Supervisory effectiveness, Sanctions and Bank risk-taking
Extended Abstract
1. Background and aim of the paper
This paper addresses the issue of banking supervisory effectiveness investigating the existence of a relationship between the banking sanctions imposed by the banking regulators and the consequent impact on the banks sanctioned, in terms of risk taking.
The topic is relevant in today's market crisis, in which several authors show substantial shortcomings in the effectiveness of regulation and sanctioning activity carried out by the Authority (among others, Blanchard, 2008; Caprio, Demirgucx-Kunt, and Kane, 2008).
It should be noted that the prudential supervision ensures compliance with the principles of sound and prudent management as well as fairness and transparency of the banks; in this context, regulatory sanctions promote the objectives pursued by the rules of prudential supervision and regulatory information. In the same light, the Pillar II (Basel 2) emphasizes the role of on-site inspections highlighting how they are able to: identify any weaknesses or inefficiencies in banking managerial processes, administration and corporate control, verify the effectiveness and the quality of internal controls, ensure that communications provided by the intermediaries for the purpose of supervision is accurate and reliable (Basel Committee on Banking supervision, 2006). Therefore, the on-site inspections and sanctions are essential tools used by supervisory authorities to ensure the stability of the financial system (Quintyn and Taylor, 2002).
Specifically, the research focuses on a sample of Italian banks in the period 2005-2011 and captures the effect of the rules on penalties on banks in terms of risk taking.
Previous studies that investigate the impact of sanctions on the behavior of banks are limited and use measures of the effectiveness of prudential regulation approximate and indirect.
For example, the International Monetary Fund (2009) has estimated, on the basis of expert opinion, the degree of compliance with the Basel Committee's Core Principles of Banking Supervision of the banking regulation in order to assess their effectiveness in terms of supervision of banks; the results of this analysis could therefore be affected by the discretion of the panelists.
Further empirical studies estimate, however, the quality of the supervisory regulations using some proxy. For example, La Porta et al. (2006) consider the degree of independence and authority of supervisors in the exercise of its control over intermediaries; Noy (2004), however, considers the degree of corruption and political freedom of the government authorities. These studies, however, focus attention on the text of legislation, rather than on the reality of the markets. Other authors (Barth, Caprio, and Levine, 2008) evaluate the effectiveness of the supervisory regulations through surveys and questionnaires to the same authority, leading to questionable results and probably affected by the opinion of those interviewed.
Based on our knowledge, there are no empirical studies that compare the impact of the types of sanctions and risk exposure of sanctioned banks.
Although thispaper is inspired by the study of Staikouras and Delis (2011), it differs from the same in many aspectsrelated to the nature of the relationship between the supervisory activities and the level of bank risk taking. In detail, the level of bank risk-taking is measured by a complete dataset.The risk measure used in Staikouras and Delis (2011) refers to the Z-score, which measure the overall risk of the bank, or the indicator NPLs / Loans that measure the credit risk only. On the contrary, this paper uses detailed risk indicators that can capture both the complex and exposure of the bank and the various possible impacts of the sanctions imposed. We use, therefore, data on credit risk, market and operational risks on the basis of quantification of regulatory capital,and others ratio suitable to capture the individual type of the bank's risk. Another element of originality of the work concerns the depth of the dataset penalties; we clusteredsanction on the type of infringement found by the on-site inspections (deficiencies in the organization, internal controls and management of credit; deficiencies in lending management and control of credit; impaired loans and provisions for losses not reported to the Supervisory Authority; lack of transparency; infringement of the rules on risk concentration) together with the personsanctioned (Director, Auditor, CEO) and the amount of the fine (in terms of amount). Combining both datasets, we intend to examine whether, in the year n, the bank showed a reduction in the level of risk taking as a result of a sanction received during an on-site inspection made in year n-1.
The study of the Italian case appears worthy of interest in the literature on banks, at least for the following three reasons: 1) during the financial crisis it has not received financial assistance of the State, unlike other European countries, so the banking risk indicators are not mitigated by contributions from third parties, other than those of supervisors; 2) the dataset of accounting data used in the analysis is extremely deep and allows the construction of risk indicators that have not been investigated by the previous studies; 3) the dataset of sanctions is unique and consistent throughout the period of the analysis.
The results of the analysis may be of interest to regulators, supervisors and professionals, interested in improving the effectiveness of the penalty rules in defense of the sound and prudent management of banks, also with a view to reducing the cost of regulation to the financial intermediaries.
In this preliminary version, the rest of the paper is organized as follows. The par. 2 describes the sample and the analysis methodology. The par. 3 illustrates the expected results and the some implications.
2. Sample and Empirical model
The sample consists of data extracted from three different sources.
The first is represented by individual bank balance sheet data for the years 2005-2011, collected by ABI Banking Data, which are 4,675 observations as a whole. The average coverage of the sample, in terms of total assets than in the Italian banking system, is over 80%.
