Large schemes 2021

Ninth annual report covering large UK pension schemes

Increase in pension schemes with over £1bn in assets

Since our first report back in 2013 there has been a 90% increase in the number of pension schemes with assets over £1bn, despite the number of schemes that have been closed to new entrants or future accrual over that period. So there are a lot more schemes and employers with more financially significant pension challenges in an environment where there is about to be a step change in regulatory expectations on governance and risk management, especially for the largest schemes.

Increase in the number of pension schemes with assets over £1bn

We’d expect this trend of increasing numbers of ‘very large’ schemes to continue into the medium term. Many schemes are still some years off reaching a cashflow negative position (i.e. paying out more than the return on the investments plus contributions) and increasingly focused regulatory guidance will see the funding targets of many schemes increase. Furthermore, while the ultimate goal for many of these schemes will be to buy-out with an insurer, the large size (and often complex nature) of these schemes may put a limit on the number of such transactions that could occur in any one year without a significant expansion of the current market.

So it is increasingly important for each of these large schemes to clearly carve out and effectively execute their own path through the ‘crowd’ and they can only do this if they ensure they have the support and advice which prioritises their specific individual needs.

Methodology: The report is based on publicly available data with year ends up to 31 March 2021. The survey covers nearly 300 schemes, with assets totalling around £1.4 trillion. Not all schemes are included in all sections of our survey results, which depend on the data available.

Trends for remaining open or closing

In 2013, around 15% of schemes included in the survey were closed to accrual. In 2021 this has increased to more than half of schemes being closed to accrual.

While the trend towards schemes closing over time is nothing new, it is perhaps notable that the position for the 2021 survey is broadly unchanged compared to the 2020 survey. This is perhaps not surprising given that the Covid-19 pandemic will have meant that management time will inevitably have been focused on other priorities. Furthermore, employer affordability of contributions to pension schemes in the short term will have been supported by TPR’s flexibility and government interventions to support businesses through the pandemic. It may also be indicative of the relative strength of sponsors supporting the larger schemes relative to the UK average.

As the Covid-19 pandemic passes, with perhaps more management time available and inflationary pressures on the rise, it will be interesting to see whether there is ‘pent up’ demand for scheme closures and whether there will be an acceleration in these over the coming years.

The outlook for wage inflation relative to price inflation is likely to be a key consideration for companies continuing to operate an open DB scheme. Some employers may have benefited from expected salary increases for active members being below the price inflation linked increases which would apply if the scheme was closed to accrual – if this were to reverse it may increase the pressure for closure to accrual. Other schemes may have caps on the salary increases which count for pension purposes – in these cases the cost of closure in relation to past service accrual may increase significantly, with the higher outlook for inflation increasing the relative value of the leaving service benefit.      

Evidence from our dataset suggests there is no discernible difference in the funding level of schemes closing to new members recently relative to those that have been closed for some time. Indeed, where scheme closure date is included in the data, the average funding level of schemes closing to new members in the last 15 years is identical to the average funding level of schemes that closed to new members over 15 years ago. So it is not simply the case that resources which have been freed up from closure to new entrants have been available to improve the past service funding position. 

De-risking investment strategies

The chart below shows the total allocation to growth and protection assets across the big schemes and how this has progressed over the past six years.

Asset split over time

There has been a sustained trend among schemes towards de-risking investment strategies for some time.  There is some evidence for an acceleration in de-risking from 2019 which has continued into the 2021 dataset. However, despite general reductions in investment risk, there is still a high degree of variability in the range of returns that are being achieved by schemes.

The charts below illustrate the 1-year investment performance for relevant schemes where the data is disclosed with the years ending 31 March 2020 and 31 December 2020.  

1-year investment performance to 31 December 2020

1-year investment performance to 31 March 2020

These show a range of 12% pa from the highest returning to the lowest returning portfolios. Although the 31 March 2020 end date corresponds to the exceptional period around the start of the Covid-19 pandemic, the range is not dissimilar at 31 December 2020 by which time there had been a substantial recovery in the markets. 

