I. Equality and Freedom in the Workplace: Contemporary Discourse

In this essay, I focus on a domain in which the conflict between freedom and equality is at stake: the world of work. Current U.S. laws impose many negative liberty restraints on employers with respect to employees. They may not discriminate by race, gender, religion, or age. They must pay minimum wages to most employees, and pay overtime to those working more than forty hours per week. They must pay wages in cash, not scrip. They must bargain in good faith with any labor union their employees elect to represent them, and may not fire any employee for joining it or urging others to do so. They are subject to workplace safety standards imposed by the Occupational Safety and Health Administration (OSHA). These laws are justified at least in part in terms of equality: antidiscrimination laws directly aim to secure equality among workers; unionization laws aim to equalize workers’ bargaining power with employers; other laws aim to correct for unequal bargaining power.

Opponents of such laws object not only that they interfere with the liberty of employers, but that they interfere with the liberty of workers. “Right to work” laws, which forbid contracts between employers and labor unions that require all employees to pay dues to the union in return for its (legally mandated) bargaining on behalf of all, are framed as defenses of workers’ liberty. Libertarian and free market political theorists support this way of framing the normative stakes in employment laws. For example, John Tomasi complains that minimum wage and maximum hour laws interfere with the “freedom of individuals to negotiate personally the terms of their employment,” and disparages labor unions for diminishing individual agency in this regard.^{Footnote 1}

This discourse continues the framing of labor law issues expressed in Lochner v. New York, the famous Supreme Court case that struck down a maximum hours law for bakers. Lochner declared the law “an unreasonable . . . interference with the right of the individual . . . to enter into those contracts in relation to labor which may seem to him appropriate or necessary for the support of himself and his family.” Bakers “are able to assert their rights and care for themselves without the protecting arm of the State” and “are in no sense wards” of it.^{Footnote 2}

This framing assumes that perfect negative liberty for all is secured by what I shall call the “laissez-faire baseline” for the labor market. The baseline is set by a pure employment at will regime. Under employment at will, employers may hire and fire employees for any or no reason, and employees may accept a job offer and quit for any or no reason. Workers and employers are presumed to negotiate over the terms of employment against this baseline. For each term the employer wants from the employee — a particular task performed, late hours, toleration of some unpleasant work condition, and so on — it is assumed that the worker obtains some compensation, either as a result of negotiation, or of competition among employers for workers. Market orderings obtain independently of state action; the state’s only proper role is to enforce contractual agreements independently reached by the parties.

Five objections to state-mandated deviations from the laissez-faire baseline hang on this framing. (1) If the state decides to reserve some right for workers, such as limited hours, safe working conditions, or payment in cash, this infringes on both parties’ freedom of contract. (2) It violates employers’ property rights by restricting their freedom to run their businesses as they see fit, and (3) violates employees’ self-ownership by restricting which rights they can alienate. (4) It constitutes class legislation against employers by putting a thumb on the scales on behalf of workers. (5) Alternatively, it amounts to paternalistic and misguided interference with workers’ liberty, denying them what they judge to be the better compensation packages they could get if they could sell more of their rights to their employers.

Liberal egalitarians often reply to this argument by claiming that the unequal bargaining power between workers and employers makes the labor contract not truly voluntary.^{Footnote 3} Vulnerable workers have no reasonable alternative to acquiescing in grueling, dangerous, or humiliating work conditions. Some degree of equality is needed to ensure truly free contracts.

Thus, the standard frame casts this disagreement as a dispute over the relations of freedom to equality. The advocates of laissez-faire see attempts to secure equality of bargaining power as a threat to individual freedom. Liberal egalitarians see some level of equality as a prerequisite to individual freedom.

In this essay, I shall argue that neither the free market argument nor the liberal reply offers an institutionally adequate representation of the stakes in this dispute. Both accept a frame in which the critical issues are played out in negotiation over the terms of the labor contract. Neither side appears to notice that little negotiation takes place in most labor contracts. The typical worker, upon being hired for a job, is not given a chance to negotiate.^{Footnote 4} Nor is she handed a contract detailing the terms of the deal. She is handed a uniform, or a mop, or a key to her office, and told when to show up. The critical terms are not even what is said, but what is left unspecified. The terms do not have to be spelled out, because they have been set not by a meeting of minds of the parties, but by a default baseline defined by corporate, property, and employment law that establishes the legal parameters for the constitution of capitalist firms. Negotiated labor contracts mostly make only minor modifications to a relationship whose normatively critical features have already been set by law independently of the will of both parties, much as prenuptial agreements make minor modifications on the marriage “contract” whose fundamental terms are set by law.

Libertarians and liberal egalitarians have overlooked this point, because they share a defective representation of the institutional structure of capitalism: they conflate capitalism with the market, and therefore imagine that the labor contract is the outcome of market orderings generated independently of the state. State regulation of labor contracts is therefore seen by both libertarians and liberal egalitarians as an “interference” with market orderings. They disagree only on whether this interference is justified.

Missing from this picture is capitalist firms, and the essential role of the state in defining their forms. Markets are not distinctive to capitalism; they exist in all economic systems more sophisticated than a hunter-gatherer economy. Capitalism is distinguished from other economic systems by its mode of production. The labor contract is not properly seen as an exchange of commodities on the market, but as the way workers get incorporated under the governance of productive enterprises. Employees are governed by their bosses. The general form of that government is determined by the laws of property, incorporation, and labor, not by contract.

