In the lead up to the 1984 election, Bob Jones’s New Zealand Party stormed onto the political scene, winning one vote in eight, but gaining no seats in Parliament.
As a result, 256,000 votes were wasted. Our First Past the Post voting system left a quarter of a million voters completely unrepresented.
Three years earlier, the situation was even worse. The Social Credit Party captured the public mood to secure the support of one voter in five. But on election night, the Party gained only two seats in a 92-seat Parliament – 353,000 votes were wasted.
Concerns about wasted votes were one of the major reasons New Zealanders chose the two-vote MMP system. We were assured that while our electorate vote might be wasted on a losing candidate, our party vote would always count towards the final makeup of Parliament – with two minor exceptions.
The first was if we voted for a party that won no electorate seats and failed to reach the five‑percent threshold to enter Parliament – then all of our party votes would be wasted.
And the second was when a party we supported won more electorate seats than its party vote entitled it to. In such cases all party votes would effectively be wasted, since it would be the electorate seats that determined the party’s total number of MPs. Those extra seats would be retained as an ‘overhang’, and Parliament would grow in size to accommodate them.
Over the years, the ‘overhang’ has become a common feature of our MMP system.
It first occurred in 2005, when the Maori Party won four electorate seats but only 2.12 percent of the party vote, entitling it to threeseats in our 120-seat Parliament. This created an “overhang” of one seat, and Parliament grew to 121 seats.
In 2008, the Maori Party created an overhang of two seats, in 2011 one seat, and in 2023, two seats.
In addition, while United Future leader Peter Dunne won his Ohariu electorate seat in 2014, his party vote of 0.2 percent was too small to entitle him to any seats in Parliament, so an overhang of one seat was created.
And in 2023, the sudden death of a Port Waikato candidate just before Election Day forced that election to be postponed. The by-election that was run later was won by National, increasing their electorate seats to 44 and boosting the overhang in the current Parliament from two seats to three.
In other words, for thirty years, relatively few party votes have been wasted: apart from those cast for parties failing to clear the threshold and some on overhang parties, the overwhelming majority of party votes have contributed to the final makeup of Parliament.
Until now.
As we face the 2026 election, a different set of dynamics is now operating.
A disillusionment with establishment parties appears to have afflicted both National and Labour. They have become stalled in the polls, seemingly unable to break through the 30 percent party vote threshold, as dissatisfied supporters turn to minor parties to provide the policy solutions they believe the country needs.
As a result, the combined party vote share for both National and Labour is experiencing an historic slump, dropping from 81 percent in 2017, to 75 percent in 2020, 65 percent in 2023, and now to an unprecedented low of around 55 percent in the latest 2026 polls.
This trend not only highlights how voters are increasingly abandoning the traditional two-party system in favour of minor coalition partners – ACT, New Zealand First, the Greens, and possibly TOP – but Electoral Commission data shows vote splitting is becoming more common.
At the 2023 election, almost 36 percent of voters gave their party vote to one party but their electorate vote to a candidate from another – an increase from 27 percent in 2020.
This pattern is particularly evident amongst the smaller parties. Of those who gave their party vote to ACT or New Zealand First, over 70 percent voted for an electorate candidate from another party – mostly National. For the Greens nearly 60 percent split their vote, and for the Maori Party, just over 40 percent.
As far as the main parties are concerned 22 percent of Labour voters split their vote and 16 percent of National.
What these figures tell us is that strategic voting is already a well-established feature of MMP elections – and since the potential governing blocs are so clearly defined at this election, with National, ACT and New Zealand First on the centre-right, and Labour, the Greens, the Maori Party, and TOP on the left, the number of voters using their two MMP votes differently is expected to grow.
That’s especially the case now that the Maori Party, which is attempting to use their reserved Maori seats to exploit the MMP overhang and change the Government, is campaigning on vote splitting. In fact, their party leaders are already claiming success: “This is the very first time the Maori roll will be determining who the next prime minister is.”
When it comes to electorate seats, 71 will be contested in 2026, one fewer than 2023. Of those electorates, National had maintained a relatively stable block of around 40 safe rural, provincial, and suburban seats over recent years, while Labour’s 20 to 30 stronghold seats were mostly urban.
However, the 2020 Covid election disrupted everything. Labour swept into power, not only winning an historic party vote majority, but producing a landslide of 46 electorate seats, including six of the seven Maori seats and seventeen National-held electorates.
The tide turned in 2023 as voters punished Labour’s mismanagement of the country by returning to National many of the electorates it had lost, along with some safe Labour strongholds such as New Lynn and Mt Roskill – with National only missing out on former Prime Minister Jacinda Ardern’s Mt Albert seat by 18 votes.
In the end, National won 43 electorate seats and 38.06 percent of the party vote. This entitled the Party to 48 seats in Parliament: 43 electorate MPs and five list MPs, with their ‘overhang’ seat taking their tally to 49 MPs – 44 electorate MPs and 5 list MPs.
While boundary reviews and population updates have altered voter demographics across many electorates, turning some “safe” seats marginal, it is still likely that National will win at least 37 electorates.
