Although I am currently chairman of the 2025 Taskforce, I am making this submission in my personal capacity. I have long had a strong interest in economic policy in New Zealand – I was a member of the Monetary and Economic Council in the early seventies, a member of the New Zealand Planning Council (and chairman of the Council’s Economic Monitoring Group) in the late seventies, the chairman of four consultative committees on taxation in the eighties (including both the committee which designed the GST and the committee which considered the taxation of superannuation and life insurance), and the Governor of the Reserve Bank from 1988 to 2002.
I acknowledge the ongoing debate about whether New Zealand has a “savings problem”, and accept that our growth performance would be substantially enhanced if the policies which influence the business environment were improved, even if the national savings rate was unchanged. Nevertheless, I find it hard to avoid the conclusion that our dangerously high net international indebtedness position, our high real interest rates, and our economic growth would all be improved if our level of national savings was higher.
At this late stage in the work of the Working Group, I do not propose to make a general submission on your work, but rather to confine my observations to four issues:
My submission does not deal with the relevance of the current parameters of New Zealand Superannuation because this issue is beyond your terms of reference, though clearly those parameters are relevant to the issue you have been asked to address. My own strongly held view is that these parameters should be changed, at least insofar as the age of eligibility is concerned, and that the longer the announcement of that change is deferred the more serious the fiscal implications of that are going to be – or the more quickly any change would have to be implemented, making it difficult for older New Zealanders to adjust.
Would there be merit in introducing a compulsory retirement savings scheme?
Predictably, there is strong support from the funds management industry for New Zealand’s introducing a compulsory superannuation scheme, perhaps by making KiwiSaver compulsory. That would certainly be a great advantage to fund managers, and to their professional advisers. However, there is little evidence that this would increase national savings; it might do little or nothing to reduce poverty among the elderly; and it would be a substantial infringement on people’s freedom to choose when to save for their retirement, how best to do that, and indeed how much to save.
It may seem strange to claim that a compulsory savings scheme would do little to increase national savings. Singapore has a compulsory savings scheme, and an extremely high national savings rate. But Hong Kong also has an extremely high national savings rate, and until recently has had no compulsory savings scheme. The very high savings rates in both countries – and of course in many other Asian countries, including China – may be a cultural characteristic of Asian societies or simply a reflection of the virtually complete absence of any form of social welfare safety net. Of course, it isn’t only Asian countries which have a high savings rate: several European countries do also. Closer to home, Australia has a compulsory retirement savings scheme and, being culturally more closely similar to New Zealand than either Singapore or Hong Kong are, is a more relevant example. Available evidence suggests that there has been some increase in private sector savings in Australia as a result of that country’s compulsory savings scheme, partially offset by the fiscal cost of the various tax incentives provided to make such saving more attractive, with the result that the increase in overall national savings appears to have been very small – in the order of perhaps 1 to 2% of GDP. And the Australian scheme has been subject to constant changes in regulations, tax treatment, and so on, providing little certainty to anyone. It is instructive that the recent Australian Budget, which proposed an increase in the required employer contributions to the compulsory retirement savings scheme from 9% of wages and salaries to 12%, estimated that the net effect on Australian savings would be just 0.4% of GDP, and that not until 2035.
Nor is it clear, on the basis of Australian experience, that a compulsory retirement savings scheme would reduce the incidence of poverty among the elderly. At the moment, OECD data suggest that poverty among those over 65 years of age is lower in New Zealand than in any other OECD country. New Zealand data also suggest that poverty among those over 65 years of age is lower than among those of any other age group. And that is not surprising: New Zealand Superannuation is more generous, relative to average incomes, than the basic wholly-taxpayer-funded pension in almost any other OECD country and has almost universal coverage, subject only to not very demanding residency requirements. It is clear that there is significantly more poverty among the elderly in Australia than in New Zealand[1], and that appears to be because, while a compulsory retirement savings scheme provides good income in retirement to higher income people and people who spend their entire adult lives in the workforce, it provides rather poorly for lower income people and those who spend a significant fraction of their adult lives out of the paid workforce, particularly women. Since it seems clear that the structure of New Zealand Superannuation can be sustained indefinitely, subject only to the need to gradually increase the age of eligibility for it to reflect ongoing increases in life expectancy, it is not at all clear why we should consider changing to a compulsory retirement savings scheme which would take decades to come to full fruition.
