Andrew Coleman is one of New Zealand’s smartest economists, one of those people I learn something from almost every time I talk to him, or read something he has written.   Andrew currently divides his time between the University of Otago and the Productivity Commission.  But we disagree, it appears quite starkly, on the place of the New Zealand Superannuation Fund.  I’ve written various posts, mostly quite critical of the Fund for a variety of reasons (some things in NZSF’s own control, others a reflection of the political choices that led to the establishment of the NZSF).    I favour winding up the Fund and using the proceeds to repay debt.

In response to a couple of posts in recent months, Andrew has posted substantive and thoughtful comments that appear to be intended as a defence of the current system, and the place of NZSF in that system.   The first set was here and the second set was posted here on Saturday night.

As I understand it, Andrew and I share a view that there should be a universal public pension scheme, that is not less generous (relative to, say, average wages) than the current system.  Where, I think, there is a difference is that I firmly believe that the age of eligibility for NZS should be increased, and that subsequent further increases in the age of eligibility should be linked to further improvements in life expectancy (there should also be rather tighter residence requirements for eligibility) .    This makes a material difference because under my preferred model, NZS spending does not keep on increasing as a share of GDP, and is a manageable expense/burden for society. By contrast, Andrew often appears to be writing in a context that treats the current eligibility rules as a given, and thus focusing on how best to finance those (political) commitments.

Andrew puts a lot of emphasis on save-as-you-go (SAYGO) funding models, as distinct from pay-as-you-go (PAYGO) models.  A funded defined benefit pension scheme is a classic SAYGO model –  employees and the employer put aside money each week for, say, 40 years, and at retirement there is, in principle, enough to finance the employee’s pension for the rest of his or her life.  The power of compound interest has been harnessed.   In principle, at least in respect of the employer’s contribution, it could have been done another way: the firm could simply have invested the money itself (including reinvesting in its own operations) and then paid its share of the pensions as they fall due.  The reason that isn’t a good model is that (a) pensions of this sort are deferred remuneration and individual employees (reasonably enough) want a secure and certain claim, and (b) firms come and go, management changes, businesses fail etc.   A separate legal entity – a superannuation fund, with a trust deed etc – is the preferred way to go, but not because one approach involves saving and the other doesn’t, but because of agency/governance/enforcement issues.

How does the NZS/NZSF model fit in to this sort of picture?

First, as I noted in some earlier comments to Andrew

Our difference is around the specific place for the NZSF. Personally, I see any connection between it and NZS as just political branding. NZSF is just a set of govt-owned financial assets, and one can’t really put ribbons round particular pots of money.

NZSF does not make any future NZS promises more affordable,  If it manages reasonable returns –  as one might expect over time – it modestly improves the government’s overall financial position, and hence its ability to meet all future spending aspirations at something like current tax rates.

I’d prefer to think of managing the government’s balance sheet and income/expenditure, both now and across time, in an overall way, rather than assigning individual pots of money to individual line items.  That seems likely to be more efficient. It is also more realistic, about the nature of how government finances will end up being managed.  Governments can’t bind themselves to not use one pot of money labelled for purpose X for purpose Y if subsequent pressures change.  Probably, nor should they.  Wars happen, disasters happen, the uncertain happens.

Thus, Andrew argues the economic merits of savings, and I wouldn’t disagree with him particularly.  But in my proposal to wind up the NZSF and use the proceeeds to reduce debt, there is nothing that would reduce either public or private savings.  All that would happen is that the government balance sheet would be less leveraged (less debt, fewer financial assets, no change to the operating balance).  I’m also quite relaxed about the notion that if a government is going to take on far-future financial commitments (like an NZS) scheme, it should probably have a stronger balance sheet (more savings, less net debt) than a government that did none of those sort of things.   A balance sheet with near-zero net debt –  when, as Andrew notes, the government is very long-lived –  and extensive real asset holdings, in a country with above-average population growth, looks pretty cautious to me. Excluding a handful of countries with non-renewable natural resource extraction proceeds (Norway, Abu Dhabi etc), our government finances are among the most conservatively managed in the world.

So the issue isn’t one of whether the government should save or not, but simply of how much it should save.  I’m not sure of the answer to that question, but there are both political and economic dimensions to any answer.

