Bank Transparency, Loan Loss Provisioning Behavior, and Risk-Shifting

1. Bank Transparency, Loan Loss Provisioning Behavior, and Risk-Shifting Robert M. Bushman & Christopher D. Williams

2. Informational Transparency: A Fundamental Lever of Bank Regulation? Unresolved issue in bank regulation: How important is informational transparency in promoting market discipline of banks by outside investors. • Basel II Capital Accord based around three complementary elements or “pillars”. • Pillar 3 recognizes that market discipline has the potential to reinforce • Minimum capital standards (Pillar 1) • The supervisory review process (Pillar 2) • “Market discipline imposes strong incentives on banks to conduct their business in a safe, sound and efficient manner, including an incentive to maintain a strong capital base as a cushion against potential future losses arising from risk exposures.”

3. What Works Best in Bank Oversight? Recent Research Findings Regulatory features that foster accurate information disclosure, empower and create incentives for private-sector oversight of banks work best to: • Promote bank development (Barth et al., 2004) • Reduce obstacles for firm raising external finance (Beck et al., 2006) • Foster sound banks as measured by Moody’s financial strength rating (Demirgüç-Kunt et al., 2006) • Lower incidence of banking crises in countries (Tadesse, 2006) • Provide incentives to reduce risk of default through larger capital buffer (Nier and Baumann, 2006)

4. What Do We Do? • Primary Research Questions: • Does a country’s financial accounting regime impact bank risk-shifting behavior? Does this impact vary with key characteristics of a country’s accounting regime? • Use a large sample of banks domiciled in 28 different countries • Capture bank risk-shifting behavior by exploiting Merton’s (1977) characterization of deposit insurance as a put option • Distinguish accounting regimes by the extent to which banks in each country use loan loss provisioning behavior to smooth accounting earnings. • Control for key elements of the bank regulatory practices adopted by each country. • Estimate realizedaccounting transparency, where much previous work uses indices based on regulated transparency on the books

5. Informational Transparency and Financial Accounting Regimes • Financial accounting systems form the foundation of the firm-specific information set available to interested parties outside the bank • A logical starting point for investigating properties of information important for addressing moral hazard problems at banks • Plausible that the quality of financial accounting information is correlated with the quality of bank disclosures falling outside of a country’s formal accounting rules. • Accounting transparency may enhance ex-ante discipline on bank risk taking activities by allowing investors in uninsured liabilities to better observe risk-taking behavior and respond to increased risks by demanding higher yields

6. Income Smoothing as a Proxy for Transparency We posit that earnings smoothing via loan loss provisioning is likely to be a first order determinant of bank transparency, particularly with respect to transparency of bank risk-taking. The loan loss provision represents the key accrual accounting choice in a bank’s income statement. With the potential to be reflective of the fundamental risk of losses inherent in underlying loan portfolios, loan loss provisioning is a key aspect of bank financial reporting for regulators and outside investors interested in monitoring risk-taking behavior. Provides a tie in to the banking literature considers role of loan provisioning behavior in mitigating the pro-cyclical effects of business cycles on banks. If forward looking provisioning behavior leads to smoother earnings, it raises the question of whether observed smoothing behavior increases the informativeness of earnings by better reflecting risks, or whether it reflects earnings management by banks attempting to obscure their risk-taking behavior

7. Merton (1977) – Deposit Guarantee is a Put Option Merton (1977) characterizes deposit insurance as a put option issued by the guarantor to the equity holders with strike price equal to the face value of the debt that is guaranteed. Merton theoretically derives a closed form pricing model for the deposit insurance put that is a non-linear function of the volatility of the bank’s assets and the bank’s leverage. • The existence of this put option creates incentives for banks to shift risk onto the guarantee agency by increasing the risk of assets without simultaneously increasing capital adequately: • Insured depositors have low incentives to monitor risk taking activities of banks • Risk-shifting occurs when banks manage to increase the risk-adjusted cost of deposit insurance that deposit insurance agencies are unable to pass onto individual banks

