The Infrastructure New Zealand Didn’t Build

How Billions in Government Spending Went to Restructuring, Consultants, and Programmes That No Longer Exist, While the Roads, Pipes, Turbines, and Grid the Country Actually Needed Went Unbuilt

New Zealand has a $210 billion infrastructure deficit. That figure, calculated by the Treasury in its 2022 Investment Statement, represents the combined gap between what the country’s roads, pipes, power lines, water treatment plants, and public buildings need and what they have actually received in investment. The New Zealand Infrastructure Commission, Te Waihanga, estimates that closing this gap would require roughly $31 billion per year in capital investment, nearly a tenth of national GDP, sustained over three decades. The country currently spends about half that.

This deficit did not appear overnight. It is the accumulated consequence of decades of underinvestment by governments of every stripe, compounded by population growth, deferred maintenance, and a political culture that has consistently preferred operational spending over capital investment. But the deficit has widened noticeably in recent years, during a period when government spending grew at an extraordinary pace. Core Crown expenses rose from $76.3 billion in the 2016/17 financial year to $127.6 billion by 2022/23, an increase of $51.3 billion per year in nominal terms, or approximately 67 percent. Total Crown expenses reached $183.5 billion by 2024/25. Yet in April 2026, New Zealand finds itself with an electricity system that nearly ran out of capacity during the winter of 2024, fuel storage that provides only weeks of buffer against global supply disruption, water infrastructure that is literally leaking into the ground, and a national grid that cannot deliver enough power to the places that need it most.

Where did the money go? And what could have been built instead?

This is not a partisan question, though it will inevitably be read as one. Both major parties have made choices that prioritised spending on things other than physical infrastructure. The Labour government of 2017 to 2023 expanded the public service, launched restructuring programmes that were subsequently reversed, and spent heavily on consultants, engagement, and policy development that produced few lasting physical assets. The National led government that replaced it has cut public services, reduced Crown revenue by approximately $1.6 billion per year through tax cuts, cancelled infrastructure projects of its own, and is now scrambling to build the very things its predecessor deferred. The result is a country that has spent more than ever but built less than it needed.

The Spending Surge

The numbers are stark. When the Labour led government took office in late 2017, core Crown expenses were running at $76.3 billion per year, or 28.5 percent of GDP. By the time it left office in late 2023, that figure had reached $127.6 billion, or 32.2 percent of GDP. Some of this growth was unavoidable. The COVID 19 pandemic required enormous emergency spending on wage subsidies, the managed isolation and quarantine system, the vaccine rollout, and economic support packages. Superannuation payments grow automatically as the population ages. Welfare benefits are indexed to wages and inflation. Health costs rise with an ageing and growing population. Interest payments on government debt increased sharply as interest rates rose and pandemic era borrowing accumulated.

But a substantial portion of the growth went into expanding the state’s operational footprint. The core public service workforce grew from 47,252 full time equivalent positions in June 2017 to 63,117 by June 2023, an increase of 15,865 positions or 34 percent. At the average public service salary of $97,200 reported by the Public Service Commissioner, this represented an additional payroll cost of approximately $1.5 billion per year. The growth was concentrated not in frontline roles like teachers, nurses, or police (who are counted separately) but in the administrative and policy bureaucracy. The number of public service managers increased 51 percent, from 5,333 to 8,059. Policy analysts grew 50 percent, from 2,633 to 3,949. Information professionals grew 73 percent, from 5,437 to 9,426. The Ministry of Business, Innovation and Employment alone grew 87 percent, from 3,366 to 6,282 staff.

Spending on external contractors and consultants reached $940 million across public service organisations in 2023/24, on top of the expanded in house workforce. When Crown entities, the Defence Force, and Police are included, total contractor and consultant expenditure exceeded $2.4 billion. The public service’s total base salary bill reached $6.1 billion in 2023, up 11.8 percent in a single year.

Built to Be Demolished

The most visible spending with no lasting physical outcome went into institutional restructuring programmes that were subsequently reversed by the incoming government. A 2024 analysis by 1 News tallied the costs of scrapped projects and found that more than $2 billion had been invested in initiatives that the new government either cancelled outright or fundamentally reshaped.

