1. Faculty of Economics, Chair of Macroeconomics A Structural Approach to Financial Stability: On the Beneficial Role of Regulatory Governance Benjamin Mohr and Helmut Wagner Ljubljana, May 2012

2. Faculty of Economics, Chair of Macroeconomics Helmut Wagner 2 What are the effects of regulatory governance on financial stability? Introduction  Insufficient banking regulation, government intervention in regulatory process, and connected lending play central roles in explanation of banking crises during last few decades  Evidence points to governance failures as the key contributing factor in recent global financial crisis (Buiter, 2008; Igan et al., 2009; Levine, 2010; Mian et al., 2010)  Prior to recent financial crisis, many regulatory authorities lacked mandate, sufficient resources, and independence to effectively contain systemic risk and to implement early action

3. Faculty of Economics, Chair of Macroeconomics Helmut Wagner 3 Introduction  While case for central bank independence is well established, discussion of independence in the sphere of bank regulation is relatively new, but gaining momentum  Independence and accountability are seen as key building blocks of regulatory governance  Need of independence and accountability of regulatory authorities has been recognized by Basel Committee on Banking Supervision (1997; 2006)  Independence from government and financial industry is essential for achieving and safeguarding financial stability (time-inconsistent government policy, industry capture)

4. Faculty of Economics, Chair of Macroeconomics Helmut Wagner 4 Motivation What are the effects of regulatory governance on financial stability?  Not entirely clear what constitutes good regulatory framework that promotes bank development, efficiency, and stability  Any construction of index that seeks to measure regulatory governance arrangements relies on judgment to some degree  No widely accepted measure, quantification, or time series for measuring financial stability  Recent studies are not sufficiently focused on the macro- prudential dimension of financial stability

5. Faculty of Economics, Chair of Macroeconomics Helmut Wagner 5 Previous findings  Empirical evidence regarding impact of regulatory governance on financial stability is rather inconclusive  two camps  [1] Beck et al. (2003), Das et al. (2004), Ponce (2009): find a positive relationship between regulatory governance and financial stability  [2] Opposing strand does not find that financial stability is associated with regulatory governance (Barth et al., 2004; 2006; Demirgüc-Kunt and Detragiache, 2010)

6. Faculty of Economics, Chair of Macroeconomics Helmut Wagner 6 Previous findings [1] Beck et al. (2003)  Address concept of financial stability by asking to what degree firms face obstacles in obtaining external finance  Higher degree of regulatory independence seems to reduce likelihood that politicians or financial industry will capture regulatory agency Das et al. (2004)  Proxy banking sector stability by index consisting of a weighted average of CAR and NPLs  Results suggest that regulatory governance has positive impact on banking sector stability

7. Faculty of Economics, Chair of Macroeconomics Helmut Wagner 7 Previous findings [1] Ponce (2009)  Uses ratio of non-performing loans to proxy financial stability  Regulatory independence significantly reduces average probability of banks defaulting on loans, legal protection and accountability seem to be of even greater importance

8. Faculty of Economics, Chair of Macroeconomics Helmut Wagner 8 Previous findings [2] Barth et al. (2004; 2006)  Study impact of regulatory practices on development and efficiency of the banking sector, and on the occurrence of banking crises  Supervisory independence is not related to bank development, efficiency, or stability Demirgüc-Kunt and Detragiache (2010)  Use Z-Score as a proxy for soundness of banking sector  No relationship between bank soundness and BCP compliance (used as proxy for regulatory governance arrangements)

9. Faculty of Economics, Chair of Macroeconomics Helmut Wagner 9 Our approach: Structural equation modeling  Structural equation model (SEM) describes statistical relationships between latent (unobservable) variables and manifest (directly observable) variables  We model financial stability and the governance of regulatory authorities as latent variables  test whether data patterns can be fitted within data sample  provide cross-country evidence of relationship between regulatory governance and financial stability

