Hedge funds invest in CDOs with three different strategies:
1. Carry trade;
2. Long/short structured credit;
3. Correlation trade.

These strategies enable hedge fund managers to express certain views on credit markets efficiently.

Carry Trade
This is the easiest and riskiest CDO strategy and consists of buying the equity tranche of CDOs. The return generated by the equity tranche depends on the leverage, the maturity and the composition of the basket of credit forming the CDO. Purchasing CDO equity tranches is equal to selling credit protection.

In June 2003, a hedge fund bought Fixed Income Senior Notes for a par value of $5.5 million (hereafter called simply “Notes”) totaling $148 million. These Notes were part of the CDO tranche with the higher claim collateralized by a private placement of high-yield bonds carried out in 1998. The issuance also included another tranche of Senior Notes, characterized by a floating rate for a par value of $34 million.

The Notes subscribed by the hedge fund were due on August 2010 and paid a fixed 6.71% yearly coupon. Technically the issue had defaulted, and the rating of the tranche subscribed by the hedge fund had deteriorated, going from AA3 in 1998 to Baa3 in 2002, turning into a “container” receiving all the capital and interest payments, which were used to pay Senior Notes first, both at a fixed and floating rate, and Subordinated Notes next.

At the time of purchase in June 2003, the par value of the Notes was $4.26 million and the hedge fund bought them for $3.63 million. The Notes were then bought for a price of 85.27 cents, at a substantial discount on the liquidation value of 94.73 cents.

At the time of purchase, the hedge fund had made a conservative assumption according to which 30% of the high-yield bonds would default and the recovery rate would be of 28 %. As a result, the internal rate of return had been fixed at 13.64 %. The investment was expected to be paid back in 3.5–4 years.

At the time of purchase, the underlying portfolio comprised 117 high-yield bonds characterized by an average price of 74.97 cents, including defaulted securities. The initial analysis conducted by the hedge fund also included the expectation that over the short term the default rate of the collateral high-yield bonds would rise, only to decrease again during the residual life of the securities.

Two months after the Notes had been purchased by the hedge fund, $949 786 had been paid out in terms of capital redemptions for the underlying high-yield notes (accounting for more than 25% of the initial original cost incurred by the hedge fund), against an initial estimate of $500 000. Over the same period, $142 808 worth of interest were also paid out. Moreover, the price of the Notes grew significantly with respect to the initial cost incurred by the hedge fund.

The improved performance of the collateral and the higher than expected capital redemptions generated a return on investment of 15.88 %, well above the estimated 13.64 %.

Long/short Structured Credit
This strategy’s objective is to generate absolute returns, regardless of economic and market conditions such as credit spread moves and the general direction of interest rates. The strategy is naturally long credit and benefits from spread tightening. The effect of market spread widening on long positions can be offset by the manager’s ability to short credits that widen more than the general market in a deteriorating environment. The fund employs fundamental credit analysis to determine long and short relative-value positions in different corporate credits. The fund expresses long views through the purchase of CDO equity tranches, and short views through the purchase of single-name Credit Default Swaps.

The long position in CDO equity tranches locks in positive carry. Furthermore, the ability of the hedge fund manager in performing the credit analysis on the companies underlying the CDO generates the trading ideas: a CDO is a basket of credits and the manager can choose inside this basket the credits he wants to be long or short. So the manager can take advantage of opportunities ranging from sector allocation, allocation to single companies and relative-value trading. Finally, the manager can limit the downside from spread widening through the long positions in credit default swaps.

The ideal portfolio built with long positions in CDO equity tranches and with long positions in CDS can have a convexity return profile: an instantaneous equal proportional shift in all spreads, assuming correlation remains constant, should have positive returns irrespective of spread moves. The most difficult challenge for a hedge fund manager implementing this strategy is the
ability to manage the correlation risk among the CDO tranches.

Correlation Trade
CDOs are a recent innovation that enable investors to buy (or sell) limited protection on credit portfolios. The protection attaches (and detaches) once the portfolio realizes certain default losses. The proper compensation for bearing this specialized risk depends not only on the individual portfolio credits, but also on their prospective dynamics, including their prospective interplay. People use the term “correlation” to discuss this interplay, and the term “correlation trading” to capture involvement in these tranched portfolio protection products.

In a correlation trade, credit protection is sold via the purchasing of CDO equity tranches delta-hedged with credit default swaps (CDS). The correlation trade consists of assuming a long position in the equity tranche of a certain CDO and a simultaneous short position in the mezzanine tranche (or a more senior tranche) of the same CDO, in the attempt to take advantage of the spreads among the different tranches of the CDO. This way the hedge fund is long the implied correlation among the CDO tranches.

Note that the implied correlation is primarily a market-based factor, driven by the demand and supply of protection for each individual CDO tranche. This trade has a positive carry, and returns can be generated if the credit spreads in the underlying portfolio move in a parallel way, that is, widen or shrink simultaneously. The correlation trade is exposed to the risk of an unexpected default in the underlying CDO portfolio but, given that in 2005 the prevailing default rate is at historically low levels, many hedge fund managers are comfortable with assuming that risk.
Some investors are reluctant to embrace correlation trading as a model-driven arbitrage strategy.

On 4th May 2005, Tracinda Corp., a company that already owned about 4% of General Motors, offered $870 million to acquire 4.95% of General Motors, and GM shares at the end of the day jumped by 18.1 %. On 5th May 2005, Standard & Poor’s had downgraded the GM bonds (from BBB– to BB) to the status of “junk bond”, maintaining a negative outlook.

This was a surprising capital structure movement, where the riskiest part of the capital, equities, jumped at a time when the less risky part of the capital, bonds, dropped. Many hedge funds were positioned in a correlation trade: they had short positions (buy protection) in CDO senior tranches and long positions (sell protection) in CDO equity tranches, delta-hedging with single-name CDS or iBoxx index. This correlation trade expected implied correlation to increase and had positive carry.

Because of the downgrade of GM, the equity tranche of many CDOs declined sharply while the mezzanine tranche soared because of a “fly to quality” movement. This means that the spreads moved in a non-parallel way. Merrill Lynch estimated that the CDO correlation trades in April lost 12 %, and for some hedge funds this loss was worsened by the use of leverage.

The important point to note is that these new strategies are relatively untried and untested in periods of market stress, and hence investors could underestimate certain risks, i.e. correlation.

This article is a part of “Investment Strategies of Hedge Funds” ebooks by Filippo Stefanini for closed private educations only. You should nuy his books for the best completely informations.