Among those economic questions is whether, and if so to what extent, additional government savings (or even compelled private savings) actually raises national savings.   If there was full offset (every dollar of additional public savings was offset by an equivalent reduction in private savings) there would be no obvious societal benefit at all (in fact, given the deadweight costs of taxation – and intermediation costs – there would net welfare costs to society).    I see no evidence that, for example, the Australian compulsory savings system has raised national savings rates in Australia.  As for government savings itself, there seems to be plenty of sign of at least some private offset.

Among the political, or political economy, questions are ones about the durability of large tax-funded holdings of government financial assets in a democratic society.  It is one thing for governments to hold large asset pools in societies with little or no democratic accountability (Singapore, Abu Dhabi and so on) or even when the assets arise from a non-renewable natural resource (as in Norway).    It is another matter altogether when the assets are tax-funded, and governments face voters every few years, in a country no longer particularly well-off by advanced country standards.  Such accumulations of assets invite electoral auctions. They also invite political jockeying to see that the assets are used in line with the priorities and preferences of those currently in power (or, indeed, of those who happen to be managing the money).

There also arguments advanced that it would be natural for any portfolio to have some significant equity exposure to (for example) secure some of the equity risk premium for the Crown.  Against some abstract benchmark in which the government was otherwise funded by lump sum taxes on the one hand, and simply paid NZS on the other, I would agree.  But that isn’t what government finances (here or abroad) look like.  Through the income tax system, the government already has an effective equity stake in every business enterprise in the country (28 per cent of all profits go to the Crown, 28 per cent of losses can usually be written off against future earnings).  And the Crown has an extensive base of real assets (equity exposures) –  whether shares in SOEs or the extensive holdings of schools, hospital, roads etc (which don’t produce a dividend stream, but save the Crown paying user fees which would include someone else’s dividend stream).

Perhaps there is a case for more Crown equity exposures, but that case really needs to be made convincingly against the backdrop of the overall public finances, not just thought of relative to future expected NZS payments.   It should also be thought about in the context of citizens’ own “risk budgets”: increased equity exposures taken on by Crown agencies should, rationally, be offset at least in part by reduced private holdings.

In his writings in this area, Andrew Coleman puts quite a lot of emphasis on government (and, by extension, NZSF) as a long-lived agent, better placed to invest in long-term assets than the private sector, and less prone to liquidity pressures.   I think he is mostly wrong about that, for a variety of reasons.  From an anecdotal perspective, NZSF seems to have had more asset allocation changes in the last decade or so than the modest superannuation scheme I’m a trustee of.    But, and much more importantly, the government (at least in a democracy) doesn’t stand remote from its citizens and taxpayers, and taxpayers/voters don’t like large losses, and (I’d argue) especially not when their personal finances are already under greater than usual stress.   NZSF will record large losses in the next serious global recession –  the more so, as NZSF hedges back to NZD –  and that recession is also likely to put stress on the New Zealand government’s operating balances.     There is likely to be heightened pressure on the government, and on those managing the Fund, to account for their losses, and perhaps to cut those losses.  It might be silly, wrong, or in some longer-term sense irrational, but no investment strategy should ever be operated without considering the extreme loss tolerances of the ultimate investor (in this case, not some detached Treasury official, but voters).   I’m sceptical that the public is comfortable with the potential for tens of billions of annual mark to market losses (the scale we could be looking at if NZSF gets much bigger), coming at a time when (say) taxes are being raised or public spending is being cut.  In other words, even if an NZSF strategy offered possible longer-term benefits, it would do so only at the cost of concentrating periods of pain.

Another aspect of Andrew’s argument is an assumption (implicit, and sometimes explicit) that governments can be trusted, and will typically be good economic stewards.  It is far from clear why we should expect them to be so, especially when entrusted with other people’s money.  Each citizen has a strong interest in their own future financial position, and (one hopes) that of their children and grandchildren.   As individuals, politicians no doubt have the same interest.  But let lose on a whole country, politicians have interests which are rather different –  often as focused on the next election as anything longer –  and with little accountability (beyond losing office) if things go wrong.  These same governments that Andrew wants us to entrust more of our money to are the same sorts of governments that did Think Big, that turned our economy inwards for decades from the 1930s, that take us into wars, that ran us into serious debt problems (whether in 1939 or 1990), and so on.  They are same group who cavalierly talk of pursuing net zero carbon targets, even if the consequence is that (on their own numbers) GDP is cut by 10 to 22 per cent, with the costs falling disproportionately on the poor).   I’m not some anarchist who wants to get rid of all government, but I don’t think the track record is particularly good, especially when governments want to commit our money/resources for the long-term.

