Prepared for the Institute of Development Studies (IDS) and the University of Sussex by the Sussex UCU Pensions working group, in response to the USS 2020 valuation discussion document published September 2020. All Sussex UCU members and wider groups of staff were consulted throughout the preparation of this view. Employers have been asked to respond by 30 October.
1. Introduction
We thank IDS Management for welcoming staff inputs to their response to the consultation by UUK, and therefore this opportunity for the Sussex branch of UCU to provide a view on the consultation at the invitation of our IDS members. We further thank IDS Management for welcoming the presentation of the view of Sussex UCU at the IDS staff meeting on Monday 19 October.
We thank the Chair of Council of the University of Sussex for responding to our request and sharing the Sussex UCU with all Council members. We thank Sussex UEG actioning our request for a USS landing page of information for Sussex staff.
We hope there is an appreciation that the time available to understand the proposed valuation and prepare a view has necessarily been limited. We have chosen not to comment on the eight individual questions in the USS consultation document. In making this choice we note the UUK view that
The trustee’s questions are wide-ranging, and it might be difficult for employers to respond conclusively; we expect that some employers might take the view that the implication of a response in one area is entirely dependent upon another element – and then in an overall sense dependent upon the final outcome being a more sustainable scheme for employers and members. This is a point we have made a number of times in recent months, and the difficulty of seeking responses from employers without the fuller picture of a potential scheme outcome is, we believe, very apparent. [UUK Information Paper, p3]
Instead we consider the proposed valuation from three successively broader viewpoints:
(i) within the framework of the USS proposed valuation method;
(ii) in the context of the work of the Valuation Methodology Discussion Forum (VMDF), which was established as an outcome of the JEP process; and
(iii) in the context of current work by experts in ethics, risks and the regulatory framework.
We have calculated potential costs and savings to the University of Sussex of the various methodologies. These range between £5 million in annual savings to £20 million in annual costs for Sussex University. These are significant sums under any circumstances. In the current crisis they are critical and as such reinforces the importance of detailed public scrutiny of the proposals.
On opt-outs, government support and the climate crisis
We have evidence that staff at IDS and Sussex continue to be extremely concerned about the unaffordability of the October 2021 increases. Contributions rates are currently 9.6% employees (ees), 21.1% employers (ers) and due to rise to 11% (ees) and 23.7% (ers) on 1 October 2021. We continue to stress that the opt-out rate is already unacceptably high, reported by JEP as currently in the region of 15-20%, and likely to increase if further contribution increases are imposed. The contribution rates arising from the proposed valuation vary (depending on the application of different USS assumptions) from 40.8% (13.2% from employees plus 27.6% from employers) to 67.9% (22.7% from employees plus 45.2% from employers) and are clearly unaffordable for a very high proportion of members as well as many employers.
We believe there are strong arguments for a government guarantee of the University sector’s pension scheme. Such a guarantee would support the sector as a key driver of innovation and skills; and ensure recruitment and retention at the highest levels to maintain the UK reputation as a world leader in the research and higher education industries. This guarantee would completely remove the rationale on which USS relies when it insists on applying tests and metrics which assume that total closure of the scheme is a possibility and therefore that the Trustee is required to put forward deficit recovery plans that factor in this possibility. With assets valued at £75.3bn as of the end of August 2020 (compared to a value of £63.7bn on valuation date 31 March 2018 and £66.5bn on valuation date 31 March 2020), 460,000 members and a cashflow that is projected to remain positive for at least another decade (allowing the asset base to continue to grow during this period without any need to draw on it to cover retirees’ benefits), USS would not represent a burden on the taxpayer (in fact the reverse would be the case). A government guarantee would therefore be a low- or zero-cost strategy for removing one of the most important sources of financial stress on members and employers alike at a time when the sector as a whole badly needs a breathing-space in which to prepare itself to rise effectively to the challenges of a post-COVID-19 world.
