Self-sufficiency in gilts is a costly means rather than an end in itself
According to USS’s Test 1, the pension scheme must be within reach, via 7% increase in employer contributions, of “self-sufficiency”, which “reflects the required level of assets to meet all future benefit payments to a very high probability without the need for additional contributions” (p. 10, n. 5, my bold emphasis).
So self-sufficiency isn’t an end in itself. Rather, it’s a means of protecting against the risk of not being able to meet pension promises. If, therefore, there is a more cost-effective means of protecting against such risk, it should be adopted instead.
USS quantifies the aforementioned very high probability as greater than 95% (see September 2017 consultation document, pp. 20, 34, and 42). Therefore the following principle underlies Test 1: employers should have a greater than 95% certainty of never having to raise their contributions by more than 7% in order to provide promised pensions.
USS maintains that a self-sufficiency portfolio that could deliver benefits with such certainty would need to be so weighted towards gilts (or other ‘liability matching’ assets) that its returns would exceed the low gilt yield by no more than +0.75%. It also maintains that the scheme’s current growth portfolio needs to be re-balanced away from equities and into gilts, to make it possible to fund a move all the way to a gilts +0.75 portfolio via no more than a 7% increase in employer contributions. The shift towards gilts that USS is calling for would result in a decrease in the long term expected returns on the portfolio by about 0.8%.
But modelling from First Actuarial indicates (and see also p. 30 of their submission to the 2014 valuation) that the above principle (in bold font) can be satisfied in less costly fashion if USS remains invested in a portfolio that is heavily weighted towards growth assets such as equity.
On USS’s best estimate of the return on index-linked bonds (see p. 19 of USS’s September 2017 consultation document), a gilts + 0.75% portfolio would deliver zero annual real returns (i.e., returns no greater than CPI) over the next 30 years. It would stretch credulity for USS to maintain that there is at least a 5% chance that the annual returns over 30 years on their current return-seeking portfolio will be more than 1% lower than the expected returns on gilts + 0.75%, even after we add, to these returns, an annual infusion of extra employer contributions equivalent to 7% of salaries. On the far more credible assumption that there is a less than 5% chance of the above, a discount rate of 1% lower than a best estimate of returns on their current portfolio will satisfy the principle that underlies Test 1, and there is no case for USS’s proposed costly de-risking of the portfolio.
[Please see this follow-up to the above post.]