The superannuation sky is not falling

When there isn’t much, if any, political or community impetus to do anything about a looming issue, it can still be useful to be told that the sky isn’t falling –  at least if that analysis is correct – but it probably isn’t an approach likely to attract too many readers.

The New Zealand Initiative last week released just such a report on New Zealand Superannuation, under the (slightly laboured) title Embracing a Super Model: The superannuation sky is not falling. (I was among those who provided comments on an earlier draft of the report.)

There are lots of interesting charts, even if perhaps most are familiar to anyone who has been reading in this area.  And there are helpful reminders of the (very) good features of our NZS system

There is a lot to like about the NZS model:

  • Low poverty rates: The material hardship rate for the elderly is low compared to other groups in New Zealand and is one of the lowest compared with European countries. The standard hardship rate for superannuitants is 3%, compared with 11% for the whole population and 18% for households with children.
  • Relatively affordable: NZS is more affordable than public pension schemes in many OECD countries, both today and in 2050. At around 8% of GDP, the projected public expenditure on NZS in 2050 is still lower than what many OECD countries are spending today. These include oft-acclaimed systems like in Denmark, Finland, Norway and Sweden.
  • Simple and efficient: NZS does not distort incentives for employment and savings as much as means-tested systems. When an NZS surcharge was introduced from 1985 to 1998, people went to great lengths to avoid paying it by hiding their assets. The simplicity of a universal benefit also lowers administrative costs.

(although, as I noted in a post a few months ago some OECD data appear to raise questions about those relative poverty rates.)

Our system has the further merit, at least in my view, that it explicitly focuses on providing a modest level of income support, leaving the responsibility for any higher material standard of living in old age a matter for individuals and families.  (Having said that, the tax system we have had in place since 1988/89 –  taxing income on savings made out of after-tax income at least as heavily as income from labour –  is quite out of step with that particular vision.)

When it comes to recommendations for change, the New Zealand Initiative report is curiously bloodless.  I agree with some of their recommendations, disagree with others, and noticed an important omission.  But with no sense of any fiscal urgency, the author seems a little at sea.  For my tastes, there was a missing moral dimension – a sense of right and wrong.  Debates about how we care for our elderly seem almost inescapably moral in nature.  Of course, economists have little or nothing distinctive to add in that area, but the Initiative seems reluctant to even attempt to make a case.

Their first recommendation is one I’d agree with

Recommendation 1: Link the pension age to health expectancy

Doing so would save some money –  potentially quite a lot of money over time.  But to me, the stronger argument isn’t about saving money per se, but about a sense of right and wrong.    In an age when most people aged 65 are perfectly capable of working –  thanks to the changing nature of jobs and the improvements in the health status of people –  what possible case can there be for paying a near-universal living allowance, raised by taxes with all their deadweight costs, to everyone of that age?   No one argues –  the Initiative certainly doesn’t –  that people who are physically unable to work should have to, but that is true of people at any age (it is why, for example, we have the Invalids Benefit).   I also don’t have a problem with society agreeing that it doesn’t expect people past a certain age to provide for themselves (or within families), unless they particularly want to work.  But given the health status of most people aged 65 how can 65 possibly be the appropriate age now?   As a chart in the report illustrates, more than 50 per cent of men aged 65-69 are still in the labour force.

I’m much less convinced by the second recommendation

Recommendation 2: Index NZS to CPI only rather than both CPI and wages

• NZS is indexed to both inflation and the average ordinary time wage. Decoupling NZS from rises in wages is a way of ensuring productivity gains reduce the costs of NZS. The real purchasing power of NZS should remain the same while the real purchasing power of wages would increase.

Although it isn’t quite stated this way, this recommendation is an assertion that the relative living standards of a large chunk of elderly New Zealanders are too high (for the bottom four deciles of the over-65s, NZS makes up almost their income).  It is to guarantee a material increase in the relative poverty rate of older New Zealanders, substantially so over, say, a 20 or 30 year horizon. In fact, what would be likely to happen is that a whole raft of means-tested forms of assistance would be added to the system, detracting from one of the great strengths (see above) of the current system.  Also, even if the analysts recommending CPI indexation rather than wage indexation are willing to live with the full ramifications of such a system  –  in principle, 100 years from now the real value of NZS would be the same as now, even though real wages might be several multiples of what they are now –  the political system just won’t do so.   Break the link to wages now, and it is likely to be back a decade from now.

(One plausible compromise recommendation might be to lock in the real purchasing power of NZS at the point a person first receives it –  eg you might get 65 per cent of the average wage as it was when you turned 65 (or 68) –  and the person turning 65 (or 68) five years hence would get 65 per cent of average wages then.  Both would only be CPI-indexed from there forward, but future old people would get to share in the productivity gains the community manages to secure.  This approach would parallel how private defined benefit pension schemes work.)

