II. The Egalitarian Conception of Fair Exchange

Here I sketch an account of fair exchange. An intuitive and useful starting point is to think about fairness in agreements in the case where there is only one exchange between two people that will determine their whole lives. In this case, the appropriate background fairness conditions for such an exchange consist in the realization of equal capacities for that exchange. This breaks down into two components: equal cognitive conditions and robust equal opportunity for exiting or refusing entry into the arrangement. Equal cognitive conditions involve equal access to information relevant to one’s interests and concerns and abilities to negotiate desirable arrangements. The basic institutional supports for this are a system of education and systems of protections of consumers in the contexts of arrangements with great asymmetries of information. We achieve equal opportunity by making sure that people have the resources that enable them to exit or refuse transactions and enter others that advance their interests. The second determinant of power in the context of agreement-making is the value of the outside option for a person. A person with good alternatives has bargaining power. Persons with equal opportunities for exit have a kind of equal power. In this sense, I am talking about real opportunities and not merely formal opportunities. Education, basic needs provision, and other goods give people opportunities to choose among transactions, thus enhancing their bargaining power.

These conditions give each person equal power to shape the agreement with the other on terms she judges best. And this gives each person equal power to shape the social world she lives in. Giving either person less than equal capacities, at least for normal adults, would amount to treating that person’s interests as having less than equal importance.

In the usual case, of course, each person engages in a series of many exchanges with many different people. Here the same principle directs us to say that the parties to these exchanges must have equal capacities globally in the sense that they start from background conditions that ensure equal capacities for all. This equal capacity condition need not be fully maintained throughout the series, however, because earlier agreements a person enters may curtail opportunities that she will have in later agreement-making. If this is done knowingly, later agreement-making in which there may be some local inequality of opportunity or inequality of cognitive conditions is not unfair. Furthermore, individuals may choose to focus on agreements in which they think of themselves as having a lot at stake and focus less on other exchanges in which they think of themselves having significantly less at stake. So, the account does not assert that there must be equality between persons in every agreement-making context. It requires only a kind of global equality of capacity for determining whole-life prospects.

One important clarification is that when I refer to equal capacity, equal power, equal cognitive conditions, and equal opportunity more particularly, I have in mind a kind of maximal equality. This maximal equality is incompatible with leveling down. It implies that not only is the distribution of power (or opportunity or capacity) equal, but there is also as much power or opportunity as the situation allows. One could conceive of this as a combination of two values: equality and the value of whatever one is equalizing. Or one can conceive of it as based on an anti-leveling-down conception of the principle of equality that implies that inequality can be superior to equality if everyone is better off in the circumstances of inequality.Footnote ⁹ Irrespective of how one thinks of this, it is essential to my discussion that one conceive of the equality as one that satisfies a kind of Paretian constraint, which is to say that one state is better than another if someone is better off and no one is worse off in it. I will flag this for the reader on occasion by putting the term “maximal” in parentheses before the relevant equality, but I am assuming it throughout the discussion of equality.

This is meant to realize a kind of equal power in the context of decentralized decision-making that is analogous to equal power in democratic collective decision-making. This is because the two conditions in the one-shot case in effect specify circumstances in which persons have a (maximal) equal say in the structuring of their relations with each other. They specify a kind of condition of global equal bargaining power between parties such that each person has an equal say in the formation of the contents of the series of agreements they enter. And the global principle of (maximal) equal capacity gives persons a kind of equal say in the formation of their social lives together with others when they engage in a series of agreements with many people.Footnote ¹⁰

III. The Rationale for the Egalitarian Approach

The distinctive approach I attempt here is grounded in the idea that there is a fundamental analogy between democracy in collective decision-making and an egalitarian principle for evaluating the background conditions of exchange in decentralized decision-making. Ultimately, I think the standards of fairness in collective decision-making and in decentralized decision-making are grounded in one single more abstract principle. That single principle is a principle of (maximal) equal distribution of power in the context of disagreement and conflict as well as cooperation. I cannot develop that hypothesis fully here except by developing the structural similarities between the contexts of decentralized and centralized decision-making, all the while respecting the differences.

What animates the idea that there is such an analogy are the similarities between what people are doing when they engage in agreement-making with others and when they participate in collective decision-making. First, in both activities, persons attempt to shape the social world they live in. In decentralized decision-making they attempt to shape that world in the many agreements they enter by altering the rights and duties people have with respect to each other, the division of labor they enter, and the local distribution of benefits. The sum of agreements a person enters over a lifetime give shape to the social world the person lives in.

It is important to note that these two contexts are quite different so that equalization of power in collective decision-making does not ensure equalization in decentralized decision-making. Decentralized agreement-making shapes the local world in which one lives so that it suits one’s particular needs and concerns. Democratic participation in collective decision-making cannot reach into the particularities of one’s relations with others in this way.

