Skip to content
The UK's only independent think tank devoted to higher education.

University pensions: ‘If something cannot go on forever, it will stop.’ PART III: WHAT IS LIKELY TO HAPPEN

  • 11 February 2021
  • By Nick Hillman

This is the third in a series of three blogs that HEPI is running this week on the state of university pensions and particularly the Universities Superannuation Scheme (USS). The previous two blogs are available here and here.

However tough the predicament faced by any pension scheme, as the title of this trilogy of blogs asserts, ‘If something cannot go on forever, it will stop.’ This is named Stein’s Law after the economist Herbert Stein.

It is clear the current USS arrangements with a large and increasing deficit cannot go on forever. So something will have to give. My guess is that, in the end, more will need to come from the benefits side of the equation than has so far been widely accepted, even though (as yesterday’s blog showed, this could be very difficult to achieve).

The decline of defined benefit pensions

This is what has happened elsewhere in society. Pension schemes with guaranteed benefits are now rare beyond the public sector.

When I worked on pensions almost a generation ago, we used to roll our eyes at the number of press releases from the pensions sector that claimed some specific change or another was ‘the final nail in the coffin of defined benefit pension schemes’.

The argument was typically overdone when applied to any individual tweak, but cumulatively lots of individual costly changes (some wise, some less so) gradually served to kill off defined benefit pension provision.

In 1994, every FTSE-100 company ran a defined benefit pension scheme and it is thought ‘all of these were open to new employees’. In 2012, the Financial Times reported that ‘Shell is to close the FTSE 100’s last remaining final salary pension scheme’ (a generous form of defined benefit scheme that was once common).

By 2015, there were only five FTSE-100 companies with any form of defined benefit scheme open to new members of staff. Many companies are now looking to offload their defined benefit schemes altogether, with one pre-COVID headline reading: ‘Half of FTSE 100 DB pension schemes could buyout by 2028’.

A few years ago, it was proposed that the USS should close its defined benefits section too, which could have stemmed the mounting problems, though it prompted ‘the longest ever strike in UK higher-education history.’

The Pensions Regulator and the deficit

The second reason why the benefits side of the equation is likely to be affected is that the deficit remains large whatever happens. This is because you can only alter pension benefits for future service; you cannot alter the pension entitlement accrued to date by the 200,000+ existing members, nor the accrued entitlements of the 180,000 deferred members and the 75,000 retired members.

It is just not possible for institutional leaders and governors to turn their heads and ignore the problem the way that some of the organisations in Lowenstein’s book While America Aged (which prompted these blogs) were able to as the Pensions Regulator will keep twisting their heads back to focus on the problems.

Many years ago, my job included lobbying the powers that be on behalf of the insurance industry over the extraordinary number of decimal places they expected to be used when calculating elements of people’s personal pension funds (which was greater than the functionality of insurance companies’ IT systems at the time). The idea that such an intrusive regulatory regime could now be relaxed about a deficit of perhaps as much as £23,900,000,000 in just one pension scheme is for the birds. (To those shouting at me on social media in response to this series of blogs, including those who have claimed the USS deficit is entirely illusory, I would point out that it is the Regulator you need to convince of your case.)

Could the Government absorb the problem?

Some opponents of change might draw parallels with student funding and its reliance on loan-based finance. If it is okay to fund today’s education of students by appearing to mortgage their futures, why can’t pensions be funded the same way?

Some people have even argued that the best way to resolve the USS challenge is for the Government to underwrite the Scheme, just as it takes on the burden of unpaid student loans.

But student loans fund education that helps people become more productive (as well as better citizens). So the benefits hugely outweigh the costs on average. The extra taxes graduates will pay as a result of being more highly educated, combined with their student loan repayments, more than cover the costs of their extra education.

In contrast, putting the USS on the public books could prove a significant burden on taxpayers, most of whom can only dream of benefits like those on offer from the USS. No responsible government is likely to do it.

Moreover, if the Treasury ever really did have considerable amounts of spare cash to spend on future pensioners, then more disadvantaged people than retired university staff members may have a better claim – and also look a surer electoral bet.

Expecting taxpayers to absorb the cost of past pension errors in our sector would also bring huge moral hazard to the fore: why should any private sector scheme ever act responsibly if it believed the Treasury will pick up the bills if things don’t work out?


These three blogs flag problems more than provide solutions and they therefore risk breaching HEPI’s own rules, which state ‘a publication that criticises without proposing better alternatives is unlikely to be published by HEPI’.

This is intentional, given the point the debate over the USS is currently at: the primary purpose of these blogs is to try and ensure discussion about the USS is not lost in the all-encompassing COVID-created fog that is leaving less room than usual for other issues.

However, as implied above, it seems clear further change needs to happen that goes beyond ideas like changing modelling assumptions on the real state of the USS, hoping the Government will step in or trying to ignore pressure from the Pensions Regulator.

Personally, I would keep a shift to a wholly defined contribution scheme on the table, as that is what has already happened throughout the private and charitable sectors. It could be attractive to the material proportion of younger staff who refuse to join the USS – and may make it easier for institutions to offer more permanent employment, rather than insecure contracts, to early career staff. But it would be a major shift and likely cause further industrial action. Moreover, if the current defined benefit part of the USS were closed, many people think it would need to be further derisked, prompting even bigger contributions. So this is not a perfect solution. (Notably, there are many types of DC schemes and one long-time critic of the USS, John Ralfe, is delivering a free lecture on so-called Collective Defined Contribution schemes at Imperial College Business School later this month.)

