6 The Yale Law Journal [Vol.11:1 Thus,the control holder will sell only a minority interest or will sell contro as a block,but will not break up its control block-and hence concentrated ownership will persist." Similarly,Mark Roe has offered an entirely independent"political" yfor why strorket cor ent with the European political tradition of "social democracy." pressure corporate managers to forego opportunities for profit maximization in order to maintain high employment.Under circumstances that would ead firms in other political environme nents to downsize their ope rations because of adverse market conditions,firms in social democracies,he argues,are compelled to expend their shareholders'capital in order to subsidize other constituencies.Public firms are relatively more exposed than s he helie tothe higher managerial agency cost that As a result, concentrated ownership is a defensive reaction to these pressures;through nontransparent accounting. hidden reserves,and direct supervision of management,large blockholders, he claims,can better resist these political pressures to expend the firm's urces on other constituencies In overview,a common denominator runs through the theories of LLS&V,Bebchuk,and Roe:Ownership and control cannot easily separate when managerial agency costs are high.Although they disagree about the causes of high agency costs ie weak legal stand ards ve sus political pressures that cause firms sometimes to subordinate the interests c shareholders-they implicitly concur that the emergence of deep,liquid markets requires that the agency cost problem first be adequately resolved by state action. reholder's m ve fo e a highe um than public shareholders i their To the extent that they can solv coon ers may be able to match the um that the laree sh holder will aacepundeicd nd may not may obtain a higher price from the public market.as they long have on the Nasdaq(including pation of gainsnoraaibe n pay a Stndiesofinitialpbicofienings s)in conce of the kholder tainin control one es and controlled on av 68.5 of the yoting MEN abstr REV.5392000 Separating Ownership from Conirol,53 STAN.L Imaged with the Permission of Yale Law Journal
The Yale Law Journal Thus, the control holder will sell only a minority interest or will sell control as a block, but will not break up its control block-and hence concentrated ownership will persist."' Similarly, Mark Roe has offered an entirely independent "political" theory for why strong securities markets are inconsistent with the European political tradition of "social democracy." "2 In his view, social democracies pressure corporate managers to forego opportunities for profit maximization in order to maintain high employment. Under circumstances that would lead firms in other political environments to downsize their operations because of adverse market conditions, firms in social democracies, he argues, are compelled to expend their shareholders' capital in order to subsidize other constituencies. Public firms are relatively more exposed than private firms, he believes, to the higher managerial agency costs that social democracies impose. As a result, concentrated ownership is a defensive reaction to these pressures; through nontransparent accounting, hidden reserves, and direct supervision of management, large blockholders, he claims, can better resist these political pressures to expend the firm's resources on other constituencies. In overview, a common denominator runs through the theories of LLS&V, Bebchuk, and Roe: Ownership and control cannot easily separate when managerial agency costs are high. Although they disagree about the causes of high agency costs-i.e., weak legal standards versus political pressures that cause firms sometimes to subordinate the interests of shareholders-they implicitly concur that the emergence of deep, liquid markets requires that the agency cost problem first be adequately resolved by state action. controlling shareholder's motive for paying a higher control premium than public shareholders is matched by their expected loss. To the extent that they can solve the coordination costs in organizing to protect themselves from a future controlling shareholder who will divest them of control, public shareholders may be able to match the premium that the large shareholder will pay for control. Second, in the case of high-risk investments, the public market affords investors the benefits of diversification, while the incoming controlling shareholder (or any large blockholder) must accept undiversified risk (and may not be willing to do so or may discount the price it offers to reflect this risk). Although this point about undiversified risk suggests that high-tech companies may obtain a higher price from the public market, as they long have on the Nasdaq (including many foreign issuers), it does not deny that the corporate controlling shareholder may often pay a higher premium in anticipation of synergy gains not available to portfolio or retail investors. 11. Studies of initial public offerings (IPOs) in concentrated securities markets have tended to confirm this prediction: 1POs seldom distribute more than a minority of the firm's voting shares to the market, with the controlling blockholder generally retaining control. For example, one recent study of Swedish IPOs finds that in close to 90% of all privately controlled IPOs, the controlling owner did not sell shares and controlled on average 68.5% of the voting power after the IPO. PETER HOGFELDT & MARTIN HOLMEN, A LAW AND FINANCE ANALYSIS OF INITIAL PUBLIC OFFERINGS 3-4 (SSRN Elec. Library, Working Paper No. 236,042, 2000), available at http://papers.ssrn.com/paper.tafabstractid=236042. 12. Mark J. Roe, Political Preconditions to Separating Ownership from Control, 53 STAN. L. REv. 539 (2000). Imaged with the Permission of Yale Law Journal [Vol. 111: 1