Second, although cooperation and mutual advantage are central to agreement-making, so is conflict. Our aims often conflict with those we exchange with in that we desire to give less for more and so does the other. The conflict between wage earner and employer fits this quite clearly. The outcome of agreement-making activity is then partly determined by a distribution of power among the parties, which I will explain in some detail below. For now, the fact that the content of an agreement favors the person with market power over someone without it, is sufficient to illustrate the idea. I will argue that power disparities make a difference even under conditions of perfect competition. Nothing in this idea implies that agreement-making is a constant-sum activity. There is a surplus of agreement but there is also the distribution of that surplus, and the surplus and distribution are both partly determined by the distribution of power.

Third, the justifications for assigning powers to shape the social world are grounded in a set of common liberal concerns. Persons have different interests that conflict and we give each person some power to pursue those interests. Persons disagree on how best to shape their social worlds and we give each person some power to act in accord with his or her own judgment. Furthermore, there is at least a basic dimension of these issues about how best to shape the social worlds that we do not think ought to be decided by expertise. We think that people ought to be able to make the basic decisions about how their society is organized, and how their lives with others are organized, based on their own judgments. This is the common core of liberalism at the root of democracy and liberal rights. In one case, they are meant to provide people with the power to participate in centralized decision-making, and in the other they are meant to give people power to engage in decentralized decision-making.

Because the interests of persons are of equal importance and we think that each person ought to be treated as an equal in this context of conflict and disagreement, I affirm that fairness requires that power be distributed equally both in centralized decision-making and in decentralized decision-making. I cannot develop this argument further now.Footnote ¹¹ Here I hope to exploit the analogy with democracy and the strong commitment most people have to democracy to argue that the analogical variant of equal power ought to be applied to the context of exchange.

To be clear, I do not mean to imply that these decentralized settings ought to be centralized and democratized in the traditional way. My intention is to show that there is an analogy between democratic participation and the activities of persons in decentralized settings. The values involved in personal relationships and development, the distinctive values that arise from people cultivating their unique talents and ideas, and the need to give shape to the local world one lives in must be given some significant protection from collectivization. And I think, with the tradition of the economic theory of Adam Smith and John Maynard Keynes, that an open market system (a market that gives a wide scope for individual choice and that is not rigidly biased against certain forms of association) is important for putting resources to productive uses, though it usually will not involve a free-market system because legal regulation limiting markets will often be necessary to enable persons to participate as equals.

IV. The Legitimate Preservation of Equality

(Maximal) Equal opportunity is compatible with unequal outcomes of various sorts. Indeed, in some cases, it is illegitimate to bring about a kind of equality of outcome when the foreseeable results of persons acting based on their equal opportunities are unequal. For example, one person consumes all their goods, and the other doesn’t and consequently accumulates goods. This might be an unequal outcome that is acceptable. It might be unfair to try to equalize the outcome under these circumstances. It won’t always be unfair to do this, but certainly it will be unfair to equalize outcomes in cases where the equalities of opportunity were quite robust and where the outcomes are not terrible.

First, let us describe and analyze the case in which equal opportunity can legitimate the outcome in markets. This is the case in which differences in outcome are the results of people’s choices in two different ways. It is one in which the difference itself is the result of people making choices about how they want to organize their lives with others.

For example, two persons come to see that a hierarchical relation between them is desirable and choose to enter that relation because they think it is desirable. They could have chosen a different kind of relation, but they didn’t. For example, one person may simply not want to have responsibility for the products of the relationship and may want to be able to detach himself from it, while the other wants to have responsibility. They could then enter that relation willingly. They could have chosen a more egalitarian relationship in which both have joint responsibility and authority, but they didn’t. In this case, if the background conditions are ones of equal power, it is difficult to see why the subsequent inequality in authority is problematic. We should expect this kind of divergence from equality in a highly pluralistic society where people have very different preferences. This example illustrates different ways in which one can choose to enter an unequal relationship. Each chose the character of the relationship and they chose the distinctive positions they would occupy in this relationship. Let us call the first choice the structural choice and the second choice the positional choice. The fortunate situation in our example is that they both wanted the same structural choice and they had complementary desires regarding the individual positional choice. Here we have an unequal outcome, but it is fully legitimate from the standpoint of equality.Footnote ¹²

The other kind of inequality that may be justified when there is equality of opportunity is that persons with greater relevant talent should be put in places where their talents will do the most good, which usually involves some inequality in power. Such inequality of power is compatible with worker participation since the members of the firm can all recognize talent and put people with high talent in the right places in the division of labor. The analogy with democracy can be instructive here as well. Democratic politics can legitimately give more power to individuals if they are more capable of making rationally persuasive arguments or have talent in organization or leadership. The power exercised here can be called a kind of collaborative power or power exercised because and to the extent that it benefits others.