IDS and Sussex staff and students would also like to see more urgent and public action taken with regard to the USS investment strategy and voting record in line with the urgent action needed to address the Climate Crisis. We are alarmed that UUK took no public view when USS voted earlier this year against the Royal Dutch Shell shareholder proposal to set and publish < 2 °C Paris Climate Targets. We believe that an opportunity is being missed for USS to switch towards investments in green infrastructure in line with the stated aspirations of the UK government and the remit of universities as charitable institutions with a public benefit duty. The UK University sector is greater than the sum of its parts and we support mutuality.
On the USS consultation
On the specific matters on which the USS consultation focuses, we find it difficult to see how the majority of USS members will accept the proposed contribution rates in the context of concerns raised without calling on UCU to enter into dispute.
We ask that employers call for urgent work be carried out to address the following points:
As a bare minimum, the current model’s assumptions about prudence and reliance on covenant should be reconsidered and justified in depth. If these are revisited it seems possible that agreement could be achieved on a fallback position of maintaining current benefits for the current contributions before the deadline of the October 2021 increases. This has the potential to avoid the risk of further damaging industrial action and a further decline in trust and morale. Urgent work must be carried out in parallel to address the opt-out rates.
The VMDF must continue its work. It must have the funds and the resources to do this efficiently, effectively and publicly. In particular we note the VMDF modelling evidence that demonstrates 26% contribution rates to be sufficient to fully fund the scheme with existing benefits. We note the reports of modelling that indicates the scheme is robust to bad events or shocks. These models need serious attention. Detailed analysis and clear and timely communication on this work is a matter of urgency.
The University sector as a whole must show leadership and publicly support experts across disciplines who are working at the forefront of research and calling for an urgent review of the regulatory framework of the pension and financial sector. Against the backdrop of the USS dispute, now is the time to place greater emphasis on ethics, stewardship and the public good.
2. Comments on the valuation as a whole but from three distinct viewpoints.
(i) The view within the framework of the USS proposed valuation model
More time is needed to understand the new proposed model but the initial view is broadly that prudence assumptions are too conservative and covenant strength underestimated. There seems to be an alignment of views from both employee and employer representatives that the changes to the methodology proposed by the Trustee are little more than window dressing. We remain unconvinced that the change from Test 1, 2, 3 to Metric A, B, C produces a new valuation methodology that is fit for purpose. We reiterate the JEP position that ‘the Panel could not identify any circumstances in which it believed that Test 1 should be the primary driver of funding and investment strategy.’
USS has proposed contribution rates, dependent on underlying assumptions of between 40.8% and 67.9%, which are obviously unaffordable across the sector. There are compelling and convincing arguments that current benefits can be maintained, at the current contribution rate and within the framework proposed by USS. A copy of such a path described by Mike Otsuka is added as appendix 1. These points appear to have been completely ignored by the Trustee despite calls for such consideration from actuaries and VMDF members on both the employer and employee side.
(ii) The view from within the Valuation Methodology Discussion Forum (VMDF)
The VMDF held 11 meetings between 6 February and 13 July 2020, this work is covered in Appendix A of the consultation document. We thank UCU and UUK representatives for their work on the VMDF. We are encouraged by multiple reports of an ongoing collegial working environment between representatives of UCU and UUK. We believe this demonstrates what can be achieved when employers accept the need to work constructively with employees in pursuing mutually acceptable solutions rather than defaulting to a position of alignment with the Trustee. It is our view that the work of the VMDF must continue through the period of the 2020 valuation, and that this work must be well resourced and publicly available.
We are dismayed at the reported delays in VMDF being given access to USS data, as evidenced in this comment from UUK representatives of the VMDF:
Meetings 3-6 were then not particularly useful as the main information requested on 14 February 2020 was not available until [20 April 2020] meeting 7
We firmly believe that a necessary condition for the successful conclusion of the 2020 valuation involves the VMDF having timely and full access to the USS data. This is needed both to understand and communicate the proposed valuation as well as to continue to explore alternative models.
