Puzzles on Comparative Corporate Governance: Rethinking the Linkage between Law and Ownership Preliminary February 13, 2016 Hideki Kanda/*/ I. Introduction Two familiar inquiries in the comparative study of law are why laws in jurisdictions are sometimes similar and sometimes different and when laws change. A companion with the second inquiry is the familiar question of whether laws are converging or not. In this presentation, I revisit these inquiries in the area of comparative corporate governance. I present a theme with the hope to add new insights to the vast amount of the past and existing literature in this area. Specifically, I offer a simple model of differences in the corporate governance laws among European jurisdictions, Japan and the United States. My argument is that differences in the laws on corporate governance among these jurisdictions are correlated to differences in the modal ownership (or ownership and control) regime of their large publicly held firms. I hope that this model will help us better understand and examine the situations in Asian jurisdictions. II. Ownership and Control A. Presenting a Theme It is well-known that differences in the laws on corporate governance can be explained by focusing on the ownership structure in the relevant jurisdictions. In particular, whether stock ownership is dispersed or concentrated has been pointed out to be the key to understanding the differences in the laws we observe. Thus, powerful 1
"inside" shareholders play a more important role in large continental European companies than in their U.S. counterparts, where shares are held by shifting pools of "outside" shareholders./1/ It is generally assumed that differences between these two ideal types of ownership regimes have important implications for corporate governance. For example, it is said that managerialism in the U.S. is more costly because there are fewer powerful shareholders to monitor managers, while private benefits of control are larger in Europe as a result of the prevalence of concentrated ownership. It is also sometimes assumed that certain features of substantive corporate governance law either result from ownership structure or cause it. For example, it is said that jurisdictions with concentrated ownership are likely to lack minority shareholder rights or, vice versa, jurisdictions that fail to protect minority shareholders are destined to have concentrated ownership. I believe that this argument is useful but not sufficient. Indeed, the dichotomy of dispersed and concentrated ownership does not explain the U.K. and Japan well. Thus, I offer a matrix of four rather than two ideal types of ownership (or ownership and control) regimes. Using four instead of two ideal types of ownership regimes allows us to design a model that explains most (hopefully all) important legal differences in the substance and institutional structure of corporate governance law. If we have a look at ownership regimes as they stood in the late 1990s, we observe that most public continental European companies had owners with voice and large private benefits. We thus assume that controlling shareholders were the interest group shaping continental European corporate governance institutions and substantive law. Conversely, given that shareholders had a weak voice and small private benefits in large American companies, we assume that managers were the dominant interest group in the U.S.; controlling shareholders owning smaller publicly traded American companies were too dispersed to count as significant opponents./2/ 2
The U.K. and Japan are special cases. In the U.K., unlike in continental Europe, the private benefits of owners of large companies were small. However, institutional investors had a voice. We thus assume that institutional investors were the dominant interest group in the U.K. In Japan, by contrast, owners were able to extract large private benefits but had a weak voice. Perhaps this can be linked to time-honored reliance upon oligarchic mechanisms and complex webs of cross-shareholdings, which are powerful enough to permit the extraction of privileged returns, but do not provide specific incentives to either seek or exercise voice. I therefore assume that, like in the U.S., managers were the dominant interest group in Japan. Interest group dominance varies over time. There have been significant changes in ownership regimes throughout the 20th century and it would be unwise to posit that the late 1990s equilibrium is a stable one. Ownership regimes are evolving due to the increasingly global presence of Anglo-Saxon institutional investors and hedge funds, ongoing European privatizations, the growing sale of family assets and the emergence of regulatory entrepreneurs (stock exchanges and others who promulgate corporate governance "codes"). However, the dominance impact of ownership changes is not necessarily immediate. B. A New Typology of Ownership and Control Regimes Table A [See Slide] provides a matrix of ownership and control regimes (at least) as they stood in the 1990s. The U.K. stands out as the example of the classic ownership regime described in corporate finance textbooks. By and large, U.K. institutional shareholders hold shares solely for investment value. They are too small and distinct to coalesce into controlling groups, but they are sufficiently large and organized to exercise effective voice. British institutional investors are clearly the dominant partners in their relationship with U.K. CEOs. These investors are, as it were, ideal shareholders from the perspective of agency theory. They are motivated by investment value, and they are able 3