Let’s be careful about what this means. First, greater talent by itself does not entitle a person to greater rewards. Second, we need to distinguish inequality of talent from a monopoly on talent. The latter allows people to exercise outsized power. Monopoly power allows one to extract goods from others beyond the extent to which one benefits them. Here I am thinking in terms of competitive markets in which many persons supply talents. A person’s position in the division of labor is justified because they can do more than others to advance the aims of other people when they occupy that position.

V. Two Dimensions of Equality of Opportunity

Equality of opportunity in the standard sense cannot always legitimate an outcome. And the egalitarian conception can help us appreciate this. To see this, we need a further distinction. First, there is equality of opportunity in the competitive sense, by which I mean that persons have an equal opportunity to occupy different positions in the division of labor. This assumes that the division of labor, the rules of the contest, the criteria of success and the rewards for success are fixed, and individuals have equal opportunities to occupy any of the positions in that division of labor. This is the traditional picture of equality of opportunity that makes it analogous to a kind of fair starting point in a race in which there are winners and losers. Here the criteria for selection in a competition and the structure of the relations established by the division of labor loom large. This is an important dimension of equality of opportunity, but it is just one dimension.

I want to contrast this competitive dimension of equality of opportunity with what I call the constructive or creative dimension of (maximal) equality of opportunity. Here each person has equal abilities to structure her relationships with others. The egalitarian conception of equality of opportunity, which gives each person power over the structuring of the social world she lives in, points to this further and important dimension of equality of opportunity.

Theorists do not usually remark upon this aspect of equality of opportunity,Footnote ¹³ but it is crucial here. It includes capacities to shape the division of labor and to determine the rewards of the division of labor to some extent, and it enables persons to play a role in defining the qualifications that are necessary to occupy the different roles in the division of labor.Footnote ¹⁴ Persons have power to determine whether they work under a boss or whether they work in a more egalitarian arrangement. But there are many other features of their associations that they can determine and have interests in determining. They can also determine what merits are relevant to determining who should occupy what position. They can play some role in determining the distribution of benefits and burdens within economic associations. Equality of opportunity can ensure this when persons have equal power to join or not join any particular association. They can use this power to determine the exact character of the association they enter.

Moreover, there is significant empirical evidence in favor of the idea that people work harder or less hard depending on the degree to which they think their wages and other benefits are fairly distributed.Footnote ¹⁵ These are all issues on which there can be significant disagreement among persons and significant conflict of interests. So, it makes some sense for people to have the opportunities to choose different kinds of economic relationships with each other, in accord with their conceptions of fairness. I see this as an essential part of the complete realization of equality of opportunity because it gives people opportunities to structure their relations with others and not merely to compete for positions in social structures that have been fixed by others.

Here too we can use the analogy with democratic collective decision-making to see how this distinction works. One partial analogue with collective decision-making that sheds light here is the distinction between the power to vote for various alternatives and the power to set the agenda that gives the alternatives from which to choose. It is usually thought that a thoroughgoing democratic arrangement involves assigning both voting power and ultimate agenda-setting power to citizens. This distinction seems to me to run in parallel with the distinction between the competitive dimension of equal opportunity and the constructive dimension of equality of opportunity. In the case of equal voting power and competitive equal opportunity, we can select from a limited menu of options. In the case of agenda-setting power and the constructive dimension of equal opportunity, we can determine what the menus of options contain. Hence, the complete conception of equality of opportunity expands our abilities to shape the social worlds we live in beyond the merely competitive variety of equal opportunity.

To see an extreme example of the absence of the constructive dimension, let us imagine a society that is ineluctably divided into masters and slaves but in which each person has an equal competitive opportunity to become either a master or a slave. We can imagine this competitive equality of opportunity to be quite robust in the sense that everyone has equal access to the educational resources for becoming a master or a slave. The competition for these positions is entirely fair. But the criteria for becoming a master are determined in advance, and it is left to each person to strive to the greatest extent possible to satisfy these criteria. No matter how fair this competition is, there is something deeply troubling about the example. It seems that it curtails the equality of persons in an unacceptably narrow way, even though it does realize equal opportunity in the competitive dimension. Here is a partial analogy in constitutional choice. Suppose that a society may democratically make a choice among just a couple of constitutions, both of which are variants of monarchy. Here people participate as equals in creating the constitution, but they become deeply unequal once the constitution is created. This too fails to satisfy fully the democratic ideal in collective decision-making.

