Keeping above the Waterline? An Incentive-based Framework for Assessment of Value

Keeping above the Waterline? An Incentive-based Framework for Assessment of Value

Value. Much emphasis is placed on the consideration of cost, economies of scale and other factors for the upcoming Assessment of Value (AoV). Performance is included; yet to date has been given little more than a cursory reference; despite being the most challenging and subjective component of the AoV. Can then performance fees, hurdles and fulcrum fees help inform the AoV? What then is value within an investment fund? Once upon a time it was simply generalised as ‘alpha,’ but such notions now seem retired. The question goes well beyond the active-passive debate; yet at its core it defines the difference and merits between solely low cost and added value. The truth lies somewhere in accurate benchmarking, targets and generating excess compound returns persistently over frictional forces.

Cost is one side of the equation. The APFI’s ‘Game of Thrones’ series on investment fund cost governance discusses the conflicting participants, agendas and identifies both technical and practical solutions. The proximity between AoV and cost transparency will remain close.

See: https://www.linkedin.com/pulse/investment-costs-europe-game-thrones-sunil-chadda-jb-beckett-chadda

The new Assessment of Value (AoV) has 7 guiding considerations. as follows;

I. The range and quality of service provided to unit-holders
II. The performance of the scheme, after deduction of all payments out of scheme property
III. The cost of providing services
IV. Whether the AFM is able to achieve savings and benefits from economies of scale, relating to the direct and indirect costs of managing the scheme property
V. In relation to each service, the market rate for any comparable service provided
VI. In relation to each separate charge, the AFM’s charges and those of its associates for comparable services provided to clients, including for institutional mandates of a comparable size
VII. Whether it is appropriate for unit-holders to hold units in classes subject to higher charges than those applying to other classes of the same scheme with substantially similar rights

Performance is the other side of the value proposition. Of the 7 guiding considerations for the AoV test; we thus consider point 2: ‘performance’ and how it can infer value. Born out of the FCA Asset Management Market Study Final Report (MS15/2.3), and follow up remedies (PS18/8) in 2018, the “Assessment of Value” (nee Value for Money) is an attempt to improve fund outcomes via improvements in governance. It aims to do this by requiring UK Authorised Fund Managers (AFMs) to focus more on their duties as agents of investors and to ensure fair treatment for all fund investors.

The key components arising are fairly obvious (performance net of cost). Costs are tangible (Cost of X is greater than Y) but at times not transparent and do not infer value in isolation. The FCA recognised this when publishing its final report. Performance on the other hand is publicly reported net of (some) fees and thus transparent to an extent but nonetheless intangible in terms of establishing value. The performance of one fund may deliver value but the same performance for another fund might not. Stated aims may vary. The performance of the scheme is therefore both the performance of the individual sub-funds and aggregated performance of the AFM, net of fees. However less seen is the fee structure (template) applied within the outright fees quoted. That structure could frame the question. Considering how performance (thus value) might vary through these fee structures may help with an AoV.

That intangibility is in part due to a lack of structure or technical guidance on how to identify, calculate and monitor ‘value’. Unlike performance which follows prescribed accounting guidance (eg. GIPS), value has never been objectively defined or assessed. Performance returns are clearly a key input but how do we discern if returns are reasonably delivering value?

In an earlier APFI technical note entitled ‘The Guns of Navarone’, Sunil Chadda and I provided the context for the AoV and therein we discussed the concept of frictional force as “the rate at which a fund can overcome its own friction then generates value over time. If we can agree this is a simple but logical assertion, then we can apply formulae used for frictional force. There are two types; static (inertia) and kinetic (moving).”

See: https://www.linkedin.com/pulse/guns-naverone-assessment-value-new-actors-begin-climb-sunil-chadda

In that paper we proposed simply that: [AoV= Return less Friction]. Yet we clearly need to refine this further, accounting for the stated aims of the fund and any peer or benchmark comparisons. This itself drifts into issues regarding benchmarks and investment sectors.

