Feedback statement on driving Value for Money in defined contribution pensions

A feedback statement from The Pensions Regulator (TPR) and the Financial Conduct Authority (FCA) following the consultation responses to our joint discussion paper on value for money in DC pension schemes.

Published: 24 May 2022

FCA publications number : FS22/2

You can save the contents of this page as a PDF using your web browser. Open the print options and make sure the destination/printer is set as 'Save as PDF':

1. Summary

In September 2021, TPR and the FCA published a value for money discussion paper inviting views on developing a holistic framework and related metrics to assess value for money (VFM) in defined contribution (DC) pension schemes. Together with the Department for Work and Pensions (DWP), we see VFM as a key part of ensuring that savers get the most out of their pensions.

Our paper set out and invited comments on three key elements of VFM:

1. investment performance
2. customer service and scheme oversight
3. costs and charges

This feedback statement summarises the responses we received, which we are using to inform the development of proposals for consultation towards the end of this year. While we do not provide our policy response in this paper, we include next steps at the end of this chapter.

We think the time is right to encourage a more consistent and structured approach to VFM assessment that drives long-term value for pension savers. In the years following the introduction of automatic enrolment, and with the anticipated growth in the DC market, there was a necessary focus on cost. The DWP introduced the charge cap on default funds used for automatic enrolment, with the FCA implementing the cap for contract-based schemes. This was against a backdrop of some legacy schemes charging considerably more than the cap. At the same time, some commercial master trusts and providers have been competing on cost to win business from employers, with charges considerably under the cap. But an excessive focus on cost may result in the other key elements of value not being given appropriate consideration, which is why we want to encourage a shift in focus to long-term value for pension savers.

The focus on true long-term value is important for the ‘double defaulters’ of automatic enrolment. Automatic enrolment has brought over 10 million people into pension-saving, most of whom have made no choice to save or about where their money is invested. These people depend on their pension scheme or provider to make the right choices for them. For this reason, we think that those responsible for providing oversight of value should be supported in their focus on what matters most for pension saver outcomes. These are investment performance, service and oversight, and costs and charges, with the level of contributions a separate but equally important consideration.

It is not enough for pension schemes and providers to consider these factors in isolation. In our view, VFM is a relative concept. Effective assessments of VFM require external comparisons. This demands public transparency of data on key-value metrics from schemes and providers across the DC industry. As rule makers and regulators, we want to work towards a framework that allows industry stakeholders to assess and compare pension scheme value on a consistent basis, with data published in a format readily accessible to third parties such as trustees, independent governance bodies, investment consultants and employee benefit consultants. Over time, we envisage industry comparisons of VFM being published and will be learning from other countries about the best way to make this work (for example, Australia’s ‘league tables’ model).

Our VFM framework must work across all DC pensions, both workplace and non-workplace. However, our approach will be phased, taking into account the challenges of extending the framework more widely. At this stage, we are focused on VFM in accumulation in workplace schemes, and in particular default arrangements. We will consider further how best to extend the framework to self-select options in workplace schemes and to non-workplace pensions, as well as to pensions in decumulation. We remain convinced of the view that there is value in enabling consistent comparisons of VFM across all types of DC pension scheme, which will support engaged pension savers in making decisions about whether and how to transfer or consolidate their pension savings.

Our work on VFM in pensions sits alongside other regulatory and government-led initiatives. In March 2022, the DWP published Facilitating investment in illiquid assets, a combined consultation and responses to two previous consultations. In that paper, the DWP said it will not introduce any new regulatory requirements with the sole purpose of consolidating the market in 2022. The DWP’s focus will instead be on working with the FCA and TPR to create a framework that brings about consistent, member-focused value metrics to enable comparison and encourage competition on overall value. Also, in that paper, the DWP consulted on measures to require disclosure of asset allocations, which could be used in the VFM framework to support assessments of relative investment performance. The potential for including more long-term, less liquid investments in DC pension schemes has been a focus for the industry-led Productive Finance Working Group, which made recommendations in its report: A roadmap for increasing productive finance investment, published last year.

Other official sector work includes the FCA’s proposals for a consumer duty for financial services firms. In December 2021, the FCA published its second consultation on a new consumer duty, which includes an expectation that the price of products and services represents fair value. In pensions, we think that the development of a holistic framework for assessing VFM should complement the proposed requirements of the Consumer Duty as it would enable firms to compare and ultimately demonstrate that they are meeting or exceeding the fair value standard of the Consumer Duty.

VFM is one of two priority areas in the FCA and TPR’s joint regulatory strategy published in October 2018. As part of this strategy, we committed to pursue VFM throughout the pensions system and intervene where our expectations are not met. Our joint work on VFM in pensions also supports us in delivering our individual strategies.

In TPR’s Strategy: Pensions of the future, TPR has committed to putting the saver at the heart of everything it does. Our joint work on VFM supports TPR’s stated intention to work with its regulatory partners and industry to establish common standards on value for money, setting clear expectations and sharing good practice. In the FCA’s Our Strategy: 2022 to 2025, the FCA explained the four overarching outcomes it expects from financial services. Our joint work on VFM supports the delivery of one of the FCA’s consistent top-line outcomes for consumers – fair value.

Our aim for this work is to help create the regulatory conditions and disclosure requirements needed to allow those making choices on behalf of consumers to compare more effectively. Ultimately, improved disclosures should encourage better outcomes for consumers through higher standards and competition in consumers’ interests. We recognise that additional disclosure requirements incur costs, which would be borne by schemes and their savers, but we think the benefits to pension savers make this worthwhile.

The different views expressed in the feedback reflect the complexities of assessing VFM in pensions. However, most respondents agreed on the need for common and prescribed metrics and a consistent approach for meaningful comparisons. Over the coming months, we anticipate a need to work through detailed issues with stakeholders before finalising proposals for consultation towards the end of this year.

The feedback we received

We received 61 responses from a range of backgrounds including pension providers, trustees of occupational schemes, independent governance committee (IGCs), and their advisers. However, the FCA’s Financial Services Consumer Panel (FSCP) was the only consumer representative body to have responded. We held two roundtable events during the feedback period with a further roundtable being hosted by the Association of British Insurers. We also had a separate discussion session with the Investing and Saving Alliance.

We are publishing this feedback now, as an interim step, to give greater transparency around the responses we received and explain where there was a difference in views. This is to help stakeholders consider these issues in advance of an opportunity for further input in any consultation. We would also want stakeholders to consider whether they have any material that may support us and the DWP in framing a future cost-benefit analysis. We intend to seek further input from stakeholders over the coming months, including input from consumer representative bodies, before finalising proposals for consultation.

We are grateful to all those who took the time to engage with us, including the Pensions Policy Institute who TPR commissioned to carry out valuable research on VFM from an international perspective.

What we found

There was agreement that a holistic approach to assessing VFM was needed and that the three elements we had proposed were the right ones. Most respondents welcomed greater transparency and access to pensions data to support better decision making, provided that proper consideration is given to how this information would be presented to the target audience. However, there was less consensus on how to measure each of these elements.

Respondents expressed a desire for us to build on the work that has already been carried out in VFM assessment. Feedback to benchmarking was mixed, and many were, on the whole, not supportive of benchmarking VFM elements separately at this stage of our work. This was due to the importance of taking a holistic view of VFM and the complexity of developing benchmarks that can adequately account for the wide differences in scheme characteristics and investment objectives across the market. Where possible, these respondents were keen that regulators carefully consider whether there is scope to rationalise existing requirements rather than introduce new ones.

Investment performance

We proposed that public and consistent disclosures of investment performance data would better allow decision-makers to judge whether a scheme’s investment strategy was performing in line with reasonable expectations, and to consider if continued underperformance over the longer term would indicate poor scheme design and value for savers.