We would particularly like to draw attention to the following comment from UUK representatives of the VMDF :
The modelling evidence (meeting 9) shows that returns of CPI+1.2% p.a. are sufficient for the scheme to be fully funded in 20 years with a 26% contribution rate from the comparatively low asset valuation at 31 March 2020. We asked in meeting 9: Why is this an issue since CPI+1.2% looks low compared with historical returns (and many endowment targets), and since the modelling provided for meeting 7 showed the Scheme was robust to selected bad events? This question was still being explored in meetings 9 to 11 with additional modelling of more extreme scenarios, but there lacks a clear conclusion that will be easy for employers to engage with in our view. Until this can be addressed definitively, it is difficult to see how UUK can have a meaningful consultation with employers without claims that there is no problem to fix. [our emphasis, p2, UUK representatives views]
A copy of such a path described by Woon Wong is added as appendix 2. If modelling shows that a 26% contribution rate will mean the scheme is fully funded in 20 years time then we are of the view this approach to the 2020 valuation should be explored with urgency and with full transparency.
(iii) The view in the context of the regulatory framework
At this stage we recall the FT editorial board, writing in November 2019:
The excellence of the UK’s higher education offering has long been one of its greatest competitive advantages, and will be even more pertinent in a post-Brexit world. Given universities’ key role in the economy, and the potentially deeply damaging impact of the USS dispute, external mediation is needed to bring a resolution. An independent inquiry should scrutinise the handling of the USS valuation by all key players, including the Pensions Regulator, and its potential effects on university funding. [our emphasis]
The House of Lords amendment to the Pensions Scheme bill was agreed 263 to 227 on 30 June 2020. It requires support for open and immature DB schemes, noting that ‘an open and active scheme that is receiving regular, significant cash contributions is very different from a closed scheme’. The Lords debate covered issues at the heart of the USS dispute. For example:
The Government’s recent White Paper ...will be at odds with the draft DB funding code that may emerge from this legislation, which seems to want to drive DB schemes on a path to so-called de-risking, aiming for a particular date of maturity. This concept is simply inappropriate for an open scheme. [Baroness Ros Altmann]
We draw attention to the two forthcoming events. First the annual Uehiro Practical Ethics lecture series ´How to pool risks across generations: the case for collective pensions´ on 3, 10, 17 November 2020 delivered by Professor Michael Otsuka:
...to bring the best scholarship in analytic philosophy to bear on the most significant problems of our time, and to make progress in the analysis and resolution of these issues to the highest academic standard, in a manner that is also accessible to the general public.
Secondly the forthcoming Newton Gateway event on 17 November hosted in collaboration with the Royal Statistical Society on ´Mathematics and Statistics for Effective Regulation´. This will be an opportunity for representatives across disciplines to discuss pension and financial sector ethics and regulation informed by, but beyond the context of the USS dispute.
The talks will look at risks in the financial sector, ethics, handling uncertainty and achieving transparency which is important for trust. The final session will be a discussion of methods to investigate how regulation could be more efficient and effective.
The introduction will be by Jane Hutton, Professor of Medical Statistics, Expert Witness and cofounder the interdisciplinary Ethics and Mathematics 2. Jane currently has an Employment Tribunal claim against her dismissal from USS for whistle-blowing. Sam Marsh, member of the VMDF and UCU representative for the USS JNC has launched a petition calling for full publication of the report into Jane Hutton's dismissal and breakdown of legal costs related to her dismissal. This petition will form part of an internal complaint to USS. The one day event on 17 November will bring together representatives of the Pension Regulator, the Home Office, the Department for Education as well as cross discipline experts in risk, ethics, modelling, uncertainty and regulation.
We believe the position of the FT editorial board, the debate in the House of Lords and the events mentioned above demonstrate that issues at the heart of the USS dispute represent critical questions for our time. Universities need to be at the forefront of framing discussions around economy, risk, inequality, ethics and sustainable investment. The long and brutal years of the dispute, played out in the national and international media, now provide Universities a public platform to lead the conversation. This work is both necessary and urgent with Universities holding the power to drive societal and economic progress.