See: https://blog.sharingalpha.com/on-the-bench-confidence-game-or-imperfect-science-the-professional-fund-investors-view-of-benchmarks/

Consider also that friction in context to a fund comes from three main sources being; 1) costs from trading the asset (subject to cost template used), 2) operational and regulatory costs and 3) investment management costs (those charged explicitly for providing investment management). The AFM has ways to manage and control these costs to greater and lesser extents. The inclusion of some costs are highly debated. For example the FCA concluded that fund managers on average were operating a 35% profit margin overall, that is retaining 35% of the earnings of the fund, in fees, after all other costs had been deducted. This will primarily have been through investment management fees but other sources such as stock lending and box profits were identified by the FCA. These will also vary firm to firm, by size of firm, size of fund, fund sector and legal jurisdiction. ETFs for example earn profits though the intra-day subscription/redemption process. Creating an objective framework to assess performance/value goes some way to understanding those fictional forces at play.

Capturing the fund’s Net Asset Value (NAV) should capture all post-trading costs on the fund (but not necessarily pre-NAV costs) and remains the most accessible means to assess performance after costs, especially with a customer lens. Fund boards should have access to both gross and net performance from the fund manager (investment adviser) and be supported by ancillary information on frictional costs impacting the profitability and the fund’s AoV. Cost attribution may become the next biggest development in fund governance.

What sort of fee framework? For example we could apply a performance fee type structure (common among hedge funds to calculate fees) to structure the AoV. This proposal then is not about moving funds to a performance fee structure; rather using the structure for AoV to consider value being triggered above a hurdle rate. This paper also does not opine on the use of performance or fulcrum fees, which have historically come under criticism, particularly in hedge funds and advisers (RIAs) in the US.

However applying a hurdle based framework in many ways is consistent with the FCA Asset Management Market Study (‘AMMS’) which had criticised fund managers approach to benchmarks and stated performance targets. In CP18/9 the FCA states;

3.35. “Where an AFM has not set a constraint, target or comparator for a fund, we propose to require the AFM to explain to investors how they should assess the fund’s performance in the fund’s prospectus and relevant communications. The intention is that investors will gain valuable insight into how the AFM views the fund and how it thinks the fund’s performance should best be judged, and that this will promote better informed choices by investors. This ‘requirement to explain’ should also be a useful discipline when AFMs are selecting benchmarks or considering changing them.”

JB View: It is clear this ‘requirement to explain’ will have a clear bearing on the AoV and the fund board’s reporting of value.

3.40. “a fund with 2 targets (target benchmarks) - to beat the return on LIBOR, and to beat LIBOR +4%. Our proposed rules would mean that it would not be acceptable for the AFM to show the fund’s past performance against LIBOR alone - it must show it against both targets.”

JB View: The use of performance targets, which have ostensibly been used as marketing aims to this point, look likely to become a key hurdle for any future AoV.

4.2. “AFMs, in accordance with our existing rules, need to make sure that all fee structures are ‘fair’. This includes any ‘performance fees’ taken only after a fund has achieved a stated objective(s), for example fees taken only once a target benchmark has been exceeded. ”

JB View: Irrespective of fee structure applied, the notion of ‘fairness’ as it relates to AoV is clear. How fairness is applied to fee structures without a target (hurdle) is then a key consideration for the AoV.

4.4. “There is substantial innovation in fund performance fees in the UK market at the moment, especially the development of more symmetrical performance fee models. These are fees that try to better align AFM and fund investors’ interests to the risks and reward of fund performance. For example, such structures reduce the fees payable to the AFM in the case of poor fund performance. We do not wish to inhibit such innovation where it is in the interests of investors. As a result we are not proposing significant rule changes at this time. We remain focused on whether fees are fair to investors and will intervene under our existing rules where we are concerned that this is not the case, for example where it is clear that an AFM is charging performance fees in a way that investors could not understand in advance and would not expect (for example well below a stated target).”

JB View: The implications for performance fee structures are clearly in focus for the FCA. All things being equal, the delivery of value should be agnostic to the fee mechanism used; albeit in practice this is probably not the case.

