.svg)
By Norman Chait, CFA
AIG Global Investment Corporation
Today's multinational insurance companies are allocating capital to externally managed hedge funds. The major issues encountered every day by investment professionals who allocate to hedge funds on behalf of large insurance companies relate to the specific challenge of managing a large pool of capital within the context of such companies.
In this chapter, I will address risk management in the above context. First, I will focus on what I believe to be the principles of sound risk management. Then I will explore the more practical aspects gathering, aggregating and interpreting risk management data both at manager and 'fund of funds' portfolio level.
I will focus on the substantive issues and, in particular, the risks inherent to the various investments and trading strategies under evaluation. I will not be discussing other important issues, such as operational risk or key man risk, in depth. This is because key man risk, for instance, is a Pandora's Box of personal and personnel issues, including (although not limited to) bloated ego risk, excessive greed syndrome and the effect of not uncommon personal events, eg, the impact of a manager's divorce on her portfolio management.
It is important to note that while certain aspects of the management process pertain strictly to insurance company matters, a similar approach can also be applied to the management of other pools of capital, regardless of size or investor type. It is strongly recommended that one performs regular and consistent operations reviews of one's major holdings, as well as keeping abreast of industry information.
Finally, where possible, I avoid technical jargon and excessive use of Greek letters.
RISK MANAGEMENT - OVERALL PHILOSOPHY
Sound risk management harbours quantitative and qualitative aspects, plus a dose of common sense. In attempting to evaluate where the current and future portfolio risks are, the approach should be proactive rather than reactive. Analysis of historical data is only part of the process: while one may apply lessons learned from the past in developing strategy, it is almost certain that risk factors not encountered before will emerge. Therefore, it is insufficient just to interview hedge fund managers about these issues; one should follow the markets actively and be open to inputs from all sources.
The major issues to focus on are liquidity, leverage and counterparty risk.
Liquidity: Before entering, be sure of your exit!
A thorough understanding of liquidity is perhaps the key factor when it comes to designing any risk management system or approach. Prior to investing, one should have a clear idea of when and under what circumstances the investment ought to be harvested. Many people - and especially hedge fund managers - forget that the main purpose of investing is to generate a cash-on-cash return over some period of time - ie, real money rather than a percentage statistic. In other words, what goes in should come out.
In addition, one should evaluate the possibility of harvesting the investment under adverse market conditions. It is important to ensure that the underlying assets in a particular hedge fund are liquid enough to fulfill all obligations to investors. All hedge funds prospectuses contain terms giving the investment manager the right to suspend the payment of redemption proceeds to investors in a number of loosely defined scenarios, regardless of the general liquidity terms offered. The following theoretical example best demonstrates this.
Example 1
Manager X has created a hedge fund that has only two investments in it. One investment is a private unsecured loan that is not callable for the first year, where the borrower may elect to distribute cash or in-kind payments. The second investment is an unlisted private convertible bond that may be converted into the equity of the company at a 15% discount to market value at the time of conversion. This may only be done after an initial lock-up period of six months. In order to raise capital to participate in these enthralling opportunities, the manager sets up a hedge fund vehicle that offers monthly liquidity, without any initial lock-up period. The reality is that this liquidity provision is meaningless, because its execution cannot been forced in the first months of the new fund's existence.
Therefore, in Example 1, there is no asset-liability match. The next level is to determine the impact of forced liquidations in various market scenarios. While this is difficult to determine on an absolute basis, one can do so in relative terms. Therefore, in Example 2, let us compare two hedged equity managers.
Example 2
The first manager manages US$1 billion in large cap US equities. The second manager manages US$200 million in mid-to-small cap US equities. Assuming current market conditions, one could ask both managers how long it would take them to liquidate 75% of their portfolio without a price impact. For this purpose, the assumption is that they cannot trade more than 20% of the six-month average daily trading volume of each security, as this would create downward price pressure. It is possible and often the case that the large cap manager is able to liquidate in less time, despite having more assets under management.
As a related rule of thumb, if a manager claims it will take 25-30 business days to liquidate the majority of her portfolio, she should not offer monthly liquidity to clients.
