Background
When we published our superfunds guidance in 2020, we committed to review it within three years in respect of whether profits could be extracted before members’ benefits are bought out in full.
The interim regulatory regime has now been in place for three years. There is currently one superfund that has completed our assessment successfully and no transactions have taken place to date.
We have also seen some material changes in the market conditions since our guidance was published in June 2020. Given this backdrop, in March 2023 we felt it appropriate to undertake targeted engagement with key external stakeholders. Our engagement included seeking input on the approach to setting the discount rate for technical provisions (TPs) and the gateway principles.
We received a number of responses from key stakeholders, including pension consultants and industry trade bodies.
We also undertook our own analysis, and commissioned and reviewed external modelling from Mercer (PDF, 10,444 KB, 32 pages) (as we did when we developed our 2020 guidance).
During the period of our review of the guidance, the Department for Work and Pensions (DWP) published its consultation response confirming that it will be legislating for superfunds and setting a clearer direction of travel for that legislative framework. However, it is likely that a permanent legislative regime will take some time to develop.
We have now published our updated superfunds guidance and our updated guidance for prospective ceding trustees and employers.
This document summarises the key issues we have considered as part of the guidance review. In particular, it sets out each area where we asked stakeholders for input, a summary of their responses, and our conclusions.
Summary of key issues raised
Overall, stakeholders agreed with the proposed areas of change, particularly in relation to the discount rate and gateway, although there were varying suggestions on the best way to achieve the proposed changes. A small number of respondents, concentrated in the insurance sector, were keen for us to maintain our original position in the interim period.
The majority of respondents felt that adopting a more dynamic approach to the discount rate would be helpful, citing that the fixed nature of the discount rate in the 2020 guidance did not enable superfunds to price on a realistic basis following market movements and that a dynamic approach was necessary to make it more attractive for new entrants to enter the market. Respondents provided a variety of suggestions for possible approaches to do this.
Most respondents were also clear that a longer time period was needed for a transaction to take place, once the gateway tests were passed. It was felt that doing this would allow time for negotiations and exclusivity arrangements with schemes, and would help reduce the likelihood of wasted time, money and effort by ceding schemes and avoid any need to re-run tests. Respondents also asked for further clarification in the guidance on the gateway process and clearance. Some respondents highlighted concerns around potential capacity in the buy-out market and that some schemes could find that the market was not accessible for them.
Finally, the ability to extract profits before members’ benefits were bought out in full was seen by some as an important feature to enable superfunds to establish in the market as commercially viable products. However, there were mixed responses as to how best to formulate a process to allow profits to be extracted in practice.
1. A dynamic approach
We asked:
Do you think it is appropriate to amend our guidance from the current 'gilts + fixed margin' to a dynamic approach for the TPs discount rate?
What advantages and/or disadvantages do you see with a dynamic approach? Are there any other approaches to setting the discount rate you would suggest are considered — if so, please describe?
You said:
Most respondents were in favour of a dynamic approach to set the discount rate. Issues they felt we should consider included creating something that works for all types of superfunds, is consistent with our defined benefit (DB) funding code and is not to the detriment of members or the Pension Protection Fund (PPF).
The advantages they identified included:
- responsiveness to market conditions
- providing more stability between superfund and insurer pricing
- removing any market distortions in the run up to a transaction
- supporting innovation and potentially making it easier for superfunds to find/keep investors
Respondents also noted that because of the current fixed discount rate and the level at which it is set, some insurers can offer cheaper pricing than superfunds, which materially undermines any potential superfunds market and the policy intent.
Despite being largely supportive of adopting a dynamic approach to the discount rate, respondents identified that a fully dynamic discount rate could be challenging from both a regulatory and governance perspective. Some believed that, depending on how it is implemented, dynamism could have undesirable consequences, with the potential for ‘herding’ around particular types of assets and additional demand being created for certain types of assets where the supply was already limited.
