Defenders of the NZSF

After the flurry of coverage a couple of weeks ago over the remuneration of Adrian Orr, chief executive of the New Zealand Superannuation Fund, debate seems to have turned towards the more substantive issues around the role of the Fund.   The chief executive has been out, in his usual feisty and rather opportunistic style, defending the Fund, advocating for it to be given more money to invest and so on.   One could reasonably question whether the latter in particular is the role of a public servant.  His job, surely, is to invest the money the government has chosen, under the terms of the legislation, to place with NZSF.    Wisely, in my view, no more money has been placed with Fund since 2009 (although even then I thought it was shame the government didn’t simply wind up the Fund).  But this is now an election campaign issue, with the Labour Party vocally calling for an immediate restart of contributions.   That is perhaps understandable from a party which now, once again, favours keeping the NZS age at 65 indefinitely –  which was pretty much their position back when the Fund was first set up.   But it is a debate senior public servants shouldn’t be participating in in public.  Whatever one’s view of the NZSF itself, there are simply fewer grounds for it if one thinks the age of NZS eligibility should be increased over time, as life expectancy improves.

What has also interested me is the vocal support Orr has had from various journalists. In the Orr interview I listened to, on Radio New Zealand’s Nine to Noon on Monday, Kathryn Ryan seemed to see her role as being to help Orr get “the truth” across, cheering him on as he went, rather than asking any searching or challenging questions.  And this morning, in the Dominion-Post, Vernon Small is channelling the Orr lines, but in even more strident tones.  He concludes that we’d be “barking mad” not to be putting more money into the NZSF each and every year.

Well, perhaps. But Orr in particular was guilty of downplaying quite a lot of important considerations.

First, for all the breathless excitement about the NZSF’s investment returns (around 10 per cent per annum, pre-tax, over the life of the Fund), there has been no hint from chief executive, or his media supporters, of the rather more disciplined approach official NZSF documents, presumably adopted by the Board, take:

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.

By design, it is a highly risky Fund (“high octane” was Orr’s description) and performance can only seriously evaluated over quite long periods of time –  20 year periods in the organisation’s own telling.  But Orr (or Ryan or Small) didn’t make that point.

Kathryn Ryan’s breathless praise of the investment performance included, a couple of times, “even over the global financial crisis”.     That period was certainly pretty dramatic, but for equity markets it just doesn’t look that unusual in the longer sweep of history.  Here is a chart of the S&P500, in real terms and on a log scale.

sp-500-historical-chart-data-2017-03-15-macrotrends (1)

The fall in equity prices over 2008/09 looks like the sort of fall one might expect every 15 or 20 years –  and that one was shorter-lived than most.  If you take lots of risk in equity and bond markets, the last 15 years haven’t been a hard time to make money.  As far as I can see, the NZSF has more or less been compensated for those risks (it has forced us citizens to bear) but no more than that.  (I was however amused by the shameless attempt of the Fund’s head of asset allocation, in response to a recent OIA request, to suggest that returns from January 2009 –  near the trough of global markets –  was a meaningful number to use in evaluating the Fund’s performance.)

So the problem with the Fund isn’t that its investment management choices (both those they would loosely classify as “passive” and those equally loosely classified as “active” –  the distinction is pretty arbitrary) have been particularly bad, or good.  They’ve probably been about what one might reasonably expect over that period.  And if we closed the Fund now, or shifted all the money back to low-risk assets, we could crystallise the gains and be thankful that, despite all the risk run, we made money rather than lost money.     But nothing in the Fund’s investment strategy will protect taxpayers if, and when, markets turn bad again.

Orr continues to misrepresents the NZSF as a “sovereign wealth fund”.  It simply isn’t.  We aren’t Norway or Abu Dhabi, managing for an intergenerational perspective, oil wealth that has been turned into cash.   All the money put into the NZSF has either been raised from taxes or borrowed.     There isn’t a pool of money that naturally needs investing.  Rather, the government has established a high-risk investment management subsidiary to punt on world markets.     That simply isn’t –  and never has been –  a natural business of government.

