CHAPTER 4

The Risk Management of Asset Management

Dennis Cox

Risk Reward

Introduction

Within asset management, risk management has always been a well-­established discipline, even if it has not been seen in that light. With ­performance being king and the need to show good results, asset ­managers have needed to be aware of the risks that they are running and to convert these into superior reported statements.

In this chapter, all of the key elements of risk within an asset manager will be addressed and future developments that are likely to impact the industry will be considered.

The key risks within fund management are:

• Market risk;
• Credit risk;
• Operational risk;
• Strategic risk;
• Reputational risk; and
• Liquidity risk.

In this, asset managers are no different from other financial service companies. It is the nature of the relationship with their clients that makes a difference to the approach.

With performance at all counts being of such importance, for most asset managers risk management was synonymous with management of the market risk within client portfolios. Other risks were generally not considered—as a consequence other risks could easily be taken on that actually had the impact of undermining performance.

Risk departments in fund managers rarely existed—there were ­performance measurement teams looking at specific market-based performance risks, but little else. Operational risk was within operations, credit risk was rarely considered and other risks hardly modeled. All of this has changed with the Basel Accord, which has encouraged asset managers also to implement a higher level of control.

Market Risk

Market risk within an asset manager is different from other financial institutions in that it is normally taken on for the benefit of investors. The depositors in an asset manager provide funds for the asset manager to invest for the benefit of the depositor. If the assets go up in value then the client’s portfolio increases in value. Likewise, if the investment decreases then the client’s portfolio also decreases.

Most of the market risk is not maintained by the fund manager, so what is the market risk that they continue to maintain? Within asset management, the relationship with the client is specified within a client agreement. This sets out the parameters within which the fund will be managed going forward, including the benchmarks against which performance will be measured, perhaps the FTSE 100 index, for example.

If a fund manager manages within the requirements of the client agreement then it will pass all of the market risk on to the client. It will then receive a fee for the management of the client assets based on the terms of the agreement. This normally provides a fee in two parts, one being the service element and the other being a performance element.

So long as the agreement is complied with, the client picks up gains and losses, so there is no market risk as such with the fund manager. If the fund is not managed in accordance with the agreement, however, this position changes. Then the fund manager has a potential risk that the client will come back to them and state that the losses should be made good.

Effectively, there is a moral risk to manage the fund in accordance with the client agreement that the fund manager has failed to comply with.

Thus, it is important for the risk management function within an asset manager to ensure that the fund is managed in accordance with the agreement with the client. This is done as follows:

• By measuring the performance of the fund against the benchmark, and calculating the tracking error caused by the fund not having the same constituents as the benchmark itself.
• By hard-coding into the system a series of parameters that cannot be broken, for example, reflecting that a fund has a prohibition on investing in a certain type of asset.

By measuring and modeling tracking error and evaluating the difference between the client’s portfolio and the benchmark, the fund manager is able to ascertain the level of risk of over-or underperformance that it is actually running. Increasingly narrow allowances for tracking error are now being implemented which could change a fund into one that effectively works little better than a tracker. The risk of failure being so much greater than the reward for success within the industry further contributes to this natural risk aversion.

Investment Styles

Fund managers will state that they have a particular investment style. Excluding hedge funds, which are dealt with elsewhere in this book, the main styles are:

• Value;
• Growth;
• Passive; and
• Lethargic.

A value house looks to identify individual stocks that are underpriced relative to the market and the firm’s models. These will then be acquired and it is hoped that performance will ensue. Actually, value principles normally lead the market when the market is in decline, but can lag behind in a growing market.

A growth fund, rather than looking at such pricing inefficiencies, is actually looking for companies that are growing in the market and will seek to invest in them. This tends to work very well in a rising market, but can be a problem in a falling market.

Both of these styles will suggest a level of tracking error if a fund is to be measured against a benchmark, the FTSE 100, for example. There is, of course, the problem with the selection of the benchmark. A client may want a fund to follow a growth or value path, yet also wish to compare performance against the FTSE 100 benchmark. The FTSE 100 will include growth and value stocks along with declining and lagging stocks, so this is a difficult comparison to make.

The consequence of this is that the fund manager will temper their stance to take account of the level of risk aversion of the client. As a result, they may become either passive or, at worst, lethargic in their approach.

The passive manager follows the market whereas the lethargic manager only plays at the edge of the fund and holds positions for much longer than would be the case for the value or growth manager.

The risk manager will continually review the tracking error and will seek to involve the investor in the decisions to be taken on the fund. If there is an investment bubble of some form in progress that contradicts the style of the house, an involvement of the investor in the decision-­making process will ensure that the moral risk does not become an actual risk.

Asset Pricing

Another aspect of the market risk in a fund is in the pricing of the underlying assets. Clearly, if there is a liquid market for the assets that are held, then the price at a point in time is clear and will reflect the price quoted in the market. This price is readily available and can easily be reflected in the portfolio valuation. But what about illiquid stocks or other investments where there is no clear price?