The FCA consultation gives more clues to the AoV. More here: https://www.fca.org.uk/publication/consultation/cp18-09.pdf

It is worth noting the two main fund fee structures that exist today are a flat fee Annual Management Charge (AMC) and a base AMC plus incentive fee (also called the performance fee). Therein a range of share class terms and subsidies can exist varying the cost and thus any ‘value’ consideration. Here I disagree to an extent with the summation of the Gartenberg rule that defines the assessment of fees in the US (the basis for a future APFI technical note).

The premise of any performance-based (incentive) fee structure is that the investor pays more when the fund manager has generated excess returns and less when they fall behind a target. Two immediate observations are that the investor assumes the manager is generating value when paying the performance fee (otherwise would not be economically incentivised to pay up) and the fund manager recognises they have not delivered the objective ergo not delivered value and therefore prepared to accept lower fees. Outside of whether Value is being delivered (or not) there is no immediate inference of the Fund Manager not being fit and proper or failing to apply best execution. Value and compliance are not the same.

From a behavioural finance standpoint, one can conclude that both sides should be economically incentivised for the fund to deliver excess returns. The fee relationship however may be skewed at outset if the two parties are not equally matched in terms of information during the fee negotiation. For example a soft performance target may pay the fund manager excessive fees when the investor did not understand what was obtainable from the market. This I have previously explored here;

“Depending on how the fee structure is set-up, a manager can become susceptible to a bias to not add incremental risk (to achieve the higher performance target) but rather add sufficient to create a continually growing cumulative return over the High Water Mark and modest outperformance above the lower performance hurdle quarter on quarter. What can then happen is that the fund manager may double their earnings above the quoted annual charge but the investor may be facing into returns 50% of what was expected.”

https://blog.sharingalpha.com/on-the-bench-confidence-game-or-imperfect-science-the-professional-fund-investors-view-of-benchmarks/

The difficulty here for performance fee structures is that the backdrop market return continually fluctuates and past performance may not necessarily repeat in the future. Nonetheless both parties are giving away some certainty for prospect of either paying lower or higher fees, versus higher returns. The fund manager will assume they can earn higher fees through delivering excess return. The investor will assume they will pay lower fees if the manager cannot. This assumption commutes well into the assumption that a fund manager can deliver value (or not). Both fund manager and holder thus strive for excess returns and identify value when the fund exceeds the hurdle; whilst the fund may continue to be fit for purpose below the hurdle but not necessarily delivering value. Both parties accept then that above hurdle performance may not be certain or persistent.

It is worth noting that above the performance hurdle the investor pays the fund manager a percentage of the excess returns. Meanwhile when the fund manager fails to hit that target then the fund manager accepts lower fees and the investor accepts to continue to pay the fund manager a reduced fee. How the performance fee is structured will then determine the level of risk-taking (risk budget) the fund manager is prepared to take in order to outperform the hurdle/watermark. We explore the structure more fully below.

Subsequently CASS Business School has proposed a ‘symmetrical AMC’ using a performance fee and reclaim structure. Again this structure will drive certain behaviours and with it the perception of value, specifically how a clawback facility effects manager and buyer behaviour.

More can be read here: http://openaccess.city.ac.uk/16840/

“Our study identifies a clear 'incentive mismatch' between the best interests of investors and managers, more specifically, there is no single structure that simultaneously maximises both the investors’ and the managers’ utility. In fact, the results show that the most prevalent fee structure currently in the UK market (a fixed fee as a proportion of AUM) is generally the best structure for the manager and the worst for the investor! To verify the robustness of our results, we stress-test the model parameters, however, none of these model variations change the base results and our main conclusion. The results in this paper give rise to a natural question: "Since investors would prefer symmetric performance-based fees, why don't more fund managers offer such fees?"”

View: We can observe the tenets of symmetrical fees and incentive matching, whilst radical, are core to the AoV. Most simply it asserts a mismatch can arise when funds continue to charge fees when not delivering value.

We will not tackle the AoV of funds with performance fee-based structures, here, since the performance is not reported net of any performance fees. The AoV will be empirically based on fees paid versus excess returns. In these cases the U.K. Fund Board would require additional internal fund accounting data, to create an adjusted total annual charge, to then estimate the adjusted annual cost and value assessment.

