Power in the employment relationship: Why contract law should not govern at-will employment

Download PDF

Unequal Power

Part of the Unequal Power project, an EPI initiative to reestablish the understanding in law, politics, economics, and philosophy, that equal bargaining power between workers and employers does not exist. Recognizing this inherent workplace inequality will bolster freedom, economic fairness, workplace protections and democracy.

Executive summary

At-will employment is the most common kind of employment relationship in the United States. When employment is “at will,” the employer and employee have a right to walk away for any or no reason at any time. The common law—the law made by judges when deciding cases—presumes that employment is at will unless the parties provide otherwise.

This paper critiques the common tendency of courts to label and attempt to treat at-will employment as a “contractual” relationship when resolving employment disputes. At-will employment is not contractual, and labeling and trying to treat it as such makes it harder to regulate employment, reinforces and legitimizes the inherently unequal power between employees and employers, and disguises political issues as doctrinal ones.

The nature of at-will employment: Unequal power and power struggles

At-will employment reflects systematic power imbalances between employers and workers because it is a creature of capitalism. The rules of capitalism both create and defend a distinction between the propertied and the propertyless, and they condition life opportunities on this distinction. Most workers enter at-will employment relationships because they do not have any property apart from themselves to bring to the market. They must bring something, however, since access to most of life’s necessities requires market exchange, and so they bring themselves. In exchange for pay, they offer to work to increase the value of another’s property, under the other’s command, for so long as the other wishes. The employee has a formal right to quit, but this right seldom carries with it a proportional, countervailing power to that of the employer.

At-will employment is also characterized by power struggles. The employee formally agrees to “work” under the authority of the employer, but to work is simply to exercise one’s faculty for taking voluntary action toward some end. The object of this agreement is not separable from the individual, so it is not capable of being exchanged through a market transaction. The individual’s ability to take purposive action—whether it be the physical, cognitive, or emotional dimensions of this ability—is an inalienable part of the person. Within employment then, the more that the employer attempts to treat this ability (its “purchase”) as an object of its acquisition and dominion, the more the worker will be deprived of autonomy, dignity, and sometimes health and life. The worker often resists.

At-will employment is not a contractual relationship

At-will employment did not develop out of contract law. The appellation of “contract” was attached to employment in the 19th century as a palatable and expedient means of extending the master’s authority over a servant to an employer’s authority over the commercial enterprise. Employers sought authority over workers that neither contract law nor property rights could justify. Thus, the U.S. legal system created a sui generis category to accommodate the employer.

The common law defines at-will employment as an agreement by one party to work for another, under the other’s authority, in exchange for pay, for as long as either party wishes the relationship to continue. By definition, at-will employment is nearly the opposite of a contract:

  • First, it lacks a “promise,” meaning a commitment by at least one party. Since either party may quit the relationship at any time for any reason or no reason, at-will employment is missing a promise.
  • Second, it lacks an ex ante bargain, meaning some agreement, entered at the outset of the relationship, regarding what the parties are going to trade or do for one another. For an agreement to be enforceable, contract law requires some minimal certainty—“definiteness”—as to what the parties are going to exchange. The parties must reach agreement on this at the time they enter the relationship, not later. Otherwise, if the parties later dispute their obligations, a court will not be able to determine if the parties lived up to their agreement. However, at-will employment is more “indefinite” than other agreements the law recognizes as contracts. In agreeing to work under the employer’s command, the employee agrees to subject her very capacity for contractual agreement to the employer’s authority. In other words, the employee says, “I permit you to choose the bargain for me at a later time.” The employer chooses the bargain by commanding the employee in her work as the relationship proceeds. The exchange is for whatever labor the employer decides to, and is able to, extract from the employee. For no other kind of contract does the law assume or permit that the bargain was not to have a bargain.

To address the first issue, the lack of a commitment, many jurisdictions posit that at-will employment is not really an at-will relationship but rather a series of one-day contracts. Under this construction, which appears to resolve the oxymoron of an “at-will contract,” the employer does not have the authority to terminate an employee at will. The employer simply has the right, like anyone else, to choose not to enter a new contract with the employee the next day.

Constructing at-will employment as a series of one-day contracts does not resolve the misfit between a contract and at-will employment, however. For one thing, it does not touch on the problem of the missing ex ante bargain. Furthermore, the law does not take seriously the one-day construction; it resorts to this construction as an expedient fiction only in certain kinds of employment disputes, and the outcome is usually favorable to the employer. This construction also fails to provide a realistic picture of how most at-will employment is organized in practice. At-will employers often structure work practices and compensation in anticipation of an ongoing relationship. The relationship is rarely what economists would call a “spot transaction.”

The consequences of labeling at-will employment a contractual relationship

Legal, political, and mainstream discourse often identifies at-will employment as a contractual relationship. By doing so, it usually sides with the employer in the power struggles within at-will employment.

The contractual designation surrounds at-will employment with the ideological aura of “freedom of contract.” The precise content of this ideology carries no consensus; however, it usually involves a suspicion of state regulation, at least where the regulation would disrupt a status quo that benefits the powerful. The association of at-will employment with the clarion call of freedom of contract has at least two consequences:

  • It deters regulation. Associating at-will employment with freedom of contract is likely a deterrent to enacting regulations to protect workers, to enforcing them adequately, and to ensuring that decision-makers interpret them broadly.
  • It justifies the employer’s at-will power. By associating at-will employment with freedom of contract, the contractual label tends to legitimize the employer’s power. Judges have invoked freedom of contract to justify the employer’s at-will authority.

The consequences of trying to impose a contractual framework on at-will employment in legal disputes

In several kinds of disputes between employers and employees, courts try to impose a contractual framework on at-will employment to resolve the dispute. This attempted framing offers little benefit to employees and instead reinforces and legitimizes the employer’s power.

Offering little benefit to employees

Attempting to treat at-will employment as a contractual relationship does not really disturb the imbalance of power between employees and employers or lead to more equitable terms and conditions of work. Contract law is not designed for these purposes. Nor does it guarantee employees the key benefit of a contract—legally enforceable expectations about the future conduct of another. Under the attempted contractual framework, there is:

  • No limit on the employer’s power of termination.
  • No right to expect the employer to abide by its policies and practices. Employers often issue written documents to at-will employees setting out performance expectations, compensation, benefits, discipline, and other employment policies. The employer generally expects employees to abide by these policies, and employees often rely on them, particularly if an employer has regularly followed them in the past. However, the principles of contractual interpretation usually enable the employer to defeat employee claims that it has assumed enforceable obligations to employees. In most jurisdictions, the employer can defeat these claims by including a disclaimer confirming that the relationship is at will and/or that its policies are not contractual. The use of such disclaimers is now common practice among employers. Often, the employer’s only enforceable obligation is to pay for completed work.
  • No right to expect that the terms and conditions of work will be stable or predictable. The point of a contract is to fix the parties’ obligations to one another within their bargain. However, trying to impose a contractual framework on at-will employment generally does not disturb the employer’s power to make up and alter the main terms of the relationship as it goes along. This is the flipside of the employer’s ability to avoid assuming contractual commitments to employees. The at-will employer can generally dictate and alter working schedules, the pace of work, the kind of work, and other important conditions when it pleases. The court’s usual authority to determine what goes into contractual “gaps” is largely inapplicable in at-will employment.
  • No right to expect the employer to act honestly, rationally, or without opportunism. Contract law imposes a duty on all contractual parties to act in “good faith.” This generally means that the parties should not be irrational, dishonest, or opportunistic toward one another in carrying out their contractual duties. However, in most jurisdictions, the duty of good faith is inapplicable to at-will employment. For example, the at-will employer may terminate an employee based on false accusations.
  • No right to expect the employer to abide by binding commitments. Sometimes employers do assume binding obligations to at-will employees. However, contract law does not usually disturb the employer’s unilateral power to alter its commitments on a prospective basis. Contract law does regulate how parties can modify their contracts. For example, a party seeking to modify a contract must act in good faith, which often requires the party seeking the modification to offer the other party something additional to what it already promised. When courts apply these requirements to at-will employment, the usual conclusion is that they do not, in effect, constrain the employer’s attempt to modify its obligations to at-will employees. The employer may unilaterally eliminate a bonus to which it contractually committed, lower employee pay, or even declare that non-at-will employees are at-will employees going forward. It may act dishonestly, opportunistically, and irrationally in making these changes.

In sum, despite the “contractual” designation of at-will employment, the employer is obligated to the at-will employee only to the extent that it desires (with the exception of its duty to pay for completed work) and only for the time it desires. It may act dishonestly, opportunistically, and irrationally toward at-will employees. It is a “contractual” party and yet exempted from the requirements of a commitment, an ex ante bargain, and good faith.