However, although National remains strong in the electorates, it has lost ground in its party vote. The effect is that the 2026 election is on track to deliver an unprecedented MMP result: National’s party vote is expected to produce no list MPs at all, since the electorates it is likely to secure will exceed its party vote entitlement.
This week’s NZCPR Guest Commentator, the Executive Director of The Initiative, Dr Oliver Hartwich explains the situation and what it means for MMP:
“On current polling, close to 900,000 people will give National their party vote, and none of those votes will yield an additional seat.
“National is on track to win about 37 electorates, one more than its share of the vote would give it. Its seats are already settled by the electorates alone. Whatever its supporters do with the party vote, National ends up with the same 37 seats.
“A National voter who understands this needs no instruction from anyone. Her party vote is dead where it is but alive anywhere else on the centre-right.
“The numbers are startling. If one National voter in nine reasons this way, the governing coalition wins a majority that current polling denies it. If one in four does, the coalition wins 69 seats in a Parliament of 132, even though the opposition parties, between them, won more votes.”
Let’s look into the detail to better understand why this election is so different.
The country’s two major newspapers Stuff and the Herald are each running a 2026 ‘poll of polls’, which combines all major political polls into a single, rolling average to theoretically provide a more accurate picture of voter support.
Stuff has now added an electorate flip calculator HERE, which indicates the voter changes needed to switch electorates between parties. It builds on the Electoral Commission’s interactive website resource HERE, which can be used to model potential election outcomes.
While both polls have National and Labour under 30 percent, their calculations are slightly different. We’ll examine the Herald’s predictions to see what they tell us.
The poll shows National on a party vote of 29.9 percent, which would entitle it to 36 seats in Parliament, Labour is on 27.6 percent with 33 seats, New Zealand First on 12 percent with 14 seats, the Greens on 11.6 percent with 14 seats, TOP on 8 percent with 10 seats, ACT on 7.9 percent with 10 seats, and the Maori Party on 2.1 percent with 3 seats.
If those numbers represented the election result, both the centre-right bloc and the left bloc would win 60 seats, and New Zealand would have a hung Parliament.
Let’s now look at how the result would change if National retained all 44 of the electorate seats it now holds. Since National’s party vote only entitles it to 36 MPs, the extra 8 electorate seats it won would create an overhang of eight in the next Parliament.
That means in a Parliament of 128 seats, winning parties would need to secure 65 seats between them. Using the above numbers, National, ACT and New Zealand First would gain 68 seats, while Labour, the Greens, TOP and the Maori Party would remain on 60 seats.
On these assumptions, the centre-right bloc would therefore hold a majority.
Under this scenario National would gain no list seats at all – and it is unlikely to do so unless its party vote climbs into the mid-30s before election day. As a result, it is highly likely all of National’s party votes will effectively be wasted.
As Dr Hartwich explained, this raises the possibility of strategic voting between the parties within a political bloc. If some National party supporters transferred their party vote to a Coalition partner instead of wasting it, National would retain its electorate seats, and the Coalition partner would gain additional representation, strengthening the centre-right.
The same principles apply to electorate voting. When parties support the same governing bloc, strategic voting could help Coalition partners win seats.
For instance, there are six electorates – Mt Albert, Nelson, Wigram, Taieri, Christchurch Central and Christchurch East – where the combined ACT–New Zealand First vote in 2023 exceeded Labour’s winning margin. If Coalition supporters voted for the National candidate, these seats could potentially be flipped from Labour to National.
There are six National seats with narrow majorities that could be retained by the centre-right if Coalition parties voted strategically: Banks Peninsula, West Coast–Tasman, Hutt South (Chris Bishop), Mt Roskill, Wairarapa and East Coast.
And the West Auckland boundary changes that have replaced New Lynn, Te Atatu and Kelston with three new electorates – Waitakere, Glendene and Henderson – could be competitive through centre-right strategic voting.
The point is that under MMP a voter is entitled to use their two votes in whatever lawful way they choose. They may double-tick the same party, or they may decide that their preferred outcome is better served by splitting their vote. The polls are telling us that the dynamics have changed at this election and 900,000 centre-right votes may be wasted. That’s something all centre-right voters should consider.
MMP was designed to ensure that votes translate into parliamentary representation. Yet under the circumstances we are facing today, the system appears likely to produce the opposite effect: substantial numbers of party votes that may produce no additional seats for the party receiving them. This is the strategic-voting dilemma facing voters in 2026.
With less than four weeks before overseas voting begins, the expression ‘the election is in the hands of voters’ has rarely been more meaningful. The way voters decide to use their two votes could materially affect the balance of power in the next Parliament.
Please note: To register for our free weekly newsletter please click HERE.
THIS WEEK’S POLL ASKS:
*At the 2026 election, are you considering splitting your electorate and party votes between different parties?
No comments:
Post a Comment
Thank you for joining the discussion. Breaking Views welcomes respectful contributions that enrich the debate. Please ensure your comments are not defamatory, derogatory or disruptive. We appreciate your cooperation.