Thus, the fundamental normative issues concerning the relation of freedom to equality at work do not lie in arguments about the voluntariness of the labor contract. They lie in arguments about the legitimate form of government of productive enterprises. Since the state is needed to set that form, when it does so by reserving specific rights to workers in labor contracts, this does not amount to “interference” in free labor markets. It amounts to setting parameters for the constitution of labor governance within the firm, limiting the scope of legitimate powers that employers may wield over workers. When the state acts in this mode, it is like the Commissioner of Baseball specifying the rules that define the strike zone, not like a fan interfering with a fielder’s attempt to catch a fly ball.^{Footnote 5}

Once we recognize that workers lie under a kind of government in the workplace, we can ask: What is the form of a legitimate government for productive enterprises? What could make workers free under that government? Does that freedom require any kind of equality? To investigate these questions, I shall reconsider republicanism as a theory of freedom under government, particularly the egalitarian form of republicanism that arose in the mid-seventeenth century and continued for two hundred years. Republicans offered a sharp (and utterly non-Marxist) critique of the governance of workers by their bosses. Their critique was largely forgotten because they failed to offer a feasible remedy to the problems they identified in the system of wage labor. Nevertheless, republican principles of constitutional design offer some insights into possibilities for a constitution of liberty in the workplace.

In the following sections, I outline some principles of republican constitutional design (Section II) and discuss the connections between property regimes and different types of constitution for productive enterprises and the state (Section III). I argue that the case for the laissez-faire baseline cannot be grounded on first premises about private property. I then discuss the economic theory of the firm, which highlights the distinction between markets and firms and explains why capitalist firms are hierarchically organized (Section IV). I argue that, while economic theory explains why efficient firms need a hierarchy of offices, this fact underdetermines the constitution of the firm and provides no support for organizing it as a dictatorship. I then discuss the role of the state in determining the constitution of the firm through property, corporate, and labor law. I argue that these laws are a kind of public good and consequently are properly subject to democratic control in the public interest (Section V). The laissez-faire baseline cannot be justified from this point of view, since it makes the default form of workplace government a dictatorship. This lacks justification in any credible principles of liberty, property, efficiency, or justice. I conclude with some speculations about the ways republican principles could help us redraft a constitution of liberty and equality for the workplace (Section VI).

II. Republican Freedom and Equality

“Radical” republicanism is a type of egalitarianism that informed the political thought of the Levellers, Commonwealth men, John Locke, Jean-Jacques Rousseau, Thomas Paine, Emmanuel Joseph Sieyès, Chartists, and the Republican Party before the Civil War. It emerged from classical republicanism, which is inegalitarian. All republican theorists define freedom as nondomination. In the standard formula, “liberty solely consists in an independency upon the will of another, and by the name of slave we understand a man, who can neither dispose of his person nor goods, but enjoys all at the will of his master.”^{Footnote 6} This formula originated in hierarchical societies defined by distinct classes of persons: free on the one side, slave on the other, where “slaves” include chattel slaves, serfs, servants (employees), apprentices, wives, and children. Freedom refers to the social status of a person who is not subject to the arbitrary, unaccountable will of another — who can act without having to ask anyone else’s permission. An unfree person is anyone subject to another’s dominion: someone who must obey another’s arbitrary orders, whose liberty is enjoyed only at the pleasure of a master who can take it away without notice, justification, process, or appeal.

Classical republicans defined a free society as one in which there is a substantial class of free citizens who govern themselves, as in ancient Greek and Renaissance Italian city-states. Absolute monarchy, or any kind of rule by decree, is incompatible with freedom, because the ruler can take away anyone’s liberties for any or no reason. Recognizing the necessity of states but the dangers to free citizens of state power, republicans focus on questions of constitutional design: How can a stable state be instituted that will promote the public good, while securing the freedom of citizens against rulers’ domination? Republicans highlight three institutions: the rule of law, private property, and elections.

The rule of law imposes procedural constraints on governments: they may only act in accordance with laws, not personal whim; laws must be passed by constitutional procedures; enforcement must follow due process, with judgments subject to appeal. Republicans do not consider law as such to be a restraint on liberty. They draw the line between freedom and unfreedom not between license and law, but between the rule of law and arbitrary rule. As Locke put the point,

A primary function of law in republican theory is to secure the freedom of citizens by protecting their property. In contrast with negative liberty theorists, private property was not fundamentally important to republican citizens because of the things they were free to do with it. It was important for securing their status as free persons, by yielding an income sufficient to support themselves and their families, without having to work for someone else. To labor for another made one unfree, because workers are subject to the arbitrary, unaccountable orders of their employers. Thus, the private property that mattered for a free society was the property of the independently wealthy. According to classical republicanism, free citizens should not have to work at all, even for themselves, as self-employment in a trade focused the mind on base private interest and supposedly made people unfit to consider the public interest. Hence, members of the free class should support themselves through rents and the labor of their chattel slaves and servants.^{Footnote 8}

Elections aim to hold state officials accountable to free citizens, and make the state a public thing directed to the common good. The dependence of workers on the will of their employers, and of wives on their husbands, justified the traditional republican restriction of the franchise to male property owners. Before the institution of the secret ballot, wives and workers were not free to cast their votes independently of their husbands and employers, because their access to the means of subsistence depended on following their superiors’ orders. To enfranchise them would in effect give extra votes to the wealthiest, and turn the government into an oligarchy. That would violate the republican principle that freedom would be secure only if no distinct class among the electors could determine outcomes all by itself.