Likewise, despite being a party to a “contract,” the at-will employee has few contractual rights. She generally has no right to expect that the employer will abide by its own policies, or even its contractual obligations. She generally has no right to expect the employer to act honestly or rationally, let alone to treat her fairly. Her main “right” is to quit. If she cannot adjust child care arrangements for the employer’s new schedule, for example, or if she is unhappy about the elimination of a benefit, her primary remedy is to leave her job. Contractual parties get enforceable expectations. At-will employees do not. 

Imposing binding terms on at-will employees

Despite contract law not delivering employees the basic benefit of a contract—enforceable expectations—courts usually apply the rules on forming and modifying contracts to permit the employer to impose enforceable obligations on at-will employees. These include mandatory arbitration agreements and restrictive covenants, like noncompete agreements. Again, the at-will employer normally does not need to act in good faith in taking such actions. In most jurisdictions, for example, the employer may require an at-will employee to sign a noncompete agreement as a condition for keeping a job she has held many years, and then terminate the employee the day after she signs.

The contractual designation also helps to rationalize an employer’s attempt to hold an at-will employee liable for violating the “duty of loyalty.” This duty restricts the employee’s activities that she might take during employment to prepare to compete with the employer following the end of the relationship.

Obscuring, reinforcing, and legitimizing employer power

Not only does the contractual designation of at-will employment fail to disturb the employer’s extra-contractual discretion, it also offers an affirmative justification for this discretion. The main consequence of trying to apply contract law to at-will employment is to obscure, reinforce, and legitimize the employer’s largely unchecked power over its at-will employees.

In some of the examples above, for example, courts rationalize the employer’s power through the fiction that at-will employment is a series of one-day contracts. This construction usually ensures that applying contractual rules will preserve the anti-contractual elements of at-will employment—the employer’s at-will authority and lack of an ex ante bargain. Two features of contract law lend themselves to this paradoxical preservation: (1) its wide notion of assent; and (2) its general indifference to the fairness of an exchange. First, contract law defines in very broad terms what it means to assent, or to voluntarily agree, to a contractual offer. Very few transactions are involuntary under contract law. Second, an enforceable contract requires that the parties exchange something; however, it does not generally regulate whether the parties are exchanging things of roughly equivalent value—one may contract to purchase a tropical island for one dollar.

To illustrate how this works in many jurisdictions, consider the case where the employer seeks to impose a mandatory arbitration agreement on its at-will employees. The employer sends an email to employees attaching the arbitration agreement and stating that, if an employee does not quit, the employer deems that she has assented to the agreement.1 Since each day is a new contract, there is no contract to change. By sending the email, the employer is proposing a new contract that includes an arbitration agreement. Therefore, it need not follow the rules of good faith. The employee may not like the arbitration agreement, but she also cannot afford to lose her job. By not quitting, under contract law’s broad notion of assent, the employee voluntarily consents to the agreement. The employer also satisfies the requirement that it provide something to the employee in exchange for her “promise” to abide by the arbitration agreement: The employer does not fire her immediately. Thus, by construing at-will employment as a series of one-day contracts, the law makes contractual rules, like the rules on modification, applicable to at-will employment while it simultaneously rationalizes the employer’s power not to follow them. Contract law is used both to explain the employer’s power to terminate the employee whenever it wants for any reason, but also to rationalize the employer’s power to commit the employee to obligations that outlive the employment relationship.


At-will employment is not a contractual relationship. By designating it as such, the law drapes the employer’s power with the legitimacy of legal authority and associates it with the ideology of freedom of contract. We should stop identifying at-will employment as a contractual relationship in political and social discourse. And decision-makers should stop recognizing at-will employment as a contractual relationship in legal disputes. Still, the above measures would be inadequate in themselves to dismantle the employer’s private rule over workers. Rather, we need to strengthen the capacity of workers to organize and collectively bargain.


It is common to hear people call employment a “contract” or a “contractual” relationship. This is, however, an incorrect classification of the most prevalent kind of employment in the United States—at-will employment. When employment is at will, the employer and employee have a right to walk away for any reason, at any time. The common law—the law made by judges when deciding cases—presumes that employment is at will unless the parties provide otherwise.2

This paper is not a critique of at-will employment as such. Others have taken on that task, demonstrating that employers wield enormous power over at-will employees (Anderson 2017). Likewise, others have critiqued the at-will presumption, correctly suggesting that it recruits the law to the political cause of maintaining the employer’s power (Bagenstos 2020b; Summers 2000).

Rather, this paper focuses on a narrower issue: the labeling and attempted treatment of at-will employment as “contractual.” Judges, legislators, scholars, and others often identify at-will employment as a contractual relationship, and in many kinds of disputes between employees and employers courts try to apply contract law to resolve the issue. However, at-will employment is not contractual as a matter of legal doctrine, and designating it so confirms and legitimizes the inherently unequal power between employees and employers.3

Affixing the contractual label tends to deter needed regulation by signaling that at-will employment reflects the parties’ “freedom of contract” and therefore should be left undisturbed by the legislature and judiciary. Thus, the contractual designation makes it more difficult for policymakers to increase wages or mandate benefits and for judges to interpret legislation protecting employees in a manner consistent with the legislation’s purpose.

Further, trying to impose a contractual framework to resolve legal disputes between employers and at-will employees provides little benefit to employees. It does not disturb the imbalance of power that underlies the employer’s “private government” (Anderson 2017) over the employee or lead to more equitable terms of employment. Redressing power imbalances is not the purpose of contract law, so this is unsurprising. However, the contractual designation of at-will employment does not even afford employees the most basic benefit that a contract has to offer—enforceable expectations about the future conduct of another:

  • The contractual designation usually does not give employees a right to expect the employer to abide by its own policies regarding working terms and conditions. This is the case even where the employer expects employees to abide by these policies and promulgates them to obtain their loyalty and hard work. Employers can often avoid assuming any enforceable obligations to at-will employees (apart from a duty to pay for completed work) by issuing disclaimers stating as much.
  • It does not give employees a right to expect that their working terms and conditions will be stable or predictable. Unlike parties to a contract, at-will employers need not commit to a contractual bargain and can invent the terms of the relationship as they go along. These terms might deal with the nature and intensity of the work, and they might also deal with matters that have little or nothing to do with performing a job well or realizing the employer’s commercial goals, such as whether the employee supports the employer’s political party. The court’s usual role in determining what goes into contractual gaps is largely inapplicable when it comes to at-will employment.
  • It does not give employees a right to expect that the employer will act honestly, rationally, or non-opportunistically toward them. In contrast, contractual parties must act in “good faith” toward one another in carrying out their contractual obligations.
  • It does not give employees a right to expect that employers will follow the usual rules for modifying contracts. Thus, even where the employer has assumed binding obligations to at-will employees, the employer can alter them unilaterally on a prospective basis.

While employees are usually left without the benefit of enforceable expectations, the attempted contractual treatment of at-will employment enables the employer to impose binding obligations on its at-will employees. In many jurisdictions, an employer can impose an arbitration agreement or restrictive covenant on its at-will employees without committing to anything in exchange. Many courts will hold the employee to these obligations even where the employer terminates the employee shortly after introducing them.

Paradoxically, where courts try to fit at-will employment into a contractual framework, they usually end up preserving the anti-contractual features of the relationship. How does this happen? The defining features of at-will employment are the opposite of the defining features of an enforceable contract. For example, an enforceable contract requires a commitment by at least one party. By definition, an “at-will” relationship lacks a commitment. Given the anti-contractual features of at-will employment, courts cannot both preserve the employer’s power and apply contract law coherently. Courts must therefore resort to doctrinal contortions and empirical fictions to create a semi-plausible case that at-will employment is, in fact, contractual. They go through this trouble only to conclude that contract rules do not regulate the employer’s behavior in any meaningful way, because the effectiveness of these rules depends upon features that at-will employment lacks. Thus, the exercise of trying to construe at-will employment as a contractual relationship rationalizes the employer’s power as legal authority. It legitimizes the employer’s anti-contractual power to terminate employees at will, invent the terms of the relationship as it goes along, disregard its own commitments, and act in bad faith.

Two features of contract law also play a role in ensuring that contract rules do not disturb the employer’s power: contract law’s wide definition of assent and its indifference to the equity of an exchange. For example, under the former, many courts will find that an employee assents to an arbitration agreement by failing to quit her job after the employer notifies her that it is eliminating her right to go to court.4 And the employer satisfies the requirement that it provide something to the employee in exchange by not terminating her immediately.

Given the consequences of trying to impose a contractual framework on at-will employment, contract law should not govern this noncontractual relationship. Legal actors and policymakers should stop referring to at-will employment as a “contract.” Nonetheless, refusing to recognize at-will employment as a contractual relationship is not, in itself, enough to check the employer’s power. We must support workers’ rights to organize and build collective power. This is also a precondition for contract law to play an effective role in regulating employment.