Three Waters

The Three Waters reform programme, launched in response to genuine infrastructure failures including the 2016 Havelock North water contamination that killed people and made thousands ill, consumed over $1.2 billion before being repealed. The programme aimed to transfer management of stormwater, drinking water, and wastewater from 67 local councils to four new regional water entities. Whatever the merits of the policy, the money spent on establishing those entities, hiring staff, commissioning consultants, and developing transition plans produced no new pipes, no new treatment plants, and no new reservoirs. The two chief executives of the water entities received $710,000 between them in redundancy payments after being in their roles for less than a year. An estimated $165 million went to consultants and contractors for Three Waters alone. The underlying infrastructure deficit that prompted the reform, estimated at $120 billion to $185 billion over 30 years, remains unaddressed.

Te Pūkenga

The merger of New Zealand’s 16 polytechnics and institutes of technology into a single mega entity, Te Pūkenga, cost at least $121 million in establishment and transition costs before the current government announced it would disestablish the entity and return to a model of more autonomous regional institutions. The national office grew to 204 permanent and fixed term staff. Software licensing costs at the national office alone went from $104,000 in 2020 to $4.4 million in 2023. A $220 million Crown loan for digital transformation saw none of the money actually spent before the programme was reversed. The polytechnics that existed before the merger still exist. They will continue to exist after the merger is unwound. The net physical infrastructure gained from the entire exercise is essentially nil.

Auckland Light Rail

The Auckland light rail project consumed $229 million in planning, advisory, and investigation costs across both governments without a metre of track being laid. The project was studied, debated, redesigned, referred to a new agency, studied again, and ultimately cancelled. For context, $229 million is roughly the cost of a mid sized wind farm that would generate clean electricity for decades.

Other Reversed Programmes

The reform of the Resource Management Act consumed $136 million in development costs for legislation that was then replaced by the incoming government with its own approach. The Māori Health Authority, established as part of the health system restructuring, cost $44 million before being disestablished. The Let’s Get Wellington Moving transport programme absorbed $167 million. The Lake Onslow pumped hydro investigation consumed $36 million in feasibility studies without reaching a construction decision. Each of these programmes may have had defensible rationales. But collectively, they represent billions of dollars that produced organisational charts, consultant reports, transition plans, and legislation rather than physical assets.

What the Money Could Have Built

The question is not hypothetical. The costs of physical infrastructure are well documented, and the gap between what was spent on restructuring and what could have been built is measurable.

A large geothermal power station, such as Contact Energy’s Tauhara plant that came online in 2024, costs in the range of $600 million to $900 million and adds approximately 150 to 250 megawatts of reliable, baseload renewable generation that operates around the clock regardless of weather. The $1.2 billion spent on Three Waters establishment costs alone could have funded one to two geothermal power stations, adding enough generation capacity to power roughly 300,000 to 600,000 homes.

A major wind farm costs $300 million to $500 million. The 176 megawatt Harapaki wind farm and the 222 megawatt Turitea wind farm each fell within this range. The $229 million spent on Auckland light rail planning could have built a wind farm generating enough electricity for approximately 70,000 homes, every year, for the next 30 years.

Grid scale solar has become the cheapest form of new electricity generation in New Zealand, with installed costs now under $2 million per megawatt. The $940 million spent on public service contractors and consultants in a single year (2023/24) could have funded approximately 470 megawatts of solar capacity, more than doubling the country’s entire installed solar base at the time.

The additional diesel storage that the Labour government planned but did not execute, and that the current government deferred before scrambling to fund during the 2026 Strait of Hormuz crisis, was budgeted at $84 million. That is less than the amount spent on redundancy payments across the public service in the year to June 2024.

The Cook Strait HVDC upgrade, which would unlock the ability to transmit more South Island renewable generation to North Island demand centres, is estimated in the billions but would deliver value for decades. The inter island ferry replacement (iReX), which was needed to maintain the freight link between the islands, was cancelled by the current government after hundreds of millions in sunk costs, leaving the country dependent on ageing vessels.

None of this is to suggest that the money spent on restructuring would automatically have been redirected to infrastructure. Government budgets do not work as simple transfers between categories. Operating expenditure and capital expenditure flow through different channels and are subject to different constraints. But the opportunity cost is real. Every dollar committed to a consultant writing a transition plan for an entity that would be disestablished within three years is a dollar that was not available for a turbine foundation, a transmission tower, or a water main.

