Non-Boring Thoughts On Credit RatingsDownload PDF
Bond investors that rely heavily on the ratings provided by agencies will be troubled by the Working Paper recently published by the Bank For International Settlements. Without drawing the conclusion directly, the BIS study shows clearly why those few investment managers that do their own independent and continuous research on the cash flows underlying each investment and its unique security structure can invest for the substantial benefit of their clients.
Most bond investors (and their clients) gained their professional experience in a period of massive government bond issuance and profound changes in inflation rates. Between 1965 and 1982, interest rates rose with rising inflation and ballooning government deficits, subsequently falling in sustained and dramatic fashion. In this environment, corporate debt became a mere addendum to the investment policies of most investors. On corporate securities, the potential return increment was perceived to be virtually insignificant in relation to the gains for making a correct “duration bet” in the huge government market, and no justification for the sacrifice in liquidity.
With relatively little corporate debt outstanding, a minor role in portfolio strategies and virtually no defaults, there was no payoff for the effort needed to investigate, monitor and manage corporate securities. Few investment managers made the effort.
For those few investors who included corporate debt in their portfolio mandates, constraints on sector weights controlled the commitments their managers were permitted to undertake. Typical in a pension plan policy mandate would be a set of statements such as: “The maximum permitted weight in AA securities will be 10%”. Thus, the rating assigned to each security became the controlling feature in determining not only its coupon rate, its trading yield to maturity, but also its quality in virtually every portfolio manager’s mind.
Recently bond portfolio investors and their clients have been forced to give much greater role to corporates in their portfolio strategies. Since 1988, the proportion of corporate debt in the Canadian bond universe has risen from 6% to 26% of total outstandings. Even more astonishing is the increase in the volume outstanding of securities rated BBB. During the same period, the percentage of the index represented by BBB has gone from 1% to 5.5%.
Today, the dominant role of credit ratings assumes critical importance. Few investment managers have the experience or skill set to analyze the intricacies of deal structure. Most are forced to rely heavily on the work of the credit rating agencies. Central to the bond investment problem then are questions about the quality, analytical time horizon, stability and usefulness of the credit ratings provided by the rating agencies.
A working paper published in February, 2003, by the Bank for International Settlements (the Central Bankers’ central bank) sheds some provocative light on these questions. Interested readers can find this article, Are Credit Ratings Procyclical?, at http://www.bis.org/publ/work129.pdf. Be forewarned that the analysis is dense with mathematics that will be a challenge to all but the most mathematically literate.
The authors’ conclusions are however both comprehensible and useful. No doubt unsettling to those many investors who rely on the ratings, the authors found that most ratings change infrequently and that changes are insensitive to business cycle conditions, which raised the possibility in their minds that such stability “might reflect a lack of continuous monitoring” (Page 12). The authors suspected that lack of resources at the agencies is the likely cause, but there are probably others of a more cynical mindset who might feel that the agencies would be unlikely to downgrade a credit before absolutely necessary since it is the rated company that pays the rating fee.
The authors carry on. “Even if irregular monitoring is not an issue … our results still point to an overly procyclical reaction by the agencies when rating changes are indeed made.” Three points leap out of this tight little phrase.
First, the agencies react. There is no suggestion in the data that the agencies allow any judgment about future developments to sneak into their considerations. Indeed, the authors believe that “the behavior of ratings agencies might be captured by a threshold model”, meaning that the modifications to the underlying business fundamentals must be overwhelming before they will change a rating.
Secondly, the reactions are almost invariably in the direction of the business cycle. One pernicious effect of this is to raise the cost of capital to the company at the worst point in the business cycle, intensifying the downward pressure.
Thirdly, the agencies tend to “overreact” or “overshoot”. Once the sanguine fog of trust has been breached, the very human tendency to feel betrayed tends to take over. The rating agency will naturally begin to worry about their own credibility and assess all statements and action of the company with great skepticism. (See related articles on behavioral finance.) The unfortunate consequence is that the trend of downgrades (and upgrades in the optimistic case) will often persist for longer than warranted by the fundamentals.
Investment managers can organize themselves to take advantage of the behaviors illuminated by this study for the benefit of their clients.
With most investors reliant on the ratings for their pricing information on corporates, changes in ratings almost invariably lead to substantial and immediate changes in the price of the bond. After the announcement, there will be no liquidity in the market until a new equilibrium level is established. Since the rating agencies react to the publication of the financial information in order to justify a change in their assessment, they are always acting after the fact.
This delayed reaction phenomenon means that analysts who continuously monitor financial developments within the issuing companies and who are prepared to proactively anticipate future developments will have the opportunity of placing virtually riskless bets on possible credit rating changes. If a situation begins to look like a ratings change might be coming, the proactive investment manager can buy or sell in advance of those who wait for the ratings agency to act. If no ratings change is forthcoming, no harm has been done, while if the change happens, the reward for correct positioning can be substantial.
The same analytical framework will regularly offer opportunities to act contrary to the spirit engendered by ratings changes when they go into “overshoot” mode. The BIS Working Paper saw evidence that the agencies naturally make their ratings changes in the direction of the cycle of corporate fortune, but that they also tend to change even more than is warranted by the underlying financials. Since the agencies will always be late in recognizing an upturn in the company’s cycle, another opportunity exists at this stage, although anticipating a reversal in an existing trend is not riskless. There is always the possibility that the trend will continue.
The investment management skills required to profit from these opportunities are the ability to look through the fog of financial numbers to focus on the cash flow that supports the interest payments required for specific securities and the willingness to do the detailed work to understand the structure of each deal. The clients of investment managers that use these rare skills will be the big beneficiaries.