2001] The Rise of Dispersed Ownership This Article dissents.Although it does not doubt that"law matters,"it finds that a transition toward dispersed ownership is already well advanced and seems likely to continue,even in the short-term absence of legal change.Part I surveys this evidence,which reveals increas g signs within the world of wership.Despite th asserted barriers,securities markets are growing across Europe at an extraordinary rate,entrepreneurs in civil-law countries are making use of IPOs at a rate equivalent to that in the common-law world,and the market for corporate control has become truly in rnational.Som thing is destabilizing the old equilibriu but how far it will progress remains an open question. Part II then analyzes the claim that securities markets require a strong legal foundation that protects the minority shareholder in order to become deep or liquid.Although the association between minor protection an liquidity se ms real,Part II will argue that the cause and effect sequence is backwards.Much historical evidence suggests that legal developments have tended to follow,rather than precede,economic change.3 Specifically,Part II will examine the early development of the New York Stock Exchange (NYSE)and the London Exchange (LSE),and contrast experiences with the arrested development of equity securities markets in France and Germany over the same period.Although securities exchanges have existed since the seventeenth century,exchanges primarily traded debt securities up until the mid-nineteenth century.The over a relatively brie period and time when the private benefits of control were unquestionably high,dispersed ownership arose in both the United States and the United Kingdom-largely in the absence of strong legal protections for minority sharcholders,which came afterward.Viewed in re rospect,this beheof mivated sonstimuency that will be se makes obvio s political sense: Legal reforms are enacte ed at the perceives that it will be protected)by the proposed reforms.Hence,the constituency (here,dispersed public shareholders)must first arise before it can become an effective lobby ing force and an in of legal change But do liquid markets develop if minority shareholders are systematically exposed to expropriation by controlling shareholders because of inadequate legal protections (as LLS&V conclude they are xposed)?A problem with much recent law the na ural pre ted ownership has been its ahistorica 13.Stuart Banner has made the in m that,ovr the ast300 yers.most major auses new seo and eventual c ane the this findi not there are Imaged with the Permission of Yale Law Journal
The Rise of Dispersed Ownership This Article dissents. Although it does not doubt that "law matters," it finds that a transition toward dispersed ownership is already well advanced and seems likely to continue, even in the short-term absence of legal change. Part I surveys this evidence, which reveals increasing signs of fission within the world of concentrated ownership. Despite the asserted barriers, securities markets are growing across Europe at an extraordinary rate, entrepreneurs in civil-law countries are making use of IPOs at a rate equivalent to that in the common-law world, and the market for corporate control has become truly international. Something is destabilizing the old equilibrium, but how far it will progress remains an open question. Part II then analyzes the claim that securities markets require a strong legal foundation that protects the minority shareholder in order to become deep or liquid. Although the association between minority protection and liquidity seems real, Part II will argue that the cause and effect sequence is backwards. Much historical evidence suggests that legal developments have tended to follow, rather than precede, economic change.13 Specifically, Part II will examine the early development of the New York Stock Exchange (NYSE) and the London Stock Exchange (LSE), and contrast their experiences with the arrested development of equity securities markets in France and Germany over the same period. Although securities exchanges have existed since the seventeenth century, exchanges primarily traded debt securities up until the mid-nineteenth century. Then, over a relatively brief period and at a time when the private benefits of control were unquestionably high, dispersed ownership arose in both the United States and the United Kingdom-largely in the absence of strong legal protections for minority shareholders, which came afterward. Viewed in retrospect, this sequence makes obvious political sense: Legal reforms are enacted at the behest of a motivated constituency that will be protected (or at least perceives that it will be protected) by the proposed reforms. Hence, the constituency (here, dispersed public shareholders) must first arise before it can become an effective lobbying force and an instrument of legal change. But how do liquid markets develop if minority shareholders are systematically exposed to expropriation by controlling shareholders because of inadequate legal protections (as LLS&V conclude they are exposed)? A problem with much recent law and economics commentary on the natural predominance of concentrated ownership has been its ahistorical 13. Stuart Banner has made the interesting argument that, over the last 300 years, most major waves of securities regulation have followed a sustained price collapse on the securities market. Stuart Banner, What Causes New Securities Regulation?: 300 Years of Evidence, 75 WASH. U. L.Q. 849, 850 (1997). It is not surprising that "bubbles" and eventual crashes produce victims and hence a political demand for reform. But perhaps the deeper meaning of this finding is that the reform of securities regulation has not been associated with any broader political movement. Thus, this evidence is in tension with Professor Roe's claim that there are "political preconditions" to the growth of securities markets. See Roe, supra note 12. Imaged with the Permission of Yale Law Journal 2001]
