The dealer’s spreads widen sharply as it stock falls despite negligible default risk.
One of the most persuasive criticisms of credit derivatives is that they destroy information. That is, they release lenders from the necessity of understanding a borrower’s credit. According to this view, they are more of a “wisdom of the masses” poll—like electoral futures markets—rather than true barometers of a reference entity’s credit. And, like electoral futures, they are usually wrong. One data point supporting this view is the sharp widening of Goldman Sachs’ CDS spread this morning, despite no appreciable increase in its default risk and its recent blowout results.
This isn’t an academic issue for corporate treasurers. Increasingly, backstop revolvers are incorporating CDS spread-linked pricing, leaving borrowers potentially exposed to the vagaries of the CDS market. Two vectors of contagion from the equity markets show how CDS spreads can be pushed around by technical factors; there are a host of other spurious variables that impact the relationship between a company’s credit and its CDS spread.
First, there’s the widespread use of firm-value or Merton-model credit analysis tools like KMV. These use equity prices as a fundamental input to estimate the option value of the equity, and therefore the likelihood a company will default. But if the equity gets pushed around by technical factors, so will the debt. Also, while this theoretical approach looks reasonable (and to be fair to KMV, its system is far more sophisticated than this description), stock price and credit are different beasts.
Take Goldman again. Its stock had declined 3 percent by mid-day on April 26. Meanwhile, Markit reported that Goldman’s CDS spread has widened by nearly 20 basis points, or about 13 percent. Now, the average price of a share of Goldman (theoretically) should be the market’s estimate of the future earnings that it will throw off, which, due to the Securities and Exchange Commission fraud charges, uncertainty over Lloyd Blankfein’s and Fabulous Fab’s testimony tomorrow, and the financial reform bill’s brightening prospects, might reasonably be trending lower today. But the CDS spread is meant to reflect probability of default. And there is no justification for believing that there’s a 13 percent greater chance that Goldman will eventually default than there was on Friday.
The second vector of infection is the bilateral nature of the CDS market, worsened by the information-destroying nature of the product. One credit trader explains it thusly: If, say, Citigroup’s CDS desk, in September 2008, saw Morgan Stanley’s stock fall, and had a big counterparty exposure to Morgan, it would have immediately shorted Morgan’s stock and/or bought CDS protection on it.
This would have caused the stock to fall further, or the CDS spread to widen, freaking out CDS traders at other banks, who would have done the same thing. John Mack’s protestations that shorts were targeting his firm was, in effect, right. But that was largely a function of the need to hedge bilateral exposures by entities that cannot do their own fundamental credit analysis because they’re trained to think the market knows best.
The financial reform bill currently being debated in the Senate could solve the counterparty issue. But if the CDS market continues to facilitate pass-the-parcel lending to entities that have no understanding of borrowers’ credit fundamentals, and exacerbate this by importing equity market noise through firm-value models, corporate treasurers should redouble their efforts to extinguish CDS-linked pricing.