Consultation outcome

Chapter 3: Diversification, performance fees and the default fund charge cap

Updated 21 June 2021

Background

1. Chapter 4 of the ‘Investment Innovation and Future Consolidation’ consultation considered the interaction between the charge cap for default arrangements^{[footnote 20]} used for Automatic Enrolment, illiquid investments and performance fees. It also proposed a new method for charge cap compliance to give trustees more flexibility to pay performance fees.

2. Here, we summarise the feedback we received to that chapter. We set out the details of proposed amendments to the charge cap to accommodate performance fees in order to facilitate investment in illiquid investments, and to put the exclusion of physical assets on a statutory footing. We also announce our intention to develop a further alternative option for schemes to use in calculating performance fees, to facilitate investment in less liquid assets such as venture capital.

3. We also propose updates to the non-exhaustive list of charges for charge cap guidance which indicates what we consider to be in or out of scope of the cap.

Stakeholder responses

The use of performance fees in illiquid investments

4. We asked about the extent to which performance fees are used or required for funds which offer illiquid investments, and whether market practice is changing. Responses to this question presented a mixed picture. It was clear that schemes can access illiquid investments without paying performance fees. However, respondents explained that the prevalence of performance fees varies depending on the type of underlying asset.

5. Those respondents that elaborated further generally agreed that performance fees are most common in private markets, such as venture capital, private equity and growth equity. They are less common in infrastructure and are rare in property investments. A couple suggested that this variation was due to the differing investment strategies associated with certain asset classes.

6. A few respondents working in the venture capital or alternative investments industry suggested that performance fees are necessary for those types of investments. Reasons given focused on their use having been accepted practice globally over a long time, alignment of interests between investors and investment managers and the level of skill needed from investment managers. Other respondents stated that, in practice, the terms of performance fees are not always effective at aligning the interests of investors and investment managers.

7. Several of the large scale pension schemes and master trusts that responded are already invested in illiquid investments, within the existing charge cap, and without paying performance fees. One large master trust stated that they have been investing in direct illiquid UK property for years. Their scale has enabled them to negotiate access for competitive base fees, at the level needed to keep costs within their investment cost constraints, which are significantly below 30bps, without paying performance fees.

8. One respondent stated that the majority of their illiquid assets have no performance fee and that it is possible to find managers who are willing to manage illiquid assets on an ad valorem basis, while still being within the charge cap. They also emphasised that when considering illiquid assets, the issue of an illiquidity premium and alpha should not be confused. Trustees should seek to harvest diversified illiquidity premia for scheme members, rather than pay for alpha which may not exist.

Changes in the use of performance fees

9. On the question of whether practices are changing, several respondents pointed out that the scale of assets in DC is still a fraction of those in the Defined Benefit (DB) market. Private market funds are often over-subscribed, meaning that there has been little need for managers to adapt their charging structure to attract DC investment.

10. However, pension schemes and their advisers told us that as the DC market matures, they are seeing more options for accessing illiquid investments without performance fees. Some pointed out this has the potential to create a higher annual management charge than if it were alongside a performance fee.

Performance fees and the charge cap

11. There is a charge cap on the default arrangements of DC schemes used for Automatic Enrolment^{[footnote 21]}. Double defaulters – individuals who make no choice when they are automatically enrolled about joining a pension or the fund to which they contribute – will be put in to a default arrangement. The cap was introduced to protect these members from excessive charges and to present a straightforward offering to those new to financial services.

12.The charge cap is 0.75% of the funds under management each year, or an equivalent combination charge. It includes most charges, but excludes transaction costs (the costs of trading) and the costs associated with holding and maintaining assets whose value is based on their physical form, such as property and infrastructure. There is guidance for trustees of occupational pension schemes on how the charge cap works^{[footnote 22]}.

13.There are currently 2 options for trustees to measure charges for funds under management to confirm they are compliant with the cap. Both methods apply over a calendar year, referred to as a charges year.

The first option requires the value of the members’ fund to be measured at regular reference points throughout the year and the average of those figures calculated at the end of the year.

The sum of charges throughout the year is then assessed as a percentage of that average fund value. The guidance refers to this as the retrospective method.

