A report recently published by Standard & Poor's that contains wildly diverging actual recovery rate data for defaults in collateralized debt obligations has left many credit derivatives professionals scratching their heads over the huge variations. The data shows the U.S. dollar amount CDO investors were able to recover on corporate credits after the names had defaulted and therefore how much money they lost.
One credit derivatives professional expressed amazement at the disparity in recovery rates on the same corporate credit. For example, the S&P report shows nine different recovery rates paid out on Enron, ranging from USD11.87 and USD23.76 (see chart, below). Officials at S&P did not return calls by press time.
The disparity in recovery rates highlights a flawed cash settlement mechanism, according to another CDO analyst. The variables affecting the recovery rate include the timing of delivery of a credit notice, the selection of dealers that are polled, the size of the obligation and the underlying obligation being quoted. Although the difference in timing of quotes might only be a few weeks, because the bonds will be in default at this stage their prices will likely be volatile.
A similar study carried out by Fitch Ratings carried out for DW found recovery rates to be more tightly clustered. Notably, recovery rates on Enron were USD12-16, SwissAir was USD5-7 and Teleglobe was USD2-3. However, WorldCom was USD12-20. Andrew Jackson, director and head of credit derivatives at Fitch Ratings in London, said the valuation process is the most important detail in CDOs. "A deal would need more credit enhancement to get the same rating if the recovery rates were expected to be low due to a sub-optimal valuation mechanism," explained Jackson.
One CDO pro said it would be in the market's interest to set up an independent fix similar to the British Bankers' Association LIBOR fix used in the interest rate swaps market. This would be run by a credit default swap broker, which keeps a record of traded prices. Using an independent price source would eliminate all the differences except the timing differential. For hard credit events, such as filing for Chapter 11 bankruptcy protection, the price could be calculated a set number of days after the event.
However, other analysts said CDO structurers are unlikely to agree to an uniform price as it would be hard to implement and may not show prices the receives of the obligations could actually trade at.
CDO analysts said they would have expected variations of around five dollars. With little transparency of recovery rate data many investors in CDOs may not be able to compare their settlement against others for the same credit, and recipients of lower settlements may feel robbed, he added. However, one CDO analyst said the difference in deliverable obligations and the timing of recovery rate values allows the most sophisticated institutions to minimize losses.
William May, managing director-derivatives at Moody's Investors Service in New York, said Moody's is working on its first report investigating the recovery rates in synthetic CDO, which is scheduled for release in the first quarter. While Moody's has been accumulating data on credit events for several years, the report is its first attempt to analyze the data in detail, said May. Thus far findings show a wide variation in the recovery rates made on single corporations, he said. May declined to offer specific figures prior to release of the report. Whether or not recovery rates of credit events in synthetic CDOs are more variable than in their cash equivalents has not yet been determined, he added.