8. Leverage & Risk : Can These Values Be Chosen Independently? Linear approximation to Merton(1977): are treated as independent parameters: (1) The coefficients (partial derivatives) in (1) assume that leverage can be held constant as risk changes. However, with supervision and market monitoring this is not likely to hold! Outside disciplinary forces may create incentives for banks to decrease leverage in response to increases in risk. Captures intensity of disciplinary response to risk (2) where Assume in equilibrium: Leverage is a linear function of risk

9. Embedding the Equilibrium Relation Between Leverage and Risk Into the Deposit Insurance Option Linear approximation to Merton(1977): (1) Substitute in for D/V (2) where 9

10. Estimating Risk-Shifting (Duan et al., 1992) The relation between the ‘fair premium’ and risk captures the equilibrium outcome of struggle between risk-shifting banks and disciplining forces. After substitution: captures outcome of the conflict between risk-shifting incentives and disciplinary responses (3) where (+) (+)) (−) Incentives to increase risk Disciplinary Response <0 Observed Risk-Shifting Risk-shifting Incentives offset by disciplinary response

11. Empirical Investigation of Smoothing and Risk-Shifting: Basic Research Design is consistent with Smoothing increasing risk-shifting by banks.

12. Estimation of IPP (a) (b) Solve (a) and (b) for V (the implied market value of assets)and σV(the implied volatility of assets), then drop the values into the theoretical pricing mode for the value of IPP : where V and σvare both unobservable where as D, E, and σE are observed Two equation approach (Ronn and Verma(1986) w/ forbearance) solves two simultaneous equations dictated by standard option pricing theory:

13. Baseline Estimation of the Three Regression Framework

14. Empirical Estimation of Smoothing BE/Assets: Control for capital management Incentives Smoothing Loan portfolio performance Control for change in credit risk at individual bank level Macro-Economic Conditions Beginning loan loss reserve Loans/Assets Type, Year > 0 is an indication of Smoothing behavior Equation is estimated by country over the sample period

15. Pooled Estimation of Observed Smoothing Laeven and Majonni (2003)

16. Empirical Investigation of Smoothing and Risk-Shifting is consistent with Smoothing increasing risk-shifting by banks.

17. Controls for the Institutional Features Bank Regulatory Environment Supervisory power: measures of the power bank regulators have over banks Regulations on capital adequacy: stringency of capital requirements Bank regulations on private-sector monitoring of banks Explicit Deposit Insurance State ownership of banks Contracting Environment Investor Rights: investor protection rights in a country Creditor Rights: rights of creditors are given in a country Judicial Efficiency: efficiency and integrity of country’s legal system. Bank Type Effects

18. Estimated Relation Between Bank Risk-Shifting and Smoothing

19. Estimate and for each country

20. Channels Through Which Smoothing Impacts Transparency Discipline Intensity is consistent with Smoothing decreasing the intensity of the disciplinary response of leverage to an increase in risk 20

21. The Effect of Smoothing on the Sensitivity of Leverage to Risk

22. Channels Through Which Smoothing Impacts Transparency Pure Incentives to Increase IPP (partial derivative effects) => More Smoothing, higher partial derivatives, greater incentive to risk-shift Option more sensitive to changes in risk and leverage 22

23. The Effect of Income Smoothing on the Partial Derivatives

24. Other findings/Robustness Smoothing and Timeliness are negatively correlated (50%) We find similar results on both the Beta and Incentive regressions using Timeliness Simultaneity between leverage and risk w/o US and Japan; elimination of extreme countries in the Timeliness regressions Inclusion of only commercial banks Only countries w/ Explicit Deposit Insurance

25. The Relation between Timeliness on Observed Bank Risk-Shifting

26. Conclusion We document : 1) A positive (negative) relation between Smoothing (Timeliness) and observed Risk-Shifting 2) More Smoothing the less sensitive leverage is to risk 3) Smoothing increases the incentives to increase IPP Overall our evidence is consistent accounting transparency as proxied by Smoothing (and Timeliness) playing an important role in creating informational transparency which facilitates more effective bank oversight.