Edition: Sept - Nov 2013
Editorials

ASSET MANAGERS

Fee complexities

National Treasury has analysed them critically. A basic guide, from an industry perspective, is offered by Steve Price of Investment Solutions.

A significant element of the total costs incurred by investors is the fee paid to the asset manager. He’s typically rewarded by:

  • Fixed fees, charged as a percentage of assets;
  • Performance fees, paid additionally to the fixed fee, are to remunerate a manager for performance in excess of a predefined benchmark. Designed to reward an active manager’s skill, they can be useful to craft an arrangement between manager and client that aligns interests to the benefit of both. If poorly conceived, however, they can result in destruction of value to clients. Performance fees often account for more than 75% of the total fees paid by clients.

At present the only segment of the industry which requires disclosure of performance fees is unit trusts. This is governed by industry guidelines issued by ASISA. They set out the minimum information that must be provided to clients in respect of performance fees in a unit trust portfolio.

The guidelines do not limit or prescribe a manager’s flexibility in determining the basis for performance fees. Investors need to make a case-by-case comparison of how performance fees operate in various funds. That performance fees in unit trusts are taken as a direct reduction of performance and are only disclosed as part of the total expense ratio (costs divided by assets, known as TER), makes them difficult to determine. Rarely, if ever, do clients see performance fees in rand terms separately on their statements.

Pooled life funds, offered through a life licence, are widely used by pension schemes and retail investors. Unlike unit trusts, which are required to make certain limited disclosures, life funds are under no obligation to disclose any information relating to performance fees either earned or paid to underlying investment managers.

Segregated portfolios, managed specifically for a client, afford greater protection as the basis for performance fees. They need to be explicitly agreed between client and asset manager (usually with the assistance of a professional asset consultant). Clients large enough to operate on a segregated basis are typically advised by an asset consultant.



Price . . . through the morass

The advantage of segregated mandates is that, at a minimum, asset managers will be required to invoice or deduct performance fees directly from the client. This makes it easier for clients to identify the quantum of performance fees being paid.

Performance-fee arrangements have a number of elements to be considered for evaluation:

  • Performance-fee benchmark is the benchmark against which performance is measured. It should align with the objectives of the mandate in order to ensure that the fees are paid for manager skill and not market ‘beta’. For example, in an equity mandate, the benchmark should be a relevant share index and not cash or inflation. A common practice in SA is the use of peer benchmarks where performance fees are payable simply for managers doing better than one another. This results in the fees being paid for underperformance of the market simply because a client chose the “least bad” manager;
  • Hurdle is the amount by which a manager needs to beat the performance-fee benchmark before earning performance fees. Managers should at least outperform the benchmark to cover the cost of the fixed fee paid to them before earning performance fees;
  • Participation rate is the share from performance taken by managers. They have a free option in that the worst managers can do is earn a fixed fee (but possibly be fired for underperformance), whereas their actions can permanently erode a client’s capital. Global best practice suggests that participation rates should never be more than 20% of outperformance. Why should managers be paid excessively for risking other people’s money?
  • Cap is the outperformance level after which no further performance fees are paid. Argument could be made for capped fees to limit needless risk-taking in pursuit of performance fees;
  • Measurement period is the time time over which the managers’ performance is measured. They should be measured over extended periods. There’s no correlation between a manager’s skill and performance against the hurdle over the short term (at least a year). Market cycles and investment time horizons are longer as the riskiness of the asset class increases. Measurement periods should align with the investment time horizon. Ideally, measurement periods should be cumulative from inception of the mandate;
  • Many measurement periods are unfortunately based on so-called “rolling” periods, typically 12 months. Here, performance fees are based on the last 12 months’ performance. The problem is that the next month’s rolling 12-month performance is determined equally by how well a manager does in the next month versus the same month 12 months ago, which now drops out of the measurement-period window. Rolling periods thus forgive a manager for underperformance, whereas the client is never forgiven. Cumulative measurement since inception is preferable;
  • High water mark ensures that managers do not earn performance fees unless they are generating new outperformance over and above the highest outperformance (watermark) generated previously. This avoids clients paying for performance more than once where managers go through outperformance and underperformance cycles.
  • Clawbacks are a mechanism whereby the performance fees paid to a manager are kept current. If a manager outperforms, a portion of assets is set aside for him. If he underperforms, this pot should be reduced or “clawed back”:
  • Cross subsidisation occurs when the performance fee is structured so that clients joining a fund pay for performance received by the fund prior to their investment i.e. performance in which they did not participate. Similarly, clients exiting a pool can often do so without paying for performance they may have received (leaving remaining investors to pay);
  • Fee at benchmark performance gives clients an indication of the fees payable if a fund or mandate performs in line with its stated performance-fee benchmark. This is usually a range around the benchmark. Performance fees often start being earned at levels well below the benchmark.Retail investors and smaller pension funds frequently have little control, and even less awareness, of performance fees and how they can affect long-term returns. The FSB’s ‘Treating Customers Fairly’ initiative many address some of these issues. A few other suggestions:
    • A minimum requirement to disclose total expenseratios on a quarterly basis, regardless of the vehicle or wrapper through which the mandate is offered. The standard for the determination of TERs is well established under the Collective Investment Schemes Control Act;
    • An industry standard on performance-fee methodologies in pooled vehicles offered to the public directly;
    • Improved financial education.