The Yale Law Journal [Vol.11l:1 character.A closer look at the experience of U.S.corporations in the late nineteenth century shows that,even in the absence of adequate minority protections and even in the presence of high private benefits of control. private actors could bond themselves in ways that credibly signaled to the minority shareholders that they would not be exploited.Both through such bonding measures and through self-regulation,as implemented by the NYSE,investors were assured that their investments would neither be expropriated by the firm's founders nor,once ownership had become dispersed ct to low-premit m takeover by an incc seeker.That the United States led the way toward dispersed ownership seems best explained not by the state of its nineteenth-century corporate law,but by a more basic fact:As a debtor nation facing the need to develop highly ital-intensive industries (e.g. railroads, steel,and electrical power),the United States was more dependent upon forcign capital,and i had to strive harder to convince remote foreign investors of the adequacy of the safeguards taken to protect their investments Dispersed ownership did not,however,arrive in France or Germany even though the Paris Bo se was the leading intern riva I to the I SE during the last quarter of the nineteenth century.Why not?The "politica thesis"offered by Roe clearly cannot explain the failure of securities markets to develop in france and Germany during the late nineteenth entury,because neither cou ntry ached being social democracy in this era.A possible explanation could be that offered by LL&,namely that French and German law provided insufficient protections for minority shareholders.But the LLS&V explanation has a serious problem:The specific "anti-director''rights that they identify as the central factors distinguishing common-law m civil-law sys ems strike many 1leg野 commentators as only tangentially related to effective legal protection for minority shareholders.The possibility thus surfaces that the observed legal differences identified by LlS&V may serve as a proxy for something de What,then,is the hidden variable that at least historically distinguished common-law from civil-law systems?Part III suggests that the principal variable accounting for the earlier development of dispersed ownership in the United States and the United Kingdom than in Continental Europe was 4.Fo ck.the leadi German state the nir nth.is chara trans.1995):see also infra notes 208-210 and ac lext (discussing this ilier.the first majorir ent bank Imaged with the Permission of Yale Law Journal
The Yale Law Journal character. A closer look at the experience of U.S. corporations in the late nineteenth century shows that, even in the absence of adequate minority protections and even in the presence of high private benefits of control, private actors could bond themselves in ways that credibly signaled to the minority shareholders that they would not be exploited. Both through such bonding measures and through self-regulation, as implemented by the NYSE, investors were assured that their investments would neither be expropriated by the firm's founders nor, once ownership had become dispersed, subjected to a low-premium takeover by an incoming control seeker. That the United States led the way toward dispersed ownership seems best explained not by the state of its nineteenth-century corporate law, but by a more basic fact: As a debtor nation facing the need to develop highly capital-intensive industries (e.g., railroads, steel, and electrical power), the United States was more dependent upon foreign capital, and it had to strive harder to convince remote foreign investors of the adequacy of the safeguards taken to protect their investments. Dispersed ownership did not, however, arrive in France or Germany, even though the Paris Bourse was the leading international rival to the LSE during the last quarter of the nineteenth century. Why not? The "political thesis" offered by Roe clearly cannot explain the failure of securities markets to develop in France and Germany during the late nineteenth century, because neither country approached being a social democracy in this era. 4 A possible explanation could be that offered by LLS&V, namely, that French and German law provided insufficient protections for minority shareholders. But the LLS&V explanation has a serious problem: The specific "anti-director" rights that they identify as the central factors distinguishing common-law from civil-law systems strike many legal commentators as only tangentially related to effective legal protection for minority shareholders. 5 The possibility thus surfaces that the observed legal differences identified by LLS&V may serve as a proxy for something deeper. What, then, is the hidden variable that at least historically distinguished common-law from civil-law systems? Part III suggests that the principal variable accounting for the earlier development of dispersed ownership in the United States and the United Kingdom than in Continental Europe was 14. For example, no matter how Prince Otto von Bismarck, the leading German statesman and politician of the last half of the nineteenth century, is characterized, he was not a social democrat. See WOLFGANG J. MOMMSEN, IMPERIAL GERMANY, 1867-1918 (Richard Deveson trans., 1995); see also infra notes 208-210 and accompanying text (discussing this period). Correspondingly, the dominant figure behind French efforts to develop a system of international investment banking in the late nineteenth century was Napoleon III, who was the sponsor of Crddit Mobilier, the first major investment bank organized on a corporate basis. See infra notes 160-162 and accompanying text. His motives were, however, largely statist, rather than economic. 15. For a description of LLS&V's "anti-director" rights, see supra note 6. Imaged with the Permission of Yale Law Journal [Vol. I11 : 1
