Equity Ownership and Market Socialism

The problems with exploitation identified in chapters 6 and 7 raise the ques­tion of whether there are forms of market socialism that would not face these difficulties. The purpose of this section and the next is to address this ques­tion in a systematic way.

One of the most serious weaknesses of the worker control-state owner­ship model is that the capital provider is not the ultimate decision-making authority and residual claimant. By contrast, one of the salient advantages of the characteristic organizations of a free enterprise system is that these three functions are all joined. In other words, the latter organizations, unlike the organizations in the worker control-state ownership model, have equity own­ers. Thus, in the classical capitalist firm and the open corporation, those who are ultimate decision makers and residual claimants cannot exploit the pri­mary providers of capital because they are the primary providers of capital.² By contrast, in the cooperative, although the workers are both the ultimate decision-making authorities and the residual claimants, they are not the providers of capital. The state exercises that function or responsibility. Since this division of functions or responsibilities is a source of numerous oppor­tunities for exploitation, a natural proposal would be to join these three roles in a market socialist system to create equity owners. In principle, this could be done in one of two ways: either the workers in the various cooperatives or the state could be the firm’s equity owners.

If the workers own most of the firm’s capital, they are unable to siphon wealth from the capital providers into their own pockets, since they are the capital providers. This proposal effectively solves the so-called horizon problem because the workers’ equity ownership requires them to have a time horizon that is as long as the ultimate decision-making authorities in capitalist orga­nizations. If capital equipment is not maintained properly, if adequate reserves are not set aside to replace used up or worn out capital goods, if intangible assets are sold off, or if the quasi-rent value of the firm itself is dis­sipated, then the workers in their capacity as the firm’s equity owners suffer the full wealth consequences of their decisions on these matters. There would be no need for armies of auditors to fight pitched battles with cooperatives over the valuation of assets. Indeed, small firms would not have to be audited at all, and large firms would be audited to gather the same kind of informa­tion as auditors seek from corporations in a free enterprise system.

This model has another advantage over the worker control-state owner­ship model. Since the workers are the equity owners, they receive the returns to capital; assuming those returns are not taxed away, the cooperatives will have both the means and the incentive to finance most new investment. If this model puts control of new investment in the hands of the cooperatives, it closes down or minimizes the opportunities for exploitation that come from total state control of new investment. There would be no opportunistic bureaucrats in poorly monitored state organizations with control over vast portions of social wealth. Failing cooperatives would have no better chance of being bailed out by the state than do firms in a free enterprise system. Groups of workers would be putting their own wealth on the line in making investment decisions, which precludes the need for state monitoring of these decisions.

Despite the advantages of equity ownership by the workers in compari­son to the worker control-state ownership model, the former remains infe­rior to a free enterprise system with respect to exploitation in at least three respects. First, there is still the opportunity for the unskilled (or some other majority of) workers to exploit the skilled (or some minority of) workers within the cooperative. Since the workers are the ultimate decision-making authorities, pay differentials would still reflect majority opinion (assuming a one person-one vote rule) about the value of different workers’ contributions. For reasons explained in the second section of chapter 6, this means that it is likely that the value of some workers’ contributions will be systematically underestimated relative to the value of the contributions of others. Labor markets would also be more sluggish for the same reasons they would be in the worker control-state ownership model: firms are reluctant to expand and contract their membership in response to changing economic conditions; it is harder to start new firms because entrepreneurial gains are widely dis­tributed; and because of variations in a firm’s income, it is difficult for prospective members to evaluate job offers from other firms. Because labor markets do not function as well as under free enterprise systems, workers who are not getting the value of their contributions would be less likely to have anywhere else to go and thus would be exploited.

Second, monitoring arrangements in small-to-medium-sized cooperatives would suffer the same systematic weaknesses of adverse selection and moral hazard those organizations suffer in the worker control-state ownership model, since, in both cases, the workers are the ultimate decision-making authorities. This means that nonshirking workers would be exploited by shirking workers and shirking monitors. These problems are less serious in the classical capitalist firm for reasons indicated in chapters 5 and 6.

