New Zealand votes on 7 November. There’s been talk of this being “the KiwiSaver election”. This has prompted me to reflect on KiwiSaver and share some of my thoughts on the topic.

The general thrust behind media coverage about the proposed KiwiSaver policies is positive. With some exceptions, I feel the same.

When it comes to KiwiSaver, however, I always feel cautious. My big concern is that as we push towards KiwiSaver being framed as the big solution towards how retirement is funded in the future, there’s a risk that we’ll head towards a system that resembles those from other parts of the world, such as Australia and the US.

I don’t think that’s a good thing. Broadly speaking, I think that New Zealand’s system is great. It’s something that we can be proud of as Kiwis, and I think it’s something we should protect. In this article, I’ll explain why.

We need to be careful about opening the way for changes that move us towards other nations. This includes following the recommendations by the OECD (more on this later).

What’s on offer

For reference, here is what some of the parties are campaigning on:

  • National: compulsory for all workers from 2028, with default contributions rising to 6% each from employee and employer by 2032.

  • Labour: compulsory employer contributions, rising to 6% by 2032.

  • NZ First: increasing employee and employer contributions to 8% initially and eventually to 10%, with enrolment for citizens at birth, along with a $1,000 initial contribution.

  • TOP: a separate compulsory scheme alongside KiwiSaver which appears designed to eventually phase the current regime out. 6% each from employee and employer, with contributions exempt from income tax and fund earnings tax-free after twenty years.

  • ACT: end the government contribution (up to $260 a year) for members who receive employer contributions, and remove tax on KiwiSaver earnings instead.

I’m relatively supportive of National’s, Labour’s and NZ First’s policies as they stand. I’m concerned about the TOP and ACT policies.

Providing tax concessions that don’t currently exist is a form of spending through the tax code. The net effect is the same as providing someone a benefit. It’s just more opaque, easier to forget, and harder to change. I think the idea of making contributions or earnings tax-free will benefit higher income earners and tend towards benefiting people with more wealth. These are tax changes that are regressive. More on this later, when I explain why the OECD is wrong.

(To be clear: none of this is a final judgement on ACT or TOP as a whole. I’m talking about these specific policies. Both parties have other policies that I like, and in one of those cases the other policies make up for it. My rule of thumb is that if there’s a party whose policies you don’t disagree with somewhere, you haven’t looked hard enough.)

Many of the headline numbers in this debate are misleading

Is NZ Super as expensive as it’s painted?

Last year, Treasury projected that the cost of NZ Super on current settings would rise from 5.1% of GDP to around 8% by 2065. (At the same time, health would rise from 7.1% to around 10%.)

On the face of it, that’s kind of scary!

If the economy stayed exactly as it is today for 40 years, that’s what it would look like per person: everything else (all other spending in the economy, public and private) gets squeezed.

If the economy stood still
GDP per person, NZ$ thousands in 2025 dollars.
NZ SuperHealthEverything else
$0k$50k$100kToday (2025): Everything else $72kToday (2025): Health $6kToday (2025): NZ Super $4k$72k leftToday (2025)2065 shares, today's economy: Everything else $67k2065 shares, today's economy: Health $8k2065 shares, today's economy: NZ Super $7k$67k left2065 shares, today's economy
Show as table
NZ SuperHealthEverything elseTotal
Today (2025)$4.2k$5.8k$71.7k$81.7k
2065 shares, today's economy$6.5k$8.2k$67.0k$81.7k

The thing is, over time we create more goods and services with fewer resources. That’s called productivity, which drives economic growth.

And whatever the headlines might imply, the economy tends to grow because we keep being able to make more from less. The headlines are usually about the rate of economic growth, and not whether there is growth at all (except for during recessions).

Treasury projects that NZ Super and health expenditure will take up a bigger share of the economy by 2065. This gets a lot of attention.

What gets less attention is that Treasury projects that the economy will grow between now and 2065. Even after we factor in increased NZ Super and health costs, Treasury’s projections suggest that we will still be significantly better off.

It’s true that the costs take up a bigger slice of the pie. But it will be a bigger pie. The chart below is a better representation of these projections.

