When the Music Stops...Hedge Funds - Looking Behind the Numbers
Hedge funds received a lot of press in 1998—does anyone still remember Long Term Capital? In September, the funds received a $3.6 billion injection from a consortium of the world's largest banks and securities firms. Then there were a host of other collapsed funds—all with good track records and reputable management. The losses primarily occurred following Russia's default, turmoil in the emerging markets and subsequent interest rate volatility throughout the fixed-income markets in September and October.
Could these debacles have been avoided? Some answers might lie in better risk management procedures, more disclosure and stronger regulation. Even with improvements in these areas, however, the likelihood of another major mishap is probably certain. Therefore, what do bank credit managers need to review in considering exposure to hedge funds? What assets do hedge funds actually own? Is the exposure secured, and if so, to what level? Further, while this article focuses on hedge funds, there are several types of entities to which the issues addressed herein similarly apply, like mortgage REITs.
Audited financials reveal little with respect to the fund's risk management procedures, investor base, strategy, valuation methods or position details. In addition, by the time the year-end financials are released, the information is likely already outdated given changes in strategy and valuations. Month-end releases to creditors or investors are also likely limited.
Therefore, one has to look behind the numbers and ask the right questions to the right level of management to gain insight into the inherent risks.
Strategy vs. Return Goals
While the lines can be blurred, most funds fall into two style categories; they can be directional or macro bets (e.g., on movements of the Yen vs. U.S. $), or relative value plays (e.g., long and short similar instruments to profit from mispricings). Subcategories include event-driven, global-macro, market-neutral and sector strategies. The credit manager must understand the relationship between these strategies and returns. For example, is the expected return consistent with historical returns for this type of strategy? A close review of the strategy specifics may reveal inconsistencies and exposure not otherwise considered.
Certainly a hot topic of late, risk management procedures vary widely among funds and are not necessarily standardized. For example, a lot depends on the management structure—i.e., checks and balances. One needs to understand the roles that the principal(s), CFO, treasurer, and controller play in the day-to-day and strategic decisions being made by the fund. This includes the negotiation of terms and conditions for legal documentation—does the fund engage counsel for this purpose or are the decisions being made by the traders? This being said, the background and experience of these individuals is equally important.
In achieving the aforementioned return goals, risk models/metrics are typically utilized to approximate market swings and sector movements, and assist in asset allocation. This is where some credit folk may falter—buzzwords and lingo can create a false sense of security. The complexity of these risk models can often be overwhelming; however, understanding the primary parameters is not. The point is, if you are going to ask for details, be prepared to dig deeper beyond the buzzwords to comprehend where the risk is being borne as well as the size of the fund's risk appetite. Understand the outside range for risk relative to historical market performance and to your own credit policies.
A key variable in these models is the volatility assumed over a given time horizon and expected deviations from the norm. Another is correlation, which is used to help diversify market risk exposure. Take note—recently volatility and correlation have not maintained their historical relationships (e.g., between stocks and bonds), which can throw off the models considerably. Another thing models cannot accurately measure is market liquidity risk. Recent catastrophes in the bond markets could never have been predicted within the normal assumptions of models/metrics.
Models often assume that portfolios can always be liquidated and/or deleveraged, even in periods of high illiquidity, which can only add to the downward spiral.
Positions, Exposure and Leverage
With reference to return goals and strategy—what exposure is being taken to achieve these goals? What levels of concentration are necessary (i.e., sector, maturity, type, currency or credit quality)? Does the fund have internal controls to ensure that these concentrations are limited? Are stop/loss rules in place to get out of positions following a specified loss? How are the exposure classifications defined? The questions are limitless. The key point is to remain focused on how the dictated strategy translates into exposure for the fund. Is this within your institution's credit risk appetite?
Measuring total leverage is another complex matter that is also not generally reflected accurately in the financials. While repos and reverse repos are the primary sources of leverage, swap derivatives are off-balance sheet instruments and should also be considered in calculating total leverage. A commonly used derivative is a total-return swap, which mirrors the mechanics and risks of a repo, but without the balance sheet leverage. Therefore, the total fair value of assets relative to invested capital, both on- and off-balance sheet, is a better measure of gross leverage. However, net leverage can also be measured by excluding short and long positions on the same instrument or similar instruments (e.g., treasuries).
The next big question is understanding how the positions are financed—i.e., under repo, prime broker financing or total return swap. Unlike lending to a corporation, where assets generally do not vary much from day-to-day, hedge fund assets are marked daily to reflect their inherent market price, and can vary substantially over a very short period of time. The securities under repo (or swap) can be of many types and qualities, such as:
- long vs. short maturities;
- investment grade vs. high yield;
- domestic (U.S.) vs. foreign;
- U.S. dollars or foreign currency denominated;
- issuer—corporate, sovereign or municipal; and
- structured/collateralized, including CMBS, ABS, etc.
Noting the substantial differences in risk based on these factors/types, one should consider exactly how your credit lines are being utilized—i.e., what types of securities are financed. If the fund defaults, one would most definitely prefer not to be holding the bag with the most thinly traded securities. These securities have fewer market-makers, which generate fewer bids, and therefore likely lower blow-out prices in a liquidation. Simply stated, U.S. treasuries are more likely to generate proceeds to fully cover repo loans than structured mortgage-backed notes. Even though a security is dedicated as collateral against a repo (i.e., "secured"), its value in a liquidation may be substantially below the amount of the loan, particularly if it is thinly traded or if the market becomes suddenly "illiquid."