Finally, this system requires the equity owners of the smaller cooperatives to monitor the entrepreneurship of managers in the latter’s roles of central contracting agents and directors of the firm’s product. This is a problem because it is difficult for the membership of the cooperatives (or indeed for anyone) to evaluate the entrepreneurial contributions of managers in these two roles. Because of the (partial) separation of equity ownership from man­agement in the smaller cooperative, there is a monitoring problem in the smaller cooperative that does not exist in the classical capitalist firm, since in the latter the equity owner is the entrepreneur. This can result in productive entrepreneurs being exploited by being forced to share the extra residuals they bring in; on the other hand, ineffective entrepreneurs can exploit their fellow workers in the cooperatives by forcing them to pay for their mistakes.

But what about larger cooperatives in which the workers are equity own­ers compared to the open corporation? Are they more vulnerable to exploita­tion than their corporate counterparts? To answer this question, it is neces­sary to investigate the monitoring arrangements of the large cooperative; this requires us to confront the issue of equity markets in this model of market socialism. A central question in this connection is whether or not ownership of equity shares is restricted to members of the cooperatives. If the sale of equity shares is not restricted, then outsiders could acquire ultimate decision­making authority in the cooperative.

One is what has been called “the portfolio problem.” It has been widely remarked that a system in which the workers have equity shares only in the firms of which they are members presents them with risks that could be avoided by portfolio diversification (e.g., Jensen and Meckling 1979, 485-88). To put the point informally, if workers have all of their nonlabor assets tied up in their own firm, they are taking risks that could be avoided by owning equity shares in a diverse portfolio of companies.³ Since it would be rational for them to diversify their risks, it is reasonable to suppose that they would do just that. It is easy to see how a process of diversification could reproduce a free enterprise system over a relatively short period of time.

David Ellerman has called attention to another reason why these firms are likely to degenerate into open corporations (Ellerman 1984, 263). At some point, older workers (and perhaps managers) in successful firms would want to realize the capital gains on their equity accounts. Those accounts would have been built up to such an extent that new workers would not be able to purchase their shares; only outside investors, perhaps by pooling their resources, would be able to do that. As time goes on, these outsiders would acquire a controlling interest in the firm. Indeed, the more profitable and better managed the firm is, the more quickly this would happen. All of the workers, not just the older ones, would want to realize at least some of their capital gains before they retired or left the firm and thus would want to sell some of their shares. However, allowing outsiders to have voting rights, which is implied by equity ownership, is incompatible with worker self-management (Bonin and Putterman 1987, 61-62; Estrin 1989, 174). This means that unre­stricted equity ownership could not be endorsed by any socialist for whom self-management in the workplace is instrumentally or logically related to the socialist vision of the good society.⁴ But even those socialists for whom self­management is not central to their conception of a socialist system could not endorse this proposal, since this system would effectively permit widespread private ownership of the means of production, thereby violating a necessary condition for any economic system to be socialist.

Suppose, therefore, that only workers in a firm could buy equity shares. The market for equity shares, then, would be restricted to present workers and those who wished to join the firm. In effect, the latter would involve a market for memberships in the cooperative. As Putterman points out, these would be a fairly thin markets, that is, there would be very little trading going on (1988a, 258-59). The reason for this is that these markets would really be labor markets, and the latter are sluggish in this type of system for the same reasons they are in the worker control-state ownership model. While current members of the firm might buy and sell shares among themselves, those mar­kets and the markets for memberships would certainly not be very brisk in comparison to equity markets in a free enterprise system. This suggests that the superior monitoring afforded by an active stock market in which equity shares are frequently traded would be lost. In addition, the system precludes a market for corporate control, which allows bad management to be ousted as the result of a hostile takeover. Internal rebellion would be the only option short of quitting for dissatisfied workers. As explained in chapter 7, this is less likely to be effective than hostile takeovers or proxy fights. All this suggests that the monitoring of management would be less effective in the large coop­erative than in the large open corporation. Less effective monitoring creates opportunities for poor managers to exploit other workers in the large coop­erative and for especially good managers to be exploited by other workers in cooperatives.