A Value-Performance-Fee Framework

The basis of our value calculation can follow different structures:

• The value can simply be the total increase in NAV
• The value can be the increase in NAV after adjusting for management fees
• There could also be a benchmark or a hurdle rate. Value only occurs when the fund’s annual rate of return exceeds the hurdle rate.
• Apply a feature called high water mark.
• The ‘reset’ typically occurs once or twice per annum and sets the new hurdle (NAV) for the next performance period.
• A ‘fulcrum fee’ is a performance-based fee that can adjust up or down based on outperforming or underperforming a benchmark.
• A ‘symmetrical fee’ as proposed by CASS, and used by likes of Orbis since 2014, is a performance-based fee that can adjust up, or clawback previous fees, based on outperforming or underperforming a benchmark.

The high water mark is where the fund manager will only receive an incentive fee IF the fund exceeds the highest NAV it had previously reported. The idea is to ensure that incentives are not paid on the profits that simply offset the losses of previous years. In other words reward consistency rather than boom-bust performance. Aside from the vagaries of market cycles; in context to creating long-term value, a water mark approach has merit.

alt

In addition to the incentive or fulcrum fee a fund will also have a base annual fee, which is paid irrespective of performance. The proportion of base and incentive fee is important. In a symmetrical based fee there is typically no base fee (zero % AMC).

Excess returns above the hurdle trigger a higher fee and implies value generation. Conversely a lower (albeit likely still positive) barrier the fund generates less fees.
Morningstar discuss the Fidelity fulcrum model here: http://www.morningstar.co.uk/uk/news/163514/what-are-fulcrum-fund-fees.aspx/)

Common to all incentive-based fee structures then is an explicit hurdle linked to a ‘good outcome’. That hurdle might be absolute positive returns, above cash, inflation or other measure. In context to the AoV the hurdle rate is the target which, until met, Value is not being achieved (or not being achieved fully). The return above the target is the amount of value being added. Below the hurdle the fund may be operating effectively and fit for purpose but is not delivering value to the investor based on the stated target of the fund. It identifies that value occurs above the zero bound and therefore a positive return does not necessarily deliver value. How sub-hurdle returns are then reported by the Fund Board may be as something as simple as stating the hurdle was not achieved and that the fund delivered: ‘50% of target or ‘50% value’ over X period or ‘delivered value for 66% in 24 out of 36 months’.

Over time (t) Net performance (or risk-adjusted return) after the Annual Cost is deducted and, when below the hurdle, is not generating value left of the axis (shaded red) but generating value to the right of the axis (shaded green). To this a waterline (high water mark) could be added.
Note: To the hurdle, a high water mark has been added above to measure incremental value on an ongoing basis (eg. Quarter to Quarter).

Value at what Risk? Establishing value above the hurdle is on the clear assumption that the fund manager is taking risk that is commensurate with the fund aims, investor and/or mandate. Riskiness is therefore relevant. However an important consideration being that risk measures are not unjustifiably used to establish value when performance falls below target or if the aims of the fund have no relevance to those measures. In that instance risk measures should be used as ancillary inputs. To this extent reviewing performance alongside risk metrics is important. Excessive risk-taking is not necessarily generating value for the fundholders. A mixture of traditional (eg. Standard deviation, VaR) promoted by the CFA alongside more holistic risk metrics (eg. Factor, concentration, liquidity) is encouraged. In the ‘Guns of Navarone’ APFI technical note I debunked some of the variance base measures currently being proposed. Risk measures do not necessarily confer value but are controls to prevent incorrect risk-taking being misconstrued as value generation (be that deploying too much or not enough).

