The article by the USS fails to make much of a case for conditional indexation (CI). It does not properly discuss the key issues that have to be addressed first before CI can be adopted. 

It is not clear as to precisely which problem CI is the solution to: Is it a chronic tendency for the scheme to seem to be often in deficit – as we saw throughout the period after 2010 when the scheme was always in deficit and many commentators were telling us it was therefore not sustainable as DB and that it would have to go the way of other private sector schemes and become DC? That situation persisted until just before the 2023 valuation.

Incidentally the UCU’s actuarial advisors First Actuarial were arguing that these deficits were the result of over-cautious valuations of liabilities due to the Scheme Actuary on gilt yields to represent investment returns in the calculation. This was at a time when gilt yields were extremely low in the aftermath of the Great Financial Crash of 2008 when governments were trying to stimulate the economy using low borrowing rates. It would have been more realistic to calculate valuations using  scheme-specific expected investment returns – and that would have largely removed the problem.

The problem was caused by very low gilt rates due to government policy and NOT poor performance.

Or is it that there was a significant regime change in 2022 and since then the scheme is generally going to be in surplus – even if valued using gilts – with maybe one or two exceptional years of deficit. There is a big difference between these two scenarios and I fear they may have got mixed up in some people’s thinking.

My view is that as a general principle it does not seem unreasonable to require that pension increases should only be paid if the funds are available, and therefore CI would be fair. What is difficult  to accept is an iinflation cap applied whether or not the funds are available in the scheme. The so-called soft cap pays the full increase up to 5% inflation but only part of it above that. Why is that not conditional? Why has the UCU accepted the cap when bouts of high inflation can occur so easily, eg, as a result of supply chains being disrupted in our increasingly unstable world.

But the main issue with the idea of CI, that is only mentioned in passing in the USS paper, is what the actual condition is. There is a lot of heavy lifting still to be done on this, the arguments set out at length by Dooley Harte. 

You could not simply base the CI policy on a gilts-based valuation. That is, using a gilts-yield discount rate to calculate the liabilities would mean the valuation is misleading as an indicator of the performance of the scheme because its investments are not mainly in gilts. The correct methodology uses a discount rate based on a best-estimate of the expected returns from the scheme’s actual investments. That way the valuation can be assumed reasonably to reflect performance whereas a valuation based on gilt yields would be a reflection rather of government interest rate policy and market movements in the bond market.  

Not only is a gilt-based valuation misleading (biased) it is also highly variable because gilt yields are market determined by bond prices which vary daily. Gilts based valuations tell us little about the scheme’s performance.

By law the discount rate must be chosen by the Trustees but in doing so they have wide discretion (so long as they act prudently with actuarial advice). Our task as a union and as members now is to persuade the Trustees to use Best Estimate discount rates instead of gilts. These are fundamentally different.

The good news is that  our union’s Superannuation Working Group has now adopted,  as a main policy priority to follow the advice of our actuarial advisor Derek Benstead of FA, to persuade the Trustees to radically change the valuation methodology and do just this for the next valuation. See https://tinyurl.com/yswpz9u4.

 This is the key issue for the USS and if adopted it would enable CI to be introduced in the manner described for the Canadian schemes which I understand use Best Estimate valuations.