Liquidity Creation and Bank Capital

See Basel Committee on Banking Supervision (BCBS) (2010, 2013, 2014) for more details on the changes to capital requirements and the definitions of the liquidity ratios as well as the implementation timetable.

Total regulatory capital now comprises the sum of the following elements: (i) Tier 1 capital (going-concern capital), that includes common equity Tier 1 (CET1) and additional Tier 1, and (2) Tier 2 capital (gone-concern capital). The CET1 capital must be at least 4.5% of risk-weighted assets (RWA) whereas Tier 1 capital must be at least 6% of RWA. Total capital (Tier 1 capital plus Tier 2 capital) must be at least 8.0% of RWA at all times. Moreover, Basel III establishes a capital conservation buffer comprised of CET1 that implies that banks, at least in normal times, should operate with a “minimum” capital of 10.5% of their total RWA. In addition, national authorities may require a countercyclical buffer. Finally, the Committee also agreed to introduce a simple, transparent, non-risk based leverage ratio, which is calibrated to act as a credible supplementary measure to the risk-based capital requirements.

The Eurozone is a geographic and economic region which consists of all the European Union (EU) countries that have adopted the euro as their national currency. As per 2018, it is a monetary union of 19 of the 28 European Union member states. The original 11 eurozone countries are: Austria, Belgium, Finland, France, Germany, Ireland, Italy, Luxembourg, the Netherlands, Portugal and Spain. Greece joined in 2001. Since then, Cyprus (2008), Estonia (2011); Latvia (2014); Lithuania (2015); Malta (2008); Slovakia (2009) and Slovenia (2007) also joined the monetary union.

See Directive 2013/36/EU of the European Parliament and of the Council of 26 June 2013 on access to the activities of credit institutions and prudential supervision of credit institutions and investment firms.

Note that a deposit insurance system can limit the negative effect of capital on liquidity creation, as suggested by Diamond and Rajan (2000, 2001).

We restrict the analysis to countries that adopted the euro during the sample period (firm-year observations are included only if the country is a Eurozone member in that specific year). We thus exclude the cases of Latvia (joined in 2014) and Lithuania (joined in 2015).

If unconsolidated financial statements were unavailable (this is sometimes the case for small-size banks, which only report on a consolidated basis as the number of subsidiaries is small to warrant separate audited statements), then consolidated data were used to avoid dropping the banks from the sample.

To identify potential outliers, we use both the leverage statistics (Belsley et al. 1980) and the Cook’s distance analysis (Cook 1977, 1979).

We check endogeneity with the Durbin score and the Wu-Hausman tests. Both tests have as a null hypothesis that the variable under consideration can be used as exogenous.

Although Bankscope does not give the required level of detail necessary to calculate the net stable funding ratio according to the BCBS (2014) guidelines for most of the banks in our sample, we can use this database to consistently approximate this ratio. Nevertheless, some assumptions are needed (Chiaramonte and Casu 2017).

Unfortunately, empirically testing the effect of off-balance sheet activities is hampered by the absence of detailed data in Bankscope. Although Bankscope gives information about off-balance sheet activities for some of the banks in our sample, considering only those banks would excessively reduce the sample size.

Liquidity creation is positively associated with liquidity risk and may be positively associated with credit risk if the high liquidity creation is caused by high business loans and commitments (Berger and Bouwman 2016).

Nevertheless, a recent body of literature suggests the possibility that the relationship between liquidity risk and credit risk in banks might be negative (see e.g., Wagner 2007; Gatev et al. 2009; Acharya and Naqvi 2012).

In response to the 2007 financial crisis, the BCBS imposes global systemically important banks (G-SIBs) additional loss absorbency requirements, which range from 1 to 2.5% CET1, depending on a bank’s systemic importance.

