Top 400: A better deal on fees
Fee structures are imperfect and may be poor value. Nick Sykes outlines ways they could be improved for institutional investors and investment managers
Today’s most common investment management fee structures for traditional long-only products have their shortcomings. But there are ways in which fees might be structured to benefit investors and investment managers. Today, almost all investment manager fees are structured in one of two ways:
• Flat: meaning ad valorem or a percentage of assets, often on a sliding scale providing a lower average fee for larger investors, such as 0.6% on the first £50m (€67m), 0.5% on the next £50m, 0.4% on the balance.
• Performance-related fees: almost always structured as a base fee (on an ad valorem basis, such as 0.25% of assets) and a share of outperformance, for instance 20% of outperformance above a certain benchmark or target level.
These fee structures are flawed, whichever way you look at them.
For the CEO of an asset management business the problem is that income is neither stable nor predictable. Fee income is a function of market levels (highly volatile, especially in equities); manager outperformance to the extent that investors have chosen performance-related fees (also likely to be volatile); and new business gains or losses. It is quite easy to imagine that an equity manager could experience variability in fee income of 30-40% a year without changing what they do or how they do it.
This makes an investment management business difficult to manage in a controlled and predictable way and, inevitably, leads businesses to seek higher levels of fees to counteract the volatility in fee income. This seems bad from a business management perspective.
In general, active management strategies ultimately suffer from diseconomies of scale, and the ability to add value eventually diminishes as the pool of assets being managed increases. But an asset management business benefits from economies of scale as profit margins increase significantly with asset growth because a chunk of costs are fixed in nature. In other words, the incremental cost of adding an extra billion of assets under management is likely to be minimal, which is not reflected in fee scales. If anything, the opposite is the case – fees for new business are often higher than the fees on the existing book of business.
This means there is an inevitable conflict between managers, who want to maximise the volume of assets under management, and investors, who may desire the opposite. For a given investment team and process, confidence in the ability to add value is likely to be higher when the volume of assets managed is smaller.
• Value for money: Both of the issues above point in the same direction – investment management fees are structured in a way that is potentially detrimental to the end-investor in terms of value for money. First, fees build in a margin to allow for volatility. Second, for new business, the benefits to the manager of gathering more assets put at risk the value creation that accrues to the investor;
• Fixed monetary amounts: You would expect the CEO of an asset manager to be delighted to have a proportion of the business on fees fixed in monetary terms with annual uplifts – RPI inflation would be easiest. This would bring greater stability to revenue with significant benefits for the management of the business. For the investor, such a fee structure would also bring greater certainty to budgeting, as investment manager costs would become much more predictable. Would such a fee structure not represent a win for investors and managers?
• Loyalty fees: Retaining existing clients is cheaper than spending time and money on winning new ones. So why should long-term clients not be rewarded for retaining the same manager for a long time? A reduced fee for long service as a client – for instance more than five or seven years – would be an appropriate reward. This could encourage longer-term, improved relationships between clients and managers;
• Fees and limited capacity: Past a certain point, the greater the volume of assets under management, the less likely performance objectives are achieved. A manager with a capacity of $10bn (€14bn) could legitimately charge a higher fee for the first investor because the likelihood of alpha capture is at its greatest. As more investors are recruited, the fee charged should be reduced for all investors equally, so all benefit. So the first $200m investor could be charged 1%, and by the time the last of the $10bn is filled, all investors would be charged 0.5%. This weakens the link between asset growth and revenue growth for the manager, and to an extent reflects the spread of manager alpha more thinly across a wider investor base. This approach would probably work best in the case of a tried-and-tested investment team and process;
• Share of alpha: Most sophisticated institutional investors recognise that manager skill (alpha) is worth paying for, not only for the additional return but also for the diversifying effect of the revenue stream it offers.
The problem is determining a share of alpha that is fair for the investor and investment manager. Performance-related fees are often structured with a base fee and performance participation of 10-20%. If investors are genuinely convinced of the value and scarcity of alpha, they might be prepared to pay a higher participation rate, for instance 25%. But in exchange they would want the base fee to be as low as possible, with the rate for index-tracking management of the asset a starting point for negotiation – in practice the agreed base level should be higher than this to reflect the higher costs of active management. Under this structure the investor collects a reduced share of the added value, but can justify this by paying a minimal fee if the manager does not perform. This tilts the risk/reward balance a little more in favour of the investor;
• Cost of capital: Taking this share of alpha argument to its logical extreme, the investor could argue that a performance-related fee should only be paid when a hurdle rate of return has been earned. This hurdle could be related to the underlying market return – the beta – of the asset class, or even to the investor’s cost of capital if that could be defined, and if the manager accepted this as reasonable. Think about it this way: if a bank were to provide capital to a proprietary trading desk, it would presumably charge the bank’s costs of capital (possibly including an appropriate risk premium) to the proprietary trading desk, with profits made from trading shared equally between the traders and bank. Why should an institutional investor not structure fees in a way that mimics the investor’s cost of capital, so that the manager pays a charge to trade the investor’s capital in exchange for a higher share of the excess return.
In conclusion, there may be better ways to structure fees that suit the needs of investors and managers.
Nick Sykes is a partner at Mercer