How New Zealand Can Fix Its Productivity Problem

New Zealanders work some of the longest hours in the developed world and have less to show for it each year. The problem is not effort. It is productivity, the amount the economy produces for every hour worked, and it has almost stopped growing.

Newswire analysed the official Stats NZ industry productivity series and the slowdown is stark. Labour productivity in the measured sector grew an average of 1.75 percent a year between 1996 and 2008. Since 2008 it has averaged 0.70 percent. Over the five years to March 2024 it managed just 0.20 percent a year, and multifactor productivity, the part of growth that comes from working smarter rather than simply adding more machines and people, went backwards.

With the general election set for 7 November, the campaign is being fought over tax, health and the cost of living. Underneath all three sits the same stalled engine. This article looks at what actually moves the needle, starting with the last time New Zealand tried to fix its economy with a big build, and working through the five levers that matter most now.

The productivity slowdown in numbers

Productivity sounds abstract but it is the whole game. It is the only durable way a country raises wages without raising prices, and it is why Australians doing the same jobs as New Zealanders take home considerably more. When it stalls, living standards stall with it, which is exactly what the past few years have delivered. New Zealand recorded the worst real wage fall in the OECD as living costs outpaced pay.

Period	Labour productivity growth (average per year)
1996 to 2008	1.75%
2008 to 2024	0.70%
2019 to 2024	0.20%

Multifactor productivity tells an even harder story. It measures how much output grows beyond what extra workers and extra capital alone would explain, and since 2019 it has fallen an average of 0.39 percent a year. The economy has been adding inputs and getting less out of them, a pattern that persists even as most New Zealand businesses adopt AI tools.

Two caveats belong up front. The measured sector excludes most government services, so these figures describe the market economy rather than everything, and the Covid years make any short window noisy.

But the direction is not in dispute, and it is the same picture the Productivity Commission painted for over a decade before it was disestablished in early 2024. The diagnosis has never been the problem. The politics of acting on it has.

Lessons from the Think Big projects

New Zealand has tried a state-led fix before. The Think Big programme of the early 1980s is remembered as a byword for failure, but the record is more useful than the caricature, a case Newswire made last year in Build things again. The projects split cleanly into two groups, and the line between them is instructive.

The Clyde Dam was a mess during construction, with cost blowouts, a geological scare that forced expensive slope stabilisation, and the bitterness of flooding old Cromwell to form Lake Dunstan. Yet as a standing asset it has quietly paid for itself many times over, a large block of cheap, low-carbon hydro power that can be turned on whenever the country needs it, worth far more in an electricity-hungry 2026 than anyone imagined in 1982. New Zealand Steel at Glenbrook is the other survivor, built around a technology designed for the one resource New Zealand genuinely has in abundance, North Island ironsand that conventional blast furnaces cannot easily use. It is still the country’s only steelmaker.

The failures were different in kind, not just in luck. The Motunui synthetic petrol plant and the Marsden Point refinery expansion were bets that the oil price would stay high. When oil collapsed in the mid 1980s the rationale evaporated, and Marsden Point eventually stopped refining altogether in 2022.

This is a reading of a handful of projects, not an economic law. But the pattern it suggests is consistent with what worked elsewhere. Public capital tends to pay off when it builds long-lived platforms the rest of the economy runs on, or backs a genuine local advantage.

It tends to fail when the government is effectively punting on a commodity price it cannot control. That rule of thumb is worth keeping in hand for everything that follows.

Housing, tax and the nation’s savings

If New Zealand’s productivity problem has a master key, it is housing. Expensive houses do not just hurt the people trying to buy them. They pull the nation’s savings into land instead of businesses, push workers away from the productive cities, and turn the whole economy into a debt-fuelled bet on property. Newswire’s house prices and income tool shows how far the cost of a home has moved beyond what households earn in every major centre.

Tax is on the ballot in November. Labour proposes a 28 percent capital gains tax on profits from selling commercial and residential investment property from 1 July 2027, excluding the family home, farms, KiwiSaver, shares, businesses and inheritances, with the revenue tagged to three free GP visits a year. National has ruled out a capital gains tax and argues growth should come from cutting regulation and lifting investment instead. Voters will settle that argument, but the economic case for taxing property gains is less about house prices, which overseas evidence suggests a CGT changes only modestly, and more about no longer steering the country’s savings toward land by taxing it more lightly than everything else.

The larger lever is supply. Research by Ryan Greenaway-McGrevy and Peter So at the University of Auckland’s Economic Policy Centre found Auckland rents were around 28 percent lower six years after the city rewrote its planning rules to allow more homes on each section than they would have been if rents had kept following their old track, one of the clearest real-world tests anywhere that letting people build brings costs down. Keeping land-use liberalisation moving, and fixing infrastructure funding so councils gain from growth rather than carrying its costs, would do more for productivity than any ownership restriction. Rules that cap how many houses someone can own or ban companies from owning them mostly reshuffle a scarce stock, and would choke off build-to-rent, one of the few models that adds homes at scale.

Electricity investment

Here the good half of the Think Big lesson applies directly. New Zealand has a genuine natural advantage in renewable generation, with wind, geothermal and hydro storage potential at a moment when cheap clean electricity is becoming a magnet for industry and data centres. Generation, grid and storage are exactly the kind of long-lived platform assets that a Clyde Dam turned out to be, not a punt on a commodity price. Newswire’s NZ Grid Live map shows how the current system runs in real time.

The obstacle is not engineering, it is the pipeline. Projects are announced, cancelled and re-announced with each change of government, and infrastructure bodies have long argued that the stop-start cycle itself raises costs by destroying the construction sector’s capacity to plan, invest and keep skilled crews together. A multi-decade build-out, financed through vehicles that share risk with private capital and durable enough to survive elections, would attack the productivity problem from the supply side while decarbonising the economy at the same time.