The Fund itself doesn’t have to worry where its money comes from.  But citizens do.   Who knows what governments would have done with the money if the NZSF had never been established.  I suspect much of it would have been wasted on increased government spending.  But it would have been possible to have cut the most distortionary taxes –  those on business income –  quite heavily, which would probably have given risen to a lot more business investment in New Zealand.  None of the analyses of the returns of the Fund ever seem to take into account, for example, the deadweight costs of taxes.  On mightn’t expect NZSF to do so –  they are just investment managers –  but if they don’t they aren’t really in a legitimate position to be calling for more public money to be steered their way.

Orr cites a number of other advantages for the Fund.   He argues that it has contributed to developing New Zealand capital markets.  I’d be interested to see the evidence for that claim.  Most of the Fund’s assets are invested abroad, by intentional design.   What would New Zealanders have done with money if they’d had it as individuals?   And I don’t quite see how sweetheart inside deals, whereby ACC and NZSF  – neither with any particular expertise in retail banking –  take chunks of Kiwibank from NZ Post, enhance New Zealand capital markets.  No doubt having a hulking behemoth (by New Zealand standards) like NZSF generates more activity –  NZDMO got to issue more bonds than otherwise, and then NZSF buys more (mostly overseas) assets –  but are the capital markets really better –  and by what standards – for the presence of the Fund?

I also heard him argue that the NZSF somehow reduces risk and improves certainty. I wasn’t quite sure what he was referring to, but it seemed to be about the future of NZS itself.    But again, he is really talking beyond his pay grade.  The future of NZS is a political issue that really has little or nothing to do with the NZSF size/performance.   It isn’t like a contractural funded defined benefit pension scheme.  Presumably the overall state of government finances, overall tax burdens, and a community consensus on what is fair and reasonable are more likely to shape the future of NZS than the presence (and investment returns) of NZSF.    And in thinking about the overall government balance sheet –  something Orr doesn’t seem to, and isn’t paid to, think much about –  NZSF is currently only about 10 per cent of total assets.

He is on similarly shaky ground when he talks about save as you go approaches beating out pay as you go approaches.  I’m sure we can all agree that saving for the future typically makes sense –  the power of compound interest and all that  –  but that insight doesn’t help at all in deciding what, if any, role NZSF should have.   After all, we could wind NZSF up today, use the proceeds to repay government debt, and nothing would change about accumulated public sector savings.  Higher public sector savings is mostly a choice between taxing more and spending less.   As it is, I’d probably be happier if overall government net debt (including NZSF assets) was quite a bit lower than it is (ie build up government savings a bit more), but at least until all the gross government debt is repaid the government simply doesn’t have to be in the financial investment management business –  it is a pure discretionary choice.    We haven’t been there so far in the 14 year life of the Fund, and it doesn’t look likely that we will be in the next few decades either.  And Orr gives no weight at all to the failures of government, which often see additional Crown revenues wasted rather than saved.

Orr, and his defenders, have also been keen to scoff at any analogies with how a household might approach decisions.  I heard him say something along the lines of “if governments could act like households, we wouldn’t need governments”, which is true, but irrelevant in this context.  One role governments play, on behalf of households collectively, is to absorb collectively some of residual risks that individuals aren’t well-placed to handle.  That might tell you that often governments can’t, or won’t or shouldn’t cut spending in severe downturns, because some of their obligations increase then (and they are willing to let automatic stabilisers work).  For that reason, governments should be wary of revenue sources, or investment returns, that are very highly positively correlated with the economic cycle.  For example, one reason to be wary of capital gains taxes is that they tend to flatter government finances in good times, only for the revenue to dry up just when governments need it most (see Ireland last decade for a classic case study).    The same might well be said of highly risky asset portfolios –  even if they do quite well over the very long haul, they will look particularly poorly at just the times when government finances are under most pressure for other reasons.  In fact, if the pressures get serious enough, governments might come under pressure to liquidate those risky holdings right at the bottom of the cycle.  Those aren’t issues Orr has worry about –  he is simply paid to maximise returns on his little chunk of government resources, subject to acceptable risk –  but citizens, and people worry about overall government finances through time should do.

