Pre-election polls understated right in NZ, with overhang skewing preliminary results

Preliminary results from Saturday’s New Zealand election indicate there are a total of 11 overhang seats, with parliament expanded from its nominal 120 seats to 131.

Overhangs occur as a result of electors having two votes – one for their local MP and one for their preferred party (which determines overall seat entitlements).

When a party wins more electorate seats than it would be entitled to based on its proportional party vote, more seats have to be added to parliament. An overhang was predicted before the election, but its size has surprised most observers.

This will be the largest overhang in a New Zealand election since the mixed member proportional system (MMP) was first first used is 1996.

National’s party vote of 29.2% was just behind Labour’s on 29.4%, which would normally entitle National to 35 seats.

But of the 71 electorate seats that are elected by first-past-the-post ballot, National is leading in 42, Labour in 21, Te Pāti Māori in five, the Greens in two and ACT in one.

Te Pāti Māori was entitled to one seat on its party vote of 0.6%, but is leading in five of the seven electorates for those on the Māori roll.

On the preliminary results, there’s a seven-seat overhang for National and a four-seat overhang for Te Pāti Māori.

On current counting, right-wing parties have 67 seats, one more than the 66 needed for a majority. Left-wing parties have 57 seats, and the centrist Opportunity Party has seven.

Poll performance

On the preliminary results, right-wing parties (National, ACT and NZ First) were 6.7 points ahead of left-wing parties (Labour, the Greens and Te Pāti Māori) on party list vote shares. This preliminary result has been marked on my poll graph below.

On current counting, the right has over-performed all the pre-election polls, with the Curia poll for the Taxpayers’ Union the most accurate.

I expect the right’s lead to drop after publication of final results, but the Reid Research poll – the only pollster that had the left leading since July – will be the least accurate of this election.

Opportunity – which could potentially hold the balance of power between left and right when final results are in – won 6.0% of the party list vote on the preliminary results.

That was well clear of the 5% threshold needed to enter parliament without winning an electorate. Polls had given Opportunity between 5.1% and 7.2%.

If there were no overhangs for either National or Te Pāti Māori, the right would still have won a majority on the preliminary results, as its 6.7-point lead over the left exceeded Opportunity’s 6.0%.

Special votes could swing final results

At the 2023 election, National won the party vote by 38.1% to 26.9% over Labour, an 11.2-point margin. This gave National 44 of the then 72 electorates at that election.

Despite trailing Labour by 0.2 points so far, National has only lost two electorates at this election.

I expect an aggregate of the electorate votes would be better for both National and Te Pāti Māori than their party list votes.

This is not the final count. No further updates will be released until final results are published on November 27. In the past, left-wing parties have gained one or two seats from right-wing parties from the preliminary to the final results.

National leads Labour in two electorates by fewer than 500 votes, in Hamilton East by 456 votes and in Wairarapa by 139 votes.

If National eventually loses both seats, the right would have a bare majority with 65 of 129 seats. Shifts to the left on the party vote could put the right below a majority.

Te Tai Tokerau is another electorate that could flip on later counting. Te Pāti Māori is leading the Greens by 436 votes. The Greens wouldn’t gain any overall seats if they won this electorate, as it would simply replace a party list seat.

That means it would be better for the left-wing parties overall for Te Pāti Māori to win in Te Tai Tokerau.