Australian equities funds and fair performance fees

Research Review is compiled by PortfolioConstruction Forum in association with Money Management.

This month PortfolioConstruction Forum asked the research houses:

What is an appropriate fee structure and level for an Australian equities fund for which performance fees are charged?

Lonsec

Lonsec believes the adoption of a performance fee addresses the key issue of aligning interests of fund managers and investors.

Performance fees change the incentive structure for a fund manager to make it congruent with the interests of investors, by aligning the fund manager’s fee earning potential with the performance outcomes for investors.

Capacity management provides an example of the way performance fees may positively influence fund managers.

We consider a high level of funds under management (FUM) to be counter to the interests of unitholders, because the ability of a fund manager to outperform over time increases in difficulty as FUM increases.

A fund manager may potentially be motivated to gather FUM to maximise revenue at the expense of investment performance.

The potential to earn performance fees may deter this kind of behaviour by providing them with the opportunity to share in the upside alongside investors. 

Consequently, all Australian equity funds could potentially benefit from a performance fee structure. The strategies that may benefit the most are those that more frequently encounter capacity issues – small-cap strategies, some long/short, and concentrated ‘best ideas’ strategies.

While we believe in the general concept of a performance fee, the appropriateness of a performance fee is determined by a number of factors that may vary between funds, including:

1. The performance hurdle – hurdles observed include:

• a benchmark (eg, ASX-300); 
• a benchmark plus the base management fee or other margin; 
• the cash rate; 
• zero. 

Lonsec has a preference for benchmark linked hurdles for strategies that are fully invested and hurdles other than zero for more benchmark unaware strategies.

2. A high watermark – we have a strong preference for the use of high watermarks, where managers may not charge performance fees until previous losses are recouped, and we prefer not to see resetting high watermarks. 

3. The level of performance fee and the level of base fee – the introduction of a performance fee should be met by a reduction in the base fee.

The range of performance fees observed across the Australian equities spectrum has ranged from 10 per cent to 20 per cent. In general, the more capacity constrained the strategy, the more defensible the level of performance fee. The level of performance fee should preferably have an inverse relationship to the base fee.

In practice, the heterogeneity of available strategies and the potential number of moving parts in a performance fee structure means that an ‘apples to apples’ comparison of fees between competing funds can be difficult. The inclusion/omission of any of the above factors also does not automatically mean a performance fee is appropriate or otherwise. 

For example, the performance fee on Aviva Investors High Growth Shares Trust is considered by Lonsec to be appropriate, although it does not employ a high watermark and has a performance fee of 20 per cent, which is at the top of the range. However, its performance hurdle of ASX-200 plus 5 per cent is considerably more onerous than other funds.

Mercer

An appropriate fee structure is one that is aligned with the long-term objectives of the investor, truly reflects manager skill, fosters sustainable alpha and is conducive to a long and cost controlled relationship between manager and investor.

But no one size fits all.

The ideal fee structure will be specific to the investor and the situation, and the engagement of a fund manager should be based on strategy first, fee structure second.

It is ideal to avoid assessing investment managers on the basis of short-term performance as this amplifies the risk of making decisions that are not in the long-term best interest of the investors.

That is because of the inconvenient truth that performance data contains high levels of statistical noise – making it very difficult to distinguish between luck and skill.

The shorter the period in review, the less useful the information we can deduce.

Skilled managers can struggle with their performance for material periods primarily on account of ‘bad luck’, and the reverse applies for unskilled managers who experience a run of better fortune.

This situation has direct application to the method of remunerating fund managers – in particular, via (traditional) performance fee models. Fund managers generally seek, and rightly so, to be judged over a full market cycle as that is the period over which skill can materialise.

However, fee structures often stipulate the payment of a performance fee over a much shorter period, with one year being the most common.

It is fair to ask – is this logical and is it compatible with aligning investor interests?

The flow-on effect of using relatively short time periods for assessing and rewarding manager performance is that it encourages portfolio managers and analysts (and in turn, corporate executives) to focus on similar timeframes when evaluating market opportunities.

This is likely to be detrimental to maximising long-term returns.

Within this picture is the reality that if an investment manager is replaced, costs are incurred.

The more obvious costs are the transition costs, including product entry/exit spreads or brokerage and adviser fees to assist in the selection of replacement candidates.

Less obvious costs are time deliberating on the issue, negotiation of new contracts including legal fees, the rewriting of fund documents, and communicating with stakeholders about the change.

There may also be tax implications and market risks to manage as part of transitioning the portfolio. Such costs, ideally, are best avoided – unless the change has a strategic imperative.

So in summary, performance fee models should reflect payment based on: 

• outcomes reflecting skill, not luck; 
• encouraging alignment with the investors’ long-term objectives; 
• rewarding business practices that foster sustainable alpha; and
• promoting an enduring partnership and controlling total costs. 

Focusing on these factors offers several benefits, including signalling that the investor is serious about focusing on longer term goals, which encourages the manager to develop and maintain a long-term mindset.

