Susan St John says the issue of whether the NZ Super Fund should pay tax is largely one of smoke and mirrors

TBF Kiwisaver is so relatively young that at most you likely had your annual salary in your account, whereas the amount people have at retirement is going to exponentially increase over the next 25-30 years as people have been in it for longer. To the point where people will reach 65 and could have (very roughly) 10-12x their salary in Kiwisaver. 

Not taxing it could incentivise this to be even higher but we need to look at who can then access Super (particularly as any tax incentive would proportionately benefit higher income earners) as this lower tax should come with the benefit of not having to provide universal Super. 

This is an area where Susan and i largely disagree. As i wrote in my series, I strongly prefer an EET system for the taxation of retirement savings, and as I mentioned on my podcast there are good reasons why the government should not tax the NZ Superannuation Fund. In both cases, i suspect Susan and I would agree that  these policies would lead to an increase in government debt. But the evidence suggests it is also likely to lead to an increase in the Government net asset position, which would allow a reduction in overall tax rates (although it could also allow an increase in government expenditure). I have no difficulty with an increase in government debt if it is associated with a larger increase in high yielding assets, and thus an increase in net government wealth. This position is common internationally, but successive NZ governments have been afraid of debt, even if it finances productive assets. 

First, the  taxation of the NZ Superannuation Fund. Suppose the government collects $1 billion in tax from the NZSF this year. If it didn't tax, it would issue an additional $1billion debt to make up the lost revenue, but it would also have $1 billion invested in the NZSF. The NZSF typically earns a return on capital rk that is higher than the interest rate the government pays on bonds. This is likely to occur for two reasons - (i) government debt is considered a very liquid safe asset by most investors, so it offers a lower return and (ii) the NZSF has an advantage over most private investors because it does not have to hold liquid assets and therefore should outperform. Indeed the return to NZSF has been very high over an extended period of time, and can be expected to continue to be much higher than the return to government bonds. it follows that if the government ceased to tax the NZSF it  would increase its debt while initially keeping its net asset position the same; but over time the likely excess return will increase the net asset position, ultimately requiring less tax revenue. It seems like a good deal. 

A related argument can be made about the EET tax treatment of Kiwisaver funds. There are actually many reasons why EET is the system that is preferred internationally (it is adopted in a majority of OECD countries including almost all of the large ones and almost of the countries we associate with progressive tax systems and low inequality such as Germany and Sweden) and there are very good theoretical reasons why countries like it: these were thoroughly discussed in the 2010 United Kingdom "Mirrlees Review" of the tax system by Nobel prize winning economist James Mirrlees.) Under an EET system, the government would exempt income placed in KiwiSaver accounts from tax when it is earned. Thus the government gets less tax revenue initially. It also exempts dividend and interest income earned as the funds accumulate. For these two reasons, the government would  initially collect less tax. However, it collects tax on the whole sum when the person withdraws the sum. If the KiwiSaver account earsns more than the government rate of interest, and if the tax rate on the Kiwisaver withdrawal is the same as the tax rate the person would have initially paid, then the government will ultimately collect more revenue from EET than the revenue collected under our current system, even adjusting for the interest the government pays on the bonds it initially issues to make up the shortfall. This can allow taxes to be reduced (or government services to be increased). The difference can be immense: given the amount a balanced portfolio will return in excess of the government interest rate, the increase in returns over the lifetime of an individual can realistically be 75% higher. (This is a good reason for individuals to prefer EET - the returns on their retirement investment can realistically be 75% higher - and the government can also take advantage of this excess return). Now it is quite likely that tax on Kiwisaver withdrawals will be a bit lower than the tax a person may have initially paid on earned income , but as NZ has a fairly flat tax schedule   they are unlikely to be so low that the government will earn less tax overall (because of the 75% increase in returns) . 

This can be put differently. Under an EET system, the government taxes less initially, but it has a claim on the whole sum in your Kiwisaver account. This makes you individually better off as you can take advantage of the power of compound interest at a high (pre-tax) rate. It also means the government has a contingent tax asset. This contingent tax asset is not included in the government accounts but it is real. It is also very large in many countries.  If a government does not mind higher debt offset by this contingent asset, it can improve its net financial position just as it could if it defered tax on the NZSF (although it is likely that the NZSF will produce higher average returns than a typical KiwiSaver account). This means it can ultimately reduce taxes or raise spending. Most OECD governments   find themselves in this position - they have higher debt than NZ, but they also have this contingent asset that is not counted against the debt in their official accounts. They seem to be comfortable with the position. The NZ Government got spooked in the late 1980s by the then size of its debt position and by adopting TTE moved to a system that reduced debt at the expense of reducing its long term asset position. i don't believe this was a very clever outcome. It is one of the reasons why the NZ government has less debt than other countries - it collects tax on Kiwisaver contributions early, and thus does not have this contingent asset. Perhaps we can blame the accountants for not recognizing this asset.

There are  many other reasons why EET is a good system. But they are a different set of arguments. I outlined them a bit in my series. One of them is the current TTE system penalises lenders when inflation is positive, by charging much higher taxes on real (inflation-adjusted) interest than on any other investments.  EET is neutral towards inflation. I have no idea why the NZ government thinks it is a good idea to tax the simplest saving product - a loan - at a penalty rate when there is inflation. A second is that EET taxes retirement savings in a manner that is similar to the way we currently tax owner-occupied housing, and so reduces the tax disadvantage of KiwiSaver savings relative to owner-occupied housing. Susan also recognizes this tax wedge between owner occupied housing and other assets, and has come up with a scheme that taxes the implicit income of owner-occupied housing, so the tax rate would be raised to the TTE tax rate on other investments. My preferred scheme is to adopt EET so the tax on retirement savings is reduced to the tax on owner-occupied housing, a principle forwarded by Lord Kaldor and adopted in most countries.  So in some sense we have similar objectives - reduce the tax advantages of owner-occupied housing and thus reduce the extent that house prices are artificially inflated by the tax system - but different methods. Or, at least, that is my interpretation of the difference. Susan may have a different interpretation. 

AC