14. The second option requires the scheme to have a predictable charges regime. Because the level and frequency of charges are known in advance trustees can confirm at the beginning of the charges year that no member will be charged more than the cap. The guidance refers to this as the prospective method.

15. Both methods for measuring charges require that the maximum permitted charge is pro-rated when the cap applies to a member for less than a full charges year. This can happen when a member leaves or joins during the charges year.

Issues with the mechanism of charge cap compliance

16. We asked whether complying with the charge cap acts as a barrier to accessing funds which charge a performance fee and whether, in turn, this acts as a barrier to accessing certain asset classes. Several respondents from across the investment industry, pension schemes and their service providers told us that the way that the charge cap is calculated does act as a barrier to paying performance fees.

There were 2 issues identified with the mechanism of charge cap compliance:

17. The first issue identified was the requirement to pro rate the cap for scheme members who are invested for less than a charges year. The charge cap of 0.75% of assets under management applies to scheme members invested for the full charges year.

A scheme member invested for four months, or a third of a year, would have a cap of 0.25%, a third of the annual cap.

18. As performance fees are based on investment returns and are likely to vary throughout the year, trustees would find it difficult to ensure that for any combination of leaving and joining date within a charging year, no scheme member will pay more than the prorated charge cap.

One response explained that, even where their fund had applied a cap on maximum permitted performance fees within the charges year, the pro rating requirement meant that compliance could not be guaranteed.

19. The second issue identified with the mechanism related to the interaction between a fixed charge cap and an uncapped performance fee.

A couple of responses suggested that in order to avoid a breach of the cap, trustees will want to leave enough headroom for any potential performance fee. This leads in turn to a lower allocation than might be preferred.

Level of the charge cap

20. Although not the mechanism for compliance, several respondents did mention the level of the cap as a potential barrier to investing in certain asset classes. This concern arose where the assets under consideration may have higher management costs than traditional equities due to the need for more intensive, specialised management.

Although the view that this made them inaccessible within the cap was not universal it was a perception held by several respondents.

12. Further issues were raised about the level of the charge cap. A few respondents suggested that trustees would be reluctant to increase spending on investment to make use of the unused headroom in the charge cap.

We are reviewing the scope and operation of the charge cap, and part of the aim of that review is to ensure trustees are encouraged and enabled to invest in a wide range of assets, including illiquids, while maintaining clear and transparent protection for savers. We will publish that review towards the end of 2020.

Other barriers to DC schemes paying performance fees

21.The ‘Investment Innovation and Future Consolidation’ consultation identified that there are specific considerations about how performance fees work in relation to DC schemes.

Members of DC schemes can leave and join funds at any time, including during a period over which a performance fee is assessed. It may not be possible to accurately distribute performance fees levied on a pooled fund to the scheme members who benefited from the investment growth. This concern was shared by some respondents.

22. A couple of respondents went further and pointed out that many of the features of a well-structured performance fee do not work as intended when scheme members can transfer in or out during the lifetime of the fund.

Examples of such features include high watermarks, which prevent investors paying for the same performance multiple times as the value of the fund fluctuates, and hurdle rates, which require a minimum level of return before any performance fee is paid. If a scheme member joins after the high watermark has been passed, they may not pay a performance fee on the returns they receive in the future.

If a scheme member joins after a hurdle rate has been passed, they will not benefit from its protection.

23. Several of the responses from the investment industry described how traditional multi-year funds, which are most common across venture capital and growth and private equity, are structured.

They explained that performance fees, or profit sharing as they are sometimes called, are only fully known at the end of the fund’s life once all of the investments have been realised. This extended assessment period ensures that the fees are accurately calculated at fund level, and is designed to encourage sustained performance. However, it exacerbates the difficulties with fairly allocating fees to the appropriate members, as is necessary in DC.

24. In addition, a couple of respondents suggested that investment platforms specifically struggle to incorporate funds with performance fees because of difficulties with fairly apportioning them, raising concerns that they may create Treating Customers Fairly issues.

25. Whilst no one presented a definitive solution to these concerns about fairness, a couple of ways of minimising the impact were suggested.

26. First was the regular calculation and accrual of the performance fee, which was discussed in the consultation and incorporated into the proposal we put forward in 2019.

