The number of long/short CDO structures, or strategies that generate alpha in a deteriorating credit environment, is on the rise. There are very few managed CDOs currently in the market that do not try to include some sort of short strategy. Incorporating efficient long/short techniques into CDOs has become one of the most challenging goals because extracting every last basis point of extra yield is no longer generally perceived to be the best strategy in credit.
It is easy to understand why the consensus over going long corporate credit which was generally the correct credit strategy over the past four years has evaporated. The momentum behind a four/five-year bull market is facing a serious check in the form of sub-prime woes and slowing US economic growth.
In the CDO world, an efficient long/short structure is something of a holy grail. It is one thing to make money from taking a leveraged exposure to a CDO portfolio, but creating value when spreads also widen is clearly an attractive proposition.
In recent years the rating agencies have accepted the incorporation of short strategies in CDOs but tend to penalize short buckets in CDO structures that is, a ring-fenced portion of the deal used by the manager to express a negative view on individual credits. This is because short positions are generally costly. Because the mechanics of a long/short structure create less subordination for the CDO, there is a greater requirement to increase leverage on the long portion of the portfolio. Also, rating agencies are conservative around giving too much trading flexibility to managers in this form.
There is no consensus on the best structure. In the cash US ABS CDO market, for instance, TCW Asset Management via Deutsche has sold two hybrid long/short deals called Dutch Hill, where long mezzanine cash RMBS positions are complemented by buying protection to individual credits views as vulnerable. This fundamental credit approach was remarkably prescient the first was priced in late 2005, the second followed up approximately a year later coming ahead of the woes in the sub-prime sector.
But some of the innovations currently taking place seek to position CDOs for systemic widening without losing too much in terms of carry. One strategy was undertaken in March of this year by Solent Capital via BNP Paribas, in a deal called Waypoint. This 260 million transaction was created to make money in the event that credit markets widen, but took an innovative look at how to get around the limitations of existing long/short strategies.
"This structure enables Solent to change the maturity in response to changes to spread. Its a mezz CDO and they can mismatch or match maturities depending on where spreads are going. A reserve account uses the BNP Paribas platform. It allows trading, not just on credit, but also on maturity," explains Joe Lovrics, head of structured credit sales at BNP Paribas.
This structure effectively takes a view on future credit curves, offering the manager flexibility through choosing a portfolio where some reference entities had shorter maturities than the tranche maturity date leaving room for future replenishment.
BNP Paribas called the structure a multi-maturity CDO. The initial portfolio includes four-, six- and eight-year assets. The bank argues that because of the roll-down effect over time in a normal long/short CDO, the benefit of having a short bucket is diminished. The multi-maturity CDO gives Solent the ability to extend the average life of the portfolio after spreads have widened.
Another interesting avenue is to go long and short on different maturities of the portfolio. It appears to overcome some of the problems inherent in traditional long/short strategies. There is a rating agency/economic arbitrage because using two different maturities makes it possible to take advantage of possible differences between the spreads and the historical default rate for a particular maturity. So it might be efficient (from a rating agency perspective) to have shorts with shorter maturities and longs with longer maturities.
But pricing these CDOs using the standard base correlation model is not straightforward. Arrangers need to find the best fit for correlation between several maturity dates. It appears that new correlation modelling enabling innovation such as these.
Earlier this year Goldman Sachs approached Axa IM with a structure that tackles the traditional CDO long/short limitation in a unique manner. Axa IM takes an equity tranche position in a long/short portfolio comprising 80 names. The manager selects a long and short portfolio that is equal in size. It is called Credit Outperformance Notes, is 155 billion in size and closed last month.
"It still gives Axa flexibility to create alpha but there is positive carry right from the start," says Pierre-Emmanuel Juillard, head of Axa IMs structured finance division.
Juillard says that while it is not super-innovative it avoids issues like adverse selection prevalent in some other long/short strategies. Axa is managing a long portfolio and a short portfolio of equal size, the latter consisting of credit or sectors that are trading tight for instance the monolines. The long portfolio has a wider spread than a short portfolio, about twice as wide. The aim is to benefit from possible spread widening rather than actual default.
"Its a superior transaction to what we have seen before to play short strategies in the context of CDOs. In a traditional CDO, the size of the short bucket tends to be 10%, which means that managers are tempted to short a few expensive names with a view to default. But unless managers get both the credit view and the timing right, the short creates a drag on the return of the equity. Because we manage an equity on a long/short portfolio with a one-to-one notional between the long and the short portfolios, Axa is able to do multiple tiny bets on the short portfolio and to take a view on spread widening. It allows us to generate alpha on the short as well, through multiple small gains, as we have done consistently on the long portfolios of our previous transactions," says Fabrice Rault from investor relations at Axa IM.
The big question is whether or not other structures are fundamentally more suitable for expressing these turn-of-the (credit)-cycle objectives: namely managed CPDOs and CPPIs. Unfortunately the general fixed-income market acceptance of these techniques is tentative, and the fact remains that the best-performing CDOs in recent years have been long only.