A bigger slice, of a bigger pie
GDP per person, NZ$ thousands in 2025 dollars. Treasury's baseline projection.
NZ SuperHealthEverything else
$0k$50k$100k$150kToday (2025): Everything else $72kToday (2025): Health $6kToday (2025): NZ Super $4k$72k leftToday (2025)2065 (Treasury baseline): Everything else $96k2065 (Treasury baseline): Health $12k2065 (Treasury baseline): NZ Super $9k$96k left2065 (Treasury baseline)
Show as table
NZ SuperHealthEverything elseTotal
Today (2025)$4.2k$5.8k$71.7k$81.7k
2065 (Treasury baseline)$9.3k$11.7k$95.8k$116.9k

Of course, this doesn’t make NZ Super free. But even with a higher dependency ratio (ie, a smaller working-age population relative to others), we can still afford it and are still likely to be better off. NZ Super’s current settings are not going to bankrupt the country, and even on the Treasury projections that are painted as scary, we will be significantly better off without any changes.

In the RNZ episode it was noted that an average KiwiSaver balance is about $40,000. It also places this against a figure of $1 million being a figure that a couple needs at retirement.

Both of those figures deserve great big asterisks.

For one thing, the average balance of all Kiwis isn’t a very useful figure.

It reminds me of statutory trustee companies promoting the fact that only 50% of Kiwis have a will, while omitting that around 90% of Kiwis who die have one. Most people who don’t have a will aren’t at the stage of life where it matters yet.

If we’re concerned about what Kiwis have at retirement, we should consider the group that is near retirement, and not include children and people at very different stages of their lives. For people aged 61 to 65 the mean balance is about $78,000.

A good portion of these people’s working lives were before KiwiSaver started, so their balances will be lower than what future balances would be, all else being equal.

Furthermore, this figure (whether you choose to use $40,000 or $78,000) refers to the average KiwiSaver balance of an individual when the $1 million figure refers to couples.

A couple (two people) with a balance of $78,000 comes to $156,000. Admittedly, it’s still well short of that $1 million figure. But at least we’re using more representative numbers.

And regarding the $1 million figure: it’s informed by the Massey Retirement Expenditure Guidelines. And for all of the good things you can say about this work, it has limitations. This figure is derived from what people who receive NZ Super spend, and to my knowledge it doesn’t separate out people who are still working. People who have only recently turned 65 and are still working are likely to be spending more than everyone else and be driving these figures up. Furthermore, I don’t believe these figures distinguish between households where people own their own home or have to pay rent.

I’m not saying that this is a case where there’s intentional deception. But the simple stats aren’t telling the whole truth.

It’s difficult to compare New Zealand’s system with other countries

I’ve written before about how New Zealand differs from Australia. Australia is more typical of how other countries approach retirement funding.

Superficially, Australia’s system seems better than New Zealand’s. On face value, it appears more equitable and cost-effective. But when you look deeper, the situation is entirely different.

Let’s start with the steel-man for Australia.

In Australia, the Age Pension costs about A$62 billion a year. As a share of GDP, that’s less than what New Zealand spends on NZ Superannuation.

The pension bill, as a share of each economy
Share of GDP. Australia 2025-26, New Zealand 2025. Approximate.
Australia: Age Pension (2025-26)Australia: Age Pension (2025-26): 2.3% of GDP2.3% of GDPNew Zealand: NZ Super (2025)New Zealand: NZ Super (2025): 5.1% of GDP5.1% of GDP

Furthermore, the Australian pension is means-tested, so less wealthy people get more from it than those who don’t need the support. Australia’s 2020 Retirement Income Review modelled how much Age Pension people get over a lifetime, from the lowest lifetime earners on the left to the highest on the right. It falls away as income rises.