And all this is before we look at the specifics of the way NZSF has actually been run:

• overselling its investment returns in a rising market, while quietly noting that it takes 20 years of data to seriously evaluate their sort of risky strategies (which may do no real direct harm, but speaks to integrity),
• used (together with ACC) to solve the previous government’s Kiwibank capital issues, in ways that inject no additional expertise to Kiwibank, while corroding effective accountability for this risky government-owned assets (no doubt at favourable pricing for NZSF, but at cost to the system),
• the decision to reduce carbon exposures, purportedly as a normal risk-return call on business prospects, but nonetheless implemented in a way where the consequences can’t be monitored, and thus looked more like virtue-signalling and playing politics than a serious neutral investment stance,
• the opportunistic bid to own part of all of the new light rail proposals. NZSF has little or no apparent experience in such assets, which themselves appear uneconomic, and thus the approach again smacks of politics and lobbying, more than pursuit of citizens’ longer-term interest.
• the latest attempt to lobby for huge tax concessions (adding new distortions to the system) for projects (and project partners) they want to get involved in.

These problems will only get more serious if the Fund is allowed to grow larger. One experience which shook my confidence was involvement a decade ago in the then-government’s Jobs Summit, which occurred at the trough of the last recession.  The NZSF fund was small then, and the pressures were resisted, but it was likes bees round a honey pot as people (well-motivated and not) emerged with ideas of how the moneypot could be used to help.  Those pressures will return next time.

And all that is before the ethical investment question.    We all have exposures to all industries (legal, moral or not) through the tax system, but NZSF involves active choices to put our money in individual companies. You might not be comfortable with whaling companies, tobacco companies, arms companies, or even financial institutions like AMP. I’m not comfortable with exposures to hospital chains that do abortions, or conglomerates that produce pornography, and I’m not keen on funding McDonalds either. My point isn’t that my preferences are better than yours or vice versa, but investment is participation, it is support, and those investment choices are neither a natural nor necessary part of a New Zealand government.  (Neither is a large leveraged investment fund.)

In many respects, the governance provisions of the NZSF aren’t badly set up, if one is going to have a body of this sort.  But rules of that sort can only take one so far.  All Board members will have their own futures in mind –  and governments have lots of apppointments in their gift.  The same goes for the CEO.    And, of course, so many people now have business dealings with NZSF, including competing for investment mandates, that it is hard to ensure that ongoing robust scrutiny an asset of that size deserves.

As I’ve noted previously, one way to reduce some of the risks around NZSF would be to amend the legislation to prohibit the NZSF dealing with New Zealand or local governments (to buy or sell assets, or to invest in proposals floated by government agencies) –  or perhaps even just to restrict exposures to, say, 5 per cent of any project/deal.    It would restrict NZSF’s opportunities, but it would also restrict the scope for logrolling, sweetheart deals, and all the sort of stuff that simply shouldn’t happen in the idealised world some supporters envisage for NZSF.

Finally, in his most recent comments, Andrew posed this point

So here’s a question, in the interest of debate: Do you have similar issues with the ACC fund? And if not, what is different about the ACC fund that makes it better than the NZSF fund?

Actually, a few months ago I noted that I thought it was worth putting ACC onto a PAYGO basis –  and to the extent there are very long-term commitments on the Crown balance sheet, that should influence the overall structure of the Crown finances, including the extent to which the Crown saves (rather than have an individual ACC moneypot).  As it happens, the ACC investment performance has been better than that of NZSF.   But my views on ACC are influenced more by my long-term doubts about the merits, or the fairness, of treating all accident victims differently from those with very long-term illnesses or disabilities.

In conclusion, I think there two quite separate issues to evaluate, and we don’t help either conversation by conflating them (as the previous Labour government attempted to when it set up NZSF).  There is the question of what an appropriate NZS policy should be.  But then there are the, largely separable questions of:

• what the appropriate overall shape of the government balance sheet, and income statement should look like, and
• what, if any, role a standalone leveraged global investment fund has to play in such a balance sheet.

Answering either question needs to range widely, and consider likely private sector responses to public sector choices, governance constraints, and the long track record of ambitious government interventions here and abroad.