The capital stringency index is built by adding two measures of capital stringency: overall and initial capital stringency. Overall capital stringency indicates whether risk elements and value losses are considered while calculating the regulatory capital. It is based on the following questions: (i) Is the minimum capital–asset ratio requirement risk weighted in line with the Basel guidelines? (ii) Does the minimum ratio vary as a function of credit risk? (iii) Does the minimum ratio vary as a function of market risk? (iv) Are market values of loan losses not realized in accounting books deducted from capital? (v) Are unrealized losses in securities portfolios deducted from capital? (vi) Are unrealized foreign exchange losses deducted from capital? (vii) What fraction of revaluation gains is allowed as part of capital? Initial capital stringency refers to whether certain funds may be used to initially capitalize a bank and whether they are officially verified. It is based on the following questions: (viii) Are the sources of funds to be used as capital verified by the regulatory or supervisory authorities? (ix) Can the initial disbursement or subsequent injections of capital be performed with assets other than cash or government securities? (x) Can the initial disbursement of capital be performed with borrowed funds? We assign a value of 1 if the answer to questions (i), (ii), (iii), (iv), (v), (vi) and (viii) is yes, and 0 otherwise, while the opposite holds in the case of questions (ix) and (x). In addition, we assign a value of 1 if the fraction of revaluation gains that is allowed to count as regulatory capital (question (vii)) is less than 0.75. Otherwise, we assign a value of 0. By adding all these values together, we create the variable capital stringency index, which ranges in value from 0 to 10, with higher values indicating greater stringency.

This database is based on four surveys conducted by the World Bank. Survey I was released in 2001, and, for most of the countries, the information corresponds to 1999. Survey II describes the regulatory situation at the end of 2002. Survey III describes the regulatory environment in 2005–2006. Finally, Survey IV provides information on bank regulation and supervision in 125 countries for 2011 (with some corrections in 2012) (Barth et al. 2013). This database is available from the World Bank website at http://​go.​worldbank.​org/​SNUSW978P0.

We find similar results when Total Capital and Tier 1 are used as the measure of bank capital.

Berger and Bouwman (2016) state that there is little research to date on this topic, suggesting that examining this relationship would be a promising line of future research.

For details about the correspondence of the “cat non-fat” and “mat non-fat” metrics of Berger and Bouwman (2009) with our liquidity creation indicators, LC (cat non-fat) and LC (mat non-fat), see Appendix Table 17.

Unfortunately, the number of observations is drastically reduced from 7275 to 2396 when LC (cat non-fat) is used as the liquidity creation proxy. This reduction is mainly because Bankscope does not provide detailed date about loan categorization (e.g., corporate and commercial loans) for all the banks in our sample. Similarly, when we use LC (mat non-fat) as the liquidity creation indicator, the number of observations drops to 2176. Now, this decrease in the number of observations is mainly due to the absence of exhaustive data on loan maturity (e.g., loans and advances to costumers less than or equal to 12 months).

The close monitoring highlighted by Diamond and Rajan (2000, 2001) may be more important in smaller banks, which deal more with entrepreneurial-type small businesses (Distinguin et al. 2013). Additionally, the “crowding out” effect is more likely to affect smaller banks as they tend to be more funded by deposits than larger banks.

Concentration is calculated as the sum of the squares of all credit institutions’ market shares within a country in terms of total assets (in percentage). It is often said that a market is highly concentrated when the index exceeds 1800 (or 0.18, if we use units instead of percentages) and is unconcentrated when the index is below 1000 (or 0.1). Therefore, a higher value for Concentration would correspond to a lower market competition. The data on Concentration in the euro area countries were obtained from the Banking Structural Financial Indicators database of the ECB at http://​sdw.​ecb.​europa.​eu/​browse.​do?​node=​bbn2869.

We added to the original sample (17 Eurozone countries over the period 1999–2013) observations from Bulgaria (65), Croatia (117), the Czech Republic (65), Denmark (92), Hungary (43), Latvia (9), Lithuania (14), Norway (222), Poland (80), Romania (66), Sweden (32), Switzerland (685) and the United Kingdom (204). The extended sample includes banks from 30 EU, EEA and EFTA countries.