Do Rating Agencies Behave Defensively for Higher Risk Issuers?

We examine whether rating agencies act defensively toward issuers with a higher likelihood of default. We find that agencies’ qualitative soft rating adjustments are more accurate as issuers’ default risk grows, as evidenced by the adjustments leading to lower type I and type II error rates and bett...

Full description

Saved in:
Bibliographic Details
Published in:Management science 2023-08, Vol.69 (8), p.4864-4887
Main Authors: Bonsall, Samuel B., Koharki, Kevin, Kraft, Pepa, Muller, Karl A., Sikochi, Anywhere
Format: Article
Language:English
Subjects:
Adjustment
credit rating agencies
Default
Ratings & rankings
Risk assessment
Risk factors
soft rating adjustments
Stock prices
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We examine whether rating agencies act defensively toward issuers with a higher likelihood of default. We find that agencies’ qualitative soft rating adjustments are more accurate as issuers’ default risk grows, as evidenced by the adjustments leading to lower type I and type II error rates and better prediction of default and default recovery losses. We also find that soft adjustments’ relevance increases with issuers’ default risk, as evidenced by the adjustments being more predictive of initial offering yields and leading to a greater market reaction to rating changes. Further, we find that the rating agencies assign better educated and more experienced analysts to higher-risk issuers, providing evidence of one mechanism used by the rating agencies to generate more accurate and relevant soft adjustments. Overall, our study suggests that as the likelihood of issuer default grows, the threat of reputational harm from discovered rating failures increasingly mitigates the rating agencies’ strategic behavior incentivized by the issuer-pay model. This paper was accepted by Brian Bushee, accounting. Funding: P. Kraft and K. A. Muller gratefully acknowledge funding received from the HEC Paris Foundation and Labex Ecodec [Grant ANR-11-LABX-0047] and the Poole Faculty Fellowship, respectively. Supplemental Material: Data are available at https://doi.org/10.1287/mnsc.2022.4537 .
ISSN:0025-1909
1526-5501
DOI:10.1287/mnsc.2022.4537