Many respondents agreed that consistent disclosure of performance metrics is necessary to enable better VFM assessment, although they also emphasised that the approach to disclosure must take into account the intended audience.

There were mixed views as to whether investment performance should be disclosed as net or gross of costs. Some respondents thought that while reporting performance net of costs would better reflect member experience, it could lead to consideration of investment costs twice if also assessed as part of the costs and charges element. In multi-employer schemes where employers might be charged differently for identical funds, some felt that net performance should be disclosed at the individual employer level, while others considered a range across employers more appropriate. Several noted that the variation in charges between employers is a result of different administration costs. Therefore, splitting administration from investment costs would enable consistent comparisons of net performance between employers.

Generally, respondents supported the mirroring of the DWP’s approach in terms of reporting periods. However, some cautioned against reporting performance using one-year intervals due to the risk of driving short-term decision making. Others pointed out that time horizons over 20 years were less relevant as strategies would have significantly evolved over time.

Since savers at different ages within the same default scheme are generally assigned different asset allocations to reflect their changing risk profile, they may experience different investment performance depending on where they are along their pensions journey. There was broad support for publishing performance metrics based on age cohorts. Many respondents agreed that reporting investment performance by age cohorts would help facilitate comparisons between schemes whose members have similar investment objectives.

Others felt that simple metrics based on age cohorts could be misleading for savers unless sufficient context was provided, for example, to explain how investment performance is also influenced by schemes using different de-risking periods and approaches to lifestyling. Respondents had different and sometimes strongly held views on what the age cohorts should be. Some considered the time to target retirement date, or state pension age if that was not known, to be more useful than the DWP’s approach of actual age, whereas others expressed a preference for retaining consistency with the DWP’s approach.

There was some support for disclosing the performance of self-select options, but a proportionate approach was suggested due to the relatively small number of members who are invested in self-select funds.

Most respondents agreed with the use of risk-adjusted returns to reflect the value added through risk management, although some argued that qualitative risk descriptions might be more appropriate for illiquid assets. Simple measures such as the Sharpe ratio were preferred, and some advocated for the risk metric to be considered alongside other measures, such as an annual geometric average, to provide further context.

There was no clear consensus around the use of benchmarking. Some felt it would help prevent schemes from only comparing themselves against poorer performing schemes, whereas others were concerned it could drive schemes to ‘cluster towards the middle’. Most agreed that other characteristics such as scheme size, investment objectives as well as member and employer demographics would need to be accounted for by benchmarks. There was general agreement that a commercial benchmark was likely to emerge over time, although concerns were raised about the possible quality of such a benchmark.

A number of respondents highlighted the importance of governance to investment returns going forward. They thought that, while hard to measure, the quality of governance could become the primary indicator of whether or not a scheme was likely to show good investment returns in the future. The assessment of governance should include whether it was forward-looking, which would help move the VFM debate away from a focus on past performance. We note the different views on how best to disclose investment performance. In preparation for our consultation, we will continue to engage with stakeholders to explore the following non-exhaustive list.

• Arguments for disclosing net investment performance and consider further which costs should be excluded under the calculation.
• Whether the metric should be backward-looking only or include consideration of expected future returns.
• Whether the publication of performance metrics by cohort is the best approach. If so, whether the criteria for cohorts should be based on age or years to retirement.
• The merits of prescribing a risk adjustment methodology and publishing other metrics, such as asset allocations, to support decision making.
• How comparisons and benchmarking should reflect broader elements of value, including ESG considerations and long-term investing.

Customer service and scheme oversight

We proposed that the VFM framework should also account for the qualitative elements of a provider’s offering, which contribute toward good long-term outcomes for members. We sought views on how regulators could support decision-makers in assessing the quality of services and scheme oversight.

There was strong support for a holistic view of VFM, which recognises the added value of factors other than investment performance and costs. Respondents generally felt customer service and oversight metrics will need to be wholly or partially qualitative and that this will complement the quantitative assessments. On the whole, respondents agreed that the service standards we had identified from TPR’s DC code were the right ones.

Most respondents emphasised that how we define good quality in service standards such as member communications and investment governance would be crucial to the effective measurement of VFM and ensuring that these metrics reflect member experience rather than devolving into mere tick-box exercises for schemes. There was consensus that focusing on the outcomes for members should be central in assessing the value of services provided. Accordingly, respondents suggested that assessments should be based on measurable metrics and expressed their support for the creation of benchmarks to allow for relatively easy comparisons of service levels to be made between schemes.

However, several pension providers expressed concerns over limiting the assessment of services to a minimum standard which would encourage a ‘race to the bottom’ as providers would be incentivised to only provide these minimums to keep their costs low. As a result, they emphasised that appropriate consideration should be given to the design of a service and oversight benchmark.

There were mixed views on whether the regulators should develop common service standards or whether these should be developed by industry with the assistance of a neutral convener. However, there was agreement that building on the work that has professional organisations have already carried out in this area would be desirable. There was a similar division around the introduction of independent certification.

As with the discussion around investment performance, we note that while many respondents agree that greater comparability of services would improve value, there is no consensus on how this should be implemented in practice. Therefore, as part of our consultation, we intend to explore further how best to capture the quality of service delivered, including communications that support pension saver understanding and decision making. Additionally, we need to consider how to assess the value of other important components of services, such as administration and governance.

Costs and charges

To promote consistent reporting of costs and charges across the market, we proposed to use existing definitions which underpin current disclosure requirements that already apply to DC workplace pension schemes. We sought views on how this information should be presented or broken down to facilitate meaningful comparisons and drive better outcomes for members.

Most respondents were in favour of continuing to use the current definitions of administration and transaction charges. They argued that existing criteria are adequate, and any changes would result in confusion. Some pointed out that a further split would be difficult to achieve for schemes with bundled arrangements or products.

However, some supported the option to split the definition of administration charges into two components: fund management and pensions administration. They argued that clearly showing the components that make up the charging structure of schemes is not only important for understanding where value could be improved and where schemes deliver exceptional value but also in ensuring that savers develop trust in the industry.

Most respondents felt that creating a benchmark for costs and charges might result in a disproportionate focus on that area and drive a race to the bottom. They felt that clear disclosure of costs and charges should be sufficient.

There were also calls to build upon the work that has already been carried out by professional organisations in this area, such as the Cost Transparency Initiative (CTI) framework.

While most agree that using existing definitions for administration charges and transaction costs would prove helpful, we need to consider whether effective comparisons require more detailed information. We also need to give further thought as to whether it is necessary for investment costs to be disclosed and assessed separately if they are already netted off under-investment performance, or whether comparisons of costs and charges should only focus on administration.

Next steps

The number and quality of responses have given us a foundation to advance our consideration of the issues raised in the Discussion Paper (DP). However, there are a number of areas where there is no clear consensus as to the way forward. We will continue to develop our thinking on how to address these issues and consider which metrics schemes should be disclosing to facilitate more effective VFM assessments by professional audiences such as IGCs, trustees and employee benefit consultants. We also need to consider how this disclosed information could be adapted to assist engaged savers in choosing self-select options and non-workplace pension products.

We will review these issues alongside other related initiatives such as the DWP’s work on small pots consolidation, proposals for asset allocation disclosures, the FCA’s new Consumer Duty, the pensions dashboard programme and industry-driven solutions. As we develop the VFM framework, we want to ensure that:

• stakeholders take a holistic view of VFM in the interest of long-term saver outcomes
• appropriate and comparable reporting metrics are developed which help drive better decision making on behalf of members in workplace pensions and which can also be extended to non-workplace pensions at a later stage
• meaningful comparisons between schemes are facilitated
• consideration is given to how this approach could be aligned with VFM assessment of pensions in decumulation

The questions we need to consider are complex. It is important that we make progress. Nonetheless, it is also important that any changes we ultimately decide to make have a lasting impact, so we need to give these questions the careful consideration they demand. We will start a programme of further work to examine the options we think are likely to be the most effective and proportionate. This will mean that we need to engage further with the industry to enable us to fully understand the potential impact of our proposals on the industry and savers.

