Research: The Bond-Rating System Isn't Broken

Certified agencies may "miss" based on short-term news, but they still serve their purpose for large institutional investors.

April 01, 2005

| by Alice LaPlante

Is the fact that Moody’s Investors Service was still touting WorldCom debt as “investment grade” just four days before the firm went bankrupt proof that the bond-rating system is broken? Not at all, argue the authors of a new research study, who say that “certified” bond-rating agencies such as Moody’s and Standard & Poor’s serve a highly specialized institutional clientele with needs that are markedly different from those of other investors.

“Certified firms serve a purpose in the bond market, as do non-certified firms. They just happen to possess different properties,” said William Beaver, the Joan E. Horngren Professor of Accounting at Stanford GSB, one of the authors of the study. He compared the study of the bond ratings market to an analysis of the pain reliever market. “Aspirin is aspirin,” he said. “There’s a medical test that can determine just that. But do other kinds of pain relievers offer something different, something worthwhile, to a different kind of customer? That, in effect, is what we attempted to find out.”

The answer, say the researchers, is yes.

The question has enormous ramifications. In 2004, more than $1.2 trillion of debt was issued, according to Thomson First Call. Since 1973, the Securities and Exchange Commission effectively has required all such public bonds to be rated by a certified agency. At present, there are just four such agencies: Standard & Poor’s, Moody’s, Fitch Ratings, and Dominion Bond Rating Service. These companies are charged with acting as “information intermediaries” and improving the efficiency of securities markets by increasing the transparency of the securities themselves.

Those in support of the system say that the certification establishes a high standard of quality for bond ratings. Others — including members of Congress and, increasingly, representatives of the SEC — argue that the certification requirement serves as a barrier to entry for new competition. Moreover, critics of the existing system argue that because certified firms collect their fees from the companies being rated, there’s a possible conflict of interest and a weakening of incentives to produce high-quality ratings.

At issue is whether certified firms — officially called Nationally Recognized Statistical Ratings Organizations (NRSROs) — should be criticized for failing to meet the basic needs of investors. After all, there are a number of non-certified bond-rating agencies that appear to do a better job of responding to the ups and downs of the bond market in real time — and which avoid any potential conflict of interest by collecting fees from investors, not debt issuers.

To determine the answer to this, Beaver and colleague Mark Soliman, assistant professor of accounting, along with Catherine Shakespeare of the University of Michigan Ross School of Business, compared the bond ratings of Moody’s — a major certified rating agency — with those of a credible non-certified agency, Egan-Jones Ratings Co. (EJR). The goal: to see if one did a better job — or whether they fulfilled different roles for different types of investors. “To extend the aspirin analogy, are we dealing with two separate, identifiably different products, or are the certified and non-certified agencies attempting to perform the same function?” asked Beaver.

The researchers examined the two company’s ratings according to three criteria: correlation of their ratings with the stock market; correlation of their ratings with the bond market; and measuring which company led in predicting upswings and downswings in bond ratings.

The results were instructive: EJR’s ratings more closely corresponded to both stock and bond market returns and appeared to be more timely and to lead Moody’s ratings by a significant margin in reflecting positive market news. But Moody’s did a better job of reflecting negative news, of explaining non-investment grade bond yields, and of predicting bond default.

This made sense given that there are two distinctly different clienteles for bond-rating information: large institutional investors, many of whom are bound by predetermined “prudent investor rules”; and investors who are more immediately “valuation-oriented” and take a more fluid approach to buying and selling bonds.

“Large institutional investors, particularly those governed by prudent investor rules, have contracts that specify the conditions for holding or selling a bond,” said Soliman. “If a bond falls below investment grade, they must sell. The decision is taken out of their hands.” Because of the high costs associated with liquidating bond portfolios, this type of investor doesn’t want the bond rating jumping up and down with each market tremor. Rather, they want the bond rating changes to be more conservative and to reflect long-term expectations about the safety and value of the securities in question.

In the stock return test, for example, EJR appeared to change its rating at the same time that new information about a firm was released to the market. To the extent that the market reacts to new information in a timely manner, this suggested that EJR is more responsive to the needs of valuation-oriented investors who rely on its reports to buy and sell securities.

But because EJR is so quick to change, its ratings fluctuate far more than Moody’s ratings — something that can actually be costly for institutional investors. For example, EJR more frequently drops bond ratings below investment grade in the interest of perfectly reflecting market dynamics. “Institutional investors can’t handle that kind of volatility,” said Soliman.

This makes sense, given that Moody’s and other certified agencies are fulfilling a quasi-regulatory role, not just an investment advisory role, said Beaver. The companies are, in effect, offering two different products appealing to two different populations. “Moody’s ratings tend to be more conservative, which is consistent with its role of being a quasi-regulatory agency,” said Beaver. Agreed Soliman: “If all you want is for bond ratings to follow the stock market, it’s not that hard to do. There are difference forces at play here.”

Indeed, both Beaver and Soliman called criticism of Moody’s and S&P’s inability to predict the Enron and WorldCom scandals as “misfocused.”

“Over the years, Moody’s has given lots of companies a chance to ‘ride it out,’” said Soliman. “In many cases, Moody’s could argue that, had they downgraded a firm below investment grade, it could have forced premature bankruptcy on the business.” In other words, he said, it’s not unreasonable to assume that the act of downgrading the WorldCom stock had something to do with the firm’s bankruptcy declaration just four days later. “Moody’s isn’t just fulfilling an advisory role; its actions can have negative impact on companies,” said Soliman.

“It’s easy to use hindsight to say that something should have been flagged earlier, but that’s not the way at all to evaluate the overall effectiveness of a rating agency. It’s simply not representative,” said Beaver.

For media inquiries, visit the Newsroom.

Explore More