the vertical lines representing the total errors.

The latest scenario analysis graph is also shown below:

Here we see that National have only a 78% probability of winning an outright majority in an election were it held today, with a further 19% chance of leading a National-ACT or National-ACT-United Future coalition for a total of 97%. The final 3% corresponds to the probability that the Maori Party will hold the balance of power in Parliament, albeit with a center-right numbers advantage. Part of the reason for this is the small drop in support for National in the latest poll, but it is largely due to the errors on the polling averages blowing up due the sparsity of recent polling data — a problem we’ve seen before, notably around the Xmas/New Year’s period. This time though, the implications are a bit different, as you can see from some of the additional simulation results shown on the Graphs page. In particular, the following stands out:

The above histogram shows the expected distribution of the number of seats won by New Zealand First in Parliament: most likely none, but with a small 9% probability of crossing the 5% threshold and receiving representation.

It is interesting to consider the implications of having the New Zealand First party back in Parliament, and we can determine them by simply running the election simulation as usual and then only counting up the results where New Zealand First gets seats. Under this condition the number of seats won are as follows:

The graph resembles the right tail of a normal distribution, with a little bit of rounding error due to eccentricities charachteristic of the Sainte-Lague method which is used to decide the seats, which is more or less the behaviour that would be expected. We see that even in these cases New Zealand First are unlikely to get more than seven seats, indicating that even if they were to surpass the 5% threshold it is unlikely at this stage to be by a large margin.

When we re-plot the results of the scenario analysis study for just the above election results (where New Zealand First get seats) we get the graph below:

This is where things start to get a bit more interesting. Labour still have virtually no chance of leading a centre-left coalition without the Maori Party. The reason for this is that New Zealand First’s seats have come roughly half from either side of the aisle, meaning that even with New Zealand First passing the 5% threshold there would still only be a 3~4 seat turnaround in favour of the centre-left. Nevertheless, the Maori Party now have a roughly 28% chance of holding the balance of power in Parliament. What are the chances of seeing a five-way coalition between Labour, the Greens, the Progressives, New Zealand First, and the Maori Party? And what are the chances the Maori Party would chose such a coalition over a centre-right coalition with National, ACT and United Future? Alternatively, it may be time for National to consider whether they would prefer to go into coalition with the Maori Party or New Zealand First if they were put into the position of having to chose one or the other.

Of course at this stage the results are largely hypothetical; all we’ve managed to show is that if an election were held today and New Zealand First were to pass the 5% threshold (two big ifs) then there is a non-negligible probability that the Maori Party would hold the balance of power. But if New Zealand First were to pick up 2% in the polls at the expense of the National party then we would be in for quite an unpredictable election.

These results are interesting given that the Roy Morgan Research polling was conducted before two important pieces of news recently came to light; firstly there were rumours of Whanganui mayor Michael Laws teaming up with Winston Peters to help get New Zealand First back into Parliament, and secondly in a recent interview Prime Minister John Key refused to rule out a coalition deal with New Zealand First like he did prior to the last election.

It will be interesting to watch New Zealand First’s polling numbers over the next twelve months.