The Government has expressed its willingness to legislate to introduce the VFM framework for the schemes regulated by TPR, and we will work closely with DWP to achieve this. We aim to consult on our proposals towards the end of 2022.

Who should read this feedback statement?

This feedback statement will be of interest to those who are involved in the workplace and non-workplace pensions markets. This includes:

• pension providers and asset managers
• the governance bodies of pension schemes, such as trustees of occupational pension schemes, Independent Governance Committees (IGCs), and their advisers
• scheme savers and their advisers
• saver and employer representative groups

2. Measuring and reporting investment performance

This chapter summarises the feedback we received to the questions asked in our discussion paper around the issue of measuring and benchmarking investment performance.

The case for disclosing performance data

In the discussion paper, we said that long-term investment returns are a key component of delivering good outcomes for savers. Having access to investment performance data, which indicates the returns achieved on a scheme’s behalf to date, is important to decision-makers in pensions. Public disclosure of this information will enable them to make judgements as to whether a scheme’s investment strategy is performing in line with reasonable expectations. We also said that it is important to achieve genuine comparability for groups of similar savers.

We asked the following question:

Q1: Do you agree that consistent disclosure of performance is necessary to enable better decision making?

The majority of respondents, including a consumer representative body, agreed that consistent disclosure of performance is necessary. However, many wanted more clarity around the intended audience, as this would greatly influence the nature of disclosures and VFM metrics.

Many agreed that the intended audience for disclosures on performance metrics should be pension professionals such as governing bodies, providers and consultants, as they are better equipped with the skills and knowledge to assess investment performance in context. Such information may be less useful to scheme members, at least initially.

Respondents also noted that a producing a single performance figure for each scheme would not provide insight into the broader VFM picture, such as: the design of the strategic asset allocation, member risk profiles and investment objectives, ‘lifestyling’, and long-term considerations such as Environment, Social and Governance (ESG) or illiquid asset holdings. This point is discussed further under Question 5. Several also commented that a performance metric should be produced against carefully considered time periods that apply consistently to all schemes, with thought being given to the frequency of investment reporting. This would prevent scheme providers from selectively reporting across periods of favourable returns.

Many also reiterated that investment performance metrics are backward-looking and do not serve as reliable indicators of future performance, especially over long investment horizons. Taken out of context, they felt these metrics could drive short-term decision-making and poor consumer outcomes.

Developing comparable metrics

In the discussion paper, we suggested that past performance data should be disclosed net of costs and charges because this reflects the savers’ experience of investment return. We recognised that varying charges for the same funds within multi-employer schemes mean that different employers investing in identical funds could have different net investment returns.

Recently enacted legislation already requires all relevant occupational DC schemes to publish their net investment returns on an annual basis. It is one of the factors that must be considered by schemes with less than £100m AUM when carrying out the more prescriptive annual VFM assessment.

We asked the following questions:

Q2: Do you agree that comparisons should be of net rather than gross investment performance?

Most respondents agreed that it is net rather than gross investment returns that will ultimately drive member outcomes; therefore, performance should be reported net of the total costs. Several respondents, however, thought that a professional audience would benefit from both net and gross returns. Many also noted that this audience would generally benefit from more nuanced and detailed information for the assessment of value, including on investment performance.

Several wanted further clarification on the definition of net and whether this should include administration, investment, and transaction costs, as well as how contribution charges, fixed fees and asset management charges should be included in the metric. They also noted that it is the administration cost that varies from scheme to scheme and not the investment costs.

A group of respondents also noted that comparing investment performance net of fees as one component and costs and charges as another would result in double-counting of costs and charges. To avoid this, they recommended that investment fees should be considered when calculating net performance while administrative and service charges should be considered when assessing costs.

Q3: Do you have any suggestions on how to make disclosure of net investment returns effective, given that there may be varying charges for the same funds within multi-employer schemes? For example, displaying a range, or requiring disclosure of each different level of net investment performance

On this question, the feedback was more diverse, with several different options suggested.

One group consisting of pension providers and consultancy firms supported the requirement for multi-employer schemes to display a range for investment performance net of the range of charges across all schemes by quartiles, noting that this would be a more pragmatic and proportionate approach.

Another group of respondents, including the consumer representing FSCP, argued that reporting each level of net performance by employer would provide more information than a range. These respondents noted that if a scheme prices differently by employer it should report those costs differently. Without transparency on fees and pricing models, they felt it would be difficult to make a comparison between the schemes, which would ultimately erode competition. Others, however, noted that disclosures at the employer level could become onerous and add to the cost of running the scheme.

Several noted that splitting the investment and administrative costs would enable consistent comparisons and provide transparency where administration charges vary between employers.

Others suggested disclosure of net investment returns within cohorts of similar employers/schemes defined by size, salary, contribution levels and staff turnover, rather than at the individual employer level.

Q4: Would it be helpful to mirror the DWP’s approach in terms of the reporting periods?

We proposed aligning reporting periods with DWP’s legislation requiring that performance of schemes be disclosed over at least one and five years, and where possible, also over 10, 15 and 20 years. The feedback to this question was mixed. Most respondents agreed that aligning potential periodic reporting across the market with the DWP’s approach would result in administrative efficiency and lower cost. They also noted that savers would be highly unlikely to recognise the difference between FCA and TPR-regulated pension providers, so from a consumer perspective, it makes sense to disclose performance over a consistent set of reporting periods, regardless of the regulatory environment providers operate in. One scheme provider disagreed with this statement, noting the significant difference in the way trust and contract-based schemes operate, necessitating different approaches.

Many also argued that, due to the long-term nature of pensions, the comparisons should be made over longer periods than one year, with a minimum of five years being suggested and longer periods being included where possible. These respondents thought that short-term reporting could drive the wrong behaviour from savers by causing them to prioritise short-term returns.

However, another group, mostly consisting of scheme providers and consultancy firms, felt that longer time horizons above 10 years have less relevance as, for example, investment strategies and approaches evolve over time. It was also noted that, in practice, some funds will not have long term performance track histories, especially out to 20 years.

One scheme provider said that discretion on reporting should be allowed depending on the age of the scheme and the date that a specific fund or investment strategy was launched. Several others, including some governance bodies and consultancy firms, pointed out that performance is currently reported at each scheme’s year-end to align with the annual chair’s statement, and they suggested introducing a universal reporting period so that all schemes would report at the same time.

Developing comparable metrics: the case for disclosing more than one metric per scheme

In the discussion paper, we recognised that most default funds are ‘lifestyled’. This means that different savers at different ages in the same scheme and in the accumulation/growth phase of their savings journey may experience different performance from their pension product’s overall investment strategy. During the pre-retirement phase of savers’ pension journeys, even those of the same age with the same pension provider may experience very different performance, for example, where savers have the option of selecting a lifestyling strategy other than the default.

We suggested that scheme investment performance data provided on age cohorts would bring reporting into closer alignment with the investment return outcomes that pension savers experience, based on where they ‘sit’ within pension products.

As an example, we suggested cohorts that group members by the number of years from their target retirement date: zero to five years, five to 20 years, and more than 20 years away. Alternatively, we said that we could follow the approach used in the DWP’s statutory guidance, which requires relevant schemes to publish age-specific results for savers aged 25, 45 and 55, where net returns vary with age.

