JEP report, new research findings and fiduciary duties of trustees

Below is my letter to the USS trustees. The letter is copied to the Pensions Regulator

Professor Sir David Eastwood
Chair, Universities Superannuation Scheme
Royal Liver Building
L3 1PY
Via email to:

21 November 2018

Dear Chair and Members of the USS Trustee Board

JEP report, new research findings and fiduciary duties of trustees

First, let me express my sincere thanks to the board of trustees for making available information, some of it commercially confidential, to the Joint Expert Panel (JEP). As the JEP acknowledged, this information has been invaluable in its assessment of the 2017 valuation of USS. I firmly believe both USS scheme members and employers share the same common interest, and I hope my new research findings together with the JEP report will help the board of trustees to fulfil the fiduciary duties, as required by the trust law, to act in the best interests of the scheme beneficiaries.

Industry debate and USS’s flawed actuarial assumptions

While the 2017 valuation of USS reports a huge deficit of £7.5bn, the First Actuarial, the UCU’s actuary, finds the scheme in a healthy state with no funding crisis. The contrast between the two is a predictable outcome of the industry debate pointed out by the Pensions Regulator in its 2017 Annual Funding Statement for Defined Benefit Pension Schemes. Specifically, the debate focuses on whether the historical relationships between gilt yields and returns on other asset classes still hold true for the future. It may be difficult to argue against either of the two vastly different valuation outcomes if there is no clear answer to the debate, and hence it is understandably acceptable for the trustees to continue to use the current USS’s actuarial assumption that interest rates play a crucial role in its valuation outcome. I would like to draw your attention to my research paper completed on 7 May 2018, which shows that the relationships no longer hold because gilt yields are driven by factors that are entirely different from those that determine the returns on equities, the most important asset class. In the 1970s and 1980s, inflation was high, which meant that interest rates and the returns on equities were comparable. Since then, successful monetary policies and other economic developments have brought stability to prices, resulting in the long-term decline of interest rates. But because the returns on equities are the result of productivity of firms in the real economy, they remain high despite current low interest rates.

Please be informed that problematic actuarial assumptions and/or practices are not new. Day (2004), for example, pointed out that ‘many standard modes of actuarial thought are, in fact, indefensible when examined with the tools and techniques of financial economics.’[i] It is thus not surprising for our former Bank Governor Mervyn King and Oxford economist John Kay to write in a recent article that the main risk to the USS and its trustees is ‘the possibility that in three years’ time the technical valuation will incorporate even more pessimistic assumptions.’

Holistic risk management and optimal portfolio to reach self-sufficiency

Risk management should not be fixated to only holding low risk assets and/or liability matching since it is the nature of financial markets to charge inordinate prices when they think you do not have a choice. A holistic approach to risk would also consider the more traditional method of risk reduction through long time-horizon holding. As the JEP report indicates: ‘insufficient weight has been given to the fact that USS is a large, open, immature scheme which is cash-flow positive and can adopt a very long time-horizon.’ Consistent with the JEP statement, my other research paper completed on 31 Oct 2018 shows that the risk of investing in equities can be effectively mitigated through long time-horizon holding or the principle of time diversification. Moreover, the paper investigates how the allocation of funds between stocks and gilts affects the risk of underfunding. Because of current low interest rate, the blue curve in Figure 1 shows a high certainty of underfunding if the scheme’s portfolio holds only gilts. As more funds are invested in stocks, the probability of underfunding decreases to the point of self-sufficiency, thanks to the principle of time diversification.

The x-axis is the proportion of funds allocated to equities. ‘0’ on the left is all gilts whereas ‘1’ on the right is all equities. Left vertical scales are probability of underfunding. Right vertical scales are expected funding ratio in percentage.

The JEP report also points out that ‘the hypothetical move to a low-volatility, low return portfolio is only one of many paths available for … USS.’ The implication is that self-sufficiency needs not be achieved only by means of a low risk low return portfolio. To illustrate this point, Figure 1 also shows the expected funding ratio (yellow curve). It can be seen that, by virtue of time diversification, high funding ratio is achievable with low or acceptable risk by allocating more funds to equities.

Figure 1 does not mean that the USS’s portfolio should hold only equities as this will not be the risk preference of its members and sponsors. But it is vital for USS to share the above information fully with its members and employers, for the JEP finds that ‘employers have been asked questions in consultations and questionnaires that have not fully explored the consequences or trade-offs of the issues under investigation.’