Perhaps a more serious difficulty with this form of market socialism is that it seems to be unstable. If a worker leaves the cooperative and a new mem­ber is not there to buy that equity stake in the firm, it is not clear what would happen. If the firm had to pay out the accrued equity value of workers who leave, it would create a serious strain on firms in which many workers are approaching retirement around the same time or in which there is high labor mobility (a fact that would not be lost on the more opportunistic workers who can credibly threaten to leave). There is also a problem if workers in espe­cially profitable firms want to get access to some of the capital gains their firm has enjoyed before they quit or retire.

An interesting way of dealing with these difficulties has been suggested by David Ellerman (1984; 1986; 1990, 81ff). To understand Ellerman’s suggestion, it is necessary to explain some of the rather novel details of his model. This model is especially interesting because it may be the most plausible way to com­bine self-management and equity ownership by the workers. In this model, members of the cooperative have individual equity accounts and the firm itself has a collective equity account.⁵ Together, these accounts are credited on a quarterly or yearly basis with the returns to the firm’s capital and any undis­tributed residuals. (Workers also receive a conventional wage.) Both types of equity accounts serve to secure the cooperative’s debt obligations, both short and long term. New investment is financed internally from these accounts and from external debt; individual and collective equity accounts appreciate in value to reflect the firm’s growth and other increases in the value of its assets.

Workers establish their individual equity accounts by paying an entry fee when they join the firm. These accounts are periodically “paid down” to the workers, though they are not fully paid out. For example, each year after an initial lag-time of say, five years, workers would receive the equivalent of their share of the returns to capital (and any other appreciation in the value of the firm’s assets) for one year, less any amounts credited to the collective equity account. This “pay-down” prevents members from building up very large equity accounts that the firm might have trouble paying out when workers leave; it also equalizes the risks between older workers and younger workers who would otherwise have significantly different amounts of equity at risk in case the firm were unable to meet its debt obligations. When a worker left the firm or retired, the balance of his individual equity account would be paid out in the form of a perpetual debt instrument that would be fully marketable (Ellerman 1990, 82-85). Equity is converted to debt because debt holders have no control rights, and this system restricts control rights to the firm’s current members (i.e., the equity owners). On the other hand, the collective equity account is never paid out. This means that some of the returns to cap­ital are “lost” to workers as individuals. The collective equity account is quite important and must be maintained, however, since it serves as part of the security for the cooperative’s external debt—short term, medium to long term, and perpetual.

The principal advantage of this structure of equity rights and this method of financing new investment is that it give the worker an essentially unlimited time horizon, since he must be concerned with capital gains or losses that he will suffer on retirement or when he leaves the firm (Ellerman 1986, 65-66). Any debt instruments that he receives will be fully marketable and thus will tend to reflect the value of the firm. Not coincidentally, this proposal bears a striking resemblance to arrangements in the Mondragon cooperatives.

Two questions can be asked of this system: (1) is it in fact stable? and (2) is this a solution to the more serious problems of exploitation that face ver­sions of market socialism in which there are no equity owners? With regard to question 1, it may not be possible for the firm to keep a large enough equity cushion to meet the equity (i.e., internal) financing needs of the coop­erative and to secure all of its debt, including the perpetual debt instruments issued to retiring workers. One reason for this might be relatively high turnover. Recall that the Mondragon cooperatives have very low turnover among their members, which gives their equity base a stability that could not exist in a more mobile society. Cooperatives in more mobile societies may become highly leveraged, especially if the collective equity account is allowed to dwindle or if it is not adequately built up. They might also become highly leveraged because of high capital-to-labor ratios and the passage of time. Consider, for example, how much perpetual debt might be outstanding for a large, capital-intensive firm, such as a mining company after forty or fifty years. Highly leveraged firms are very vulnerable in difficult economic times, since debt holders must be paid periodically no matter how slow business is. Thus, it is unclear whether an economic system composed primarily of these types of cooperatives would be able to survive the ups and downs that any firm in a market economy faces over the long term.