to exercise their control rights or liquidate their holdings to protect shareholder value. This was not always so, since founding families dominated most U.K. listed firms as recently as 1960. By the 1980s, however, institutional investors had almost entirely replaced these families and became the dominant shareholder group in the U.K. In continental Europe, we enter the insider realm of controlling shareholders. Controlling shareholders traditionally dominated in the continental European corporate landscape. From the early to the late 1900s, boardrooms of most German, French, Italian, Spanish or Swedish listed firms have been controlled by wealthy individuals, families, parent companies or the state. Moreover, shareholder voice has been going hand in hand with large private benefits. Controlling shareholders have been getting such benefits through self-dealing (especially within the group of companies), control premia, company perks or social status. Thus, albeit with some exceptions, continental European firms can be categorized as insider shareholder regimes. For the U.S., we remain in the realm of outside shareholders but enter the region of managerial control. At least since the New Deal and probably from the turn of the 20th century, large listed American companies have tended to be widely-held. In contrast to their U.K. counterparts, however, American institutional shareholders have traditionally had limited influence in the boardroom for a variety of legal and historical reasons, many dating from the conscious policies of the New Deal. Although the balance of power between institutional shareholders and managers has changed with the rise of corporate governance movement, America was traditionally a jurisdiction in which outside owners had little voice in the governance of large listed public companies, and the modal public company operated under an ownership regime with both "outsider" and "managerialist." Finally, in respect of Japan, we enter the quadrant of insider managerialist regimes. From the mid-1940s to the 1990s, managers of Japanese listed companies 4
typically enjoyed compliant shareholder support in the form of stable crossshareholdings. Through cross-shareholding alliances, companies enjoyed a dependable shareholder base, and also participated in the shareholder base of other companies with which they were affiliated. In this regime, companies rarely intervened in governance affairs of the firms in which they invested, since investment returns were not the point to begin with or at least not the only point. Rather, companies bought stock to cement commercial relationships and enjoy the reciprocal benefits of recruiting loyal shareholders to their own investor bases. These ownership structures should not prove more stable in the future than they have been in the past. In fact, ownership structures are currently evolving due to the global presence of Anglo-Saxon institutional investors and hedge funds, significant European privatizations, the growing sale of family assets and the emergence of regulatory entrepreneurs. C. Explaining Laws with the Matrix of Ownership and Control a. Employee Participation The dispersed ownership jurisdictions do not recognize employee participation in the board. In Germany, a representative of the concentrated ownership jurisdiction, certain large-scale companies must provide "co-determination" by which board members are elected both by shareholders and by employees. Japan, also a concentrated ownership jurisdiction, does not recognize such system in corporate law, but under lifetime employment practice almost all directors of publicly held firms were former employees of the same firm at least until the early 1990s, and this is still true today for most directors. The dimension of control in my matrix is not necessary here, but one might be interested in why concentrated ownership jurisdictions developed employee representation on the board. 5
b. Separation of CEO and Chairman In the U.K., stock exchange rule provides separation of CEO and chairman of the board, although firms can opt out from this (and from other rules as well) under the "comply or explain" procedure (i.e., the firm must explain if it opts out). Such separation has never been adopted as a rule in the U.S. Thus, it is interesting that the U.S. and the U.K. are different, and the control dimension in the matrix works rather neatly here. Japan does not have such a rule, and generally, it is said that continental European countries have such a rule. c. Private Litigation by Shareholders It may be misleading if I take up the amount of litigation here [please see the slide], inasmuch as I discuss legal rules throughout this presentation. Both in the U.S. and Japan, the system of shareholder derivative action is recognized. Japan, however, does not have the class action system. Both the U.K. and continental Europe have been unfamiliar to shareholder derivative action, although some form of private litigation by shareholders has been recognized in recent years. In the managers jurisdictions, unsatisfied shareholders sue, and in the dispersed ownership jurisdictions (with managers), those shareholders sue more often than in the concentrated ownership jurisdictions. In contrast, it should not be surprising to find that shareholders do not sue in the shareholders jurisdictions because they exercise control and do not need to sue, but in the concentrated ownership jurisdictions (with shareholders), those who exercise control are controlling shareholders, and minority shareholders sue. d. Defenses against Hostile Takeovers 6