Thus, in republican theory, the rule of law secured citizens against arbitrary rule by the state; private property secured free individuals against the arbitrary rule of any private person; and elections made rulers accountable to citizens and thereby made the affairs of state a matter of public interest, rather than the private concern of state officials. Classical republicanism presupposed social hierarchy, class privilege, slavery, and subjection. To the extent that it cared for equality, it was only among the free, who jealously guarded their independence against threats of domination by other free people.

In the seventeenth century, economic changes in England enabled an emerging class to appropriate republican theory for egalitarian ends, and thereby make it radical. The rise of cities and commerce, along with enclosures leading to expulsion of feudal tenants from the land, created a motley class of “masterless men” — individuals who did not owe obedience to any particular social superior, but who were not masters themselves.^{Footnote 9} Lower members of this class included day laborers, itinerant peddlers, entertainers, squatters in forests, heaths, and wastes, cottagers, and vagabonds. The more advantaged members of this class were the self-employed who earned a steady income from a fixed establishment: yeoman farmers and long-term leaseholders, artisans, shopkeepers, printers. These individuals were masterless in economic affairs. Many were also masterless in spiritual affairs, having joined Protestant sects, some of which, such as the Baptists and Quakers, adopted democratic self-governance and a lay ministry, thereby abolishing clerical authority.^{Footnote 10} They demanded a share in government. If the republican justification of disenfranchisement was dependence on the will of a superior, the rise of masterless men posed a challenge to the property qualification for voting. Here was a class that, although not exercising dominion over anyone else (save their wives), was free of subordination. Their spokesmen, the Levellers, demanded a (near) universal male franchise.^{Footnote 11} They also demanded equality under the law — one law, passed by the House of Commons, for lord and commoner alike, with all accused subject to trial in common law courts. This abolished the Lords’ privileges, since the latter were exempt from arrest for most crimes and subject to trial only in the House of Lords. Locke’s egalitarian version of the traditional republican formula expresses the same demand — “to have a standing rule to live by, common to every one of that society” (emphasis mine).

Ideological support for radical republicanism drew from a confluence of republican reasoning with natural law, which asserted the equality of all persons in possession of a robust set of natural rights, and popular sovereignty over the constitution of any government to which they were subject. Locke played a pivotal role joining the republican and natural law traditions.^{Footnote 12} This tradition paid close attention to the connections among property, production, and freedom under government. This requires a closer examination of varieties of property regime.

III. Property and Governance

Property regimes may be classified according to their relations to the governance of persons. The state is only one type of government. Government exists wherever some have the authority to issue orders, backed by sanctions, to others. Thus, masters govern their slaves; feudal lords govern their serfs, tenants, and retainers; men in patriarchal marriages govern their wives; employers govern their employees; clergy govern their parishioners; parents govern their children. In classifying property regimes, we should consider how property rights relate to all forms of government, especially of work. This perspective allows us to sketch ideal types of the following property regimes: feudal, mercantile, republican, capitalist, and socialist.

The core principle of feudalism is that private property in land confers the right to govern its denizens. Feudalism is a complex system of fragmented, devolved sovereignty based on personal loyalty between sovereign and subject, lord and tenant. Formally, all the territory of the state is the monarch’s private property. The state secures its power by granting tenure in land and accompanying rights of dominion to lords, in return for their loyalty and service to the monarch. Lords obtain subordination from their denizens by granting rights of occupancy and use in return for duties of personal service to the lords. Government, whether of the monarch over his subjects, or of lords over their tenants, is private, arbitrary, unaccountable, and opaque. Each denizen may be subject to different rights and obligations, depending on the idiosyncrasies of the terms of his tenancy.

Whereas the state secures its power in feudalism by grants of land and privilege to landlords, it does so in a mercantile property system by grants of monopoly, tariff protections, subsidies, charters of incorporation, and other privileges to merchants and manufacturers, including rights to govern and control trade with private colonies abroad. The state secures loyalty of the rich from the fact that the latter owe their wealth to the state’s grants of privileges and protections against competition. Government, whether of the monarch over his subjects, or of the mercantile company over its colonies, is private, arbitrary, and unaccountable to its subjects. As in feudalism, different laws apply to different subjects, depending on the ad hoc deals made between sovereign and subject, but only the wealthiest get to strike such deals. Inferior subjects must accept whatever government is imposed on them.