Those who were not in favour of dynamism suggested that, if implementing a more dynamic system, we should base it on Solvency II. They also suggested that changes to the discount rate should not be made in isolation and that a full, holistic risk assessment should be carried out and that mitigations should be implemented where necessary.
Our response:
We agree that moving to a more dynamic rate would better account for market movements, allowing superfunds to price accordingly and remain competitive. While we may draw on some particular elements of Solvency II, the DWP is clear in its consultation response that it wishes to base the legislative regime on 'a pensions-based funding approach'. Our interim approach mirrors the DWP’s intent.
While we felt that gilts + 0.5% per annum was appropriate when we published the interim regime in June 2020, it is evident that, following market movements, it is now too low to allow the superfund market to develop. Following receipt of some updated modelling analysis from Mercer, we have decided to change the discount rate to gilts + 0.75% per annum with immediate effect. We believe this increase will be sufficient in the current market to allow superfunds to price appropriately, while ensuring that security of savers is not unduly compromised.
As can be seen in pages 15 to 23 of the Mercer analysis, we recognise that the probability of meeting benefits (POMB) for superfunds running on at just this level of funding (ie not including the capital buffer) is reduced relative to gilts + 0.5% and the modelling we considered for the 2020 guidance. However, this new discount rate when taken together with the capital buffer, means the POMB remains very high and in the region of 99% based on the investment strategies we have modelled (which is not materially different to our 2020 modelling). Any superfunds that fall to this level of funding may also look to adjust their investment approach to maximise the POMB — as highlighted in pages 4 and 5 — meaning that the POMB may not be reduced relative to our 2020 modelling.
Further Mercer’s analysis also highlighted how, without this change, the superfund market is unlikely to be able to price at a level below buy-out. It will therefore not enable schemes that can’t afford buy-out to benefit from moving to a superfund.
We accept that developing and adopting a fully dynamic approach would have the potential to introduce significant complexity and significant challenges from a regulatory and governance perspective. In anticipation of a longer-term legislative regime being established by the DWP, we believe that implementing a fully dynamic regime in the interim may not be appropriate.
However, to enable the discount rate to respond to material market developments, including pricing in the buy-out market, we will be implementing a review system. Future reviews of the discount rate will be triggered when either:
- the PPF changes its s179 basis or s143 basis; or
- an index of credit spreads, smoothed over rolling six-month periods, falls outside pre-set ranges
In deciding whether to change the discount rate following a review, we will consider whether superfunds continue to be able to price appropriately in the buy-out market and balance this with ensuring security of savers remains appropriate. We will change the discount rate only in 0.25% per annum increments to limit the burden of frequent changes. We believe this approach will provide us with enough flexibility to be able to respond to material market movements and will help to facilitate the development of the superfund market.
Its consultation response has indicated 'recent modelling suggests that up to gilts plus 1% may be appropriate under recent market conditions'. While we await further details of how the discount rate in the legislative regime will be set, we are confident that currently gilts + 0.75% strikes an appropriate balance between member security and superfund development, while being consistent with the DWP’s direction of travel.
We have commented on how the discount rate (and review mechanism) we have set fits with our DB funding code under question 2, below.
2. Suggestion on implementing a dynamic process
We asked:
If yes to question 1: Do you have any suggestions for how a dynamic process — that is practicable to implement (for a superfund) and to define and monitor (for TPR) — could operate in practice?
For example, if yes, why and to what level? What evidence do you have to support this? How frequently should it be reviewed and what circumstances should trigger a further review by TPR?
You said:
Most respondents provided suggestions on how a dynamic process could work, including:
- basing it on Solvency II (matching adjustment), but not as stringent or complex
- ensuring it is consistent with our forthcoming DB funding code
- setting an individual discount rate based on superfunds’ own assets, therefore allowing each superfund to set out its discount rate and then demonstrate to us that it is appropriate
- allowing for a minimum and maximum discount rate (suggested range 0.75% to 1.5%), with the leeway to adjust to reflect the composition of a reference portfolio
Regarding the frequency and triggering of reviews, some respondents suggested it should be linked to the PPF valuation assumptions or recalibrated based on the 2023 PPF valuation assumption consultation. It was also suggested that industry should be allowed to participate in any review.
