Business — Banking — Management — Marketing & Sales

Ratings rank the credit standing of debt issues using coded letters for the ratings from agencies. Ranks are ordinal numbers, not absolute values of the level of risk, by contrast with default probabilities whose value quantifies the likelihood of default over a given horizon. Internal ratings refer to ratings assigned by banks to their borrowers and their loans. Unlike agency ratings, which use public scales, internal ratings use proprietary scales that vary across banks. Both types of ratings serve as the foundation for the Internal Ratings-Based (IRB) approach of the New Accord and should play an increasing role for differentiating credit risk of loans.

External ratings are those of debt issues, not of issuers. The ratings assigned to senior unsecured debt are close to issuer ratings since the debt defaults only if the issuer does. Subordinated debt might default without a default of senior debt. The secured debts benefit from various guarantees, acting as a shield between the debts risk and the pure issuers credit risk. The ratings of issues of facilities capture the severity of losses, which is a combination of default probability and expected recovery.

Rating schemes use various criteria, from qualitative factors, such as strengths or weaknesses of firms, up to financials of corporate borrowers. Internal rating systems should use a rating scheme isolating the various ingredients of credit risk, default probability and recovery rates. This implies distinguishing the intrinsic rating of a borrower, the role of a supporting entity if any and the recoveries resulting from the guarantees attached to each facility. To move back and forth from internal ratings to default probabilities, as the IRB approach of regulators requires, a mapping between external ratings and internal ratings is necessary. It allows us to move from the internal to the external ratings and assign the default frequencies attached, through averages across debt issues, to external ratings. This mapping remains conventional. The alternative route, using directly models of default probabilities, is detailed in Chapter 37.

This chapter provides details about the philosophy underlying ratings and how they help in quantifying risks. The first section details the nature of external ratings. The second section describes the basic scheme underlying internal rating systems. The third section provides an overview of rating criteria. The last section explains how to move from internal ratings to default probabilities, one of the major credit risk drivers in risk models.