10. Faculty of Economics, Chair of Macroeconomics Helmut Wagner 10 Our approach: Structural equation modeling  A SEM is often used to estimate size and development of shadow economy (e.g., Bajada and Schneider, 2005)  Other studies treat corruption as latent variable directly related to its underlying causes (e.g., Dreher et al., 2007)  Recent work of Rose and Spiegel (2009; 2010; 2011) uses structural equation modeling approach, treating the recent financial crisis as a latent variable

11. Faculty of Economics, Chair of Macroeconomics Helmut Wagner 11 Advantages of our approach  Structural equation models can provide information about relationship between variables that have observable causes and effects but cannot themselves be directly measured (or are difficult to measure)  Fit measures can provide summary evaluation of complex models that involve a large number of linear equations, so that no separate “mini-tests” of model components conducted on an equation-by-equation basis  Methodology allows for a number of indicators that reflect different dimensions of multidimensional variables, enabling a better estimation

12. Faculty of Economics, Chair of Macroeconomics Helmut Wagner 12 What are the effects of regulatory governance on financial stability? Financial stability  No consensus on what best describes the state of financial stability, neither theoretical nor conceptual  We take a systemic view, emphasizing resilience of financial system as a whole to financial or real shocks and its ability to facilitate and support the efficient functioning and performance of the economy  To proxy financial stability, we use the IMF’s financial soundness indicators (FSIs)

13. Faculty of Economics, Chair of Macroeconomics Helmut Wagner 13 IMF’s financial soundness indicators (FSIs)  Have a high degree of international comparability  Measure the soundness of financial system as a whole  Include 5 variables from the core set: regulatory capital to risk- weighted assets, bank provisions to non-performing loans, return on assets, return on equity, non-performing loans to total loans  Additionally, we include bank capital to assets ratio from the encouraged set

14. Faculty of Economics, Chair of Macroeconomics Helmut Wagner 14 Determinants of financial stability We use  Regulatory governance  Structure of banking sector  Macroeconomic conditions  Economic freedom

15. Faculty of Economics, Chair of Macroeconomics Helmut Wagner 15 Determinants of financial stability  Regulatory governance  We build indices to proxy independence and accountability of regulatory agency by using data from Barth et al. (2006)  We take two indices from Masciandaro et al. (2008), and  Indicator variables reflecting degree of central bank independence (Arnone et al., 2009)

16. Faculty of Economics, Chair of Macroeconomics Helmut Wagner 16 Determinants of financial stability (2) Structure of banking sector comprises openness, competitiveness, and ownership structure of the banking sector As indicator variables we consider  Degree of consolidation in banking system  Proxy for foreign share of banking sector assets and degree of foreign bank entry  Share of bank deposits held in privately owned banks  Degree to which banks are allowed to engage in securities, insurance, and real estate markets

17. Faculty of Economics, Chair of Macroeconomics Helmut Wagner 17 Determinants of financial stability (3) Macroeconomic conditions Here several macro variables are considered in the relevant literature:  Rate of inflation, real interest rate, fiscal balance  GDP growth and GDP growth volatility  Credit and money growth  Deposit rate and deposit rate volatility  Financial openness

18. Faculty of Economics, Chair of Macroeconomics Helmut Wagner 18 Determinants of financial stability (4) Economic freedom Here institutions are regarded as consistent with economic freedom when personal choice, voluntary exchange, freedom to compete, and protection of persons and property are promoted We use  World Bank’s Good Governance Indicators  Indicator variables from the Database of Political Institutions  Democracy measures taken from Polity IV

19. Faculty of Economics, Chair of Macroeconomics Helmut Wagner 19 Empirical Strategy  Sets of manifest variables are used to capture hypothetical, difficult-to-measure constructs  Latent variables are interpreted as hypothetical constructs (are the “true” variables underlying the measurable indicator variables)  Structural equation model consists of two parts: the structural model and the (exogenous/endogenous) measurement models  Aim of procedure is to obtain values for parameters that produce an estimate for the models’ covariance matrix that will fit the sample covariance matrix of the indicator variables