Our response:
As stated under question 1, we acknowledge the complexities in implementing a truly dynamic discount rate and question whether it would be appropriate to do so at this time while we await further details on how this might be developed as part of the longer-term legislation. We are also aware that the Institute and Faculty of Actuaries have considered[1] this issue but are yet to highlight any potential approaches that may form a helpful input in developing our approach.
As some respondents identified, there are a number of ways to implement dynamism, each with strengths and weaknesses to consider. Creating a fully dynamic system would be a time intensive undertaking and it could mean that legislation is underway by the time we could introduce it. In addition, before adopting any fully dynamic system, we believe it would be appropriate that we consult further with our stakeholders. We are also mindful that anything we do under the interim regime should be broadly aligned with the direction of travel that a future permanent legislative regime might follow, ie with allowance for some form of dynamism in the discount rate.
In light of these circumstances, we believe our update of the discount rate (with the addition of a pragmatic review mechanism) will enable the rate to be reviewed and adjusted in the future, will address the previous lack of market responsiveness, and will support innovation while maintaining appropriate security for savers.
Lastly, we note comments about the need for consistency with our DB funding code. We are satisfied that a higher superfund discount rate of gilts plus 0.75% per annum is fully consistent with the principles in the draft code for low dependency. We also see no conflict between the superfund discount rate and the proposed low dependency discount rate for Fast Track, which is currently lower. While there are similarities, the latter is set with a different objective to the former, namely to help us triage valuation submissions. Additionally, the discount rate for superfunds provides a broad reflection of the level of mainly credit linked market yields that might be available today for a suitable superfund investment portfolio, whereas fast track provides a broad reflection of the level of yields that might be available once schemes have transitioned to a low-dependency asset allocation in the future. Also, it is used to discount over a future period, starting at 'significant maturity'.
The discount rate for superfunds cannot be viewed in isolation. Taken together with the minimum capital adequacy requirements, which are driven by the level of investment risk taken in the investment strategy, we continue to ensure a high level of security for prospective members.
3. Gateway test
We asked:
Our guidance for prospective ceding trustees and employers requires, specifically in relation to Gateway Test 1 (and by extension Gateway Test 2), that 'the ceding trustees’ assessment of whether buy-out is affordable should be based on the ceding scheme actuary’s estimated buy-out funding level at a date no more than one month before the date of the clearance application.'
Do you think a period longer than one month would be appropriate here? Could you highlight advantages and disadvantages, including from an accepting superfund’s perspective?
You said:
Most of the respondents agreed that the one-month period should be extended for the following reasons (among others):
- Gateway tests are undertaken early on in any transaction and the current one-month period does not provide enough time to conclude negotiations and prepare for clearance.
- A longer period would be fairer and would help to ensure that costs are kept to a minimum if the market moves and a gateway test subsequently fails.
- It could help superfunds enter exclusivity deals with schemes, meaning the transaction is much more likely to go ahead.
Some respondents also stated that the test should not just focus on affordability of buy-out, but also take into consideration whether schemes could access the buy-out market. This could include, for example, whether schemes could obtain a fair and competitive quote due to potential capacity issues with insurers or whether the scheme needed to undertake significant data cleansing despite having sufficient funding levels. It was also suggested that buy-out estimates should be completed by specialist de-risking advisers and a set of clear principles for these advisers/scheme actuaries should be developed so there is consistency in how a buy-out estimate is undertaken.
Our response:
We understand that providing some flexibility with regard to the timing for the tests would be helpful, while still ensuring that adequate protections for members remain in place. Therefore, our amended guidance states that if the gateway tests are demonstrated to be met at a particular date, we will deem them to be met for nine months afterwards.