The Think Big Precedent

New Zealand has built at scale before. The comparison that is both instructive and uncomfortable is Robert Muldoon’s Think Big programme of the late 1970s and early 1980s. Responding to the oil shocks of 1973 and 1979, the Muldoon government embarked on a series of major energy and industrial projects designed to reduce New Zealand’s dependence on imported oil and create industrial capacity using domestic energy resources.

The projects included the expansion of the Marsden Point oil refinery, the synthetic petrol plant at Motunui, the methanol plant at Waitara, the Clyde Dam on the Clutha River, the electrification of the North Island Main Trunk railway, the expansion of the New Zealand Steel mill at Glenbrook, the ammonia urea plant at Kapuni, and additional gas processing facilities in Taranaki. The programme was controversial at the time and remains so. Several projects, particularly the synfuels plant, were economic failures that cost the taxpayer dearly. The programme contributed to a fiscal crisis that required the painful economic reforms of the 1984 Labour government.

But the infrastructure that survived has delivered extraordinary value. The Clyde Dam still generates electricity today, more than 40 years after construction began. The Marsden Point refinery expansion operated for 36 years before the refinery closed in 2022. The Methanex methanol facility at Motunui continues to operate, producing methanol for export from Taranaki natural gas. The Kapuni ammonia urea plant, now operated by Ballance Agri Nutrients, still produces 40 percent of New Zealand’s domestic urea demand. The electrified railway still carries freight. New Zealand Steel still makes steel at Glenbrook.

The lesson is not that every Think Big project was wise. The lesson is that governments which are willing to invest in physical infrastructure, even at the risk of some projects failing, create assets that outlast the political arguments about them. A wind farm built today will generate electricity in 2056. A geothermal plant will run for 50 years or more. A water treatment plant will serve a community for generations. A consultant’s report has a shelf life measured in months.

The Opportunity in Front of Us

The irony of New Zealand’s current position is that the country has never had a better set of investment opportunities in front of it. The economics of renewable energy have shifted dramatically in favour of building. The demand case is strong and getting stronger. And the competitive advantages that New Zealand possesses, primarily its renewable electricity resource and its geographic position, are exactly the advantages that the global economy is increasingly willing to pay a premium for.

Electricity Generation

New Zealand’s electricity system generated 43,879 gigawatt hours in 2024, with 85.5 percent from renewable sources. MBIE projects that electricity demand will grow substantially as transport electrifies and industrial processes switch away from fossil fuels, with approximately half of all energy demand met by electricity by 2050. Meeting this growth will require thousands of megawatts of new wind and solar capacity. Transpower’s generation pipeline lists over 11,000 megawatts of solar projects alone. The resource is there. The question is how fast it can be built.

Government co investment in generation capacity, through mechanisms such as long term power purchase agreements, concessional financing for new builds, or direct equity participation in renewable projects, could accelerate the build rate and reduce the risk for private developers. The Think Big model, updated for modern conditions, would involve the government identifying the generation capacity needed over the next 20 years and committing capital to ensure it gets built, rather than waiting for the market to deliver at its own pace. The winter of 2024, when wholesale electricity prices exceeded $800 per megawatt hour and the aluminium smelter had to curtail production to keep the lights on, demonstrated what happens when the market does not build fast enough.

Grid Infrastructure

The national grid, operated by Transpower, is the circulatory system that connects generation to demand. The Cook Strait HVDC link, the critical connection between the South Island’s hydro generation and the North Island’s population centres, is approaching its capacity limits. Upgrades planned for completion around 2031 will expand this capacity, but in the meantime, the mismatch constrains how much South Island power can reach North Island consumers. Data centre developers, who could bring billions in investment to the South Island where renewable power is abundant and cooling costs are lowest, are being told they may need to wait for grid upgrades before proceeding.

