1. Bank Credit Ratings: What Determines their Quality? 1 Harald Hau University of Geneva and SFI http://www.haraldhau.com Sam Langfield UK FSA David Marques-Ibanez European Central Bank

2. © Harald Hau, University of Geneva and Swiss Finance Institute 2 Why look at bank ratings? Spectacular rating failures in the 2007/8 crisis expression of a general problem? Why do bank ratings matter? Cornerstone of bank regulation (ratings determine Basel credit risk weights for interbank exposure) Determine financing costs and competitive advantage (resource allocation role across and to bank sector) Specific rating problems: Opaqueness and interconnectedness Governance: First and second moment of cash flow matter Leverage implies insolvency might always be close

3. © Harald Hau, University of Geneva and Swiss Finance Institute 3 How to measure rating error? Cardinal Ratings: Translate into absolute measures of default risk Ordinal Ratings: Imply only relative assessment: banks with high ratings have relatively less default risk Rating agencies exacerbate confusion about the meaning of their ratings Dilemma of cardinal ratings: Default risk of all banks explodes in a financial crisis Financial crises are largely unpredictable

4. © Harald Hau, University of Geneva and Swiss Finance Institute 4 Expected Default Frequencies (EFDs)

5. © Harald Hau, University of Geneva and Swiss Finance Institute 5 How to measure rating error? Any cardinal rating quality measure will be heavily tainted by non-predictability of financial crisis Bank rating systems do not incorporate systemic risk measures Ordinal rating is what they produce Quality metric for ratings requires a nonparametric or ordinal approach Define Ordinary Rating Quality Shortfall (ORQS)

6. © Harald Hau, University of Geneva and Swiss Finance Institute 6 How to measure rating error? High rating quality: CR rank and EDF rank are strongly related Scattered along the 45 degree line in a CR-rank EDF rank plot Low rating quality: CR rank and EDF rank shows no correlation Uniform distribution in the CR rank – EDF rank plot

7. © Harald Hau, University of Geneva and Swiss Finance Institute 7 Bank rating data End quarter bank rating data from Moody’s, S&P and Fitch for 1990-2011 on 369 banks headquartered in the U.S and EU15; ignore subsidiary ratings Uniform rating scale across agencies Further subdivide each grade by rating outlook (if possible) Use EDF data from Moody’s (measured two years later) EDF calculations are based on the Merton model Drawing on Moody’s data spares us any parameter choices Use accounting data from Bankscope Obtain 21,131 ORQS measures, 75% fall into the 2000- 2011

8. © Harald Hau, University of Geneva and Swiss Finance Institute Credit rating rank and EDF rank Uniform distribution in the investment grade range (AAA to BBB-) Correlation only for speculation rating range (BB+ to C)

9. © Harald Hau, University of Geneva and Swiss Finance Institute Credit rating rank and EDF rank Weak correlation between rating rank and EDF rank also for investment grade range The ORQS is distance from the 45 degree line

10. © Harald Hau, University of Geneva and Swiss Finance Institute Rank correlations Investment rang grades (top and middle tier) contain no information about future EDF Basel II and III grant steep risk weight reduction Credit Assessment AAA to AA-A+ to A-BBB+ to BBB-BB+ to B-Below B-unrated Risk Weight20%50%100% 150%100%

11. © Harald Hau, University of Geneva and Swiss Finance Institute 11 Alternative measures: TORQS and DORQS Use Box-Cox Transform of ORQS to make data more normal: TORQS Use directional measure of rating quality to capture rating bias:

12. © Harald Hau, University of Geneva and Swiss Finance Institute 12 Hypotheses about Rating Quality H1: Different in Crisis and after Credit Booms? H2: Different across Agencies? H3: Do Bank Characteristics Matter? H4: Bank Fee Income Creates Rating Bias? H5: Competition Improves Rating Quality? H6: Is Public Disclosure a Remedy for Underrating?

13. © Harald Hau, University of Geneva and Swiss Finance Institute 13 H1: Rating quality in crisis and after credit booms? Ratings contain slightly more information (in an ordinal sense) during crisis and after strong credit growth (over the last 12 quarters); STD of TORQS = 0.43

14. © Harald Hau, University of Geneva and Swiss Finance Institute 14 H2: Rating quality differs across agencies? S&P ratings show less positive rating inflation

15. © Harald Hau, University of Geneva and Swiss Finance Institute 15 H3: Do bank characteristics matter for bias? Focus on large positive rating errors: Bank Size creates significantly more rating error STD log assets = 1.8; STD of DORQS = 0.42; Higher Loan share (relative to assets) correlates with less error

16. © Harald Hau, University of Geneva and Swiss Finance Institute 16 H4: Bank fee Income creates rating bias? S&P ratings inflated for banks with high fee incomes

17. © Harald Hau, University of Geneva and Swiss Finance Institute 17 H5: Competition improves rating quality? Decreasing industry concentration (lower HH index of market concentration) after 2001 decreased ”small” rating errors (log of ORQS)

18. © Harald Hau, University of Geneva and Swiss Finance Institute 18 H6: Is public disclosure a remedy for underrating? DORQS correlates negatively with Accounting Data Completeness for all quanitiles: Disclosure weakest for banks with most inflated ratings Accounting Data Completeness = percentage of 40 Bankscope data fields available for each bank

19. © Harald Hau, University of Geneva and Swiss Finance Institute 19 Policy Implications Bank credit ratings contain very little information; more in the speculative grade range, and more in crisis periods and after credit booms Basel II and III allow a strong discrimination in risk weights in the investment grade range; this regulatory privilege has no empirical justification; looks more like another example of regulatory capturing Large banks get significantly better credit ratings [“too big to fail” effect should lower EDF and create opposite effect] Competitive advantage to large banks Capital misallocation

20. © Harald Hau, University of Geneva and Swiss Finance Institute 20 Policy Implications Conflict of interest between banks and credit rating agencies likely explanation for rating bias in favour of large banks [need agency specific rating business proxy for each bank to nail this] Competition (lower HH index) correlates positively with rating quality; “small errors” diminish

21. © Harald Hau, University of Geneva and Swiss Finance Institute 21 Policy Implications Rating Agency Reform: Extending Liability (Dodd-Frank act) seems have failed (SEC withdrew proposal on ABS) Low quality of bank ratings make it impossible to create pecuniary incentives for better ratings Rating pay by user unlikely to work if buy-side has additional agency problems (Calomiris, 2011, Efing 2012) Corporate Ratings: Bloechlinger, Leippold and Maire (2012) show that better ratings can be constructed based only on public data

22. © Harald Hau, University of Geneva and Swiss Finance Institute 22 Policy Implications Bank data disclosure is inversely related to the rating bias; underrated banks report considerably more: Public data disclosure and ratings are substitutes Regulators should see it as their prime objective to enhance bank disclosure; effective way to decrease the exorbitant power of rating agencies Europe: Collusion between regulators and banks to minimize bank disclosure in mutual pursuit of non- accountability