The risk manager will be seeking to ensure that the valuation of the investment was fully considered when the asset was acquired and that an independent price is available. In the case of venture or private capital deals, this may involve an independent third party coming up with a valuation. For derivatives, the problem requires the risk management function actually to perform a recalculation of the price. In the case of property, while there are indexes available that are becoming of increasing quality, there is still a problem with independent valuation. Yet, it is these asset classes that can often provide a fund with outperformance.

Performance Attribution

The regulations make it clear to any Board that it is now inappropriate to reward excessive risk taking. Indeed, any performance measure which runs contrarian to effective risk management should probably be rewarded. With bonuses now being either deferred or being capable of being repaid much change is afoot.

For a risk manager in asset management performance attribution is fraught with danger. As a fund manager takes on greater volatility they increase the risk of negative performance and increase tracking error. Clearly fund managers should not be rewarded for achieving benchmark returns—that is what they are expected to achieve. Likewise, any returns that they do achieve should be discounted if they have taken on excessive tracking error as a measure of volatility. However, I know from bitter experience that this is a minefield foe a risk manager. Front office professionals fail to buy into this proposition and consider that they should be rewarded for all gains but not penalized for failure.

To be successful in this area the risk manager needs the Board to really buy into what we are seeking to achieve and to develop a complete set of risk based performance metrics which have been properly explained to the investment professionals, gaining their support to the fairness of the approach adopted. A great objective but hard to achieve in practice.

Model Risk

Where models are used to provide valuations, this creates additional problems. The risk management function needs to be able to both understand and recalculate the asset calculation, understanding the assumptions that are inherent in the primary model and replacing them with an independent model. This needs to take data feeds for pricing that are as independent of that of the primary model as possible.

The question is whether the risk management functions possess the skills necessary to undertake this work. In many cases, I doubt that this is the case. Similarly, such skills are unlikely to be available within either an internal or external audit. The requirement is for an advanced risk management team that fully understands such valuation principles. Such people will have the skills possessed by the front office dealing personnel supplemented with a detailed understanding of risk, perhaps a long shot.

Without this risk management back-up there is inappropriate model reliance. Models are only as good as their underlying assumptions. If these assumptions, for example, liquidity or volatility, are invalid, then the output from the model will be at best unreliable. Generally, we would expect any report on a model to include a statement of key assumptions to be assessed.

Credit Risk

Just as funds do not take direct market risk, nor do they take direct credit risk. The issue is rather different here, however. Generally, funds do not take on credit risk with the intention of making money from taking this risk. It emerges as a consequence of the investment ­process through settlement risk and through placement of excess funds in the market. It also arises in the case of derivatives—mark-to-market ­positions and collateral.

Funds are increasingly building credit rating-based limit monitoring systems to address the risks that arise from counterparty credit risk and also to comply with the requirements of the client agreements. This requires fundamental credit analysis to be undertaken prior to a fund manager utilizing a financial institution for business.

In practice, however, many of the institutions are less than forthcoming with the information required by a fund manager to complete this activity. In the case of derivatives, collateral limits are placed to reduce the exposure to the counterparty and any consequent risk, but this then requires collateral management and collateral mark-to-market to be conducted.

Using credit limits for limit setting is helpful and allows a system to be implemented promptly, but it must be recognized that credit ratings by their nature are more likely to follow a market than to lead.

The risk management function is increasingly leading in all of these areas and providing assurance to the fund manager that there will be no moral losses arising from counterparty credit risk.

Operational Risk

Operational risk encompasses operations risk, together with people, legal and compliance risks. As such, historically it was not considered separately from the normal operations risks that a firm would run. This has now changed and operational risk is seen correctly as a discipline in its own right.

All of the key elements of operational risk should be undertaken within a fund manager:

• Risk appetite calculation;
• Development of a risk register;
• Modeling of business processes;
• Identification of key controls that add value to the institution;
• Development of an internal loss database;
• Utilization of external loss event data;
• Development of key risk indicators;
• Exception-based reporting;
• Implementation of control and risk self-assessment;
• Development of sensitivity analysis;
• Development of stress testing and scenario modeling; and
• Development of business continuity plans for both normal and severe events.

In this way, fund management operations are no different to other financial institutions and similar systems and controls can be implemented.

One area of difference is within security lending. This is an activity where assets held by a fund manager are lent out to a third party to meet obligations that they might have. Within the industry, this is an area where higher levels of vigilance are required. A single security lending loss can easily wipe out a year’s profits from a relationship and this is unlikely to be capable of being passed on to the client. Times when this is higher risk includes the lending of securities around corporate events, for ­example, a reorganization. The risk management function needs to keep such areas under close observation.

Strategic Risk

In common with other types of business, strategic risk within the fund management industry is still not modeled on a coordinated basis with other risks. Key strategic risks arise in the course of corporate activity, product development and sectoral stances taken by the firm. Yet, in practice such decisions are often made independently of a consideration of other types of risk.

