臺大管理論叢 NTU Management Review VOL.29 NO.1

261 NTU Management Review Vol. 29 No. 1 Apr. 2019 securitization makes loan growth (especially business loans) less sensitive to cost of fund shocks. Therefore, we infer that there is a negative relationship between bank securitization and liquidity reserves. Banks with higher levels of securitization are better able to transform their loan portfolios into assets with higher liquidity, and thus need less liquidity reserves compared to those with lower levels of securitization. 2.1.3 Bank Capital The literature holds two different theories about the association between bank capital and liquidity creation. One is the “financial fragility-crowding out” effect, which predicts that bank capital has a negative effect on liquidity creation because additional capital reduces the fragility of the bank’s capital structure, so banks have less incentive to commit to monitoring, inducing lower levels of liquidity creation (Diamond and Rajan, 2000, 2001; Gorton and Winton, 2000). The other theory is the “risk absorption” hypothesis, which predicts a positive effect of bank capital on liquidity creation for the reason that capital absorbs bank risks and expands banks’ risk-bearing capacity. Hence, higher bank capital increases the capacity to take risk, which induces higher liquidity creation (e.g., Bhattacharya and Thakor, 1993; Repullo, 2004; Von Thadden, 2004; Coval and Thakor, 2005). Berger and Bouwman (2009) conducted an empirical test to investigate how bank capital affects liquidity creation. They found that the “financial fragility-crowding out” effect is stronger for small banks whereas the “risk absorption” effect is stronger for large banks. Berger and Bouwman (2009) argued that larger banks have more ability to offer large loans or engage in other off-balance sheet activities, and they are always exposed to intense monitoring by regulatory agents and market participants; therefore, higher capital works to absorb risk and expands banks’ risk-bearing activities. On the other hand, small banks operating in less segmented markets may have a significant overlap between those that invest in bank equity and those that invest in bank deposits, so the increase in capital is more likely to crowd out bank deposits. Besides, small banks are less able than large banks to absorb other liabilities such as subordinated debt. Adding these two effects together, they suggested that capital is more likely to crowd out deposits at small banks than at large banks, decreasing their ability to create liquidity. Berger and Bouwman (2013) reemphasized the importance of bank capital and argued that bank capital can enhance a bank’s survival probability and market share. Their empirical results support the claim that bank capital helps small banks avoid being crowded out by large banks, and it enhances large banks’ performance during banking crises.