By Joseph Darby*
Governments across the developed world are struggling to pay their bills. Populations are ageing, which means more healthcare and more pension spending with fewer working-age taxpayers, pandemic-era borrowing has left a larger debt pile and a larger interest bill, infrastructure everywhere needs renewing, and voters punish anyone proposing spending cuts or increasing taxes to cover the gap.
Alongside the problem sits an obvious pile of money. Decades of compulsory and semi-compulsory saving have built private retirement pools worth trillions, and finance ministries from London to Canberra have begun treating them as part of the answer.
New Zealand fits the description uncomfortably well. There is $142.3 billion in KiwiSaver on the Reserve Bank’s count, while taxpayers are paying close to $9 billion a year in finance costs on government debt, and Treasury expects the pension (NZ Super) bill to climb from 5.1 percent of GDP to around 8 percent by 2065. The same statement projects government debt reaching around 200 percent of GDP by then if policies remain unchanged.
Sébastien Betermier, executive director of the International Centre for Pension Management, describes what happens next in systems around the world.
“At a time when pension funds have become really big, and the governments themselves are cash-constrained, it’s becoming increasingly tempting for governments to go: maybe I can redirect it for other policy goals.”
Betermier calls the pattern pension fund nationalism and sees it across Canada, Britain, Europe and Australia.
Be clear about what coming for KiwiSaver would mean, because it is highly unlikely to be a raid on anyone’s account. The realistic form is a government using KiwiSaver settings to serve objectives beyond members’ retirement outcomes, most plausibly by requiring default funds to hold a set share of New Zealand infrastructure or housing assets.
KiwiSaver assets are scheme property held for members, and no minister can withdraw or borrow the pool; what a government can reach is the settings. Governments here have used the settings for other objectives repeatedly. A limited version of directing where the money goes was previously put out for consultation in Wellington and has been legislated as a reserve power in London, and in my opinion the conditions for it are strengthening.
A retirement scheme with a growing job description
The KiwiSaver Act 2006 states the scheme’s purpose: a long-term savings habit and asset accumulation, for individuals not positioned to enjoy retirement living standards resembling their working ones. Retirement saving remains the stated purpose. The scheme has also picked up other jobs across its first two decades, some written in at the start and widened since, others added by later governments.
The scheme is now leant on to ease the retirement income problem NZ Super cannot carry alone, even as Budget 2025 halved the government contribution to a maximum of $260.72, removed it entirely above $180,000 of income, and lifted default contribution rates. The Retirement Commission’s analysis of the package found continuously contributing wage earners should generally end up with more, while the self-employed and some lower earners come out behind.
A larger balance does not currently reduce anyone’s NZ Super entitlement, which is paid at a flat rate regardless of savings. What moves is the expectation. KiwiSaver is increasingly the vehicle expected to cover the distance between NZ Super and the retirement people want, with more of the load on employees and employers and less on the taxpayer. KiwiSaver incentives can be reduced by future governments while the restrictions on accessing the savings remain.
KiwiSaver also functions as a housing deposit scheme. First-home withdrawals have been available since 2010, and successive governments have widened the facility. Withdrawal is member-directed, so members choose to convert retirement savings into housing equity, which costs them diversification and liquidity while it can also lower their housing costs later in life. Inland Revenue’s count shows $1.9 billion withdrawn for first homes in the year to June 2025 by about 43,600 members, against $257.8 million a decade earlier.
Default product settings were pointed at climate change in 2021, when the terms appointing default providers excluded fossil fuel production, which the responsible minister said reflected the Government’s commitment to addressing a warming climate. Whatever its merits, and ministers argued stranded-asset risk alongside the climate case, it demonstrated the terms can carry public-policy objectives.
Some managers combine investment and social objectives voluntarily. One large KiwiSaver provider invests members’ money through a connected build-to-rent programme it expects to earn competitive returns, and describes the programme as helping to solve the housing crisis. Voluntary impact investment and a compulsory allocation are different things, though the example shows how readily retirement capital is drawn into arguments about national problems. And now some fund managers and commentators are arguing the pool should help fund infrastructure.
Then there is the direct approach. In 2022 a proposal to apply GST to fund management fees appeared inside an omnibus tax bill with official advice behind it, on FMA modelling worth $103 billion off KiwiSaver balances by 2070, around 4.7 percent of the projected 2070 pool, on the FMA’s own high-level assumptions. Officials framed it as standardising GST treatment; it also happened to raise revenue from retirement savings. It was reversed within 24 hours, by public reaction rather than any legal obstacle.