What of the third recommendation?

Recommendation 3: Contributions to NZ Super Fund should not be at the expense of paying down debt

The Super Fund should not be relied on to reduce the future costs of NZS (it cannot do that), and contributions to the Fund should not come at the expense of paying down debt.

I get the impression that the New Zealand Initiative isn’t very keen on the New Zealand Superannuation Fund, but is reluctant to call a spade a spade and call for its disestablishment.  There is an analytical point to be made-  NZSF doesn’t materially affect the future affordability of NZS –  but there is an at least equally important debate to be had about whether runnning a highly-leveraged (wholly leveraged) investment fund trading world markets –  and making politically convenient plays whether around climate change, light rail, or whatever – is any sort of natural or appropriate role for government.   I don’t think so, and I doubt the Initiative does either, but they seem strangely unwilling to say it (I guess they want to keep on good terms with the government).  Note that the existence or not of the NZSF is a different issue from the question of whether governments running a welfare system, especially for old people, should also run much lower levels of net debt (even net assets) than some stylised government doing only law and order and infrastructure might. I think they should.

And what of the fourth recommendation

Recommendation 4: Productivity growth will make NZS – and everything else – more affordable

Faster rates of productivity growth relative to increases in the real interest cost of government borrowing can allow increased government spending without falling into a public debt spiral. Raising productivity growth is a way of making NZS (and everything else) more affordable, and gives future governments more options and flexibility to adjust to changing economic and political circumstances.

Well, of course, although on the narrow NZS point this is a less-strong argument than it appears.  When net debt is near-zero, debt servicing costs aren’t a particularly important consideration.  The Initiative argues for CPI-indexing partly so that productivity gains will improve the fiscal position, although that seems to me to put the emphasis in the wrong place.  Faster productivity growth –  and recall that ours has been lamentable for decades –  offers the prospects of better material living standards for almost everyone  including, as they note, flexibility about support for the elderly.

But in practice, this isn’t so much a New Zealand Initiative recommendation as an aspiration.   We’d all prefer that productivity growth had been, and would be in future, faster. But wishing it doesn’t make it so, and the Initiative hasn’t been particularly strong on identifying the key factors, or policy issues, that might explain that failing and offer credible New Zealand-focused pathways out of it.

Finally, turning back to NZS itself, it was striking that –  as far as I could see –  there was no discussion in the report of the rather weird aspect of our system: that in a country with so many immigrants and emigrants, we offer a universal benefit to anyone who has lived in New Zealand for 10 years after turning 20, including 5 years after turning 50.  It is made worse by the fact that we have Social Security Agreements with various countries, notably Australia and the UK, which mean that residency in those countries counts as residency in New Zealand for NZS purposes.  Is this affordable?  Perhaps so in the same sense the New Zealand Initiative notes that the overall NZS system could be afforded. But is it right?  Well, that seems like a moral question –  informed no doubt by analysis –  and one where I’m pretty clear what the answer should be. It is simply wrong.

Welfare systems should be about “looking after our own”, and if you went to Australia at 20 and spent your entire working life there, I don’t see any good reason for New Zealand taxpayers to support you back here in retirement (of course, we don’t know how material these numbers might be).  Or if you happened to come to New Zealand first at 55.    A graduated system, in which NZS payments are proportional to the time spent in New Zealand between 20 and 65, seems both fair and fiscally prudent.   Take the 10 year residency (real residency) as a starting point at which you might get, say, a third of the standard NZS at 65, and scale it up so that after say 30 years you get the full benefit.  (Will there be a few hard cases? No doubt, but that is where charity and family support should be expected to fill the gaps.)

It will be interesting to see what, if any, NZS policy the opposition National Party comes out with.  The previous government, at the very end of its term, and having changed leaders, did promise to phase in –  very very slowly – an increase in the NZS eligibility age to 67.  But only if they were re-elected, which they wern’t.  And doing nothing about other features of the system –  dealing with any life or health indexation (in full or in part) –  or the very short residency requirements.  With Labour and New Zealand First seemingly fully committed to the current parameters of the system, it would be a brave Opposition to campaign for change, especially from a party with little obvious sense of an ability to engage on matters of right and wrong.   One should probably never wish for a recession –  especially now given the limited capacity of the authorities in so many countries to respond –  but perhaps it will take a recession to get our leaders to more seriously address the NZS issue.  That was, after all, what it took in 1989 and then 1991 when Labour started, and National greatly accelerated, the move back to 65.


21 thoughts on “The superannuation sky is not falling

  1. Your analysis of our super arrangements is bang on, economically and morally. The indexation to CPI post-retirement is an interesting and sensible idea and the linking of entitlement to residency post age 20 is important. Abolition of NZS is also the right thing to do – it was really a device to constrain spendthrift, generally Labour, governments. Michael Cullen was afraid he would not be able to control the spending ministers in the Clark Government.