We asked the following questions:

Q5: Would publishing a set of metrics based on age cohorts bring investment performance reporting closer to the saver’s investment performance experience of a pension scheme/product? If not, is there a better alternative we have not considered?

The feedback to this question was mixed.

Many respondents broadly supported publishing the metrics based on age cohorts, noting that it would result in better comparison across schemes and products. Given that there are age-related differences in asset allocation, this, in theory, should result in the performance of groups with similar asset class mixes being compared.

However, some pension providers and consultancy firms disagreed, arguing that this approach could result in inconsistent asset class mixes being compared. They pointed out that lifestyle strategies may have different de-risking periods and non-linear changes of asset allocation. Schemes could have different approaches to blending the different stages of the glidepath, making comparison between schemes ineffective.

Several respondents felt that, in addition to age considerations, earning and contribution levels should also be factored into assessments of performance, as this could further support decision makers in determining whether the level of risk is appropriate to the scheme’s membership. This information could also be used for more targeted assessments to help identify value for money issues in certain saver cohorts, such as for lower-paid members.

Several others highlighted that context was critical to ensure like-for-like comparisons. They also noted that short-term comparisons should not become the focus of VFM assessments. For example, they suggested that investment in ESG funds, illiquid investments or productive capital are factors that are likely to affect short-term performance and may lead to inconsistencies in comparisons.

One pension provider cautioned that the disclosure of performance by age cohorts could allow trustees or IGCs to justify a particular investment strategy by citing strong performance achieved for a particular age cohort while downplaying poor returns experienced by another cohort.

Q6: When considering which age cohorts to consider, is the example we have provided appropriate? Alternatively, would it be more effective to mirror the DWP’s approach?

The DWP’s approach requires publication of net returns for savers aged 25, 45 and 55, where returns vary by age. Several respondents argued that consistency with this approach would be helpful for minimising complexity in the market and administration burdens on schemes.

Many others preferred the approach proposed in the discussion paper to use years from target retirement date. These respondents noted that, in most cases, the investment and de-risking strategy is based on the time to target retirement rather than the member’s specific age. However, they felt that proximity of the member to retirement date would need to be more precisely defined, as the fund performance over any reporting period will be dramatically different for a member 20 years from retirement compared with a member five years from retirement. One respondent suggested the use of a spread of points, so returns are measured during accumulation, during de-risking, and after de-risking.

Several respondents also noted that the most appropriate method would depend on the target audience. The DWP’s approach could be more meaningful for individual savers, whereas disclosure in terms of years to retirement is more appropriate for IGCs, trustees and professional advisers.

Another respondent suggested a similar approach but allowing some flexibility with a ’stage’ framework, so schemes could prepare the performance information for early growth, mid growth, pre-retirement, and at-retirement phases.

Another respondent noted that consideration should also be given to a ’zero years to retirement’ cohort, referring to those who choose not to undergo any de-risking, as reflected by many members who remain invested in preparation for drawdown.

One respondent noted that in the interests of making the framework simple and intuitive, the number of cohorts or ‘model points’ should be minimised.

Q7: What disclosures, if any, should be made for self-select options?

Responses were evenly split between those in favour and those against requiring investment performance disclosure for self-select options.

Some respondents, including the FSCP, supported the view that the approach applied to default schemes should be required for self-select funds to ensure consistency and allow for comparison of all pension products. They pointed out that trustees and IGCs are responsible for monitoring and reviewing all investment options, and therefore not including self-select funds may result in these funds being overlooked as part of the review process. Although the proposals at this stage are largely intended to support decision making by pension professionals, a few respondents also noted that disclosures might prove useful to members in self-select funds, who are more likely to be engaged in monitoring the performance of their fund. Some respondents also noted that this would have the potential to highlight the impact of self-selection on savers and enable them to see whether their choices were better value than the default.

Several respondents also noted that the DWP’s regulations currently require publication of the performance of self-select options as part of the VFM assessment included in the chair’s statement. Alignment with the DWP’s approach was welcomed. However, respondents also noted that disclosure for self-select options could present practical challenges for providers due to the volume of variables within such a small population. It would also potentially increase reporting complexity by introducing another metric. The FCA and TPR were encouraged to work with providers to determine the reporting requirements for this population and ensure it can be achieved in a proportionate way.

To address the issue of proportionality, some respondents recommended that disclosure for self-select options should only be required in relation to the most popular option.

They suggested that analysis would be needed to determine whether the most popular option should be based on the percentage of members invested or amount of assets, as well as the appropriate level of disclosure (employer or scheme). They also felt that this could be a suitable approach for non-workplace schemes.

Many other respondents, however, disagreed, arguing that it would be disproportionate to reflect the huge variety in self-select options.

A few respondents suggested that the regulators should leverage the existing Undertakings for the Collective Investment in Transferable Securities Key Investor Information Document (UCITS KIID) and (Packaged Retail Investment and Insurance Products Key Information Document) PRIIPS KID regimes which mandate performance disclosures for retail investment product providers. They felt that the annual chair’s statement, factsheets and investment guides were sufficient for member decision making. As such, they argued against layering further requirements to promote consistency and avoid confusing members.

Several respondents also suggested that there could be an opportunity to align self-select options with the required member illustrations included in the chair’s statement to recognise the importance of these options for some savers.

Developing metrics: risk adjusting return figures

In the discussion paper, we stressed the importance of capturing the risk element of performance to allow for effective comparison. We believe it is important for decision-makers to be able to assess whether one investment strategy produces an equivalent or superior return per unit of risk to that of another over a defined period. This is especially true if data is collected in a way that allows for the experiences of similarly aged investors with similar glidepath targets to be compared side-by-side.

The DWP’s statutory guidance requires net investment returns to be presented as an annual geometric average but allows schemes to show risk-adjusted returns as well if they believe it would be helpful to do so

We recognised there would be challenges in determining a standardised approach to assessing risk. However, we pointed out that the disclosure of net investment returns without risk adjustment could be misleading or lead to firms being incentivised to take excessive risk.

We asked the following questions:

Q8: Do you think reporting based on age cohorts would be enhanced through the use of risk-adjusted returns as an element of a scheme’s VFM assessment, or would risk adjustment then be unnecessary?

The response to this question was mixed.

Many respondents agreed that risk-adjusted disclosures give a more accurate indication of how a scheme generates its returns and whether its strategy is suitable for its members. They noted that risk adjustment will reflect the value added through risk management, especially in periods of increased volatility. However, many respondents noted that risk-adjusted returns would only support decision making for professional audiences and would likely confuse retail consumers.

Others felt that reporting risk-adjusted returns represents little value for members and is likely to add complexity. These respondents noted that in the absence of a consistent methodology prescribed by the regulators, comparisons would not be achieved. Some believed that risk-adjusted ratios would not act as a reliable proxy for the comparison of schemes’ individual investment strategies, as they vary according to the specific needs of their membership.

Several respondents noted that longer-term illiquid assets cannot be valued frequently. Therefore, most of the risk-adjusted measures suggested in the discussion paper would not be applicable to them. Instead, they argued that qualitative risk descriptions may be more appropriate for illiquid assets.

Alternatively, several financial services firms suggested aligning with the risk measures reported under PRIIPs or UCITS as this would facilitate easier comparison with investments held through other channels and provide a consistent metric for savers.

Q9: If risk-adjustment is used, what risk-adjustment metric(s) would you suggest? For example, the Sharpe ratio as i) a standalone factor, or ii) in combination with other risk metrics?

Many respondents broadly agree that the Sharpe ratio is the most widely known, best understood risk-adjustment metric across the industry and should be used for the VFM assessment.