This paper proceeds in several parts. Part I explains the features of at-will employment as a social relationship. Part II explains at-will employment as a legal relationship. Part III demonstrates that at-will employment is not contractual as a doctrinal matter. It also explains the mismatch between the organization of at-will employment in practice and a common legal construction that courts use to make it appear contractual. Part IV examines the consequences of labeling at-will employment a contract. Part V examines the consequences of trying to construct at-will employment as a contract in legal disputes. It shows that purporting to treat this relationship as a contract does not generally provide at-will employees with enforceable expectations, but it does enable employers to impose binding obligations on at-will employees. Part VI argues that imposing a contractual framework on at-will employment tends to obscure and legitimize, and thus reinforce, the employer’s power. Part VII proposes that worker organization is the best way to counter the employer’s power.

Two caveats are in order before continuing. By necessity, this paper generalizes the law of contracts and of employment. Each U.S. jurisdiction5 has its own common law of contracts and employment, and this paper focuses on the majority and plurality positions. Keep in mind, however, that the law of a particular jurisdiction may differ.6 Statutory contract law is more uniform, but can also vary.7 Second, the focus of this paper is on the relationship between contract law and the common law at-will employment relationship. The paper deals little with statutory employment law, except where highlighting how the contractual designation of employment tends to limit statutory worker protections. The reader should nonetheless be aware that statutory law places some limits on the employer’s power over the employment relationship.8

Part I. At-will employment as a social relationship

To understand the at-will employment relationship, we first need to understand the employment relationship. At first glance, employment appears simple. To be an employee means to work for another, under the command of another, in exchange for remuneration. But what does it mean to “work”? To work is to take voluntary action toward some end. It means choosing and acting upon those choices toward an end. Thus, in working under the command of another, the employee places her ability to take purposive action at the employer’s disposal.

To understand employment in the U.S., we also need to understand capitalism and its main features. Employment is a creature of capitalism and one of its defining characteristics.

Capitalism is a social system in which price-setting markets dominate many forms of economic activity. As in other places, the development of capitalism in the United States involved the assimilation and subjugation to price-setting markets of various kinds of economic activity—meaning activity by which humans interact with and shape their material environment to satisfy wants and needs. A market is a meeting of buyers and sellers, and these buyers and sellers need not be bound to one another through feudal or kinship ties. In a price-setting market, buyers and sellers determine prices through voluntary exchanges. The price mechanism governs the use of social resources by regulating their production and allocation in the commodity form. A commodity is simply something of use to humans that is produced for market sale and capable of exchange via a market transaction.

The philosophy behind the development of capitalism was that price-setting markets would govern the use of finite economic resources in a more efficient manner than other systems. Prices would equilibrate the supply and demand of commodities, directing them to those who valued them most.

What distinguishes capitalism as a peculiar kind of market system is its attempt to treat the ability to work—the ability to take purposive action—as a commodity. In other words, capitalism seeks to commodify this ability. Why?

Another feature of capitalist systems is the appropriation and monopolization of social resources by a small group of persons. This was originally accomplished through extra-legal means (Marx 2005, 521) and labor practices that have since been outlawed (Steinfeld 1991; Orren 1991). Once appropriated, the social resources become productive, private property, or capital. Through the law of private property, the state protects this group’s possession of capital and conditions access to these social resources—including life necessities—upon market exchange (Hale 1923, 472–73; Weber 1961, 277). But the monopolization of capital by a small group means that the majority of persons have no property to bring to the market and nothing to exchange for the commodities they need to survive and live with dignity (Weber 1923, 277, 306–7). Thus, they bring themselves to the market, “compelled” to work for those who possess capital (Weber 1961, 277). As law professor and economist Robert Hale wrote: “The employer’s power to induce people to work for him depends largely on the fact that the law previously restricts the liberty of these people to consume, while he has the power, through the payment of wages, to release them to some extent from these restrictions” (Hale 1943, 627). By applying her labor to the employer’s capital, the employee makes the capital more valuable for market sale. Today, most persons in the U.S. make their living by working for someone else (Bureau of Labor Statistics 2020).

Why, however, does the capitalist insist that the propertyless individual agree to work under its command—to subjugate to the capitalist her ability to take purposive action? Why not instead agree to an exchange of some quantity and kind of work (the results of purposive action)? As a capitalist, the employer seeks to get as much as possible from the propertyless individual, and its survival depends in large part upon its competitive production of market value (Marx 2005; Weber 1961). To obtain as much value as possible from the employee, the employer does not want to commit itself to terms specifying the quantity or quality of work the employee should provide. It would rather have free use of the employee’s very ability to work, at least for some number of hours (Marx 2005). In this way, the employer can seek to maximize the rate at which work is converted into market value. The employer dictates the nature of the work, its intensity, its pace, and even disciplinary rules to maximize this rate of conversion.

By definition, price-setting markets allocate resources efficiently only to the extent that resources are commodified. The resources must be produced for sale on the market if the price mechanism is to regulate their supply. And the resources must be exchangeable through market sales; that is, they must be capable of being alienated from the seller and acquired by the buyer. If a resource cannot be separated from the seller, we cannot claim that the market is allocating resources to where they are valued most. Apart from the resource not changing hands, we no longer have a subject on the selling side to value the use.

The ability to work, however, is a “fictitious” commodity (Polanyi 1957).9 It is not produced for market sale, nor is it exchangeable through a market transaction. These two dimensions of labor’s status as a fictitious commodity are the object of power struggles between workers and capital.

An individual’s faculty to take purposive action is not created for market sale. When the state seeks to regulate the supply of this ability via the price mechanism (e.g., the level of wages), deleterious social consequences follow: Decreases in labor demand increase poverty, starvation, degradation, and other forms of abject human misery (Polanyi 1957).

Economics uses somewhat different language to explain the harmful consequences of treating labor as a commodity. Labor supply is less elastic than labor demand, meaning that the supply of labor is less responsive to decreases in the price of labor or labor demand than the demand for labor is sensitive to increases in the price of labor or labor supply (Offe and Wiesenthal 1980; Silver 2008). In an ideal market, levels of supply and demand adjust until they reach equilibrium. Suppliers of a good or service, when faced with a decrease in demand, respond by withdrawing supply from the market (e.g., producing less) or decreasing the price at which they are willing to sell. Buyers of a good or service, when faced with an increase in price, demand less of the good or service or bargain for a lower price.

Workers, as suppliers of labor, are limited in their ability to withdraw labor from the market or accept price decreases (lower wages). In some cases, workers are able to withdraw labor from the market to go to school. A tiny number of workers can rely on savings until the market improves. Unemployment insurance in the United States is generally only an option if the employee loses her job through no fault of her own. Moreover, compared with other wealthy industrialized states, the United States makes it difficult for workers to withdraw labor from the market in order to raise a family or pursue other socially valuable activity. As an empirical matter, labor markets rarely clear, meaning there are always more individuals seeking work than there are positions for them (Bivens and Zipperer 2018).

Employers, as possessors of capital, are better positioned than workers to lower their demand for labor when the price of labor increases. Capital’s concentration, flexibility, and mobility make several fixes available to employers that are not generally available to workers. These include relocating production, switching lines of production, moving to another industry, or leaving commerce altogether to profit via financial activity (Silver 2008).

Anyone who has taken an introductory course in macroeconomics is familiar with the vertical labor supply curve. This verticality is not natural: It is a consequence of legal and policy choices to treat labor as a commodity. We see power struggles over the regulation of labor supply through price-setting markets in many forms. These include living-wage movements, as well as disputes over family leave, funding for education and the arts, and other policies that would incline the labor supply curve and give workers some meaningful autonomy. The power struggles also appear in worker efforts to leverage the price mechanism through collective power, for instance, by unionizing for better wages.

The second obstacle to commodifying the ability to work is that the ability cannot be exchanged. Exchanging a commodity by definition means alienating that commodity so that the buyer can appropriate it as its own private property. An individual’s ability to choose and act upon her choices is inalienable, however. Contrast this to work that is not carried out under the authority of another and is exchangeable in its objectified, completed form. For instance, a computer programmer’s work is alienable from the programmer in the form of the program.10 However, the employment relationship by definition involves the sale of one’s ability to work, not work itself. This ability is inalienable from the individual who possesses it—it is her vitality, life force, and means of self-determination (Marx 2005).11

What happens when the employer seeks to grasp and command the worker’s means of self-determination for its own ends? Workers generally seek to preserve some means of self-determination from the employer’s commandeering. Therefore, attempting to treat the ability to work as a tradeable commodity generates power struggles. These struggles appear in various forms, including political contests over privacy protections at work and the regulation of working hours. Workers also wage these struggles by organizing and using collective power to establish limits on the employer’s commandeering. For instance, union contracts often include limits on the pace and kind of work the employer can command.