But what is more interesting than the answer to this "75/20" question is how long it takes the manager to answer it. Many have never focused on this issue. If a manager is unaware of the liquidity traits of portfolio investments, this should raise a red flag.
One should try to understand the underlying liquidity of each of the asset classes that hedge fund managers typically invest in. This is more difficult to ascertain for over-the-counter securities such as convertible bonds, mortgage-backed securities and high-yield debt. However, one can gauge the relative liquidity of these securities by talking periodically to investment professionals both on the buy side and the sell side, and by gleaning insights from sell-side publications on these asset classes.
Liquidity: practical experience
Certain US convertible arbitrage managers do offer monthly liquidity to clients. But unless the managers use limited leverage or stick with investment grade issues, this could create an asset-liability mismatch. Convertible arbitrage entails the purchase of a convertible bond and the concurrent short sale of the correct ratio of equity of the same issuer. While there are many ways to make money in this strategy (price appreciation of the bonds, positive carry coupon clipping, volatility trading, etc), in most instances the convertible bond is traded less frequently than the equity hedge. Moreover, some convertible arbitrage managers will lever up their investments six to ten times, and will purchase over 20% of some issues.
In cases where managers are running a large balance sheet way in excess of their capital base, it is clear that it will take some time to unwind positions. One leading US manager with a focus on non-investment grade convertible bonds admitted that it would take her up to six months to liquidate the majority of her portfolio. Nevertheless, the fund offered monthly liquidity to investors. Thus, in the event that most investors wanted to exit the fund at the same time, this would have a severely negative price impact should the manager honour redemptions. In a similar instance in 1998, another manager limited the amount of capital that could exit the fund in a given calendar quarter. Redemption requests were prorated and, while the fund offered quarterly liquidity, it took most investors up to a year to get their money back. A large portion of the portfolio was invested in Regulation D convertibles, which tend to become extremely illiquid in times of market stress.
Another issue to examine is the proportion of the manager's assets from any individual investor. If, for example, a hedge fund has 50% of its capital from a single investor who can sell along the same lines as everyone else in the portfolio, this could potentially impact investment results on account of the execution slippage caused by liquidating half the portfolio. It is therefore important to determine the diversification of the investor base. Moreover, it is preferable not to exceed 10% of the assets of a fund because, if an exit is necessary, it can then be effected without impacting the manager's entire portfolio.
Leverage: ROA beats ROE
Leverage is not always undesirable and, when used prudently, can be a useful way to enhance returns. For example, a significant portion of US families run levered investment portfolios because they have mortgages on their houses. But in these cases, there is usually sufficient collateral backing the loan, which can be seized in the event of default. Also, among businesses, there is usually sufficient cash to service both interest and principal debt payments.
The above principles should apply to any levered asset class. In hedge fund land, the problem arises when an excessive amount of leverage is used. For a given strategy, if two managers are able to generate similar returns, the manager who uses less leverage is preferable. One should look for return on assets (ROA) rather than returns on equity (ROE). According to the classic Du Pont analysis, the difference between ROA and ROE is financial leverage.
Let's take a step back and examine hedge fund managers for what they advertise themselves to be. While there are many efforts to try to institutionalise the business, hedge funds are still considered a talent pool rather than an asset class, and the managers therefore promote themselves as the best and the brightest. With the global financial markets as dynamic and diverse now as ever, talented investment managers should be able to discover enough attractive investment opportunities without having to over-lever their balance sheets.
For each hedge fund asset class, one can determine maximum leverage boundaries. As a rule of thumb, regardless of strategy, the net exposure(long minus short) should be lower than 100% of investor capital. The higher the gross leverage (long plus short), the lower the net exposure should be.
Counterparty risk: do not operate on a business plan that puts you at the mercy of others
Counterparties include both debt and equity claims against a manager's portfolio. It is essential to determine whether the hedge fund manager really understands the nature of all these counterparties, and how they will behave in times of stress. Moreover, one should always ask what steps the manager has taken to mitigate and diversify counterparty risk. A hedge fund is a business and should be run as such; it is not just about sitting in a room and picking stocks or trading spreads.