There are three distinct problems of fairness in the example. The first problem is that there is an extreme inequality in the positions that people can occupy, which has the effect of curtailing the ability of the slaves to play a role in shaping the social world they live in after they have been placed in their roles. A second, related problem is that even competitive equality of opportunity will normally diminish after people are placed in their positions. The role of slave diminishes the ability of the slave to move to the position of master. Slaves are stuck in their positions to some degree. Third, there is no choice as to the structure of the division of labor. I will argue that these problems are closely associated. Now, to be sure, they are different because one can imagine the conditions described in the first and second problems not obtaining while the third does obtain. For example, one can imagine a society in which only the completely egalitarian forms of relationship are possible so that there is no extreme inequality. Nevertheless, there is no choice over the structure of the division of labor. Furthermore, one can imagine, as a logical possibility at least, that equal persons have chosen the master-slave division of labor. Though this latter is highly unlikely, and I shall make use of this unlikelihood later in my argument, we can see that it is distinct.

There is already a great deal of plurality of economic organizations in modern societies, as Henry Hansmann in his classic work has been at pains to point out.Footnote ¹⁶ There are firms that are controlled by owners, firms controlled by boards of directors (who are accountable either to shareholders or to some combination of shareholders and workers), worker-controlled firms, consumer cooperatives, seller cooperatives, as well as many others. This suggests that there is already some element of the constructive dimension of equal opportunity at work in contemporary societies. But I will argue that there is significant rigidity in the markets that make some forms of economic association very difficult to achieve in modern societies without legal regulation.

VI. Complete and Perfect Markets

Under perfect competition with complete markets, persons simply plan out everything in their relations with everyone else. In the case of uncertainty, due to external shocks to the system, they agree on various securities or options.Footnote ¹⁷ Participants have universal opportunity since there are no transaction costs and no limits to information regarding all the other parties. There is no adverse selection or moral hazard, two main barriers to the acquisition of credit and insurance, especially for poor people. Persons simply find the places in which they can make the best contribution and which appeal to them. And there is no market power, or power to determine the price and conditions of agreement. Each person is a price taker. No one is subordinate to anyone else; everything is by agreement against a background in which many options are available. In this sense, persons do not exercise power over others in complete markets. When A and B meet to make a contract, A cannot simply determine the terms and say take it or leave it. A cannot simply improve the terms for her own benefit at the expense of B.

In this sense, complete and competitive markets realize incompletely the ideal of (maximal) equality of capacity. The one remaining obstacle to equal capacity is inequality of initial endowments. Equality in these would be necessary to achieve full equality of capacity. And a key feature of competitive analysis is that the efficiency and other desirable properties of perfectly competitive and complete markets are entirely compatible with different distributions of initial endowments.

Furthermore, under conditions of perfect competition and complete markets, there should be nothing to favor capitalist-owned firms over worker-owned firms. In general, there is nothing to favor capitalists renting workers over workers renting capital. The two types of firms can exist together and either type of firm can be efficient in competitive equilibrium, as is demonstrated in the equivalence theorems.Footnote ¹⁸ Hence, the presence of a kind of constructive equality of opportunity is evident in complete markets. This is an important result because of the central role of complete markets in economic theorizing and because perfect and complete markets realize the ideal of equal capacity (if initial endowments are equal). Equality of capacity can be used as a kind of benchmark. The results we see in existing markets are the consequences of distortions or imperfections in markets.

VII. Firms

So far, I have characterized the democratic conception of markets and the idea of equality of capacity that defines it. I have distinguished the competitive from the constructive dimensions of equality of opportunity and suggested that there is a defect in equal opportunity in societies whose organizations are ineluctably authoritarian. I have observed that equal power is realized in complete and perfect markets with equal initial endowments and that worker-controlled firms are as likely in these markets as capital-controlled firms. Here I want to discuss the problem of structural rigidity in market societies and regarding firms. But first we need to arrive at an understanding of what a firm is.

A firm is a group of people organized in a nonmarket way for the purpose of producing something to be sold to others in return for a reward. Usually, this involves incomplete contracts and some degree of hierarchy that enables the members of the firm to act in ways not fully specified by the contract. Incomplete contracts mark an important departure from a perfectly competitive and complete set of markets in which all contracts are state-contingent contracts (i.e., they cover all possible eventualities).

In complete or comprehensive contracts, A and B make an agreement that holds for all possible eventualities, and all the contractually agreed elements are fully observable and enforceable. There are some uncertainties about how natural events may impact agreements. And in imperfect markets there may be uncertainties about what A and B will do since they may not know each other’s types and they may not be able to verify the levels of effort each puts in. But the contract is written for all the different possible outcomes.Footnote ¹⁹

Incomplete contracts do not have provisions for all possible eventualities. They leave gaps and they leave some provisions missing. Obligations are undetermined in some cases and are vague or ambiguous in other cases. The incompleteness arises from the various costs of contracting. These costs arise from the inability of agents to foresee all the things that might arise for the relationship and from the complexity of the negotiations that are necessary for a complete contract.Footnote ²⁰