On the question of exploitation, the answer would seem to be mixed. Equity shares are not freely alienable, and (as noted) there is no market for corporate control. Some aspects of management (notably, entrepreneurship) would likely not be as well monitored as they are in the open corporation, since there are no outsiders who can make a great deal of money by buying and selling equity shares or by making takeover bids. This means that it would be easier for bad managers to exploit cooperatives, and good man­agers of cooperatives would be more easily exploited than their corporate counterparts. However, this problem might be mitigated to some extent by the market for corporate debt.

On the other hand, in theory at least, Ellermans model avoids the moral hazards of substantial debt financing by cooperatives. Recall that these haz­ards involve opportunities for exploitation having to do with maintenance and replacement of capital goods in a system in which the ultimate decision makers have a limited time horizon. This model elegantly solves this horizon problem in a way that is compatible with the requirement that only present workers are ultimate decision-making authorities. If the opportunities for exploitation created by a combination of substantial debt financing and the workers’ potentially limited horizon are the really serious defects of the worker control-state ownership model, then assuming that substantial debt financing can be avoided, the other forms of exploitation identified might look like a price worth paying for the benefits of self-management. The only problem is that this is a price that a socialist, qua socialist, cannot afford to pay. Fundamentally, there are two reasons for this. One is that it is arguable that this economic system is not really socialist. The other is that it cannot realize a number of important elements of the socialist vision of the good soci­ety. These are logically distinct issues.

The social character of this form of ownership might be called into ques­tion on the grounds that the disposition of the net social product (the returns to capital plus the residuals) is done to serve private interests, not the public interest. In other words, decisions about what to do with the firm’s earnings, including, most notably, new investment decisions, still represent the inter­ests of disparate groups, even though in this case, the groups are the respec­tive memberships of the cooperatives. In this type of system there is no eco­nomic organization or family of economic organizations that is intended to represent the interests of society as a whole.⁶ As Saul Estrin has said, “The concept of collective ownership must... preclude any direct ownership... by the workers of the machines upon which they work. Ownership of coop­eratives in a market socialist economy must therefore be social in the sense defined earlier” (1989, 185). Social ownership is explained a few pages ear­lier by reference to the Yugoslav system, in which the state, as the represen­tative of society as a whole, owns the cooperatives or at least their capital (p. 172). Similarly, in the worker control-state ownership model, the social char­acter of ownership of the means of production is located in the fact that the state effectively owns the means of production because it receives and dis­burses the returns to capital, thereby controlling most new investment. Assuming a highly participatory democratic state, the latter is supposed to represent, albeit imperfectly, the interests of society as a whole.

Ellerman shows some sensitivity to this issue and argues for the social character of ownership in his model on the grounds that each worker has a voting right in the firm of which he is a member; because of this, the firm rep­resents the interests of its members (1990, 51-58). He compares the cooper­ative to a government in this respect:

Governments are “all-inclusive” in that they represent everyone who legally resides in a certain geographical area.... But the management of a demo­cratic firm is also “all-inclusive” in that it represents everyone who works in the enterprise.... Why shouldn’t a grouping of people together by com­mon labor be just as “social” as the grouping of people together by a com­mon area of residence? The genuinely “social” aspect of a democratically governed community is that the community itself is not a piece of property. (1990, 75)

This might establish the social character of the ownership of the cooperative’s capital and even possibly the socialist character of the cooperative itself. How­ever, it does not establish the socialist character of the larger economic sys­tem of which it is a part. To infer the latter on the basis of the former is to commit the fallacy of composition. The legitimacy of this worry about the socialist character of the entire system is strengthened by the important sim­ilarity that this model has with a free enterprise system: in its control of the social product, this system seems to be guided by a swarm of essentially pri­vate interests instead of a collective expression of the public interest that state control is supposed to represent.

Ellerman might have a response to this, however. In his discussion of the democratic principles that underlie his model, he advocates something he calls the affected interests principle, which states that “everyone whose rightful interests are affected by an organization’s decisions should have a right of indirect control (e.g., a collective or perhaps individual veto) to constrain those decisions” (1990, 47). This is a principle that I suspect most contem­porary socialists would assent to. Ifan economic system follows this princi­ple, then the interests of others are taken into account in all decisions, includ­ing those concerned with control over the social product.