Investment in grid infrastructure has a multiplier effect. Every dollar spent on transmission capacity unlocks multiple dollars of private investment in generation and industrial activity. The electrification of the rural power network in the early twentieth century, when local power boards issued loans totalling the equivalent of $1.7 billion in today’s dollars to build distribution networks, required voter approval through referendums. Every single referendum passed, with an average of over 85 percent support. Te Waihanga has noted that this would be the equivalent of current Auckland residents voting overwhelmingly for a $2.9 billion piece of infrastructure paid for solely by residents. New Zealanders, when presented with a clear case for infrastructure that would improve their lives, have historically been willing to pay for it.

Water

The water infrastructure deficit is perhaps the most urgent and least glamorous challenge. Estimates from the Water Industry Commission for Scotland, commissioned by the New Zealand government, put the cost of bringing the country’s drinking water, wastewater, and stormwater systems to an acceptable standard at $120 billion to $185 billion over 30 years. Pipes are leaking. Treatment plants are operating on expired consents. Over a third of wastewater plants will require re consenting within the next decade. People died in Havelock North because the water infrastructure failed. The Three Waters reform was an attempt to address this, but $1.2 billion was spent on the organisational restructuring and essentially nothing on the actual pipes. The current government’s replacement framework, Local Water Done Well, has not yet demonstrated that it will deliver faster or better results.

Transport

New Zealand’s transport infrastructure faces competing demands. The Roads of National Significance programme aims to expand highway capacity in key corridors. Rail infrastructure, particularly the inter island link and the North Island Main Trunk, requires ongoing investment. Auckland’s transport network remains chronically congested, with no mass rapid transit solution despite decades of debate and hundreds of millions spent on planning. The inter island ferry fleet is ageing and the replacement programme was cancelled, leaving a critical freight link in an uncertain state.

Industrial Capacity

The most transformative opportunity, and the one most reminiscent of the original Think Big vision, is using New Zealand’s renewable electricity advantage to attract and develop energy intensive industries that produce exportable products. The Tiwai Point aluminium smelter, for all the controversy it has generated, demonstrates the model. It uses renewable electricity to produce a high value physical product (aluminium) that is exported for approximately $1 billion per year. The smelter contributes $406 million to the Southland economy annually and employs over 3,000 people directly and indirectly.

Data centres represent the next generation of this model. Amazon Web Services has committed USD 7.5 billion to its New Zealand cloud region. Microsoft has launched a hyperscale facility powered entirely by renewable energy. These investments are attracted specifically by New Zealand’s clean electricity grid. But they are constrained by the grid’s capacity to deliver the power they need, particularly in Auckland where transmission into the city is tight. Building generation and grid capacity ahead of demand, rather than waiting for demand to arrive and then scrambling to supply it, would make New Zealand a more competitive destination for exactly the kind of high value, high employment industries the country needs.

Green hydrogen production, direct air carbon capture, advanced manufacturing powered by clean electricity, and sustainable aviation fuel production are all industries that could locate in New Zealand if the generation capacity existed. The proposed biorefinery at Marsden Point, which would use domestic biomass feedstock to produce sustainable aviation fuel and renewable diesel, is exactly the kind of project that a modern Think Big programme would support. Channel Infrastructure’s CEO has emphasised that unlike the old fossil fuel refinery, the biorefinery’s feedstock would be domestic, avoiding import dependence.

What Both Sides Got Wrong

The failure to build is genuinely bipartisan. The Labour government of 2017 to 2023 expanded operational spending dramatically, launched institutional restructuring programmes that consumed billions without producing physical assets, banned offshore oil and gas exploration without developing alternative energy sources at the pace needed, allowed the country’s fuel storage position to deteriorate, and deferred decisions on critical infrastructure while pouring resources into policy development, consultation, and engagement processes that produced reports rather than results.

The National led government that took office in late 2023 has its own list. It cancelled the Auckland light rail project after $229 million had been spent. It scrapped the inter island ferry replacement after hundreds of millions in sunk costs. It cut the Clean Car Discount scheme, causing electric vehicle uptake to collapse from over 50 percent annual growth to under 10 percent, slowing the very transition that would reduce fuel dependence. It implemented tax cuts that reduced Crown revenue by approximately $1.6 billion per year. It deferred the additional diesel stockholding that Labour had budgeted for, then scrambled to fund emergency storage when the Strait of Hormuz crisis erupted in February 2026. The cost of reversing the previous government’s reforms, disestablishing Te Pūkenga, restructuring the water entities, reshaping the health system, itself consumed hundreds of millions that could have gone to infrastructure.