The Age Pension on its own, by lifetime income (modelled)
$'000 per person, 2019-20 dollars, discounted. Modelled; read by eye off the Review's chart.
$0k$200k$400k$600k$800k10th percentile: Age Pension about $310k10th20th percentile: Age Pension about $305k20th30th percentile: Age Pension about $280k30th40th percentile: Age Pension about $255k40th50th percentile: Age Pension about $220k50th60th percentile: Age Pension about $180k60th70th percentile: Age Pension about $140k70th80th percentile: Age Pension about $85k80th90th percentile: Age Pension about $10k90th
Show as table
Lifetime income percentileAge Pension
10th~310
20th~305
30th~280
40th~255
50th~220
60th~180
70th~140
80th~85
90th~10

Compare that to New Zealand. Here, NZ Super is universal, so the richest person over 65 in the country gets the same pre-tax fortnightly payment as the poorest. At today’s rates, a single person living alone receives about $29,000 a year after tax and each member of a couple about $22,000 ($44,000 all up). A 65-year-old today can expect to live about another 20 years. Over that retirement NZ Super comes to roughly $450,000 to $580,000 of government support per person.

New Zealand on the same footing: NZ Super over retirement, by income
$'000 per person, today's dollars, undiscounted, couple rate, 20-year retirement. Everyone gets the same.
NZ Super, each member of a couple
$0k$200k$400k$600k$800k10th percentile: NZ Super about $444k10th20th percentile: NZ Super about $444k20th30th percentile: NZ Super about $444k30th40th percentile: NZ Super about $444k40th50th percentile: NZ Super about $444k50th60th percentile: NZ Super about $444k60th70th percentile: NZ Super about $444k70th80th percentile: NZ Super about $444k80th90th percentile: NZ Super about $444k90th
Show as table
Income percentileNZ Super
10th~444
20th~444
30th~444
40th~444
50th~444
60th~444
70th~444
80th~444
90th~444

Furthermore, NZ Super benefits probably skew towards wealthier people. Wealth is correlated with longevity. If the same annual amount is paid over a longer number of years, then NZ Super benefits wealthier people. The chart below is based on some back-of-the-envelope calculations correlating longevity with wealth.

New Zealand on the same footing: NZ Super over retirement, by income
$'000 per person, today's dollars, undiscounted, couple rate. The tilt is my assumption, not a published figure.
NZ Super, each member of a coupleAssumed extra from living longer (estimate)
$0k$200k$400k$600k$800k10th percentile: NZ Super about $388k at the shortest-lived decile's life expectancy10th20th percentile: NZ Super about $388k at the shortest-lived decile's life expectancy20th percentile: assumed extra $14k from living longer20th30th percentile: NZ Super about $388k at the shortest-lived decile's life expectancy30th percentile: assumed extra $28k from living longer30th40th percentile: NZ Super about $388k at the shortest-lived decile's life expectancy40th percentile: assumed extra $42k from living longer40th50th percentile: NZ Super about $388k at the shortest-lived decile's life expectancy50th percentile: assumed extra $56k from living longer50th60th percentile: NZ Super about $388k at the shortest-lived decile's life expectancy60th percentile: assumed extra $69k from living longer60th70th percentile: NZ Super about $388k at the shortest-lived decile's life expectancy70th percentile: assumed extra $83k from living longer70th80th percentile: NZ Super about $388k at the shortest-lived decile's life expectancy80th percentile: assumed extra $97k from living longer80th90th percentile: NZ Super about $388k at the shortest-lived decile's life expectancy90th percentile: assumed extra $111k from living longer90th
Show as table
Income percentileNZ Super, assumed
10th~388
20th~402
30th~416
40th~430
50th~444
60th~458
70th~472
80th~486
90th~500

On those assumptions, the top decile collects about $110,000 to $145,000 more NZ Super over a lifetime than the bottom decile. Put the two pensions side by side, and Australia wins not just on spending less but also on equity.

