in Comparative Perspective: Prospects for Convergence

Sanford M. Jacoby The Anderson School UCLA Los Angeles, CA 90095-1481 [email protected]

October 2001

Contribution to the “Symposium on Employees and Corporate Governance” to appear in Comparative Labor Law & Policy Journal. 2

During the last decade, corporate governance has become a contested and controversial issue. The rules for governing public corporations are being challenged by a diverse group of critics ranging from institutional investors to labor unions and social activists. Some hope to give greater weight in corporate decision-making to shareholders; others hope to bolster the influence of other stakeholders such as employees and customers. Less organized but nevertheless influential are the professional managers who make most of the strategic and operating decisions for the corporation. Since the early twentieth century, corporations have come under the control of these professional managers. 1 Yet managers typically own only a relatively small portion of the assets they manage. These facts give rise to two key issues in corporate governance: first, how to make managers accountable to those who have made investments in the corporation, and second, what constitutes an “investment” in the corporation that warrants the right to monitor and direct management.. There is stunning international variety in the rules by which these issues are resolved; no two countries handle corporate governance in precisely the same way. These differences—and the international debate that swirls around them—have recently attracted an outpouring of scholarly research on the characteristics and consequences of various corporate governance systems. 2 Despite a welter of differences, it’s possible to identify two distinctive patterns among the advanced industrial countries. 3 First there is the “shareholder” system, which also goes by such names as the Anglo-American system, the market-outsider system, or simply stock-market capitalism. The other is the “stakeholder” system, which has also been called the relational-insider system, the dedicated-capital system, and welfare capitalism. This is the model that prevails in , and some other countries. 4 The former may be thought of as an “exit” model, where shareholders sell shares to

1 Alfred D. Chandler, Jr. and Herman Daems, eds., Managerial Hierarchies: Comparative Perspectives on the Rise of the Modern Industrial Enterprise (1980). 2 For a magisterial overview, see Klaus J. Hopt, Hideki Kanda, Mark J. Roe, Eddy Wymeersch, and Stefan Prigge, Comparative Corporate Governance: The State of the Art and Emerging Research (1998). 3 Some separately classify he French and Scandinavian systems. See Rafael LaPorta, Florencio Lopez-de- Silanes, Andrei Shleifer, and Robert Vishny, “Law and Finance,” 106 Journal of Political Economy (1998), 1113-1155. 4 Sanford M. Jacoby, “Employee Representation and Corporate Governance” 3 Univ. of Penna. Journal of Labor and Employment Law 449 (2001); Ronald Dore, Stock Market Capitalism: Welfare Capitalism; Japan and German versus the Anglo-Saxons (2000) 3 express dissatisfaction with management. 5 The latter is a “voice” model, where shareholders—as well as employees and other parties--are less mobile and more inclined to express concerns by communicating directly with management. As the papers and comments in this Symposium indicate, these systems assign to employees very different roles in corporate governance—little or no role under the Anglo-American system, moderate (by custom) in Japan, and important (by statute) in Germany. The Anglo-American or shareholder system is one in which equity is a relatively important form of corporate finance; corporate shares are widely held and easily traded; and shareholders are privileged to elect the board of directors because of their status as the corporation’s residual claimants. Shareholders who are dissatisfied with management can seek to replace it or sell their shares. The latter option includes the possibility of selling shares to a group seeking to replace the incumbent management; this is the “market for corporate control.”6 The stakeholder or relational system is one in which shareholding is more concentrated, with large blocs held by banks, other corporations, and families. These blocs tend not to be actively traded. The greater reliance on debt financing and a governance role for banks. Banks hold equity either in direct or depositary form, and may be represented on the board of directors (a standard practice in Germany but one occurring chiefly in crisis situations in Japan). 7 Shares also are held by key customers, suppliers, and allied corporations, often on a reciprocal basis. Concentrated, dedicated holdings make it difficult to establish a market for corporate control. Finally, employees play a role in corporate governance. In Germany, this occurs through a legally-mandated system of works councils and co-determination. In Japan, most large companies have enterprise unions and joint committee with access to senior management. There is also a managerial ethos that employees are an important corporate stakeholder and that

5 Albert O. Hirschman, Exit, Voice, and Loyalty (1972). 6 An excellent description of the U.S. system may be found in Margaret Blair, Ownership and Control: Rethinking Corporate Governance for the Twenty-First Century (1995). On the United Kingdom, see Jonathan Charkham, Keeping Good Company: A Study of Corporate Governance in Five Countries (1995). 7 Steven N. Kaplan and Bernadette Minton, “Appointments of Outsiders to Japanese Boards” 36 Journal of Financial 225 (1994); Ulrike Schaede, “Corporate Governance in Japan” (CCC Working Paper no. 92-12, 1992). . 4 management’s job is to mediate between shareholders, employees, and other stakeholders such as business groups members and suppliers. 8 In short, there is considerable cross-national diversity in corporate governance and the impact that employees have on it. But how is that national systems achieved such diversity in the first place? And what are the prospects for convergence in corporate governance systems? These questions are considered in this essay. Pundits of globalization predict the eventual demise of the relational governance model and its replacement by the Anglo-American model. Were this to occur, it would sharply change the relationship of employees to their employer in many parts of the world. Yet it’s not obvious that convergence is inevitable. As I will show, there is considerable inertia and persistence of national governance models. And while one can find evidence of institutional diffusion across borders, it occurs in multiple directions; not all roads lead to Wall Street. I. Origins of National Diversity An essay of this brevity cannot do justice to the vast literature on comparative corporate governance. Suffice it to say that the usual explanations of national diversity in economic institutions—law, politics, culture, and resource endowments—also come into play in the area of corporate governance. a. Law Legal explanations for corporate governance patterns start with the observation that the “typical” U.S. pattern of a corporation being owned by a multitude of investors is actually rather atypical among advanced economies.9 The pattern is limited to common law countries. The association between dispersed ownership and the common law has been attributed to the British courts’ shift of allegiance from the Crown to Parliament and property owners. The courts eventually came to protect investors not only against

8 Keisuke Nakamura and Michio Nitta, “Developments in Industrial Relations and Human Resource Practices in Japan” in Richard A. Locke et al., eds., Employment Relations in a Changing World (1994), 347-50; Jacoby, supra note 4, at 462-3; Masahiko Aoki, Information, Incentives, and Bargaining in the Japanese Economy (1988). A recent survey found strong support among Japanese managers for the notion that a corporation has multiple stakeholders including employees, and little support for the idea that corporations are the property of their shareholders and employees merely a factor of production. Takeshi Inagami and RIALS, eds., Gendai Nihon-no Koporeto Gabanansu (Corporate Governance in Contemporary Japan), table 2-18 (2000). 9 Rafael La Porta, Florencio Lopez-de-Silanes, and Andrei Shleifer, “Corporate Ownership Around the World,” 54 Journal of Finance 471 (April 1999). 5 monarchial expropriation but also against inside dealing and other forms of “crony capitalism.” In civil law countries, however, the state maintained its control of corporations and resisted surrendering its power , leaving investors relatively unprotected. Statistical analyses show that today, civil law countries, especially those following French legal traditions, offer relatively weaker protection to investors in areas such as limitations on incumbent directors (e.g., mail proxies) and creditor rights (e.g., automatic stays on secured assets). Hence in civil law countries, there is a tendency toward concentrated ownership, which can be interpreted in two very different ways: either as an expression of cronyism or as a form of monitoring to prevent abuses that in common law countries are subject to legal enforcement.10 There are, however, problems with the argument that civil law countries offer less protection to small investors and that they lack extensive equity markets.. First, it doesn’t well apply to Japan, a country influenced by German legal traditions. Despite corporate cross-shareholdings and bloc ownership, Japan nevertheless has a huge and liquid market for corporate equities. It also does not appear to be true that common law countries consistently do a better job of protecting small investors. In the early-to-mid nineteenth century, U.S. corporations as well as those in Western Europe had voting schemes that favored small investors by allowing one vote per investor—or some kind of graduated scale--rather than one vote per share. But these democratic voting schemes began to disappear in the in the mid-nineteenth century, giving disproportionate power to large owners, whereas in France, Germany, and other countries, graduated voting scales persisted far longer than in the United States, even into the twentieth century. 11

