Governance and Performance of Banks in Prewar Japan: Testing the "Organ Bank" Hypothesis Quantitatively

More than forty years ago, Kato[1957] posed the organ bank hypothesis. Namely, he stressed that in prewar Japan, many of the banks were tightly connected with certain industrial companies, and those banks loosely gave loans to the connected companies, which eventually resulted in the Showa Financial Crisis. This view has been widely accepted by economic historians. However, there has been no attempt to test the organ bank hypothesis quantitatively. In this paper, we tested the organ bank hypothesis using quantitative data and econometrical methodology. First, we compiled a comprehensive database of company directors and auditors, based on Zenkoku Shogaisha Yakuinroku 1926 issue(Shogyo Koshinjo[1926]), and using the database we identified the interlocking of directors and auditors between banks and non -banking companies. Interlocking of directors and auditors between banks and non -banking companies was very pervasive. Nearly 90% of ordinary banks had more than one directors or auditors who were at the same time directors or auditors of non -banking companies, and average number of interlocking per bank was as large as 7.85. Observing by banks scale, we found that interlocking was more pervasive in the large-sized banks. Second, using the interlocking variables, we examine the influence of interlocking on bank performance. Through regression analyses we found that interlocking tended to give negative effect on the liquidity performance and profitability of banks, and increased the probability of bank closures in 1927. Also, the interest rates of the deposits of those banks with interlocking were relatively high, while the interest rate of loans were not, and consequently profit margins of those banks were relatively small. It implies that the banks with interlocking should offer relatively high interest rate to gather deposits, because of the low evaluation of the financial market. Those findings support the organ bank hypothesis. In prewar Japan, banks' business practices based on connection of the directors and auditors made the banking system unsound, and eventually caused the Showa Financial Crisis in 1927. On the other hand, many of the literature on the Asian Crisis in 1997 stress exploitation of the minority shareholders by the core members of the family-based companies. In this sense, the Showa Financial Crisis was a predecessor of the Asian Crisis in 1997.