New Zealand Superannuation is akin to a Ponzi scheme, whose demographic underpinnings are starting to crumble.
The Treasury estimates that as the population ages the cost of NZ Super, measured against the size of the economy which has to support it, will almost double by the middle of the century.
It is not an insoluble problem, though, just one which requires two things that tend to be in short supply in New Zealand: a willingness to save and a sense of intergenerational fairness.
Mark Twain said: "It ain't what you don't know that gets you in trouble. It's what you know for sure that just ain't so."
Among superannuitants it is a deeply cherished belief that they are only drawing out of the system what they have paid in over their working lives.
But some careful analysis by Andrew Coleman, an economist at the Wellington think tank Motu, shows that on average what we pay in as taxpayers only covers about half of what we can expect to get out as pensioners.
The rest is a transfer from each generation to the one before.
Like a Ponzi scheme it depends on an ever-growing number of people joining the scheme (taxpayers in this case) to fund an outsized payout to their predecessors.
But, as is well known, the combination of rising life expectancy and falling fertility is changing the age structure of the population, relentlessly raising the ratio of superannuitants to those of working age.
New Zealand is almost unique among developed countries in the extent to which it relies on a taxpayer-funded, one-size-fits-all, pay-as-you-go (PAYGO) scheme to fund retirement incomes and how little we rely on savings and investment.
In research outlined in a recent lecture, Coleman looked at whether the PAYGO system could over time be turned into a prefunded save-as-you-go (SAYGO) system which would deliver the same entitlements at a lower cost to future taxpayers.
"I'm not saying this is what we should do. But would it work? I'm saying it looks feasible and it probably isn't even unfair."
He argues that on some plausible assumptions about rates of return to capital (3 or 4 per cent per annum real) a SAYGO scheme, when mature, is about twice as efficient as PAYGO.
That is, it will fund the same pension at half the cost in taxes or twice the pension at the same cost.
Coleman's modelling suggests that under a wide range of circumstances the tax rate under a SAYGO system will eventually be lower than under PAYGO by between 2.5 and 5 per cent of GDP, or by 4.5 to 9 per cent of labour income.
"There is no real magic going on there. It's the power of compound interest," Coleman says.
"If prefunding were to happen the economy, or at least gross national product, would be larger because people are delaying their consumption and the capital stock will be increasing. There would be a larger pie, so a smaller slice of the pie would be required to fund retirement benefits."
The problem is how to get there from here.
The difficulty has always been that a transitional or sandwich generation has to pay twice.
They have to support their parents or grandparents under the existing scheme while contributing to their own super, perhaps through payments into an expanded version of the Cullen fund.
But Coleman argues that we have a lot of options in designing that transition.
The end point does not have to be a pure SAYGO system. It could be a hybrid of the two.
And there are choices about how long the transition period is, and how the prefunding contributions are funded. But the essential trade-off is that taxes rise in the near term so that they can be permanently lower in the long term.
One of the scenarios Coleman models is a transition to a scheme that is three-quarters SAYGO and one quarter PAYGO in 60 years time.
It ends with NZ Super costing the same share of GDP as it does now.
But it requires taxes to rise by 2.8 per cent of GDP immediately, instead of in 25 years time, and it takes 37 years before the tax burden required to fund the system starts to fall below the PAYGO track.
To put that in context, the increase in taxation up front is about double what will be required in any case when the Government resumes contributions to the Cullen fund - as is its policy, fiscal circumstances permitting.
Though the transitional generation needs to pay twice, that does not mean paying twice as much but rather paying for both systems.
And it is arguably not unduly hard on them, given than under the status quo they can expect to get twice as much out as they contribute in over a lifetime of paying tax.
"If New Zealand makes the transition to a SAYGO-funded pension scheme, the biggest beneficiaries will be generations of yet-to-be-born New Zealanders who will have a pension scheme that provides them the same payment stream at half the cost of a PAYGO system," Coleman says.
"If it does make the transition, the Government would still need to decide the extent the scheme was public or private. However, if it wishes to make the transition it should act soon, for, as Australia has discovered, the demographic baby-boom makes the transition considerably easier than it would otherwise be."
One potential difficulty politicians might face in selling a higher level of prefunding is what might be called the dancing Cossacks problem.
That is a reference to the calamitously successful campaign Robert Muldoon used to defeat the Kirk government's early attempt at a form of SAYGO, raising fears that the fund would end up owning everything and it would be communism by stealth.
If we kept the existing structure, including the New Zealand Superannuation Fund, a mature, fully SAYGO system would require a fund of the order of $600 billion, Coleman calculates.
For a hybrid system it would be correspondingly smaller.
It would need to be politician-proofed and to a substantial extent invested abroad.
While an asset pool of this size is unprecedented in New Zealand, Norway already has a sovereign wealth fund of around $500 billion, he says.
New Zealand is in no danger of saving too much.
On the contrary the size of our current account deficit (rising again) and resulting conspicuously high stock of foreign debt testifies to the extent to which national saving falls short of funding our investment needs.
It leaves us vulnerable and creates a wedge between what we produce and what we get to keep.
In addition our internationally eccentric way of funding retirement income rests on increasingly outdated demographic assumptions - a foundation which will only deteriorate faster as people figure out there are not only higher incomes to be had overseas but far better pensions too.