    I don’t understand, however, why you think a state-guaranteed retirement income is inconsistent with taxing income from savings. The large gap in the income tax base due to exclusion of owner-occupied housing eviscerates taxes on capital income. It implies you think taxes on consumption and/or labour income should be relatively higher.


    • THanks.

      My point was mostly about the combination of a system that provides a universal modest income (while expecting people to provide for themselves beyond that – thus unlike many European systems where the state system offers a (personal) income related pension) and an undue tax burden on capital income. But my starting point is that capital income is generally overtaxed (and I’m not fully convinced by the economic or – more especially moral – case for taxing imputed rents). We have among the highest share of GDP taken in capital income taxes of any advanced country (and, perhaps not coincidentially modest savings rates and low rates of business investment). With the fiscal surpluses that we are currently running I would be happy to use the margin to markedly reduce capital income taxation (and, of course, the rates at which interest can be deducted). Without doing the numbers (and not assuming in dynamic gains from eg more business investment) I don’t know how deep a fiscal hole that would leave, but there are expenditure saving opportunities (including on NZS) to look too before raising consumption taxes further.

      (Not, of course, that this is ever likely to be very feasible politically, but we are a bit of an outlier in how onerously we tax many forms of capital income, including because we don’t have – labour only – social security taxes).


    • The big problem with taxes is the misallocation as we have seen with Shane Jones, Pablo Escobar style of dishing out cash to anybody in the regions prepared to put their hands up and note in their feasibility analysis that they will vote for NZFirst at the next election. $3 billion in wasted tax dollars to buy votes for NZFirst. As Shane Jones would regularly tell us “To the victors the spoils” and how he chooses to throw away NZ tax dollars is entirely at his personal discretion.


  2. Social Security Agreements help address the residency issue you have described by deducting any other country, tier one pension entitlements from a person’s New Zealand Super. There is a bill before the House at the moment which proposes to increase the residency criterion to 20 years but if this were to pass it would shave less than 2% off the cost of NZS while probably having a number of negative consequences


    • But if you leave Australia at 65 having cashed out your compulsory private super, large enough to have no entitlement to the Aus age pension there would be no such deduction would there?

      2% of a large number is still worth having, but I’d argue the case for change (as per the model in my post) as more a moral one than a purely fiscal one.


      • That’s probably the case but I wonder if it might be possible to address the anomaly re the Aussie pension directly rather than via a more sweeping measure (changing the residency criterion).


  3. Link the pension age to health expectancy. As you say it is a moral issue and a sensible response. It will become more of a moral issue if in the near future millionaires are able to buy DNA tweaks that extend their life. Effectively we already are in that situation with anyone over 65 needing Keytruda and paying for it themselves.
    There is a moral minefield with life expectancy of women being longer than men and desk-workers being longer than manual workers.


  4. Buying a house on mortgage and paying it down over one’s working life while accumulating tax free gains is the commonest means of personal lifetime saving in NZ and arguably the soundest as it hedges the cost of retirement housing. Retirees owning their own home supplemented NZ Super are comfortably well off as you note whereas retirees that rent are much less so but qualify for other income-tested welfare support, if in need. That’s the core of our system and it works tolerably well.

    Given that taxing owner-occupied housing is off the political agenda (note however local authority rates are a significant tax on all property) it is hard to justify tax breaks for any other capital income let alone subsidies for private saving via Kiwisaver. Most other capital income is already subject to income tax, however, so the capital gains tax currently being floated by the TWG is really only a shift in timing and will be administratively challenging to say the least. Tax experts to my knowledge question whether the complexity is worth it. to the extent that there are long term capital gains on investment properties (essentially appreciation in land values) the touted capital gains tax may divert private investment into other avenues (financial assets) and result in more limited supply and higher rental yields but I suspect the impact would be marginal.


    • The much touted capital gains tax also applies to all other personal investments and not just isolated to investment property.


  5. Thanks Michael. Don’t particularly disagree without your additional proposals, but I’ve a hard time seeing their making that substantial a difference to the overall costs of NZ Super. I’m certainly not opposed to Super being scaled to years of work in New Zealand so someone coming in at age 50 and working for 15 years here receives a smaller entitlement than someone who starts work here at age 25 before taking retirement. Could that possibly make more than rats-and-mice difference to the costs at 2050 though? And it looks like the government is already moving to do it anyway.

    I would quibble about your suggestion that the report does not make a moral case about raising the age of entitlement. The report argues that maintaining the current eligibility age will come at greater and greater opportunity cost where the funds devoted to the could-be-working healthy older population could be put instead to other more pressing needs. That’s a moral case, no?