After all, it is not as if governments don’t already have other income and investment returns that are quite tied to the economic cycle.  Even on the investment front, for good or ill the government has quite large commercial holdings (those SOE stakes), and on the revenue front the tax system effectively makes the government an equity stakeholder in every business in New Zealand.

Orr and his defenders also scoff at household comparisons because, so they note, the government can borrow more cheaply than households.  More flamboyantly, here is Vernon Small’s take

As a comparison it may be politically effective but it is about as useful as a chocolate teapot.

Show me the household that can tax, has a central bank to set interest rates and biff around the exchange rate paid at the corner dairy, can borrow more cheaply than any business at rates below any mortgage offered by banks – and can live on for decades past the final days of its family members – and I’ll show you the household that has much to learn from a central government or vice versa.

Actually, the typical government (as distinct from the idealised one no one has ever seen) has operated with a horizon considerably shorter than that of most households.  And that is understandable:  I care about my kids and potential future grandchildren, who will still be there in decades to come.  Politicians –  who run governments –  face elections quite frequently, and in the course of a single lifetime successions of them run policy all over the show.

And quite how do people think that governments borrow so cheaply?  Because they have the power to tax, and that power is mostly exercised not over random stateless aliens, but over New Zealand households.  Every debt the New Zealand government takes on involves risks for New Zealand households – risks that at times of stress, governments will disrupt household and business plans by unexpectedly making a grab for a larger share of our incomes/wealth.  That risk limits the other risks households can afford to take, and is why I keep stressing that an accountable government can’t think of the cost of funds as simply the government bond rate; it has to price the implicit equity, bearing in mind that the coercion involved in the power to tax is more costly and distortionary than (say) a large company having to issue new debt if times get tough.    People like to say that governments can’t (usually) go bust, and so are subject to fewer “bankruptcy constraints” in thinking about undertaking possible long-term activities –  but that is typically true only to the extent that they ignore the perspectives and finances or their citizens, who ultimately bear the risks.  Ignoring citizens isn’t what governments are supposed to do.

My bottom line remains that NZSF hasn’t done badly what it has been asked to do (if you want a high risk fund that is).  Equally, it hasn’t really been put to the test.  They probably made some quite good calls at times, but the risks they assume for the taxpayer are very considerable.  In that OIA response I referred to earlier, they attempt to rebut some of my arguments, by suggesting that the appropriate hurdle rate of return should depend on the riskiness of the project.  I read that and thought: “yes, and that is really to concede my point”.    Over the life of the Fund, the standard deviation of annual returns has been almost 13 per cent.  Those are really large fluctuations –  by design –  and in considering establishing (or retaining) a government leveraged investment fund –  effectively a business subsidiary of the government – taxpayers need a lot of compensation for that risk.    Especially as that risk –  in the extremes, which are what matter –  is pretty correlated with other risks to government finances directly, and those of household sector finances indirectly, so there isn’t much –  if any –  overall risk reduction taking place.   When typical Australian companies uses hurdle rates in excess of 10 per cent, we shouldn’t be that comfortable  in our government running such a risky investment management operation for returns that, over a good 14 years, have only just matched 10 per cent.  I’m not suggesting anyone could have done much better than NZSF managers have, just that it wasn’t worth doing at all, evaluated by the sort of standards firms and households apply to their own finances.     And all that on a Fund that at present is only around 13 per cent of GDP.   The risk dimensions of the Fund become even more important if contributions are resumed and we envisage a Fund that could become a much larger component in the overall Crown balance sheet.