b. Politics

10 10Rafael LaPorta, Florencio Lopez-de-Silanes, Andrei Shleifer, and Robert Vishny, “Investment Protection and Corporate Governance,” 58 Journal of Financial Economics 3 (Oct.-Nov. 2000), passim. 11 Colleen A. Dunlavy, “Corporate Governance in Late 19th-Century Europe and the U.S.: The Case of Shareholder Voting Rights” in Hopt et al., supra note 2, at 5-39. The demise of democratically-constituted voting schemes and the absence of employee voice are of a piece with the relatively weak U.S. constitutional protections against disenfranchisment. The U.S. is one of the few countries that does not constitutionally guarantee its citizens the right to vote. See Alexander Keyssar, The Right to Vote: The Contested History of Democracy in the United States (2001). 6

Another way of classifying corporate governance systems is to distinguish between those with bank ownership and involvement, as in Germany and Japan, and those where banks play a relatively small role, as in the United States and Britain. To account for this difference, scholars like Mark Roe have emphasized the importance of political traditions. In the U.S., populist suspicion of concentrated financial power (and other large institutions) led to statutory restrictions on the creation of nationwide banks and, later on, to the Glass-Steagall Act’s separation of commercial and investment banking. Hence U.S. banks never became as large as their foreign counterparts, as able to invest in equities, or as motivated to monitor corporate managers. 12 Now the shoe is on the other foot: there is an implicit presumption that politics may have interfered with the development of an “efficient” governance system in the United States, whereas in the legal explanations favored by neoclassical the presumption is that common- law countries have governance systems that are more efficient and conducive to economic growth. How, then, did Europe and Japan come to include not only banks but other “stakeholders,” including employees, in their corporate governance systems? On this question, a more general argument has been made about societal differences in the sequencing of key historical events. Simply put, the U.S. was a country in which big business emerged before big government, whereas the order was reversed in Europe and Japan. Prior to the New Deal, the U.S. had the weakest national government in the developed world, giving business a monopoly on political and institutional power. Attempts to regulate the economy or to harness business in the furtherance of public objectives were strongly resisted by businessmen, up to and including the New Deal era, when the federal government vastly expanded its regulatory efforts and sought to compel the involvement of employees in corporate governance through collective bargaining13. In other industrialized countries, strong, centralized governments existed before the rise of big business. Hence when big business emerged in the late nineteenth and

12 Mark J. Roe, Strong Managers, Weak Owners: The Political Roots of American Corporate Finance (1994). 13 David Vogel, “Why American Businessmen Distrust Their State: The Political Consciousness of American Corporate Managers,” 8 British Journal of Political Science 61(1978); Thomas K. McCraw, “Business and Government: The Origins of the Adversary Relationship,” 26 Management Review 33 (1984). 7 early twentieth centuries, it adapted to the existing reality of a semi-autonomous state deeply involved in national economic development. In France, Germany, Japan and other countries, the state promoted industrialization and wielded extensive regulatory powers that required the mobilization of business in the pursuit of national objectives— economic, military, and social. Social objectives included the maintenance of industrial harmony and stability, as reflected in the German case by social-insurance legislation and by a series of pre-World War I mandatory worker committee laws (1891, 1904, 1908).14 A third approach is to view a developmental state and a bank-centered corporate governance system as mechanisms to boost catch-up industrialization in late-developing countries. (Late developers ostensibly require banks to take the place of underdeveloped financial markets.) A virtue of the late-development argument is that it helps to account for the peculiarities of the British, a country that does not easily accord with the other political explanations. Although Britain had fewer banking restrictions than the United States, it developed neither a bank-centered financing system nor bank-centered governance. And although its government was relatively strong and centralized, it asked less of its businessmen than its European neighbors. Why? Simply put, Britain was the first industrial nation and as such never felt compelled to play the catch-up industrialization game, unlike Germany and Japan. And its early and prolonged industrialization allowed it to commit social “mistakes” that follower countries were determined not to repeat. 15 Recasting political theory in late-development terms creates an opening for the neoclassical argument, however. A relational corporate governance system might have certain advantages in the early stages of industrialization, just as infant-industry tariff protection may have certain advantages over a free-market approach when an industry initially develops. But, so the argument goes, these advantages turn into liabilities when an economy (or industry) matures. In fact, this is precisely the argument that has been made by contemporary critics of Japan, Inc. and Modell Deutschland: that their

14 Gerald Feldman, Army, Industry, and Labor in Germany, 1914-1918 (1966). 15 Peter Mathias, The First Industrial Nation: An Economic History of Britain, 1700-1914 (1969); Alexander Gerschenkron, Economic in Historical Perspective (1962); Ronald Dore, British Factory-Japanese Factory: Origins of Diversity in .Industrial Relations (1973). Also see Gregory Jackson, “The Origins of Nonliberal Corporate Governance in Germany and Japan” (Max Planck Institute, Cologne, Working Paper, 1999). 8

“stakeholder” systems have outlived whatever advantages they may have conferred in the past. But recent work by Neil Fligstein has shown that distinctive features of the U.S. economy – notably, the dominance of the finance-driven, diversified (M-form) corporation—themselves emerged at a relatively late date and did so not in response to efficiency incentives but to regulatory policies such as anti-trust law, up to and including the Celler-Kefauver Act of 1950.16 Regulatory policy in the United States had the unintended consequence of pushing U.S. companies in the direction of unrelated diversification, whereas in Germany and Japan it continued on a prewar trajectory of discouraging mergers in favor of cartels and of promoting corporate growth through internal expansion rather than acquisitions. In other words, modern regulatory policy in the U.S. produced corporations who relied on markets to acquire ideas and talent, whereas in Germany and Japan it produced corporations whose primary emphasis was on production and on the internal generation of ideas through development of human capital and organizational learning. The implications for corporate governance are straightforward: corporations favor shareholders in the U.S. so as to obtain capital for diversification and acquisitions; they favor managers and employees in the Germany and Japan so as to create internal organizational competencies. 17 Here again, as with Roe, characteristic institutions of the U.S. economy—weak banks, diversified companies—are traced back to distinctive American political and regulatory traditions rather than to efficiency incentives. 18 c. Social regulation and social norms Sociologists like Fligstein increasingly are straying across the disciplinary fence to analyze economic institutions. The common theme struck in their writings is first, that