Debt holders
These include Prime Brokers, who provide financial leverage and stock loan to hedge funds. They have policies pertaining to how much each type of hedge fund can borrow, and the margins they are willing to provide. Managers should be fully aware of the notice period given for changes in lending policies or margin requirements. One should also examine the hedge fund manager's knowledge of the intricacies of the stock loan markets. It is useful to know how often the manager is subject to stock loan recalls (which force her to cover a short position at an inopportune time),and whether she has multiple sources of stock loan. Established hedge funds have their own internal policies regarding contractual arrangements with debt counterparties. Funds that employ more than Regulation T leverage (2:1) should demonstrate expertise in documentation of over-the-counter derivative contracts such as swaps.
Equity holders
These are the investors in the fund. Unstable "hot money" investors who want to sell can often affect the remaining shareholders if their actions cause the manager to liquidate attractive positions at unattractive prices. Therefore, it is important for the investor to evaluate both how well the manager knows her clients and whether the client base is sufficiently diversified. This is less important if the fund offers tougher liquidity provisions, such as a limit to how much of the fund can be sold in any given quarter.
LIQUIDITY, LEVERAGE AND COUNTERPARTY RISK ARE ALL INTER-RELATED
While it is important to understand each of these risks separately, liquidity, leverage and counterparty risk are inter-related and cannot be viewed in isolation. The Long-Term Capital Management (LTCM) crisis in the autumn of 1998 best illustrates this. There were many fixed-income arbitrage hedge funds that were affected. They also ran highly levered portfolios - often with similar trades to LTCM. When LTCM had to liquidate, these hedge funds found that they were unable to sell positions at attractive prices. Liquidity evaporated and counterparties became nervous. Investors put in redemptions and, critically, some lenders either increased their margin requirements or refused to roll over short-term loans.
The end result was that a number of funds other than LTCM either suffered double-digit losses or had to close down. Most significantly, this all happened at a time of relative global prosperity, evidenced by the fact that all the US Federal Reserve had to do to effect a full recovery was to inject some liquidity into the system by relaxing short-term rates. Those managers who had properly managed their liquidity, leverage and counterparty risks not only survived 1998 intact, but enjoyed excellent1999 performance, because the number of players in the arbitrage markets decreased.
RISK MANAGEMENT PROCESS
Now we move from the theoretical to the practical aspects of risk management for hedge fund portfolios. These can be divided into three primary stages:
- data collection;
- interpretation of data collected at the manager or strategy level; and
- data aggregation and interpretation at the portfolio level.
Data collection
Contrary to common belief, it is not too difficult to obtain relevant risk management information from hedge fund managers. The key is to know what information to obtain, and how to integrate it into one's own risk management systems. Here are some pieces of information one can usually obtain from managers:
- assets under management, including monthly client inflows and out-flows;
- monthly profit and loss (P&L), with some level of attribution analysis;
- gross and net exposures;
- sector exposure, or, in the case of multi-strategy arbitrage or event-driven managers, breakdown of long and short dollars by asset class;
- for equity managers, details of the top 10 or 20 longs, and the percentage of long market value they comprise (individual shorts are generally not disclosed in writing);
- breakdown of the long portfolio by market capitalization, (eg, percentage of companies in the portfolio with market caps of over US$10 billion,US$1-10 billion, and under US$1 billion);
- percentage exposure to foreign securities; and
- options and futures hedges.
It is not necessary to have full disclosure of a manager's entire portfolio. The utility of analysing whole portfolios is limited relative to the time spent doing so. It is better to receive the necessary information in aggregated form. Some managers will allow their prime brokers to provide clients with the relevant statistics - however, this practice is still not widespread. Below are some of the guiding principles used in determining what risk management information to obtain and how to go about doing so..
- Try to understand the major exposures a manager has, and what the key factors are that keep her up at night.
- Do not ask managers for intricate trade secrets. They are irrelevant to the risk management process. Rather, one should be able to build a sound risk management platform based on non-confidential information. Moreover, the biggest secret sometimes is that there are no secrets. It is no use grilling managers on issues they may be unwilling to disclose when they may be more than comfortable discussing other more important matters.
- Disclosure of individual positions is not important unless the manager runs a concentrated portfolio. However, most equity managers provide atop ten long list, which is helpful in tracking portfolio turnover and idea overlap amongst managers.
