Quality of Bank Capital, Competition, and Risk-Taking: Some International Evidence

ABSTRACT

This study empirically investigates how bank capital and competitive conditions affect bank risk-taking. Using financial data of 7620 banks on 118 countries from 2001 to 2016, we show that banks with a higher Tier 1 ratio and a lower Tier 2 ratio are lower risk-takers. A bank with greater market power in a banking system tends to reduce its risk-taking activities. Our findings also highlight that the negative relationship between Tier 1 ratio and bank risk are more pronounced in more competitive conditions. During the financial crisis, Tier 1 capital acted as a stable funding source and reduced bank risk, but the evidence on Tier 2 capital shows that a higher Tier 2 ratio results in a higher level of risk and increases bank instability.

KEYWORDS:

• Tier 1 ratio
• Tier 2 ratio
• Lerner index
• risk-taking

JEL:

• G21
• G28

Acknowledgments

The author would like to thank the editor Ali Kutan and two anonymous referees for their valuable comments and suggestions.

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.

Notes

1. KPMG, Impact of Basel 4 on EU banks, 2018. https://home.kpmg.com/xx/en/home/insights/2018/10/impact-of-basel-4-on-eu-banks-fs.html.

2. In the bank regulation and supervision information provided by Barth, Caprio, and Levine (2013), however, the surveys were conducted in 1999, 2002, 2006, and 2011. Following the study of Anginer, Demirguc-Kunt, and Zhu (2014a) and Anginer, Demirgüç-Kunt, and Mare (2018), we employ the data from previous surveys until the newer survey data became available for matching the bank regulation and supervision variables with bank-specific variables and country control variables, that is, the data from the surveys conducted in 1999, 2002, 2006, and 2011 for the periods from 1998 to 2001, 2002 to 2005, 2006 to 2010, and 2011 to 2016, respectively. For a detailed description of the bank regulation and supervision information, see Barth, Caprio, and Levine (2013).

3. Based on earlier studies, we define the output price $P_{it}$ as the ratio of total revenue to total assets; see also Angelini and Cetorelli (2003), Koetter, Kolari, and Spierdijk (2012), Fu, Lin, and Molyneux (2014) and Li (2019).

4. Bikker, Shaffer, and Spierdijk (2012) indicate that a scaled revenue function leads to a significant upward bias and incorrectly measures the degree of competition; therefore, in our analysis, we employ the unscaled revenue equation to reduce the estimation bias.

5. For detail information of feasible generalized least square (FGLS) regression, see Wooldridge (2010) and Bikker, Shaffer, and Spierdijk (2012).

6. See Baum, Schaffer, and Stillman (2007) and Roodman (2009) for detail discussion of system GMM.

7. Gropp et al. (2018) confirm that banks increase their capital ratios by reducing their risk weighted assets, not by raising their levels of equity, and this finding is consistent with debt overhang. Berger, Öztekin, and Roman (2019) find that charter value is the most important factor in explaining target capital results, and also help explain the speed of capital adjustment results.

8. For brevity, we only report the results of FGLS estimations in Tables 6 and 7. The results for the GMM estimations are consistent with those findings and are available upon request.

9. For brevity, we only report the results under FGLS, the results for the GMM estimations are largely similar and are available upon request.

10. The bank-level control variables, country control variables, and bank regulatory variables are included in the regression but not report for brevity.

11. The timeline for the adoption of Basel III regulatory framework by Basel Committee members can be find in the progress report progress report on adoption of the Basel regulatory framework provided by Basel Committee on Banking Supervision. See also https://www.bis.org/bcbs/implementation/rcap_reports.htm.

12. The results of the interaction terms for Crisis dummy are dropped because of collinearity.

Additional information

Funding

This work was supported by the National Natural Science Foundation of China [71801040]; Key Project of Philosophy and Social Science Research in Colleges and Universities of Jiangsu Province [2019SJZDA024]; Fundamental Research Funds for the Central Universities [2242021S30013].