Canso Investment Counsel

Canso Corporate Bond Newsletter

2006 May Newsletter

The recent strength in the U.S. economy is pervasive. Unemployment fell to 4.7% in March in the United States which is a five year low. Wage increases are running in excess of inflation and show no signs of slowing with the growth in manufacturing. Canada’s economy is strong as well. In March of 2006, the Canadian unemployment rate dropped to a 32 year low of 6.3%. This is the lowest level since 1974, another period of rising commodity and oil prices. Wage gains are running at 3.5% in Canada, well above the inflation rate of 2.2%.

This economic strength in the U.S. gives lie to the market consensus that the current round of Fed tightening is nearly finished. It is also causing long term bond yields to increase relative to short term yields. In early April, U.S. 30 year treasury yields moved beyond the critical market technical level of 5% on the very strong March employment data. This puts the 30 year yield .25% above the Fed target short term rate of 4.75% and returns the yield curve to “normal” configuration. Since a key portion of the market consensus was based on the thesis that an inverted yield curve was a precursor of economic weakness and recession, the normalization of the curve contributes to the growing realization that the economy is strong and the Fed could be raising interest rates for some time to come.

Patently Stupid Investment Products

The long period of very low and negative real interest rates has engendered a blatant disregard among investors for capital risk. The bond market in particular has become a magnet for patently stupid investment products. Credit derivative swaps (CDS), collateralized debt obligations (CDO) and all manner of credit products have been sliced and diced into highly levered and risky products. The preference for risk in these areas has been feeding back into the cash markets for credit which is driving credit spreads into very expensive levels.

Bankers Behaving Badly

Bankers are not immune from the hysteria to accept capital risk that has accompanied this long period of easy money. In a research piece called “Loan Volumes Surge, Covenants Shrink in 2005” (April 5, 2006), FitchRatings examines the “steady loosening of lending standards among banks on commercial and industrial loans”. FitchRatings concludes that the rush to lend and standardize loans for sale has led to a substantial deterioration in the covenant standards in bank loans. Since loan covenants are the contractual way that banks reduce a borrower’s ability to worsen their credit condition, this means increased likelihood of default, especially among lower quality borrowers. As FitchRatings puts it:

“One result of this risk receptivity and “borrower friendly” funding environment has been visible erosion in covenant usage. This trend has been most acute among non-investment grade loans despite a steady decline in the credit quality of newly originated deals.”

Since bankers exhibit the most herd-like behaviour of all investors, this to us is another confirmation of the peak of the credit cycle. It usually takes a few years for banks to forget the pain of defaults and emphasize higher income at almost any cost. The loan losses of 2002 are now ancient history to bankers now incented to increase volume through their bonus structure. Of course, their current credit models are also encouraging institutional stupidity. To paraphrase Forrest Gump; Stupid is as Stupid doesÉ when following the dictates of a quantitative model. And Stupid has never been so legitimized by mathematics and credit rating agencies.

Probability of a Market Top and Structured Stupidity

Recently, the credit rating agencies have lowered their Probability of Defaults (PD) on bond rating categories. As in the past, this probably indicates a market top. As firm believers and exploiters of the credit cycle, we think it is a necessary precondition for a credit market top to have a consensus view that credit risk is vastly decreased. This time around the credit rating agencies have made it easy for us by formalizing their benign and market top view of credit risk with their declining “PDs”.

Unfortunately, these PDs are a major input into the quantitative models which are used to rate and create CDOs. The lowering of PDs fed into these models necessitates less equity and generates higher profits from securitizing corporate bonds and bank loans. We would encourage thoughtful investors to read through some of the latest research pieces out of the Structured Products area of major investment banks. After running the new PDs through their models, Merrill Lynch wrote an acronym filled piece positively gushing about the prospects for tighter investment grade corporate spreads and the higher profits in corporate debt structured products.