- If a manager is reluctant to provide certain information that her peers do provide, it does not hurt to remind her of this fact.
- One should always explain to managers why the information is needed and how it will be used.
- Try to automate the process as much as possible. Often, managers provide sufficient information on their web sites or in monthly fax updates. If not, then it is best to deal with someone in the organisation other than the investment manager, where possible. This could be the Chief Operating Officer or Director of Investor Relations. Discussions with the manager should focus on substantive investment issues. Managers are extremely responsive to focused questions based on previously collected risk management data and portfolio holdings information.
Risk management - evaluation at the individual manager level
Once the information is collected as outlined above, one may perform periodic variance analysis on each fund. Any significant variations provide a reason to speak with the investment manager directly. Indeed, the following factors may raise an eyebrow.
- A large influx of capital into the fund that makes it exceed the capacity level at which the manager had previously said she would feel comfortable.(Note how managers see things differently once they have raised lot of capital - the discussion usually moves from the benefits of nimbleness to the virtues of economies of scale!).A significant increase in gross leverage (longs plus shorts). This should also be viewed in conjunction with capital outflows.
- An increase of net exposure to a level in excess of previously determined net long limit. This is a common occurrence with equity managers who cannot find sufficient short ideas, but who do not want to sell their longs in order to avoid a short-term taxable event.
- A large shift in sector or asset class allocation, especially where this is not indicative of the manager's style.
- For convertible and fixed income arbitrage managers, an increase in leverage when credit spreads are narrowing.
- Special openings either in mid-month or mid-quarter for those funds that only have quarterly openings. While these are sometimes due to truly attractive investment opportunities, one should still, when they occur, take note of the fund's current financial position (high gross leverage and so on.
Risk management in the portfolio context
In managing a large portfolio of hedge funds, one cannot evaluate each hedge fund in isolation. Therefore, when evaluating prospective investments, the following two key questions should be asked.
1. Does the manager meet the quality test? If not, it probably is not worth pursuing the investment, even if it is in an asset class where one would like to increase exposure.
2. If "yes" is the answer to the first question, how does the manager fit within the portfolio context? While many managers may present themselves as the best thing in the investment world since the discovery of the tulip bulb, the reality is that many are really executing the same strategies as their colleagues across the street. Often, two fund managers may share a summer rental in the Hamptons 1 and their investment ideas at the same time. In some cases, an investment professional may have crossed the street to start up a new shop, having been dissatisfied with the terms offered by her former employer. If one invested in both the old fund and the start-up founded by the renegade former employee, the chances are that no real strategy diversification would be achieved (while key man and operational risks may either be enhanced or diminished).
Therefore, it is crucial to ensure efficient diversification at the strategy level. To put it simply, one should invest with a lot of talented managers who all do different things.
Once a prospective manager has passed both tests, then one should determine the appropriate size of the investment by considering the following aspects.
1. The manager's total assets. Typically, one should not exceed 10% of the assets of a fund, because having more than this may make it more difficult to exit.
2. The maximum size limit for each individual investment. Depending on the size of the overall hedge fund portfolio (or fund of funds), a single manager should not exceed 5-10% of total portfolio assets. The larger the portfolio, the smaller the position size should be, leading to a more diverse portfolio.
3. Optimal and maximum exposure to each asset class. It is not necessary to determine minimum exposures to any asset class. Accordingly, one should not invest in a particular asset class, however attractive, if there are no good managers. Rather, a maximum asset class exposure level should be determined. The next step is to aggregate long and net exposures by asset class to see if these boundaries are exceeded. As many managers are able to shift assets between strategies, one should monitor exposure levels on a monthly basis. This is very important in helping to decide where to place the incremental dollar.
The idea is to add one's exposure to each major asset class on a gross and net dollar basis. Therefore, if manager Z accounts for 5% of the fund of funds portfolio and has 150% long dollar and 100% short dollar exposure to merger arbitrage, then assume, on a portfolio basis, that the manager adds 7.5% gross long merger arbitrage exposure and 2.5% net merger arbitrage exposure. If the manager also has 50% of investor assets long in convertibles and 40% short, then assume this adds 2.5% gross long and0.50% net exposure to convertibles on a portfolio basis. One can add up these exposures on a dollar basis for each fund, and then aggregate the data from all the funds at the portfolio level.