Another less radical idea would be to retain the current defined benefit arrangement but to lower the salary cap below its current level of almost £60,000. Salaries above this threshold are still pensionable, just on a defined contribution basis.

I explained in a history of the USS that HEPI published in 2019 that the original motivation for a defined benefit scheme for university staff came from the top. It is clear that many senior managers are keen to maintain some defined benefit provision. But it needs to be affordable at a time when university incomes are being squeezed.

This is the final of three blogs run the day before yesterday, yesterday and today.

HEPI’s previous output on the challenges faced by the Universities Superannuation Scheme includes a history of the Scheme from its foundation in the mid-1970s, a piece on why the COVID pandemic is likely to lead to pension reforms and a response to some of our commentary on the USS by Matt Waddup of the UCU.


  1. 1. “It could be attractive to the material proportion of younger staff who refuse to join the USS – and may make it easier for institutions to offer more permanent employment, rather than insecure contracts, to early career staff.”

    Rather than making a series of assertions, it would have been useful if you had bothered to engage in the details of this debate. It’s widely accepted that the impact of the changes you’re advocating for (shifting to DC) will have a larger impact on younger members of staff. Using the USS’s assumptions, we’ve created a modeller that you can use to estimate the impact on benefits and contributions for staff members of different ages here: I would encourage those of you who are concerned about your pension to see how the USS proposals affect you – we will update as we get more information.

    What is clear from these models is that the shift from DB to DC imposes huge losses on younger members of staff. A 25-year-old staff member on £35k would have lost £366k under UUK’s 2018 proposal to shift to DC. The USS 2020 proposal would require contribution increases of £258k. So it’s slightly unclear to me why cutting younger members of staff’s benefits or increasing their contributions by hundreds of thousands of pounds would be attractive to anyone, let alone younger members of staff.

    2. Re the government underwriting the scheme – this is unnecessary, even under the assumptions deemed prudent and acceptable in 2018, the scheme is unlikely to have a deficit. Since the valuation date of March 2020, the scheme’s assets have increased by over £14bn. The USS Trustee is proposing changes to the valuation from 2018 assumptions that increase the scheme’s deficit by £7.4bn to £9.5bn. The deficit you cite above is from the scheme’s financial management plan that requires assuming a discount rate of -0.7% after inflation – i.e. that we expect the scheme’s cumulative return over the next 30 years to be -19%. This is possible, but is it likely? Has the USS provided any credible evidence to support these claims?

    3. “if the current defined benefit part of the USS were closed, many people think it would need to be further derisked, prompting even bigger contributions” When the DB scheme is open, it is cashflow positive and will not need any of its assets for 30+ years, therefore its investment time horizon is very long, and the scheme can invest in diverse and productive assets. Whereas you are advocating for closing the DB scheme, which will force it to divest its assets rapidly. This will require the scheme to divest its productive assets (e.g. investments in companies), and shift to investing primarily in government bonds or similar. Government bonds currently yield -2.2% after RPI, i.e. for every £1 we invest today, we can expect to get 49p back in 2050. It is not possible to fund a DB scheme using government bonds, and attempting to do so, will likely leave the scheme underfunded. The push from tPR and others to force DB schemes to invest in bonds or similar is a form of financial repression and stealth taxation and is one of the primary reasons for the closure of DB schemes you describe above. Why are you in favour of forcing otherwise viable schemes like the USS to the making loss-making loans to the government, rather than making productive investments in the economy?

    It’s great to see people discussing the USS valuation. But in order to meaningfully contribute, you do have to engage with the details. What discount rate do you think is plausible? What “level of prudence” should the scheme be using? How should the USS Trustee account for experience? What evidence would be sufficient to justify the changes the USS is looking to impose?

    To paraphrase reviewer 2 “the arguments in this post are flawed, incoherent, and demonstrates a weak grasp of the theoretical and empirical literature and should be rejected without further review.”

  2. Martin Evans says:

    I am left with a sense of disappointment. I appreciate that in some institutions that USS is all that there is. But what your three articles have failed to take into account is that Universities are participating employers in Teachers Pension Scheme where employer contributions rose 44% and that isn’t over. Local Government Pension Scheme where funds vary from Greater Manchester at one end to Brent at another and the NHS pension scheme. All these schemes have issues. Some universities have taken steps to mitigate their exposure . So an interesting exercise which will hopefully educate many of your readers but disappointing in that you do not look beyond USS contrary to what your headline may state

  3. Rob Kett says:

    Starting almost 20 years ago the DB pension issue has largely been resolved in the private sector. Unfortunately, under the leadership of VCs who have significant personal interest in maintaining defined benefits, the HE sector has been unable to follow the same path. Indeed, some of their comment during previous valuations seemed ill-informed and showed a willingness to ‘kick the can down the road’, perhaps to the point when they could draw their own pension. I pity the early career academics whose contributions will be providing the funding. Perhaps it is time to take the issue out of the hands of universities’ executive management and pass it to non-executive board members who, potentially, have experience of dealing with the problem – successfully!

Leave a Reply

Your email address will not be published. Required fields are marked *