UK university pensions suffer from misplaced prudence

UK universities face a fresh wave of strikes over pensions. That was the indication from reports warning that the students who have recently returned to university may face more disturbance from striking academics. It is easy to condemn them, but I do not. They are the victims of unduly risk-averse decision making at the Universities Superannuation Scheme, under the influence of misconceived regulation.

The USS is a large funded scheme, with 476,000 members and £82.2bn in assets. Universities are also more or less immortal institutions. If they cannot afford the benefits promised in their scheme, nobody can, apart perhaps from the government.

Given the impossibility of total safety, how “safe” would a prudent person want such long-term promises to be? Two recent papers illuminate this question.

Raghavendra Rau of Cambridge analyses a portfolio weighted 55 per cent in global equities, 20 per cent in global bonds, 20 per cent in UK bonds and 5 per cent in UK bills from the year 1900. He compares the USS’s “prudent return” assumption of zero real return over 30 years with the worst realised return of such a portfolio in the last 120 years, which covers two world wars and the great depression. It turns out that the USS’s prudent assumption is far worse even than these disasters.

Is this prudence or extreme pessimism? The former, argue David Miles and James Sefton of Imperial College. They estimate risk in terms of a standard probability distribution and conclude that “substantial investment in riskier assets (equities) makes the average outcome one in which the scheme is comfortably able to pay accrued benefits. But the risk of having far fewer funds than needed to pay existing pension promises is significant and the chances of large deficits is very substantial.”

Thus, even if returns on equity are mean reverting, there is a 23 per cent chance that the USS would run out of money by 2100. If one wishes to avoid this possibility, something has to change, they insist.

This raises two big questions. The first is over the likelihood of such dire outcomes. The second is over whether one can insure a pension fund against such disasters.

In correspondence with me Miles argues that “there is a big risk of sample selection bias: the US and UK did not lose wars, did not have hyperinflation, were not invaded, did not have political breakdown. Returns on Argentine assets, on German (stocks wiped out in mid-century) etc look very different and are massively more volatile.” It is easy to imagine catastrophic developments — thermonuclear war, communist revolutions, collapse of the state and so forth — that would wipe out any equity portfolio.

Nevertheless, it defies belief that a higher contribution rate and more cautious investment strategies would protect wealth against such catastrophes. These are uninsurable, certainly for a large number of people. If the UK and the US were hit by 1,000 nuclear warheads, capitalism collapsed or governments defaulted and taxed away most wealth, no investment fund would survive. Wealth destruction on this scale cannot be avoided by a little prudence today. That gives an illusion of safety, but not the reality.

Forget this notion. What does make sense is to have a sensibly invested fund (that is, one predominantly in equities) that is structured in ways that limit the downside risk to both members and sponsors in the event of things going wrong, on a relatively manageable scale. What would then be needed is some adjustment of benefits and contributions. This flexibility is what all pension funds need, not to protect themselves against extreme disasters, but against the downsides of actual performance in the real world.

In a sensibly managed collective defined contribution scheme, that would happen. Since the sponsors of the USS do not have infinitely deep pockets, it makes sense for the liabilities they bear to be capped. As Miles also writes: “The question is whether the USS with its strange structure (large numbers of different institutions, many with limited ability to put more in) is able to make promises guaranteeing fixed pension payments. Risk needs to be spread better.”

That makes good sense. But the answer is to consider precisely that structure. It is not for the USS to try to insure itself against the end of our world. The end of the world is uninsurable. No sensible fund should attempt any such thing. In the normal course of events, the USS is more than properly funded, whatever the regulations may say.

martin.wolf@ft.com

Follow Martin Wolf with myFT and on Twitter