The second type of data is represented by the sanctions, which are derived from the census of 84 bulletins (issued monthly) by the Bank of Italy in the years 2005-2011. During the seven years of analysis are detected 3,588 persons sanctioned in banks which are subject to a penalty (they are distinguished in: Board of directors, Directors, Chief Executive Officers). In this regard, it should be remembered that the sanctions are imposed, usually (though not exclusively) by inspections, asupervisory banking tool that provides to collect information and data on the results of intermediaries (Bank of Italy, 2013).
The third type of data is that of the regional GDP, taken from ISTAT.
The empirical analysis is based on multivariate regression model(time-series and cross-section). The following equation illustrates the relationship between banking supervision and bank risks:
Yit = β0 + β1SanctionCRit-1 + β2SanctionMRit-1+ β3SanctionORit-1 + β4Boardit-1 + β5Auditit-1 + β6CEOit-1+β7SAmmountit-1 + β8GDPit-1+ β9TAt + β10ROAt +εit + ui
wherei identifies the individual bank belonging to the sample (i = 1, 2, 3,..., 4,675); Yit is the nth-bank’s measure risk; t expresses the time variable (t = 2005, 2007 and 2011); β1,β2,… β10are the parameters to be estimated. Also in the model are indicated the constant (β0) and the error terms (εit; ut).
The three variables, Sanction CR, Sanction MR andSanctionOR, are dummies that take the value 1 when it was imposed a sanction related to Credit Risk, Operational Risk and Market Risk or 0 otherwise. Variables Board, Audit and CEO are also dummies taking value 1 when the sanction was imposed to the Board, the Audit (CollegioSindacale) and the CEO and 0 otherwise. The variable SAmmount is equal to the natural logarithm of the total amount of the penalty. In the model are also included some control variables related to the bank's characteristics (Total assets, TA and Return on Asset, Roa), in view of the fact that the imposition of sanctions by the Supervisory Authority will be assessed taking into account the size of the company, in accordance with the principle of proportionality (Bank of Italy, 1999 and 2012), and finally, the model takes into account the economic scenario (GDP).
3. Expected Result and Implications
The results of the empirical analysis could confirm the existence of an inverse relationship between the supervisory sanctions imposed on banks and the consequences of risk-taking intermediaries themselves in the year following the penalty. Thus, the results could confirm the effectiveness of the penalty rules of banking supervision.
The results of the analysis are relevant from the point of view of regulators, supervisors and practitioners. With regard to the regulators and supervisors, the analysis could detect any possible improvements of the system of penalties in terms of, for example, the nature and frequency of inspection work on intermediaries that may reduce the cost of regulation at the level of the overall system. With reference to the practitioners, we believe that the analysis could be of interest for the banks where highlights the extent and nature of the sanctions and, in this way, finds that failings or deficiencies at all levels of the organization that generate the enforcement actions by the supervisory authority.
Ultimately, the paper aims to contribute to improving the effectiveness of supervisory regulations on the banking system, capable of detecting and punishing behaviors related to banking "excessive" risk-taking at the expense of the stability of the system as a whole.
References
Banca d'Italia (1999), Circolare del 21 aprile 1999, n. 229, e successivi aggiornamenti (Istruzioni di vigilanza per le banche), limitatamente al Titolo VIII, Capitolo 1.
Banca d'Italia (2003), Provvedimento della Banca d’Italia del 3 settembre 2003, Procedure per l’applicazione delle sanzioni amministrative agli intermediari non bancari.
Banca d'Italia (2013), Le sanzioni amministrative della Banca d’Italia: principi generali, procedura e tempi, Roma, Febbraio.
Banks for International Settlement (2006), Basel II: International Convergence of Capital Measurement and Capital Standards: A Revised Framework, March.
Barth J., Caprio G., Levine, R. (2004) Bank regulation and supervision: What works best?, Journal of Financial Intermediation, Vol. 13, pp.205–248.
Barth J., Caprio G., Levine, R. (2008) Bank regulations are changing: for better or worse?, Washington DC, USA: World Bank Policy Research Working Paper 4646.
Barth J., Caprio, G., Levine, R. (2001) The regulation and supervision of banks around the world: A new database. Washington DC, USA: World Bank Policy Research Paper 2588.
Basel Committee on Banking Supervision. (2002). The Relationship Between Banking Supervisors And Banks’ External Auditors, Basel, Switzerland.
Berger A. N., Davies S.M. (1998), The Information Content of Bank Examinations, Journal of Financial Services Research, Vol.14, 2, pp. 117-144.
Delis M.D., Staikouras P.K. (2011), Supervisory Effectiveness and Bank Risk, Review of Finance, doi: 10.1093/rof/rfq035.
International Monetary Fund (IMF) (2009), Lessons of the Financial Crisis for Future Regulation of Financial Institutions and Markets and for Liquidity Management, International Monetary Fund: Washington, DC.
La Porta R., Lopez-de-Silanes F., Shleifer A., Vishny, W. (2006), What works in securities laws?, Journal of Finance, Vol.61, pp.1–32.
Noy I. (2004), Financial liberalization, prudential supervision, and the onset of banking crises, Emerging Markets Review, Vol. 5, pp.341–359.
Quintyn M., Taylor M. (2002), Regulatory and supervisory independence and financial stability, International Monetary Fund Working Paper No. 02/46, International Monetary Fund.
1