The important distinction to make here is that actual assets allocated to each asset class are counted, regardless of how the manager is classified. This is because both multi-strategy arbitrage and event-driven managers invest in merger arbitrage, and both macro-oriented and equity-hedged managers invest in stocks. It seems that in today's investment climate, the three major areas where a large portfolio may easily become over-exposed are fundamental equities, merger arbitrage and convertible arbitrage. Smaller portfolios may also become over-exposed to distressed debt, statistical arbitrage, fixed income arbitrage, trend-following CTAs, volatility traders and emerging markets.
Let us illustrate how one may determine maximum exposure to the three larger asset classes.
Equities
This area focuses only on fundamental stock pickers, whether hedged or long only, and not on other strategies that may use equities - eg, short-term trading or systematic statistical equity arbitrage. (Assume at this stage that the latter two asset classes have little market exposure - they will be accounted for later on.)
The fundamental equity book should first be examined in isolation. The investor should determine a comfort level with regards to maximum gross and maximum net exposure. By determining this on a portfolio basis, one can use a limited number of long-biased managers (often without having to pay a performance fee) if they are offset by more hedged players who will bring the aggregate net equity exposure to the desired level. As a rule of thumb, the aggregate gross leverage should not exceed 150% of the assets of the managers that invest in this asset class. Funds with gross leverage in excess of this level often tend to be too volatile, and may test margin requirements. Net exposure should be well below 100%. There is no point in giving up 20% of the performance in the form of an incentive fee for a levered long-only portfolio. A comfortable range net exposure level is between 35% and 65%, depending on how much market exposure one is willing to tolerate.
One may further fine-tune this to determine maximum geographic orsector limits, but this entails a lot more data collection. It is usually better to invest with equity managers who invest in many sectors as opposed to sector specialists. The latter tend to have a net long bias, often fall in love with their sector, and do not hedge sufficiently in times of sector weakness. Generalist managers have the advantage of shifting assets to the more attractive sectors. While it is not always practical to track sectors systematically every month, one can gauge this once every 3-4 months, or on amore random basis. For example, when a number of interesting new technology-related funds knocked on our door recently, I immediately added up the long technology exposure with our existing managers and came to the realisation that the technology exposure was sufficient.
Let us now assume, that one desires a maximum average net exposure of 60% to equity managers. The next stage is to determine how much net exposure to fundamental equity strategies one is willing to accept on an overall portfolio level. If the answer is 20%, then one could only invest up to a third of one's portfolio with managers who invest in this asset class.
One should not forget that this is really a rule of thumb exercise, which does not address issues such as tracking error in hedging or sector bias. However, it does go a level further than just determining that if, for example, one gave US$10 million to Manager X, then she is 10% of aUS$100 million fund. The key driver of the system is the tracking of actual portfolio dollar gross and net exposure every month. This allows for the making of informed portfolio management decisions. The following example is indicative.
In April 2000, following the NASDAQ correction, managers became more defensive and their average net exposure dropped by almost 10%. As a result, one could see that there was a potential opportunity to add to some existing hedged equity managers. This triggered analysts to call these managers to get a sense of how they saw opportunities in the coming months. It became clear that most were afraid of the Federal Reserve raising interest rates and would remain defensive through the summer. With one small exception, it was decided not to beef up exposure.
Here, the process is more important than the result. By having the information at hand, one is able to perform timely research that allows for informed portfolio decisions.
Merger arbitrage
The reality today is that most managers invest in the same deals. An effective risk management model therefore assumes that if a well-known large merger breaks, all managers who do invest in merger arbitrage will be in the deal. This limits the marginal utility of adding another merger arbitrage manager to the portfolio over and above a core number, as little additional diversification is achieved.
There are also managers whose main focus is not merger arbitrage but they nevertheless, still invest in the strategy. Therefore, dollar merger arbitrage long and net exposure should be tallied up manager by manager. Assuming that only 10% of a portfolio as counted by investment dollars is invested with pure merger arbitrage managers, total dollars invested in merger arbitrage may amount to over 30% of the fund of funds portfolio assets. This is a function of both leverage and merger arbitrage investments by multi-strategy funds.