A radical republican property regime aims to abolish private government by (1) breaking the link between private property and rights to govern other people, and (2) securing for all (male) citizens sufficient property that each can be self-employed, and thereby escape dominion by a landlord or employer.^{Footnote 13} Radical republicans therefore oppose feudal property, corporate monopolies,^{Footnote 14} serfdom, and slavery. Locke designed his labor theory of property acquisition to counter claims to feudal property by sovereign grant, and claims to govern others in virtue of private property rights. Radical republicans also aim to minimize resort to wage labor, as this undermines the independence citizens need to participate as equals in public affairs. The property regime therefore aims at a wide distribution of property and limits to its concentration, so no one is consigned to wage labor, and the rich do not capture the political process.^{Footnote 15} This ideal came closest to realization in the United States in the North before the Civil War. In the Republican Party’s vision, wage labor would be merely a temporary stage of young adulthood, with wages high enough to permit saving to attain self-employment as a yeoman farmer, artisan, shopkeeper, or independent professional. The promise of homesteading in a free soil frontier kept the republican hope alive longest in the United States.^{Footnote 16}

Commercial republicanism offers a moderate variant, more optimistic than Rousseau about the compatibility of commercial society with republican values.^{Footnote 17} Less concerned with extension of the franchise and popular political participation than radical republicanism, it extolls the promise of commerce for securing people’s independence from subjection to arbitrary government, whether of the state or of private parties. Adam Smith was its leading advocate. Smith opposed all forms of involuntary servitude, including slavery, serfdom, and apprenticeship.^{Footnote 18} His chief defense of commercial society was that it promoted good government and the personal independence of workers by liberating them from subjection to arbitrary lords. Before the rise of commercial society, lords had no other way to spend their income than to directly maintain their serfs, tenants, servants, and retainers. They received the pleasures of dominion in return. As commercial society made more consumer goods available, lords shifted their spending to luxuries, and people left their lords’ estates to become traders and artisans, thereby gaining personal independence. As workers moved from agriculture to cities, the remaining tenants worked on larger plots, while the lords raised rents. Tenants accepted this only in return for longer leases and release from duties of personal service. Thus the farmers too cast off subordination to the lords. Workers’ liberation from “servile dependency upon their superiors,” along with good government, is “by far the most important” of the effects of commercial society.^{Footnote 19}

Commercial republicanism secures widespread personal independence through a property regime that supports self-employment. This requires free markets in consumer goods and land, because state-granted monopolies and privileges, and property rules such as entail and primogeniture,^{Footnote 20} concentrate the means of production in a few hands and thereby force the rest into dependency. Four features distinguish Smith’s vision from laissez-faire capitalism. (1) Economies of scale are rarely significant. The great virtue of free trade — the abolition of state-granted monopolies, tariffs, and other protections — is not merely that it allocates resources more efficiently, but that it dissolves concentrations of wealth and thereby multiplies opportunities for independent producers. (2) The corporate form, and consequent importance of stock markets for raising capital, is sharply limited in scope. Smith criticized joint stock corporations for negligent mismanagement of stockholders’ capital, conspiring to restrain trade, rent-seeking, and provoking foreign wars. He thought they were justified for only four types of routine, non-entrepreneurial business: banking, insurance, canals, and water utilities. These were the only businesses that required the huge concentrations of capital that joint stock corporations raise.^{Footnote 21} (3) Labor markets are small. While not as hostile as radical republicans to wage labor, Smith’s leading argument for the value of free markets in commercial society depends on their support for self-employment. Furthermore, his critique of the stultifying effects of a fine-grained division of labor^{Footnote 22} raises doubts about the value of scaling up production too far. The famous pin factory that Smith praised for its productivity-enhancing division of labor had only ten workers.^{Footnote 23} Such small-scale enterprises could still support a robust republican culture of workers’ independence, since they could be run on a collaborative basis.^{Footnote 24} (4) Smith supported pro-labor state regulation: when a regulation of “the differences between masters and their workmen . . . is in favour of the workmen, it is always just and equitable.”^{Footnote 25} His example of a just labor regulation — requiring employers to pay workers in cash rather than in kind — illustrates the importance of regulating labor contracts for securing workers’ independence. To be paid in goods chosen by one’s employer is to submit to the employer’s regulation of one’s private life.

Capitalism and socialism are distinguished from these older property regimes by the large scale of productive enterprises, requiring a fine-grained division of labor within the firm. These property regimes facilitate capital concentration and vertical and horizontal integration. Both systems aim to reap the benefits of the Industrial Revolution, which realized immense productivity gains from increasing economies of scale. The distinction between capitalism and commercial republicanism is thus found not in free markets, but in the scale and structure of production. Capitalism is marked by the ubiquity of corporations (and similar forms of capital conglomeration, such as trusts), capital markets, and labor markets. Capitalism undermined the radical and commercial republican ideals by destroying their material basis in a self-employed workforce. It dramatically diminishes opportunities for individuals to attain independence by founding their own businesses. The overwhelming majority of workers are subject to their employer’s governance.

State socialism replaces market orderings for factors of production and produced goods with direct state ownership and control of large-scale productive enterprises, and centralized planning of production and distribution. The remaining varieties of socialism and capitalism rely on markets to guide (mostly privately owned, for-profit) firms’ decisions. Varieties of market socialism and capitalism may be defined by their constitutions of firm governance. The government of workers is democratic when workers own and control the firm. This is market socialism.