We believe a nine-month period will provide more certainty for schemes and ceding trustees and enable a smoother transfer process to operate. Schemes and ceding trustees will have more comfort that the transaction will still be able to take place without the need for additional gateway tests and the associated burden of further costs. There will be more time for the different elements of the transaction to be properly completed.
We recognise that there is a risk that the markets move during this nine-month period, which may have an impact on the scheme’s funding position and any subsequent assessment of the gateway test. However, trustees will still need to ensure the transaction is in their members’ best interests before they complete the transaction.
We chose a nine-month period because we allowed for the time it may take to prepare a clearance application, submit the application form and complete the transaction after clearance. We believe that in most cases these three phases should take no longer than nine months.
Some respondents highlighted that there may be a capacity crunch within the insurance market due to the number of schemes that are now able to afford buyout due to improvements in their funding positions. This has been highlighted as a risk by the Work and Pensions Select Committee and supported by the evidence they have received. We agree that this is a risk and acknowledge that smaller schemes in particular may find that they have more difficulty in accessing buy-out as insurers may prioritise other schemes to transact with.
We believe that allowing some schemes that cannot access the buy-out market to be able to transfer into a superfund could give them the opportunity to safeguard members’ benefits. This is particularly true if a scheme’s covenant is weak, and it would enable them to lock in and likely improve their current funding position without the worry of running the scheme on and funding levels worsening.
Our guidance now states that 'The trustees’ assessment of whether the scheme can access buy-out should be based on the response by an appropriate insurer (ie one indicated by professional advice to be potentially interested in taking on schemes with similar characteristics to the scheme in question). If the insurer declines the scheme or advises that buy-out could not be completed within a reasonable period appropriate for the circumstances of the scheme, then it would be reasonable to conclude that the scheme cannot currently access buy-out.' This change aims to prevent disadvantage to schemes and members if there are capacity issues in the insurance market.
We understand that, while it may be helpful to develop further principles for schemes to consider when determining what a reasonable period might be, we are not minded to do this. This is because any consideration of what a reasonable period appropriate for the circumstances of the scheme might be would need to take account of the market at that time and the individual scheme’s circumstances. These circumstances are likely to include, for example, the outlook for the scheme’s sponsoring employer/covenant. Consideration would also be supported by appropriate independent advice in relation to what was likely to be in the best interests of the members.
4. Timing to complete a transfer to a superfund
We asked:
More widely, in relation to timing, do superfunds, ceding trustees and ceding employers, collectively, have sufficient time to negotiate, obtain clearance for and ultimately complete a transfer to a superfund? If not, how might the current requirements be adjusted and how might our guidance and/or the role we play in the process be amended in light of other potential risks?
You said:
Several suggestions were made to amend the process as follows:
- Streamline the clearance process to give more certainty for participants that the transaction will be approved and, following on from question 3, a longer period would be helpful.
- As superfund transactions are similar to insurer transactions and an insurance transaction may take six months, further time should be allowed.
- We could provide more guidance around the clearance process (for example, around investigations for material detriment). One option could be to separate the process — undertake the gateway assessment first and then do detailed work on the transfer later.
- Consider giving 'clearance in principle' as this would give confidence to transact.
Several respondents raised the issue of cost, as a lot of expense and time is incurred before clearance can be obtained without the certainty of knowing whether clearance or gateway tests will be met or given. This is perceived as a barrier to entering the superfund market.
Our response:
As noted in our response to question 3, we are amending our approach so that if the gateway tests are demonstrated to be met at a particular date, we will deem them to be met for nine months afterwards.
In addition, we are introducing a similar nine-month period for demonstrating that the capital requirements have been met in relation to a particular transfer. We still expect our capital requirements to be highly likely to be met at the date of transfer, if reassessed at that date. The guidance explains how the agreement between the parties should reflect this expectation.