See: https://www.linkedin.com/pulse/guns-naverone-assessment-value-new-actors-begin-climb-sunil-chadda

Downside mitigation: Funds that can justifiably use risk measures more readily in the value assessment include absolute return and risk target funds.
Some funds also purport to add value through capital preservation and downside protection. Here a second hurdle could be set based on absolute loss value, when the benchmark is below a certain level. In operation this might not be dissimilar to setting a structured product’s ‘barrier’. However care should be taken to ensure hurdles and waterlines are calculated that are fair to the investor. For such products setting a multi-year hurdle, calculated on a rolling basis for the AoV, might work if consistent with how such funds are marketed. Boards would then need to consider how to establish and report on AoV over shorter periods. Alternatively these funds continue to base the hurdle on a cash or inflation plus target.

Net Value? Calculation of value Net of fees appears consistent with the FCA’s corresponding view of performance fees in CP18/9;

4.5. “Our current guidance says that an AFM should not charge performance fees on gross performance, i.e. before other charges such as the AFM’s fee have been reduced. However, this is not prohibited by a rule. In theory, if an AFM can show that taking fees in this manner is fair, then it could be permitted. In our view it is not in the interests of investors for AFMs to take performance fees on a gross basis. This is ‘a fee on a fee’ and is unlikely to be understood by investors. We propose a rule to prevent AFMs charging performance fees on a gross basis. This is consistent with the 2016 International Organization of Securities Commissions (IOSCO) Report ‘Good Practice for Fees and Expenses of CIS’. ‘Good Practice 4’ states that performance fees should be taken on performance that is net of other fees.”

View: The fund’s end investor has little control of fees levied, by the AFM, or visibility of gross performance and so reporting value on a Net basis appears most aligned to the investor. Fund boards will again seek to see gross and net performance to understand the difference and what is detracting from the AoV.

Let’s take an example to understand the calculation of the hurdle and value. Let’s say the hurdle rate is 6% (roughly Cash +4% or the long run return of global equities) and the value is calculated on gains net of management fees.

Example 1. The fund began with a £100 NAV.

Year Unit NAV Hurdle ~6% AoV per Unit
1 £100 £106 =0%
2 £140 £112.36 =20% of £27.64=£5.528
3 £150 £148.4 =20% of £1.6=£0.32

Example 2. The following example illustrates adding a high water mark.

Year High-Water Mark Unit NAV AoV per Unit
1 £100 £140 =20% of £40=£8
2 £140 £120 No value as the NAV is below the high-water mark
3 £140 £160 =20% of £20=£4

Time value: How a fund delivers value should be in context to the Recommended Holding Period (RHP), stated aims and investor expectation (Treating Customers Outcome #5). The RHP is particularly useful where the fund manager is unaware of its’ investors average holding period (for example if held through omnibus accounts and nominees). If there is a mismatch between the fund’s return horizon and the investor holding period then value might prove elusive. Correctly assessing value over the correct holding period (rolling basis) makes sense. It draws on the studies relating to the Fidelity Magellan Fund, which showed holding period mismatches can arise. Whilst the fund manager often had little control over the buy/sell behaviour of investors; it can rely on the RHP as a common reference point.

Ref. The Magellan example is discussed here: http://www.tuveinvestments.com/documents/Disipline-Akeytosuccessinbusinessandinvesting.pdf

Similarly, just with performance based fees, any hurdle, fulcrum or water mark approach to identifying value should similarly be over the correct holding period (typically the RHP) to ensure good investor alignment.

Other considerations: This paper has not considered the effect of share class variance and also variable term deals, and ‘break-points’ which are often scaled based on the size of investment and AUM accrual. These are captured both by Gartenberg and Assessment of Value and require attention. These will effect the value assessment as the fund grows and fee changes. One difficulty with variable or unquoted deals is that they are not reflected in the public NAV or performance of a share class. Therefore the most pragmatic solution is to apply the AoV to the most expensive NAV visible. If it fails then it is worth estimating for the effect of discounts and cheaper share classes. However prudence is needed from tripping into the subjective.

Conclusion: Many fund managers and fund boards will struggle to objectively identify value from performance (after costs) without a framework. Using the basic assumptions of performance-based fees; holding periods, hurdles, high water marks and fulcrums, Fund Boards can more easily identify value both in terms of excess returns and persistency. I look forward to working with Fund Boards and investment advisory committees to help them traverse these issues.