Pension only: Australia and New Zealand side by side, by income
$'000 per person over retirement. NZ: today's dollars, undiscounted, couple rate, tilt is my assumption. Australia: 2019-20 dollars, discounted, modelled. Similar ballpark, not strictly comparable.
Australia: Age PensionNew Zealand: NZ Super
$0k$200k$400k$600k$800k10th percentile, Australia: Age Pension: about $310k10th percentile, New Zealand: NZ Super: about $388k10th20th percentile, Australia: Age Pension: about $305k20th percentile, New Zealand: NZ Super: about $402k20th30th percentile, Australia: Age Pension: about $280k30th percentile, New Zealand: NZ Super: about $416k30th40th percentile, Australia: Age Pension: about $255k40th percentile, New Zealand: NZ Super: about $430k40th50th percentile, Australia: Age Pension: about $220k50th percentile, New Zealand: NZ Super: about $444k50th60th percentile, Australia: Age Pension: about $180k60th percentile, New Zealand: NZ Super: about $458k60th70th percentile, Australia: Age Pension: about $140k70th percentile, New Zealand: NZ Super: about $472k70th80th percentile, Australia: Age Pension: about $85k80th percentile, New Zealand: NZ Super: about $486k80th90th percentile, Australia: Age Pension: about $10k90th percentile, New Zealand: NZ Super: about $500k90th
Show as table
Income percentileAustralia: Age PensionNew Zealand: NZ Super
10th~310~388
20th~305~402
30th~280~416
40th~255~430
50th~220~444
60th~180~458
70th~140~472
80th~85~486
90th~10~500

The thing is, that doesn’t show the full picture.

What the charts above don’t show is that Australia provides enormous tax incentives to encourage people to put money into superannuation. Many people don’t get the Australian pension, but they receive retirement-oriented tax benefits over their working life.

There are two main breaks. One is lower tax on money paid into super: that is, instead of being taxed on your top marginal tax rate, you are taxed at a discounted rate for what you put into super. The second is lower tax on the investment returns inside of super.

Money that could have been taxed, but isn’t, is the same as money spent. So we need to factor this in. When looked at in this way, Australia’s superannuation tax concessions cost about A$60 billion a year. That’s almost the same as what it spends on the Age Pension (A$62 billion).

Australia: as much on tax breaks as on the pension
Cost, A$ billions. Different years, so approximate.
Super tax concessions (2025-26)Super tax concessions (2025-26): A$60.7bA$60.7bAge Pension (2024-25)Age Pension (2024-25): A$62.2bA$62.2b

When you look at who benefits from these tax concessions, they go to the highest earners. People earning above the median get about 83% of the benefit. The top tenth alone gets about 35% of the benefits.

Who gets Australia's super tax breaks
Share of the two concessions combined, weighted by cost. 2022-23 incomes.
Money paid inReturns inside super
0%10%20%30%40%Tenth 1, money paid in: -0.1% of all super tax breaksTenth 1, returns inside super: 0.7% of all super tax breaks1Tenth 2, money paid in: -0.1% of all super tax breaksTenth 2, returns inside super: 1.3% of all super tax breaks2Tenth 3, money paid in: 0.8% of all super tax breaksTenth 3, returns inside super: 2.1% of all super tax breaks3Tenth 4, money paid in: 2.0% of all super tax breaksTenth 4, returns inside super: 2.7% of all super tax breaks4Tenth 5, money paid in: 4.1% of all super tax breaksTenth 5, returns inside super: 3.0% of all super tax breaks5Tenth 6, money paid in: 4.9% of all super tax breaksTenth 6, returns inside super: 3.1% of all super tax breaks6Tenth 7, money paid in: 5.8% of all super tax breaksTenth 7, returns inside super: 3.6% of all super tax breaks7Tenth 8, money paid in: 7.5% of all super tax breaksTenth 8, returns inside super: 4.8% of all super tax breaks8Tenth 9, money paid in: 11.1% of all super tax breaksTenth 9, returns inside super: 7.7% of all super tax breaks9Tenth 10, money paid in: 16.9% of all super tax breaksTenth 10, returns inside super: 18.0% of all super tax breaks10Equal share (10%)Tenths of earners, lowest to highest taxable income
Show as table
Tenth of earnersMoney paid inReturns inside superTotal
1-0.1%0.7%0.6%
2-0.1%1.3%1.2%
30.8%2.1%2.9%
42.0%2.7%4.7%
54.1%3.0%7.1%
64.9%3.1%7.9%
75.8%3.6%9.5%
87.5%4.8%12.4%
911.1%7.7%18.8%
1016.9%18.0%35.0%

Tax concessions are a form of hidden spending. Once you see this clearly, you realise that the people who don’t need the help are getting it.

