نظارت و مقررات سهامداران بزرگ: تجربه بانکداری ژاپن

During a period where Japanese banks operated under a less restrictive regulatory environment, 1986–1988, we find positive relations between bank risks and ownership concentration. This empirical evidence suggests that shareholder monitoring is present when the potential return to monitoring is high (the “shareholder monitoring hypothesis”). During the periods before and after this particular period, we do not observe evidence of shareholder oversight. These results are consistent with the argument that regulation (an external governance mechanism) and shareholder monitoring (an internal governance mechanism) are substitutes for one another (the “substitution hypothesis”) because the pre- and post-periods are characterized by stricter regulatory environments. Finally, tests on bank performance lend supporting evidence to both hypotheses.

Researchers contend that regulation and shareholder monitoring are substitute governance mechanisms. For example, Demsetz and Lehn (1985) suggest that “systematic regulation restricts options available to owners” and “regulation also provides some subsidized monitoring and disciplining of the management of regulated firms”. Therefore, for example, financial firms and utility firms probably do not have active shareholders. Black (1998) also argues that regulation obstructs the potential for effective shareholder oversight. When a firm operates in a regulated industry or environment, the gains to shareholder monitoring may be limited. The implication of these contentions is that an internal governance mechanism (e.g., shareholder monitoring) can be a substitute for an external governance mechanism (i.e., regulation). Concentrated ownership is often used to identify shareholder monitoring (e.g., Demsetz & Lehn, 1985; La Porta, Lopez-de-Silanes, Shleifer, & Vishny, 1998; Shleifer & Vishny, 1986). When a public firm's ownership is concentrated into the hands of a few large shareholders, then these large shareholders should have both the desire and the power to monitor the firm's operations and management. For example, according to Shleifer and Vishny, 1997 and Shleifer and Vishny, 1986, large shareholders have the incentive to collect information and monitor management for the purposes of profit maximization. If large shareholders are inactive, then there is an obvious cost as their inactivity (or shirking) is significantly borne to large shareholders (Demsetz & Lehn, 1985). As such, Demsetz, 1983 and Demsetz, 1986 argues that a shareholder's ability to exercise oversight and control “must be the primary explanation for ownership concentration”. In a similar vein, Grossman and Hart (1986) contend that observed ownership concentration is, in effect, the same as observing the extent to which large shareholders have control. However, the benefits to control are unlikely to be equal across firms. Demsetz and Lehn (1985) contend that large shareholders will be present in risky firms, as the payoff potential to active monitoring of risky firms is high. The mere presence of large owners in risky firms strongly suggests they are monitoring, as the downside potential of not monitoring is also especially high in risky firms. In empirical tests, Demsetz and Lehn (1985) find that firms in regulated industries have lower ownership concentrations. For unregulated firms, they find the relation between ownership concentration and firm risks to be positive. Their finding that ownership concentration and regulation are negatively related supports the “substitution hypothesis”. Their finding that ownership concentration and firm risks are positively related supports the “shareholder monitoring hypothesis”. Holderness, Kroszer, and Sheehan (1999), and Himmelberg, Hubbard, and Palia (1998) also find a positive relation between ownership concentration and firm risk. They too suggest that this evidence is indicative of monitoring by large shareholders. Other researchers find support for the substitution hypothesis. Kole and Lehn (1999) find that controlling shareholders in the US airlines industry instituted numerous governance mechanisms after the industry was deregulated. Anderson and Fraser (2000) study US banks during more regulated and less regulated periods and find managerial shareholdings and risk-taking are positively related only during the less regulated period. Konishi and Yasuda (2004) study risk-taking by Japanese banks after its capital adequacy requirement was increased. They find that risk-taking by the banks’ stable shareholders declined after the capital adequacy requirement was increased. All of these papers’ findings can be viewed as being consistent with the substitution hypothesis. Owners are more active when there is less regulation. Booth, Cornett, and Tehranian (2002) find that two internal governance mechanisms, board quality (proxied by the fraction of independent directors) and officer/director stock ownership, are substitutes for each other. However, for regulated industries (banks and utilities), they find that independent boards and officer/director stock ownership are not substitute governance mechanisms. Therefore, they also contend that an external governance mechanism (i.e., regulation) serves as a substitute for internal governance mechanisms (i.e., independent directors and stock ownership). La Porta et al. (1998) conduct a cross-country test of the substitution hypothesis. They find that firms in countries with relatively weak shareholder protection laws have relatively high ownership concentrations, which support the substitution hypothesis—shareholders look out for their own interests when laws do not. Caprio et al. (2003) also conduct a cross-country test while specifically focusing on the relation between bank governance and bank value. They find that banks in countries with strong legal environments are often widely held, which is consistent with La Porta et al. (1998) and the substitution hypothesis.1 In general, the literature has tested the substitution hypothesis in one of three ways. Papers have either (1) contrasted periods before and/or after a change in the regulatory environment (e.g., Anderson & Fraser, 2000; Kole & Lehn, 1999; Konishi & Yasuda, 2004), (2) contrasted firms from regulated industries to firms from nonregulated industries (e.g., Booth et al., 2002; Demsetz & Lehn, 1985), or (3) contrasted the internal governance of firms from countries with different regulatory environments (e.g., La Porta et al., 1998 and Caprio et al., 2003). We offer additional evidence on the substitution hypothesis taking the first approach mentioned above. However, in our paper, we study an industry that experiences a significant deregulation, and then experiences a significant increase in regulation (we will discuss other aspects of our study that differentiates our study from others, shortly). Therefore, we will be able to test the substitution hypothesis surrounding two shifts (in opposing directions) in the regulatory environment, within the same industry. Specifically, we study large shareholders of Japanese banks under three different regulatory regime periods. According to the substitution hypothesis, when deregulation occurs we should observe oversight by large shareholders, which, in turn, implies a positive relation between bank risk and ownership concentration during the deregulated period. When an increase in regulation subsequently occurs, then the positive relation between risk and ownership concentration is expected to disappear.2

We examine ownership concentration and bank risk during the period from 1983 to 1996 for a sample of Japanese commercial banks. During the event period of 1986–1988 when the capital ratio requirement significantly decreased and deposit insurance significantly increased, a positive relation is exhibited between three measures of bank risks and ownership concentration. This finding reveals shareholder oversight during a period where the regulatory environment was less restrictive. We also find a stronger relation between ownership concentration and bank performance (accounting profits) under the less restrictive regulatory environment. Both these risk and profit findings do not persist in subsequent periods, which are distinguished by the Basle Accord regime. Overall, our research reveals that large shareholders of banks can potentially be important for bank governance (shareholder monitoring hypothesis). Perhaps most importantly, our findings indicate that bank shareholders will exert oversight when the regulatory environment provides more flexibility and more freedom to shareholders. As such, our findings lend support to the substitution hypothesis that regulatory oversight (an external governance mechanism) and shareholder oversight (an internal governance mechanism) substitute for one another. While our paper does not attempt to argue that one governance mechanism is better than another, our results do echo the sentiments of Chami et al. (2005), who stress that the balance between bank regulations and shareholder oversight (i.e., market discipline) is a delicate and complicated issue, and thus likely to be a difficult one to address in the on-going deliberations on the future of bank regulation.