IMAGINE two kinds of investment funds, both of which have the same aim: to provide pensions for their employees. You might think that they would invest in a similar way. But when it comes to American pension funds, you would be wrong. It turns out that public funds have a rather different, and more aggressive, approach to risk from private ones (and indeed from their counterparts in other countries).

As a recent paper* by Aleksandar Andonov, Rob Bauer and Martijn Cremers shows, that different approach is driven by a regulatory incentive—the rules that determine how pension funds calculate how much they must put aside to meet the cost of paying retirement benefits. Usually, the bulk of a pension fund’s liabilities occur well into the future, as workers retire. So that future cost has to be discounted at some rate to work out how much needs to be put aside today.

Private-sector pension funds in America and elsewhere (and Canadian public funds) regard a pension promise as a kind of debt. So they use corporate-bond yields to discount future liabilities. As bond yields have fallen, so the cost of paying pensions has risen sharply. At the end of 2007, American corporate pension funds had a small surplus; by the end of last year, they had a $404 billion déficit.

American public pension funds are allowed (under rules from the Government Accounting Standards Board) to discount their liabilities by the expected return on their assets. The higher the expected return, the higher the discount rate. That means, in turn, that liabilities are lower and the amount of money which the employer has to put aside today is smaller.

Investing in riskier assets is thus an attractive option for a public-sector employer, which can tap only two sources of funding. It can ask its workers to contribute more, but since they are well-unionised that can lead to friction (after all, higher pension contributions amount to a pay cut). Or the employer can take the money from the public purse—either by cutting other services or by raising taxes. Neither option is politically popular.

Unsurprisingly, therefore, the academics found that American public pension funds choose a riskier approach. Theory suggests that as pension funds mature (ie, more of their members are retired), they should allocate their portfolios more conservatively, because the promised benefits need to be met sooner and funds cannot risk a sudden decline in the value of their assets. That is the case with private-sector pension funds, but public funds take more risk as they mature—putting more money into equities, alternative assets (like private equity) and junk bonds.

Public pension plans have also increased their allocation to risky asset classes as interest rates and bond yields have declined. Again, this does not make sense in theory. The expected return on both risk-free and risky assets should decline in tandem. But a fall in ten-year Treasury-bond yields of five percentage points has been associated with a 15-point increase in public funds’ allocation to risky assets.

The academics also look at the trustees, the people who make the investment decisions. They find a relationship between the riskiness of a fund’s assets and the proportion of political trustees (such as state treasurers) and worker trustees elected by scheme members. Neither group will want to see contributions rise in the short term. So it makes sense that both groups bank on achieving higher investment returns, leaving any scheme shortfall to be cleared up later.

Some people might argue that this is all for the good. Pension schemes have long-term liabilities and thus should take more risks. If public funds take more risk and earn higher returns, that is saving both their employees and taxpayers money. Alas, the academics find that, even if you allow for their asset-allocation decisions, public pension funds underperform their benchmarks by more than half a percentage point a year. This underperformance is greatest in the alternative-asset categories such as private equity. Even if their asset allocation might have beaten a bond-only portfolio, the overall strategy isn’t working. For example CalPERS, the Californian state pension fund, has underperformed its targeted 7.5% return over the past 3, 5, 10 and 20 years.

The paper demonstrates convincingly that American accounting regulations have created perverse incentives for public pension funds. And that can mean only one thing. The rules need to change.


* “Pension fund asset allocation and liability discount rates”, March 2016