The new limits on carbon forestry

One fix is already law. Whole farms were being sold into pine for carbon credits, converting productive land that employed people into permanent forests that employ almost nobody, driven by the profit to be made from carbon credits rather than anything the land produces. The legislation restricting farm-to-forest conversions took effect on 31 October 2025.

It bans registering exotic forest in the Emissions Trading Scheme on better farmland, classes 1 to 5 on the scale used to rank land quality, and caps conversions on medium-quality class 6 land at 15,000 hectares a year, allocated by two ballots of 7,500 hectares. The first ballot opened on 19 June and closes on 22 July 2026, so the new system is being tested right now.

The nuance worth keeping is that trees are not the enemy. Steep, erosion-prone class 7 and 8 country is often better in forest than in marginal grazing, and conversions there remain unrestricted. The policy problem was pine swallowing good land, and on that specific failure the door has now been closed.

KiwiSaver and the savings gap

New Zealand workers have less machinery, software and infrastructure behind them than workers in most rich countries. Economists have long pointed to a shallow pool of domestic capital as part of the reason. KiwiSaver is the main tool for deepening it, and it is moving, slowly. From 1 April 2026 the default contribution rose to 3.5 percent from employees matched by 3.5 percent from employers, on its way to 4 percent plus 4 percent in April 2028, while the government contribution was halved to 25 cents per dollar and removed entirely for incomes over $180,000, according to Inland Revenue.

The scale gap remains the story. Even at 8 percent combined, New Zealand will sit well short of Australia, where employers must pay 12 percent into superannuation, and where three decades of steady increases built a pool so large it now finances infrastructure, property development and growing companies as a matter of course. The thin layer of capital behind each New Zealand worker has several causes, including distance and a small market, so deeper savings alone will not close the gap.

But it is the least glamorous and most reliable lever on the list, provided the climb is timed against wage growth, since employers claw back part of their contributions through lower pay rises. There is also an awkward local wrinkle in that a large share of KiwiSaver withdrawals go into first-home deposits, sending the retirement savings pool into the very asset class the country is already over-invested in. Our guide to growing your KiwiSaver faster covers what the settings mean for individual savers.

Immigration and the export base

The hardest argument to referee is immigration. Former Reserve Bank economist Michael Reddell has long argued that New Zealand ran some of the highest population growth in the OECD into an economy whose exports come substantially from a fixed amount of land. More people do not create more hectares, so export earnings per person get diluted, and every new arrival needs a house, roads and schools, absorbing capital that might otherwise have deepened what stands behind each worker. It is a serious thesis and it deserves the attention it gets.

Two things push back against the full conclusion. The land is fixed but the value extracted from it is not, as decades of dairy conversion, horticulture and wine have shown, with kiwifruit and vineyards earning far more per hectare than the sheep they replaced. And the real escape from the trap is decoupling exports from land altogether. Technology is now the country’s third-largest export sector, with export earnings passing $15 billion in 2025, and it is also the sector most dependent on skilled migration.

A blunt cut to immigration protects the arithmetic of the old economy while starving the new one. The defensible middle course is to moderate the total intake to what housing and infrastructure can absorb while staying aggressively open to high-skill migrants, through wage thresholds and sector targeting rather than a single national number.

The three-year political cycle

Almost everything above takes ten to thirty years to pay off, and New Zealand governs in three-year slices. Infrastructure plans, tax settings, water reform and planning rules are routinely reversed before they bite, and the reversal itself is widely argued to carry a productivity cost, in cancelled projects, wasted design work and a construction industry that cannot commit to capacity. Australia’s superannuation system compounded for decades largely because it became politically untouchable.

A longer parliamentary term was one proposed remedy, but the government scrapped plans for a referendum on a four-year term in February 2026, citing time constraints. That leaves the harder, less mechanical version of the fix, which is genuine cross-party agreement on a small number of long-horizon commitments, an infrastructure pipeline, savings settings and education among them, so that thirty-year investments can be made against a stable backdrop. It is the least concrete item on this list and probably the most important.

What a serious productivity fix would look like

Pulling the threads together, the package that follows from the evidence looks like this.

• Stop steering national savings into land, through tax settings that treat property gains like other income and through continued land-use liberalisation so homes get built
• Commit to a multi-decade electricity and infrastructure build-out that survives changes of government, the Clyde Dam lesson applied deliberately
• Hold the line on carbon forestry limits so good farmland keeps producing and employing
• Keep lifting KiwiSaver step by step toward Australian-scale contributions, timed against wage growth
• Target immigration by skills and by what housing and infrastructure can absorb, rather than a headline number
• Build cross-party durability for all of the above, because none of it works in three-year pieces

None of this is quick, and countries that have climbed out of comparable holes did it with sustained, boring consistency rather than a single dramatic reform. The encouraging reading of the numbers is that New Zealand’s problem is not mysterious or cultural. It is an accumulation of specific, fixable underinvestments, and the fixes are sitting in plain sight.

Productivity will not appear on the ballot paper on 7 November, but nearly everything on it flows from these choices. Newswire’s voting tool compares where the parties stand. What would you fix first? Tell us in the comments below.

Method note. Productivity figures are Newswire calculations of average annual growth from the Stats NZ measured-sector labour productivity and multifactor productivity index series (Productivity statistics 1978 to 2025), years ended March. The measured sector excludes most government services, and Covid-era years make short-period averages indicative rather than precise.

This article was written by AI, briefed to report the facts, hopefully without some of the bias people bring to the job 🙂