27. The other suggested approach was for funds to have monthly or annual share classes within which the performance fees were assessed. Once the performance fees had been levied, the share classes could be merged in to a single fund. One respondent acknowledged the additional administrative burden this presented.

28. The consequential effects of paying a variable fee were also discussed. This included, most notably, the costs disclosure framework which applies to DC schemes and how to communicate performance fees in a way that allows members to assess value.

Additional burdens for schemes in comparing the costs of different funds and monitoring how performance fees have been applied were also raised.

Other barriers to accessing illiquid investments

29.Some respondents identified other barriers to DC schemes accessing illiquid investments as being more pressing than the charge cap and performance fees. The most prominent barrier put forward was that the investment platforms which the vast majority of DC schemes use are often only compatible with investments which are priced daily.

This means DC schemes may struggle to gain access to the existing products.

30. Several respondents raised behavioural reasons that meant those running schemes were reluctant to spend more on investment. They told us that competition between schemes is mainly based on costs and schemes were concerned that by investing in a more expensive investment strategy, they would appear to offer worse value for money than other, cheaper schemes.

31. A couple of respondents identified that – in times of market stress – schemes with exposure to more illiquid investments may find it difficult to rebalance the portfolio to keep charges within the cap.

Proposed amendment to the charge cap compliance mechanism

32. The ‘Investment Innovation and Future Consolidation’ consultation proposed an additional method of charge cap compliance to make it easier for trustees to invest in funds which levy performance fees.

33. The proposed method required that schemes first use the existing prospective method of assessment to establish that the charges, excluding performance fees, are compliant with the charge cap for any combination of joining or leaving date.

Then, just for scheme members invested for the full year, the difference between the charges for the full year and the charge cap would be calculated. This difference is the budget that can be spent on performance fees.

34. If a default arrangement has a charges regime that would result in members invested for a full year paying 0.5% of the value of their pot in fees then, under the proposed method, that arrangement could pay performance fees of up to 0.25% over the charges year.

35. By only considering the performance fees in relation to scheme members invested for the full charges year, this method avoided the prorating issue.

As long as the maximum potential performance fees, plus the % funds under management fees would be compliant with the charge cap for scheme members invested for a full year, and the % funds under management fees would be compliant with the prorated cap for partial year members, then the charges regime would be permitted.

36. We sought views on whether this additional method of charge cap compliance should be permitted, whether it was likely to lead to problems with complying and whether it may disadvantage members.

Minimising the potential for disadvantage to individual members

37. The consultation identified the risk that members of DC schemes could pay performance fees which are not related to the investment returns they have received.

38. In simple terms, this is because the performance fee is typically levied on a pooled fund, but passed on to individual scheme members’ pots. As scheme members of a DC scheme may transfer in and out of the fund at any time, their period of membership may not align with the time period over which the performance fee is assessed.

39. To reduce the potential impact of this unfairness, the consultation proposed that statutory guidance could specify that DC schemes should only accommodate performance fee structures which accrue^{[footnote 23]} the performance fee each time the value of the investment fund is calculated.

Although the consultation did point out that trustees would want to consider the frequency with which those performance fees are paid out to the managers and whether that might encourage short term decision making.

40. It also proposed that the additional flexibility around performance fees only be permitted alongside a funds under management charge and not alongside combination charges, which include either a contribution charge, or a flat annual fee.

This reflected one of the principles set out when the charge cap was introduced to simplify charging so that scheme members are able to compare the value of different schemes.

41. The additional method of assessment received broad support and the majority of respondents who answered this question thought we should go ahead.

42. Respondents did, however, identify some problems with introducing the additional method of assessment. The fact that it relies on a capped performance fee which is not standard practice was raised.

As the additional method is linked to the prospective method of charge cap compliance, it would require trustees to confirm at the beginning of the charges year what the maximum possible performance fee would be. It was pointed out that capping performance fees could mean that the full charge cap budget is rarely used in practice because the headroom has to be reserved for potential high returns.

43. We spoke with pension lawyers and service providers who challenged the view set out in the consultation that most schemes use the prospective method of assessment. They pointed out that where schemes pay any ad hoc fees within the year, such as paying for auditing services, they do not operate a regular charges regime.