From a risk management perspective, the key issue is to decide how much exposure one is prepared to lose on a total portfolio basis if a large deal breaks. As a rule, managers try to limit individual deal break risk to3% portfolio loss (one can fine-tune this by asking each manager what their risk limit is, and then aggregating this on a weighted basis). In addition, if one does not want to lose more than 1% on a fund of funds portfolio basis, then one must cap the total long dollars allocated to merger arbitrage to no more than 33% of fund of funds assets.
If one wanted the above 33% overall portfolio cap on merger arbitrage, and the weighted average long dollar exposure to merger arbitrage per manager that invested in the asset class was 150%, then one could only allocate 22% of the portfolio to these managers.
Based on the experience of merger arbitrage managers (especially in 1987and 1989, when many deals broke simultaneously), one cannot rule out four deals breaking at once and causing each manager who has invested in merger arbitrage a 12% loss. All the managers trying to close out the position at the same time may compound this loss.
Convertible arbitrage
The rules here are similar to merger arbitrage, and one should add up one's gross and net dollar exposure to the asset class. For example, if there are only four managers in the portfolio with a 5% weighting each and they are levered 3:1 on the long side, then the portfolio exposure to long convertibles would be 60% of the total capital allocated to all hedge funds. If the managers were short 70 cents for every dollar long, then the net dollar exposure would be 18%.
As a rough rule of thumb, one can assume that half the short exposure is related to the delta hedge of the convertible, while the other half is the premium dollars at risk. The premium is the value of the convertible over and above the sum received if the bond is converted to equity. The premium exists as the convertible bond typically pays an interest coupon in excess of the dividend of the underlying stock. Some managers will disclose the exact delta-premium breakdown upon request. Therefore, in the above example, if convertible premiums were to collapse to zero, the overall fund of funds portfolio could lose 9%. While this is an extreme scenario that is unlikely to occur, it does allow one to place an outer boundary on the maximum portfolio loss.
One could also determine one's maximum desired exposure to convertibles based on either portfolio sensitivity to sharp moves in interest rates, the equity markets, or based on past performance. Convertible arbitrage is a cyclical strategy, which has tended to under-perform every four years or so (1990, 1994 and 1998 are good examples). One should also take heed that this asset class is fairly illiquid, owing in part to the fact that hedged convertible traders comprise such a large part of the market now. There is thus crowded exit risk.
Where possible, one should ask for a portfolio breakdown between European, Asian and US convertibles, because each of these markets has different characteristics. For example, the US market is far more credit sensitive than the European or Japanese markets. Such a breakdown can give one a better understanding of portfolio diversification.
Evaluation of market risk in the portfolio
This can be determined first by examining the net market exposure of all the fundamental stock pickers on a portfolio basis, as described above. Whereas all the other strategies in the portfolio are supposedly market neutral, there is always some long bias, especially in merger arbitrage and other event-driven strategies (even when they are hedged with S&P futures). Therefore, it is essential to estimate the market exposure of these other strategies through qualitative judgement, with the help of historical statistics. One may over-estimate in order to be conservative. So, if net long equity exposure is 20%, a market risk factor of 10% could be added, so that the expected net market exposure of the entire portfolio is 30%. This is not an exact science, but it is a useful exercise all the same.
QUALITATIVE RISK MANAGEMENT ISSUES
The above quantitative analysis provides a platform for understanding manager and portfolio risks, and in so doing allows one to pose informed questions to managers when detecting deviations from the normal state of affairs.
However, quantitative analysis alone is insufficient, meaning qualitative facets must be part of any sound risk management system. Here are a few items to focus on.
Timeliness of replies
When calling a manager to obtain an answer to some factor highlighted by the risk management system, one should note the time and the manner in which these queries are answered.
Understanding the underlying investment philosophy of each manager
This provides a framework for future evaluation of how the managers go about executing their strategies. A sound approach is to write detailed investment reports prior to each new investment and revisit these reports periodically to check whether the managers have deviated from their investment style or mandate.