2001 The Rise of Dispersed Ownership 9 the early separation of the private sector in the common-law world from the close supervision and control of the central government.In the absence of direct governmental regulation,relatively strong systems of self-regulation arose in the United States and the United Kingdom which were administered by private bodies(most notably,private stock exchanges)that sought to regulate their members'conduct in their mutual self-interest. Although these exchanges may not have been optimal regulators,they were at least entrepreneurial entities that adapted quickly to ne nditions and opportunities.Fran e and,to a lesser extent,in Germany,the state intervened constantly in the market,sometimes to protect it and sometimes to chill it,but the degree of paternalistic supervision that was imposed froze the development of Continental markets and left little room for enlightened self-r Viewed in this light,the critical role of law in the separation of ownership and control was not that it fostered minority sharcholders (in common-law countries)or abandoned them (in civil-law countries),but rather that the common-law world was,for a variety of reas mor spitable than the civil-law world to private self-regulatory institutions. the sommon a hasa more decentralized character that co private law-making,while the civil law tends to be more centralized and hostile to private law-making,this difference transcends the field of comparative law and has conte mporary relevance for planners and regulators in transitional economies.As will be stressed,it suggests that private action,through bonding and signaling measures,may be the critical first step toward stronger securities markets. This proposd which deemphasizes theo of forma law grees ith LLS&V that i not coincidental that liquid equity securities markets arose in the United States and the United Kingdom,but not in France or Germany,but disagrees with them that the key explanatory variable was the impact of the substantive law on shareholder right Because this interp tion fo es less on substantive law,and more on th structure for lawmaking within the broader society,it is not confounded by the special case of the Netherlands,where securities markets first arose in Amsterdam well ahead of London.Although the Netherlands is a civil-law the citic fact xp of markets there in this Article' pluralistic,decentralized society in which the private sector was relatively ORIGINS THE SH INDIVIDUALISM(1978)[her MACFARLAN Imaged with the permission of Yale law lournal
The Rise of Dispersed Ownership the early separation of the private sector in the common-law world from the close supervision and control of the central government. In the absence of direct governmental regulation, relatively strong systems of self-regulation arose in the United States and the United Kingdom, which were administered by private bodies (most notably, private stock exchanges) that sought to regulate their members' conduct in their mutual self-interest. Although these exchanges may not have been optimal regulators, they were at least entrepreneurial entities that adapted quickly to new conditions and opportunities. In contrast, in France and, to a lesser extent, in Germany, the state intervened constantly in the market, sometimes to protect it and sometimes to chill it, but the degree of paternalistic supervision that was imposed froze the development of Continental markets and left little room for enlightened self-regulation. Viewed in this light, the critical role of law in the separation of ownership and control was not that it fostered minority shareholders (in common-law countries) or abandoned them (in civil-law countries), but rather that the common-law world was, for a variety of reasons, more hospitable than the civil-law world to private self-regulatory institutions. 6 If the common law has a more decentralized character that encourages private law-making, while the civil law tends to be more centralized and hostile to private law-making, this difference transcends the field of comparative law and has contemporary relevance for planners and regulators in transitional economies. As will be stressed, it suggests that private action, through bonding and signaling measures, may be the critical first step toward stronger securities markets. This proposed interpretation, which deemphasizes the role of formal law, agrees with LLS&V that it was not coincidental that liquid equity securities markets arose in the United States and the United Kingdom, but not in France or Germany, but disagrees with them that the key explanatory variable was the impact of the substantive law on shareholder rights. Because this interpretation focuses less on substantive law, and more on the structure for lawmaking within the broader society, it is not confounded by the special case of the Netherlands, where securities markets first arose in Amsterdam well ahead of London. Although the Netherlands is a civil-law country, the critical fact explaining the early appearance of securities markets there in this Article's view was that it was, much like England, a pluralistic, decentralized society in which the private sector was relatively 16. This thesis that decentralization encouraged economic growth has been developed on a grander scale by the British historian and anthropologist Alan Macfarlane. See generally ALAN MACFARLANE, THE ORIGINS OF ENGLISH INDIVIDUALISM (1978) [hereinafter MACFARLANE, ORIGINS]; ALAN MACFARLANE, THE RIDDLE OF THE MODERN WORLD (2000) [hereinafter MACFARLANE, RIDDLE]; infra notes 225-229 and accompanying text. Imaged with the Permission of Yale Law Journal 2001]