A charges regime is the basis of the prospective method of assessment. The use of the retrospective method is, therefore, much more common in practice than we first realised and those schemes would not benefit from the flexibility offered by the proposed additional method.

44. Of those respondents who disagreed with the proposal, the reasons given were split. A couple from the investment industry raised the point that the method does not reflect the way that many performance fees are charged, in that it would only permit capped performances.

The others, from pension schemes and a trade union, argued that the change is not necessary and may encourage performance fee use, which they considered to be inappropriate for DC schemes.

45. There was also broad support for the proposal that statutory guidance be produced which would require performance fees to be accrued each time the investment is valued.

The proposal to limit the additional method to schemes which use a single charge structure was also well supported as a means to reduce complexity for members and trustees. There were a couple of concerns that this would restrict two of the largest master trusts, who do use combination charges, from making use of the additional method.

One respondent argued that even where a combination charging structure is being used, the performance fee would be reported within the funds under management charge and so would not add a third type of charge for scheme members.

Government response

46. The evidence we received demonstrated that trustees can develop a well-diversified portfolio without paying performance fees. The investment industry is beginning to develop charging models which reflect the structure of DC schemes and it is realistic that this will continue as the DC market grows and consolidates.

47. It seems that the extent to which the charging models are changing does vary based on the underlying asset class. However, we have not seen any evidence to indicate that there are asset classes which are inaccessible to DC schemes due to the charge cap.

48. Closed-ended investment trusts are pooled investment vehicles which invest in a portfolio of assets and issue shares which can be traded on a stock exchange, like a listed company.

Because there is a fixed number of shares, investors can only cash out of the trust by selling their share. This means that the trust itself does not have to be able to offer redemptions and can, therefore, invest in illiquid assets, but the investors can trade the shares like any other equity.

This characteristic can make investment trusts an attractive option for DC schemes seeking exposure to less liquid asset classes.

49. We looked at available investment trusts in the infrastructure, renewable infrastructure and venture capital sectors and found that they all had options both with and without performance fees^{[footnote 24]}.

One of the 7 listed infrastructure trusts and one of the 12 listed renewable infrastructure trusts each levied a performance fee. Of the ten largest venture capital trusts, 6 levied a performance fee.

Of the 3 sectors, it was only in infrastructure where the highest 5 year returns were from a fund with a performance fee.

50. The fact that competition between schemes is based mainly on the level of charges may well lead to a perception that cheapest means best value. This is not a new phenomenon.

Before the introduction of the charge cap, when employers did compare pension schemes it was likely that they would prioritise low charges rather than scheme quality or governance^{[footnote 25]}. We want to make it clear that good value for money is about more than just low fees, and that investment returns must always be considered as part of assessing value.

This is why we are proposing that net returns are included in the updated value for members assessment. Detail of this proposal is covered in Chapter 2.

51. Where the trustees of large DC schemes decide to invest the default in illiquid assets, they will generally make up no more than 10% to 15% of the portfolio. Therefore, we are satisfied that a diversified portfolio can be achieved by blending those assets with lower cost ones to keep overall costs within the cap.

52. We understand that trustees may be cautious about making use of the available headroom in the charge cap while there is a review of the level of the cap pending. We are reviewing the level of the charge cap this year, as well as considering the permitted combination charging structures and the treatment of transaction costs.

In reporting on that review, we will seek to give trustees sufficient assurances so that they are able to make appropriate long term investment decisions.

Amendment to the charge cap compliance mechanism: an in-year adjustment to prorating performance fees

53. While we are satisfied that schemes can access a broad range of investments without paying a performance fee, it is for trustees to decide whether a performance fee represents best value for their members.

We also understand that the current charge cap compliance mechanism restricts the performance fees which schemes can pay and that greater flexibility could be afforded to trustees, while maintaining member protections.

After considering responses to the consultation we plan to proceed with an amendment to the way that compliance with the charge cap is measured.

54. The requirement to prorate the charge cap works for charges which are spread evenly throughout the charges year. However, we recognise that for performance fees, which can significantly vary throughout the year, the requirement to be able to meet the prorated cap level for any combination of joining or leaving date is difficult to satisfy.

Our proposed draft regulations will provide an easement to the requirement to prorate performance fees when assessing compliance with the charge cap.