The government of workers is despotic when workers are slaves or peons, the legal or virtual property of masters, and have minimal rights. Marx supposed that slavery is incompatible with capitalism, because his stage theory of history tied unfree labor regimes to pre-industrial modes of production. He was wrong. The first commodity created by mass industrial processes was sugar, produced by slaves. The slave plantations of the U.S. South, the Caribbean, and Brazil were not throwbacks to a quasi-feudal era. They were large-scale, entrepreneurial, commercial enterprises, producing cash crops for a global market, vigorously introducing new technologies to increase productivity and maximize profits. Capitalism does require factor markets, including labor markets. But it does not require that the sellers of labor are the workers. In the capitalist form of slavery, competitive labor markets exist, but labor is capitalized and not self-owned.

The government of workers is dictatorial under laissez-faire capitalism. Its core principle is that private property in capital confers the right to govern employees by fiat. Nonexecutive workers have no voice in the governance of the firm, but must follow their bosses’ orders, on pain of demotion or discharge. Because employment at will entails that bosses may fire or demote employees for any or no reason, firm governance is private, arbitrary, and largely unaccountable to the workers. As in the feudal and mercantile systems, different orders apply to different people. While elite workers may strike ad hoc deals with their employers (as knights did with lords), the rest (if not represented by a union) are subject to dictatorship. However, workers are formally free and hence consent to enter the firm and may freely quit. Workers may choose their Leviathan, but only Leviathans are in most people’s opportunity set. Other varieties of capitalism may be defined by their modes of worker governance. In non-laissez-faire forms of capitalism, the law may reserve rights to specific work conditions, benefits, or equal treatment to workers, and provide for union representation or worker rights to a voice in management.

Whenever advocates of one mode of production have challenged another, each side (other than state socialists) has appealed to the rights of private property in defense of their system. Given the variety of private property regimes, it is arbitrary to attempt to justify any particular mode of workplace governance by taking intuitions about rights of private property as first premises. Free market advocates err in supposing that the principles of private property advanced by Locke or Smith support capitalist modes of government. Their principles were republican, not capitalist, and are skeptical about deriving rights to govern other people from private property in land or capital.^{Footnote 26}

IV. Markets and Hierarchies

Most current discussions of labor freedom neglect questions of enterprise governance, because they represent the critical normative features of this problem as situated in market contracts rather than enterprise organization. Did workers enter the contract under duress or deception? May they freely exit? This frame falsely assimilates the employee to an independent contractor, and thereby effaces the power relations entailed by employment. Independent contractors own their own tools, set their own hours of work, work for a variety of clients of their own choosing, and decide how to do their jobs without direction from their clients. This is not a governance relation. When employees accept a job, they leave the market and enter a hierarchy within the firm, subordinate to their boss. This is a governance relation.

Why do such relations exist in production? First, consider why firms exist. Why isn’t production managed by self-employed independent contractors? Why doesn’t every worker at each stage of production contract with others to buy inputs, and produce an output sold to a worker at the next stage of production? Indivisible capital goods bar such a system. Modern production involves the use of large-scale equipment and infrastructure, such as assembly lines, airports, and banks, that cannot be divided up and independently operated by individual workers, but which can only be used by teams of closely cooperating workers. No set of contracts, however detailed, can successfully coordinate all stages of production.^{Footnote 27} In a production process with a complex division of labor, innumerable contingencies arise that require workers to alter their routine. Who should do what if the machine breaks down, if a coworker fails to show up, if too many customers are waiting in long lines? It is not merely costly but impossible to specify all contingencies in detailed labor contracts. Firms arise at the point where production requires closely coordinated and open-ended cooperation, and complete contracts cannot be drawn.^{Footnote 28}

These considerations help explain the boundary between the market and the firm, between contract and governance. They do not explain why that governance is hierarchical. Why aren’t firms run as an egalitarian participatory democracy, where no one has authority over others, and all work decisions are made collaboratively? Nonhierarchical groups face huge transaction costs in allocating contractually unspecified tasks to particular workers and ensuring that they get done. Without authority relations, responsibility for dealing with unforeseen contingencies is diffused. If several problems call for multiple deviations from routine, who should do what? While joint agreement may be reached if everyone is oriented toward the collective good, the costs of reaching agreement may be high. Moreover, teamwork raises problems of shirking that require monitoring and sanctions, which are difficult and costly for egalitarian groups. Authority relations can overcome these problems. Make managers responsible for determining how to divide the labor among workers and orchestrate their cooperation, for noticing problems and issuing orders to fix them, and for detecting and sanctioning shirking.^{Footnote 29} For this to succeed, the labor contract must be open-ended, not fully specified. At its simplest, it is an agreement to obey managerial orders, whatever they may be.

Given such efficiency considerations, one might suppose that the emergence of the hierarchically governed firm can be explained as the outcome of free market competition, without requiring any state intervention. Let workers and capitalists bargain freely over the terms of employment, and workplace dictatorship emerges as the most efficient mode of firm governance. This does not follow. While efficiency considerations may require some form of hierarchy, they do not entail that those in authority exercise arbitrary power over their workers, entitled to issue any orders other than to commit crimes, on pain of job loss. While participatory democracy may be inefficient, it does not follow that the workplace cannot be governed as a representative democracy, with workers electing managers, and managers limited by rule-of-law constraints on the orders they issue to workers. Efficiency considerations underdetermine the constitution of workplace governance.