Evaluation of past mistakes
One should get a sense of whether or not the manager has readily applied lessons learned from the manner in which questions regarding past mistakes are answered. In many cases, managers shrug off these questions. Sometimes, one may ask for examples of losing trades, only to be given situations where the manager did not make or lose money. This is not the same thing. A solid manager should have nothing to hide.
Understanding the process of how managers mark unlisted securities
For unlisted securities, it is important that managers receive quotes from parties who are willing to execute at the prices quoted. For example, there was a situation once where three managers owned a private security. One of the managers marked the security at 75 cents on the dollar, while the other two marked it at 30 cents on the dollar.
Understanding how the portfolio is priced
Does the manager price the portfolio, or does someone else in the organisation do it?
MANAGING RISK WITHIN THE CONTEXT OF A LARGE INSURANCE COMPANY
Fully invested portfolio
Hedged fund investments are generally managed as part of the insurance float. The float is the pool of capital that resides at the insurance company from the time the insurance premiums are collected to the time they are paid out. As insurance underwriting and claim activities occur continuously, this pool of capital is permanent.
The majority of the float is generally invested in investment-grade fixed-income instruments. The other, riskier investments typically constitute a yield enhancement overlay on the bond portfolio. Bonds are sold when hedge fund investments are made and bought back when such investments are sold. Bond positions are not counted as part of the hedge fund portfolio.
As such, the hedge fund portfolio is always fully invested. No cash or cash equivalent positions are held. Therefore, in a market downdraft, the portfolio could possibly take a larger hit than a similar portfolio able to raise cash. Therefore, the only way to manage risk is to reduce the size of the portfolio on a dollar basis during risky market periods, or to purchase portfolio insurance. The latter is expensive and should typically be avoided. Therefore, one must be especially vigilant in selecting managers who will protect their portfolios in tough times. It is theoretically possible to be fully invested with managers who have 100% cash positions. One has to rely on the managers to raise cash when necessary, as it is not possible to do so at the portfolio level.
Managing a large pool of capital
For liquidity reasons, it is preferable not to be more than 10% of the total size of each invested fund. This prevents one from investing in smaller funds as investments of under US$10 million do not make a meaningful impact on portfolio returns. As mentioned, a maximum individual position size limit should be applied. As such, the portfolio is more diversified than many smaller portfolios.
The ideal would be to find twenty-five 4% positions that one would never have to sell. Unfortunately, it takes much time and active monitoring to determine whether a manager has the ability to generate consistent returns going forward. Therefore, positions tend to be built up slowly.
Finally, if a position exceeds the maximum position limit, one should consider taking profits to pare it back. An exception to this would be in the event of the particular fund closing to new investment and the insurance companies overall hedge fund allocation being expected to increase. As there are limited times in which one can invest and sell hedge funds, it is difficult to fine-tune portfolio exposures unless the portfolio is more diversified.
Liquidity issues
The underwriting activity of an insurance company is influenced by the degree of liquidity of the investment float. Therefore, one should strive to improve the liquidity terms offered by the hedge funds. In most cases, at least quarterly liquidity should be made available. Less liquid strategies should be deferred to other alternative assets classes, such as a private equity portfolio.
Taxation issues
A major issue encountered is the necessity to reduce corporate interest expense. Thus is it prudent to avoid highly levered strategies, such as fixed-income arbitrage. This also makes sound investment sense, as the highly levered strategies are most susceptible to blow-up risk.
As insurance companies have corporate form, they are less sensitive to issues that affect individual investors, such as capital gains tax. Therefore, where possible, it is not advisable to invest alongside individual investors, on whose behalf the hedge fund manager may try to tax manage the portfolio. Tax management strategies often lead managers to hold to investments for longer than they would have had after-tax returns been less of an issue. This can impact pre-tax performance.
CONCLUSION
This article has spanned the philosophical and practical aspects of risk management for hedge fund portfolios. While this is not rocket science, it is useful to develop a consistent, rigorous and comprehensive risk management system based on hard data provided by managers on a regular basis. A successful system, combined with experience and sound judgement, will allow the portfolio manager to make timely data-driven risk management decisions.
Taken from 'Managing Hedge Fund Risk' by Virginia Reynolds Parker, published by Risk Books ©, www.riskbooks.com, Tel: +44 (0) 20 7484 9757 E-mail: books@riskwaters.com