The problem—and it is an enormous one—is that Ellerman nowhere explains how this principle is to be implemented in a way that goes beyond respect for market relations. Indeed, he admits (with considerable under­statement) that this principle would be difficult to implement (1990, 47). This principle is an end or a goal, not a means. Unfortunately, he provides no details about the organizations or institutional structures beyond the cooper­ative that would implement or realize this principle. Without some idea of how this principle is to be implemented (especially when its realization con­flicts with the decisions of the cooperatives about, for example, pricing, investment, distribution of residuals), the whole system of property rights in the means of production must be judged radically underdetermined. In con­sequence, both the socialist character of this system and its feasibility remain in doubt.

But even if these doubts could be resolved, there is another family of rea­sons for calling into question the socialist character of this proposal. These reasons can be approached by considering why it even matters whether this is a socialist economic system. The answer argued for in chapters 1 and 2 is that socialists believe that a socialist economic system is responsible for real­izing a vision or conception of the good society, a minimal version of which was identified in chapter 2. Unfortunately, the Ellerman model is unlikely to realize three crucial elements of this vision of the good society. Two of these elements or goals that a socialist economic system is supposed to achieve are (1) collective control of the rate and direction of economic growth and (2) the prevention or correction of the social irrationalities of the market (e.g., unem­ployment and pollution) in a way that is categorically more effective than what a state in a free enterprise system can do.

It might be argued that this is what the affected interests principle is sup­posed to do, but to reiterate, the principle is an end, not a means. Absent some account of the institutional means by which this principle is to be real­ized, there is no reason to believe that either of these elements of the social­ist conception of the good society could be achieved in this type of system. After all, the rate and direction of economic growth cannot be the subject of collective choice because control of new investment is left to the individual cooperatives. It is also unclear how or why this type of system would be in a better position to deal with the social irrationalities of the market than is the state in a free enterprise system. This is not to deny that states in free enter­prise systems can affect the rate and direction of economic growth by a com­bination ofsubsidies and discriminatory tax policies. Nor can it be denied that these and other instrumentalities could be used to treat various social irra­tionalities in a market socialist system. The problem is that there is no reason to believe that the state in such a system would be categorically more effec­tive in these matters than it is in a free enterprise system. The promise of socialism is not one of modest improvement on one or more of these fronts; instead, it is the promise of truly radical beneficial change on all of these fronts. Nothing short of that Wouldjustify the enormous institutional changes that socialists advocate. How such changes could occur in a system in which control of new investment is highly decentralized is unclear, at best. Certainly the current state of the art in macroeconomic analysis and policy gives no cause for optimism.

Finally, perhaps the most serious problem of all from a socialist perspec­tive is that it is very doubtful that this type of system would be able to realize the achievement of relative equality of material condition. The reasons for this can be explained as follows. As a technical matter, firms differ widely in their ratio of capital to labor. Oil refineries, for example, are very capital­intensive whereas truck farming is very labor-intensive, ff workers received the returns to capital, as this proposal implies, workers in highly capital­intensive industries would be much wealthier than those in less capital-inten­sive industries. Thus, the refinery workers would be much wealthier than the vegetable growers, and both would be paupers in comparison to the mem­bers of some large financial cooperatives.

To this it might be responded that the state could adopt a progressive income tax to equalize incomes or at least to reduce inequality of incomes (though this would leave inequalities of control over productive resources relatively untouched). There are two problems with this. First, it is unlikely to succeed for roughly the same reason it has not succeeded in democratic societies with a free enterprise system: one of the things that people do when their wealth or income is threatened by the state is to invest resources in the political process to ensure that the state does not expropriate them. The oil refinery workers would undoubtedly form a political action committee or something similar to help elect people or parties who believe that workers are entitled to the returns to capital that they control. Instead of “All Power to the Workers’ and Soldiers’ Soviets,” their slogan would be, “All Returns to Capital to the Workers’ Cooperatives.” Marx (not to mention Lenin) could hardly be pleased with these developments. Second, this policy would under­cut the whole point of giving equity ownership to the workers, namely, to give them a financial stake in maintaining the value of the firm and in making wise investments for the future. One cannot maintain the form of equity owner­ship while eviscerating its reality by high capital gains and/or income taxes.