And to cap it off, a compulsory savings scheme would represent a significant infringement on the right of people to decide when, how, and how much, to save for their retirement. Those who advocate a compulsory savings scheme are often self-interested participants in industries which would directly benefit from compelling citizens to use their services. And the other advocates of compulsion are convinced that they know better than their fellow citizens how much they should be saving. But the evidence is that, on average, and once New Zealand Superannuation is taken into account, most New Zealanders have made adequate provision for income in their retirement. That makes it even less appropriate that the state should be telling people when, how, and how much to save. For most young adults, possibly with children and a mortgage, by far the most sensible form of saving involves paying down the mortgage and/or investing in their children’s education. No fund manager can promise to earn a consistently higher rate of return, without risk, than is implied by the after-tax rate of interest on the average mortgage, so that paying down the mortgage makes perfect sense for a young adult. The Working Group’s Terms of Reference mention New Zealand’s economic vulnerabilities: one of those vulnerabilities arises from the very high level of household debt. Compelling families to save, in vehicles they could not get access to until retirement, would seem particularly ill-advised in a climate with a large overhang of debt and fairly subdued prospects for income growth over the next few years.
Would it be desirable to change the tax system to encourage more savings?
The Consultative Committee on the Taxation of Superannuation and Life Insurance, which I chaired in the mid-eighties, recommended that locked-in retirement savings should be taxed on an Exempt-Taxed-Taxed basis – in other words, that contributions into such schemes should be from pre-tax incomes but that the income in the savings vehicles and payments from them in due course should be taxed as ordinary income. The argument was that, while this was not a concessional tax treatment as compared with the alternative of Taxed-Taxed-Exempt on the assumption that income tax rates remain constant over time, it would feel like concessional treatment to savers, and would not require savers to assume that a future government would exempt from tax a flow of cash from a retirement savings vehicle which might look like income. Whether for fear of the initial loss of tax revenue or from a desire to tax all income from savings on a consistent basis, the government of the day rejected this recommendation.
There have been various attempts to modify the tax system to encourage saving over the decades since the mid-eighties, with the most recent being the introduction of the PIE regime.
Taken in isolation, the introduction of the PIE regime was, in my judgement, a step in the right direction. But it was a timid step in the light of the latest thinking on the taxation of capital income. The economics literature is now increasingly clear that, to the extent that the focus in tax design is on overall economic performance, the taxation of capital income should be minimized. A recent paper by Professor Greg Mankiw, a professor at Harvard and former chairman of the US Council of Economic Advisers, summarised the point as follows:
“The logic for low capital taxes is powerful: the supply of capital is highly elastic, capital taxes yield large distortions to intertemporal consumption plans and discourage saving, and capital accumulation is central to the aggregate output of the economy.”[2]
Indeed, the literature increasingly suggests that the optimal tax rate on capital income, particularly in a world of free capital movements, is much lower than the optimal tax rate on labour income, perhaps even zero. I understand that, on his recent visit to New Zealand, the new Nobel laureate, Peter Diamond, recently nominated by President Obama to be a member of the Board of Governors of the Federal Reserve Board and a man who has done extensive work in the area of savings, indicated his support for taxing capital income less heavily than labour income. By contrast, in New Zealand taxation on capital income is unusually large as a share of GDP – on some estimates, the second highest in the OECD. It’s hard to avoid the conclusion that there may be a connection between our relatively low level of capital per worker and this relatively heavy taxation on capital income.
There are two arguments usually adduced to reject a radically lower rate of tax on capital income (often referred to as the Nordic system). The first comes from the Inland Revenue Department, which is understandably concerned to minimise the compliance and enforcement complications of the tax system. These complications certainly exist whenever there are differences between the tax rates applied to different sources of income, and of course already arise from the need to determine the division of income between company income and personal income where the owners of a closely-held business are also the main employees of that business – as is the case now, with a 28% company tax rate and a 33% top personal income tax rate. These challenges have been handled in the countries that have adopted the Nordic system. Doing so might require more resources for the IRD, and those costs would have to be weighed against the benefits of a tax system which better reflected the differential elasticities of supply in respect of labour and capital.
