Time to wind-up the NZ Superannuation Fund

In their print edition last Friday, NBR ran a piece from me suggesting that it was now time to wind-up the New Zealand Superannuation Fund.  For those with NBR subscriptions, it is now available on line.

I didn’t assign my copyright to NBR or anything of the sort, but I won’t reproduce the full column here.  It was, largely, a much shorter version of a post I did here a couple of weeks ago (and in the process of generating it, I proved to myself again that one reason I write long posts is that short posts take much longer).  But this was the final paragraph.

There is a political debate to be had about both NZSF and about the future parameters of New Zealand Superannuation.  But the two simply aren’t very logically connected.   What NZS policy we run in future depends on all manner of things, including the overall state of the government’s finances.  But the performance of one part of the balance sheet simply isn’t a key consideration.    If it makes sense for governments to run speculative investment management operations, it does so whether or not life expectancies are increasing.   But short of Norwegian quantities of oil and gas being discovered here –  which give the government cash that simply has to be invested –  it is just isn’t a business the government should be in.  Fortunately, the Fund hasn’t done badly over its life to date, but let’s bank the gains, and wind-back the considerable future risks by closing the Fund and using the proceeds to repay government debt.   

In both my recent posts on NZSF (here and here) , I’ve tried to emphasise that those who run it probably haven’t done a bad job in the time they’ve been managing our money.  (Here I should note that the NZSF have pointed out, via a response to someone else’s OIA, that in my first post I made a mistake in annualising the rates of return on the New Zealand stock exchange, a number I had mentioned in passing in that post.  NZSF and I agree –  on this if little else –  that the New Zealand stock returns are not a particularly meaningful benchmark for these purposes.)

NZSF has been charged by Parliament with running a high risk fund, they’ve done that, and over a period in which global asset markets have done rather well, they’ve made quite a bit of money for the taxpayer.  But if the relevant benchmark isn’t the NZSE50, neither it is (as NZSF and its chief executive keep claiming) the Reference Portfolio that the Fund Board themselves construct and adopt.  The latter has some uses inside the organisation, but it is largely irrelevant for taxpayers and citizens trying to decide whether NZSF has (a) done a good job, and (b) should exist at all.

In their own official documents, the relevant performance benchmark is

It is our expectation, given our long-term mandate and risk appetite, that we will return at least the Treasury Bill return + 2.7% p.a. over any 20-year moving average period.

That expectation is hardly ever mentioned when, for example, the chief executive takes to the media.  It didn’t appear in the OIA response I linked to earlier either.

As it happens, the returns to date have been (quite a bit) better than that benchmark –  which is why I’m quite open about the fact that NZSF has probably done a reasonable job, perhaps as good as most other managers with their opportunities might have done.   But after only 13 years, their own official documents tell us that we can’t yet know how much of those returns is down to luck, and how much down to skill.   As they note, when you take as much risk as they do –  really big year to year fluctuations –  it takes a long time to tell.   When global markets are rising strongly it isn’t hard to make money.

Over the life of the Fund, their total returns have averaged around 10 per cent per annum.  As I’ve noted previously, over most of that period the Treasury required government agencies looking at investment projects to use a discount rate of 8 per cent real (or around 10 per cent in nominal returns).  Of course, what discount rate one should use depends on how risky the proposed project is, and what it does to the risk profile of the whole of the owner’s business.  A “high octane” (Orr’s words) investment management subsidiary is risky –  it might not be Wellington aiport runway extension risky or INCIS risky, but it will be riskier than, say, a new school in a growing city.   And these sorts of investment management returns tend to add to, rather than reduce, the variance of the Crown’s overall financial position –  in particular, the biggest losses tend to become starkly apparent in periods when there is the greatest pressure on other government finances (ie in severe global economic and financial downturns).  A 10 per cent (pre-tax) return over the period from 2003 to now –  even though it is better than they expected –  just isn’t that impressive.  And NZSF tell us that even they believe those sorts of returns can’t be expected to be replicated in future –  they tell us we should expect 2.7 per cent above the Treasury bill rate.  Who knows what a “neutral” New Zealand Treasury bill rate is, but even if it is getting up to around 4 per cent, that suggests expected future pre-tax returns of less than 7 per cent.   And there is no reason to think that the variances, or covariances, of the portfolio will be less in future than they have been.    (In recognition of the lower interest rate environment, Treasury now encourages government agencies to use a real pre-tax discount of 6 per cent –  or 8 per cent in nominal terms.)