Australia’s 2020 Retirement Income Review put the pension and the tax breaks side by side over a whole lifetime. The Review’s own conclusion is that “higher-income earners receive more lifetime Government support in dollar terms than lower- and middle-income earners”. Its modelling suggests that, once the tax breaks are counted, people at the top of the lifetime income range (the 90th percentile and above) end up with something like $600,000-700,000 of government support. People at the bottom get about the same as those in the middle, roughly $400,000, and far less than those at the top.

Lifetime government support in Australia, by lifetime income (modelled)
$'000 per person, 2019-20 dollars, discounted. Modelled; read by eye off the Review's chart.
Age PensionTax breaks
$0k$200k$400k$600k$800k10th percentile: Age Pension about $310k10th percentile: Tax breaks about $55k10th20th percentile: Age Pension about $305k20th percentile: Tax breaks about $90k20th30th percentile: Age Pension about $280k30th percentile: Tax breaks about $115k30th40th percentile: Age Pension about $255k40th percentile: Tax breaks about $135k40th50th percentile: Age Pension about $220k50th percentile: Tax breaks about $160k50th60th percentile: Age Pension about $180k60th percentile: Tax breaks about $190k60th70th percentile: Age Pension about $140k70th percentile: Tax breaks about $230k70th80th percentile: Age Pension about $85k80th percentile: Tax breaks about $370k80th90th percentile: Age Pension about $10k90th percentile: Tax breaks about $625k90th
Show as table
Lifetime income percentileAge PensionTax breaksTotal
10th~310~55~365
20th~305~90~395
30th~280~115~395
40th~255~135~390
50th~220~160~380
60th~180~190~370
70th~140~230~370
80th~85~370~455
90th~10~625~635

Let’s put total benefits for Kiwis by percentile against total benefits by Australia:

The whole picture: lifetime government support, Australia and New Zealand, by income
$'000 per person over retirement. Australia is Age Pension plus super tax breaks; New Zealand is NZ Super only (government KiwiSaver contribution omitted). NZ: today's dollars, undiscounted, couple rate, tilt is my assumption. Australia: 2019-20 dollars, discounted, modelled. Similar ballpark, not strictly comparable.
Australia: pension plus tax breaksNew Zealand: NZ Super
$0k$200k$400k$600k$800k10th percentile, Australia: pension plus tax breaks: about $365k10th percentile, New Zealand: NZ Super: about $388k10th20th percentile, Australia: pension plus tax breaks: about $395k20th percentile, New Zealand: NZ Super: about $402k20th30th percentile, Australia: pension plus tax breaks: about $395k30th percentile, New Zealand: NZ Super: about $416k30th40th percentile, Australia: pension plus tax breaks: about $390k40th percentile, New Zealand: NZ Super: about $430k40th50th percentile, Australia: pension plus tax breaks: about $380k50th percentile, New Zealand: NZ Super: about $444k50th60th percentile, Australia: pension plus tax breaks: about $370k60th percentile, New Zealand: NZ Super: about $458k60th70th percentile, Australia: pension plus tax breaks: about $370k70th percentile, New Zealand: NZ Super: about $472k70th80th percentile, Australia: pension plus tax breaks: about $455k80th percentile, New Zealand: NZ Super: about $486k80th90th percentile, Australia: pension plus tax breaks: about $635k90th percentile, New Zealand: NZ Super: about $500k90th
Show as table
Income percentileAustralia: pension plus tax breaksNew Zealand: NZ Super
10th~365~388
20th~395~402
30th~395~416
40th~390~430
50th~380~444
60th~370~458
70th~370~472
80th~455~486
90th~635~500

To be clear: we aren’t comparing apples with apples here. Currencies and calculation methodologies differ. The thing to take away is how much is directed towards the different percentiles.

In Australia the benefits skew heavily towards the wealthiest, and the poorest get the least assistance. In New Zealand, the wealthiest get about $110,000 to $145,000 more, on the assumptions related to longevity. So both systems tilt towards the rich. But ours tilts much more equally.

