New Zealand's tax system is exceptionally distortionary with respect to housing - but the way to correct that is not to tax housing more harshly, but retirement saving more lightly.
That is the thesis advanced by Otago University economist Andrew Coleman in a recent paper, Housing, the 'Great Income Tax Experiment' and the intergenerational consequences of the lease.
He argues that there is nothing unusual about the way New Zealand taxes owner-occupied housing. People buy, or pay off, their homes out of after-tax income but neither the imputed (avoided) rents they gain nor the capital gain upon sale are taxed. In the jargon, it is TEE - for taxed-exempt-exempt.
But that is in stark contrast to the way the government has taxed retirement saving since 1989. The money put into retirement savings schemes comes out of after-tax income, then the income those savings earn while in the hands of a fund manager is taxed and it is only tax-free when paid out to the saver upon requirement.
This TTE (taxed-taxed-exempt) model is consistent with the way retail bank deposits are taxed, but is very unusual by international standards.
Most developed countries - Coleman lists 22 - tax retirement saving on an EET or expenditure basis. That part of a person's income is taxed not when it is earned, but when it becomes available to the saver to spend. Money put into government-approved savings vehicles is exempt from income tax and so is the income it generates as it accumulates. It is only taxed when it comes out.
If owner-occupied houses were taxed on the same basis, they would have to tax capital gains as they accrue and also imputed rents (as Gareth Morgan's Opportunities Party proposes).
That, as Coleman drily notes, "does not appear to be politically palatable". So if we want to reduce the distortionary consequences of the current system on housing markets, we should consider moving to the standard back-loaded EET model for taxing retirement savings accounts.
The distortionary consequences are serious and largely arise from the second T in TTE. Taxing the income generated by savings as they accumulate lowers the opportunity cost of instead putting hard-earned money into the roof over your head.
You would expect the consequences to be higher house price inflation, a higher house price to rent multiple, larger new dwellings and a higher premium for land close to amenities.
And, indeed, Coleman finds that the stylised facts are consistent with what theory would predict.
Between 1990 and 2014, house prices rose by an average 5.7 per cent a year, or 3.5 per cent a year when adjusted for consumer price inflation. Most of the real price increase occurred after 2000: 4.2 per cent a year compared with 2.5 per cent in the 1990s and a real decline of 0.8 per cent a year between 1975 and 1990.
Many developed countries have experienced high rates of house price inflation since 1990 but it has been especially virulent in New Zealand.
Between 1990 and 2016, real house prices in New Zealand rose 221 per cent compared with 157 per cent in Australia, 39 per cent in the United States and just 1 per cent in Germany.
"There are many possible explanations for the house price increase, of which tax changes are but one," Coleman says, but the data do indicate that it has been large and persistent by international standards.
The data also show a steep rise, at least since 2000, in the ratio of house prices to rents.
And the average size of newly constructed houses has increased faster than in Australia or the United States, and is amongst the largest in the world, Coleman says.
"However, it is not possible to definitively attribute these changes to the tax changes, as other factors that affect housing markets such as the international decline in interest rates and rising incomes have also changed since 1989."
He has tried using econometric techniques to unpick the relative importance of the various influences, but concludes it is not possible to estimate the role of the tax changes.
"The evidence is consistent with the conjecture that the tax changes are part of the cause of the change, but cannot prove it." But on the plausible assumption that tax has been and still is a major driver of house price inflation, the intergenerational effects are dire.
If the same tax distortions which have propelled house prices to levels that shut many young people out of the market and create a Generation Rent also discourage the accumulation of financial assets, that is a particularly cruel combination.
The solution Coleman advocates is to admit that the great income tax experiment of 1989 has not worked as intended because it proved too difficult to tax capital income properly, and undo the reforms by taxing income placed in retirement income schemes on an EET basis, like most other developed countries.
The tax man should not get his cut until the income is available to the saver to spend.
EET for retirement saving and TEE for housing are not exactly equivalent, because realised capital gains are taxed under the latter but not the former, and different marginal tax rates may apply. But scrapping the second T is the crucial thing to avoid distorting investment choices.
No longer front loading the taxation of retirement savings would come at a fiscal cost.
Some of that might be offset by revisiting some of the tweaks to the tax system which have been introduced over the years to nibble at the distortion. One is the modest tax credit available under KiwiSaver. Another is the PIE regime which caps the tax rate of the second T at 28 per cent when the top marginal income tax rate is 33 per cent.
It may also be necessary to grandfather existing TTE schemes.
Such transitional issues fall outside the scope of Coleman's paper.
But he is surely right to observe that if the Government has fiscal surpluses at its disposal this would be a useful thing to do with them.