16 Neil Fligstein, The Transformation of Corporate Control (1990). A related argument is made by William Roy: that the dominance of the limited liability joint stock corporation in the U.S. resulted from contingent historical events: anti-trust laws, state corporation laws, and the difficulty of administering regional cartels on a national basis. William Roy, Socializing Capital: The Rise of the Large Industrial Corporation in America (1997). 17 A similar notion can be found in Wendy Carlin and David Soskice, “ Shocks to the System: The German Political Economy under Stress,” 159 National Institute Economic Review 57-76 (1997). 18 The problem with these approaches is that they make regulatory policy exogenous, a construction that many historians would disagree with. See, for example, Martin J. Sklar, The Corporate Reconstruction of American Capitalism, 1890-1916: The Market, Law, and Politics (1988). 9 corporate structures are a product of social and cultural norms as much, if not more than, economic forces; and second, that the distinction between “social” and “economic” is fuzzy given that markets are socially embedded processes varying across time and place. Hence organizational forms are contingent; market forces do not select for the most efficient form. Frank Dobbin’s comparative analysis of railroad industrial policy in Britain, France, and the U.S. in the nineteenth century shows how each nation’s pre-industrial political culture shaped its response to industrialism and its choice of appropriate regulation. In France, a bias toward central bureaucracy led to railroad regulations and a railway system that were centralized and closely coupled to nation-building. In the U.S., that kind of central coordination repeatedly was rejected in favor of local control by towns and states. Toward the end of the century, federal regulation came to play a larger role but its logic was that of bureaucratic minimalism via market competition and a court- centered system of legal enforcement. The argument is consistent with the historical sequencing concept but Dobbin goes beyond it in showing how norms shaped economic institutions over long periods of time. Government officials, financiers, and managers enacted cultural scripts that caused them to interpret common economic “facts” in very different ways. 19 Social norms have also been advanced as an explanation of differering governance structures, as in Roe’s recent work on social democracy and corporate governance.20 Roe argues that European managers give weight to employee interests, even when they do not have to, because they do not wish to violate deeply-held social democratic norms regarding the corporation’s responsibilities to its employees. Roe quotes an American union official involved with the recent Chrysler-Daimler merger, who said, “it is amazing to me that in Europe . . . the corporations feel that they have [an ethical] obligation to their employees . . . Th[is] comes naturally in the European culture.” 21

19 Frank Dobbin, Forging Industrial Policy: The United States, Britain, and France in the Railway Age (1997) 20 Mark Roe, “Political Preconditions to Separating Ownership from Control: The Incompatibility of the American Public Firm with Social Democracy,” 53 Stanford Law Review 539 (2000). 21 Id. at 555. 10

Why this should come naturally in Europe and not so readily in the United States is the same conundrum that led Werner Sombart to ask, over ninety years ago, “Why is there no socialism in the United States?” Since then social scientists have advanced a variety of explanations: the ethnic and racial homogeneity of the European nations, late mass enfranchisement, low rates of social mobility, and a dominant value system that is more collectivist and less individualistic than in the United States. 22 There are problems with this line of argument. Social historians have challenged the claim of American exceptionalism and demonstrated that, despite the absence of a mass working-class party, socialist ideologies did sink deep roots in the American working class. 23 Second, it doesn’t well handle the British situation, since, although Britain fits the continental pattern on social democracy, its corporate governance mechanisms do not. The other country that doesn’t fit the attribution of European governance structures to a social democratic ethos is Japan. Although not a social democracy, Japan’s approach to corporate governance has strong similarities to the stakeholder system of continental Europe. In Japan there is concentrated ownership by banks and other companies; a preference for voice over exit; and expectations (albeit not regulations) that a company should be responsible to employees and suppliers as well as shareholders. What’s important to note is that these practices are not legally required yet are regularly complied with. Perhaps a key factor here is not social democracy so much as the “all in one boat” mentality that prevailed in postwar Japan and Europe as ethnically homogeneous societies with a discredited business class and restive labor movements sought to rebuild their economies along new and more cooperative lines. Moreover, in Japan the Occupation created a postwar legal muddle, which led to various private rules regarding shareholder rights, takeovers, and banks; this is consistent with a line of legal scholarship that views norms as low-cost substitutes for public rules. 24

22 Seymour Martin Lipset, American Exceptionalism: A Double-Edged Sword (1997); Sanford Jacoby, “American Exceptionalism Revisited: The Importance of Management” in Jacoby, ed., Masters to Managers: Historical and Comparative Perspectives on American Employers (1991), 173-200. 23 Sean Wilentz, “Against Exceptionalism: Consciousness and the American labor Movement, 1790-1920,” 26 International Labor & Working Class History 1 (1984). 24 Kathleen A. Thelen, Union of Parts: Labor Politics in Postwar Germany (1991); Andrew Gordon, Wages of Affluence: Labor and Management in Postwar Japan (2001); Curtis J. Milhaupt, “Creative Norm Destruction: The Evolution of Nonlegal Rules in Japanese Corporate Governance” 149 U. Pa. L. Rev 2083 11

II. Signs of Convergence? During the 1990s, as the U.S. stock market boomed and financial markets globalized, there were signs that the “relational” governance systems of continental Europe and Japan were becoming more like the Anglo-American “shareholder” system. Seeking to take advantage of the prestige and the funds associated with U.S. equity markets, a growing number of foreign firms sought listings on the U.S. exchanges. To meet exchange requirements, the firms had to adopt U.S. standards such as having at least two outside directors and providing equal opportunity for all shareholders to participate in tender offers. 25 At the same time, U.S. investors have sought stronger protection for their ever-larger holdings in overseas corporations while U.S. financial companies are trying to build U.S.-style equity markets and an “equity culture” outside the United States. Another force for convergencehas been economic stagnation in Europe and Japan, especially as contrated to strong U.S. economic growth. This has led policymakers in all three countries to conclude that disparities in growth rates are due, in part, to corporate governance factors, and that the remedy is to emulate the U.S. system. Hastening this conclusion has been the spread of neoliberal ideologies that question the current validity of the social compromises and “social market” systems forged at the end of the Second World War. The combined result has been a variety of statutory changes seeking to remold “relational” governance systems more along Anglo-American lines. In Germany, there is currently a proposal to permit negotiated changes in the co-determination system. 26 In Japan, the 1990s saw passage of a plethora of governance-related laws, such as those permitting shareholder class-action suits and stock repurchases by corporations, as well as laws that hastened restructuring by reducing the number of board meetings required to approve mergers and by permitting equity swaps in mergers. Financial deregulation and the growth of foreign ownership are forcing Japanese managers to pay

(2001); Sanford Jacoby, “Risk and the Labor Market: Societal Past as Economic Prologue” in Ivar Berg and Arne Kalleberg, Sourcebook on Labor Markets: Evolving Structures and Processes (2001); Cass Sunstein, “Social Norms and Social Roles,” 96 Columbia Law Review 903 (1996). 25 John C. Coffee, Jr., “The Future as History: Global Convergence in Corporate Governance and Its Implications,” 93 Stanford L.R. 641, 687. 26 See Sadowski, Junkes, and Lindenthal in this Symposium 12 greater attention to shareholders and have induced voluntary changes such as smaller boards of directors.27 Governance reform in Japan and Europe also has domestic origins, such as the aging of the population, which has led to a renewed interest in private pension provision and to pension-fund managers seeking higher returns on their equity investments. Efforts by the European Union to reduce unemployment via early retirement and to harmonize retirement laws across Europe also are leading to growing emphasis on private pension provision. 28 Similar events have occurred in Japan, a society traumatized by the prospect of an aging population. 29 Meanwhile, domestic financial corporations have been pursuing dissolution of corporate cross-shareholding in Germany and Japan. Large German banks and insurance companies have sought greater liquidity in equity markets in the hopes of making the same juicy profits as their U.S. counterparts; they are seeking to trade voice for exit- related returns. 30 Similar events have transpired in Japan, where banks are in the process of liquidating shareholdings, although this has had less to do with new perspectives on governance than a desperate need to raise cash. 31 Finally, both German and Japanese firms are experimenting with heavier reliance on stock options in executive pay, 32 a change intended to satisfy shareholders by aligning their interests with those of management, and also, perhaps, a change intended to salve managements’ envy of highly-paid U.S.executives. 33 But while there is evidence of movement in the direction of a U.S.-type system, change is slow and sometimes illusory; hence there remains considerable distance between the relational and the shareholder governance models. Germany still is an economy with a predominance of concentrated corporate ownership through cross-