These benefits aren’t just provided at a point in time. They compound. The earnings concession is a discount on the tax that would otherwise be paid on returns, so the bigger the balance, the bigger the discount, every year, on a balance the discount itself helps to grow. Look at actual balances. Most Australian households have modest super. A small group has a great deal, and that is where the concession does most of its work.

What Australian households actually have in super
Mean household super balance by quintiles of households by net worth, A$, 2019-20.
$0k$200k$400k$600k$800kLowest fifth by wealth: mean super A$22,000$22kLowestSecond fifth by wealth: mean super A$85,000$85kSecondThird fifth by wealth: mean super A$123,000$123kThirdFourth fifth by wealth: mean super A$241,000$241kFourthHighest fifth by wealth: mean super A$677,000$677kHighestAll households ($230k)Fifths of households, lowest to highest net worth
Show as table, with the split by income
Fifth of householdsBy net worthBy income
Lowest$22k$64k
Second$85k$141k
Third$123k$213k
Fourth$241k$262k
Highest$677k$500k

With this in mind, NZ’s system is more equitable. Australia means tests the pension, but in lieu of this it provides tax benefits over its residents’ working lives. These benefits compound. And in doing so – in an opaque way – it benefits those who need it least.

The tax benefits don’t show up as spending. They show up as tax not collected. This is hard to explain to people. It’s hard to get people excited about. And it’s very hard to unwind.

The whole retirement bill, as a share of each economy
Share of GDP. Australia 2025-26, New Zealand 2025. Approximate.
PensionSuper tax concessions
Australia (2025-26)Age Pension: 2.3% of GDPSuper tax concessions: 1.7% of GDP4% of GDPNew Zealand (2025)NZ Super: 5.1% of GDP5.1% of GDP

Australia may spend less than we do as a percentage of GDP, even after we factor in these tax benefits. But on a per-person basis, in absolute terms, it probably spends about the same, since Australia has a much higher GDP per capita than New Zealand, and the pie is bigger.

Some other thoughts

Locking money away

Making it compulsory to put large amounts of income into KiwiSaver might be good in general, but it’s not good for everyone. It has a real cost to people who already have enough.

If you’re on track to retire comfortably, being made to keep contributing means being forced to lock up money you might put to better use now: paying down a mortgage, helping your kids, running a business, spending more and enjoying your life.

Add tax breaks, and the people with the most spare money aren’t just forced to save, but are also incentivised to lock it away. There’s something perverse and market-distorting about this.

As KiwiSaver balances grow, it makes KiwiSaver more likely to be tinkered with in the future. I’m not a fan of making predictions, but here is one: as an increasing portion of people’s wealth gets locked up in KiwiSaver, they will want to have more choices and flexibility about how they can invest the funds. We will start to see people – and especially people who have more than they need – agitating for arcane policy changes that give them more options and choices. We’ll start to see rumblings for something resembling Australia’s Self-Managed Super Fund (SMSF) regime, which has its advantages but also represents a massive bump in flexibility and opacity.

This isn’t an immediate concern. But it is something we need to look out for.

Will larger KiwiSaver balances really represent increased investment in New Zealand?

One argument for a bigger KiwiSaver is that it creates a pool of capital to invest in New Zealand. For example, into infrastructure.

That’s often stated as if these investments in New Zealand follow automatically. Whether a bigger pool of KiwiSaver-based capital would actually end up in productive local investment, and on what terms, is a question mark.

I don’t find this argument compelling. There’s a lot of money to be invested out there already, whether from New Zealand or abroad. If these are compelling investments – and they would need to be for an investment manager to justify directing capital in this way – why aren’t we seeing these investments already?

KiwiSaver is part of the picture, not the picture on its own

Which brings me to the thing that bothers me most about calling this “the KiwiSaver election” and what triggered me to write this article after listening to the RNZ episode.

The framing treats KiwiSaver as the retirement system. But it’s a part of the system, and should be considered in this way.

If we provide significant tax concessions for KiwiSaver, it will make means-testing NZ Super significantly more likely. Whatever you think about it, those two things are intertwined. Maybe that’s fine, but you cannot look at one of these things without looking at the other.

KiwiSaver is one way to save for retirement. It’s a great system for a lot of people. But ultimately, it’s just a managed fund with a special lock on it.