The second objection is a political one: any move to radically reduce the tax on capital income – whether to something like the 12.5% rate the 2025 Taskforce suggested might be possible in time[3], or to something lower still – runs up against an understandable reluctance to advantage the owners of capital at the expense of wage and salary earners. But this is an illusion: the evidence internationally is clear that the overwhelming bulk of the tax on capital income is not, in economic incidence terms, finally paid by the owners of capital but by wage and salary earners – most especially by the least mobile wage and salary earners. In other words, because a lower tax on capital income would lead to more investment, resulting in more capital per worker, it would also result in higher wages and salaries. By contrast, a higher tax on capital income leads to less investment, less capital per worker, and lower wages and salaries. In a similar vein, it is sometimes argued that this model may be good for some countries, but that labour is so mobile in New Zealand as to render the difference between capital and labour moot. But since wage differentials, even between New Zealand and Australia, are both very large and persistent (perhaps 35%), while the typical gap between interest rates in the two countries is around 1 percentage point, it is clear that the data back the commonsense intuition that capital is considerably more mobile than labour.
Accordingly, I believe the Working Group should give serious consideration to recommending a radically lower tax on capital income, as applies in Nordic countries.
Is the New Zealand Superannuation Fund part of the solution?
In a modest way, Michael Cullen may have tried to do something to increase (or at least maintain) our national savings rate by setting up the New Zealand Superannuation Fund. When the Cullen Fund was first set up, the government was running substantial surpluses and had been since 1994. The government’s debt was falling fairly rapidly, and the Minister of Finance may have been under pressure from within his own caucus to increase government spending rather than to use the large surpluses to further reduce debt. So I suspect he hit upon the idea of setting up the NZSF in which to store away some of the surpluses in order to preserve those government surpluses and mitigate the future fiscal cost of New Zealand Superannuation.
Of course, while this may have made political sense – pointing to a gradually filling piggy bank is easier than pointing to a gradually diminishing debt, and is doubly attractive when that piggy bank can be erroneously described as in some sense guaranteeing New Zealand Superannuation for decades ahead – this made no economic sense at all. No prudent household with a mortgage would use temporary cash surpluses to buy shares in New York or petrol stations in New Zealand.
Michael Littlewood has noted that at the end of June last year, the government’s total debt (including that of the SOEs) was some $62 billion, so that the $16 billion of assets in the Cullen Fund was in effect all funded by borrowed money.
“If the Guardians (of the Fund) do not achieve a return of at least the cost of the Government’s most expensive debt each year, the NZ Superannuation Fund has lost taxpayers real money. Over the six years to last June 30 (the last period for which audited accounts are available), the fund’s Guardians have missed that most basic target by a significant amount. The accumulated loss was about $2.6 billion at June 30 (2009). This is on the basis that the contributions to the fund could have been allocated instead to repay the longest-dated (10-year) government debt instead of investing it through the fund. The recovery in investment markets to last November reduced that estimated loss to ‘only’ $1.4 billion.”[1]
And of course, as Michael Littlewood noted, the Fund should have earned a margin above the cost of government debt, to reflect the greater risk of investing in equity markets.[5] Even against the performance criteria set for the Fund itself, there is no evidence that it has yet managed to add any positive investment return, once allowance is made for risk (exemplified by the extreme volatility of returns on the Fund assets).
When politicians call for the Government to continue contributing to the Fund even today, when the government is borrowing a net $12 billion a year, they simply demonstrate that they either don’t understand basic fiscal arithmetic or have supreme confidence in their ability to dupe the public. Why anybody would recommend increasing our borrowing still further in order to invest in (mainly) foreign assets absolutely escapes me – especially at a time of widespread concern about the level of sovereign debt. And increasing government borrowing to buy additional New Zealand assets – as some advocate – has other problems, including the very large holding of New Zealand business assets which central and local government already own. Even when the government gets back into surplus, the case for contributing to the Superannuation Fund while the government still has a large stock of debt (and at a time when age-related expenditure will already be rising strongly, with the first of the baby boomers claiming New Zealand Superannuation next year) seems weak at best.
One of the big risks of the Superannuation Fund is that political pressures will grow to invest the funds in projects which, while politically popular, simply don’t stack up on the basis of any kind of careful cost/benefit analysis – Transmission Gully, a tunnel for trains under Auckland, and a bunch of low-yielding dairy farms have all been suggested by politicians as good projects for the Fund. If we want to reduce the country’s growth potential, it’s obviously vital that we keep squandering scarce capital on low return projects!
As the 2025 Taskforce urged in its first report last year, the New Zealand Superannuation Fund should be wound up without delay, and the proceeds used to repay government debt.[6] It certainly isn’t, in any way, a solution to increasing our national savings rate. Even if the original intention when the Superannuation Fund was set up was to restrain growth in government spending, in the end an explosion in government spending occurred anyway.