The above discussion illustrates that the employment relationship reflects systematic power imbalances between employers and workers and that it is a site of power struggles. Most individuals are “compelled” to work for employers. Laws and policies that attempt to subject labor supply to price-setting markets and to commodify the ability to work sustain the employer’s power and generate conflict between workers and capital. One such law is the employer’s common law right to terminate employees at will, explained further in the following section. Parts IV through VI show that, where the law designates at-will employment a “contractual” relationship, it usually sides with the employer in the power struggles between workers and capital.

Part II. Employment as a legal relationship

The employment relationship in the United States is a combination of common law and statutory law.

A. Common law development: The departure from contract

The common law employment “contract” is the legal expression of the attempt to commodify the ability to work. It defines employment as an agreement to work for another under the other’s right of control, in exchange for remuneration.12

This definition derives from the pre-industrial, household relationship of master and servant. The master-servant relationship was not contractual; it was in pre-industrial times one of several kinds of legal relationships involving an exchange of services for pay. Apart from slavery and indentured servitude, it was perhaps the most one-sided in favor of the hiring party. The master had a property right to the personal services of the servant (Tomlins 1992), a right that entailed a broad authority to command the servant on anything touching the master’s interest and even beyond, reaching the servant’s private life (Morris 1946).

The reformulation of the master-servant relationship as a contract was the accomplishment of judges and legal treatise writers in the 19th century (Tomlins 1993). They transplanted the master’s authority over an individual servant into the employer, giving the employer discretionary authority over a business enterprise (Tomlins 1992, 83).

The categorization of the master-servant relationship as a contract appeared to solve the challenge of accounting for the employer’s power through law while making it politically acceptable. The employer’s de facto power over its workers derived from its property in the means of production and from the law of private property, which denied to workers both the means of life and means of production. Sometimes, legal decisions suggested a property rationale to explain the employer’s power, but the problem was that property rights could not adequately account for the employer’s command over workers (Tomlins 1993, 295). For example, the employer’s property rights could not give the employer a right to discipline employees.13

Employers insisted upon a legal right to devise and change working conditions and work rules as they went along, and they sought the courts’ assistance. Employers asked the courts to declare that their power issued from an expanded form of the master’s authority over the servant; however, they also asked the courts to ascribe a contractual appearance to this power-cum-authority. The courts obliged, and “rules and working conditions were incorporated into the employment contract as implied terms. The employee is presumed to have given his assent to the rules, and this assent is the font of their legitimacy” (Selznick 1969, 131). Thus, 19th-century cases expanded the employer’s legal prerogatives on the basis of employee consent to an employment contract rather than resting these prerogatives on property ownership (Tomlins 1993). This interpretation entrenched the master-servant character of employment in the law and blocked its development along contractual lines, even though the relationship was denominated as “contractual” (Tomlins 1993, 343). “By the end of the nineteenth century the employment contract had become a very special sort of contract—in large part a legal device for guaranteeing to management the unilateral power to make rules and exercise discretion” (Selznick 1969, 135).

The common law employment relationship of today retains an additional feature from the 19th century: It presumes that employment is at will.14 Under this presumption, either the employee or employer can terminate the relationship at any time, for a good reason, a bad reason, or no reason.15

B. The at-will presumption and employer power

The employer’s authority to terminate employees at will translates into considerable power. Consider the following example:

At the time of her discharge, the plaintiff was the mother of a young son whom she cared for herself and supported entirely from her earnings. She commuted from Cape Cod to work for the defendant in Canton. When she was hired in April, 1991, she was told that her hours of work would be 8:15 A.M. to 5:30 P.M., with the need to work late on one or two days each month. The plaintiff arranged child care accordingly. In fact, the requirements of her job kept her until 6:30 P.M. to 7 P.M. from the outset and even later as the job progressed. In late July, 1991, the plaintiff was told that she would have to work until 9 or 10 P.M. each evening and all day Saturday for at least several months. The plaintiff informed her employer that she would not be able to work such hours because of her responsibilities as a mother. She was discharged two weeks later.16

The court found that the above discharge was lawful. Within the limits imposed by wage-and-hour legislation,17 the employer may direct the hours the employee is to work, and it may terminate the employee who refuses to work over 12 hours a day due to child care responsibilities.18 In another case, a court upheld a newspaper’s right to terminate an editor for placing an advertisement for her ice cream and hot dog stand in a rival newspaper in addition to her employer’s newspaper.19 It did not matter whether the employer’s reaction was “unreasonable,” because the employer is under no obligation to act reasonably when exercising its authority to terminate an employee.20 Under the at-will rule, it is also lawful for an employer to terminate a breastfeeding employee for taking unauthorized breaks to pump milk.21 And courts have upheld terminations following an employee’s good faith reporting of safety concerns to the employer.22

The common law recognizes few limits on the obedience the employer can demand and back by its at-will authority. Employers are generally free to direct the tasks that employees are to perform, the pace of the work, and its intensity. The employer may determine the kinds of work the employee provides, the hours the employee must work, and other terms and conditions, such as when and whether the employee may take bathroom breaks (within the limits of statutory requirements). Commands that are even more intrusive are permissible, such as when to speak, what to say or not to say, what emotions to convey, how to dress, and how to style one’s hair. Walmart, for instance, has prohibited employees from chatting with one another while on duty (Anderson 2017). The employer has a right to surveil the employee at work for compliance with its commands, and, subject to some limitations, to search the employee’s belongings. Of heightened salience today, the employer may demand that employees perform dangerous work, including work that will expose the employee to a risk of infection during pandemics.23

The employer’s authority extends to employee behavior unrelated to work.24 The employer may dictate elements of the employee’s civic life, including what political candidate the employee publicly supports (Anderson 2017; VanderVelde 2017). The employer may use its power to control the employee’s romantic life, for instance, determining who the employee may date or not date, or marry or not marry (Anderson 2017; VanderVelde 2017).25 The employer may control other aspects of the employee’s social life, such as the employee’s social media use and choice of recreational activities or hobbies.26

The at-will feature of employment is not a mandatory term but rather a presumption. The employer and employee may agree that the relationship is not at will. For instance, their agreement might provide that the employment is to last a certain term, like two years. In this case, terminating the employee early where the employee has not done anything wrong would make the employer liable for breach of contract. Instead of providing for a definite term, the agreement might provide that the employer can terminate the employee only under certain conditions, for instance, when the employer has “cause” or “good cause.” Certain employees commonly have non-at-will relationships. These include high-level managerial employees, like corporate officers, and employees with exceptional talent, knowledge, or expertise, like professional athletes. Further, collective bargaining agreements usually include a “just cause” provision. If the parties dispute the at-will nature of the relationship, the party arguing that the relationship is not at will (usually the employee) has the burden of proving this to the court.

C. Noncontractual, common law limits on the employer’s at-will authority

A majority of jurisdictions recognize a common law exception to the employer’s at-will authority in the form of an action for “wrongful discharge in violation of public policy.”27 In most jurisdictions with this exception, the cause of action is not under contract law but under tort law, the body of common law dealing with civil wrongs.28 When an employer terminates an employee in violation of public policy, it does not breach its agreement with the employee. The law deems the discharge to harm not just the employee but also third parties or the public. The elements and breadth of the tort claim differ across jurisdictions, but the tort commonly prohibits employers from terminating employees for exercising certain legal rights, like filing a workers’ compensation claim, serving on a jury, or cooperating with a law enforcement investigation.29

D. Statutory regulation

Statutory regulation also defines the employment relationship. Although these interventions fail to adequately constrain employer power, they play a critical role in protecting some of the human, civil, and political rights of workers.

Federal statutes and rules regulate various aspects of employment. They establish a minimum wage and overtime pay (Fair Labor Standards Act); prohibit discrimination on the basis of protected characteristics, such as race, religion, and sex (Title VII of the Civil Rights Act of 1964), and disability (Americans with Disabilities Act); establish unemployment insurance and workers compensation schemes; grant unpaid family leave (Family and Medical Leave Act); regulate employer-provided retirement benefits (Employee Retirement Income Security Act); protect employee privacy (for instance, under the Genetic Information Nondiscrimination Act); and give employees the right to refuse to perform unsafe work (Occupational Safety and Health Act).

The National Labor Relations Act (NLRA) protects employees’ right to organize, form unions, and bargain collectively with employers. It also gives employees a right to engage in “concerted activities for the purpose of…mutual aid or protection” whether or not they are in a union or seeking to form one. Thus, the NLRA protects an employee who approaches the employer with a co-worker to complain about unsafe work requirements during the pandemic. In addition, under the NLRA employees have a right to refuse to perform the unsafe work if the employer does not adequately address their concerns.

State and local government laws also regulate employment and sometimes go further than federal legislation, for instance, by requiring a higher minimum wage or paid family leave.