We believe that by adding this responsiveness to market movements, in addition to limiting the risk of transactions being 'undercapitalised', parties involved in superfund transactions will be more easily able to agree in advance the amounts they will contribute when the transfer takes place. This should enhance the smoothing of the process of transferring to a superfund that follows from the nine-month gateway period.
In relation to the comments on clearance, while we are not making changes to the actual process, we have made some clarificatory changes to the guidance which provide insight into the process.
5. Profit extraction
We asked:
Would it be appropriate to allow profit extraction at this point in time on an ongoing (annual) basis and, if so, what would be the appropriate requirements around it (for example, level of funding)?
You said:
Responses to this question were split. Those who felt it would be appropriate gave the following reasons:
- It is essential for the purpose of this product to be successful (ie establish superfunds as a commercially viable marketplace, as opposed to a 'feeder for the insurance market'). It would improve competition and therefore the marketplace.
- Superfunds offer solutions where insurers cannot, so it might be a solution to extract profit once the capital has reached the equivalent capital requirements of the insurance regime.
Those who disagreed cited the following reasons:
- Only allow after benefits secured, as still little evidence.
- Legislative uncertainty.
- TPR should focus on strong pensions, not weak insurance.
Some suggestions on the approach were provided, as follows:
- There needs to be distinction between the extraction method allowed for arrangements that are 'bridge-to-buyout' and those which focus on run-off.
- The percentage at which run-off profit extraction should be allowed should be based on reaching a certain level of funding surplus that has been risk-assessed to not impact the likelihood of members receiving full benefits.
- The circumstances under which profit can be extracted need to be clearly defined for superfund investors to invest.
Our response:
We have listened to your feedback and agree that allowing a proportion of surplus profits to be extracted in some circumstances could help the market to start transacting, without putting members at additional risk. The DWP, in its consultation response envisages allowing for profits and that a specific trigger will need to be established in the legislated regime. We agree with this approach and are mindful that any approach we adopt should not unduly compromise the security of members’ benefits.
This is a complex area and as a result, we intend to engage further with industry to establish a profit trigger mechanism for our interim regime. We will update our guidance accordingly.
6. Clarifications on any other aspects
We asked:
Are there any aspects of the current superfund guidance (and information requirements) that could be clarified to make the application process more streamlined?
You said:
Respondents provided a number of suggestions for changes to the application process, including the following:
- The template application for clearance is only available on request. It would be helpful to have this in advance to help schemes prepare.
- TPR could clarify what the gateway test means in practice. However, it is expected that this will get easier as the market gains experience in applying the test.
- It would help to have greater clarity around the application process, the documents required and what due diligence should be completed.
- The opportunity to receive mid-stage non-definitive feedback from the Regulatory Decisions Committee would significantly enhance the application process.
Our response:
Assessment and ongoing supervision
We have made a number of minor clarificatory changes to the guidance on the assessment of superfunds that reflect both our experience of operating the interim regime and the feedback we have received during the engagement.
Although our expectations have not changed, some sections have moved to improve clarity and readability, like our new appendix on ‘Ongoing supervision and superfund transactions’. We have included a section on ‘Collective Competence’, which has restructured some information from the Fit and Proper section and provides further information on an existing part of the assessment. Also, in relation to the Fit and Proper assessment, we have taken the opportunity to further explain our process. As part of our assessment, we will also assess whether there are other individuals who should be included in the assessment.
We have provided examples in some places to bring our expectations to life and we are being more explicit about what we expect to see. For example, in relation to IT systems used in administration, for the avoidance of doubt, we have explicitly stated that we will consider how superfunds would respond to fraud and data breaches. This better aligns our guidance with assessment forms that we would provide to superfunds under assessment.
We have further clarified how we run our assessment of systems and processes. We will expect a superfund’s systems and processes to be developed enough so they can be assessed. We will need evidence to satisfy us that there are sufficient systems and processes in place, and that they will be ready to go live at the point of a first transaction. The further assessment of systems and processes in the context of a transfer into a superfund is now described in Appendix C.