10 The Yale law lournal [Vol.111:1 autonomous and free from direct state supervision.Moreover,if legal protections of minority shareholders were the indispensable precondi for the growth of securities markets,as LLS&V posit,the successful U.S experience would seem inexplicable.As will be seen,in the late nineteenth cen tury,U.S.law was characterized by a high level of judicial corruption was demonstrably vulnerable to regulatory arbitrage (as participan corporate control battles regularly played one court and one state off against another),and wholly lacked any federal law on securities regulation.Given that the private benefits of control were high and realistic minority protections were weak,the LLS&V model would edict that disperse ownership could not arise in such an environment.But it did That dispersed ownership was able to arise in this era derived in large measure from the ability of private actors to develop functional substitutes for formal law.Ove time systems of secu rities regulation in the United States and the United Kingdom functionally converged.Only late did legislative changes bring about formal convergence.That functiona convergence should precede formal convergence is even more predictable in a rapidly globali world in which increasingly international capital and product markets el fi adapt and penalize those firms that have a higher cost of capital Thus Part III predicts that functional convergence may be the principal mechanism by which the separation of ownership and control will come both to Europe and,more slowly,to transit onal economies Specifically, t suggests that some recent developments in Russian corporate governa are functional parallels to the bonding and signaling devices used in the United States in the 1870s and 1880s.and that some European stock exchanges are beginning to show today the same activism that the NYSE int that the United Kine and the Neth nds had a similar social lega John C.Coffee. smultiple jurisdictions on legal rul and prac nce can also be ach ved as the result of private actio such as b 1es0 on the NYSEin United Srates market (that is. vate class actio )Such bon d appears to Kraakman,supra note3. Imaged with the Permission of Yale Law Journal
The Yale Law Journal autonomous and free from direct state supervision.' 7 Moreover, if legal protections of minority shareholders were the indispensable precondition for the growth of securities markets, as LLS&V posit, the successful U.S. experience would seem inexplicable. As will be seen, in the late nineteenth century, U.S. law was characterized by a high level of judicial corruption, was demonstrably vulnerable to regulatory arbitrage (as participants in corporate control battles regularly played one court and one state off against another), and wholly lacked any federal law on securities regulation. Given that the private benefits of control were high and realistic minority protections were weak, the LLS&V model would predict that dispersed ownership could not arise in such an environment. But it did. That dispersed ownership was able to arise in this era derived in large measure from the ability of private actors to develop functional substitutes for formal law.'" Over time, the systems of securities regulation in the United States and the United Kingdom functionally converged. Only later did legislative changes bring about formal convergence. That functional convergence should precede formal convergence is even more predictable in a rapidly globalizing world in which competitive pressures in the increasingly international capital and product markets compel firms to adapt and penalize those firms that have a higher cost of capital.' 9 Thus, Part III predicts that functional convergence may be the principal mechanism by which the separation of ownership and control will come both to Europe and, more slowly, to transitional economies. Specifically, it suggests that some recent developments in Russian corporate governance are functional parallels to the bonding and signaling devices used in the United States in the 1870s and 1880s, and that some European stock exchanges are beginning to show today the same activism that the NYSE 17. This same point that the United Kingdom and the Netherlands had a similar social structure, but different legal origins, has been well made by Alan Macfarlane. See MACFARLANE, RIDDLE, supra note 16, at 279-80. 18. In earlier work, I distinguished "formal convergence" from "functional convergence." John C. Coffee, Jr., The Future as History: The Prospects for Global Convergence in Corporate Governance and Its Implications, 93 Nw. U. L. REV. 641, 657 (1999). Formal convergence requires multiple jurisdictions to enact common legal rules and practices. Functional convergence can arise, however, because of the use of functional substitutes that look dissimilar but have equivalent effects. Functional convergence can also be achieved as the result of private actions, such as bonding devices or related actions that deliberately limit managerial discretion. For example, a firm in a country with weak legal rules and disclosure standards might deliberately list on the NYSE in order to subject itself voluntarily to its higher disclosure, accounting, and market transparency standards and to the enforcement mechanisms that apply to firms that enter the United States market (that is, private class actions and SEC enforcement). Such bonding through cross-listing on a foreign exchange has recently become common and appears to increase the firm's stock price. Id. at 673-75. 19. Professors Hansmann and Kraakman emphasize this point at some length, arguing that as a result a norm of shareholder primacy is becoming dominant worldwide. Hansmann & Kraakman, supra note 3. Imaged with the Permission of Yale Law Journal [Vol. I111: 1