Alternative scheme design and USS: both universities and their UCU branches call for working together to explore a new scheme design for USS that could lead to a much enhanced benefit structure

  • Cambridge, Oxford and their respective UCU branches have published a joint statement calling for exploration of a new approach to USS scheme design
  • The statement calls for a fully-resourced team to be set up to look into alternative scheme designs and to report in six months
  • We believe that it should be possible to provide a good pension for 25-30% of salary (combined member and employer contributions) and that conditional indexation – a form of member risk-and reward-sharing – is one of the designs that should be examined and considered, given the constraints of the current regulatory and actuarial approach
  • Any viable conditional indexation design would need to ensure there is not the discretion for the trustee to withhold indexation out of a desire (without clear justification) to increase the level of prudence in the scheme
  • Any alternative scheme design should only be considered if extensive modelling demonstrates it is not likely to result in lower pensions for members
  • Any alternative scheme design will not be part of the solution to the 2020 valuation

We, the undersigned, consider that a contribution rate of 25-30% of salary should be sufficient to secure a good pension for staff who are members of the Universities Superannuation Scheme (USS). 

However, we believe that the current regulatory and actuarial approach to risk for traditional defined benefit schemes – including the current USS scheme – makes it difficult to obtain good value for money. Successive valuations have locked in higher contribution rates and/or reduced benefits for a scheme that can afford to operate under a very long-term investment horizon. Members see their benefits eroded on the basis that something bad – but ultimately unlikely – might happen (USS will not be able to make good on its future pension promises) but see little flexibility in the scheme to benefit from improved market conditions. Absent a change to the current regulatory and actuarial approach, we believe this can only be achieved by redesigning the scheme. Moreover, until this happens, it is difficult to see an end to the problems that USS, its members and employers, have experienced since at least 2016. The outcome of the 2020 valuation is likely to do little to address these underlying issues and break out of this cycle.

We therefore seek to build on the benefits of engagement between UUK and UCU since the last valuation via the Joint Expert Panel and Valuation Methodology Discussion Forum. Scheme design and valuation need to start with the assumption that USS will invest in the assets that will deliver the best overall return over the long term, without excessive concern for volatility along the way (much like a university endowment). This should provide the best pensions for members with the lowest overall level of contributions. We should only move away from this efficient, long-term portfolio if it can be demonstrated under plausible scenarios that the scheme will be unable to pay benefits, and then only when other mechanisms to manage the risks identified (such as affordable employer and member risk and reward sharing) have proven to be unworkable. 

At present, USS uses a standard, and overly simplistic approach to risk that may be appropriate for single employer closed schemes but is not appropriate for USS. Employers and members need to work together with a well-resourced and transparent USS management to optimise the scheme and find ways to manage risk that do not over-rely on the ‘gap to self-sufficiency’ measure. USS should focus on how the scheme would actually manage its risk in various downside scenarios. This should be overseen by a trustee board with expertise in maximising long-term returns. 

This approach to valuation is fundamentally different from the current scheme design and the recent UUK proposal. Under this alternative, scheme members would obtain better pensions in the vast majority of expected outcomes. 

We therefore reiterate the need to explore an alternative scheme design as quickly as possible to provide a good pension for 25-30% of salary. We recognise that developing a new scheme is a complex process with a number of design and regulatory issues to be addressed and the timeframe is too short to implement such a solution for the 2020 valuation. However, we think that a team mobilised rapidly to explore the options could allow any benefit reductions/increased costs that flow from the current valuation process to be in place for the minimum amount of time. 

We  therefore ask USS, UUK and UCU to put together a fully resourced team to investigate alternative scheme design based on employer and member risk and reward sharing. We briefly illustrate below, as one possible example, how this might be achieved using conditional indexation. This team should report on progress and emerging potential options within six months.
 

Michael Abberton, President, Cambridge UCU branch

Dr David Chivall, President, Oxford UCU branch

Professor Louise Richardson, Vice-Chancellor, University of Oxford

Professor Stephen J Toope, Vice-Chancellor, University of Cambridge

On 20 September 2021, Loughborough University and Loughborough UCU also published a joint call for the exploration of alternative scheme design.


Conditional indexation: how it could work

Conditional indexation is a form of pension scheme design where members share some of the risk (and reward) of changes in asset performance. This can allow pensions to be provided for a lower overall contribution level than is needed under traditional defined benefit schemes where trustees consider the extent to which the scheme itself could bear all of the risk of employer failure, and typically build extremely high levels of prudence into their calculations. 

Under  a conditional indexation scheme, new benefits would continue to accrue as now. Taking accrual rates currently in place in USS as an example, an employee earning £37,500 would receive a pension in retirement of £500 a year (1/75th) for each year worked (and £1500 lump sum on retirement).

However, the benefit would not – as it does now – automatically increase with inflation until retirement. Instead, increases will be conditional on market performance – within defined parameters that should protect members from significantly deleterious outcomes. If financial market conditions turn out to be broadly as expected, then the benefit would increase with inflation. If market conditions were materially worse than expected, the benefit would not increase at all. And if market conditions were materially better than expected, the benefit would increase by more than inflation. Once the benefit had increased by inflation, it could not be decreased again.

Conditional indexation would apply to new benefit accrual only. Benefits earned in USS to date would continue to increase with inflation according to the scheme’s existing rules. After retirement, all benefits must by law increase at the lesser of CPI inflation or 2.5% per year. Therefore, after retirement, benefits would increase by a minimum of CPI inflation or 2.5% per year, but might increase by more than that if market conditions were materially better than expected (i.e. conditional indexation adjustments would still apply, but with a ‘floor’ of 2.5%/CPI). 

Over the medium-long term, we would expect benefits to appreciate as much or more under this approach as in the current inflation-linked scheme. The investment in a diverse mix of long-term growth assets should protect against the erosion of benefits even in conditions of high market volatility such as the high-inflation 1970s. Of course, in finalizing the design of the scheme, extensive modelling should be undertaken to demonstrate that the probability of a material reduction in pension compared to CPI indexation is low .

It would also be important that scheme design ensures USS continues to invest a significant proportion of the fund in growth assets and that there is not the discretion for the trustee to withhold indexation out of a desire (without clear justification) to increase the level of prudence in the scheme.


Creative Commons License
The text in this work is licensed under a Creative Commons Attribution 4.0 International License. Images, including our videos, are Copyright ©University of Cambridge and licensors/contributors as identified.  All rights reserved. We make our image and video content available in a number of ways – as here, on our main website under its Terms and conditions, and on a range of channels including social media that permit your use and sharing of our content under their respective Terms.