The landscape about legal rules concerning takeover defense is messy. The U.S. and the U.K. show a sharp contrast. The U.K. does not permit board discretion and requires shareholders' decision when the firm wants to adopt defensive measures against hostile takeovers. The U.S. generally permits board discretion, subject to judicial review under the "reasonableness" standard (known as the enhanced business judgment rule). In both Japan and continental Europe, until several years ago, hostile takeovers were almost not possible (with few exceptions). In these jurisdictions, however, legal rules are evolving today. With a few hostile takeover attempts in recent years, Japanese courts have been struggling in search for the appropriate standard of judicial review. In Europe, the EU Takeover Directive tries to bring the member states in line with the U.K., but it allows opt-out. Thus, it is likely that board discretion will remain to be permitted at least in some continental European jurisdictions. III. Preliminary Conclusion In this presentation, I have offered a theme with a matrix of ownership and control with the hope to better understand the linkage between ownership and corporate governance laws around the world. In my view, in addition to the dichotomy of dispersed and concentrated ownership, the control dimension should be paid more serious attention in the research on comparative corporate governance law. Notes /*/ Professor of Law, University of Tokyo. Some of the ideas in this presentation were developed with Professors Gerard Hertig and Reinier Kraakman. /1/ See e.g. Rafael La Porta, Florencio Lopez-de-Silanes, and Andrei Shleifer, Corporate Ownership Around the World, 54 Journal of Finance 471 (1999). See also Lucian Bebchuk, The Elusive Quest for Global Governance Standards, 157 University of Pennsylvania Law Review 1263 (2009). 7
/2/ See Ronald J. Gilson, Controlling Shareholders and Corporate Governance: Complicating the Comparative Taxonomy, 119 Harvard Law Review 1641 (2006) (pointing out that the size of the economy prevents families from playing a Europeanstyle national role). 8
Puzzles on Comparative Corporate Governance: Rethinking the Linkage Between Law and Ownership February 22, 2016 Hideki Kanda University of Tokyo 9
Current Scholarship: Legal rules (and their enforcement) on corporate governance in publicly held business corporations are explained by stock ownership patterns: dispersed or concentrated. Problem: It is difficult to locate (and explain) the U.K. and Japan. My Claim: The dimension of control should be added to that of ownership. The Issue: What was the landscape in the 1990s? Is the picture changing? 10
Current Scholarship (Coffee, Gilson and others) OWNERSHIP dispersed concentrated U.S. Continental Europe It is difficult to locate (and explain) the U.K. and Japan. 11
Adding Control to Ownership control ownership dispersed concentrated managers U.S. shareholders Continental Europe 12
Adding the U.K. and Japan control ownership dispersed concentrated managers U.S. Japan shareholders U.K. Continental Europe 13
Example 1: Employee Board Representation control ownership dispersed concentrated managers no (U.S.) yes (Japan) shareholders no (U.K.) yes (Continental Europe) 14
Example 2: Separation of CEO and Chairman control ownership dispersed concentrated managers no (U.S.) no (Japan) shareholders yes (U.K.) yes (Continental Europe) 15
Example 2a: Definition of Independent Director in the 2010s Is Controlling Shareholder Excluded? control ownership dispersed concentrated managers no (U.S.) yes (partially) (Japan) shareholders yes (U.K.) yes (?) (Continental Europe) See Puchniak and Lan (2015) 16
Example 3: Private (Shareholder) Litigation control ownership dispersed concentrated managers many cases (U.S.) some cases (Japan) shareholders few cases (U.K.) some cases (Continental Europe) 17
Example 3a: Enforcement Comply or Explain Approach for Independent Directors control ownership dispersed concentrated managers no (U.S.) yes (from 2015) (Japan) shareholders yes (U.K.) yes (Continental Europe) 18
Example 4: Defenses against Hostile Takeovers control ownership managers shareholders dispersed board discretion (U.S.) board passiveness (U.K.) 19 concentrated manager discretion (Japan) control shareholder discretion (Continental Europe)
Note on Takeovers and Defenses mandatory bid rule no yes board passiveness yes U.K. no U.S. Germany 20
The 1990s versus the 2010s control ownership dispersed concentrated managers U.S. Japan shareholders U.K.??? Continental Europe 21
Conclusion: A matrix of ownership and control may enrich our understanding of the variety of corporate governance laws around the world. 22