27 Jacoby, supra note 4, at 468-69. 28 Mary O’Sullivan, Contests for Corporate Control: Corporate Governance and Economic Performance in the United States and Germany (2000), 262-4, 279. 29 Noriyuki Takayama, The Greying of Japan: An Economic Perspective on Public Pensions (1992). 30 Martin Hopner and Gregory Jackson, “Political Economy of Takeovers in Germany: The Case of Mannesmann and its Implications for Institutional Change” (Working Paper, Max Planck Institute, Cologne, February 2001). . 31 Douglas Ostrom, “The Keiretsu System: Cracking or Crumbling?” 14A JEI Reports, April 7, 2000, at 10. 32 “Ban Lifted on Stock Option Systems”, Japan Labor Bulletin, August 1997, at 2; O’Sullivan, note 28, supra at 285. 33 See O’Connor in this Symposium. 13 holdings, bank holdings, and family ownership. In fact, despite public rhetoric to the contrary, German banks in the 1990s increased the number of 5% blocks they held in leading German companies. 34 While the proportion of equities owned by German households rose over the course of the 1990s, it remains very low by international standards. 35 Similarly, shareholding by institutional investors (insurance companies, pension funds, and investment companies) grew a bit in the 1990s but, as a percentage of GDP it remains at about one-twelfth of the U.S. ratio.36 Despite some well-publicized instances of hostile takeovers—like Krupp’s bid for Thyssen, which was abetted by large banks—these remains at levels far below hostile M&A activity in the U.S. and the U.K. 37 Finally, co-determination and works councils remain deeply embedded in law and in corporate practice.38 The Japanese situation displays a similar mix of change and inertia. The biggest change has been in the main bank system. Until recently, bank shareholding and bank monitoring were important characteristics of the Japanese economy; the percentage of shares held by financial institutions fell modestly from a post-1970 peak of 42 percent in 1988 to 39 percent in 1997, which is where it stood in the early 1980s. Since 1997, however, bank shareholding has unwound rapidly, with banks dumping more than 3 trillion yen of corporate shares during 2000. In addition, there has been a spate of mergers of major Japanese banks. All of this has called into question the continued viability of the main bank system for corporate monitoring. However, banks continue to remain involved in governance. And bank reform by the Koizumi government may yet

34 Lucian A. Bebchuk and Mark J. Roe, “A Theory of Path Dependence in Corporate Governance and Ownership, 52 Stanford Law Review, 775, 793. Also see Ulrich Jurgens and Joachim Rupp, “The German System of Corporate Governance: Characteristics and Changes” (Working Paper, Wissenschaftszentrum (WZB) Berlin, January 2001). 35 O’Sullivan, supra note 28 at 284. 36 Mauro F. Guillen, “Corporate Governance and Globalization: Is There Convergence Across Countries?” (Working paper, The Wharton School, University of Pennsylvania, January 2000), 45. Also see Sigurt Vitols, “The Reconstruction of German Corporate Governance” (Working paper, WZB,. Berlin, June 2000), which argues that German firms espousing “shareholder value” are rare and not performing especially well. 37 Ibid., 49; O’Sullivan, note 28, 285. 38 See Sadowski et al. in this Symposium. 14 lead to a restabilization of main bank monitoring, perhaps with holding companies playing a supplemental role. 39 Japanese corporate cross-shareholding also has fallen but the decline here has been more modest than for bank shareholding. Within the “Big Six” keiretsu, there was more cross-ownership in 1999 than in the late 1980s. 40 As in Germany, while M&A activity has risen, it stands at low levels as compared to the United States and hostile takeovers remain rare. 41 And despite legal reforms that make it easier for companies to utilize stock options or to engage in management buy-outs (MBOs), such practices are atypical, which reinforces the claim that governance practices are held in place by norms and not merely through legal compulsion. 42 A recent study identifies six main features of Japanese-style corporate governance: stable shareholders, an emphasis on steady payment of dividends, heavy reliance on banks for debt financing, internal labor markets for managers, managerial pay cuts to protect employee jobs, and a stakeholder ethos that includes employees as a key bloc. Currently, eighty-two percent of Japanese firms have four or more of these characteristics; only 5 percent have but one or two of these characteristics.. 43 While Japan never went so far as Germany in giving employees a voice inside firms, nearly all large Japanese companies today continue to have enterprise unions that are involved to varying extent in strategic issues such as restructuring and technological change. Although there is more job switching than in the past, it is not common. Most senior managers are “lifers” who, earlier in their career, were enterprise union members and union leaders; there is little American-style animus toward unions. 44 In addition, there are financial mechanisms aligning managers and employees. Managers who safeguard employee jobs are rewarded with higher salaries, while directors’ pay rises and

39 Ostrom, note 31, supra; Carla Koen, “The Japanese Main Bank Model: Evidence of the Pressure for Change” (WZB Working paper, 2000). 40 Hideaki Inoue, “The Accelerating Dissolution of Stock Crossholding,” NLI Research No. 133, March 2000, at 39. 41 “Japan: Land of the Hostile Takeover?” Business Week, April 10, 2000 at 66; 149; Guillen, supra note 36 at 49. 42 “An Exercise in Shareholder Capitalism” Nikkei Weekly, January 31, 2000, 14. 43 Inagami et al., supra note 8, tables 2-1, 2-2., 2-3. 44 Nitta and Nakamura, supra note 8. 15 falls with the size of employee bonuses. 45 In addition to internal corporate norms and incentives, there remain in place strong legal protections against the dissolution of career- type jobs. Recently some large companies like Toshiba and Fujitsu have made sizeable job cuts but these will be instituted over several years via attrition and early retirement; mass firings have not occurred. 46 III. Limits to Convergence What accounts for the persistence of diversity despite globalization and other pressures for convergence? There are several explanations: the institutional rigidities created by path dependence; the absence of a microeconomically superior model of corporate governance; and doubts about the necessity of reforming corporate governance to improve macroeconomic performance. Each of these possibilities is examined below. a. Path dependence Path dependence originally was a concept applied to technological change, the most celebrated, if contested, example being the QWERTY keyboard. Although QWERTY is not the most efficient keyboard available (typist productivity is higher under other layouts), it was the earliest design. Typewriter firms producing this keyboard took ever-larger shares of the market, achieving increasing returns to scale as production expanded. When more efficient alternatives emerged at a later date, few users made the switch due to huge sunk costs in typewriters. They were also held back by the availability of a pool of trained typists, whose existence is a network externality: as more users adopt QWERTY, it creates a benefit to existing users that they do not have to pay for. Thus, contingent initial conditions set a path in motion. Once it is established, there occurs a nexus of related, adaptive innovations and these, along with network externalities, make it prohibitively expensive to switch to another technological path. This gives rise to what is called “lock-in.” But while a given path is locally efficient (i.e., cheaper than nearby alternatives) it is not necessarily globally efficient. There are myriad other examples of