It is clear that while there has been a substantial move to protection assets in general over recent years, there is still significant variability in investment outcomes. Some of this will reflect the fact that liability matching assets will differ between different schemes, for example due to different maturity profiles and levels of inflation linkage in the benefits, but the range is nevertheless still significant.

This highlights the importance of ensuring that the risks inherent in some of the increasingly complex investment strategies are fully understood and therefore the importance of independently reviewing investment/fiduciary manager performance – both against scheme targets and external benchmarks. 

Drilling down into the allocations, the charts below illustrate the broad split of assets held for both the largest schemes with over £5bn in assets and those with assets in the range £1bn to £5bn.

Asset allocation - above £5bn

Asset allocation - £1bn to £5bn

For the largest schemes, over 50% of assets are now held in bonds, with the figure around 40% for the £1bn to £5bn range.

The largest schemes have higher allocations to property.

The £1bn to £5bn schemes have a greater proportion in ‘non-traditional’ asset classes ('Other' in the chart above), although the overall split between growth and protection assets is broadly similar.

Our data shows that 23 schemes in the 2021 survey now have at least 25% of their assets in a buy-in policy and we expect this number to increase significantly in the next few years as schemes implement their endgame strategies.

Increasing focus on end game planning

With the long trailed introduction of the revised code of practice on DB funding expected next year, the industry is increasingly focusing on planning for the end game – whatever this may look like on a scheme-specific basis.  

Barnett Waddingham has recently introduced the DB End Gauge index – a monthly index estimating the average time for UK pension schemes to reach a sufficient level of funding to buyout their liabilities with an insurance company. At the end of September 2021, this reflected an average period of 10.2 years. Consistent analysis for schemes with over £1bn in assets suggests a slightly longer average period to buyout of 10.8 years, although the average for schemes with assets of more than £5bn is 9.9 years. 

A total of 15% of the over £5bn schemes are estimated to have an average time to buyout of less than two years. While there are many different endgame options, and buyout may not be the destination for all of these schemes, this does illustrate how well funded some of these schemes are. Such strong funding positions are likely to provide trustees with more options as they consider the endgame and this highlights the need for a clear plan and system of monitoring to ensure the best outcomes for members are realised as soon as possible.


All over £1bn

All over £5bn

While the above commentary covers the average period to buyout, in reality there are a number of risks which will undoubtedly affect the timescales which might be achieved in practice by individual schemes. A prominent example at the current time is the outlook for future inflation. Over the last year, long-term inflation expectations have risen significantly. We estimate that the total increase in liability as a result of changing inflation expectations over the last year for the schemes included in our survey alone is of the order of £100bn. While this is expected to be substantially offset by a corresponding increase in inflation hedging assets which are held by these schemes, most schemes are unlikely to be fully hedged against inflation movements, certainly as far as full buyout costs are concerned.

Case Study: the Old British Steel Pension Scheme

In October 2020 the Old British Steel Pension Scheme completed a £2bn transaction with PIC, securing member benefits at or above PPF levels of compensation. Barnett Waddingham provided transaction, actuarial, investment and administration consultancy to support the Trustee in achieving the best result for members.

Large schemes often have complex legacy arrangements and in such cases it is vital for administrators, actuaries and transaction specialists to be working collaboratively to ensure nimble, pragmatic, cost effective and successful decision making. As part of the transaction, Barnett Waddingham advised on and implemented complex benefit rectification activities including GMP equalisation across over 20 distinct benefit sections, equalisation of normal retirement ages, and share of fund calculations (including allowance for recent court judgements on PPF benefits), as well as running a member option exercise for those with small benefits in the scheme which included more than 12,000 eligible members.

TPR and ‘significant maturity’

In its consultation on the revised code, TPR comments that “an investment portfolio with a higher level of growth assets than 20% is unlikely to have a prudent expected return consistent with the low dependency funding basis at significant maturity” with significant maturity defined as a duration of 14-12 years.