Statutory worker protections and rights do not derive from or depend upon the characterization of employment as a contract. Rather, as discussed in Part IV, this characterization has posed an obstacle to the enactment of worker protections. It likely deters additional regulation and encourages courts to interpret existing regulations in ways that limit their reach (Bagenstos 2020b).

Part III. At-will employment is not a contractual relationship

At-will employment is not a contractual relationship from the perspective of contract law. Like a contract, at-will employment involves an exchange—an exchange of work for pay. But contract law does not recognize every exchange as a contract. Further, the construction that many courts adopt to make at-will employment appear contractual neither resolves the doctrinal misfit nor provides a plausible picture of how most at-will employment is actually organized.

A. Doctrinal mismatch

By definition, an at-will relationship is not a contract. The sine qua non of a contract is commitment. Further, a contract must include an ex ante bargain—some minimal agreement, determined up front, as to what the parties are going to exchange or otherwise do for one another. At-will employment is missing both a commitment and an ex ante bargain.

1. Absence of a promise

A contract must include at least one promise to be binding. The Restatement (Second) of Contracts defines a contract as a “promise or set of promises for the breach of which the law gives a remedy, or the performance of which the law in some way recognizes as a duty.”30 The term “promise” indicates two essential things about contract law.

First, contract law deals with expectations about future conduct. It lays out rules by which someone can assert a legal right to expectations about how someone else will behave. If the parties meet certain requirements, contract law provides one party a remedy for the other party’s breach of her expectations.

Second, the promise requirement indicates that contract law deals in commitments: One party’s expectations about the future conduct of another receive no legal protection without a commitment either to or from the other. The Restatement (Second) of Contracts explains, “Words of a promise which by their terms make performance entirely optional with the ‘promisor’ do not constitute a promise.”31 Contractual promises cannot be “illusory” (Farnsworth 1990, 75–76). In other words, “One who states ‘I promise to render a future performance, if I want to when the time arrives,’ has made no promise at all” (Corbin 1995, 175–76).

At-will employment lacks a contractual promise because neither party makes a commitment to the other. Either party may terminate the relationship at any time for any reason or no reason. At first glance, it may appear that the employer and employee make promises to one another—the employee promises to work for the employer in exchange for compensation, and the employer promises to provide work. However, in legal terms, the at-will employee says, “I will work for you if and for as long as I feel like it.” And the at-will employer says, “I will give you work if and for as long as I feel like it.” An at-will relationship is, by definition, the opposite of a contract.32

Contrast at-will employment to employment for a definite term. In this case, the employer makes a promise to provide work and to pay the employee for completed work for the term of the agreement. The employee can get contractual damages if the employer terminates the employee or stops providing work or payment before the agreement expires.33

In one respect, at-will employment involves a promise. The employer must pay the agreed-upon rate of pay for completed work. Courts tend to treat this as a contractual commitment by the employer, meaning an employee can bring a breach of contract claim against the employer to recover unpaid compensation.

However, from the perspective of contract law, this is not one promise but an undetermined number of promises. In contractual terms, the employer makes an enforceable promise only upon providing work to the employee and each time thereafter that the employer provides work, or, in other words, each time the employer instructs the employee to do something. This arrangement becomes apparent when we look at how courts treat supply agreements in which a buyer agrees to pay for any orders it decides to place with a seller, but where the parties do not otherwise commit to place or fulfill orders. The courts tend to treat each buyer order and seller acceptance as a separate contract; as the judge makes clear in the following excerpt, the overarching agreement is not a contract:

At no point do the Terms and Conditions specify any obligation on the part of either party sufficient to create an enforceable contract—not a minimum number of units to be ordered, a minimum duration of exclusivity, nor a fixed price at which goods can be purchased.34

Courts treat service agreements in the same way: Without more, an agreement to provide services on an as-needed basis if the buyer has need for such services is not enforceable as a contract.35 The employer does not make a promise or undertake a commitment by hiring the at-will employee—at this point, the commitment is illusory.

Employment is the only “contract” carrying the at-will presumption. Where a true contract is silent about its duration, courts do not presume that the contract is at will. Instead, contract law says that a court can insert a duration term into the contract. For instance, the court might provide that the agreement was to last a “reasonable” length of time.36 The Uniform Commercial Code (UCC) expressly recognizes sales contracts of indefinite duration. In contrast to employment, however, the UCC generally requires the party terminating the agreement to provide “reasonable” notice to the other.37

2. Absence of an ex ante bargain

At-will employment is also not a contract because the object of the agreement—working under another’s control—is incompatible with the bargain requirement. At-will employment is the only “contract” where courts assume or permit that the “bargain” is not to have a bargain.

The ex ante bargain requirement

Most contracts must include an ex ante bargain, meaning some agreement as to what the parties are going to exchange or do for one another, reached at the outset of the relationship.38

The bargain must include the essential terms of the deal. Where the parties leave essential terms open and the court cannot fix this by inserting terms itself, the contract fails for being “indefinite.” The standard for definiteness is that the terms must be “reasonably certain,” meaning “they provide a basis for determining the existence of a breach and for giving an appropriate remedy.”39 The UCC requires evidence that the “parties have intended to make a contract” so that the court has a “reasonably certain basis for giving an appropriate remedy.”40

The parties must agree upon this bargain ex ante, meaning at the time they enter the agreement. A common refrain by the courts is that they will not enforce “agreements to agree.” In an agreement to agree, the parties not only omit an essential term but also provide that they will make up their minds about the term at a later time.41 Agreements to agree suggest that the parties did not agree to the essential terms of their deal at the time they formed the agreement.

One mantra of contract law is that “all contracts are incomplete.” The parties cannot possibly account for every scenario that might arise and address it in their contract. Contract law does not expect this. However, courts do not recognize vague commitments as binding contracts.

What do courts look for to determine that the parties “intended” to enter a contract? What terms are essential to the existence of a bargain? Given the salmagundi of matters covered in contracts, this is difficult to generalize; however, the UCC and case law provide guidance. Essential terms tend to answer basic questions about the exchange: “What, for how much?” For example, agreements lacking a quantity term (a “what”) tend to be unenforceable unless they satisfy an exception.

Output and requirements contracts illustrate this point. In an output contract, a buyer promises to purchase all of a seller’s output of a particular product. In a requirements contract, a seller promises to sell as much of a certain product as a buyer requires. Thus, outputs and requirements contracts have open-ended quantity terms. Further, the seller’s promise in the output contract and the buyer’s promise in the requirements contract appear illusory. For instance, the buyer in a requirements contract may decide that it no longer “requires” the selling party’s product, but that it instead “requires” the product of a cheaper seller. Thus, courts recognize these contracts only if they require the buyer and seller to deal exclusively with one another with respect to the product.42 

The courts sometimes take it upon themselves to fill in a contractual gap, and we can better understand the limits of the law’s tolerance for open-ended agreements by examining the limits of the courts’ willingness to do this.

In the practice of “implication,” the court implies something into the parties’ agreement that is not expressly articulated.43 For example, both the common law44 and UCC authorize courts to insert missing terms under certain circumstances where the parties later disagree on what they intended. The UCC provides for a court to insert a missing price term into a sales agreement so long as the agreement is otherwise clear. In this case, the court can supply a term that the contract price will be a “reasonable price.”45

The main justification for the court’s authority to imply content into agreements is that the inserted content reflects the “reasonable expectations” of the parties at the time they formed the agreement, given the context of the agreement. Thus, where the court inserts a price term under the UCC, the justification is that it was within the parties’ reasonable expectations that the contract would include the statutory term in the absence of the parties including their own term.

The court assesses the parties’ reasonable expectations from an objective standpoint. The court is unconcerned (formally) with the subjective intentions of the parties. Instead, it looks to the empirical manifestations of their agreement and its context and asks, “What would a ‘reasonable’ person take this agreement to be?”46 Judges are not positioned to derive the attributes of a reasonable person or the relevant context of the agreement through social scientific methods. As many have observed, the objective account of reasonable expectations is an alibi for judicial policymaking. Below I examine how courts have used it to delineate the outer contours of the bargain requirement. 

Consider how courts deal with contracts that afford one party discretion to determine some aspects of the bargain. Commercial contracts often grant one party the discretion to exercise certain terms, supply certain missing terms, or resolve certain ambiguities in the interpretation of terms. For example, a buyer might delegate to the seller the right to determine the manner of delivery. Limited grants of discretion are not a problem for meeting the bargain requirement.

What poses a problem is when the agreement seems to delegate so much discretion to one party as to enable it to deprive the other party of its deal. These agreements, by their terms, appear to permit one party to determine essential terms of the deal ex post.47 In such cases, the court generally has two options. First, it may find that the parties did not intend to enter a contract—they did not agree on the essential terms.48 In other words, the court refuses to countenance that it was within the reasonable expectations of the parties that one party would have such discretion. Second, the court may imply terms that restrict the party’s exercise of its discretion, again, on the theory that this is what the parties reasonably expected when they entered their agreement.