This approach also fosters a partnership that can benefit both parties, and avoid manager switching costs. It places the emphasis less on rewarding noise/luck (shorter periods) and more on skill (longer periods) and achievement of the primary goal. Finally, it can promote moderate manager trading activity/portfolio turnover, and hence, lower transaction costs.

Standard & Poor's

Standard & Poor’s believes an appropriate performance fee structure can help to align the interests of investors and fund managers by reducing the manager’s incentive to gather assets at the possible expense of alpha. This can be particularly relevant in the small-cap sector, where capping funds under management can help ensure a manager retains appropriate flexibility to continue to meet its performance objectives.

Where a manager is charging a performance fee, we prefer to see it is doing so on the basis of a reduced management fee.

The manager should be prepared to back its ability to deliver superior returns. 

Other features to consider when assessing a performance fee are the hurdle rate above the benchmark, the presence of a high watermark (including any built-in reset clause), and the calculation and frequency of payment. 

• Hurdle rate above benchmark – most managers require a fund to outperform the benchmark after fees before a performance fee is charged.
  However, we would like to see a greater trend towards performance fee hurdles reflecting outperformance objectives above an appropriate benchmark. It is also critical to understand the risks a manager has assumed to generate the excess return upon which a performance fee may be payable. For example, is outperformance simply due to leverage of benchmark returns in the case of a hedge fund manager?
• A high watermark – this ensures a manager cannot accrue performance fees until a fund’s unit price is above its previous highest level or previous underperformance has been recouped. This is a vital component of performance fees and should be mandatory.
  High watermarks should not have a built-in reset clause. If a manager fails to meet its performance objectives, it should not be given a second chance until investors have recouped their losses.
• Payment – how frequently does the performance fee accrue and when is it payable? Performance fees should be paid at a suitable frequency. Short-term payments may encourage managers to focus on short-term gains, which may not be in the interests of investors.
  An important consideration for open-ended funds is how the performance fee is calculated. The fund-level approach, where all fees are calculated at the overall fund level, can cause inequities between investors with different application/redemption timings. In the hedge fund space, performance fees are often calculated through either the equalisation or series method.
  These approaches are designed to ensure incentive fees, to the extent possible, are fairly allocated between each investor. 

In the domestic small-cap space, we regard the performance fee applying to the Ironbark Karara Small Companies Fund to be one of the better incentive fee structures in place.

It includes a meaningful hurdle above the S&P/ASX Small Ordinaries Accumulation Index, and is only charged when fund returns are above zero.

S&P is particularly wary of performance fees applying to broader-based, large-cap Australian equity strategies in the absence of a significantly reduced management fee or a meaningful limit on stated capacity.

Morningstar

In a perfect world, we would like to see fulcrum fees adopted in Australia.

A fulcrum fee is a perfectly symmetrical fee structure which rewards fund managers on the upside, but equally penalises them on the downside by reducing the size of the base fee.

This relatively simple fee structure is fair to both fund managers and investors, and does away with many of the complicated facets of existing performance fee structures.

Unfortunately, current methods of charging fees are generally considerably more profitable for the fund manager – so don’t expect to see the adoption of fulcrum fees unless the regulator gets involved.

In the meantime, Morningstar believes there are a number of key areas to consider when selecting a fund with a performance fee.

1. Make sure a fund’s performance fee is benchmarked to an appropriate index. This is especially important in rising markets: funds with absolute fee structures can take advantage of a rising tide and charge fees for relative underperformance.
2. Ensure the fee structure has a high watermark that cannot be reset. This stops investors being charged a fee when a fund manager is merely making up previous underperformance.
3. A decent hurdle is a must. The amount of outperformance the manager needs to post before it begins taking fees is significant. The bigger the hurdle, the better – but it should at least cover the base fee. If it doesn’t, the fund manager can take a fee even if it underperforms.
4. Take note of the actual performance fee quantum, as all else being equal, the higher the fee, the more you pay. However, you may be better off paying a higher performance fee to ensure the overall structure has an equities-relative benchmark, a high watermark, and a satisfactory hurdle, as the omission of any of these can prove even more costly.
5. Favour longer crystallisation periods. This is the frequency at which the fund manager can charge the performance fee. Managers with quarterly crystallisation periods may, for example, be tempted to focus on shorter term performance chasing.

In Australia, it is rare to find a manager who ticks all of the boxes. We believe Maple Brown Abbott has a very good performance fee structure. Those with notable drawbacks include:

• PM Capital, which uses an absolute benchmark; 
• Aviva, which tends to run without high watermarks (but does have very large hurdles); and
• Alphinity, whose decision to reset its high watermark was a big negative. 

Nevertheless, in the vast majority of cases, the base fee that a manager charges will by far be the most significant cost to the investor.

It has to be paid, rain or shine – so if looking for value for money, look for a manager which has a low base fee, and therefore a real incentive to outperform.