There is a political debate that should be had about NZSF.  There is a debate to be had about the future parameters of NZS.  But the two aren’t really very logically connected –  despite the words in the legislation.  If speculative investment management is a natural function of government, it is so regardless of baby boomers ageing, life expectancy or the parameters of any element of the welfare system.  Short of New Zealand discovering Norwegian quantities of oil and gas, I suspect it is no appropriate business of government.

 

 

 

 

 

27 thoughts on “Defenders of the NZSF

  1. The crucial questions from my point of view are: (1) Can we reasonably expect that the Fund will obtain a pre-tax return significantly above the cost of borrowing, over the long term? (2) Secondarily, is there a major risk that its value will negatively deviate below that track at some point prior to the long term by enough to risk the Government’s ability to borrow?

    If the answers are Yes to (1) and No to (2) for a given level of contribution to the fund, wouldn’t the rational thing to do to keep on increasing contributions until the answer to the second question starts getting shaky? Otherwise we are, as a country, foolishly leaving money on the table by leaving this anomaly unexploited.

    Like

  2. setting aside all the governance issues etc, then I might broadly agree with that sort of framework. but i’d redefine your first criterion as “can we reasonably expect the Fund will obtain a pre-tax return at least as good as returns citizens might expect to obtain on other long-term projects” – which is where the private sector hurdle rates come in. so far, they’ve barely met that sort of level of returns – and that on the basis that (as they say in the OIA response) they’ve done better than they expected. That doesn’t make me optimistic about the ability to sustain average 10% returns in future.

    Like

  3. Another comment – admittedly not a major one. You say: “All the money put into the NZSF has either been raised from taxes or borrowed. There isn’t a pool of money that naturally needs investing. Rather, the government has established a high-risk investment management subsidiary to punt on world markets. That simply isn’t – and never has been – a natural business of government.”

    My favourite lecturer would have called the last sentence “simple assertion.” It is not a surprising assertion since you have a centre-right economic viewpoint. But a less ideological view might be that there are benefits that come from being a nation state, such as seigniorage, and that using those for the benefit of citizens is indeed a natural business of Government. If the cost of borrowing is below the risk-adjusted return, and by enough so that one can be confident that it will be a nett boon ex-post, then why not? Now, your point about hurdle rates is entirely fair, and that goes to the question of whether you cna be sure, and to be honest I’d fret about it if I were our Treasurer; and I’d also fret about a future Treasurer cancelling the whole enterprise and selling at a low point.

    Like

  4. Simple assertion, or reasonably historically accurate description? I was intending the latter, rather than (primarily) making a philosophical point. Happy to engage in the debate with anyone who thinks it should be a natural business of govt.

    Re nation states, yes they exist for a reason, and in the modern incarnation seignorage is usually one of the gains (tho not say in Panama or other dollarised places). I put quite high risk global business investments in a different category, but I’m sure there is a spectrum where it might sit.

    Like

  5. I didn’t hear Adrian on the radio, but he may not have been completely wrong in claiming that the fund “somehow reduces risk”. One (and probably the only) advantage of such a fund is that it allows for a more diversified portfolio: you can invest in foreign assets whose returns may be at least partly independent of NZ economic fortunes, whereas a pay-as-you-go scheme is necessarily 100% invested in NZ Inc.

    It’s hard to agree with any of his other justifications though. And in evaluating performance, which I agree is really only par-for-the-course over the last 8 years, it’s easy to forget that the Fund hasn’t had to deal with inflows of new money during that time, which is what brings all active fund managers down eventually.

    As for Ryan and Small, well, this just reflects that they aren’t economists and so have no built-in bullshit detector. If they’re told something is good, they believe that even more of it would be better still.

    Like

    • Yes, it is a possible interpretation and a plausible story to some extent (altho the correlations all tend to go to 1 in the extreme events that probably matter most, and NZ does collect – some – tax revenue on NZers’ overseas assets).