Incomplete contracts raise the question of who has control at a particular point in time after the contract has been made. Who gets to fill in the blanks left by the incomplete contract? And who gets to determine when the contract has been fulfilled or not? A standard way of filling in the gaps is through the construction of a firm with hierarchical relationships between some persons and others. There is a kind of limited authority relation established. For example, the contract between A and B is incomplete, but the firm gives A the power to tell B what to do where the contract has gaps, is vague, or ambiguous. A tells B what to do when the contract has not initially specified the exact behavior expected of B. A fills in the blanks at the appropriate moment. And the contract may give A the power to decide what to do when B has not acted as A wants him to act. This may involve insubordination on the part of B if B does not act as A has told him to act. Or it may involve something less well defined in advance, such as absence of quality work or insufficient effort, which of course contracts cannot normally fully specify. A simply doesn’t think that B has put in the effort that A wants to see. Or A thinks the quality of B’s work is not as good as he wants it to be. Or maybe A thinks some other person, C, who is not now employed by A, would do a better job, so A lets B go and hires C on the same incomplete contract.Footnote ²¹

To be sure, B has some power too here. B can leave. If we assume employment at will, A or B can terminate the contract at any time. But within the context of the firm, an authority relation is created with real power to enforce.

The presence of incomplete contracts and the partial authority of the firm do not require that the firm be an authoritarian institution. An authoritarian institution is one in which some people tell other people what to do and the subjects never have such power.Footnote ²² A nonauthoritarian hierarchy is established when the collectivity of the members of the firm makes decisions for the members. This might be true for the partners of a law firm, for example. It is true for the workers in a worker-managed firm and it is partially realized in collective bargaining. The characteristic of this kind of firm is that each person rules and is ruled in turn.

VIII. Consent

Consent is often thought to make A’s exercise of authority over B legitimate. But consent here takes on a kind of blank-check type character. It is one thing to consent to do a whole variety of things, each of which is specified in detail and can be clearly enforced. It is another thing to consent to someone telling you what to do when you don’t know in advance what that is.

Two conditions are normally thought to increase the acceptability of this authority relation. First, persons earn wages for their efforts and for the willingness to subordinate themselves to others. There is a certain limited protection from risk here because they are owed the wages even if the firm doesn’t make as much money as it thought it would. Second, most employment has a kind of at-will character for employees. That is, they can withdraw their consent at any time. They may consequently lose subsequent wages and the job, but they do have the liberty to withdraw from the relationship.

The egalitarian conception tells us that consent is not enough to make this relation fully legitimate. The consent must be given against the background of equality of power, which is specified in terms of robust and maximal equality of opportunity and equal cognitive conditions. The consent of a person who has very few options and little knowledge does not fully legitimate the relationship with someone who has a lot of options unless the latter person gives up the bargaining advantages she has in determining the relationship.

IX. Rigidity

Many have observed that markets tend to be biased toward authoritarian firms and against worker participation in any of the forms that I have discussed. A market is rigid in the sense I have in mind when it forecloses possibilities of social organization. Rigidity does not make different possibilities impossible; it makes alternatives very difficult and rare. But we need to look beyond rigidity into the explanations for rigidity. Different explanations make a difference to whether the rigidity is acceptable or not. Some sources of rigidity may make the difficulty of realizing alternatives legitimate.

There are several sources of rigidity in markets that incline them toward more authoritarian structures. (1) There is significant background inequality of capacity among persons in the society so that some have very little bargaining power in determining their roles in the firm. This is due to serious background inequalities of resources and education. (2) Inequality of capacity arises also because there is a significant degree of monopsony in labor markets. (3) Owners and managers also have the ability to make sure that the return on capital to them is as high as possible and thus may be able to block a more equitable distribution of control of, and returns to, the firm. (4) Some structures of firms are much more productive than others and may for that reason survive more frequently in a competitive environment.Footnote ²³

(1) Background inequalities and imperfections in the market. One main reason why firms are structured in an authoritarian way is because the members of the firms have radically unequal opportunities and cognitive conditions, and nothing in the society makes up for that. The consequence of radically unequal opportunities is that some have little power to determine their relations with others while others have a great deal of power. The difference in power enables the latter to say to the former, “Take it or leave it,” and since the former have little alternative, they take it. Surely this plays a significant role in determining the relations between employees, managers, and owners in modern society. It is difficult to imagine that equals would put themselves in such positions of subordination in modern societies except under the most extraordinary circumstances.

Another related reason is that societies at present do not offer the same quality of education to all people, so some are woefully underprepared for complex tasks while others are very well prepared for them.

Another case is when workers will not choose worker ownership because they rely heavily on their endowment to supply consumption and they are thus risk averse with it, so they are not optimal investors in capital. In addition, lenders tend to add a premium to the interest rate because of the danger of default since the borrowers are not well off and do not have a great deal of collateral.Footnote ²⁴ This is a complex case, but it does suggest the absence of equal capacity since workers generally have to be worse off than others to have this problem.