Several respondents noted that risk and return should always be presented alongside each other, as it is important to understand what the relative impact of both excess return and risk management has been. They also noted that the key parameters will have to be clearly defined, especially the frequency of volatility calculation (e.g. daily, weekly and montly) and also what should be used as the risk-free rate of return.

Additionally, different risk-adjusted metrics may be more suited to different member cohorts. They pointed out there are some complexities associated with calculating performance and risk-adjusted metrics for lifestyle strategies.

Among the alternative solutions respondents suggested such options as:

• the PRIIPs or UCITS risk indicators
• absolute performance
• historic volatility
• maximum drawdown
• annualised standard deviation of monthly returns, matching each of the periods for which investment returns are reported – i.e. three-year, five year and ten years
• a standardised format where ex-post risk (realised volatility) and ex-post return are presented in a two-dimensional chart, produced over different time frames.
• a simple return/volatility measure that is easier for members and employers to understand, although it can be problematic when dealing with negative returns
• information ratios as the simplest measure of value for risk

Q10: Is there any reason why it would be impractical to report on risk-adjusted performance metrics in addition to providing a metric based on actual performance returns?

Respondents generally did not raise concerns around reporting risk-adjusted performance in addition to absolute returns. Many also stressed the importance of striking the right balance between long and short-term value assessment. Several noted that it is important to maintain a consistent approach to reporting across schemes and therefore thought that the choice and calculation of a risk-adjusted metric should not be left to individual asset managers.

Some respondents did not see the need to report on these metrics because, in their view, trustees and scheme sponsors can access this information through advisers and pension providers.

Q11: What are your views on presenting returns as an annual geometric average to provide consistency with the DWP’s requirement?

Most respondents supported alignment with existing DWP requirements, noting that returns are commonly presented as an annual geometric average on documents provided to investors. However, they highlighted the same overarching comment that any measure of investment performance should not be used in isolation and that decision-makers need to take a more holistic view of individual scheme characteristics and demographics. Some noted that the longer-term nature of these metrics would be preferable to short-term metrics and emphasised that a consistent approach to time periods would be needed when carrying out comparisons.

Another group of respondents disagreed, highlighting that extreme volatility may present some issues for risk reporting. There would also be challenges with incorporating risk-adjusted performance numbers in conjunction with a geometric average. A few pointed out that members join and leave schemes at different stages in their working lives, and so any one average will not be representative of the returns experienced by most members.

Comparison and benchmarks

In the discussion paper, we said that disclosure of performance would drive better comparison between schemes. We expressed a desire to foster market-driven solutions through transparency rather than reporting to us. We felt that benchmarking is important and makes comparisons easier. Although the use of specific benchmarks for comparing investment performance could be prescribed, we felt that commercial benchmarks would emerge. We provided a number of different options for discussion.

We asked the following questions:

Q12. We would welcome views on how you see this developing. Would it be helpful/possible to establish a benchmark, or would you prefer to compare cohorts against a market average or against a few selected similar schemes? If so, how would that selection be made?

While some respondents agreed that a common benchmark would be useful for preventing firms from selectively comparing themselves against poorer performing comparable schemes, there was no consensus as to whether benchmarking investment performance would improve decision making and member outcomes.

Many respondents from the industry thought that the use of benchmarks was unlikely to improve decision making as it will be difficult to account for different scheme characteristics such as scheme size, member and employer demographics in a single benchmark. Therefore, using a single comparator such as a market index as an industry-wide benchmark could lead to unintended consequences. For instance, such a benchmark might stifle innovation by causing investment strategies to ‘cluster towards the middle’ or drive schemes to focus on short-term performance in order to exceed the benchmark. Investor confidence might also be shaken where the chosen industry-wide benchmark performs poorly. However, several others noted that performance evaluation cannot be conducted in a vacuum, and the performance of a reasonable benchmark should be included.

Many respondents, predominantly from the industry, said that the non-homogeneity of scheme members, charging structures and the shape of investment strategies would require an unwieldy number of carefully designed benchmarks to facilitate any meaningful comparison, and thus benchmarks should not be the focus at this stage.

A few respondents recommended further exploring the trade-offs between consumer-level policy objectives and broader macro-prudential policy objectives, as asset owners could potentially be driven to adopt similar investment strategies as a result of benchmarks.

Several respondents sought further clarification around the policy objectives of benchmarking investment performance, and whether our aim is to facilitate discretionary or prescribed:

• comparisons against market indices
• comparisons against peers using a market or cohort average, which would require further consideration around the reporting frequency due to the significant lag between data disclosures and the calculation of benchmarks
• comparisons against economic trends and bespoke benchmarks (e.g. Consumer Price Index + X% to achieve the Pensions and Lifetime Savings Association Retirement Living Standards), in which case any such targets would need to be reviewed frequently to ensure these appropriately reflect changing market conditions

A few respondents argued against prescribing requirements to compare investment performance between schemes, given that the role of IGCs and trustees is to consider the prospective capability of default investment strategies to generate returns that are appropriate to the scheme’s membership profile and to support this with a robust rationale.

These respondents argued that given the importance of flexibility to the trustee model, IGCs and trustees should be allowed to select their own benchmarks or, where necessary, to engage an external evaluator to compare their scheme against the benchmarks that most closely align with their scheme’s investment objective, strategic asset allocation and member demographic.

To facilitate meaningful comparisons using benchmarks, respondents felt that this approach would require schemes to report performance over the same evaluation periods and to apply a consistent risk adjustment methodology. Some respondents also suggested that additional supplementary information would be required to support informed decision-making. These include:

• asset mix, strategic asset allocation or investment objectives
• stage of lifestyling or maturity of investment
• scheme benefits (protected pension age, annuity rates or protected tax-free cash allowance)
• employer contributions/subsidies
• average pot size/contribution rates

Q13. Do you think a commercial benchmark is likely to emerge if these data are made publicly available?

Many respondents agreed that commercial benchmarks are likely to emerge over time, pointing out that aggregate fund level benchmarks have already emerged for certain markets such as master trusts and group personal pensions. However, respondents pointed out that the quality of benchmarks would strongly depend upon the appropriateness of disclosure requirements and whether these requirements make computation and collation manageable and commercially viable.

A few respondents highlighted the possibility of aligning any VFM data publication requirements with incoming requirements relating to Pension Dashboards while also stressing the need to carefully consider how such information might drive consumer behaviours and decision making.

Others emphasised the importance of having a robust framework and methodology to ensure data is processed consistently to prevent gaming or manipulation regarding the development and subsequent maintenance of benchmarks.

Some respondents questioned the independence and utility of commercial benchmarks for supporting VFM comparisons, given how these would be skewed to cater for the needs of specific market segments.

Several respondents stated that commercial benchmarks would not lead to better consumer outcomes due to the risk of driving undesirable behaviour from schemes to deliver investment returns which match the benchmark instead of focussing on the specific needs of members and qualitative elements of VFM. Some firms also commented that such benchmarks might discourage investment into innovative ESG funds or long-term illiquid assets, thus making them incompatible with the government’s recent aspirations.

3. Measuring and reporting customer service

This chapter summarises the feedback we received to the questions asked in our discussion paper around customer service and scheme oversight.

Member communication

When discussing what constitutes good value for pensions service, we stated that timely member communications provided in an accessible format are vital as a means to encourage members to save more and engage more deeply with choices that lead to good retirement outcomes. For instance, we highlighted how the provision of effective channels for members to access information about their pension and accounting for member contact preferences might constitute good communication.

We asked the following question:

Q14: Do you agree the quality of communication is a relevant factor to consider in VFM assessments?