      Like

  6. I’m getting lost amidst the rhetoric

    You are against a Sovereign Wealth Fund? Why

    I was not in NZ when the NSF was set up.
    Wikipedia describes the Cullen Fund as a Soverign Wealth Fund
    The New Zealand Superannuation Fund is a sovereign wealth fund in New Zealand.
    https://en.wikipedia.org/wiki/New_Zealand_Superannuation_Fund

    I was resident in Australia when The AU Government set up its Future Fund

    Initially to preserve the proceeds of the sell-off of Telecom-Telstra
    The reasoning made sense
    It still makes sense
    http://www.futurefund.gov.au/

    It is now managing $150 billion

    It’s a goer

    Like

    • Compare that to NZ’s treatment of the sell-off of the Power Electrcity Assets, the proceeds of which have disappeared into the ether

      Like

    • lost amidst the rhetoric

      You are against a Sovereign Wealth Fund? Why

      I was not in NZ when the NSF was set up.
      Wikipedia describes the Cullen Fund as a Soverign Wealth Fund
      The New Zealand Superannuation Fund is a sovereign wealth fund in New Zealand.
      ”’en.wikipedia.org/wiki/New_Zealand_Superannuation_Fund

      I was resident in Australia when The AU Government set up its Future Fund

      Initially to preserve the proceeds of the sell-off of Telecom-Telstra
      The reasoning made sense
      It still makes sense
      ”’www.futurefund.gov.au/

      It is now managing $150 billion

      It’s a goer

      Like

  7. One potential benefit of nzsf – the increased government borrowings from funding it, arguably somewhat constrain them, relative to if they didn’t have the debt. I tend to think constraints on government are good, as it limits how much of our money they waste

    Like

  8. hmmm; if there are private sector investment projects with +10% returns floating about, you would expect NZ business investment to be steaming along, yet, doesn’t seem to be the case (maybe in part due to household preference for real estate investment…with support from the credit system); but, perhaps beside the point: paying down debt is risk free and guarantees an increase in equity – somewhat relevant given circa 60% of central/local government debt is held by non-residents….

    Like

  9. My take on this is that you can have three kinds of funds
    1. an accounting device. Govt contributes to fund – 100% invested in govt bonds. very cheap to run, no impact on risk – benefit is through altering govt spending etc. to more intergenerational efficient levels.
    2. Passive equity fund – exploits the ‘excessive’ equity risk premium over time -get paid more than the true risk of the investment – if you believe the excessive risk premium theory.. Still very cheap to run -out of DMO. ( note that the bulk of NZSPs equities are in indexed fund). NZSF only takes active positions where they think they can add value. No need for performance benchmarks – the excess returns and potential losses will be what they will be.
    3. Actively managed equity fund . promise of superior returns exploiting the fund’s superior patience. More expensive to run. Performance benchmarks become more important

    Which leads us to the performance benchmarks.
    Performance re t-bills ( should probably be more like av. govt funding costs since whole of the funding will not be in t-bills but that is a minor point) is really irrelevant. Since a strategic decision has been made to be in equities the outcome will be the outcome. It is interesting to know the difference to funding costs but the outcome will be largely luck.

    Performance re the passive benchmark is the test of the active component of the investment decision. ( it implicitly accounts for risk if you believe that market risk pricing is a good proxy for govt risk) On the face of it the fund has performed very well with excess returns, of, from memory, about 1.2 % – way up in the fund performance rankings
    But there may be little bit of a cheat here. The passive fund is 80% equities, 20% bonds so the natural way to beat the benchmark is to go more heavily into equities. Over time it will generate ‘superior’ performance given the equity premium and the fact that the benchmark does not account for the additional risk. As there is less of reason to borrow in NZ govt debt to invest in hedged foreign govt debt the benchmark could be 100% equities. As with most passive benchmarks there are probably other soft components that can be exploited to obtain ‘superior’ returns because embeded risks are not fully identified and priced.