The securities held as collateral under repos are not available to general creditors in a default or liquidation scenario, except to the extent that the liquidation of the securities by the repo counter-party results in proceeds in excess of the amount owed—an unlikely scenario. In fact, under the U.S. Bankruptcy Code, repo lenders may generally set off mutual debts and claims arising under or in connection with repo agreements without regard to the automatic stay imposed by the Code. Similarly, the "avoidance" powers granted to a bankruptcy trustee or a debtor-in-possession do not apply to margin payments made by or to a repo participant in connection with a repo agreement or to transfers made under swap agreements. Case law suggests that repo transactions will be treated as buy-sell arrangements rather than "secured" lending arrangements. As such, the non-defaulting party does not have to lift the automatic stay to access/liquidate the collateral. In addition, most master agreements include automatic setoff rights to the non-defaulting party for all of its repo close-outs. For example, if one repo close-out results in an amount due to the fund and another results in an amount due from the fund, these proceeds can be set-off against each other.
Swap close-outs are similar and can also often be set-off against one another, if contracted into the International Swaps and Dealers Association (ISDA) master agreement. However, swaps vary in that they are off-balance sheet in the first instance, and are only reflected as unrealized profits/losses in the income statement. When they are liquidated, their realized profits/losses can vary widely relative to the fund's historical mark-to-market value.
The fund may also have master set-off agreements with some counter-parties that permit set-offs across types of trades (e.g., swaps and repos), as well as across legal entities (e.g., ABC Bank Inc. and ABC Finance SA). This is particularly noteworthy when considering prime brokerage arrangements. For example, a credit balance at the time of liquidation in a prime brokerage account may not necessarily be protected against set-off from amounts owed to another legal entity of the same broker. Even if not contracted, in some legal jurisdictions such as the U.K. or Cayman Islands, set-off may be permitted against monies in prime brokerage accounts. In fact, this right sometimes cannot even be contracted out.
Documentation/Default Valuation Methods
Governing master agreements are rarely interesting and often overlooked—except in a default. The Public Securities Association (PSA) or PSA/International Securities Market Association Master Repurchase Agreements are generally used for repos, while the ISDA Master governs swaps. Under a default scenario, the non-defaulting repo or swap counter-party has the right to liquidate the trades to cover its exposure without the consent of the fund or the creditor pool.
For example, the assets under repo are generally liquidated by methods that the non-defaulting party deems "commercially reasonable," using the "best available offer price," or using prices "obtained from a generally recognized source." Needless to say, these methods are not clearly defined, and in a liquidation they rarely generate excess proceeds available toward the general creditor pool. While the ISDA language is more clearly defined (i.e., generally under the Second Method and Market Quotation), substantial flexibility remains with the non-defaulting party when terminating the trades. It is vital to understand the risks associated with these definitions—both from the fund's and counter-party's perspective.
Mark-to-Market Valuation Methods
The net asset value (NAV) of the fund is only as meaningful as the values assigned to each of the trades/positions. Valuation methods vary for different types of securities—from dealer-based quotations to complex models and matrices. It is important to consider how the fund's methods compare with market convention and the default valuation methods under the master agreements. Generally, the fund should obtain quotes from two or three dealers when collecting its marks for non-exchange traded positions. In addition, consider if the fund marks its entire portfolio daily, or only at month's end. Does it rely on the brokers as calculation agents for the large majority of positions? The likely answer may be some combination of the above. More important, however, may be the quality of these sources—are these dealers the market makers for these securities? Are there any inherent conflicts between these dealers and the fund—i.e., are they also investors?
Let's face it—the term "hedged" is often a misnomer. Be wary of how this is defined. For example, "hedged" can mean that the fund is long and short the same amount of the same security for the same term (i.e., arbitrage). But if these two trades are not with the same counter-party, each counter-party may value the underlying security differently, thereby demanding different levels of margin. And in a default scenario, the liquidation prices are not likely going to be identical. So in short, while the fund may be hedged with respect to market risk, counter-party risk remains. However, "hedged" may also mean "partially hedged"—e.g., long $200 million and short $150 million. This translates to market risk exposure. Hedging can also employ long and short positions of different but related securities—e.g., those in the same sector, currency or quality. Any dislocation in the original structure of the hedge can create exposure. There is rarely a "perfect" hedge.
Another case in point is the using of indices or treasuries as a hedge. In theory, both of these instruments should move in tandem with the fixed income instrument being hedged, and hedge the basis risk. However, in very high periods of market and interest rate volatility, they may not be very responsive or may actually move against the hedge. For example, a long position hedged with a comparable index or treasury might move well ahead of the hedge. Unless the hedge is adjusted, significant market exposure exists.
A third example might be using currency-forward rate contracts to hedge sovereign price/credit risk. While these trades, when used properly, can reduce sovereign risk, they also can carry unique contractual legal risks. Emerging market trades, in particular, often include specific language either in the confirmation itself or built into the governing master agreement, which can leave either party widely exposed to a potential contractual dispute. This would include market disruption events and government currency restrictions. What might therefore appear fully hedged is suddenly an exposure. As noted above, the documents can be ambiguous and contain several "outs" and plenty of "wiggle room."
"Let the Lender Beware"
We can only begin to touch upon the lack of transparency with respect to the reporting and disclosure of hedge funds. While this may improve over time, the long-term capital debacle has begun to fade from memory. A well-known but rarely followed idiom, "know your client," is something that should remain close at hand, however. Get close, very close, to management, and involve traders/business people in understanding how the fund's strategy is expected to translate into returns—and ultimately, repayment. If nothing else, we hopefully will have learned something from the recent debacles to better manage credit exposure.