Appendix 1
A case for ‘no detriment’ from the status quo:
By Mike Otsuka, Professor in LSE's Dept of Philosophy, Logic & Scientific Method
11 October 2020 [shared with permission via UCU pensions email list]
Here’s a case for retaining status quo benefits and 30.7% contributions. It adheres to the limits of Metric A (the new Test 1), while also pushing against some of the less defensible elements of the USS valuation, in which the trustee is choosing to be more conservative than past practice. It also refrains from dipping into the same investment risk well twice – as Bill Galvin warns against in his THES piece.
In other words, here’s how 30.7% contributions can be justified, from within the confines of USS’s current parameters and past practices:
We can get to 30.7%, either by (i) going to the limits of USS’s risk metrics for the pre-retirement discount rate, but not also pushing on deficit recovery plan or calling for smoothing of future service, (ii) or by staying within USS’s more conservative assumptions for a pre-retirement discount rate, while pushing them on deficit recovery plan and smoothing of future service.
Note that neither path (i) nor path (ii) dips into the same investment risk well twice (see Galvin above).
In the rest of this note, I’ll simply sketch path (i) to 30.7%.
The Aon paper (see p. 11) reveals that Metric A (the new Test 1) is consistent with a £5bn (rather than a £9.8bn) TP deficit, which corresponds to a gilts+4.5% pre-retirement discount rate, rather than the highest +3.5% USS appear to be willing to entertain in the consultation document. (Neither Metric B nor Metric C is sensitive to discount rate for TP or future service, so they’re not relevant here.)
Such a discount rate also gives rise to a 25.1% future service contribution rate, according to p. 11 of the Aon paper.
Based on some preliminary calculations, it appears that 15 years with no assumed investment outperformance of that discount rate would be sufficient to get DRCs to pay off the £5bn deficit down to 5.6%, and hence within the overall 30.7% contribution rate.
Under path (i), we emphasise that we are not, therefore, dipping into the deficit recovery investment risk well, since we're assuming no outperformance if the discount rate. (Nor are we calling for the smoothing of future service contributions.) Rather, we’re simply making a case for the same pre-retirement discount rate of gilts+4.5% for TP, DRCs, and future service.
In making this case, we note (see above) that gilts+4.5% is within the limits of USS’s risk metrics.
We also note, as USS notes, that gilts+4.5% is at the current 67% level of prudence that USS employs, when applied to a 55% growth portfolio under the new dual discount rate.
Hence, USS is able to get down to their preferred upper limit of gilts+3.5% only by ramping up prudence from the current 67% to 78%, even though such ramping up is not called for by the new risk metrics.
In condemning such ramping up of prudence, we can draw on the following Aon commentary:
The concept of taking a more prudent approach to confidence intervals is supported by the Pensions Regulator in its letter to the VMDF (page 50), and we were interested to see this comment as it is not obviously in keeping with more general guidance to industry through the 2020 Annual Funding Statement. We are not convinced that a higher percentile approach is justified by there being greater uncertainty about future investment returns (since this should be reflected in the modelling through a greater assumed volatility of returns). However, increasing the percentile would in our view be justifiable if there is less certainty about the future employer covenant.
The following Aon commentary is also relevant:
The JEP provided three illustrative discount rates: Gilts+2.5%, Gilts+3.0%, and Gilts+3.5% p.a. At the time, it was noted that Gilts+2.5% gave a similar deficit to the 2018 valuation approach (and a lower future service rate). On the face of it, these are the three rates illustrated by the Trustee for the Strong covenant case.
However, the JEP suggested that the pre-retirement discount rate is expressed as a CPI fixed margin, rather than relative to gilts, given that growth assets are held notionally for this part of the liabilities (rather than gilts or bonds). Since the JEP report was published on 13 December 2019, gilt yields have fallen by about 0.5% p.a. relative to CPI. In our view, the illustrative JEP discount rates would increase by around 0.5% (so Gilts+2.5% would become Gilts+3% etc.) when updated using the CPI+ approach. (Our expectations for growth asset returns over government gilts have also increased between 2018 and 2020 – on our assumptions, by over 1% p.a. So, the CPI+ approach is supported by how return expectations have evolved.)