The rest of the world got there first
This is already happening overseas. The record abroad shows the full menu, from polite persuasion through to compulsory transfer, along with the occasional government moving in the opposite direction.
Australian superannuation, KiwiSaver’s closest institutional cousin, is an A$4.5 trillion pool its politicians openly court for housing and energy. Parliament legislated the counterweight in 2021, inserting the word financial into the trustees’ covenant so they must act in members’ best financial interests. Persuasion toward commercially justified domestic assets survives such a duty comfortably; sacrificing members’ returns for national objectives does not, which is the whole design.
The encouragement has become more explicit. In July 2026 the Prime Minister described the pool as a national asset capable of delivering returns for the nation as well as its members, and urged funds to lend more at home. Paul Schroder, who runs AustralianSuper, the largest fund in the country, called the prospect of government involvement in investment decision-making an “utter disaster”.
Britain escalated. The voluntary Mansion House Compact of 2023 aimed at 5 percent in unlisted equity; the Mansion House Accord of May 2025 doubled the ambition: at least 10 percent of main default funds in private markets by 2030, including at least 5 percent of the total in British private markets, expressly subject to fiduciary duty and investable supply.
In case the accord underdelivered, section 40 of the Pension Schemes Act 2026, which received Royal Assent on 29 April 2026, created regulation-making powers for an asset allocation requirement: heavily fenced, confined to default auto-enrolment funds, carrying a savers’ interest test, and needing secondary legislation before it operates. The unexercised power can be used only from January 2028, is capped at 10 percent of a main default fund with up to 5 percent of the total in British assets, and lapses in 2032 if it goes unused. Three years, from dinner-speech ambition to law.
Canada capped how much of a pension could leave the country, at 10 percent from 1971 rising to 30 percent, enforced through a penalty tax until abolition in the 2005 federal budget. Research at the time estimated full removal was worth in the order of C$1.5 billion to C$3 billion a year to members through better diversification, a number worth holding onto whenever a domestic quota is proposed anywhere.
Poland went furthest. Its second-pillar funds were mandatory but privately managed, holding individual member accounts with private firms. On 3 February 2014 the government transferred about 153 billion zloty, roughly US$50 billion and 51.5 percent of the funds’ assets, to the state social insurance institution, and cancelled the government bonds among them. Public debt fell by around 8 percentage points of GDP on contemporary measures. Members received credits in the public system, and marketable assets in their accounts were replaced by a claim on the state. Thankfully nothing proposed in New Zealand resembles the Polish transfer, which marks the far end of the intervention spectrum.
The bills keep arriving
Every force pushing those governments toward their pension pools is present here, and several are growing. Te Waihanga, the Infrastructure Commission, projects infrastructure investment rising from just over $20 billion a year to more than $40 billion by the 2050s, across central government, councils and private owners alike, and argues more of the network should be paid for by the people using it.
A government facing that list has a menu of unpopular options: raise taxes, borrow more, cut services, sell assets. Next to them sits a pool of locked-in savings growing by billions a year, with settings a minister can adjust. Directing private retirement capital can appear politically easier than raising taxes, cutting services or adding visibly to public debt.
The practical machinery already exists. Under section 132 of the KiwiSaver Act the Minister appoints default providers, after FMA advice, on the terms and conditions the Minister thinks fit, subject to the Act and ordinary public law limits. Those terms bind default products rather than every KiwiSaver Scheme, and they already carry investment exclusions imposed through the appointment process rather than through any new Act. Enabling private assets, encouraging domestic investment and compelling a domestic allocation are separate policies, and only the last is the concern here.
The 2019 default provider review consulted on requiring providers to put up to 0.5 percent of a default fund into alternative New Zealand assets, an option never adopted. Officials also excluded a broader domestic quota from the options altogether, on the grounds it addressed no identified problem, risked over-exposure to the New Zealand economy, and could compromise prudent investment. The paper also recorded the reputational cost: default members could feel their savings were being used to achieve additional government objectives. Those objections prevailed in 2019. The next default provider process will show whether they still prevail under heavier fiscal and political pressure.