Most respondents agreed that the quality of communication is essential to improving long-term member outcomes and is a key element in the assessment of VFM. They stated that communication influences member behaviour and decisions at every stage of the retirement journey. For example, several respondents thought that communication could positively influence the level of member contributions, specifically in the workplace pensions market where members are less engaged with their contributions, thereby driving improved member outcomes. They said that the VFM framework should require schemes to develop and assess their communication strategies against those of other schemes, which will result in improvements across the industry and ultimately better outcomes for members.

However, many noted that ‘quality’ is subjective, and therefore, it will be difficult to objectively measure, assess and quantify it to produce meaningful benchmarks and comparisons. Measuring ‘quality’ becomes more difficult because of the diversity in membership profiles that each require different content and methods of communication.

Most respondents said that the key point to measure and assess is not the communications themselves but whether or not they drive better member outcomes. This could look like whether members, as a result of the communication, have increased their contributions, pursued further guidance or advice, or taken an active investment decision to align their investments with their needs.

To this end, some respondents suggested measurable outcomes against communication objectives, including the volume of members who increase contributions, update/confirm their selected retirement date, confirm how they wish to take benefits, update their expression of wish form, and sign up online and regularly log on to monitor their investment performance. These quantitative metrics directly assess the impact of communications on effectively engaging members in key areas that could have substantial consequences on member outcomes. This approach to ‘quality’ would provide decision-makers with the necessary insights to robustly judge performance and design strategy for improvement.

Similarly, another group of respondents suggested measuring the levels of engagement, as the preferred output from quality communications, instead of focusing on individual communications, based on the theory that communication is effective as long as members are actively engaging with the scheme. Other respondents disagreed and suggested that effective communication should include evidence of an individual’s understanding and changes in member behaviour to eradicate any biases specific to a scheme’s membership.

Respondents suggested that any proposals to establish standards and measurable metrics for the quality of communication need to consider the following.

• A wider definition of communication, including nudge communications, guidance, facilitating regulated adviser support, and financial education support. Communication should also cover how the scheme responds to member feedback and takes into consideration members’ views on a range of issues such as ESG matters
• Measurement against criteria that include the availability of different channels, accessibility, format, usefulness, timing and understandability.
• A requirement for schemes to tailor their communication to the specific needs of the different membership groups.
• The importance of employers as a critical source of communication.
• Avoiding becoming a ‘box-ticking exercise’, and the framework to not discourage schemes from going beyond the minimum standards if it results in material value creation for members.

Respondents would also like the regulators to consider the insights generated by recent initiatives in this area, including the FCA’s Consumer Duty as well as the FCA and TPR’s joint work on the Consumer Journey. Many respondents argued for greater convergence in communication regulations.

Scheme administration

In the discussion paper, we pointed out that all schemes are expected to meet certain basic administrative requirements, which most trustees tend to outsource. However, firms cannot outsource responsibility for the quality of administration, and we explained how poorly administered schemes and delays to core transactions could result in material consequences and a loss of confidence from members.

Even if interactions between a scheme and its members are infrequent, we explained how better administrative quality could provide crucial value for members at key moments, such as when transferring out of a scheme or when accessing funds at retirement. We asked:

Q15: Do you agree administration is a relevant factor that contributes to long-term VFM?

Almost all respondents agreed that administration is a relevant factor that contributes to long-term VFM. Efficient administration is critical to ensuring a positive experience for employers, trustees and members.

They pointed out that high-quality administration results in added value through accurate record keeping, effective and timely execution and investment of contributions, and the availability of informative and engaging servicing tools. Poor administration can affect the accuracy of member data and the ability of members to interact effectively with a scheme because of the errors and delays, thereby resulting in value detraction. Additionally, administration makes up a significant proportion of costs, and therefore inefficiency results in higher costs, which are passed on to members. Poor administration also results in loss of confidence, which could affect engagement and contribution rates. Therefore, respondents explained that high-quality administration alongside high-quality communication, as proposed in the discussion paper, will ultimately result in better member engagement and long-term value.

However, a few pension providers highlighted that schemes operating on a retail investment platform have higher customer expectations for administration standards compared to others. For instance, it was pointed out that savers using retail investment platforms tended to interact much more frequently with the scheme. Administration is also undertaken directly by the investment platform provider and covers a broader range of investments and span of activities than those typically associated with most workplace pension providers, such as live valuation of investments as well as access to investment analytics tools.

There was a consensus among respondents that the VFM framework must include a section to reflect the quality of administration, and therefore the outcome metrics need to be established. Quality should consider, among other things, accuracy, timeliness and member feedback. As with the quality of communication, respondents expressed that it would be challenging to assess the quality of administration using measurable metrics. However, they noted that some administrative services lend themselves to quantitative measurement, and these could provide a starting point for the regulators. For example, some schemes report on the time it takes to execute core financial transactions and other service metrics, including waiting time for members on the phone.

A group of respondents argued that all schemes should be expected to at least meet a minimum standard of service. This approach distinguishes between the standard of service and level of service. The level of service refers to the features that members choose to pay for. This approach, arguably, takes into account the different expectations of members, which will vary between different memberships and at different life stages within the same membership.

A few consultancy firms highlighted that the current method of assessing and evaluating administration might need to be modified. Although the traditional service level agreement (SLA) is useful, it fails to consider member experience, and as a result, other metrics like net promoter scores are becoming more widely used rather than SLAs.

Other respondents also expressed their support for cross industry initiatives like STAR and Origo to drive transparency and comparability, such as through agreed timescales for completing pension transfers. Finally, a number of respondents highlighted that this area falls within the new FCA ‘Consumer Duty’, which will challenge firms to measure outcomes through every stage of the consumer journey, and any proposals should take that into consideration.

Governance and oversight

In the discussion paper, we stated that scheme governance bodies for workplace pension schemes have a duty to properly account for the specific characteristics, needs and objectives of savers when assessing the design and monitoring the effectiveness of default investment strategies. Appropriate consideration should be given to factors such as the need for diversification, the choice of asset classes and level of allocation, the risks and expected returns of underlying investments, ESG factors, glidepaths and lifestyling.

Although effective governance and oversight should be demonstrable and visible from the scheme’s long-term investment performance, we sought views on whether trustees and IGCs may require more qualitative factors for making effective VFM comparisons in the short term.

Q16: Do you agree the effectiveness of governance is a relevant factor that contributes to long-term VFM?

Almost all respondents agreed that the effectiveness of governance is a relevant factor that contributes to long-term VFM. Respondents consider it self-evident that a poorly governed scheme will not deliver long-term VFM for its members. Many respondents argued that governance is the most important single factor in delivering VFM because good governance is capable of addressing all the other issues in the long run.

A number of respondents highlighted that the quality of governance could be the prominent indicator of whether or not a scheme is likely to show good investment returns in the future. The assessment of governance must include an element of looking forward, and this would, in fact, help move the VFM debate from focusing on past performance.

Overall, the majority of respondents agreed that the value-add of governance oversight should be established and monitored against a recognised and consistent benchmark. However, there was a consensus that it would be challenging to reduce the governance dynamic to a quantitative metric.

Many respondents suggested that the quality of governance is already reflected in other aspects of the VFM assessment, namely investment performance, costs and charges and administration standards, and therefore governance should not be included as a separate standalone component of the VFM assessment. A number of respondents argued that the current legislative and regulatory requirements already assess the effectiveness of governance, and therefore there is no need for further transparency in this area.

Q17: In your opinion, are there any obvious service standards missing from the above list? Please explain how your suggestion contributes to scheme value.

There was a consensus among respondents that there were no obvious material standards missing from what has been proposed in the discussion paper and that any other factors would likely be considerably more nuanced and near impossible to measure and compare consistently.