The pattern is not unique to New Zealand, but it is particularly costly here because the country is small, geographically isolated, and dependent on infrastructure that must function without redundancy. When a larger country underinvests in its electricity grid, it can import power from a neighbour. When it runs low on fuel, it can truck supplies from an adjacent state. New Zealand has no neighbours. If the grid fails, there is no backup. If the fuel runs out, there are no trucks coming over the border. The margin for error is smaller, and the consequences of underinvestment are felt faster and harder.

The Country We Could Be Building

Te Waihanga has calculated that New Zealand’s annual capital investment in infrastructure will need to increase from just over $20 billion today to more than $40 billion by the 2050s to meet projected demand. The Commission’s analysis suggests that infrastructure investment will need to average around 6 percent of GDP annually over the next 30 years, within the bounds of what New Zealand has historically been willing to spend. The challenge is not affordability. It is priority.

The early twentieth century New Zealanders who voted in power board referendums to fund rural electrification understood something that recent governments appear to have forgotten. Infrastructure is not a cost to be minimised. It is an investment that generates returns for generations. The farmers who voted to borrow the equivalent of $1,600 per person (in today’s dollars) to electrify their districts were not being reckless. They were building the foundation for the agricultural productivity that still underpins the economy a century later. The government that built Manapouri to power an aluminium smelter was not engaging in corporate welfare. It was creating an industrial asset that has contributed billions to the economy over five decades.

New Zealand in 2026 has the renewable energy resources, the geographic advantages, the institutional capacity, and the capital market access to build at scale. What it has lacked, across successive governments, is the willingness to prioritise physical infrastructure over organisational restructuring, policy development, and operational expansion. The evidence is now overwhelming that this priority setting has been wrong. The electricity system is tight. The fuel system is vulnerable. The water system is failing. The grid cannot support the industries that want to locate here. The infrastructure deficit is growing, not shrinking.

The country that built the Waitaki hydro scheme, the Clyde Dam, the Manapouri power station, and the national highway network should not be struggling to build a wind farm or upgrade a water main. The resources exist. The technology is proven. The economics work. The only thing missing is the political will to stop spending billions on things that do not last and start building things that do.

The Think Big projects of the 1980s were not all successes. But the ones that produced physical infrastructure are still delivering value 40 years later. The restructuring programmes of the 2020s are already being dismantled. In another 40 years, which investments will New Zealand wish it had made?

Sources and References

• New Zealand Treasury, Financial Statements of the Government, years ended 30 June 2017, 2023, 2024, and 2025
• New Zealand Treasury, Fiscal Time Series Historical Indicators
• Te Kawa Mataaho Public Service Commission, Workforce Data, 2017 to 2025
• Te Waihanga New Zealand Infrastructure Commission, New Zealand’s Infrastructure Challenge
• Te Waihanga, National Infrastructure Plan and Pipeline Snapshot, December 2025
• Treasury, 2022 Investment Statement
• 1 News, Government’s 100 Day Plan Scrapped Projects and $2 Billion Invested, March 2024
• New Zealand Initiative, Public Service Cuts Needed to Help Rein in Government Spending, January 2024
• RNZ, How Many Public Sector Roles Are Going and From Where, May 2025
• NZ Herald, Public Service Headcount and Pay Increases, November 2023
• Water Services Reform Programme, Department of Internal Affairs
• IWA Publishing, The Three Waters Reform in New Zealand, December 2025
• Reseller News, Undoing Te Pūkenga, October 2024
• MBIE, Energy in New Zealand 2025
• Infrastructure New Zealand, Budget 2022 and Addressing the National Infrastructure Deficit
• Rio Tinto, NZAS Tiwai Point power agreements, May 2024
• Channel Infrastructure, Our History and Terminal and Pipeline Services
• Electricity Authority, Tiwai Point Smelter Demand Response, November 2024
• AWS, Microsoft, and Goodman Property Trust data centre investment announcements, 2024 and 2025
• Sense Partners, New Zealand’s Infrastructure Challenge, commissioned by Te Waihanga

What do you think New Zealand should be building? Share your thoughts in the comments below.