I could add into the mix the point that few private companies will be using hurdle rates of return of less than 10 per cent in deciding on investment projects, or the acquisition of a subsidiary.  And that is what NZSF is –  a fairly aggressive investment management subsidiary of the New Zealand government (with deteriorating governance/transparency –  eg when a large chunk of a government-owned local retail bank is in the portfolio).

And so my bottom line is that we should be thankful for the reasonable returns we’ve had from NZSF to now (through some mix of luck and skill), but that since we can’t count on anything like those sorts of returns in future (even NZSF say so), and even the returns to date are really only sufficient to compensate us for the risk run on our behalf, we’d be better off locking in the gains we’ve had, closing down the Fund,  liquidating the assets over a couple of years, and using the proceeds to repay public debt.    Our government does not need to be in this game –  unlike, say, the governments of Norway or Abu Dhabi, with genuine wealth to manage and smooth –  and the returns to doing so don’t look that attractive.   As the Crown is already heavily exposed (both through the tax system and its other extensive asset holdings) to the ups and downs of the domestic economy and global markets, strategies that reduce risk, rather than increase it, seem intuitively more appealing.  The NZSF logic is the opposite of that.

To accompany my NBR piece, they did an NBR Radio interview.  The one question that got me thinking –  and really the reason for this post, which to some extent is traversing old ground –  was along the lines of “wouldn’t winding up the NZSF leave ordinary people worried about whether their NZS pension would be there for them when they got old?”.

That the question slightly rattled me probably just reflected the fact that I’m a macroeconomist not a political scientist.    Perhaps it is also a testimony to Michael Cullen’s political skill in constructing the NZSF, initially to safeguard surpluses from spending ministers as anything else, and linking it to the NZS regime.      But why shouldn’t people worry?

  • NZSF is only a small part of the government’s overall balance sheet,
  • Our government finances are –  thanks to a succession of Labour and National led governments –  among the stronger in the advanced world,
  • Even the staunchest advocates of NZSF never saw it contributing more than a moderate proportion of future NZS costs,
  • We’ve had a universal government pension since around 1940 –  with brief interruptions in the means-testing years of the 1980s and 1990s – and a government old-age pension for those in need since 1898,
  • Our universal old-age pension is high enough to keep most old people out of poverty (even those who earned low enough incomes through their lifetimes that they’d have had little effective capacity to save), but doesn’t (of itself) deliver high incomes to people beyond basic needs (so it isn’t like some state schemes elsewhere in the OECD),
  • And there is nothing about NZS that is unaffordable in the long run, provided that sensible, well-foreshadowed, adjustments in the age of eligibility are made, in line with continuing improvements in life expectancy.  Ideally, we’d have the age at 67 already, and be increasing (by statutory formula) a little each year as the life expectancy rises.   Simply shifting the age to, say, 67 doesn’t resolve the issue, but a suitable resilient formula-based adjustment would.

In modern times, the greatest period of uncertainty –  by far –  about future state pension support was during the years from around 1985 to 1998 when a means-testing regime was instituted and then constantly revisited.  That had to do with some combination of the overall stretched state of the government’s finances, as well as essentially ideological debates about universality vs targeting.

If one believes –  as I, somewhat reluctantly (but for good second-best reasons) do –  in a modest but universal state age pension, the best protections of that over the long-term, and the best way to provide predictability to individuals (avoiding jerky discontinuous policy adjustments) are

(a) sound overall government finances (low debt levels, and tax and spending shares of GDP that don’t push historical or international limits), and

(b) agreed, or at least broadly accepted across the main parties, statutory provisions that lay down well in advance how the age of eligibility will adjust in future as life expectancy increases.

NZSF really isn’t relevant to either.    It is just a high-risk investment management business, and that isn’t a natural business of government.

And veering slightly further off topic, I’m still puzzled by that Labour-Green commitment I discussed yesterday to keep core Crown spending to around 30 per cent of GDP while doing nothing to alter NZS parameters.  This was the chart from the Treasury’s long-term fiscal statement released late last year.

ltfsThey would be okay for the next few years, but once you get even 10 years out from here, there would need to be a lot of other expenditure cut to keep spending to around 30 per cent of GDP, even if (a) the two parties resume NZSF contributions soon, and (b) investment returns proved to be pretty good.   I’m genuinely puzzled how they propose to square that circle.