Centre for Economic Policy Research

Rating Agencies: Conflicts of Interest in Credit Assessment and Consulting 45

---

to legal risk. Companies that pay for ratings may have additional access to rating
agency personnel to supply additional interpretation about their future prospects. 
Another aspect of the evolution of the rating industry that deserves mention is
the increased reliance of regulation on external ratings and the accompanying
move to certify certain agencies as approved for such purposes. As already noted,
external ratings had been used for supervisory purposes as early as the 1930s.
Initially, there was no formal guidance about which ratings would be acceptable
for these purposes. With the increasing use of ratings, however, and growing 
regulatory reliance, there was an obvious danger that ratings quality would vary
across agencies, and the temptation to ‘shop’ for ratings would intensify. For this
reason, the SEC introduced in 1975 the concept of the ‘Nationally Recognized
Statistical Rating Organization’ (NRSRO). At the time, the SEC did not 
contemplate its designation being used by other regulators, as in fact turned out
to be the case (Securities and Exchange Commission, 2003a).
The fact that some agencies are recognized, even though recognition is 
intended to reflect a market reality rather than convey approval, has influenced
the competitive structure of the industry. It adds to the value of a rating from a
recognized agency and further increases barriers to entry in the industry. The
greater the range of regulatory purposes to which a rating is put, the greater the
value of recognized status, and correspondingly, the harder it is to distinguish the
intrinsic value of a rating from its regulatory value.
In fact, the range of regulatory uses to which ratings are put has tended to
widen substantially in recent years. Various regulators in the United States have
enshrined the SEC’s NRSRO designation into their own procedures, and regulators
in other countries have also developed practices for the regulatory recognition of
ratings (Bank for International Settlements, 2000). A substantial further increase
will take place when the new Basel Capital Accord comes into effect (Bank for
International Settlements, 2001). The new Accord responds to criticism that the
previous Accord (Bank for International Settlements, 1988) was insufficiently 
risk-sensitive, by classifying bank assets into a larger number of credit risk classes.
For this purpose, banks following the so-called standardized approach to calculat-
ing risk weights will rely on external credit assessments, which will, in practice,
mean ratings from approved agencies.
The regulatory use of ratings may have contributed to the striking degree of
concentration in the rating industry. White (2001) notes that the number of 
rating agencies in the United States has always fluctuated between three and five,
and that other countries have even fewer. Moreover, the incumbent agencies
seem to have substantial staying power. There can be few, if any, industries in the
United States where the three remaining recognized firms in 2003 were direct 
lineal descendants of the dominating firms of 80 years earlier.
39
Concentration is partly a reflection of economies of scale and scope, and