If the Gilts+3.5% pre-retirement discount rate is adopted, then we would see this as being towards the middle of the JEP range (updated using the CPI+ approach). This would lead to a cost of new benefits of 29.4% of pay, and a corresponding deficit of £9.8Bn.
According to Table 8.2 of the USS consultation document, a best estimate of Gilts+4.47% is equivalent to CPI+2.77%. Assuming that this Gilts-CPI equivalence still holds in the case of a 67% chance of achieving this return, the CPI+ equivalent of a Gilts+4.5% pre-retirement discount rate would be CPI+2.8% or thereabouts. It might be useful to note that this is the CPI+ pre-retirement discount rate that would be assumed, on the above proposal for retaining contributions at 30.7%.
Appendix 2
On 2020 TP Consultation: Adopt CPI+ assumption, risk management based on economically relevant metric, and self-sufficiency by investing growth assets
By Woon Wong, Reader in Financial Economics, Cardiff Business School
5 October 2020 [shared with permission via UCU pensions email list]
Slides to accompany this appendix also shared with permission. [PDF slides]
A very different, positive message
In sharp contrast to the 2020 TP consultation (the “Consultation”), a contribution at 26% of payroll is sufficient to fund USS’s liabilities. Moreover, if the scheme continues to invest significantly in growth assets, its funding ratio will continue to rise toward self-sufficiency at which point little reliance is required of the employers. While it may not be a statutory requirement, self-sufficiency achieved by investing growth assets has several benefits in the waiting for stakeholders: a future contribution lower than 26% of payroll, less volatile contributions which make financial budgeting easier, a cost- effective pension few other sectors or countries can rival thereby raising the competitiveness of the higher educator sector.
The rest of this article is organised as follow. Section 1 critiques the valuation assumptions presented in the Consultation. Section 2 provides evidence for the optimistic outlook on USS. Section 3 briefly discusses risk management. Finally, Section 4 points out that legal action may be required to rid the pension industry of the practice that has resulted in closures of defined benefit (DB) schemes in this country.
Section 1
Gilts aligned assumptions are the key problem
USS’s valuation assumptions are highly aligned with gilts. However, such assumptions have little economic relevance to USS because (a) there is no plan to increase the amount of riskless assets held by the scheme; (b) there is no Test 1 to impose a hypothetical ‘de-risking’ valuation; (c) there is little correlation between gilt yields and future asset returns; and (d) self-sufficiency (SS) can be achieved by accumulation of surplus assets rather than by a low risk portfolio.
Point (d) is particularly relevant since lowering gilt yields by quantitative easing (QE) is now an established monetary policy tool. Without QE, gilt yields would have been significantly higher and USS’s liability significantly lower, though the economic recovery could take longer. With QE, the economy is helped by low interest rates and yields on productive assets remain high.
Gilts aligned assumptions are the key problem of the Consultation that needs to be resolved; otherwise, the current low interest rates brought by QE will render USS’s DB more expensive than the defined contribution (DC) alternative.
Examples are given below on how gilts aligned assumptions in the Consultation have impacted the technical provisions.
The self-sufficiency (SS) liability and discount rates
Gilts-plus discount rate is used to obtain the SS liability, which has been driven to very high values because of QE. The high value of SS liability renders USS’s tending-to-strong covenant inadequate, thereby forcing a lower discount rate to be set under the risk management framework. According to page 26 of the Consultation, the discount rates are lower by 0.2% and 1.0% for the post- and pre- retirement liabilities, respectively. The problem with the described approach is that the SS liability is a hypothetical construct that has little economic relevance to USS; see points (a) – (d) above.