Why do workers in capitalist societies work under a dictatorship, rather than a democracy or some less authoritarian form of government? The next section argues that the answer is not that workers accept it because they can get better pay and benefits from firms organized this way. It is because state laws make it so.

V. Capitalism, the State, and the Laissez-Faire Baseline

People can create property conventions and operate markets without relying on a state. It does not follow that capitalism needs the state only to enforce the property conventions and contracts that people devise independently of the state. Markets and property can exist without state action, but capitalist property and markets cannot.

This fact is evident from what spontaneous property conventions and markets look like when they arise without state action. Across the developing world, millions of poor people house themselves in extralegal squatter settlements, obtain their necessities in extralegal markets, operate extralegal services, from dentistry to transport, and conduct most of their economic lives outside the law. Rates of self-employment are high, entrepreneurs are constantly innovating, yet nearly everyone is poor. They are poor in part because informal property and contract conventions hold only locally; they cannot support economies of scale or take advantage of the gains from trade with distant strangers. Hernando de Soto makes the connections among formal (state-sanctioned) property, capitalism, and scale explicit in arguing that to enable their property to function as capital, as an asset that can be utilized to build wealth, people need the state to formalize their property rights.^{Footnote 30}

Formalization entails two operations: (i) a written record of what people own, and (ii) standardization of property rights to fit the legal conventions of the larger society and create a single integrated property system. Both are needed to enable people to conduct mutually beneficial trade with people outside their face-to-face communities.^{Footnote 31} Written records enhance the certainty of outsiders that the people they are dealing with are entitled to what they say they own. They locate property in a representative scheme that facilitates transactions with strangers — for example, by means of legal addresses. Addresses make people eligible to be billed for services, subject to being tracked down and held accountable for their contracts, and so make people more trustworthy. Records enable property transfers at a distance, without having to take physical possession or meet face-to-face with the other party. They enable abstract transactions such as borrowing against a mortgage, and dividing property into shares. Standardization involves overriding idiosyncratic property norms that arose locally, to enable integration of property into the larger formal system. In the United States, this was achieved by numerous Congressional preemption statutes, which replaced varied local laws with uniform federal law.^{Footnote 32} Standardization enables comparison of different properties, and thereby makes a given parcel fungible, more easily assigned a market price.

These advantages of formalization concern scale. “The problem with extralegal social contracts is that their property representations are not sufficiently codified and fungible to have a broad range of application outside their own geographical parameters.”^{Footnote 33} Unified property records and standardization enable networks of cooperation and trade to be dramatically scaled up. The gains from secure and informed trade between distant people tend to be greater than between neighbors, because strangers are more likely to have access to different information, skills, and resources, and because they are more likely to have different tastes and face different relative prices for goods.

Corporate law provides the key to scaling up productive enterprises by enabling the concentration of capital and close coordination of many workers. By standardizing the parameters of corporate governance — the rights and obligations of shareholders, boards of directors, and executive officers of a firm — corporate law enables multiple strangers to invest their money with confidence that their share will be protected, without having to pay high transaction costs in negotiating ad hoc governing arrangements with other shareholders, or retaining lawyers to advance their interests. It enables shares to be sold without buyers having to check the details of the corporate arrangement. Employment laws further define governance relations between managers and employees. Thus, the state is needed to supply the framework for the constitution of government for employees in a capitalist system.

What is that constitution? Corporate law vests the right to govern the participants in a business enterprise in the owners or trustees of its capital. However, given the separation of ownership and control in publicly held corporations — also embedded in corporate law — day-to-day governing authority is vested in the CEO. For most purposes, the CEO is the firm’s dictator, who delegates governing authority over workers to subordinate managers, much as a feudal king delegates governing authority over denizens to lords. Limits on sanctions and laws securing rights for workers limit the de jure authority of CEOs compared to lords. Exit rights, negotiation, and competition for workers reduce their de facto authority, to different degrees for different workers. Nevertheless, feudal and capitalist systems alike ground the private governance of persons in ownership of property, and deny the governed a voice in their government.

In the laissez-faire baseline for governance of workers, employment at will entitles management to fire workers for any or no reason. The resulting authority of employers is sweeping. Unless a right is specifically reserved to workers in law or a negotiated contract, employers enjoy legal power to govern their employees’ lives both on and off the job. Workers can and have been fired for having a gay partner, for speaking a language other than English, for (potentially) exercising their right to sue their daughter’s rapist.^{Footnote 34} Confirming republican worries about relations of domination corrupting elections, bosses can and have ordered their workers to attend rallies in support of political candidates the workers oppose (at the loss of a day’s wages),^{Footnote 35} and have fired them for supporting political candidates the bosses oppose.^{Footnote 36} They may fine workers who refuse to share their sexual histories and reproductive plans with third parties, and who do not diet and exercise according to corporate wellness programs.^{Footnote 37} Until recently, bosses could and did order their workers to have sex with them, and forbid them from urinating during working hours.^{Footnote 38} While the maximum penalty employers can impose for disobedience is job loss, the cost of job loss to most workers is severe.