For most people, KiwiSaver won’t be enough on its own. This is the case even with the proposed higher contribution rates. KiwiSaver might end up being a useful part of the picture. But it shouldn’t be the full picture.

In my experience as a financial adviser, KiwiSaver tends to act as a supplement to a broader retirement plan. If you’re relying on KiwiSaver alone to provide for a high quality retirement, then I’d be concerned.

A retirement policy discussion that focuses myopically on KiwiSaver is likely to give the impression that the answer for retirement is a bigger managed fund. It’s not always the best use of the next dollar for someone with a mortgage, someone who needs capital to grow their business, or for someone who already has more than they will ever need.

Housing

On this note, I think the focus on KiwiSaver can redirect us from something that might be even more important when it comes to retirement. Which is housing.

NZ Super generally works for people who own their home. It helps to cover living costs, but if you’re using it to pay rent, things get hard.

According to Berl, around a fifth of over-65s already rent. The Retirement Commission expects that to reach 40%, about 600,000 people, by 2048.

A retiree who owns their home and one who rents can get exactly the same NZ Super and live very different lives.

All else being equal, home ownership at retirement is what makes the difference for a lot of people. None of the KiwiSaver policies touch this. The episode didn’t touch on this.

Even the OECD agrees that housing is the problem, in its own way. The same report that recommends tax breaks for KiwiSaver (more on that below) describes New Zealand as having one of the most housing-biased tax systems in the OECD, and says this has pushed households to hold a disproportionate share of their wealth in property rather than in diversified investments. Yet its only housing recommendation is a tax on windfall gains from land rezoning. The rest of its answer is a tax break for KiwiSaver. If housing is the distortion, it seems odd to fix it by distorting something else.

Housing policy doesn’t just impact young people trying to get on the property ladder. It impacts people at all stages of life – and notably, people who are past their earning years.

Why the OECD is wrong

Earlier this year, the OECD recommended that New Zealand change its taxation settings in relation to KiwiSaver.

In simple terms, there are three moments when tax can apply in relation to retirement savings. I’ll describe them in relation to KiwiSaver:

  • when the money is earned and goes into KiwiSaver;

  • while the money is being invested; and

  • when it is withdrawn.

New Zealand taxes the first two and leaves the third alone. That is, your income is taxed at its normal rate before it goes into your KiwiSaver fund. And each year your KiwiSaver’s income is taxed in one way or another. When you take money out of KiwiSaver, you’ve already paid your tax, and it’s all yours.

Most OECD countries do the opposite. They charge no (or less) tax for income that goes in, and no (or less) tax for funds while it is invested. Tax is only paid when it is taken out of the retirement savings vehicle.

The OECD has suggested that New Zealand follow this model. It recommends taxing retirement savings on the way out rather than on the way in. Let people contribute from pre-tax pay, leave the returns alone, and tax the money as income when it is withdrawn. (It also recommends means-testing NZ Super, but I’ll put that to the side for now.)

The OECD’s proposal is tidy on paper. But it has a number of problems in practice.

  • It seems tidy on paper but would complicate things in practice. The appeal of the OECD’s model is that retirement savings are taxed once, at the time it is withdrawn. But our system already does the first two. At the point of withdrawal nothing special needs to happen, because KiwiSaver money is treated like money in any other fund. The OECD’s version would require an exemption for contributions, a second exemption for returns inside the scheme, and a new regime for taxing withdrawals: rules about what counts as a withdrawal, when, at what rate, and what happens on death or emigration. Each of those is a new change to the tax system, and a new thing for people to argue about and plan around. We would be adding when our system is already very simple.

  • With this approach, the government has to wait decades for tax revenue from a given investor, rather than receiving tax revenue gradually over time. In a sense, tax accrues to a retirement account, and is like a loan from the Treasury to the saver, which isn’t repaid for forty years or more. By the time the money comes due, it will be a different Government, a different budget, with different priorities. In the immediate term it would probably shrink New Zealand’s tax base, just as its population ages and the bill for NZ Super, which this is meant to address, comes due.