In sum, these agreements usually fail as contracts unless the court intervenes to save them by implying limits to one party’s discretion. To illustrate, consider an agreement in which one party promises the other that it will have an exclusive right to deal in the first party’s products, and where the other agrees to deal in the first party’s product. The UCC requires that the second party use its “best efforts” to market the first party’s product.49 Similarly, under the common law, where a commercial agreement fails adequately to specify one party’s obligations, the court will sometimes imply a term that the party use reasonable efforts or best efforts, thus limiting the party’s discretion and revealing an ex ante bargain.50 As illustrated above, output and requirements contracts also seem to provide one party with a quantum of discretion that would enable it to destroy the expected deal for the other party. In addition to implying a term of exclusive dealing into an otherwise valid output or requirements contract, the UCC imposes another requirement on these contracts: an implied term forbidding the party from demanding or tendering an “unreasonably disproportionate” quantity compared with what it demanded or tendered earlier under the agreement.51

Another means by which the case law and UCC circumscribe agreements that give too much discretion to one party is through the implied covenant of good faith and fair dealing (“good faith”). Each party to a contract owes a duty of good faith to the other party in the performance of its contractual duties. The law implies the duty of good faith into every contract and makes it nonwaivable.52 This means that the duty attaches to contractual obligations regardless of whether the parties expressly provide for it and even if the parties try to limit or exclude it. As with other implied terms, the dominant rationale for implying a duty of good faith is to realize the parties’ reasonable expectations at the time they entered the agreement.

The duty of good faith varies by jurisdiction.53 It usually requires that parties perform contractual duties honestly. Consider the following example where a requirement of honesty limits one party’s discretion to destroy the other party’s expected benefit. An artist agrees for a price to paint a portrait of her client to her client’s “satisfaction.” The artist paints the portrait, and the client is pleased with the portrait. However, the client tells the artist that she is not satisfied and refuses to pay the contract price. The client’s dishonesty breaches the duty of good faith.54 Depending on the jurisdiction and circumstances, the duty of good faith may prohibit sabotaging the other party’s ability to perform its contractual duties, willfully shirking one’s own contractual duties, abusing the power to specify terms, and exercising one’s contractual discretion to deprive the other party of its expected bargain (Summers 1968; VanderVelde 2017). For example, to fulfill the duty of good faith, the UCC requires that the purchaser in a requirements contract not drastically change the quantity it demands.55

At-will employment and the missing ex ante bargain

At-will employment far exceeds contract law’s tolerance for open-ended agreements.56

By definition, to form an at-will employment relationship, the parties need not reach any agreement that contract law would otherwise recognize as an ex ante bargain. Recall that employment is an agreement to work for another under the other’s right of control. And to work is to exercise one’s ability to take purposive action, or to make choices and act upon those choices. Thus, in agreeing to work under the employer’s right of control, the employee agrees to subject her very ability to reach a bargain to the employer’s command. The employer chooses the bargain—what the employee is to provide—by commanding the employee in her work as the relationship proceeds: what the employee does, how it is done, the effort exerted, the benefits the employee is to receive, and, as explained in Part V, even the employee’s obligations after the relationship has ended. At-will employment is an “agreement to agree” to whatever the employer decides upon ex post.

Take the example of an employee paid an hourly wage. The employer decides ex post the terms that answer the questions, “What, in return for how much?” It decides these questions by commanding the nature, pace, and intensity of the work. The employee agrees to provide an indefinite amount of work—whatever the employer can extract in an hour—in return for a definite amount of pay (Fox 1974, 190).57

Selznick (1969, 135) illustrates the absence of an ex ante bargain in this rhetorical passage:

Just what does the employer purchase? A reasonable amount of labor? So much as the employee is willing to do? Enough to keep the machinery running at a rate the employer finds proper? In purchasing labor, does the employer buy the right to regulate the employee’s working day as he sees fit? Does he purchase the right to ignore the proprieties of conduct, or must he treat the employee with decency and respect for his physical and psychological needs?

Now we can see that the employer’s “promise” to pay the employee in exchange for her obedience creates only the illusion of an ex ante bargain. At the end of the relationship, the employee has provided some quantity and quality of work to the employer, under certain conditions. However, the employee and employer did not agree on this quantity or quality or these conditions ex ante. The employer promised to pay only for whatever work it would later decide to, and be able to, extract from the employee.

An employer usually hires an employee to fill a particular position and will often lay out duties that the employee is to perform. The employer may even describe the employee’s duties in written form. However, the employer generally has no contractual right of action against the employee where the employee fails adequately to perform in the position.58 The terms are not contractual. Likewise, the at-will employee generally has no right to perform these duties, and the employer may terminate the employee if she refuses to perform other kinds of duties.59

This almost open-ended discretion afforded to one party is extraordinary from the perspective of contract law. At-will employment does not meet the same fates that befall other agreements with open-ended quantity terms or terms that otherwise afford one party an enormity of discretion. In most jurisdictions, at-will employment does not fail as a contract for indefiniteness. However, the law does not provide for the implication of terms to restrict the employer’s discretion. For example, courts do not impose a legal requirement that an employer and at-will employee deal exclusively with one another for a particular kind of work. In this regard, courts sometimes recognize the exceptional nature of employment as a “contract”:

[T]here is one area of contract law which is strikingly idiosyncratic. That is the law of employment contracts. It has developed contrary to all of the standard, modern contract principles….If the parties to an employment contract do not specify the duration of the contract, a court will not imply a reasonable duration. The contract is considered terminable at will.60

Nor does the law generally provide for courts to use the duty of good faith to delineate a bargain in at-will employment. The law imposes no duty on the employer to use “reasonable efforts” to provide work to an at-will employee, even where the employee is relying on her job to earn a living and the employer expects the employee to be available full time. Several jurisdictions do not recognize the duty of good faith in at-will employment at all.61 No jurisdiction subjects the employer’s at-will authority to an independent duty of good faith.62 In some jurisdictions, the duty applies to a limited category of dismissal decisions—the employer cannot dismiss an employee with the intention of depriving her of earned compensation or vested benefits.63 For example, if an employee earned a contractual annual bonus to be paid in the following year on the condition that the employee still be employed then, the employer cannot avoid paying the bonus by terminating the employee before the payment date.64 Thus, the duty in these cases is available only where the court first finds that the employer and employee agreed to a contractual right or benefit—it does not follow from the agreement to form an at-will employment relationship.65

It is difficult to find another “contract” where the law deems such broad discretion as that afforded to the at-will employer to be within the parties’ reasonable expectations. For no other kind of agreement that courts recognize as “contractual” can one party delegate so much discretion to the other without prompting the court to find either that the agreement fails for indefiniteness or requires the implication of additional terms.

3. At-will employment as a unilateral, day-to-day set of contracts

To deal with the conundrum of calling an at-will agreement a contract, courts often adopt the fiction that at-will employment is not an at-will agreement. They try to impose a peculiar doctrinal construction on at-will employment that involves two elements: (1) construing employment as a unilateral contract; and (2) construing employment not as one contract but as a series of contracts in which the employer and employee enter a new contract each day or for some other short duration.66

In a unilateral contract, instead of exchanging promises of future conduct the parties exchange a promise of future conduct for a performance. One party makes an offer to the other, but it is not bargaining for a return promise from the other. Rather, it is bargaining for the other party to do something.67 The structure of a unilateral contract offer is, “I promise to do X if you do Y,” whereas the structure of a bilateral contract offer is, “I promise to do X if you promise to do Y.” For instance, if a police department promises a reward for locating a wanted criminal, it does not want someone to promise to locate the criminal. It is bargaining for someone to locate the criminal. The other party can accept the contractual offer only by performing. The contract only comes into existence when the nonpromising party completes the performance, for instance, by locating the criminal. Note that the nonpromising party cannot breach a unilateral contract, because it makes no promise. If it decides not to render the performance or stops halfway through, the legal interpretation of this conduct is that the party does not accept the offer. Thus, the police department in the example cannot sue someone on the basis of a contract breach where the person tried but failed to locate the criminal. Nor can unilateral contracts be modified, since there is no contract unless and until the party completes the performance.

In jurisdictions that try to impose a unilateral contract framework on at-will employment, the employer is the promisor and the employee is the performer. The employer promises to pay the employee for following its commands. The employer is not bargaining for the employee’s promise to work, but rather for the employee to follow its directions.

The other part of this reconstruction is to depict employment not as one unilateral contract but as a series of unilateral contracts: The employer and employee renew the unilateral contract each day. Each day the employer makes an “offer” to the employee, and the employee accepts the offer and completes the contract each day by working under the employer’s direction.

