The Value for Money Framework: consultation

Background

On 8 August 2024, the FCA published a consultation on detailed rules and guidance for a new value for money (“VFM”) framework.

In this response

• General comments

General comments

We welcome and support the policy aim of ensuring that savers receive optimal value for money regardless of the nature of their pension arrangement and that investments by trustees and providers are in savers’ best interests. While we appreciate that the proposed framework is designed to achieve consistent reporting according to specified metrics to make it easier to compare schemes, we have several key concerns:

• Scope – many trust-based schemes have multiple “inadvertent” default arrangements. For example, where certain members have been moved to investment arrangements without their consent, or those where changes have been made without member agreement but in which only a small proportion of the scheme’s assets are invested. We would suggest that the VFM framework only applies to “true defaults” to reduce the burden of carrying out the VFM assessments and so that the focus is on the arrangements which affect the most members. In principle, the application of the VFM framework to legacy “quasi-defaults” is sensible, but we suggest this should be done in a second phase of rolling out the framework so that the initial focus is on active defaults receiving new money? Further, we have clients with schemes to which the proposed VFM framework could not easily be applied, or which would be difficult to compare meaningfully with “standard arrangements”, so we suggest the following schemes should be expressly excluded (at least initially):

• • legacy schemes with unusual features such as with-profits, guarantees etc
  • schemes where the DC section is provided on a non-contributory basis (ie only employer contributions)
  • DB schemes where the only DC benefits are AVCs (to the extent these are not excluded from the final definition of “default arrangement”).
  • “in-house DC arrangements” (hybrid schemes in which a notational amount is invested in a DC section alongside the DB benefits, rather than members having individual DC pots).
• Comparability of data – as the FCA will know, we have co-ordinated a value for money comparison study with all the key IGCs in the industry since 2017. The framework used by the IGCs has a number of similarities with the proposed VFM framework, although the IGC framework has been more extensive in certain areas, notably engagement/service. In our experience, no matter how closed the data point request is, even in the more “quantitative” style questions, (relating to investments and charges etc) respondents to the request have invariably answered it in different ways. This raises questions about whether the data is comparable and sufficiently reliable to formulate their assessments and has resulted in a need for follow up with the firm(s) in question to check it/they has/have correctly understood what a question is driving at. The proposed VFM framework, which contains similar questions to the framework we have been using with the IGCs, will not have a single supplier to verify the data that firms and trustees of own trust occupational pension schemes publish. We have a real concern that, particularly in the early years, the data published using the proposed VFM framework may not be reliable enough to allow trustees and IGCs to make informed decisions on value. This concern is particularly acute given the proposed approach to actions following a RAG analysis (see further below). Essentially, IGCs and trustees will be asked to use potentially unreliable information (from a fair comparison perspective) to make significant decisions relating to pensions continuity which, in turn, could potentially affect a provider’s ability to operate. We would feel less concerned if the FCA/DWP/TPR could ensure that trustees/firms answer all questions in a consistent manner.
• Volume of data – the proposed framework includes over 200 data points. This volume of data could result in information overload, reducing its potential usefulness. Further, lots of time and resource will need to be spent on data collection and comparison for little real added value (particularly given that an amber or red rating seems unlikely given the consequences – see further below). We would encourage the FCA and DWP to consider whether each proposed metric adds clear value to the process.

  We also think the 3-month timeframe for providers to collate this data (from 31 December to 31 March) is ambitious and suspect the process may take much longer, especially in the early years.

• RAG ratings – we have a number of significant concerns regarding the RAG ratings:

• • The proposed RAG rating is in our view too simplistic. Assessing VFM is a complex process and the FCA/DWP/TPR should be looking to introduce a more sophisticated mechanism which better reflects the array of factors which IGCs and trustees consider when undertaking their assessment.
  • We are concerned about the proposed approach to and consequences of the amber rating. If it is adopted, the consequences of an IGC or master trust giving an amber rating are that their provider cannot take on new business in relation to that in-scope arrangement and participating employers need to be informed. In practice, amber ratings would most likely be equivalent to red ratings; were a provider to receive an amber rating in respect of one of its principal default arrangements, even for a very short period, the commercial pressures of the market could mean that it would be at risk of losing significant business. This means it is unlikely that in practice an IGC or master trust trustee would give their provider’s in scope arrangements an amber rating. The likely consequence is that IGC and master trust assessments will be in shades of green, which could render the RAG results meaningless. We would therefore urge the FCA to reconsider its approach.
  • We support the overall policy objective to create consistency across trust and contract-based arrangements when it comes to VFM assessments. However, it is not clear how the proposed RAG framework would be operated by occupational pension schemes which are not master trusts. To take an extreme example, if the intention is to mirror the regime in a trust based context, if a trustee of a small own trust rated its default arrangement (the investment funds and administration of which are provided on a bundled basis by a large insurer) as “red” in its VFM assessment in terms of bundled administration/fund management, this would that mean that the provider in question would have to move all members in that default arrangement, in its master trust or any contract-based arrangements, elsewhere. An amber rating would require the provider to cease admitting new employers into its master trust or any contract-based schemes that utilise the same default arrangement. We assume the intention is not that a small scheme of, say £100m AUM could dictate the fate of a provider with £100bn AUM via its master trust/contract-based provision. However, we are concerned that could be an unintended consequence, unless appropriate safeguards are put in place. We would therefore strongly encourage the FCA, DWP and TPR to reconsider this aspect of the proposals.
  • Following a similar theme, it is unclear how this will work for own trust schemes who use one provider for administration and another for investment platform services and schemes which administer their pension arrangements in-house. A number of very large own trust schemes have in-house administration and/or use eg an employee benefit consultant for administration and an insurer for investment services. The Trustee must take account of the investment performance and the administration services in determining its RAG rating for an in-scope arrangement, but the value provided in respect of each of these could be radically different if the services are being provided by different parties. If the Trustee gives an amber rating, then this could cause reputational damage to both providers, even where one part of the service provision and therefore one provider was not the key cause of the rating. Also, if the in scope arrangement is used elsewhere i.e. in a master trust or contract-based arrangement (in terms of its investment strategy) but with a different administration provider, would that arrangement be permitted to take on new business? We think this requires further consideration.
  • In addition, given the reputational damage of either an amber or red rating, there is a further risk that the rating system could drive the wrong behaviour and stifle innovation. For example, schemes may be more risk-averse, leading to less divergence of investment strategies.
  • There is also a risk that if a member sees a provider has a red or amber rating in relation to an in-scope arrangement they simply decide to stop contributing/opt out of the scheme altogether (rather than moving to a different scheme or in-scope arrangement), defeating the purpose of the framework.
  • We have concerns about how the assessment will work for multi-employer schemes where it is to be considered at cohort level. The RAG rating is an overall rating even if there is a difference in VFM between different cohorts and the arrangement cannot accept new business until it is rated as green overall. This seems unfairly punitive to multi-employer arrangements.
• Choice of comparators- we are unclear what the rationale is for comparison against sufficiently sizeable providers and both contract and trust-based schemes. In practice, it will not always be possible for members to be moved from their existing pension scheme to a large master trust or to a contract-based arrangement. Many smaller schemes do not have sufficient assets to be commercially viable to large providers and several trust-based DC schemes have some kind of hybrid benefit structure which makes a transfer to a master trust tricky and bulk transfers without consent are not permitted from a trust-based scheme to a contract-based arrangement. We consider that single employer trust schemes should be able to compare against other single employer trust schemes.
• Administrative burden – meeting the requirements will be a significant administrative burden which could disproportionately impact smaller own trust schemes. Less well-resourced own trust schemes will have the burden of meeting the new requirements, with no, or limited, ability to outsource the work which will have an impact on the funds available for other projects. This could negatively impact their member offering and reduce value for money. We appreciate the relationship between scheme size, resource and value but would note that not all smaller, less well-resourced schemes provide poor value to members. If the aim is consolidation, we think there are more effective ways of achieving this; notably more direct and focussed regulatory intervention on a scheme/provider specific basis.
• Corresponding requirement on providers – smaller schemes with less leverage may face more hurdles in obtaining the necessary data from investment/asset managers, administrators and other providers. We would encourage the FCA and DWP to consider a corresponding requirement on providers to provide the necessary information in time for trustees to make their assessment as was done for the provision of costs and charges information.
• Driving high-quality schemes out of the market – there are a number of reasons why this framework might result in the unintended consequence of driving high-quality schemes out of the market. First, the administrative burden and the hurdles in obtaining the data might force smaller trust-based schemes out of the market. Second, requiring firms to disclose the total costs and charges paid by members and their employers may skew the apparent value offered by high-quality schemes with employer subsidies.
• Reducing competition – whilst we acknowledge the policy aim of encouraging consolidation, competition from new entrants can have a positive impact on how the market develops. Our concern is that the impact on smaller providers might create barriers for new entrants into the market, thus reducing competition. Taken together with smaller trust-based schemes exiting the market due to an inability to satisfy the requirements (for whatever reason), we are concerned that the market may contract too much and too quickly, causing it to converge on a limited number of FCA registered insurance providers.
• Costs and charges likely to prevail as key comparator – due to the qualitative nature of much of the assessment framework, making direct comparisons will be difficult and, in practice, it is likely that arrangements will continue to focus on benchmarking costs and charges. We are concerned that the drive for lower charges will not sit neatly with government’s wider policy drive around investing in private markets, with such investments more likely to have higher charges but with the aim of higher investment returns and net performance (see below).
• Proposed costs and charges metrics – we have several concerns regarding the proposed costs and charges metrics:

• • Whilst we acknowledge that costs and charges in themselves do not form part of the determination of a RAG rating, we are concerned that the proposed disclosure of total costs and charges paid by members and their employers (employer subsidies) may skew the apparent value of a scheme. For example, high-quality schemes run by individual employers, often with the benefit of an employer subsidy, may not score highly on costs and charges compared with large master trusts, even if member outcomes could be as good if not better. As set out above, it is important that the framework does not result in the unintended consequence of driving high-quality schemes with good member outcomes out of the market.
  • In our view, net investment performance is a much more useful metric, particularly if the policy objective is to increase the use of private markets/illiquid assets in DC investments. Higher charges might be appropriate if performance is strong, making performance net of charges a much better indicator of value than charges in isolation. As this is also one of the metrics to be disclosed, we would query the benefit of separately having to disclose the charges metrics. Alternatively, we think the total costs and charges and the split between employer and member charges should be disclosed.
  • We appreciate the merit of separating out investment and service costs in principle, but it is not clear how feasible this will be in practice. Where providers are offering a bundled service, we understand that the price is given for the service as a whole and is dependent on commercial market forces. Separating it into its component parts could be difficult and potentially lead to artificial results. One solution might be to have different approaches for bundled and unbundled arrangements, as an employer in a bundled arrangement may be more likely to compare that to other bundled arrangements. However, we recognise that this would make it more difficult for employers to compare bundled and unbundled arrangements.
• Proposed service quality metrics – we have several concerns regarding the service quality metrics:

• • How members access their benefits is a fundamental component of value and there is some overlap in terms of whether this forms part of accumulation (in scope for VFM) or decumulation (not currently in scope). However, we would suggest that consideration of retirement support is excluded from VFM assessments for now. Current offerings are unlikely to be comparable and there are expected to be changes around decumulation obligations which would also need to be considered.
  • The metrics on promptness and accuracy of core financial transactions are similar to those trust-based schemes currently report, but go into far more detail. We have reservations about how easy it will be for administrators to provide this extra information and how much it will cost to obtain it.
  • We are not convinced that the negative perception metrics will be helpful as they do not address potential differences in what each service provider counts as a “complaint” or whether the complaints which are in scope are warranted. For example, one scheme could have several serial complainers bringing complaints which are without merit, while another could close complaints (even those with merit) promptly. Similarly, escalation to the Ombudsman does not necessarily mean that a complaint is valid or will be upheld.

Few own-trust schemes will have the resources to run saver satisfaction surveys, and the additional administration burden and cost of doing so could be disproportionate to the quality of the feedback they obtain. If this cost is passed on to members it will impact value, if it is borne by employers it could serve to hasten the demise of the scheme. As noted above, members’ feedback / complaints may not always be justified and those with strong and unreasonable views may be more likely to respond than those who are satisfied with the service they are receiving.

• We have some concerns with the metric on updating or reviewing active contributions. Where a member’s salary and living costs are fairly constant/linear, we would not expect savers to update or review their active contributions. In trust-based occupational pension schemes, contributions are often prescribed and there may be little or no ability for members to adjust their contributions. We also still have clients with schemes which are non-contributory (ie no member contributions). In addition, we are not convinced that being able to make changes to pension arrangements easily is always a good thing. It may not be in the members’ interests to be able to frequently switch funds or adjust their contributions at the touch of a button.
• We do not agree that “a scheme providing a good quality service is likely to have higher percentages of members actively engaged”. Engagement is very dependent on the scheme membership and not all engagement is substantive. We also note that this framework applies to default arrangements which by their nature will contain the less engaged members.

Further, we would also not expect many savers, even engaged savers, to contact their scheme or access a secure portal or application on a regular basis. In addition, whilst a scheme might not have a secure portal or application digital solution or at least not one which works particularly well, it may offer a subsidised advice service which would be far more helpful. A scheme could be offering a quality service, despite scoring poorly on these engagement metrics. We would presume other metrics would be available for schemes which do not offer a secure portal or app and note that currently many of our clients receive data setting out, for example, how many members opened targeted emails or clicked on website pages.