45 Tomohiko Noda, “Determinants of Top Executives’ Promotion and Remuneration” in Toshiaki Tachibanaki, ed., Who Runs Japanese Business? at 36; Katsuyuki Kubo, “The Determinants of Executive Compensation in Japan and the UK” ( working paper, 2000), p. 6. 46 See Araki in this Symposium; “Japan’s Toshiba Plans to Cut 20,000 Jobs by 2004” Wall Street Journal, 28 August 2001. 16 technological lock-in, such as gas-powered automobiles or the Microsoft operating system. 47 Path dependence has been applied to less tangible innovations. For example, concentrated share ownership in Germany has been attributed to events that occurred long ago, such as the bank-based system of corporate finance. Given the environment that prevailed at that time, bank-based financing and ownership made sense. 48 Then it was followed by a series of adaptive innovations—bank-issued securities held in depositary accounts, bank technical departments for monitoring corporations, corporate reputations based on relations with a main bank--that had the effect of making a switch to more dispersed ownership so costly that parties could not change even if they had reason to do so. Also inhibiting switching was the growth of supportive practices, both regulatory and normative (e.g., anti-takeover norms and professional accounting and legal norms regarding bloc holding). Finally there were economies of scale analogous to network externalities: the greater the number of firms with concentrated ownership, the cheaper this method of financing and governance. 49 Complementarities extend to labor markets and labor relations. In Germany and Japan, enterprise unions and co-determination are a good fit with a stakeholder approach to governance that does not privilege shareholders. Moreover, a corporation’s long-term relations with banks, customers, and suppliers facilitate commitments to “permanent” employees. 50 In turn, these commitments promote extensive firm-specific training systems because employees require assurances that the returns on their training investments will be fairly divided and ultimately paid. Treating employees as stakeholders—through participation, representation, and stable employment—provides

47 Paul A. David, “Clio and the Economics of QWERTY,” 75 American Economic Review (May 1985), 332-37; W Brian Arthur, “Increasing Returns and the New World of Business” 74 Harvard Business Review, (1996), 100-109; David A. Kirsch, The Electric Vehicle and the Burden of History (2000). For a critique, see Stan Liebowitz and Stephen Margolis, "Policy and Path Dependence: From QWERTY to Windows 95,” 18 Regulation, (1995), 33-41. 48 Jurgen Kocka, “The Rise of the Modern Industrial Enterprise in Germany” in Alfred D. Chandler, Jr. and Herman Daems, eds., Managerial Hierarchies, supra note 1, 77-116; O’Sullivan, supra note 28, 233-50. 49 See Bebchuk and Roe, supra note 34; Julian Franks and Colin Mayer, “Bank Control, Takeovers, and Corporate Governance in Germany” in Hopt et al., Comparative Corporate Governance, supra note 2, 641- 58. Also see Ronald J. Gilson, “Corporate Governance and Economic Efficiency: When Do Institutions Matter?” 74 Washington University Law Quarterly 327 (1996). 17 such assurances. In the United States, by contrast, employer training investments are much lower than in Germany or Japan, employees are more mobile (especially engineers), and there are few examples of employees having a role in corporate governance except in unionized firms. 51 Or take managerial labor markets. In the U.S., executive compensation based on stock options is, at least in theory, supposed to alleviate the agency problem associated with dispersed ownership. Also, fluid managerial labor markets facilitate contests for control by making it easier for ousted managements to find new jobs after a hostile takeover. In Japan, the contest for control occurs not through hostile bids and replacement of ostensibly ineffectual managers but through career employment and the managerial promotion system. Because managers spend their entire career with a single firm, and because they are carefully evaluated through a kind of tournament system, the winners – who become company directors-- are usually the most dedicated and competent executives at the firm. They are also well known and trusted by the various long-term stakeholders at the company, including banks, customers, suppliers, and employees. These outstanding managers are rewarded with a seat on the company board, although to protect against entrenchment, there is a maximum limit on board tenure and mediocre senior managers serve shorter board terms.52 As these examples suggest, it’s difficult to disentangle the exogenous initial conditions that established a path from the ex post adaptations. The common law legal tradition was exogenous for the United States, whereas for Meiji Japan, legal and other social systems were a matter of choice. 53 Strong craft traditions were exogenous in Germany and caused employers to put a premium on training and participation, whereas

50 Mark J. Roe, “German Codetermination and German Securities Markets” 1 Columbia Business Law Review 167 (1998); Howard Gospel and Andrew Pendleton, “Financial Structure, Corporate Governance, and the Management of Labor” (King’s College, London University, Working Paper 2001). 51 Kazuo Koike, Human Resource Development (1997); John Paul MacDuffie and Thomas A. Kochan, “Do U.S. Firms Invest Less in Human Resources? Training in the World Auto Industry,” 34 Industrial Relations 147 (1995); Carlin and Soskice, supra note 18. 52 Hideshi Itoh, “Japanese Human Resource Management from the Viewpoint of Incentive Theory” in Masahiko Aoki and Ronald Dore, eds., The Japanese Firm: Sources of Competitive Strength (1994), 223- 64; Dore, supra note 4 at 78-80. 53 D. Eleanor Westney, Imitation and Innovation: The Transfer of Western Organizational Patterns in Meiji Japan (1987). 18

U.S. employers made a fateful choice to reject skill-based forms of production. 54 Thus what’s most likely to be the case is that capital markets, labor markets, legal regulations, and corporate norms co-evolved from a set of initial conditions. One thing is clear, however: Given institutional complementarities and path dependence, it’s difficult for one country to borrow a particular practice and expect it to perform similarly when transplanted to a different context. Two examples: First, despite numerous calls for the Japanese to do more in the way of venture capital, the fact is that Japan lacks the fluid labor markets, legal expertise, and equity-related compensation schemes that are the underpinnings of the U.S. venture-capital approach. The Japanese nevertheless do have high rates of innovation, but they achieve it via corporate spin-offs and big-company funding rather than venture capital. 55 Second, were the Japanese or Germans to adopt a U.S.-style corporate governance approach that relies on takeovers to mitigate agency problems, it would prove highly disruptive of managerial incentive and selection systems presently in place. 56 Hostile takeovers also would be disruptive of relations with suppliers and key customers, a substantial portion of which exist on a long- term basis. In Germany and, especially, in Japan, there is less vertical integration of industrial companies than in the United States or the United Kingdom. Rather than rely primarily on arms-length contracts to protect suppliers and purchasers from opportunism, there is heavy use of relational contracting based on personal ties, trust, and reputation. Personal ties are supported by lifetime employment; the business relations are buttressed by cross-share holding. 57 In short, imitation across path-dependent systems is inhibited by the cost of having to change a host of complementary practices that make an institution effective in a particular national system.