This day-by-day construction appears to explain why at-will employment is a contract that either party can quit without incurring contractual liability. If the employee quits, she is simply “rejecting” the employer’s “offer,” not breaching her contract. When the employer terminates the employee, it is declining to extend a new offer to the employee for the next day. Note that the common law does not impose a duty to negotiate contracts in good faith.68 Thus, a party may decide that it does not want to renew a contract for a good reason, bad reason, or no reason. Under this construction, at-will employment looks more like a commercial agreement to supply goods or services, where the buyer promises to pay for any orders it places with the seller, but where the parties do not commit to place or fulfill orders. Each order by the buyer and responsive fulfillment by the seller creates a new unilateral contract committing the buyer to pay. 

However, the comparison is inapt. This attempt at a doctrinal solution fails adequately to explain at-will employment as a contractual relationship.

First, the reconstruction of employment as a series of day-to-day unilateral contracts does not address the problem of the missing ex ante bargain. It is still an agreement to work for another under the other’s right of control, although only for one day. Thus, it is still an agreement to allow the other party to choose the bargain ex post.

Second, the law does not take the day-to-day reconstruction seriously. Courts adopt this expedient only for some kinds of disputes, where it usually favors the employer. The common law does not otherwise recognize at-will employment as a one-day contract or as a non-at-will contract for a different period. For example, almost the only way to breach an at-will employment agreement is for the employer to fail to pay the employee for completed work. If at-will employment were a unilateral, one-day contract, however, it would be possible for the employer to breach the contract by other means. The employer would breach the contract by dismissing the employee mid-morning and refusing to pay the employee for the rest of the day. This breach occurs because contract law prohibits the promisor from withdrawing its offer of a unilateral contract until the other party has had an opportunity to accept by completing the requested performance.69 Thus, if at-will employment were a one-day unilateral contract, the employer would be required to permit the employee to continue working for a day. Yet, the at-will employee is generally entitled to be paid only for completed work.70 Likewise, if at-will employment were a one-day contract, the employer would not have to pay the employee where her performance was woefully inadequate. Instead, the employer is usually required to pay for the completed work. The employer’s remedy is to terminate the employee. Furthermore, as explained in Part V, even where courts do attempt to treat at-will employment as a series of one-day contracts, they are unable to reconcile this at-will, bargain-less relationship with contract law.

The only way to construe at-will employment as a collection of non-at-will contracts and still satisfy the bargain and commitment requirements would be to decompose each hour, minute, moment, or task into an independent contract. Each employer instruction would be an “offer” of work and implied promise to pay the employee for following that instruction. This uncertain decomposition that would be necessary to fit at-will employment into a contractual framework makes it clear why at-will employment is different from the overarching arrangement between a buyer and supplier of goods or services, where each buyer order is a new unilateral contract. In the buyer-seller relationship, it is easier to identify the transaction constituting the contract. It may be an order to purchase three gadgets, to perform one delivery job, or to cater two events. Identifying the contract in at-will employment is more difficult, if not impossible, in most cases: Is the contract for the hour, the minute, the moment? For one customer served, one library book reshelved, one bundle of insurance claims processed? And, indeed, courts do not undertake this exercise, despite it being necessary to determining whether there was any bargain and to find a commitment.

At least one scholar has suggested that at-will employment is a spot contract (Epstein 1992); however, this argument also fails to fit at-will employment into a contractual framework, for two reasons. First, a spot transaction refers to an extemporary exchange, for instance, where the parties trade goods and money simultaneously (Epstein 1992, 60). By definition, a spot transaction is not a contract, because it lacks a commitment: Instead of either party making a promise, both parties exchange performances. They form and perform the agreement at the same time, so the exchange does not exist in an unperformed state. It is like exchanging $2 for a glass of lemonade at a lemonade stand.71 Second, in practice, contract law deals with what otherwise look like spot exchanges by superimposing a contractual framework that supplements the otherwise bare-boned exchange. For instance, the UCC implies certain promises into transactions for the sale of goods, like the warranty of merchantability.72 The absence of contractual supplementation in the case of at-will employment makes it stand out.

B. Empirical mismatch between at-will employment and daily contracts or spot exchanges

Construing at-will employment as a series of daily contracts provides an unrealistic picture of most at-will employment.

The typical spot market as described by economists comprises one-off transactions: The buyer and seller meet for one exchange, and the parties settle the exchange—and their relationship—within a short period. The unpredictable and decentralized currents of the market bring them together once and then separate them again. They do not anticipate repeat business with one another, let alone that they would repeat the transaction on mostly the same terms, in succession, for an indefinite period. If the typical employee and employer were truly renegotiating a new exchange every day—the equivalent of the parties wandering into a busy bazaar each day to strike up a new deal—most economists would shame this as an extremely inefficient way of doing business or immediately recognize the legal characterization as a sham.73

Most at-will employment relationships do not resemble spot transactions. Day laborers may be hired for the day, but employers and at-will employees usually do not renegotiate their agreement and re-establish the terms of the relationship each day (let alone each hour or minute). The employer and employee make “transaction-specific” investments—specific commitments of resources to the relationship—on the assumption that the relationship will (or must) continue on about the same terms and shared understandings for some time (Williamson 1981). The design of the work process and compensation formula usually indicate that the employer does not contemplate a one-off transaction. Employers normally embed at-will employees into some kind of organizational regime. Employers screen and train employees. They issue employment handbooks laying out policies like performance expectations, benefits, and disciplinary procedures. They schedule employees and locate them within a particular division of labor. The employee accrues knowledge and experience regarding the employer’s methods—knowledge and experience that the employer cannot replace without some cost.

Economists have observed that wages generally have not followed the pattern one would predict in spot markets. Namely, wages have not reflected employee productivity but instead have tended to increase incrementally over the length of the employment relationship (Schwab 1993; Lazear 1979). Thus, early in the employee’s tenure with the employer, when the employee is still learning and gaining experience with the work, wages tend to exceed the employee’s productivity. In the middle of the employee’s tenure, when the employee is proficient at the work, wages are more equal to productivity. Closer to retirement, wages again exceed productivity. The secular increase in wages rewards the employee’s continuing effort, commitment, and loyalty (Weil 2014; Stone 2004; Weiler 1990). Over the past few decades, wages have become more “market-mediated” (Kalleberg 2011). For example, performance-based pay is more common, along with the practice of paying piece rates that roughly reflect the value of employee sales (Stone 2004). Some companies tie wages to company earnings. Despite these trends, compensation practices in at-will employment do not match the spot market theory.

Many jobs have become more precarious over the past few decades (Kalleberg 2011; Doussard 2013).74 Average job tenure has decreased for men (Hollister and Smith 2014), and many workers today face fluctuating and unpredictable schedules (Lambert, Fugiel, and Henly 2014). Companies have reorganized work practices to require less training, experience, skill, and education (Weil 2014; Kristal 2013; Doellgast 2012), rendering each worker more fungible. For instance, companies have used technology to deconstruct complex tasks into simple ones and refine the division of labor, with the consequence of making each worker’s contribution easier to replace (Irani 2015). Companies have outsourced work to unstable, undercapitalized entities that face more competitive pressures than their upstream clients, rendering the positions offered by the downstream entities less secure (Weil 2014). These changes have emboldened many employers to claim that the workers performing their core business operations are not their employees at all but rather “independent contractors” (Weil 2014). Indeed, nonstandard work arrangements have increased relative to regular, full-time positions (Katz and Krueger 2016).

These changes are still a far cry from turning at-will employment into a one-day contract. Only the rare employer is indifferent to cultivating an experienced or reliable workforce. This observation holds even if we consider a company that claims the workers performing its core business operations under its command are not employees—the on-demand ride company Uber. In legal proceedings, Uber holds itself out as a company that has all but perfected the use of information technology to design its relationship with its drivers as one-off exchanges. However, even Uber designs the work process and pay formulas to encourage driver loyalty to its platform. For example, Uber has programmed its algorithm to incentivize drivers to choose Uber over other platforms (Prassl 2018). Its pay is structured to incentivize drivers to work for Uber somewhat regularly (Uber n.d.).

Part IV. Labeling at-will employment a contract: ‘Freedom of contract’

One of the consequences of designating this noncontractual relationship a contract is to legitimize the employer’s power by raising an ideological barrier to regulating at-will employment.

Contract is a convenient ideological home for employment in a capitalist system. Once the master-servant relationship becomes a “contract,” the ideology of “freedom of contract” claims it for its own and makes regulation of the relationship suspect (Bagenstos 2020a).