The background inequalities that explain firm structure are in violation of the underlying egalitarian principle we are working with. The problems posed by these background inequalities are compounded by the imperfections of the market. Asymmetries of information make it difficult for people with little wealth to get credit and insurance.

One could simply try to overcome the background inequalities in order to satisfy the principle. This would require massive redistribution of income and wealth. Alternatively, one might try to remedy the inequalities by realizing a greater degree of worker participation in the firm. This would involve legal regulation such as regulation that favors the formation of unions or regulation that requires that workers participate in selecting the members of the boards of corporations. Many societies do some of both.

(2) Monopsony as a structural cause of inequality in the market. In addition to being responsible for the rise of the firm as we know it, imperfect competition also plays an important role in creating unequal power for many workers and owners. Many have observed that the relations between ordinary unskilled or low-skilled workers and owners are very unequal in a way that gives rise to unequal bargaining power between workers and capitalists.Footnote ²⁵ Part of this consists in very unequal background conditions for workers and capitalists such as unequal educational institutions and economic resources. Another part consists in the presence of degrees of monopsony that result from there being many workers and relatively few and large firms.Footnote ²⁶ This does not require that there be only one firm and many workers. There need only be relatively fewer firms than workers with some difficulty for workers to move from firm to firm, either because of a paucity of information or because workers are attached to individual firms for a variety of reasons. There need only be some significant difference between workers and owners in these respects for some degree of monopsony to occur. Economists have noted the markdowns in wages for workers under these circumstances in the United States, for example.Footnote ²⁷ These markdowns are indicators of lesser economic power. What is especially noteworthy is that monopsony can be a systematic and necessary consequence of imperfect markets. It slots people into positions of unequal power in a way that is not explained by choice but rather by the structural characteristics of the labor market. These markdowns can occur even in the presence of a powerful welfare state that promises to help people between jobs.Footnote ²⁸

To be sure, not all relations between firms and workers are monopsonist; highly skilled workers tend to have greater bargaining power due to their more limited supply. Highly skilled workers illustrate what a market looks like when there is significant equality of power in the market. Unequal background conditions play a large role in explaining the difference in power between highly skilled and less-skilled workers, but the structural characteristics of the market also play a significant role.

In any case, these inequalities of power account for a significant tilt toward authoritarian workplaces that is the consequence not of individual choices but of the structural features of the market. As the market is structured, some people are going to have to occupy positions of lesser power whether they want to or not. Although there may be competitive equality of opportunity to occupy these positions, there is inequality of constructive opportunity among persons. The situation is like, though by no means as serious as, the society of masters and slaves discussed above. This suggests the remedy of worker participation to restore equality.

Again, it is worth remembering that in perfectly competitive and complete markets with egalitarian initial endowments, monopsony and inaccessibility to capital are not problems. The opportunities to form different kinds of association are not unequal. There is no greater difficulty in workers renting capital than in capital renting labor.Footnote ²⁹

(3) Worker participation as excluded by the need for a high return to capital. Worker participation is blocked by managers and owners who think that they will lose some of the rents of the firm.Footnote ³⁰ Rents, which are returns that are over and above what is necessary for a factor to be supplied, are ubiquitous in imperfect markets. In authoritarian firms, most of the rents go to the owners and managers. Worker participation redistributes power in the firm and has the effect of redistributing the rents of the firm, transferring income from capital and managers to labor. So even if a firm with worker participation is as productive as, or even more productive than, an authoritarian firm, there may be significant difficulties in raising the capital for the firm with worker participation. The suppliers of capital will try to avoid forms of organization with lower returns when they have alternatives that supply higher rents to capital. This, coupled with the asymmetries in monopsonist labor markets, will bias the market toward authoritarian structures, even if they are no more or even less productive than firms with worker participation.Footnote ³¹

(4) Productive efficiency. One argument against egalitarian governance is that such governance is not as productive in the market as authoritarian governance. That is, more resources and labor must be expended in order to produce the same or fewer goods for consumption. I accept this efficiency defeater for two reasons. First, I think the right understanding of the principle of equality must be an egalitarianism that is Paretian. Since equality of power matters because power matters, the principle of equality must include a ranking that places Pareto-superior states above Pareto-inferior states in terms of justice. Inequality is always unjust, but it may also be the case that a Pareto-superior inequality is less unjust than a Pareto-inferior equality.Footnote ³² The second reason is that I think of the market as a means for satisfying human needs and desires. A form of market organization that fails to do this well fails in one of its most important purposes.