  • The tax which has been deferred may never be collected at all. A large pool of money that hasn’t yet been taxed is a political target, and the politics tend to run in the direction of not collecting these taxes.

    Australia is a case study. In its 2006 Budget, it announced that from the following year, super paid out to anyone 60 or over would be tax-free. This meant that tax that had been deferred on decades of savings was never collected. The UK has changed the rules on how its pensions are taxed over and over, in both directions. A large pool of untaxed money invites tinkering. Future political parties will be tempted to make promises to a large cohort of retirees, like forgiving tax at the point it falls due. Once it’s forgiven, it’s hard to come back. In New Zealand, we don’t have that political risk. The tax is already banked.

  • These policies privilege savings that are locked-in under whatever the regulatory constraints are at the time. Deferral only works if the money cannot be touched before retirement, otherwise everyone would route all their saving through it. It comes with lock-ins, access rules, hardship provisions, and an industry of products and advice built around them.
  • It favours one kind of saving over every other. A dollar in KiwiSaver gets a tax break. A dollar paid off the mortgage, put into a business, or saved outside the scheme does not. That pushes people towards the locked box (not to mention, the people who run those locked boxes).

    One of the OECD’s arguments for this is that Kiwis invest too much in residential property, and this would orient the market away from this use of capital. My issue is that this is a market distortion in itself. And if Kiwis invest too much in residential property, I believe this should be decided and addressed directly rather than orthogonally via KiwiSaver policy.

  • Most of the time, it’s regressive!! It usually benefits people with higher incomes and larger balances significantly more than other people. These are the people who need the support and incentive less than others.

Which brings me back to ACT and TOP. Neither of them is proposing what the OECD recommends. The OECD’s model moves the tax to the end. ACT removes the tax on earnings and puts nothing in its place. TOP removes the tax on contributions, then phases out the tax on earnings over twenty years, and says nothing about taxing withdrawals. (Presumably its other taxation policies would make up for this.) When it comes to KiwiSaver, both take the exemptions and leave out the collection. That is the OECD’s model – without the part that makes it add up.

Our system has none of this. The tax is collected now, by the government that is paying for today’s services. It’s not deferred for future political parties to forgive. It’s neutral: a dollar saved in KiwiSaver is taxed the same as a dollar saved anywhere else, so people aren’t incentivised to lock money away and then push for complex changes to the regime.

The OECD is suggesting that New Zealand adopt a system that can be quietly hollowed out. In essence, ACT and TOP are suggesting we skip the system and go straight to the hollowing.

Conclusion

I don’t have a tidy one. What I have is a sense of exasperation at the quality of “the discussion” about KiwiSaver, which was a term used several times throughout the RNZ episode.

New Zealand has a retirement system that is simple, cheap to run and, in my view, more equitable than the Australian one it keeps getting compared with. The tax is collected now rather than deferred for a future government to forgive. A dollar saved in KiwiSaver is treated the same as a dollar paid off the mortgage or put into a business. And NZ Super goes to everyone, which is the part that does the heavy lifting for people on low incomes.

So the positive case is a conservative one: don’t break it. Bigger contributions, compulsion, tax breaks and longer lock-ups each sound like common sense on their own. Taken together they could see us drift, one policy at a time, towards a system that is more complex, more regressive, and more expensive to run.

Some of the questions I ask when I think about these sorts of policies:

  • What is it likely to do to the whole system, not just KiwiSaver? I believe that providing additional tax concessions makes NZ Super more likely to become means-tested over time.

  • Does it favour one kind of saving or investment over others? Remember that KiwiSaver funds are just managed funds with a special lock on them. Most of the large KiwiSaver providers offer funds that are similar or identical to their KiwiSaver funds outside of the KiwiSaver environment.

  • Does it do anything for the two in five retirees who may be renting by 2048?

That last question is where the real gap is, and no KiwiSaver policy touches it. If we want to spend political energy on retirement, housing is probably where it would do the most good.

To the extent there is a “discussion”, a lot of it comes from people with an interest in the outcome. Every one of those policies means more funds under management, and more fees. That doesn’t make them wrong. But it’s worth knowing who’s talking, and why.

I’ve said my piece (for now!). You can decide how much of that applies to me too.