The chart below illustrates the estimated starting position (in terms of both asset allocation and estimated duration) for a number of large schemes relative to TPR’s indicative low dependency benchmark.

This analysis shows that there is a large range of starting points and therefore varying degrees of work required to navigate the endgame successfully.

Almost half of large schemes already have growth assets below TPR’s notional 20% threshold, despite many being some years off “significant maturity”.

This suggests that many large schemes are well progressed in their thinking on end-game strategy, but given the average period of a decade before buyout might be reached as measured by the DB End Gauge index, this highlights the importance of balancing the need to make further progress whilst minimising the risk of going backwards. Monitoring the right metrics for each scheme circumstances will therefore be key. Increasingly trustees will want to monitor a range of more sophisticated metrics to successfully navigate their end game journey. It will be less important to simply keep an eye on the funding level, but instead other metrics will rise in prominence, for example: the projected length of any journey plan, the required return to achieve the desired end point, or the probability of reaching a given point at a given time. There will be a particular need for clear advice in presenting and interpreting such metrics and on the actions to be taken when certain thresholds might be breached.

Barnett Waddingham’s DB Navigator framework provides a bespoke and holistic Actuarial, Investment and Administration approach to navigating your DB endgame, providing the knowledge, structure and tools to set, monitor and land your endgame objectives. 

Decreasing transfer activity

In our previous analysis on large schemes we identified the slowing up in transfer activity among the UK’s largest schemes, with 60% of schemes showing a decrease in transfer activity in the 2020 survey. This trend has continued with 70% of schemes in the 2021 survey showing a further decrease in amounts transferred compared to the 2020 data.

Schemes showing a decrease in transfer activity


There are many well documented reasons for the general trend in decreasing transfer activity, including pressures on IFAs, concerns around scams, the difficulty of accessing suitable advice and general lack of management time over the period of the pandemic to facilitate bulk exercises. However, given the period of low gilt yields which would be expected to lead to relatively high transfer value quotations compared to other historic periods, the extent of the decline in amounts transferred is perhaps surprising.

Based on schemes where information is available for each of the last three years, on average, around 1% of the liabilities of large DB schemes are extinguished via the payment of transfer values each year. Transfers remain a helpful part of the toolkit for managing risk and settling liabilities while also providing flexible options to members. Increasingly, to facilitate the use of such options, we see schemes partnering with IFAs to provide advice and protection for members to support transfer activity, for example as part of a routine at-retirement option.

Funding ‘very robust’ against economic turbulence

While in many ways, the analysis from the 2021 survey shows a steady continuation of established trends, this is perhaps notable in itself given the nature of the period in question. It has been well commentated on elsewhere that the funding of large pension schemes appears on average to have been very robust to the turbulent economic markets of the past 18 months. 

Although the Covid-19 pandemic does not appear to have set back these pension schemes, many schemes still have a long way to go to get to their endgames. The demand for de-risking actions will continue to accelerate and schemes will need to be more proactive than ever in getting themselves to the ‘front of the queue’ - whether that is for IFA support on liability management exercises, with insurers for buy-in / buy-out policies or with sponsors for management time. These larger schemes will need a team of trustees and advisers who can work together to really prioritise their needs.

Barnett Waddingham is well placed to provide actuarial, transaction, investment and administration consultancy with our independent expertise, collaborative approach and proven track record. Please get in touch if you would like to discuss further how we can add value for your scheme.

Get in touch

Ali Tayyebi

Contact Ali

David Simpson
Associate and Senior Consulting Actuary

Contact David

DB End Gauge

Our monthly index provides an estimate of the average time for UK pension schemes to reach a sufficient level of funding to buyout their liabilities with an insurance company.

Latest Analysis

DB Navigator: the framework

Packed with case studies to help you navigate each confluence, takeaways to steer you on the right course and videos to guide your DB pension scheme strategically through the choppy waters of endgame planning.

View interactive framework

FM Evaluate

Providing independent evaluation and oversight of fiduciary management services

Find out more