How important is reducing the fiscal deficit in improving national savings?
The most important thing the Government can do to increase our national savings rate at the moment is to get its own deficit under control. At the present time, the primary fiscal deficit – the deficit excluding financing costs – is at its highest level in at least four decades. The Government has announced an intention to eliminate the deficit between now and 2016, but at this stage has not indicated how it intends to do that. There is a widespread view in the community, which the Government has done little to change, that the main reason for the deficit is the temporary reduction in tax revenue and increase in expenditure flowing directly from the recent recession, and that as a consequence the deficit will go away of its own accord as the economy picks up momentum. Unfortunately, that is not the case. Most of the deficit is structural in nature, which means that it will not simply disappear as the economy grows more strongly. Eliminating the deficit will take specific decisions, either to reduce government spending or to increase tax rates.
At the moment, the operating deficit in “core Crown” terms is about 5% of GDP. Eliminating that deficit would not, of course, increase national savings by 5% of GDP because to some extent the elimination of the deficit (reduction in dis-saving by government) would be offset by less saving by households and firms (partly because of the likelihood that eliminating the deficit would involve measures that directly reduced household disposable incomes). But it is likely that the reduction in private sector saving would be a good deal less than the reduction in the dis-saving of the government, so that the faster elimination of the government’s deficit would certainly contribute materially to the overall national saving rate.
Moreover, to the extent that a rapid reduction in the government deficit made possible (indeed, necessitated if inflation were to be kept above the bottom of the target range) an easing of monetary policy, it is likely that it would facilitate a fall in the exchange rate, meaning that while there would be some reduction in overall private sector savings as a consequence of the reduction in the dis-saving of the public sector, there would be an increase in the profitability (that is, in the saving) of the tradable sector, with beneficial impact on the current account deficit. Given our very high level of net international indebtedness, such a reduction in the exchange rate, and resultant encouragement of the tradable sector, would be a highly desirable development.
So to summarise:
A) Government should not introduce a compulsory saving scheme. If modeled on the Australian scheme, it would do little to lift national savings, would do little or nothing to reduce poverty among the elderly, and would be an unwarranted interference with the freedom of people to decide when, how, and how much to save for themselves.
B) As fiscal circumstances permit, Government should give serious consideration to the introduction of a radically lower tax on capital income, as applies in Nordic countries.
C) Government should wind up the New Zealand Superannuation Fund without delay, using the proceeds to retire government debt. The Fund does nothing to increase national saving, and may in fact reduce national saving by creating the erroneous impression that the Fund in some way “guarantees” the current parameters of the New Zealand Superannuation scheme. It makes absolutely no sense for the government to be borrowing to invest in shares in New York or petrol stations in New Zealand.
D) The most constructive thing the Government can do to improve New Zealand’s national saving rate is to quickly eliminate its own fiscal deficit. This would have the additional benefit of facilitating a fall in the exchange rate, to the benefit of the rebalancing of the economy towards the export sector which is so urgently required.
[1] Only 2% of New Zealand’s population over the age of 65 is classed as living in relative poverty by the OECD (the lowest fraction in the OECD), whereas the equivalent figure in Australia is 26% (the fourth highest fraction in the OECD). Pensions at a Glance 2009, OECD, page 64.
[2] “Optimal Taxation in Theory and Practice”, by Gregory Mankiw, Matthew Weinzerl and Danny Yagan, mimeo, 2009.
[3] Answering the $64,000 Question: Closing the income gap with Australia by 2025, November 2009, pages 100-103.
[4] New Zealand Herald, 10 May 2010. For a fuller discussion, see “Pre-funding a government’s future financial obligations – the New Zealand Superannuation case study”, in New Zealand Economic Papers, April 2010, or “Best Practice in Smoothing the Tax Burden: the New Zealand Experience”, a paper presented to an international seminar on social security in Malaysia on 13 July 2010. In that paper, Mr Littlewood estimates that the government’s loss on the Cullen Fund to 31 May 2010 was $1.14 billion.
[5] The Business Herald of 18 June 2010 reported that to the end of May 2010 the Fund had earned just 5.81% per annum since inception on 30 September 2003, falling a good way short of providing adequate compensation for the riskiness of the portfolio.
[6] Op. cit., pages 110-111.
Copyright © 2024 Don Brash.