54 Michael Piore and Charles Sabel, The Second Industrial Divide: Possibilities for Prosperity (1984). 55 Henry Chesbrough “The Organizational Impact of Technological Change: A Comparative Theory of National Institutional Factors” (Working Paper, Harvard Business School, 1999). 56 T. Hoshi, “Japanese Corporate Governance as a System” in Hopt et al., supra note 2, at 847-876. 57 Mari Sako, Prices, Quality and Trust: Interfirm Relations in Britain and Japan (1992); W. Carl Kester, “Governance, Contracting, and Investment Horizons: A Look at Japan and Germany” in Donald Chew, Studies in International Corporate Finance and Governance Systems (1997), 227-42. In her empirical research, Sako finds that when suppliers do not have to worry about being acquired, it adds to the credibility of the purchaser’s longterm commitments in areas such as purchasing and technology transfer. 19

b. : Costs and Benefits at the Micro Level Anglo-American finance economists are fond of touting the efficiency advantages of shareholder governance; they are convinced of the superiority of institutional arrangements in their home countries. But the fact of the matter is, each corporate governance system has attached to it a complex set of costs and benefits. Accurately toting up and comparing these sets is difficult if not impossible; claims of superiority are wishful thinking. 58 The chief benefit of the Anglo-American system is widely dispersed ownership, which brings with it deep equity markets and lower costs of capital. But this has to be weighed against the defects of dispersed ownership. Foremost is the possibility that managements are not effectively monitored, the old Berle-Means problematic, also known as the agency problem. Although the rise of institutional investors over the past twenty years has fueled hopes that their large stakes give them an incentive to monitor, until now institutional investors have made only limited forays into corporate governance. Rather than acting like long-term investors, institutional investors—under pressure to show strong performance of their pension and mutual funds--tend slightly more often to have short-term horizons and “bubble expectations” that produce distorted investor decision-making (and can have destabilizing effects when bubbles burst). As for shareholders with relatively small holdings in any particular company, they are inclined to rely on exit rather than voice and to push markets in a speculative direction. 59 Another device for solving the agency problem under dispersed ownership is the so-called “market for corporate control”—hostile mergers and acquisitions of underperforming companies. Here there is debate over whether the cure is worse than the

58Michael C. Jensen, “The Modern Industrial Revolution, Exit, and the Failure of Internal Control Systems,” 48 Journal of Finance 831 (1993); Masahiko Aoki, Information, Corporate Governance, and Institutional Diversity (2000), 1-21. 59 Blair, supra note 6 at 145-201; Robert J. Shiller, Irrational Exuberance (2000), 227; Michael Smith, “Shareholder Activism by Institutional Investors” 51 Journal of Finance (1996), 227-52. High stock turnover creates greater volatility and uncertainty in equity prices than exist under a system of more patient capital. If this is the case, and if uncertainty works against long-term projects, then it’s possible that there exists a capital-market bias against longterm investments that would be heightened by the sensitivity of equity prices to near-term information. In fact, the evidence is ambiguous—much of it based on R&D figures that are not a reliable measure of “long term”-- and one can find empirical support on both sides of the issue. For a balanced overview, see Kevin J. Lafferty, “Economic Short-Termism: the Debate, the Unresolved Issues, and the Implications for Management,” 21 Academy of Management Review 825 (1996). 20 disease. While M&A activity in the 1980s positively restructured some overextended conglomerates, it also had a downside. It was associated with reductions in plant and R&D expenditures, showed no net positive effects on productivity, and did not have clear efficiency advantages once its redistributional elements (from incumbent employees to shareholders) were factored into the analysis.60 A third way of solving the agency problem under dispersed ownership is to have an independent board of directors. This is a characteristic feature of Anglo-American governance, one that has been prescribed in recent years for Asian and European companies. Yet research has found that the presence of independent board members does not affect overall firm performance. The reasons for this are not clear—perhaps independent boards are not very independent-- but it should serve as a caution not to overstate the impact—either positive or negative--of governance arrangements on performance. As Bhagat and Black sensibly observe, “board independence may simply not be very important, on average and over time, compared to other factors that influence corporate performance.” 61 What about costs and benefits of relational governance? There surely are instances where relational governance amounts to little more than a kind of crony capitalism in which insiders loot the firm at the expense of minority shareholders, using inside information and self-dealing. But the evidence suggests that this is not the modal tendency. Blockholders with large stakes have strong incentives to carefully monitor management and they have access to high-quality information (often non-quantifiable) that smaller shareholders do not possess. In the United States, when institutional investors do exercise voice and behave like active blockholders, it has a positive effect on performance.62 In Japan, main banks play an additional role in preventing the premature

60 William F. Long and David Ravenscraft, “Decade of Debt: Lessons from LBOs in the 1980s” in Margaret Blair, ed., The Deal Decade (1993); O’Sullivan, supra note 168-9; Andrei Shleifer and Lawrence Summers, “Breach of Trust in Hostile Takeovers” in Alan Auerbach, ed., Corporate Takeovers: Causes and Consequence (1988), 33-67; Robert McGuckin and Sang Nguyen, “On Productivity and Plant Ownership Change: New Evidence from the Longitudinal Research Database” 26 Rand Journal of Economics (1995), 257-76. 61 Sanjai Bhagat and Bernard Black, “The Relationship Between Board Composition and Firm Performance” in Hopt et al., supra note 2.. at 302. 62 David L. Kang, “The Impact of Activist Institutional Investors on Performance in Public Corporations” (Harvard Business School, working paper, January 2000). 21 liquidation of temporarily depressed, but still productive, firms, 63 while in Germany, bank ownership is associated with higher returns on capital. 64 Moreover, the existence of multiple blockholders and the presence of an employee voice in corporate governance help to attenuate private rent-seeking by managers or particular blockholders. Evidence of the agency problem under dispersed ownership can be inferred from the run-up during the 1990s in salaries and options received by U.S. executives, the costs of which are only now beginning to be realized as stockmarkets cool off. 65 German and Japanese executives can only look wistfully at these compensation levels and, perhaps, write op-ed pieces urging reform of the governance systems that restrain their own pay. A more precise measure of how different governance systems handle managerial ineptitude is to examine the likelihood that chief executives will be replaced when corporate performance—earnings or share price—declines. Here, there is no difference in replacement sensitivity between Germany, Japan, and the U.S, which suggests that the three systems produce similar results despite vastly different processes for reaching them. 66 In theory, giving employees a voice in governance—either at the workplace or corporate levels-- should enhance their willingness to invest in firm-specific skills and to share productivity-enhancing ideas with the employer, thereby mitigating bilateral monopoly “hold ups.” Voice should also reduce turnover and turnover-related costs. 67 On the other hand, employee voice may deter the deployment of labor-saving technology

63 Masahiko Aoki and Hugh Patrick, eds., The Japanese Main Bank System (1994). 64 Gary Gorton and Frank Schmid, “Universal Banking and the Performance of German Firms” (NBER working paper 5453, 1996). 65 John Core, Robert Holthausen, and David Larcker, “Corporate Governance, CEO Compensation, and Firm Performance” 51 Journal of Financial Economics 371 (1999). The ratio of CEO earnings to those of the average employee in Fortune 500 firms rose from forty in 1974 to 180 in 1996. Paul Weiler, “A Principled Reshaping of Labor Law for the 21st Century,” 3 Univ. of Penna. Journal of Labor and Employment Law 177 (2001) at 184. 66 Steven Kaplan, “Corporate Governance and Corporate Performance: A Comparison of Germany, Japan, and the U.S.” in Chew, supra note 57 at 251-258. In a recent article, the head of Japan’s main business association, the Keidanren, said that he was “skeptical about claims that American companies respect shareholders’ rights and interests substantially more than Japanese firms do.” Yoshio Nakamura, “Corporate Governance in Japan,” 11 Japan Economic Currents (August 2001), 1. 67 Blair, supra note 6 at 249-59; Oliver Williamson, Markets and Hierarchies: Analysis and Antitrust Implications (1975); John Roberts and Eric van den Steen,”Human Capital and Corporate Governance” in Joachim Schwalbach, ed., Corporate Governance: Essays in Honor of Horst Albach (2001); Richard Freeman and James Medoff, What Do Unions Do? (1984). 22 and bias employment decisions in favor of senior employees. 68 Empirically, employee voice is found to have mixed and ambiguous effects on corporate performance, whether it occurs via statute, as in Germany, via custom, as in Japan, or via employee choice, as in the United States. 69 It matters what size is the organization and at which level employee voice is exercised, with more positive effects found in smaller organizations and at the workplace level (e.g., employee participation, works councils) than in larger organizations and at the strategic level (e.g., board representation). This is hardly surprising, given that employee voice usually is stronger in smaller, localized units. 70 Yet there is a great deal that is unknown about these issues or that we are just beginning to learn about, such as the ability of employee monitoring to deter managerial misfeasance. 71 Given the diverse costs and benefits associated with different governance systems, and given bounded rationality -- the near-impossibility of measuring and toting them up-- it makes no sense to proclaim either relational or shareholder capitalism is microeconomically superior to the other. In light of these ambiguities, it’s not surprising that the tendency toward global convergence has been rather weak. Inertia is consistent with the hypothesis that European and Japanese actors understand that switching costs are high and the likelihood of gain uncertain. c. Macroeconomic Issues and Outcomes While the microeconomic evidence is ambiguous, the macroeconomic evidence, at least recently, implies the superiority of shareholder governance. The strong U.S. economy stands in sharp contrast to the mediocre-to-abysmal performance of the German and Japanese economies over the last decade. There is, however, a problem