    Even so they are probably still decently ahead of the game. A good part of that probably comes from their Z-investment. the issue is whether these kind of -investment opportunities will scale up if the fund were to be materially bigger -possibly not. there is also a question of whether an equity benchmark is the right metric here. A tough one to answer – arguably these investments are riskier than a broad portfolio.

    Like

    • I don’t agree with you that performance re the benchmark is most of what matters. Their benchmark itself is a strategic asset allocation choice (for which they should be accountable), and it is also set in a way that is pretty favourable to management appearing to be capable active managers. Thus, part of the Fund’s advantage is that absence of the threat of runs, and hence the longer-term holding periods/illiquidity they can cope with. But the whole benchmark is highly marketable securities.

      But even on their own active mgmt measure, performance is ok but nothing startling, The Sharpe ratio appears to have been around .45. And as you (and Glenn above) point out, they haven’t had any new inflows for eight years, and the number of opportunities for them to add value isn’t necessarily going to increase with the volume of new funds (esp the opportunities within NZ).

      From a taxpayer perspective, I think it is performance relative to the risk-free rate that matters (mean and variance), but it is fair to observe that on its current mandate NZSF can’t be expected to adopt a low risk fund – so if they end up losing over time (get caught in a decades long weak market) we shouldn’t blame them, but rather the politicians who set up the fund and gave them the existing relatively high risk mandate.

      Like

      • philosophically i agree with you – I struggle a bit with the piggybank model. -easier just to repay the debt. If there is a free lunch here it could just as easily apply to the Ministry of Womens Affairs or any other spending programme if we pre funded that expenditure it would free up resources for other spending including pension , health etc.

        Like

  10. I read Andrew Coleman’s eight-page defence of savers you go versus pay-as-you-go.

    I must read more of the literature but it seems to be along the lines of higher taxes now has a trade-off for lower taxes later on various assumptions about the growth rate of wages and capital.

    My concern is most of the writers in this literature do not exactly have a public choice awareness.

    The real trouble is the NZ superannuation fund was set up to defend the existing eligibility parameters which is at the age of 65. Higher taxes now to avoid lower taxes later by raising the eligibility age.

    Edward Prescott has done quite good work in the USA showing the large productivity gains from turning Social Security into mandatory private savings accounts. People treat the savings accounts as wealth while they treat their Social Security taxes as taxes.

    Like

  11. Either way, the super fund now admitted that they had not written to me prior to saying that they had on air. There is an internal mixup.

    The letter they added at the end of the OIA request was simply putting a different view.

    Hopefully the Dominion Post will publish my letter to the editor tomorrow or the next day but they do not have a good record regarding any communication from the taxpayers union.

    Like

  12. It all points towards compulsory KiwiSaver so that each individual becomes responsible for their own retirement. Make KiwiSaver compulsory and then It is best to disband NZSF and distribute the fund to all existing KiwiSaver contributors.

    Like

  13. I’d simply raise the age of NZS eligibility to 67/68 and then index future increases in the age to future increases in life expectancy. That would materially and permanently lower the fiscal burden, keep the system simple, and avoid the pretence that poor people can really afford to save for a retirement income. Without more far-reaching changes, there is no public policy role for Kiwisaver. The middle classes, who want a reasonable standard of living in retirement can – and do – save for themselves, usually in more tax-efficient ways than (say) having compulsory taxed Kiwisaver accounts on the one hand, and large non-deductible mortgages (on owner-occupied properties) at the same time. Change the tax treatment of savings and it becomes less inefficient and less inequitable – but getting that sort of change will be even harder than changing NZS.

    Liked by 1 person

    • Maori would argue that their average life expectancy is lower therefore 67 does not work for them in general. This then opens up discriminatory policies based on gender and based on racial lines.

      Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s