We suffer from a tax system which for many years has told us that if you want to provide for your old age, don't save money. Instead, borrow money — as much as you can — and buy housing.

Among the questions the Tax Working Group chaired by Sir Michael Cullen is seeking submissions on, is whether the tax system should encourage saving for retirement as a goal in its own right, and if so, what changes are required.

Everything the discussion document says about the broader context for the review points to a resoundingly affirmative answer.

It points to the ageing of the population, which will see the ratio of those 65-plus to those aged between 15 and 64 doubling over the next 50 years, from one to four to one to two. That, surely, is an argument for encouraging more private provision for retirement income.


It points to the changing nature of work and the need for the tax system to reduce its reliance — unusually high by international standards — on taxing labour income, in favour of taxing capital income. But that capital income has to be earned in the first place, and our negative household saving rate is not a good sign in that respect.

It points to concern about mounting inequality of income (after housing costs) and of wealth. Rampant house price inflation has been a major driver there.

And it quantifies the infamous extent to which investment in housing is tax-advantaged compared to financial assets. It puts the effective marginal tax rates on bank deposits, shares, PIE vehicles and superannuation funds in the 47-56 per cent range, while for equity in rental property it is 29 per cent, and 11 per cent for owner-occupied housing.

If these numbers seem high, it is because it is the tax on the real return these investments generate. As tax is levied on nominal returns, the bite it takes from real returns is a lot higher.

The 2011 Savings Working Group recommended that interest income and expenses be inflation-indexed for tax purposes. It is a recommendation the Cullen review might want to consider.

In the case of superannuation funds, the paper says New Zealand's tax treatment is "distinctive". That is officialese for exceptionally harsh and forbidding by international standards.

The normal approach is to defer the taxation of income locked away for retirement until the saver is free to spend it. New Zealand has, since Sir Roger Douglas introduced changes nearly 30 years ago, front-loaded the taxation.

Not only are savings made out of after-tax income, the income those savings generate while entrusted to a fund manager is taxed as it accrues.


The taxman makes off with his share of the investment income as soon as it is earned, while the saver has to wait years, even decades, to access his or her share of it.

This means the pot of money increases more slowly than it would if compound interest were able work its magic unmolested by the taxman.

And worse, it also provides a stark contrast to investment in housing (owner-occupied or rental), where leverage amplifies the equity gains from what is almost always a rising market.

Andrew Coleman argues in his paper Housing, the 'Great Income Tax Experiment' and the intergenerational consequences of the lease that the distortions arising from the adoption of the taxed-taxed-exempt (TTE) treatment of retirement savings vehicles nearly 30 years ago could also be addressed from that side, rather than changing what is an internationally normal approach to the taxation of housing investment.

Cullen probably agrees, unless he has changed his mind since, when introducing KiwiSaver, he described Douglas' resort to the TTE regime as the biggest exercise in intergenerational theft he had seen.

Yet disappointingly, the Tax Working Group's perfunctory discussion of TTE displays a kind of cognitive dissonance from everything else in the document. It just rehashes the Treasury dogma from 30 years ago:


"This approach ensures that economic distortions to save in a retirement account instead of through other savings are minimised. New Zealand is also unusual compared to other OECD countries in that New Zealand Superannuation is universal, so additional retirement savings do not reduce the amount of New Zealand Superannuation paid out. If New Zealand were to switch to a system where income earned on the investment was not taxed, the fiscal cost would be significant."

Surely, the distortion to worry about here is not between saving in a retirement account and other forms of financial asset, but rather between it and housing.

People can sell shares or withdraw money from the bank at will (even if there is some financial penalty). They are liquid assets. It is entirely reasonable to treat long-term locked-away retirement savings differently, as other countries do.

It also simply false to assert that other retirement savings do not reduce NZ Super. It is taxable income and the tax rate which applies reflects other income, including investment income.

And clearly, any change to the tax rules will have fiscal consequence. It would be more useful for officials to quantify it in this case than to cite it as grounds for defending the status quo.

To quote the Savings Working Group again, "... the structure of New Zealand's tax system is more biased against saving than in comparable countries. Further, with no capital gains tax the New Zealand tax system is biased towards a large block of economic activity — property investment — that tends to be at the less productive end of the spectrum."


The perverse tax treatment of retirement savings is not, of course, the sole cause of runaway house price inflation and the social ills that flow from that.

It is not the sole cause of our negative household saving rate either, or the fact that our external deficit is once again growing faster than the economy.

But there is too much of a tendency in debates like these to adopt a competitive approach, where people say, "that's not the problem; this is the problem. That won't fix it; we have to do this other thing."

As if we were only allowed one problem and one solution.