20. Faculty of Economics, Chair of Macroeconomics Helmut Wagner 20 Empirical Strategy  Wide variety of indicator variables!  Testing a range of model specifications, starting from the most general specification and omitting variables, applying an iterative procedure  Choice of variables based on several criteria: statistical significance of the estimated parameters, parsimony of the model, goodness-of-fit measures

21. Faculty of Economics, Chair of Macroeconomics Helmut Wagner 21 Empirical Strategy  Goodness-of-fit measures considered: Ratio of chi-square to degrees of freedom Comparative Fit Index Root Mean Square Error of Approximation  Fit measures based on statistical information theory: Akaike Information Criterion Browne-Cudeck Criterion Bayes Information Criterion

22. Faculty of Economics, Chair of Macroeconomics Helmut Wagner 22 Model Structural model Latent variables

23. Faculty of Economics, Chair of Macroeconomics Helmut Wagner 23 Model Exogenous measurement model Indicator variables

24. Faculty of Economics, Chair of Macroeconomics Helmut Wagner 24 Model Endogenous measurement model Indicator variables

25. Faculty of Economics, Chair of Macroeconomics Helmut Wagner 25 Results (1)  Positive relationship between regulatory governance and variables indicating the degree of independence and accountability of regulatory authorities, as well as indices that proxy the strength of external audits  Positively associated with political independence of central banks; points to a more active role for central banks in banking regulation process  Regulatory governance has a positive influence on the stability of the banking sector  Relationship is robust and positive throughout all model specifications

26. Faculty of Economics, Chair of Macroeconomics Helmut Wagner 26 Results (2)  Positive relationship between structure of the banking sector and bank concentration  Private ownership of banks and foreign bank competition are positively related to banking sector infrastructure  Restrictiveness of bank activities enters with a negative sign  Positive relationship between the structure of the banking sector and banking sector stability  More open and less restricted banking systems tend to increase safety and soundness of banking system

27. Faculty of Economics, Chair of Macroeconomics Helmut Wagner 27 Results (3)  Negative relationship between macroeconomic conditions and our inflation indicator, the real interest rate, deposit rate and GDP growth volatility  Signs of coefficients for credit and money growth are negative  Financial openness enters positively  Macroeconomic conditions have a negative influence on banking sector stability  Attributed to the fact that indicator variables considered mainly represent symptoms of an unstable and adverse macroeconomic environment

28. Faculty of Economics, Chair of Macroeconomics Helmut Wagner 28 Results (4)  World Governance Indicators (control of corruption, rule of law, …) enter with positive signs  More democratic and parliamentary systems indicate greater economic freedom  Increasing government size is negatively related to economic freedom  Economic freedom seems to have a negative effect on the stability of the banking sector  Greater freedoms might imply that banks engage in activities that carry higher risks

29. Faculty of Economics, Chair of Macroeconomics Helmut Wagner 29 Conclusions  Regulatory governance contributes to sound banking sector  Performance of bank regulation could be improved by providing regulatory authorities with sufficient degree of independence and accountability so that financial stability mandate can be effectively fulfilled  More open and less restricted banking sector is associated with increased soundness of the banking system  Macroeconomic disturbances are negatively related to banking sector stability

30. Faculty of Economics, Chair of Macroeconomics Helmut Wagner 30 Conclusions  Economic freedom seems to have a negative effect on stability of the banking sector  Greater economic freedoms might imply that banks engage in activities that carry higher risks  Institutional environment may induce greater risk-taking and distort incentive structure in the banking sector so that banking stability could be undermined

31. Faculty of Economics, Chair of Macroeconomics Helmut Wagner 31 Policy implications  Policymakers should provide high degree of independence to regulatory authorities  Regulatory authorities must be accountable to executive and legislative branches of government and to the financial industry  Anyway, regulatory governance is in critical need of improvement (IMF: Vinals et al., 2010)

32. Faculty of Economics, Chair of Macroeconomics Helmut Wagner 32 Thank you for your attention ! Paper can be downloaded from my Webpage: www.fernuni-hagen.de/HWagner