However, several respondents suggested that the VFM framework should adequately consider the following additional factors that contribute to long-term VFM:

• sustainability, including ESG considerations, stewardship and carbon reporting
• decumulation and ‘at retirement’ support and product options, including ease of access to retirement options
• investment range options
• liquidity and risk management
• support for vulnerable customers
• advice and guidance solutions
• scheme security, including cyber security, operational resilience and scalability
• employer impact on VFM
• additional features and benefits, for example, the annual addition of profit share

Challenges to assessing good value in customer services through regulatory disclosures and considerations on appointing a neutral convenor

In the discussion paper, we explained how the highly qualitative nature of service metrics and the degree of subjectivity associated with assessing customer service make it challenging for us to enshrine specific methods of measurement in regulation. We suggested that we might be better placed to use our role to encourage industry efforts, such as those from the Pensions Administration Standards Association or the Audit and Assurance Faculty, to develop meaningful service standards instead of directly introducing requirements for service-related regulatory disclosures.

We sought opinions on whether employing the expertise of a neutral convenor, such as the British Standards Institution (BSI), would prove helpful. However, we understand that this option would also incur costs that stakeholders would need to consider. We explained that we would not discount a more interventionist approach where no clear consensus was found around developing a service standard.

We asked the following questions:

Q18: Do you agree this is not a role for the regulators at this stage?

Responses to this question were split into two opposing points of view.

On the one hand, some respondents argued that regulatory intervention is not necessary or appropriate at this stage and that regulators should avoid prescribing standards of service in regulation because of the challenges inherent in comparing and enforcing wholly or partially qualitative metrics. These respondents also felt regulatory intervention would not be flexible enough and would amount to box ticking. Instead, they argued that developments in this area should be led by the industry and driven by investors requiring VFM disclosure as a perequisite to their investment, with involvement from the regulators limited to areas of reporting where non-compliance or data integrity issues are identified.

On the other hand, some respondents considered that the lack of regulatory involvement means that failures in this area often come with limited remediation. These respondents stressed that the market is unlikely to make the necessary investment in developing consistent standards unless there is a regulatory requirement.

A number of respondents argued that the regulators should at the very least steer and enforce best practice by issuing industry-wide guidance on helpful metrics. Alternatively, they thought that regulators could articulate and mandate the desired objectives of these standards and then allow the industry the flexibility to innovate to achieve those objectives within a set time period, thereby allowing for a sufficient level of subjectivity and proportionality.

Q19: Would it be helpful to appoint a neutral convenor to develop a service metrics standard? If not, who do you think should create metrics on service in pensions?

Views from respondents were roughly evenly split between those who thought that it would be helpful to appoint a neutral convenor to develop a service metrics standard and those who thought that it is the responsibility of the regulators.

Respondents suggested that a neutral convener with industry expertise should oversee the development of a service metrics standard, with input from the industry and regulators to create realistic expectations and avoid unnecessary burden and significant costs. This would be a more inclusive, independent and transparent approach. Additionally, this would allow the industry to leverage existing services by commercial providers in this area and allow for the sharing of data.

Some respondents suggested that the BSI would be well suited to develop the service metrics standard not only because of its vast experience but also because it has an established process and sufficient resources to address the key concerns and deliver in this complex area of work.

However, some respondents expressed doubt that a neutral convener would be better equipped and positioned than the regulators to develop this highly contentious standard. The convener, regardless of its expertise, will lack the detailed understanding of the industry expected of the FCA and TPR. Several other respondents cautioned that any benchmark established by a single professional organisation that is not a regulator could not be effectively enforced across the entire industry.

Alternatively, several respondents suggested that focused industry working groups, such as the small pots working group, have proved to be successful in bringing the industry together to consider complex issues, and therefore a similar approach should be adopted to develop a service metrics standard.

Benchmarking through independent certification against a standard

Where a common standard for service has been agreed upon and established, we asked whether it was desirable that trustees and IGCs disclose against the standard through obtaining independent certification by an organisation accredited by the United Kingdom Accreditation Service. This standard-certification approach could be designed to raise the quality of service and governance up to a minimum level that all savers should expect.

Trustees and IGCs could then report on compliance based on whether the firm or scheme was independently certified against the standard and request an explanation where wasn’t. We explained that schemes with less than £100m AUM who have separately considered the quality of their administration and governance as part of their VFM assessment could use the findings of this assessment to reasonably justify why they have not been independently certified. We asked:

20. Do you think that, over time, independent certification against a standard is worth exploring for benchmarking service metrics? If not, what alternative arrangement would you suggest?

Respondents were divided on whether independent certification based on a ‘comply or explain’ approach would encourage consistent disclosures or whether it would merely duplicate the work of IGC’s and trustees by outsourcing their role to another party, resulting in unnecessary administration and costs borne by savers.

They argued that independent certification would provide a solution to the current discrepancy in standards and would help improve schemes with poor service levels. Well-recognised certification gives confidence to members that the scheme is well managed. However, a number of respondents expressed their concern that this would result in a one-size-fits-all approach, hinder innovation, and quickly become out of date. They also argued that the cost of the process would be disproportionate to the value added.

A number of respondents cautioned against setting a minimum threshold in this area as it may have unintended consequences, including limiting innovation and driving a ‘race to the bottom’ where schemes only focus on meeting the minimum standards to save on costs.

There was a general consensus that the industry, at this stage, should continue to focus on and direct its resources to develop concrete and viable service metrics standards, and that independent certification should not be considered until this has been achieved.

4. Definitions and benchmarks for costs and charges

This chapter summarises the feedback we received to the questions asked in our discussion paper around the definitions and benchmarks for costs and charges.

Disclosing costs at the right level and building upon existing cost disclosures

In our discussion paper, we proposed to design market-wide cost disclosures by building on the existing definitions of ‘administration charges’ and ‘transaction costs’ associated with existing disclosure requirements for DC workplace pension schemes to avoid schemes having to collect further data.

We sought to allow trustees, IGCs, and others to make meaningful comparisons while also noting that, within an overarching HMRC-registered scheme, costs were likely to vary based on the terms negotiated by different employers and that similar considerations arise in the context of a master trust.

We suggested an option for schemes to separate fixed ongoing costs from variable costs when disclosing administration and transaction costs. We also asked whether it might be helpful to further split the current definition of administration charges into fund management and pensions administration to allow trustees to better understand where costs were being incurred and to manage them accordingly.

Our discussion was focused on member-borne costs. We did not think that disclosing costs that were purely met by the employer was necessary to facilitate effective VFM comparison for savers. However, we recognised that employers might still be interested to know whether members were receiving additional value for the features which employers are paying for.

Q21: Should we use the existing administration charges and transaction costs definitions in developing VFM costs and charges metrics?

Q22: Would splitting out the administration charges be a more useful metric? If not, are there other definitions you think would be more appropriate?

Most respondents supported using existing administration charges and transaction costs definitions in developing VFM metrics. These definitions are commonly adopted across the industry and offer consistency for investment design and reporting arrangements.

Several firms pointed out that schemes with ‘bundled’ arrangements that rely upon a single provider for both administrative and investment management services could struggle to clearly distinguish between certain administrative and investment costs. These firms did not believe there was a strong case to separate these costs due to the interdependence of both activities. They argued that preserving and employing the existing definitions would result in less confusion as well as minimise implementation costs. Several also voiced concerns that splitting out administration costs could disclose sensitive commercial agreements made between providers and their investment partners.

However, some respondents saw benefits from separating investment costs from administrative costs, as well as splitting the current definition of administration charges into fund management and pensions administration further. They felt that this split would be more useful when comparing investment portfolios because investment performance is best assessed net of the cost of its delivery and not net of the bundled additional services. Therefore, the cost of the investment component should be provided separately and not as part of a bundled product charge.

This separation would help trustees and IGCs assess how their scheme administration costs compared to other alternatives and use this information to drive VFM.