The dual discount rates (DDR)
Pre-retirement discount rate: The JEP report suggested the pre-retirement discount rate be expressed as CPI plus a fixed margin and provided equivalent gilts-plus discount rate for illustrative purpose. Since its publication, gilt yields have fallen by about 0.5% relative to CPI, implying that the discount rate should increase by 0.5% if expressed in gilts-plus. The Consultation has adopted the JEP’s illustrative gilts-plus discount rate without increment.
Post-retirement discount rate: The Consultation follows the JEP’s recommendation by setting the post-retirement discount equal to a gilts-plus self-sufficiency discount rate. Given the problems with gilts-plus discount rate explained above, the CPI-plus approach should be adopted, with the fixed margin high enough to consider that retired members’ risk can be shared by active members.
Volatile gilts-aligned liability
Adopting gilts-aligned assumptions effectively mark the value of liability to the gilts market. The result is high volatility in USS’s liability. To illustrate, Table G1 in the Consultation suggests that a fall of less than 1% in gilt yield increases the liability by £14.9bn between Mar-2018 and Mar-2020. If similar liability-gilt yield relationship were assumed, then March of 2020 would witness USS’s liability fell by c. £13bn around middle of the month, and then rose by c. £9bn by end of the month after Bank of England announced to re-start its QE programme to purchase gilts.
There is no economic relevance in the wildly fluctuating liability because only a tiny fraction of the liability is required to be paid off during Mar-2020. Since DB is long term by its nature, discount rate should be set relative to a more stable variable such as expected inflation or GDP growth rate that is representative of the returns of USS’s assets.
USS and market practice
USS justifies the low level of discount rates by comparing them to the market practice. Specifically, Chart F1 in the Consultation shows that the 2020 TP discount rate is in the upper quartile of distribution of discount rates assumed by UK defined benefit schemes since 2005, suggesting that a high value of discount rate is being set. There are two problems in such comparison.
First, many of the pension schemes in Chart F1 are funded by a single employer, are either closed to new members or closed to benefit accrual entirely. Since such schemes require more cautious funding, it is only natural that USS should occupy in the upper range of the distribution, if not higher.
Second, the discount rates in Chart F1 are expressed relative to gilt yields which have fallen by about 3.8% from 4.6% in March 2005 to 0.8% in March 2020. In view of the reasons explained earlier, the discount rates in the Consultation do not reflect the required adjustment for the fall in gilt yields.
High confidence level of prudence
There is no evidence to support the high confidence level of prudence used (73% to 85% in the Consultation versus 67% in the 2018 valuation). A stochastic simulation study finds that the 1918 influenza pandemic has little impact on the price of prudence. Indeed, it may be argued that the low asset value in March 2020 has already priced in the increased level of uncertainty due to COVID-19. Using a significantly higher confidence level to price prudence on top of an already depressed market may be regarded as a double counting of risk.
Adopt CPI+, drop gilts+
Recommend replace gilts+ assumptions by CPI+, for the latter is supported by economic rationales.
Section 2
Falling break-even discount rate
In Table 1, we can see that the realised growth rate (column A) of USS’s assets are considerably higher than the discount rate (column B) of past valuations, which implies that “extra” assets (on top of what is predicted by the discount rate) have been added to the scheme. Since these extra assets help pay for the liabilities, the break-even discount rate (the “BEDR”, column C), which is the required rate of returns on USS’s assets to pay for the liabilities, falls from over CPI+4.1% in 2011 to approximately CPI+1.1% when assets are valued at £66.5bn in March 2020. The market happens to be at its bottom in March 2020. Since then it has rebounded and USS’s assets are updated to worth £75.3bn, implying a lower BEDR equal approximately CPI+0.6%.
Table 1
Note: £75.3bn (in square brackets) is the latest update of its asset value provided by USS, which implies the £60bn assets in 2017 grew at a rate of 7.6% (in square brackets in column A) to reach the updated asset value with a BEDR of CPI+0.6% (in square brackets in column C).
Expressed in real term, both the world economy growth rate and world equities returns are 2%-3% and 4%-5% respectively, which are considerably higher than the BEDR of 1.1% (or 0.6% based on latest asset value). Since USS invests significantly in growth assets, a contribution of 26% of payroll is sufficient to pay for USS’s liabilities, after allowing for prudence [1].
