New Database Supports Case For Loans

Investors in bank loans now have evidence of what they already suspected: recovery rates in loans are significantly higher than those of high yield bonds.

Introducing the latest analytical tool to the evolving loan market, Portfolio Management Data last week rolled out a new credit loss database that tracks recovery data on defaulted loan syndications.

The database, formally introduced at a PMD/Standard & Poor's seminar last Tuesday, provides default and recovery data on more than 300 obligors who first defaulted between 1987 and 1996. More current defaults were excluded because, in many cases, the bankruptcy and recovery process had not had the chance to run its course, said David Keisman, principal with PMD.

While Keisman revealed a host of interesting statistics from the database, the most important finding was disparity in recovery rates in the various asset classes. Bank debt recoveries, on average, were 84.5% - significantly higher than the recovery rate for senior secured bonds (65.7%), senior unsecured bonds (49.3%) and senior subordinated bonds (36.8%).

That distinction has been made repeatedly by players in the bank loan market in the last six months or so, especially in light of the bond market turmoil during last year's third quarter. But with the exception of a few recent studies, loan pros haven't had much outside support to verify their claims.

"It's a tool we could use in our marketing to quantify the differences in recoveries, and hence volatility, between our asset class and high-yield bonds," said Payson Swaffield, co-portfolio manager at Eaton Vance. While loan defaults have been tracked for a number of years, market players said this is the first time that someone has combined loss data with recovery information. Equally important, sources added, is the fact that the database comes from an objective non-participant.

One of the more valuable aspects of the database, sources said, is the ability to break down defaults and recoveries by industry. Keisman noted in his presentation, for example, that retailers logged the most defaults while telecom issuers were the least likely to default. - AR