The Default Probability Daisy Chain

Notably, the entire piece did not consider the underlying credit situation of issuers or the stark reality that PDs were being lowered at the peak of the credit cycle. To these quantitative analysts, PDs are just statistics to be entered into their models. How can it be imprudent to use these fancy quantitative models which give blind confidence to those who buy these sliced and diced credit bombs on credit rating alone? It’s not as if credit rating agencies make huge amounts of money rating these structures or ever change their minds quickly on companies or even whole industries. It might be a bit ironic to them that the utility industry is only now clawing its way back from a wholesale rating downgrade in 2001 after the collapse of Enron.

After reading a number of CDO and structured finance research pieces, we were left with the hollow feeling that credit doom is approaching. It also seemed a little incestuous to us that rating agencies are using their own ratings to develop the default statistics that they then use in their “proprietary” models which they sell and use to assign ratings to collateralized debt obligations. The term “daisy chain” comes to mind.

Yes, we are “old school” fundamental credit people who prefer direct credit exposure in the “underlying” bond or loan to obfuscated structures that give ratings comfort but defy analysis. We also prefer to escape the huge fees and profits buried into these transactions. Try as we may, we can’t escape the reality that our underlying cash market is being driven ever tighter by the inanity of the structured product market.

When an economic or credit event starts to increase PDs, this substantial demand for credit will quickly dry up. Given the institutional propensity for credit rating agencies to play catch up and overkill weakening credits, the damage to the corporate market could be substantial. The speculation in these products is currently extreme. Delphi automotive was reported to have $19 billion in credit default swaps outstanding against $2 billion in cash bonds. Any market that has derivatives outstanding that far in excess of the realistic economic demand for hedging has got to be volatile.

The Siren Call of M&A

Corporate bond spreads are currently narrow, reflecting the current low default rate and strong issuer fundamentals. This is usually the case at the peak of the credit cycle. The increasing merger and acquisition activity is also a telltale of a credit market peak, as improvement in internal cash flows becomes increasingly difficult and the siren song of investment bankers with M&A tales to spin becomes an attractive way to “build corporate value”.

The financial prospects of a bond issuer can change dramatically in an effort to appease shareholders whose investment horizons grow increasingly short with moderating returns. Today’s Dofasco saga is a great lesson in the perils for the unsecured bond holder in the M&A game. This fine Canadian company is caught up in the corporate machinations of Mittal’s bid for Arcelor. Since Mittal believes that Arcelor overpaid for its acquisition of Dofasco, it has promised it to Thyssen Krupp (TK) should it succeed in taking over Arcelor. A key defense strategy from Arcelor has seen Dofasco dropped into a Dutch foundation which makes it unlikely that it will be sold to TK.

The Dofasco bondholders are watching this mess unfold, with one end of their see saw an investment grade rating, as part of Arcelor, and the other end a TK downgraded below investment grade. Additional salt in the wound comes from the knowledge that the Board of Dofasco had incented management to “maximize value” with share options around the time of the new Dofasco bond issue which didn’t disclose all the expressions of affection from potential Dofasco’s corporate suitors in its public prospectus. Outraged bondholders are clamouring for their bonds to be called. We think it will take some time for the bondholders to be satisfied since Arcelor management is currently occupied with their scorched earth campaign against Mittal.

Beware Fall Spread Widening

If we’re not at a peak of the credit cycle, we are very close to it. Overwhelming enthusiasm for credit risk, tight corporate bond spreads and very speculative markets support this view. Tightening monetary policy and rising interest rates will eventually send corporate bond spreads wider. Our best guess for a sell off in credit would be the fall “spread widening” season. The fall of 2006 would be exactly eight years since the Long Term Capital credit scare in the fall of 1998 which started the last bear market in credit. It would also be four years since the bottom of the credit market in the fall of 2002.

We do not see the upside in assuming credit risk at current spread levels. We continue to upgrade quality in our portfolios and await better values.