The consequence can be that sectoral or industry stances may undermine the ability of a fund manager to really make a difference through asset selection. The risk management function should seek to evaluate the impact of such decisions and to report this to senior management, such that they are able to understand the levels of risk that they are running.

Reputational Risk

Is reputation a separate risk or is it just a consequence of other risks that the fund undertakes? While the Basel Accord does not explicitly require capital to be put aside for reputational risk, this neither means that there is no capital put aside, nor that management should not take the risk seriously.

Reputational risk within the fund management industry is paramount. Fund management companies offer similar services with reputation, size, and performance being the key determinants of differentials. The consequence of this is that the reputation of the firm must be maintained at all costs. Yet reputational risk is often seen as being the responsibility of compliance or the chief executive, rather than being within the remit of the risk management function.

Clearly, if you are to look at the totality of risk that a fund manager incurs, then reputational risk would be part of this equation. It is, therefore, imperative for the risk management team to ensure that the risk is properly considered and that the appropriate policies and procedures are implemented. Not having Basel-explicit capital calculation does not mean that there is no risk to be managed, indeed quite the contrary.

Boards do tend to consider reputational risk as one of the key risks that they are concerned about, yet we do not tend to see significant modeling, perhaps due to the absence of risk management input. So how should it be managed? Clearly, reputational risk is only incurred if a suboptimal event becomes public knowledge. If a firm incurs a loss of £1m and nobody hears about it, then the loss to the firm is just £1m. If the event becomes public, this then increases to a multiple of the loss, depending on the market’s view of the quality of management and whether the institution appears to have become accident-prone. It is, therefore, quite realistic to model reputational risk as follows:

• Identify the likelihood of suboptimal events occurring;
• Identify the loss that would result directly from the ­suboptimal event;
• Evaluate the likelihood that the event would become public;
• Assess the financial impact that would result from such ­disclosure; and
• Reduce the overall liability by the controls that can be ­implemented to mitigate this loss.

At present, however, this is not generally undertaken within the fund management industry.

Liquidity Risk

Liquidity and asset and liability management are crucial to a fund manager. Public cases have identified the problems that can occur. To obtain superior performance, fund managers are increasingly seeking alternative asset classes, including:

• Property;
• The use of derivatives;
• Private equity;
• Venture capital;
• Unlisted securities;
• Commodities; and
• Credit derivatives

Each of these carries its own risk. What they all tend to have in common is that they are not easy to liquidate in a hurry. If a fund has a substantial level of unlisted securities and the investors request their funds to be returned, then the fund manager has a problem. It is unable to get a quality price for the assets due to the absence of a liquid market in the securities. Accordingly, the fund manager tends to transfer the assets from the fund to their own account, or to reimburse the fund for consequent losses resulting from the failure to realize the assets promptly.

Accordingly, it is imperative for the risk management function to consider the likelihood of customer withdrawals from a fund and to ensure that the fund managers have maintained liquidity adequate to meet these obligations.

That being said, full asset and liability management at a fund level is not generally undertaken. Perhaps this is because most of the funds do maintain adequate liquidity for normal purposes. Risk management is not, however, restricted to normal market conditions. Rather, it is intended to deal with stress and under stress liquidity declines exponentially.

Again, it is not difficult for the risk management function to implement a series of metrics and controls in this area, yet this is again often overlooked. We still have too many risk management functions within funds management restricting their attention to market and operational risk.

The Future of Risk Management

There can be little doubt that risk management practices within funds management are undergoing a sea change. The Basel Accord has been the catalyst for many funds to implement risk management, but many still have a long way to go. Mistakes are being made, mostly through what might be considered a lack of vision.

The objective of risk management is to provide management with the assurance that they require that business is being conducted in accordance with the risk framework that they have approved and the risk appetite that drives the process. As such, a series of policies, procedures, controls and, importantly, metrics can be implemented to ensure that the executive management is fully aware of the risks that they are currently running.

Working from the total view of risk, including liquidity, strategic and reputational risk, it is possible to build a consistent structure to ensure that all risks are evaluated consistently. The best way to do this is to use some form of risk-probability approach, akin to the loss-distribution approach within operational risk. It is clear, however, that loss is not a simple proxy for risk, rather the future expectation of impact would be a better measure.

The absence of quality software is one of the main issues that we currently see. The consequence is that the executive committee receives information that has been designed for income-generative purposes (i.e., tactical data) rather than the information that it requires (strategic data). Too often, the wrong information is provided to the wrong people at the wrong time.

In the future, fund managers will maintain risk databases that consider the totality of risk and evaluate future loss expectation using a series of realistic distributions. Risk management will address all of the plausible scenarios which can be foreseen—and we will be talking about Basel V rather than Basel II.

All of this is quite possible to implement now and some of the more enlightened institutions are making moves toward this approach. The win is clear—management adequately appreciates the totality of its risk and the results of the firm therefore become increasingly predictable and certain. What is also clear is that risk management within the fund management industry will continue to grow in the coming years, both in terms of its size and also the quality of the analysis conducted.