Several protections sit in the way. Managers owe duties to act in participants’ best interests, with care, diligence and skill, which the FMA describes as overarching. Independent supervisors and FMA oversight sit above those duties, related party rules constrain connected transactions, members can move their money to another scheme, and ministerial decisions remain subject to ordinary public law limits. A statutory allocation requirement would not break those duties; it would redraw the mandate they operate inside. Nor is the industry a united line of defence. Where a manager or a related party earns fees from the private asset vehicles a domestic push would favour, its enthusiasm needs assessing for conflict as well as merit.
The FMA consulted in May 2026 on how the related party transaction provisions should apply, using a KiwiSaver Scheme investing into a connected wholesale property fund as its worked example. A domestic push could benefit members, managers, connected vehicles and policymakers in quite different ways.
Domestic investment can be commercially sensible without being politically directed. Certain mature infrastructure assets can offer long-duration or inflation-sensitive cash flows. Members may earn an illiquidity premium where entry pricing, governance, manager skill and fees justify surrendering liquidity. Housing, development and greenfield projects carry their own construction, demand, valuation and regulatory risks, and removing the regulatory barriers holding managers back from private assets is worth doing on its merits. The downside deserves equal attention.
Private infrastructure is not inherently safe: demand can disappoint, construction costs can run, regulation can change, leverage magnifies the result in both directions, and valuations stay subjective for long stretches. The test separating investment from capture starts with who chooses and whether the manager could decline, then asks whether the asset stays competitive after fees, risk, liquidity, valuation uncertainty, concentration and related party conflicts. Free selection against global alternatives is investment management. A compulsory allocation is public policy, and it should carry a heavier burden of proof, because the cost of getting it wrong comes out of somebody’s retirement.
A mandate’s drag can be sized by multiplying the share of the fund covered by the return it gives up, and at the allocation sizes anyone is likely to propose the answer is admittedly small. Concentration does not behave so politely. Owner-occupied dwellings and other directly held real estate already account for 48 percent of household assets across New Zealand, most members also depend on a New Zealand employer, some on their own small- or mid-sized New Zealand business.
For many, KiwiSaver is their principal genuinely global investment. Tilting it homeward asks people already long New Zealand to buy more of it. The shocks worth worrying about are the ones which arrive together. A potential foot and mouth outbreak, which analysis released by the Biosecurity Minister puts at roughly $14.3 billion a year in lost export revenue, would take the currency, the share market, employment, private infrastructure values, and house prices down in the same season. A major earthquake or a closed export market does the same. Global diversification exists precisely for when the local economy has a bad run, and a mandate quietly withdraws some of it.
No mandated allocation is currently proposed in New Zealand. In my opinion it is a live and growing policy risk all the same, and when it arrives it will be promoted as nation-building. One way will be through the current default KiwiSaver provider appointments, which run to 1 December 2028, and the tender before then is the clearest scheduled opportunity to attach a new investment condition. The cross-party accord the Retirement Commission called for in its 2025 review would put retirement settings on a longer political clock, and make an allocation introduced through a procurement round much harder to slip through.
The interesting question is what New Zealanders do about it, and the record offers both answers. The GST change died in roughly 24 hours because it arrived in a bill, with a number attached, and made the front pages. The 2021 investment exclusion arrived in a procurement document. Visibility will shape the reaction next time.
Any investment condition put up in the next default process should be published with the numbers attached: the assets covered, the expected return after fees, the liquidity and valuation risks, the added concentration in one small economy, and any benefit flowing to related parties. Domestic investment which wins on those numbers has earned members’ money. Anything needing a quiet procurement, or a public campaign of patriotism will be a huge red flag.
*Joseph Darby is a financial adviser and CEO of Become Wealth, a provider of financial advice and Discretionary Investment Management Services (DIMS). This article is the author’s opinion and does not necessarily reflect the views of Become Wealth. Nothing in this publication is, or should be taken as, an offer, invitation or recommendation to buy, sell or retain a regulated financial product.
2 Comments
Unprincipled Govts deciding that other people's money is theirs, no surprises there.
Using superannuation to underpin national development would make sense iff the projects selected made economic and financial sense, would have a reasonable rate of return, and the implementations were well managed. That would also require depoliticising how we develop as a nation and run things on data.
None of those conditions are typically met by government projects, and an option not listed in the article is for government to get more effective and efficient - which also seems impossible.
We welcome your comments below. If you are not already registered, please register to comment
Remember we welcome robust, respectful and insightful debate. We don't welcome abusive or defamatory comments and will de-register those repeatedly making such comments. Our current comment policy is here.