Whatever the virtues of the Ellerman model (and I think they are con­siderable) there is nothing in it to ensure relative equality of material condi­tion, at least across firms. Indeed, for the reasons indicated, wide interfirm variations in income are quite likely. And, as noted, it is doubtful whether the rate and direction of economic growth could be a matter of collective choice in the society, given that individual cooperatives have the resources and incentives to finance internally new investment through retained earnings and to finance externally by borrowing. Finally, it is unclear how the social irrationalities of the market could be handled in a categorically more effec­tive way. For all these reasons, even if this type of economic system is social­ist, it is not capable of realizing the socialist vision of the good society.

In fairness to Ellerman, he does not make a sustained effort to establish his socialist credentials. Indeed, he suggests that his proposal is a “third way” between capitalism and socialism—though the form of socialism he has in mind is state socialism (1990, 206). One of the themes of this chapter is that models of economic systems that avoid the problems of the worker con­trol-state ownership model are either nonsocialist or unable to realize the socialist conception of the good society, or both.

Another form of equity ownership for a market socialist system is full state ownership of the means of production. Making the state the ultimate deci­sion-making authority, residual claimant, and provider of capital has some obvious advantages over the worker control-state ownership model in pre­venting exploitation. First, since workers are not ultimate decision-making authorities and residual claimants, the exploitation of skilled workers by unskilled workers would not take place. Workers are not voting on the crite­ria whereby pay is determined; instead, they are hired by state-appointed managers. Labor markets would be about as robust as they are in free enter­prise systems, and the various protective devices (notably, unions) found in free enterprise systems could and presumably would persist in this type of market socialist system. Second, since the managers are not answerable to the workers, the adverse selection and moral hazard problems that plague man­agement of the small-to-medium-sized cooperative would not arise in these state-owned firms. Finally, managers who are poor entrepreneurs could not exploit the workers, nor could the workers exploit managers who are good entrepreneurs.

However, full state ownership is both inferior to a free enterprise system on the issue of exploitation and problematical from the point of view of the socialist conception of the good society. The main vulnerability to exploita­tion arises in connection with the monitoring of management. It is an impli­cation of full state ownership of the firms that managers are ultimately answerable to political authorities. The problems associated with monitoring by political authorities were outlined in the last section of chapter 7: multiple decision criteria and structural impediments to accurate monitoring (not to mention sheer political patronage) create numerous opportunities for exploitation by and through political organizations—opportunities that do not exist in the classical capitalist firm or that tend to be minimized by the superior monitoring mechanisms available to the owners of the modern open corporation in free enterprise systems.

Some of the more important avenues of exploitation in a state-owned market socialist firm are those relating to successful and unsuccessful entre­preneurship on the part of managers. Because the state is the residual claimant, if a firm’s management (perhaps with the advice and active coop­eration of its workforce) earns large entrepreneurial profits, those profits are shipped off to the state. Though the state might pay its managers perfor­mance bonuses, managers’ pay cannot be largely determined by their entre­preneurial contributions, for then they would effectively be the residual claimants. Unlike their counterparts in free enterprise systems, who can start their own firms or take their respective divisions private through a leveraged buyout, successful managers of large, state-owned firms can only work for other state organizations, which are relatively poorly monitored and politi­cized; that is, they effectively have nowhere else to go. For these reasons, the managers of successful firms would be exploited by the state. If, however, the firm suffers losses and the state pumps more resources into it to keep it afloat, the taxpayers are exploited. By separating residual claimancy from opera­tional control, the rewards and penalties that go with successful and unsuc­cessful entrepreneurship are systematically misallocated. Finally, it is hard to see what the incentive is for workers and managers to be good stewards of the portion of social wealth over which they have operational control (if not ownership). A version of the horizon problem would undoubtedly face this type of system.