I illustrate this with one of the most famous arguments that some firm structures are not very efficient and thus are unable to compete in the market with others. Henry Hansmann argues that worker-owned firms of a certain complexity are unlikely to survive in the market because the decision-making that is necessary to make good decisions in markets becomes too complex.Footnote ³³ He argues that complex firms with many kinds of workers are such that there is too much conflict among workers because there is too much diversity. This makes collective decision-making difficult, and so it tends to weed out such firms. He does not argue that all worker-controlled firms have this same problem; those that have relatively simple structures as a result of relatively simple decisions, can and do survive. Here we have an explanation of the presence or absence of worker-controlled firms that is mostly connected with the productive efficiency and the consequent ability of those firms to stay alive in a tough competitive environment. I do not want to endorse this explanation, but it does serve as an illustration of the kind of explanation I have in mind. The superior efficiency of a certain kind of firm is certainly a relevant consideration in favor of it.

There have been many discussions of the relative efficiencies of authoritarian and egalitarian firms. In the case of unions, many have pointed to the monopolistic supply characteristic of unions and have argued that they are inefficient. Others have pointed to the element of voice that is realized in unions and have argued that this can enhance the productivity of firms with unions by enhancing the circulation of information. The most sophisticated assessments I have seen are that sometimes unions detract from the productivity of firms and sometimes they enhance that productivity, depending on the institutional arrangement. On average, they do not detract from productivity but contribute to a more equitable distribution of power and income in society.Footnote ³⁴ As a consequence, though unions do not detract from productivity, they may be disfavored in market settings because they redistribute the rents of the firm away from owners and managers.

Worker cooperatives have often been criticized for having short time-horizons and for not maximizing profits but for maximizing the per capita income of workers, which would tend to slow down output.Footnote ³⁵ But empirical research on worker cooperatives has not borne out either of these predictions. It does seem to be the case that worker cooperatives are found mostly in labor-intensive and not capital-intensive industries, but it is not clear why this is so. One important observation is that worker cooperatives do not fail or turn into authoritarian firms any more often than authoritarian firms fail. The reason why they are rare has more to do with difficulty in starting up.Footnote ³⁶

Codetermination, which is a combination of collective bargaining and worker participation in choosing the people who run the company (without the workers having to own any of the capital), has also been criticized for suppressing innovation and thereby undermining productivity. But the evidence for this is mixed. In some cases, codetermination increases innovation. Also, codetermination is often credited with increasing overall productivity in firms because they tamp down conflict between labor and capital and massively increase information flows.Footnote ³⁷

So, although I think the efficiency defeater is a legitimate defeater, I do not think we have general reason to think that efficiency is a reason for not having worker participation. There is ample evidence that many firms that bargain with workers collectively and that firms that have significant worker participation are highly productive and efficient.Footnote ³⁸

These remarks are not meant to solve the issues of the productivity effects of worker participation; they are meant to suggest that there are viable institutional devices to implement worker participation. One further point to note about productivity is that in some cases, even if productivity is lower than in authoritarian firms, worker participation may still have a net welfare benefit. A firm that loses something in productivity, but that distributes power and income more equally, may enhance welfare on net because of the gain to the worse-off.

Another way to classify these rigidities is in terms of their relation to ideal theory. The rigidities in (1) are really cases of nonideal conditions at least as far as the equal capacity has it. The idea is that equality of capacity is not realized even in the background institutions, and therefore there is a rigidity in the market such that it is closed to more egalitarian arrangements. The rigidities in (2) and (3) are rigidities that can occur in an otherwise egalitarian society. They are illustrative cases of the situation in which equality of opportunity may not be adequately realized in a market, even when the background institutions are egalitarian. The case in (4) is one in which the presumption in favor of equality may be overridden, but I will discuss this more below.

X. The Argument for the Remedy in Favor of Workers’ Voice

I have been arguing as if no particular form of organization is required a priori by the principle of equal capacity for a firm in a market. If a firm is created by persons with equal opportunity and there are no relevant rigidities, then the firm is formed in an egalitarian way, even if it is authoritarian in structure. If there is no rigidity in favor of the authoritarian structure, then there is no problem in its formation. If there is a rigidity in favor of the authoritarian form, then we must make the egalitarian form available in some way by means of collective action.

The argument for this derives from the argument for equality of capacity. The first premise is that equality of capacity is a requirement of procedural justice in economic life. It accords with the egalitarian conception that says that people ought to have an equal say in the formation of the social world they live in. The second premise is that the market organization that makes unequal governance or power inevitable or very likely, makes it necessary that some people must give up their ability to shape the world around them to others. The market may not select ex ante who will be in this subordinate position, but it does say that some will have to be in this subordinate position. This implies that the society is structured in such a way that some must have a lesser say than others in the social world they live in. It undermines equal capacity. The third premise is that unregulated imperfect markets actually do realize inequality of power under many circumstances. Fourth, if there is a form of market organization that avoids unequal power by realizing equal governance without making everyone worse off, even in terms of ability to shape the social world, then the society must organize the market in that way.Footnote ³⁹ In this way, equal governance in the firm is a kind of substitute for equal power in the market.