55. We initially proposed an additional method of compliance which would only apply to the prospective method of assessment. As we are now aware that the retrospective method of assessment is widely used by schemes, we have adapted the proposal so that it will apply to both.

56. Where a scheme member is in the default arrangement for only part of the charges year, we propose that trustees will still have to calculate a prorated charge cap based on the length of time they have been a scheme member.

The effect of the proposed policy change will be that when assessing the charges that scheme members have paid against the prorated cap, trustees will exclude the performance fee element^{[footnote 26]}.

This only applies to partial year membership. When considering a scheme member who is in the default arrangement for the whole of the charges year, the performance fee would be assessed against the charge cap.

57. This proposed easement will not apply to all performance fees; the draft regulations only allow trustees to exclude a performance fee from the calculation where it is accrued each time the value of the investments is calculated^{[footnote 27]}.

This prevents a situation where a scheme member joins an arrangement after an investment has increased in value and ends up paying a performance fee on the growth which occurred before they joined.

58. We propose that a performance fee which is accrued less frequently than the tradeable value of the investment is calculated would still be subject to the prorating requirement.

59. We propose that it is more appropriate to put the accrual condition into regulations rather than statutory guidance as we previously suggested.

60. The draft regulations are available at Annex C - Regulation 6 makes amendments to the Occupational Pensions (Charges and Governance) Regulations 2015. A Keeling Schedule, indicating how the Charges and Governance Regulations will read once updated is available at Annex D.

61. We propose that the easement for performance fees would apply to the first charges year which begins after 5th October 2021^{[footnote 28]}.

Question 4: Do the draft regulations achieve the policy intent of providing an easement from the prorating requirement for performance fees which are calculated each time the value of the asset is calculated?

Further option for the charge cap compliance mechanism: creating a multi-year rolling calculation approach

62. Some stakeholders have suggested that DC schemes may require further flexibility in how they calculate performance fees within the charge cap, and that they should be given the option to calculate such performance fees over a multi-year rolling period.

It’s suggested that a multi-year option would be necessary to facilitate access to venture capital and growth equity assets that determine the performance element of their fees (known as a ‘carry’) over multiple years.

A 5 year rolling average has been suggested by the British Business Bank^{[footnote 29]} as a suitable period.

63. We therefore propose developing a multi-year approach to calculating performance fees as an alternative option for schemes that believe a long smoothing period is required to enable them to access a wider range of illiquid assets.

Such an approach would amend the calculation methodology to align with the multiyear structure by which these “carry” payments are assessed. Our proposal is that schemes will be able to select whether to use the in-year approach set out in our draft regulations or to use a multi-year approach.

64. We seek views on how a workable mechanism for enabling a rolling calculation over a number of years might best be constructed. There is a balance between creating a framework that can allow schemes to reflect the complexity of performance fees across multiple years without making that calculation overly burdensome.

We are keen to hear views from stakeholders about how such an measure might best be achieved in order to facilitate DC scheme access to the less liquid products that charge in this way.

65. We are also particularly interested in input from asset managers and distributors about ways in which their charging structures might be adapted to better fit the needs of DC schemes as customers.

Question 5: What should we consider to ensure a multi-year approach to calculating performance fees works in practice?

Question 6: We are proposing a five-year rolling period. Is this appropriate or would another duration be more helpful?

Question 7: We are proposing offering a multi-year option as an alternative to an in-year option for schemes. Do you have any suggestions for how to improve this offer?

Question 8: To what extent will providing a multi-year smoothing option give DC trustees more confidence to invest in less liquid assets such as venture capital?

Cost of holding physical assets

66. The costs of holding physical assets, such as real estate or infrastructure, are not included within the charge cap. This is already set out in the charge cap guidance^{[footnote 30]} but we propose to take this opportunity to put that exclusion on a statutory footing.

67. The draft regulations carve the costs solely attributable to holding physical assets out of the charge cap^{[footnote 31]}. They define physical assets^{[footnote 32]} and provide a non-exhaustive list of costs^{[footnote 33]} which are excluded.

Physical assets are defined as: an asset whose value depends on its physical form, including land, buildings and other structures on land or sea, vehicles, ships, aircraft or rolling stock, and commodities

Within that, commodities are defined as: any goods of a fungible nature that are capable of being delivered, including metals and their ores and alloys, agricultural products and energy such as electricity, but not including cash or financial instruments.