Independent of these problems of exploitation, this form of state owner­ship is likely to be unable to realize other elements of the socialist vision of the good society. Recall from chapter 2 that one of the motivations for self­management is that it is supposed to ameliorate alienation. In a self-managed firm, the workers are not working for someone else since they appoint the managers, or at least the managers are ultimately answerable to them. Because of this, there is a fundamental sense in which their productive lives are unalienated. Self-management is supposed to reduce alienation in the workplace in more concrete terms as well: in self-managed firms workers decide on matters such as plant layout, work schedules, and even trade-offs between income and more intrinsically satisfying work. This control over their working lives is also supposed to have other good effects, including dampening the us-versus-them mentality between managers and workers. However, no matter how democratic the state is, when the latter has complete ownership of the means of production, these benefits are no longer assured. The reason for this is that whatever the institutional details, the locus of ulti­mate decision-making authority—and thus ultimate power—shifts out of the firm and into the political arena. Politicians are effectively the ultimate deci­sion makers, and they are in turn answerable to a diverse electorate. The problems involved in monitoring politicians aside, it is clear that workers will not have as much of a say in the operation of their firms as they would in the worker control-state ownership model. Thus, there is no assurance that man­agement will be as responsive to the potential causes of alienation as their counterparts in the worker control-state ownership model would be.

This observation touches a final difficulty with complete state ownership of most firms. In light of the experience of communist societies in the twen­tieth century, full state ownership of most of the means of production may no longer be a serious option in the capitalism/socialism dispute. It is true that some socialists see the failures of the communist world as failures of a certain type of state—a centralized, nondemocratic state that has been alien­ated from the rest of society. They believe that if only these states had been truly democratic—in the sense of widespread participation in the political process—the problems of communism would not have existed or, at least, would not have been nearly as severe as they were and are. But the fact is that most people do not seem to view it that way. There is the widespread per­ception that state ownership of the means of production has been a large part of the economic problem with former communist systems. Evidence for this comes from the fact that former communist countries are experimenting with various forms of nonstate ownership and not merely various forms of a mar­ket economy. In addition, throughout the developed world there seems to be little enthusiasm for state ownership. In general, complete state ownership of most of society’s means of production just does not seem to be on the agenda anywhere in the world today.

Perhaps the most compelling reason to be deeply suspicious of complete state ownership of the means of production is that it unites in the state both political and economic power. It is plausible to maintain that one of the most important lessons of the twentieth century is that this arrangement is inher­ently totalitarian. Indeed, most socialists in the last third of the twentieth cen­tury have abandoned full state ownership in favor of some form of coopera­tives. Experience with totalitarian state socialism is surely part of the reason—and a good reason—for that.

Given the serious deficiencies of these forms of equity ownership from a socialist perspective, there seems to be not even a prima facie case for com­bining them by putting representatives of the state on the board of directors of the cooperatives in the Ellerman model. On the other hand, a case might be made for putting state representatives on the board of the cooperatives in the worker control-state ownership model to provide better monitoring of the firm’s use of society’s capital. This in turn would attenuate the opportu­nities for exploitation that would otherwise face a system in which the ulti­mate decision-making authorities are not the primary capital providers.

The problem with this suggestion is that it assumes that a board of direc­tors composed in this manner would represent a kind of blending of the var­ious interests board members represent, whereas, in point of fact, it is likely that the board would represent the interests of the dominant group. This has sometimes been called the “law of one majority” (Ellerman 1990, 47). So if the workers are the predominant ultimate decision-making authorities (and it is hard to see how any other arrangement can ensure the benefits of self-man­agement), their interests will decisively determine the policies and procedures of the firms. On the other hand, if state representatives are the ultimate deci­sion-making authorities (and it is hard to see how any other arrangement could effectively deal with opportunities for exploitation), then their inter­ests will be determinative.

To conclude, equity ownership has some undeniable advantages in pre­cluding or minimizing forms of exploitation that would flourish in the worker control-state ownership model that has been the main focus of this book, but making the workers or the state the equity owners creates other serious dif­ficulties from a socialist point of view. The alternatives are either unstable, not socialist, or incapable of realizing the socialist conception of the good soci­ety. The next section considers some other alternatives that do not involve joining the roles of capital provider, residual claimant, and ultimate decision­making authority, that is, some alternatives that do not involve equity own­ership.