XI. The Place of Capital

I have not challenged the idea that there is a place for the ownership of capital in an egalitarian society. All the forms of worker participation I discuss allow for the ownership of capital and for that ownership to give some right to control of the firm. What I do challenge is the idea that the ownership of capital gives one the right to full or even primary control over the firm that employs that capital.

If we look at matters from the standpoint of the egalitarian conception of markets and the resulting equal-capacity view, we can see here too that there is reason to confer some control on the creators of things and on the legitimate property holders. My being in control of property and of things I have created is clearly an important part of my having the opportunity to shape the social world I live in. But the egalitarian conception gives us some insight because it does not argue in an all-or-nothing way on this point. It can maintain, therefore, that while the owner should have some power in this context, others ought to have power as well. And I have argued that my being ineluctably in complete control of the uses of the things I legitimately own seems incompatible with giving a say to those who work with the things that I own.

An abstract example may help here. Suppose we live on an island. I own a piece of land legitimately and only you can farm that land. Suppose we disagree about how to do this best, and both our lives depend on it. The result here is that both of us will have a say in how to farm the land. This example displays a distribution of power that seems just and in accord with the egalitarian conception. This suggests that it is at least conceivable to have a system of equal power among owners and non-owners that is a fully voluntary system of cooperation among equals. The idea of perfectly competitive and complete markets whose participants have egalitarian endowments permits such a system of interaction among persons with equal power, whether those persons have capital or not.

Traditionally, several arguments have been advanced to support giving owners of capital exclusive control over their capital and a kind of authoritarian control over the firm. Some of these arguments work to assign to owners some control over the use of their capital, but none of them works to assign exclusive control. I will discuss them briefly to respond to potential objections to my position. I list the arguments here:

1. (1) The argument from managerial skill, expertise, and entrepreneurial ability.

2. (2) The argument from savings of legitimately acquired goods.

3. (3) The argument from legitimate creation of capital.

4. (4) The argument from the assumption of risk.

5. (5) The argument from capital as firm-specific investment.

6. (6) The argument from too many masters when other stakeholders have some control.

Let us remember what point these arguments are put forth to make. The question is whether workers ought to have power in the governance of the firm as a substitute for the lost equal power in a fully egalitarian market. That loss of power results from background inequality, monopsony, and the ability of capital to block worker participation, which is motivated by the unwillingness to share in the rents of the firm. When these three conditions do not obtain, in the case of perfectly competitive and complete markets with an egalitarian distribution of endowments, control over capital is not an issue.

The first argument points to the need for skill and ability in the running of firms to suggest that firms should be controlled by the owners. Strictly speaking, one could argue that this does not imply anything about the right to control the firm based on the ownership of capital. Experts and entrepreneurs do not need to own the capital they work with. But the main counterargument is that entrepreneurial ability and expertise are fully compatible with worker participation. Worker participation is compatible with a division of labor in the running of the firm, just as citizen participation is compatible with a division of labor in the running of a democratic state.Footnote ⁴⁰ Workers play a fundamental role in the structuring of the division of labor and with the standards that are used to evaluate occupants of the division of labor as owners do now, but they need not occupy all the positions in the division of labor.

The second and third arguments do suggest possible reasons for some control to be assigned to capital. A person who has genuinely saved up resources from her earnings and has an idea of how to use those resources for good does have some claim to control over the use of her capital.Footnote ⁴¹ And a person who has legitimately invented or created capital also has a claim to some control over its use. This follows from her interests in shaping the social world in which she lives. However, there is no argument here for complete control over her capital; the considerations are simply indeterminate when it comes to saying how much control there ought to be. And I have given reason for saying that the control ought to be significantly limited when others are otherwise ineluctably in a position of unequal power.

The argument that a capital owner assumes risk in being part of a firm seems to ground a claim to some role in governance as well, but it does not ground a claim to complete control. All the other parts of the firm also undertake risks in being part of the firm.

The same response holds for the firm-specific investment argument, which maintains that because the capital owner makes an investment in the firm, the owner cannot recoup by going to other firms.Footnote ⁴² It does ground some claim to control but clearly not a claim to complete control. Workers also make firm-specific investments.

Finally, the argument from too many masters, which alleges that including stakeholders aside from shareholders in the governance of the firm makes for a firm run by too many masters, which in turn implies all the difficulties of collective decision-making in a complex electorate, seems unlikely to succeed.Footnote ⁴³ First, shareholders are more diverse than this argument suggests. They have different risk profiles, and they often have different preferences for how the firm should be run, as we see in the current trend toward corporate social responsibility.

Is there a consequentialist argument for such complete control? I have argued in this essay that there are no general consequentialist arguments for capitalists’ complete control over their capital. The evidence suggests that worker participation is on average compatible with a very healthy economy. Indeed, one consequentialist concern favors worker participation; economies with worker participation have more equal distributions of income.