In our view, all funds with performance fees should also have lower base fees, but this is seldom the case.

van Eyk

Of fundamental importance is the degree to which van Eyk thinks the theoretical alignment of interest provided by performance fees actually brings about a different or improved outcome for investors.

Will the presence of a performance fee change a manager’s behaviour?

On the one hand, it seems unlikely that adding, for instance, a share class with a performance fee component to an existing product will materially influence an existing team’s ability to outwit the market.

At the other end of the scale, it seems equally intuitive that hedge funds may succeed in attracting a specific type of individual willing to back their own judgement due to both the quantum and type of fees inherent to the business model.

It is equally plausible that those in charge of such a firm are likely to be highly motivated to ensure the conditions are in place where excess returns are likely to be generated (be they above cash or another benchmark).

In that sense, outperformance has become mission-critical, and the performance fee is a critical part of the structure – even if that can also lead to an element of optionality and excessive risk taking under certain scenarios. 

That said, all other things being equal, van Eyk’s view is that properly designed, performance based fees can strengthen the alignment of interests between the investment managers and fund investors. ‘All other things being equal’ specifically means ‘as long as investors don’t end up paying more overall and on average’.

On that basis, we believe that to be effective, performance fee structures should at least include:

• A perpetually high watermark – this means any underperformance must be fully recovered before incremental performance based fees can be charged. This provides a more equitable structure for the fund manager and fund investors to share both the upside and downside of performance risks.
• A performance hurdle – the appropriate benchmark for performance based fee measurement should include a performance hurdle that is commensurate with the investment risk of the strategy.
• A competitive asset-based fee – the level of asset-based fee should encourage the fund manager to carefully manage fund capacity (reducing the temptation to build assets under management just to obtain higher fees).

The evidence as to whether performance fees lead to better performance is mixed.

The top two performing funds in van Eyk’s most recent concentrated Australian equity fund review did not take performance fees (Hyperion and Dalton Nicol Reid).

However, a very large proportion of the other funds that did charge performance fees also outperformed over the past three years (in a universe where consistent alpha is rare).

In some instances, we have marked down a good team and process due to the weight of fees having a significant impact on likely levels of outperformance after fees (K2 is a well publicised example in the local universe). However, the IFP Global Franchise Fund was rated ‘AA’ despite quite high fees, which it has more than justified.

For that reason, we believe that in assessing the effectiveness of performance fees, the context for the individual firm is as important as the principle of alignment, and that broad brush assumptions about the effectiveness of performance fees might merely lead to higher fees with little or no noticeable impact on performance.

Zenith Investment Partners

In principle, Zenith is comfortable with the philosophy of a manager being rewarded for strong performance via a performance fee, where the performance fee structure aligns the interests of the manager with that of the fund’s investors.

In particular, we are comfortable with an appropriately structured performance fee for funds with FUM capacity constraints, as this provides potential revenue upside for the manager when closing the fund in the interests of protecting returns for existing investors. In general, capacity constrained asset classes tend to be the domestic asset classes including:

• Australian equities – large capitalisation funds
• Australian equities – small-capitalisation funds
• Australian equities – long/short funds
• A-REIT funds
• Australian hybrid funds
• Australian bond funds

The most capacity constrained domestic asset classes/capabilities are Australian equities small-cap, Australian equities long/short, A-REIT funds, and Australian hybrid funds.

With the exception of the latter, and to a lesser extent, A-REIT funds, performance fees are common.

In principal, Zenith believes the investment management fee should be around 20 per cent lower for a fund that also has a performance fee structure.

This shows goodwill on the part of the manager in reducing the base fee charged, and further incentivises the manager to exceed its performance fee benchmark.

The actual base fee level will vary from asset class to asset class, but as an example, Zenith likes to see Australian equities small-cap managers who have a performance fee with a base fee of around 0.90 per cent per annum to 1.00 per cent per annum.

Zenith believes best practice for performance fees is as follows:

• 15 per cent of outperformance above the benchmark;
• The benchmark to be the relevant market index for the asset class for which the fund invests in plus 1.0 per cent;
• The existence of a high watermark such that a performance fee is not payable unless the fund has outperformed its benchmark AND the unit price is higher than when the last performance fee was paid. This guards against paying a performance fee when the fund may have outperformed its benchmark, but absolute returns are negative; and
• Payment frequency to be annual at the end of June each year.

It is important to understand that performance fees will not be equitable for all unitholders in a fund.

The return each unitholder experiences will vary depending on when they entered the fund. As an example, a unit holder may invest in a fund just prior to 30 June, when a performance fee will be paid if the fund satisfies all of the qualifying criteria.

That investor will not have enjoyed all of the outperformance, having just invested in the fund, but will still pay the performance fee.

The only way around this, is that the performance for each unitholder is assessed – and this is very intensive for the unit registry to administer, and as such, is not common. That is why Zenith believes an annual performance fee payable as at 30 June is best practice, so investors can elect to delay investing into a fund until after 30 June.