Reply to USS’s response on 16 October 2018

The implications of the JEP report and my new research findings can also be appreciated in my reply below to the USS’s response to Dr Sam Marsh’s finding.

According to Dr Sam Marsh’s calculation, with de-risking absent, the deficit of USS as at 31 March 2017 is £0.4bn which implies little or no deficit recovery contribution is required. The USS responded on 16 October 2018 by saying that while ‘Dr Marsh’s analysis is not wrong in isolation — but it is simply not an adequate premise on which to set the funding arrangements for the scheme.’ In particular, according to the USS, planning for contributions that are adequate on average over time is not a sufficient condition for an acceptable valuation. This is mainly because of the risk that (a) interest rates do not rise as expected; and (b) possible correction in asset markets.

With regard to (a), as pointed out earlier in the letter, future returns on other asset classes (especially equities) do not depend on whether interest rates will rise in the future. Indeed, on the basis of constant interest rates, my two research papers mentioned above find that USS is more than fully funded. My results are consistent with the analysis of First Actuarial, the UCU’s actuarial advisor.

On (b), first note that the discount rates used by the USS for 2017 valuation average only 3.27%. Despite the low discount rates, Dr Marsh finds a deficit of mere £0.4bn. The over-pessimistically low discount rates reflect the current low interest rate since the USS assumes that low gilt yield means low future returns on other asset classes. If this flawed actuarial assumption is abandoned, a higher but prudent discount rate would yield a significant surplus for USS, a result similar to my research findings and that of First Actuarial. Therefore, there should be a reasonable ‘surplus buffer’ that enables the USS to remain in a healthy funding position if there were a market correction.

The risk of a market correction can also be mitigated by sharing the risk between age cohorts. If the USS is a defined contribution scheme, its pensioner members will be hardest hit when a market correction takes place. Since the USS is predominantly a defined benefit scheme, the risk of a market correction can be absorbed by active members as their contributions will be buying assets at a lower price after a market downturn. Indeed, as the JEP report indicates, the USS is an open, large and immature scheme with positive cash-flows. If there were a market correction, no assets need to be sold at low price as the sum of contributions and cash-flows from assets are more than sufficient to pay for the promised benefits for a considerable number of years.

Finally, the Chief Risk Officer of USS remarks that ‘current active members would today be accruing benefits that cost more than the contributions that are being paid (And, on the flip-side future members would be accruing benefits which cost less.)’ It is worth looking at this issue from another perspective. That is, if the reference portfolio (roughly 70% of funds in equities and properties) can be implemented, Figure 1 shows that the future funding ratio can be high enough to allow a lower contribution rate for future members.


The past relationships between gilt yields and returns on other asset classes no longer hold true for the future, and hence interest rates are no longer appropriate for the valuation of the USS. Moreover, since the JEP report, new evidence shows that

1. The USS has no funding crisis and deficit recovery contributions are not required.
2. De-risking is the problem rather than the solution.
3. More funds should be allocated to equities, for the associated risk can be effectively mitigated through time diversification and risk sharing between age cohorts.
4. Given the current low interest rates, there are far better ways than a low-volatility low return portfolio to achieve self-sufficiency.

Let me repeat this: I firmly believe both USS scheme members and employers share the same common interest, and that relevant research together with the JEP report will help achieve it. My research findings are open and transparent. I urge the board of trustees to examine and challenge them if any flaws can be found. Otherwise the board of trustees risks breaching the fiduciary duties to act in the best interests of scheme beneficiaries if the existing USS’s valuation method and investment strategy based on flawed actuarial assumptions continue to be implemented.

Finally, allow me to remind the board of trustees that unnecessary deficit recovery contributions due to flawed actuarial assumptions do not serve the best interests of scheme beneficiaries, amongst them are (a) active members who are experiencing a cut in their real pay; and (b) prospective members who may be deterred from joining the pension scheme due to higher contributions. Indeed, I believe many of the active members belong to what our Prime Minister Teresa May describes as ‘just about managing’; some may be forced to opt out from the scheme if there is an unnecessary increase in contribution.

Yours sincerely

Dr Woon Wong, FRM
Reader in Financial Economics
Director of Trading Room Operations and Development
Cardiff Business School

Cc: Lesley Titcomb, the chief executive of The Pensions Regulator

End notes

[i] Day, T. 2004. “Financial economics and actuarial practice.” North American Actuarial Journal 8:3, 90–102.