Technical provisions and capital adequacy modelling
In Appendix A of our DB superfunds guidance we highlight, as would be expected, that actuaries using the Continuous Mortality Investigations Bureau’s mortality improvement model will need to incorporate the recently introduced 'w factor' in their calculations of TPs.
In Appendix B of the same guidance, we have also asked that a 'statement of modelling principles' is prepared, revised when the modelling approach changes and an up-to-date version is provided to us on an ongoing basis. This should summarise the approach taken to modelling (for example, process and responsibility for amending the model and selecting and updating input assumptions, sensitivity of capital buffer to key modelling assumptions) and include the current set of assumptions in use and the sensitivity of the capital buffer to these. We would expect this to be a brief document but with signposting to more detailed underlying information (which also needs to be kept up to date). This will help our supervision and clearance teams be prepared to review proposed transactions and should further smooth the process for all concerned.
Wind-up trigger
Our guidance set the wind-up trigger at 105% of S179 'unless otherwise agreed by us in exceptional circumstances'. We have clarified in our guidance that, in considering agreeing to any lower wind-up trigger for a particular scheme transferring to a superfund, we will base our decision on the balance of whether it is in the best interests of members against the risks of a call on the PPF. We would expect, in any circumstances where we might be asked to consider this, that there is suitable rationale with appropriate evidence for us to consider in consultation with the PPF. This does not create a new expectation but provides further explanation as to what you could expect from us.
7. Further feedback
We asked:
Any other feedback?
You said:
A significant number of respondents provided suggestions on changes to the interim superfund regime. Respondents also asked for clarity on areas such as ‘profit distribution’, appropriate capital adequacy risk metrics for long-term investments, as well as our expectations of a trustee in the assessment phase.
A recurring theme was also the request for further clarity and feedback from us in general, such as proactive help during the application process, more guidance for trustees on the areas to consider when looking at investment strategies, as well as prompting the DWP for a response to its 2018 consultation.
Our response:
Firstly, we would like to thank all the respondents for taking their time to provide a wide range of suggestions. We welcome those comments.
We received a few comments on investment. These included a request that we should provide more guidance to trustees on the areas to consider when looking at their investment strategy, that the limits on illiquid investments (Investment Principle 4) should be relaxed and that a letter we issued to superfunds in July 2020, to clarify some practical implementation aspects of Investment Principles 3 and 6, should be published as part of the revised guidance.
Our superfunds regime allows for an appropriate amount of risk-based capital to be provided to support the level of investment risk being taken in the scheme and capital buffer. We are aware that different superfund investors want to offer different types of superfund model, with different approaches to investment. We welcome innovation and expect trustees to obtain appropriate advice from their investment advisers on the investment strategies being run in the scheme and buffer, and the implementation of those strategies. We do not propose to provide additional guidance for trustees on areas to consider in relation to the investment strategy.
We have considered the request for a relaxation of the limits under Investment Principle 4 and, following advice from Mercer, we have made some adjustments to the previous limits. To allow for some investments typically held being categorised as Level 2 (rather than Level 1) and more illiquid assets typically held being categorised as Level 3, we have:
- reduced the minimum exposure for Level 1 assets to 30% (from 40%)
- increased the maximum exposure for Level 3 assets to 35% (from 25%)
These changes will provide additional flexibility in implementing the investment strategy
As above, we have attempted to address your comments in respect of clarifying our expectations, TPs and capital adequacy modelling. We will continue to monitor the market as it develops and may revisit some of the suggestions in light of market developments.
Lastly, while the DWP has indicated that it will continue to develop its definition of superfund for the legislative regime, we have taken this opportunity to amend our explanation of a superfund to more closely match the DWP’s current thinking. We hope that this amendment makes it clearer what we consider to be a superfund and that it will minimise any impacts on our stakeholders during the transition from our interim regime to the upcoming legislative regime.