earticle

영어

The existence of economies of scale implies that banks will be more profitable and less risky as they get larger in size. In the context of modern portfolio theory, one can reasonably expect that banks with large and diversified portfolios can save costs and afford to put less capital at risk. Extant literature, however, lacks definitive evidence demonstrating that economies of scale exist in the banking industry. In this paper, we empirically re-examine whether economies of scale exist in the banking industry. We add more risk measures to the z-score which was used commonly in the literature. Our measures of bank risks include capital adequacy, asset soundness, and liquidity. We also analyze the effects of relative bank size and absolute bank size. To incorporate each bank’s systemic risk into our analysis, we utilize global data on banks, and divide our sample banks into five groups. We sort each group by their asset size relative to the size of the GDP of each of the countries they belong to. Our sample period begins in 1997 and ends in 2012. Our sample covers data for a total of 2,153 banks operating in 43 countries. We employ panel regression methodology with the country fixed effects and the year fixed effects. The dependent variables used in this paper include return on asset, non-performing loan ratio, loan-reserve ratio, capital ratio, liquidity ratio and cost ratio. Our main independent variables include total assets, systemic asset size calculated by total asset divided by GDP, asset growth rate, GDP growth rate, and inflation rate. Our results can be summarized as follows. First, bank profitability measured by ROA (return on asset) decreases as asset size gets bigger, regardless of whether we use absolute size or relative systemic size. Second, we find mixed effects of bank size on risk measures. As a bank gets bigger, its capital adequacy ratio decreases, but other ratios, such as asset soundness ratio, liquidity ratio and cost ratio are increasing irrespective of whether size is measured by its total asset or systemic size. Third, the effects of bank size on the bank’s risks differ across countries. More importantly, they differ according to the bank’s systemic size. In the case of smaller banks, asset growth improves asset soundness and liquidity. However, in the case of systemically large banks whose asset size exceeds 80% of the GDP of its country, bank size growth increases bank risk while decreasing profitability. Fourth, economies of scale exist in all countries for all size groups of the sample banks except for systemically large banks, i.e., biga100 (which belong to the largest systemic size group and account for more than 80% of the GDPs of their respective countries). Systemically large banks (biga100) exhibit no economies of scale. In summary, we find that there are both positives and negatives resulting from growth in bank assets. As the size of bank assets grows, both the asset soundness ratio and liquidity ratio improve with a saving in cost, but the profitability ratio and capital adequacy ratio deteriorate. However, asset growth in systemically large banks is associated with increased risks, without achieving higher profits. Our results also imply that there exists an optimal asset size which can maximize profit and minimize risk. Our results indicate that the optimal asset size will lie somewhere between 60% and 80% of the GDP of the country to which a bank belongs. Investigating which factors result in cross-country differences is left for future research.

한국어

본 연구는 은행의 대형화가 수익성, 리스크 및 비용효율성에 미치는 영향을 43개국 2,153개 은행을 대상으로 분석한 것이다. 실증분석 결과는 다음과 같다. 첫째, 은행이 대형화될수록 절대규모나 시스템 위험규모와 관계없이 수익성이 하락하고, 자본적정성이 악화되는 음(-)의 효과와 자산건전성은 개선되고, 비용효율성도 개선되는 양(+)의 효과가 존재하는 이중적인 효과가 있다. 둘째, 은행 대형화가 리스크에 미치는 영향은 국가별로, 규모별로, 리스크 종류별로 다르다. 소형은행의 경우, 대형화에 따라 수익성이 감소하고 자본적정성이 악화되지만 자산건전성과 비용효율성이 개선되는 효과는 얻을 수 있으나, 대형은행이 초대형화 되는 경우에는 수익성의 증가를 수반하지 않으면서 리스크만 커지는 결과를 초래하는 것으로 나타났다. 셋째, 절대규모를 기준으로 볼 때, 은행 대형화에 따른 규모의 경제효과는 존재하는 것으로 보이나, GDP 대비 80%를 초과할 정도의 초대형화의 경우 오히려 규모의 비경제가 나타났다. 은행의 대형화로 인해 리스크가 감소하는 효과는 모든 리스크 종류에 무관하게 항상 존재하는 것은 아니며, 특히 시스템 위험의 규모가 매우 큰 초대형은행의 경우 자본적정성 하락효과가 매우 큰 것으로 나타났다. 따라서, SIFI 등 바젤 III의 추가 자본규제는 필요하나, 시스템 위험규모를 기준으로 실시하는 것이 타당하며, 국가별 GDP 대비 60%∼80% 수준에서 수익성 하락을 최소화 하면서 리스크를 최소화 시킬 수 있는 최적자산규모가 존재할 수 있다.