The non-exhaustive list of costs is:

• the costs of managing and maintaining the asset
• fees for valuing the asset
• the cost of insuring the asset in question
• ground rent charges, rates and taxes incurred in relation to the asset

68. The following are examples of charges that we would consider to be solely attributable to holding physical assets, and therefore excluded from the charge cap:

• costs of maintenance, repair, improvement and development of properties and other physical assets, including project management and contractor costs
• costs of holding the physical assets including business rates, insurance, utilities, void costs, valuation fees
• property management fees paid to third parties for day to day management

69. This proposed update to the Charges and Governance Regulations would come in to force on 5th October 2021.

Question 9: Do the draft regulations achieve the policy intent? Do you have any comment on the definitions used?

Non-exhaustive list of charges

70.The ‘Investment Innovation and Future Consolidation’ consultation included a non-exhaustive list of costs and charges and whether government considered each to be in or out of scope of the charge cap^{[footnote 34]}. This was provided to support trustees in understanding the government’s policy intent.

71. The updated list of cost and charges was overwhelmingly welcomed as providing greater clarity on the policy intent.

72. We received one substantive query from a couple of respondents. This was in relation to the costs incurred in underlying firms and when trustees should ‘look through’ to those costs.

73. We have always been clear that trustees should look through to underlying costs in investment funds, for example in fund of fund structures.

However, where the firm in which the scheme is invested is a commercial company rather than an investment company, for example a supermarket, the underlying costs of running the supermarket would not be considered when calculating charges for the cap.

74. In order to determine when trustees should look through to underlying costs, we had proposed that they do not look through to the costs ‘incurred by investee firms which have a general commercial or industrial purpose’.

Two responses asked for further clarification on how trustees can make this assessment, particularly in relation to the treatment of real estate investment trusts (REITs).

75. In response to this request for further clarification, we have amended the list of costs and charges (see Annex F).

We propose that schemes should look through all open-ended funds and all UK listed closed-ended investment funds and international equivalents. The vast majority of securities which are listed in the UK appear on the Financial Conduct Authority’s Official List^{[footnote 35]}. For the purpose of the list, securities are categorised.

If a company is categorised as a closed ended investment fund, then schemes should look through to the underlying costs.

76. Whether a firm is a REIT does not determine the need to look through to underlying costs. If a UK-listed REIT is classified as a closed ended investment fund, then there should be look through. If it is classified as a commercial company, then there is no look through.

77. Where a security is not so explicitly categorised either because it is listed in another country or it is not on the Official List, trustees will want to continue with the current practice of considering their own legal advice on the status of underlying charges.

1. Introduced in April 2015 by The Occupational Pension Schemes (Charges and Governance) Regulations 2015 (SI 2015/879) ↩

2. Introduced by The Occupational Pension Schemes (Charges and Governance) Regulations 2015 (SI 2015/879) ↩

3. The charge cap: guidance for trustees and managers of occupational schemes ↩

4. By accrual we mean that the fees are calculated and the value of the fund is adjusted to reflect them. They are not necessarily banked (crystallised) by the investment manager at this point. The fees would however constitute a charge, unless they are refunded to departing members. ↩

5. Analysis carried out on Morningstar Closed-end Fund Screener website, August 2019 ↩

6. Defined contribution workplace pension market study, September 2013 ↩

7. “performance fee” is defined in regulation 6(2)(a)(iii) of the draft regulations. ↩

8. Regulation 6(2)(c) of the draft regulation. ↩

9. Regulation 1(5) of the draft regulations. ↩

10. The Future of DC Pensions: Enabling Access to Venture Capital and Growth Equity ↩

11. The charge cap: guidance for trustees and managers of occupational schemes, see paragraph 12, page 6, October 2016 ↩

12. Regulation 6(2)(a)(i) of the draft regulations. ↩

13. Regulation 6(2)(a)(ii) and (iii) of the draft regulations. ↩

14. Regulation 6(2)(b) of the draft regulations. ↩

15. See annex to Investment Innovation and Future Consolidation ↩

16. The Official List ↩