68 Jaroslav Vanek, The General Theory of Labor-Managed Market Economies (1970). 69 Giorgio Brunello, “The Effect of Unions on Firm Performance in Japanese Manufacturing,” 45 Industrial & Labor Relations Review 471 (1992); Sadowski, Junkes, and Lindenthal, in this Symposium; Freeman and Medoff, supra note 67, passim. 70 Casey Ichniowski and Kathryn Shaw, “The Effects of Human Resource Management Systems on Economic Performance: An International Comparison of U.S. and Japanese Pklants,” 45 Management Science, 704-21 (1999): Motohiro Morishima, “Use of Joint Consultation Committees by Large Japanese Firms,” 30 British Journal of Industrial Relations 405-23 (1992); William O. Brown, Jr. “Faculty Participation in University Governance and the Effects on University Performance,” 44 Journal of Economic Behavior & Organization 129-43 (2001); John T. Addison, W. Stanley Siebert, Joachim Wagner, and Wei Xiangdong., “Worker Participation and Firm Performance: Evidence from Germany and Britain,” 38 British Journal of Industrial Relations 7-48 (2000). 71 See Kubo, supra note 45. 23 with this line of reasoning: if the links between corporate governance and microeconomic performance are ambiguous, they are even less clear at the macroeconomic level, where the number of causal variables affecting performance is much greater. Hence the presumption that variations in corporate governance can explain variations in macroeconomic performance is heroic, to say the least. The one study to examine this question found no relationship between the dispersion of share ownership—the sine qua non of shareholder governance--and per capita GDP levels.72 In the United States, the association between changes in corporate governance and changes in macroeconomic performance is similarly difficult to discern. If anything, there is a rough temporal correlation between heavy M&A activity and poor performance (in the 1980s and early 1990s), and declining M&A activity and improved economic performance (in the latter 1990s). 73 If it were the case that recent U.S. prosperity was due largely to its corporate governance system, one would expect that countries with similar systems should have enjoyed similar prosperity, or at least more of it than those adhering to the stakeholder model. But this is hardly the case. British growth rates lagged behind Japan’s in the 1970s and 1980s and lagged behind Germany’s in the 1990s. As for British per capita GDP, it remained well below Japanese and German levels through the 1990s. While London today is relatively prosperous, the rest of the country is not. 74 Northern England has a per capita GDP that is 60 percent of London levels, which is why Great Britain is the largest beneficiary of EU structural development funds after Greece, Portugal, and rural Spain. 75 Canada, another adherent of the shareholder model, has per capita income levels that exceed Britain’s, but the country has been experiencing economic stagnation throughout the 1990s. 76

72 Rafael La Porta, Florencio Lopez de Silanes, Andrei Shleifer, and Robert Vishny, “Law and Finance,” 106 Journal of Political Economy (1988), 1113-55. 73 On trends in M&A activity (but one that reaches a more sanguine conclusion about their economic consequences), see Bengt Holmstrom and Steven N. Kaplan, “Corporate Governance and Merger Activity in the United States: Making Sense of the 1980s and 1990s” 15 Journal of Economic Perspectives (2001), 121-44. 74 “A British Miracle?” , March 25-31, 2000, at 57. 75 Tony Judt in New York Review of Books, July 19, 2001. 76 Richard Freeman, “The U.S. Economic Model at Y2K: Lodestar for Advanced Capitalism?” 26 Canadian Public Policy (2000), S187-S200. 24

To put the shoe on the other foot, it’s unlikely that corporate governance was a major factor—if it was a factor at all—behind the sluggish growth rates observed in Germany and Japan in the 1990s. During the 1970s and 1980s, these countries turned in admirable economic results, despite their governance models. What, then, if not shortcomings in corporate governance, can account for recent economic problems in Germany and Japan? In both countries, macroeconomic factors loom large. Germany spent heavily on the absorption of the East while its economy was restrained by high interest rates associated with this spending and with European unification. A major factor in Japan’s stagnation has been the government’s macro-mismanagement, including its failure to solve post-bubble banking problems that have caused Japanese lending to fall to its lowest level in nine years. 77 Now that the U.S. economy has begun to slow and equity prices have tumbled, it’s becoming clearer that rapid growth in the late 1990s was not caused by a “new economy” but instead by a one-time massive investment in high technology, which, due to excess capacity, is unlikely to be duplicated in the near future. For a while, the party was self-sustaining: investment pushed up equity prices, and those high prices buoyed investment both in existing businesses and in new firms (venture capital). When the bubble burst, however, the whole cycle went into reverse, and now falling share prices (and overcapacity) are depressing investment, productivity, and new ventures. 78 Another factor that buoyed U.S. growth in the late 1990s were the juicy profits that U.S. companies reaped from their position as first-movers in the electronics and information technology industries. But while first-movers sometimes enjoy long-lasting rents—if increasing returns deter new entrants--this is hardly the norm. In fact, since the 1950s there have been several other periods when the U.S. seemed to have a first-mover advantage only to see it dissipated by overseas entrants. In the electronics industry, this

77 “Japanese banks: Fiddling while Marunouchi burns,” Economist, Jan 27, 2001, 67-69. For an optimistic take on the German—and other social democratic—economies, one that supports the economic viability of this variety of capitalism, see Geoffrey Garrett, Partisan Politics in the Global Economy (1998) 78 Productivity growth slowed from a 2.5 percent annual growth rate between 1996 and 2000 to 1.6 percent in the first half of 2001, which is only slightly higher than the 1.4 percent rate from 1973 to 1995. “A Spanner in the Productivity Miracle,” Economist, Aug. 11, 2001, p. 55. 25 happened in transistors, consumer applications of integrated circuits, and, for a while at least, in semiconductors. It may yet happen again. 79 d. Multiple Equilibria Another way of understanding these issues is through the concept of multiple equilibria—that there are is no one-best way of designing institutions to support growth in advanced industrial economies. Multiple equilibria can arise and persist due to path dependence, institutional complementarities, bounded rationality, and comparative advantage. Sometimes multiple equilibria involve functionally similar but operationally distinctive institutions, such as the use of big firms as incubators in Japan versus the U.S. approach of incubation via start-ups and venture capital. Other times different institutions create qualitatively different outcomes. That is, a set of institutions, including those of corporate governance, may be better at facilitating certain kinds of business strategies and not others. Companies—and the countries in which they are embedded— can then secure international markets by specializing in those advantageous business strategies because foreign competitors will have difficulty imitating them. For example, the emphasis on specific human capital in German and Japan is supportive of production- based technological learning, incremental innovation, and high quality production, all areas in which those economies have specialized. By contrast, the U.S. emphasis on resource mobility and on high short-term rewards directs resources to big-bang technological breakthroughs. In short, there are substantial gains to be reaped from sustaining institutional diversity and competing internationally on that basis. 80 None of this is to deny that institutional imitation is possible and sometimes can be advantageous. It’s difficult to say, a priori, what are these circumstances. We can say with greater certainty, however, that the existence of multiple equilibria means that imitation will not be unidirectional but instead will occur along multiple vectors. Pundits have been obsessed with exhorting and documenting the spread of Anglo-American practices to Europe and Japan. What has been less well observed is the movement of U.S. companies in the opposite direction. For example, institutional ownership has begun