Rising funding ratio
If USS continues to invest in significant amount of growth assets, its funding ratio would be on a rising path. This is because the prudence (in setting the discount rate) produces surplus assets, which in turn will grow by compounding each year thereby raising the funding ratio further.
Some members may not be interested in the long-term objective of self-sufficiency but want to turn all the surplus assets into lower contributions or higher benefits at each valuation. There are two drawbacks in this approach: (1) it implies a higher funding cost in the future when the scheme matures; and (2) if there were a large deficit, high recovery contributions might be needed and benefits could be reduced. Therefore, subject to agreement by stakeholders and that contributions remain at a reasonable level (say 26% of payroll), the surplus assets may be kept in a contingent support vehicle for the scheme until there is sufficient buffer of surplus assets being accumulated.
There are several advantages to USS if its funding ratio is on a rising path with increasing surplus assets. First, the surplus assets will act as a buffer to absorb investment risk, reduce chances of a need for deficit recovery contribution and make financial budgeting easier. Second, the presence of surplus assets to absorb risk allows continual significant investment of growth assets even as the scheme matures. Third, as the funding ratio rises, the accumulation of surplus assets will reach a point that contributions will have to fall and/or benefits improve. Finally, in the long-term, a cost-effective well-funded pension scheme will give the higher education sector a competitive edge other sectors and countries find it hard to rival.
Section 3
Risk management
Given the protracted low interest rates, the aim of a low risk self-sufficiency (SS) portfolio will make USS (the DB section) more expensive than a DC. On the other hand, since there is no plan for de- risking, the risk management framework (the “RMF”) in the Consultation effectively makes the price of risk management more expensive than risk itself. The problem lies in the gilts-plus SS liability (on which RMF is predicated) is measuring ‘risk’ that is irrelevant to USS.
The truth is that the DB scheme of USS has the best design of all financial structures to deal with risks. Even if its covenant were less than tending-to-strong and its portfolio invested more in growth assets, the open and immature with last-man-standing multi-sponsors structure allows USS to effectively mitigate investment risks including those due to the current pandemic.
The risk management framework in the Consultation may be useful if the SS liability is calculated using a CPI-plus discount rate. Also, a stochastic simulation study to determine an expected shortfall of payment for the promised benefits by funds from contributions and asset-generated cash flows may be carried out. The expected shortfall may then be compared with the employer covenant.
Section 4
Despite DB being superior to DC in terms of risk diversification and hence in principle more cost-effective, pension provision by the former has declined drastically in the past few decades. As a result, pensioners have become poorer; for employers, pension provision has become more expensive. It is a sign that both the Pension Regulator and the pension industry have failed in their role to the good of society.
The Consultation illustrates many of the reasons that make DB unnecessarily expensive so that DC becomes a cheaper alternative, at a loss to both employees and employers. The JEP, Valuation Methodology Discussion Forum and several other working groups (e.g. on transparency, risk, etc.) have made progress, but not enough.
The works of Academics for Pensions Justice find there is a case of breach of trust against USS. There is no ill feeling against any party, but legal action may be required to change some deep-seated traditions and ideology that have resulted in closures of many DB schemes [2]. Others advocate members to take over the scheme so that valuation can be carried out not only in compliance with the technical actuarial standards, but also enjoy the same rigour of peer-review system in academic publication – all relevant data, assumptions and details of any stochastic models are made publicly available for stakeholders to examine and duplicate.
1. Note that although contributions have gone up from 26% of payroll since 2018, the impact on the final asset value is small.
2. Note that prudence lowers the best estimate return by roughly 1%-1.2% to arrive at the discount rate in 2014 and 2017 in Table 1. The large difference between the realised growth rates and the discount rates suggests that past deficits of USS are due to excessive prudence. Indeed 2017 liability would be lower if 2014 gilts+ assumption were used. Gilts+ assumption overstates liability when interest rates fall.