  6. I was hoping the report would address the rising opportunity cost of New Zealand’s primarily PAYGO-funded Superannuation scheme on young people- a subject almost always ignored in the debates about NZ Superannuation. NZS may be a not-too-bad scheme for people born before 1970, but those born afterwards will be paying a much larger sum for the pensions they will receive than those born in earlier years. Peter Diamond showed us how to calculate these opportunity costs 50 years ago; it is strange that the Treasury has never been interested in calculating them, but maybe it is because it would show that the opportunity costs are likely to keep rising, even if the age of entitlement is increased by one or two years a decade.

    The resumption of the funding of the NZSF will help to reduce these opportunity costs, by altering the amounts current and future generations pay for the pension. (That is, if the Honourable David Parker can keep his fingers off the fund.) But it is not the only way to reduce the opportunity costs of a PAYGO pension scheme on young people and future generations, and it would be useful if NZ had a proper debate about the way to design a scheme that is in the interests of young people as well as oldies like me.

    I can’t say this too often: if you read Gokhale, Kotlikoff, Sefton and Weale(2002) you might have a different opinion about the way a pension scheme like New Zealand’s can have a detrimental effect on long-run wealth inequality even if reduces income inequality among the elderly. Forcing lower income people (who often have lower life expectancy (conditional on gender)) to provide for their retirement through a system that starves them of bequeathable assets does not generate an equitable distribution of wealth. It is possible to design schemes that reduce income and wealth inequality, but NZ policy makers have expressed little interest in thinking through this problem.

    Gokhale, Jagadeesh, Laurence J Kotlikoff, James Sefton, and Martin Weale (2001) “Simulating the transmission of wealth inequality via bequests,” Journal of Public Economics 79(1) 93-128


  7. In a similar vein to Andrew’s comments, given that our economic system is fundamentally focused on price stability (+2%) that absolutely requires a % of the population to be unemployed, effectively through fault of their own. Individuals may move in or out of unemployment but as a group there will always be a % unemployed.

    There are thus good moral and ethical grounds that this group of people should receive a similar level of entitlement as those eligible to qualify for superannuation.


    • Probably a better policy than, unemployment remuneration equivalent to super, would be a job guarantee. The unemployed miss out on much more than just income but also career development and the good mental health that comes from engaging in productive employment.

      Though the term is (probably intentionally) overloaded I expect our host thinks of full employment meaning near the NAIRU rate. If you accept that you must also agree with being able to give 110% because the economy is above it today.

      Anyway its probably correct to say actual full employment cant only be attributed to just monetary policy. It frequently requires appropriate fiscal policy to be achieved.


    • ?

      1) Yes, in a totally deregulated market with no social support everyone would have to work (or be supported by family) and minimum wages would be very very low, but rise & fall to reflect demand, or

      2) The government could be employer of last resort removing all unemployment (

      But I don’t see given empirical evidence from all other countries running price stability regimes (& a welfare system & all the other distortions in a modern economy) that any have zero unemployment, or can maintain “full employment” for little more than an instant.


      • But material levels of unemployment usually result from the interaction of regulatory/stautory provisions such a minimum wage laws, unemployment benefits rule, and rules around how easy (or otherwise) it is to hire and fire. There may very well be good reasons for all those provisions, but they are the culprit, not monetary policy.


      • Interesting read, thank you.

        I have to say that if I hear any government suggesting trying to do away with NZ Super (or devaluing it in relative terms by linking only to the CPI) anytime soon,I will be investing my savings in a Yellow Vest supply company.


      • Shouldn’t your Yellow Vest be out by now in line with the brave French protesting out in force over Fuel taxes?


  8. Peter Dunne had the sensible idea that the age for starting to collect National Super should be flexible (60 to 70 – and I would push out to 75), with those getting it a younger age early getting comparatively less and those waiting getting a little more. Giving that flexibility should also making raising the mid-point age more acceptable.

    Once you give some slight benefit for deferring then all those people still working and earning a good income have more of an incentive to keep working and putting off collecting National Super. You would probably also find more moral pressure coming on well paid people (like parliamentarians) to not collect their National Super until they actually need it.


    • The cost of super on the economy is the actual material cost of retirees consuming goods and services of the economy without producing them. On the other hand the financial cost no longer functions as a cost at the macro economic level as, pensioners generally spend their income which makes them having an income a support for the economy and taking that away (relatively speaking) depressant on the economy. Instead of a cost pensions are a way of the government providing more or less support to the economy as it goes (while giving pensioners an income).

      I don’t know of anybody who thinks NZ has a pensioner induced shortage of real goods and services.


    • Can’t see the point of deferring the age for Universal Super entitlements when Universal basic wage is being considered as future entitlements.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s