79 Leonard H. Lynn, “Technology Competition Policies and the Semiconductor Industries of Japan and the U.S.: A Fifty-Year Retrospective,” 47 IEEE Transactions (2000), 200-210. 26 to create forms of relational investing which, although they have not yet born distinctive fruit, may over time lead to more voice and less exit on the part of U.S. shareholders. Within firms, nonunion U.S. employees have somewhat more influence today due to such phenomena as the spread of ESOPs and other equity-based forms of compensation; the growing importance of intellectual capital; and the adoption of high-performance work practices that are sustained by employee voice at the workplace. 81 Of course, these gains have to be counterbalanced against the loss of voice that has occurred due to shrinking private-sector unions and a stronger shareholder-oriented ethos among senior management. e. Limits to Globalization Finally, it’s important to realize that, despite the shift to more globalized investment and trade, it is not the case that Anglo-American companies are dominating the global market such that other countries must conform to the Anglo-American model. In fact, the Anglo-American share of the world’s outward foreign investment is falling, from 66 percent in 1980 to just over 50 percent in 1997, while French, German, and Japanese investment rose sharply during this period.82 Moreover, the world’s largest economies still produce and consume much of their output domestically. As Robert Wade observes, “the world economy is more inter- national than global. In the bigger national economies, more than 80 percent of production is for domestic consumption and more than 80 percent of investment by domestic investors.” 83 Even multinational companies still have home bases in particular countries. In other words, globalization is not a fait accompli; it is regularly contending with centripetal tendencies in the world economy. The latter tendency may be enhanced in the future by more vigorous regionalization in Europe and the Americas as a result of EU, NAFTA, and similar arrangements. IV. Conclusions

80 Aoki supra note 58 at 132. 81 Blair supra note 6 at 158-59; 172-89; O’Connor in this Symposium. 82 Guillen, supra note 36, table 2. 83 Robert Wade, “Globalization and Its Limits: Reports of the Death of the National Economy are Greatly Exaggerated” in Suzanne Berger and Ronald Dore, eds., National Diversity and Global Capitalism (1996), p. 61. 27

Much of the discussion about corporate governance is framed in efficiency terms, as if one could measure national institutions in a comparative statics framework. Yet this framework is inadequate for understanding the dynamic properties of governance systems, especially those related to innovation and long-term growth. 84 Moreover, when there are multiple equilibria and bounded rationality regarding what constitutes an institutional optimum, we are operating in the world of the second best.85 In that world, there is no reason to believe that revamping a governance system will necessarily move an economy closer to an economic optimum. The economic case for the superiority of Anglo-American governance —and of the Anglo-American version of “free markets” as we know them, as opposed to a theoretical ideal—is actually rather weak.86 Another problem with the static approach to corporate governance is that it gives short shrift to normative issues, which bulk large when we consider the role that employees might play in corporate governance. What should be the purpose of the corporation? Is it a community of human beings, a nexus of contracts, or the possession of its shareholders? Do corporations incur social obligations in return for the privileges that society grants to them? What constitutes a fair distribution of rewards from economic activity? Too often questions like these are ignored or treated as second-order concerns. The fairness question is of immediate relevance to the Anglo-American system, where the assumption traditionally has been made that suppliers of capital get control rights in return for their bearing of residual risk. But, as Margaret Blair and others have been at pains to point out, employees are an important (though imperfectly measured) source of capital to the enterprise. 87 And, while shareholders are protected through limited liability and portfolio diversification, employees have neither of these risk- reducing shields. So a normative case can be made that shareholders are getting more than their due and employees less.

84 Mary O’Sullivan, “The Innovative Enterprise and Corporate Governance” Cambridge J. of Economics 343 (2000). 85 Ronald Lipsey and Kelvin Lancaster, “The General Theory of Second Best," 24 Review of Economic Studies 11 (1956). 86 Charles E. Lindblom, The Market System: What It Is, How it Works, What to Make of It (2001); Garrett., supra note 77. 87 Margaret M. Blair & Lynn A. Stout, “A Team Production Theory of Corporate Law,” 85 U. Va. L. Rev. 247 (1999). 28

This is not to say that we lack economic rationales for involving employees in corporate governance. As we have seen, many types of employee voice are associated with better firm performance, which is hardly surprising given that employees possess inside information about inefficient processes and ineffectual managers, and are at least as motivated as shareholders to see that the firm is run effectively. 88 Moreover, voice creates habits of responsibility that buttress citizenship and other civic values associated with economic growth. 89 Voice and civic values are associated with equity in the distribution of economic rewards, and equity, in turn, is associated with higher long-term growth rates. 90 Today, European and Japanese systems of corporate governance are under attack for their failure to conform to Anglo-American practices. The strongest criticism comes from financial corporations eager to propagate an equity culture and associated commissions. There is also a conservative political critique rooted in the belief that unemployment is due to corporations giving excessive weight to employee interests. At this time it does not appear likely that faddish predictions of convergence will come true. The notion that relational forms of governance will give way to shareholder capitalism is yet another case of wishful thinking. On the other hand, it’s also unlikely that the Anglo-American system will move very far in the direction of stakeholder capitalism. The same forces that hold other systems in check—path dependence, institutional complementarities, and wariness about costs and benefits—will likely keep U.S. workers in the position of being among the least influential group of employees in the advanced industrial world. Lest this sound excessively deterministic and pessimistic, keep in mind that path dependence is not an iron law of evolution. There are moments when sudden shifts can and do occur, causing the “punctuation” of equilibria and movement onto a new evolutionary path. It is not well understood what triggers these changes. 91 So, like

88 Chris Doucouliagos, “Worker Participation and Productivity in Labor-Managed and Participatory Capitalist Firms: A Meta-Analysis,” 49 Industrial & Labor Relations Review 58 (1995). 89 Robert D. Putnam et al., Making Democracy Work (1994). 90 Phiippe Aghion et al., “Inequality and Economic Growth,” 37 Journal of Economic Literature 1615 (1999); Alberto Alesina and Dani Rodrik, “Distributive Politics and Economic Growth” 109 Quarterly Journal of Economics 465 (1994). 91 Jerry A. Coyne et al., “Mechanisms of Punctuated Evolution,” 274 Science (Dec. 6, 1996), 1748-51. 29

Pascal’s Wager, it’s worth working to promote employee voice inside the corporation, in the hope that the U.S. environment someday may turn more favorable to these efforts.