Creditflux Newsletter

Defaults aren’t always low and recoveries aren’t always high. The credit crunch shows how whole sectors can go bad

Nobody will want to be reminded of this, but two years ago the buzzword in credit was the "new paradigm". The credit cycle had been abolished as a result of improved credit risk management, and spreads were set to mean-revert around a new permanent low. Defaults were a thing of the past.

Since then, of course, defaults have rocketed and spreads have mean-reverted to the levels of the 1930s. But, in a way, predictions of a new paradigm were right. Recent events suggest that credit does not behave as we once supposed. As a result, we should rethink the way we manage and model credit risk.

In their hearts, credit investors always knew that defaults were not constant and recoveries were not uniform. But almost every calculation about credit investment returns or the relationship between different credit prices was based on these simplistic assumptions.

So what do recent events teach us about default and recovery rates? Clearly, that default rates and recovery rates are inversely correlated. But also that credit losses come in rare sector-specific waves.

Divide defaults into two categories. First, there are the idiosyncratic events where a borrower gets into problems for its own unique reasons. Think of Comdisco, Argentina, Enron or Parmalat. These defaults are often sudden and sometimes technical. For these defaults, recoveries are usually high - because there are plenty of buyers for the company's assets - unless fraud is involved.

Second, and more common, are defaults that take place in a troubled industry. Usually, this follows a lending bubble. There have been numerous examples in the past, including the railroad bond boom of the nineteenth century, the Latin government borrowing boom of the 1980s which nearly brought down Citibank, and the tech and telco bubble of the start of the millennium.

These sector defaults are the ones that destroy most value. They explode like a chain reaction through the industry knocking over good companies and bad. Default rates can reach startling highs. If you think that recent high yield default rates are unprecedented, look back at the initial wave of European high yield bonds - many of them issued by telecom companies. According to Gary Jenkins, head of fixed income research at Evolution Securities in London, 56% by market value of the 2000-2003 cohort of European high yield bonds ended up defaulting.

Sector defaults can also produce very low recoveries. The recent negligible recoveries on chemicals companies Lyondell Chemical and British Vita have a precedent in the single digit recoveries of telecom operators Teleglobe and Global Crossing at the beginning of this decade.

The current global credit crisis can be seen as an extreme form of sector crisis (affecting the financial sector). It has been exacerbated by further simultaneous sector crises, some triggered by the sharp economic slowdown and some (such as in autos) that predate it.

The good news is that these sector-wide events are rare. In the future, default rates of close to zero and high recoveries will return. Losses will be minimal except for the occasional idiosyncratic event.

The recent credit problems will help us better understand credit risk. This is a world of long periods of low risk punctuated by occasional sector credit crises and idiosyncractic events. That is harder to model than a world with 1% or 2% annual default rates and 40% recoveries, but it should not be impossible.

The job of credit investors is as simple as ever. Avoid credits that smell bad. And when you see a lending bubble, get out.