Current U.S. employment law incorporates numerous deviations from the laissez-faire baseline. The deviations, however, preserve the open-ended dictatorial power over workers defined by the laissez-faire baseline in matters not specified by law.

VI. The Laissez-Faire Baseline and Alternatives

In the standard frame for evaluating employment law, whenever the state vests some pre-contractual right in workers with respect to management, this infringes on the workers’ freedom of contract as much as the employers’. Freedom of contract requires the laissez-faire baseline. I have argued that this frame misrepresents institutional realities. The laissez-faire baseline is not a product of bargaining between employer and employee. It is the baseline against which any bargaining takes place, set by state law. It is thus on a par with alternative legally determined workplace constitutions. Corporate and employment law supplies an infrastructure for capitalism. It is a public good provided by the state. As such, its structure is properly subject to evaluation and control by democratic processes.

The laissez-faire baseline establishes dictatorship as the default constitution of workplace governance. The employer can legally order the employee to do anything that is not against the law, and may make the employee submit to any condition other than criminal victimization by the boss. The critical term of the labor contract is thus what is not specified, not negotiated but legally presumed — namely, an open-ended agreement to follow orders. This is why most workers are hired without negotiation or a written contract: the law has already specified its default terms. Upon entering the government of their employer, workers under the laissez-faire baseline lose all liberties relative to their employer, except those specified in a negotiated contract, along with the right to quit.

Consider an analogy. Under the law of coverture, which defined the terms of the marriage contract until the end of the nineteenth century, men had dictatorial control over their wives. Husbands were entitled to control their wives’ movements, forbid them from engaging in wage labor, confiscate any wages they earned, and rape them. Marriages were void if women did not consent, divorce was nearly impossible, and the terms of the marriage contract could not be modified. Imagine a variation on this governance regime in which either spouse could divorce at will, and the marriage contract could be altered by a prenuptial agreement. This is analogous to the laissez-faire baseline. Suppose feminists demanded further modifications of the legal baseline, guaranteeing married women freedom to engage in paid employment, keep their wages, and move about freely, and securing their rights against marital rape. It would be absurd to argue that this further modification violates women’s freedom of contract.

It would also be absurd to argue that the modifications paternalistically interfere with married women’s freedom, or make them worse off overall. The modifications are necessary to preserve wives’ freedom from paternalistic or abusive interference by their husbands, and enable them to preserve their status as free persons. Nor is it proper to complain that the state, in making such modifications, has engaged in a form of illegitimate gender-biased legislation, playing favorites with women. Since the default reserves to men virtually all rights over their wives, and reserves only those rights for women that are specified by law, even the imagined modifications of traditional marriage law still leave immense power in men’s hands. Similarly, labor laws that reserve specific rights to employees are not biased in favor of workers. They barely make a dent in the dictatorial control that the laissez-faire baseline reserves for employers.

Radical republican theorists objected to wage labor because it subjects workers to employers’ domination. Republicanism was supplanted by liberalism in part because, in committing itself to self-employment or nonhierarchical cooperative production, it could not generate the immense economies of scale of the Industrial Revolution. Workplace hierarchy is needed for this. Furthermore, efficient labor contracts cannot be fully specified. They involve a somewhat open-ended agreement to follow orders. This creates an inherent danger in the capitalist workplace that managerial authority will exceed any bounds that could be justified in the public interest, and be used to oppress or humiliate workers, and indulge bosses’ desires to exercise dominion over subordinates.

This is a danger about which libertarians and egalitarians should agree. I have long argued that a fundamental concern of egalitarians is to abolish relations of domination and subjection.^{Footnote 39} Libertarians also find such relations suspect. Gerald Gaus persuasively argues that resentment of contemptuous bullies in defense of personal independence is the core moral sentiment driving egalitarianism.^{Footnote 40} Radical republicans would recognize and applaud the sentiments of Gaus’s hunter-gatherers. Deirdre McCloskey’s stress on interactions in which the parties acknowledge each other’s personal dignity fits into this picture.^{Footnote 41} She is right that the market is a major, indispensable domain in which individuals can realize their dignity. Commercial republicans, and many radical republicans, were free-traders for this reason.

Yet these republicans saw something that has been lost from the view of most mainstream libertarians: that hierarchical firms are distinct from markets, and often threaten the dignity and personal independence of workers. The common representation of the institutional structure of capitalism, which confuses hierarchies with markets, taints markets in the eyes of egalitarians, who are attuned to the humiliations and abuses many workers suffer on the job. It also induces many libertarians to turn a blind eye to what goes on there, on the assumption that it’s all the product of consensual, negotiated agreements between the parties, an expression of inviolable property rights, or the efficient outcome of market competition. By exposing the coercive hand of the state in constructing workplace hierarchy, I hope to spur mainstream libertarians to scrutinize the workplace with a more critical eye, and be more open to constitutional reform of workplace governance.