Additionally, some respondents believed that clearly showing the components that make up the charging structure would help members develop trust in the industry. They felt it was desirable that no charges or fees should be hidden or rolled into other categories, although many stressed that this should be done in a way that is understandable, clear, and not overly burdensome on schemes.

A number of respondents proposed other definitions and suggestions, including:

• administration charge should be split into AMC, TER and platform fee
• any measure of charges must take into account future changes to charges where relevant. For example, legacy products have structures that result in charges changing over time, and firms must be allowed to take a forward-looking approach to include these in comparisons
• schemes should disclose which fees are paid to the commercial sponsors or distributors of the plan and which are paid to external service providers to allow comparisons between in-house services and external providers
• levy fees should be shown separately so that the saver is informed about the costs of regulation and compensation

A few respondents disagreed with the view expressed in the discussion paper that the disclosure of any charges solely met by the employer would not be necessary to allow an effective comparison of VFM for savers. Respondents explained that it would be illogical to disincentivise employers from continuing to bear these costs. They argued that the focus on member-borne charges only distorts the assessment of true value and is not holistic, as the disclosures do not represent the value for the member of costs incurred by the employer that they would otherwise pay.

On the disclosure of costs and charges to savers, one respondent suggested that the proposals should not be implemented until the existing MiFID rules post cost disclosure regime has been reviewed. The respondent explained that the detailed basis of calculations is very complex and theoretical, with no clear evidence that members pay attention to them.

Some insurers and pension providers also identified a number of other considerations when comparing costs and charges.

• There are challenges around comparisons between different products, and therefore the framework must ensure like-for-like comparisons. For example, legacy products should be compared against other legacy products instead of modern products.
• Cheaper options are not always available to employers because this largely depends on size, scheme demographic and contribution levels.

Some respondents highlighted that the existing cost transparency initiative framework is a strong and consistent industry-wide standard that could be helpfully incorporated into the VFM framework.

Introducing new benchmarks to assess costs and charges

We suggested that introducing benchmarks would make the VFM achieved by different pension schemes, in respect of their costs and charges, directly comparable. We proposed a number of options for a benchmark, such as: using a market or a relevant market segment median charge, using the mean of charges in the market, using quartiles rather than just a median, and using a not-for-profit benchmark.

We asked the following questions:

Q23: Do you agree we should introduce benchmarks for costs and charges?

The majority of respondents argued that now would not be an appropriate time for regulators to introduce benchmarks for costs and charges. Respondents argued that it would be difficult for simple benchmarks to provide accurate and meaningful comparisons due to the different fee structures and business models. Without any reference to the service provided or investment performance, the benchmark in isolation provides meaningless comparisons and does not recognise value that the scheme may provide.

For example, a with profits fund has a higher charge compared to a unitised fund. However, it delivers, through smoothing, reduced volatility and other additional benefits to members, which need to be accounted for.

Respondents expressed concern that a costs and charges benchmark could risk ignoring other aspects of VFM and promote a short-term view, thus driving members to the cheapest but not necessarily the most appropriate products. Such behaviour would go against the goals of the Productive Finance Working Group and other policy objectives. Respondents believed that a benchmark would be a limiting factor for trustees who are considering more expensive investment strategies, including investment in illiquid assets. Finally, a standard benchmark would likely fail to capture the additional value associated with costs paid by employers.

Respondents also argued that clear disclosure of costs and charges would be sufficient. They noted that the DWP’s disclosure requirements, as laid out in PS20/2, will make a large amount of data on costs and charges more publicly available, which will allow governance bodies to assess their scheme against the market. The data will also allow governing bodies to challenge and negotiate with providers. The public disclosure of data will also provide opportunities for commercial data comparisons, and time should be allowed for this area of the market to mature so that regulators can monitor the developments before deciding that a public benchmark is necessary.

Respondents further explained that a disproportionate focus on the costs and charges could lead to a downward pressure on providers to charge rates that are not economically viable. As a result of this, some providers might decide to withdraw from the market, reducing competition.

Another group of respondents agreed with the suggestion to develop benchmarks for costs and charges. Some argued that any benchmarks developed should recognise the idiosyncratic nature of schemes and ensure that comparisons are made across comparable schemes. Alternatively, others argued that schemes must be required to compare against a reasonable number of schemes with varying characteristics to prevent them from only comparing themselves against other similarly poor VFM schemes.

Q24: What are your views on our suggested options for benchmarking costs and charges? If not these options, what benchmarks should be used?

Most respondents disapprove of using a not-for-profit scheme as a benchmark or comparison. One firm was concerned that the not-for-profit scheme might struggle to accept a large market movement of employers or cope with the resulting burden this would place on the systems and resources of a single not-for-profit scheme.

Many pension providers stressed that economies of scale provide schemes with certain advantages in pricing, and a not-for-profit scheme like NEST, with its access to public funding and subsidies, supplemented by a unique charging structure, makes it unsuitable as a benchmark for other schemes to be fairly assessed against. One respondent argued that this could result in a regulatory bias towards the not-for-profit side of the market.

Additionally, some respondents argued that all the suggested options failed to ensure a sensible grouping of schemes. Instead, respondents highlighted that transparency and clear disclosure of costs and charges were sufficient. One respondent argued that existing regulations, including the charge cap for workplace pensions, in addition to publishing data on charges as part of the VFM assessment, would result in meaningful comparisons without the need for benchmarks.

However, some respondents argued that there was merit to the options for benchmarking costs and charges against, for example, a not-for-profit scheme. There was a general agreement that if benchmarks were introduced, they should be scheme-specific, for example, grouping schemes into different cohorts. Most agreed that using quartiles, rather than comparing against a market average, would be the better option.

One respondent suggested that, instead of benchmarks, the following classification of charges would be more meaningful:

• low cost (<40bps)
• average cost (40 to 60bs)
• higher cost (>60bps)

Another felt that benchmarking should focus on two elements:

• the items of the service being paid for
• the effectiveness of the provider or distributor governing them for the members.

There were suggestions that a combined benchmark for costs and charges, performance, governance and administration would be more meaningful, and arguments that benchmarking should be against the overall VFM rather than just costs and charges.

Some respondents suggested that regulators must first allow data to develop and become available and monitor how governance bodies understand and act on this information before deciding whether there is a need for benchmarks.

Annex 1: list of non-confidential respondents

Aegon

AgeWage

AJ Bell

Alliance Bernstein

Aon

Association of British Insurers

Association of Consulting Actuaries

Association of Professional Compliance Consultants

Association of Professional Pension Trustees

Association of Pension Lawyers

Aviva

B&CE

Barnett Waddingham

CACEIS Bank

Capita Plc

Chartered Financial Analyst Institute

Connected Asset Management

Creative Benefit Solutions

Cushon

Dalriada

Evolve Pensions

FCA Practitioner Panel

Financial Services Consumer Panel

Hargreaves Lansdown

Hymans Robertson LLP

Investment & Life Assurance Group

Interactive Investor

Investment Association

ISIO Group

Legal & General

Lane Clark & Peacock LLP

M&G Plc

Mercer

Money Alive

Morningstar

NATWEST Group Retirement Savings Plan

NEST

Now Pensions

Partners Group

Pension Administration Standards Association

Philip J Milton & Company Plc

Pinsent Masons LLP

Pensions and Lifetime Savings Association

Quilter

Railpen

Rexroth

Royal London

Royal London IGC

Scottish Widows

Society of Pension Professionals

St. James's Place plc

Smart Pension

Standard Life

The Investing and Saving Alliance

The Pensions Institute

The Phoenix Group IGC

Threesixty Services LLP

Willis Towers Watson

Zurich

Two individual respondents

Back to top

Is this page useful?