Deregulation and governance: plight of Japanese banks in the 1990s

Abstract

The governance structure of public corporations is determined by the agency relationship between shareholders and managers, and the agency theory predicts that deregulation of an industry leads to governance adaptation. Deregulation of the Japanese banking business in the 1980s offers an interesting case study in this framework, as Japanese banks fell into serious solvency problems in the post-deregulation 1990s. This paper investigates whether ineffective governance was responsible for the plight of Japanese banks. The sample is 384 corporations that were listed on the Tokyo Stock Exchange between 1983 and 1990, which includes 59 banks. One of the findings is that bank shareholders allowed more diffuse ownership structure after deregulation, which demonstrates that the agency theory did not hold for shareholders of Japanese banks.

Acknowledgement

The author would like to thank Sumner LaCroix and Peter Lanjouw for helpful comments on earlier manuscripts. Akiko Sugimoto-Kawaura provided excellent research assistance. Financial assistance from the Zengin Foundation for Studies on Economics and Finance is gratefully acknowledged.

Notes

¹ Kole and Lehn (1999) is a pioneering work in the analysis of this dynamic relationship. Using the Airline Deregulation Act of 1978 as a natural experiment, they examined the change in the governance structure of US airlines for the 1971–1992 period, and found that governance structure generally adapted in manners predicted by the agency theory.

² For details of deregulation measures, see Endo (1996).

³ See OECD (1995, p. 116) for the account of closure of two credit unions in 1994. Gup (1998, pp. 35–42) offers an overview of bank failure incidents in Japan.

⁴ Aoki (1983) emphasizes the importance of greater shareholding concentration as a mechanism to enforce wealth maximization against managerial discretion. Shleifer and Vishny (1986) present a theoretical framework in which large shareholders bring about value-maximizing changes in corporate policy. Yafeh and Yosha (1995) find that expenditures on activities with potential for managerial shirking, e.g. advertising and entertainment expenses, are inversely correlated with shareholder concentration in a sample of Japanese firms. Weir (1997) find that executive directors of acquired firms had significantly lower shareholdings than those of non-acquired firms in an analysis of UK mergers. Morck et al. (1988) and Short and Keasey (1999) conduct statistical tests based on the managerial entrenchment hypothesis that the relationship between shareholdings by board members and corporate performance is not linear.

⁵ As of the end of March 1983, there were 589 companies with the same cycle of financial year (April–March) in the First Section of the Tokyo Stock Exchange. The sample is reduced to 558 firms after removing firms for which consistent data are not available in March 1990, which is selected as the post-deregulation point as the major component of deregulation package, interest rates liberalization, was completed by the end of the 1980s. The sample size is further trimmed to 384 (with 59 banks) in order to remove direct effects from another regulatory change. The Fair Trade Commission amended the Anti-Monopoly Act in 1977, when they prohibited banks from owning other companies in excess of 5% of stocks outstanding. Banks were required to comply with this rule during a grace period of 10 years, i.e. by 1987. Companies in which banks had more than 5% of shares in 1983 were removed from the sample as their inclusion would bias the analysis.

⁶ Data are available from company's financial statements and various issues of Rates of Return on Common Stocks published by the Japan Securities Research Institute.

⁷ Results are not sensitive to the number of firms in the ownership concentration index. Regression results obtained with the share of three and five largest shareholders as dependent variables are equally significant, and lead to the same conclusions. These results are available from the author upon request.

⁸ As an example of direct control, BOJ had a monetary policy instrument called ‘window guidance’ to dictate the volume of credit for individual banks between 1961 and 1991 (see Rhodes and Yoshino, 1999). It was not until the middle of the 1990s that BOJ had to be engaged in the publicly backed disposal of a failed financial institution for the first time since 1927. The author thanks an anonymous referee for pointing out this perception issue.

⁹ These are five categories of publicly traded banks that operated in Japan in the 1980s. Two features distinguished these banks. First, city banks, regional banks, and sogo banks were primarily engaged in short-term finance, while long-term credit banks and trust banks provided long-term credit. Second, long-term credit banks, city banks and trust banks operated on a nation-wide basis, and regional banks and sogo banks were based in the principal city of a prefecture and served clients of a particular region. For a more detailed account of the Japanese financial system in the 1980s, see Suzuki (1987) and Endo (1996).

¹⁰ A Herfindahl index is obtained by summing the squared share of stocks controlled by each shareholder. Although a complete list of all the shareholders, together with the number of shares held by each of them, is necessary to calculate an exact Herfindahl index, this information is available only for the seven to ten largest shareholders in Japanese corporate financial statements. Therefore AH is an approximation based on information on the largest owners and shareholding structure data, i.e. the number of shareholders in each range of shareholdings, such as more than 1000 unit (1000 shares/unit) stocks, 500 to 999 unit, 100 to 499 unit, and so on. For an operational purpose, this index is used after multiplied by 10⁴ in this study.

¹¹ Kini et al. (1995) provide evidence that the takeover market plays an important role in the governance mechanism for US corporations.

¹² Horiuchi and Shimizu (2001) investigate whether these former officials had any impacts on the soundness of bank management, and find that their acceptance to the management led to more aggressive risk-taking.