Not all libertarians have overlooked this problem. Kevin Carson, at the anarchist margins of libertarianism, articulates a critique of workplace hierarchy that originated in radical republican thought, and also stresses the role of the state in supporting it.^{Footnote 42} I part ways with Carson’s skepticism about efficiency gains from large-scale hierarchical production, however. While much of the abusiveness of hierarchy is an expression of bosses’ love of dominion, and may even undermine efficiency, not all hierarchy is like this. I therefore call not for abolishing but for taming workplace hierarchy. Although the republican remedy against workplace hierarchy is not viable, republican ideas about constitutional design can help us think about where and how to draw the line between legitimate managerial authority and illegitimate domination. There are three general strategies for doing this: exit, voice, and state regulation.^{Footnote 43}

Consider, first, state regulation of the substantive terms of the employment relation, via reform of the legal constitution of workplace governance. A republican reform would be guided by Locke’s definition of freedom under government: “(1) to have a standing rule to live by, (2) common to every one of that society, and (3) made by the legislative power erected in it; (4) a liberty to follow my own will in all things, where the rule prescribes not; and (5) not to be subject to the inconstant, uncertain, unknown, arbitrary will of another man” (numbering mine). Workplace governance cannot satisfy condition (4), since workers must coordinate their activities around a common end and cannot be generally free to choose their own ends at work. However, progress can be made toward subjecting the authority of managers to rule of law constraints so as to protect workers against arbitrary rule (5). Antidiscrimination laws are akin to having equality under the law (2). We may identify some additional rights reserved to workers, which employers could not transgress. This is analogous to a bill of rights against the king. A minimal set would include workers’ freedom to order their off-duty lives as they see fit, without sanction from their employers. This follows from the fact that the only justification for workplace hierarchy is productive efficiency.^{Footnote 44} The interest in productive efficiency does not require snooping into or regulating workers’ off-duty lives, since their productivity can be directly observed at work. There is also no public interest in authorizing bosses to subject workers to humiliating and degrading work conditions, such as sexual harassment or prohibitions of basic physiological functions such as urination.

Libertarians prefer exit over state regulation as a means to protect workers’ interests. Given that the legal constitution of the workplace is a state-provided public good, exit is not so much a substitute as a supplement to a bill of worker rights. Suppose we accept a bill of rights reform as the default but allow both sides to exit that arrangement by explicit agreement. This might be justified if a bill of inalienable worker rights would inefficiently bar gains that could be obtained if workers could trade their rights away. Workers’ rights to exit firms would ensure that they were fully compensated for any infringement on their rights.

I am more skeptical than libertarians of relying so heavily on exit, because workers bear substantial costs of job search, acceptance, and loss, and often lack important information about options. These labor market frictions endow employers with market power over workers, reflected in a somewhat inelastic supply curve of labor to the firm. Because they exercise some monopsonistic power, they can impose adverse work conditions without fully compensating workers.^{Footnote 45} Workers at the bottom, who enjoy no opportunities for negotiation, would simply be handed a contract of adhesion conditioning the job offer on waiving their rights. A regime of wholly alienable rights would return us to feudalism — of private, arbitrary government, with a different law for every person based on separate contracts with each individual, and benefits secured in return for personal subjection.^{Footnote 46} If this is no way to constitute a state government, it is also no way to constitute a workplace government. While the democratic state poses a greater danger than workplace governments in the severity of sanctions it can impose, workplace governments pose a greater danger in the sweeping scope and minuteness of regulation of workers’ lives on (and sometimes off) the job. A bill of inalienable rights is needed to constrain both kinds of government.

A third option would enhance the voice of workers in the constitution of legislative power within the firm (Locke’s condition 3). Numerous managerial decisions involve legitimate tradeoffs between productive efficiency and workers’ liberties that could not be handled by a bill of rights. Because employers exercise market power over workers, any workplace authority vested exclusively in management will not give sufficient weight to workers’ interests. Vesting authority exclusively in workers may not give sufficient weight to the interests of the owners of a firm. Republican theorists argued that, in societies composed of distinct classes, the best form of government would be “mixed” — that is, vest each class with distinct authority. (Thus, England mixed monarchical, aristocratic, and democratic forms of government in its constitution of King, House of Lords, and House of Commons. The U.S. Constitution, with an independent president and bicameral legislature, offers a variation on the theme of mixed government.^{Footnote 47}) Germany’s system of co-determination, in which workers elect some representatives to managerial and board positions, may offer an approximation to this model.^{Footnote 48} My point is not to endorse it, but to highlight the feasibility of experimenting with alternatives, if full workplace democracy is rejected. As a pragmatist, I do not think optimal constitutional designs can be settled by purely theoretical argument. Experimentation is needed to see what works.

Establishing limits to employer authority over workers, so it functions only in the public interest and does not subject workers to their employers’ arbitrary will, is one of the great outstanding problems of political economy. Neither the laissez-faire baseline, nor the current mix of state regulation, minimal worker voice, and heavy reliance on exit, secures workers against domination. I have argued that the prevailing discourse of liberty and equality in the domain of work misrepresents the issues, because it conflates markets with production, contracts with governance. Neither the doctrine of liberty of contract, nor a priori theories of property rights, offer sound ways to balance managerial authority and workers’ liberty. The question is about constitutional design for legitimate workplace government. I have suggested that, despite their failure to come to terms with the necessity of hierarchy in governing large-scale productive enterprises, republican theories of constitutional design, focused on limiting authority in the interest of freedom-as-nondomination, remain relevant for devising solutions to the problem of workplace governance. If this is so, then limits on social inequality are necessary for freedom.