Most references to freedom of contract are not to be taken literally. It is usually a metonym for “freedom” of the market, or for the tenet that the government should not interfere with the power of a private property owner to use and dispose of its property (Weber 1961; Bagenstos 2020a). Two key premises of this tenet are: (1) every individual is the foremost judge of her own interests, including how to weigh and rank those interests; and (2) the best evidence—the only valid evidence—of the individual’s interests are her choices about how to use and dispose of her property, including how the individual decides to dispose of her property through market exchange. Therefore, whenever the state seeks to regulate the market, it is (1) second-guessing the individual’s choices as to her interests, and, by extension, second-guessing the self-rule of the individual; and thereby (2) interfering with the efficient allocation of resources to those who value them most. Connected to this ideology is the 17th-century philosophy of possessive individualism, under which the individual possesses her ability to work as private property: Each person is user and disposer of her own labor (Gershon 2011).

Together, freedom of contract and possessive individualism form a dense dogma, since “when commerce and industry are perceived as the use and disposition of private property, there is no encouragement to legal scrutiny of the social structure that lies behind the economic act” (Selznick 1969, 65). Evidence has revealed time and again that these premises do not provide a logical or realistic depiction of the individual or establish that market “interference” is inefficient; nonetheless, the ideology discourages legal scrutiny of employer power. Affixing the inaccurate contractual label to at-will employment tends to deter needed regulation of this relationship by signaling that it embodies the parties’ freedom of contract.

For example, the contractual label has encouraged spurious defenses of the at-will presumption. Courts have invoked freedom of contract to justify and protect the employer’s at-will authority (Bagenstos 2020b).75 Other proponents of the at-will presumption have likewise argued that it protects workers’ freedom of contract (Epstein 1984) or that it is at least consistent with it (Hillman 2014). Historically, there is some support for this position: Giving the employee a right to quit meant eliminating the “entire contract doctrine,” under which an employee who left before the expiration of the term of employment (usually one year) would forfeit all payment for the employee’s work (Orren 1991; Stone 2007). But for a long time now, the freedom of contract argument has made no sense. Contract law does not require that contractual rights be mutual, including rights to terminate the agreement.76 Courts will enforce a contract giving only one party a right to terminate the agreement upon reasonable notice.77 Imposing limits on the employer’s right of termination does not mean the agreement must likewise restrict the employee’s right to quit to constitute an enforceable contract.

Furthermore, requiring parties to abide by their commitments for an agreed-upon period is consistent with freedom of contract. Every contract restrains the freedom of market trade to some extent, because every contract includes a commitment and thus looks to the future. For example, if I have a contract to purchase a bicycle from you that obligates me to pay you in two weeks, your bicycle is off the market for two weeks. The theory justifying contractual market restraints is that such restraints ultimately facilitate commerce by enabling rational planning. Contract law permits parties to agree to long-term contracts, including for personal services. It enforces lifetime contracts, contracts that last for decades, and contracts with no express termination point. Courts do not presume that a contract missing a duration term is terminable at will for any other kind of contract apart from employment. At-will employment represents freedom from contract.78

Placing at-will employment beneath the banner of freedom of contract likely inhibits statutory worker protections as well. Opponents have sought to prevent the enactment and extension of labor legislation by contending that it interfered with freedom of contract. During the Lochner era, courts regularly invoked freedom of contract to overturn statutory worker protections (Bagenstos 2020b).79 Courts still use this totem to support narrow interpretations of statutory protections (Bagenstos 2020b).

Part V. Constructing at-will employment as a contractual relationship: Little benefit to employees

In several kinds of disputes between employers and employees, courts try to impose a contractual framework on at-will employment to resolve the dispute. This offers little benefit to employees.

A. Contract law not designed to disturb power imbalances

Contract law is not designed to level imbalances of bargaining power, so trying to impose a contractual framework on at-will employment does not disturb the imbalance of power between employees and employers. It does not provide at-will employees a meaningful legal basis to challenge inequitable terms and conditions of work.

Several features of contract law add doctrinal flesh to the ideology of freedom of contract and its indifference to inequality.

First, the doctrine of “consideration” illustrates that contract law is unconcerned with the fairness of the terms of the exchange. To be enforceable against the promisor, the promise must be backed by consideration—something of legal value.80 Consideration could be a benefit to the promisor or a detriment to the promisee. For example, imagine that a seller and buyer bargain for return promises: The seller promises to give its car to the buyer in exchange for the buyer’s promise to give the seller $3,000. The promise to pay $3,000 is consideration backing the seller’s promise. It gives the buyer a legal right to insist that the seller execute its promise to give the buyer the car. Likewise, the promise to give the buyer the car is consideration for the buyer’s promise. It gives the seller a legal right to insist that the buyer pay it $3,000. Consideration may also be a promise not to exercise a legal right. For example, courts will enforce settlement agreements in which one party agrees not to sue the other in exchange for the other’s payment.

Contract law does not generally inquire into whether the consideration is “adequate”: “If the requirement of consideration is met, there is no additional requirement of…equivalence in the values exchanged.”81 In other words, so long as the consideration is something deemed to have legal value,82 courts will not otherwise scrutinize the fairness of the exchange. In keeping with the ideology of freedom of contract, interfering with a contract because the court disagreed with the terms of the exchange would breach the parties’ autonomy to determine their own interests.83 Thus, denominating at-will employment as a contractual relationship does not invite scrutiny of the terms of the exchange, for instance, of the wage level and working conditions.84

Second, contract law defines assent extremely broadly. Nearly all agreements are, by definition, “voluntary.” The commodious boundaries of contractual assent likewise reflect contract law’s basic indifference to disparities in bargaining power. The exceptions are narrow and not often applicable.85 Where the exceptions are helpful to employees, this is in part because attempting to force a contractual framework on at-will employment provides a rationale for allowing employers to impose legal obligations on employees in the first instance (see the section, “Imposing enforceable duties on at-will employees,” below).

Third, the interpretative principles are not designed to counter the adverse consequences of an imbalance of power between the parties.86 The goal of contractual interpretation is to find the parties’ intent based on the objective manifestations of this intent. The court considers the “reasonable expectations” that similarly situated parties would have about the meaning of the agreement at the time of contracting. The more powerful party can often determine that certain expectations are reasonable by clearly dictating them in a contract of adhesion.87 For example, if your contract with your internet provider states that your monthly charge is $50, it is generally unreasonable for you to claim that you expected the charge to be only $35 per month under the agreement. Contract law is concerned with preventing a party from frustrating the other party’s reasonable expectations. However, the harm caused to at-will employees due to the employer’s unilateral imposition of employment terms and conditions is not just a product of frustrated expectations; rather, it is also a product of an imbalance of power.

In sum, even if at-will employment were a contract, certain doctrinal features of contract law would foreclose most legal scrutiny of its voluntariness or the fairness of its terms. (Not only do these features help to show why employees receive little benefit from attempting to treat at-will employment as a contract, but, as discussed in Part VI, they help explain why this attempt legitimizes the employer’s exercise of power over employees).

B. At-will employment does not provide the key benefit of a contract: Enforceable expectations

Designating at-will employment as a contract does not even give employees the key benefit of a contract—legally enforceable expectations about the future conduct of another.

A key purpose of contract law is to facilitate commerce by making it possible to plan. In a complex market society, recognizing certain promises as legal commitments (in theory) enables persons to arrange their affairs in anticipation of the behavior of others. “Calculability” (Weber 1961) is an animating principle of contract law: “That is why contract as a legal device is so well adapted to the market economy. The obligor knows what he is getting into and can calculate his costs. He can maximize his freedom to make alternative decisions under changing economic conditions” (Selznick 1969, 56). A contract gives one a legal right to form and rely upon expectations about how another will act.

Attempting to impose a contractual framework on at-will employment does not give the employee a legal right to form and rely upon expectations about the employer’s conduct. It does not ensure that employers will be held to their own policies. It provides no certainty about the terms or length of employment. It does not restrict the employer’s abusive or arbitrary exercise of discretion. It does not even subject employers to the normal rules for modifying contracts.

In limited circumstances, employees succeed in negotiating contractual obligations that restrict the employer’s power, including its at-will power. Executive employees are generally able to negotiate binding, favorable terms. Star athletes and performers, and some professionals, are in a similar position. Another example is where workers negotiate a collective bargaining contract or otherwise use collective leverage to negotiate terms.88 In all of these cases, however, bargaining power is necessary to achieve these contractual arrangements. Contract law alone is inadequate.

The purpose of a contract is to “restrict…its expected future effects to those defined in the present, i.e., at the inception of the transaction” (MacNeil 1977–1978, 863). The employee, however, secures no commitments as to the future when she enters at-will employment.


1. No enforceable expectations that employers will abide by their own policies

Courts do not necessarily approach disputes over the interpretation of employment agreements differently than disputes over other agreements; however, the employer can often avoid assuming any contractual obligations to at-will employees (apart from a duty to pay for completed work), even where it issues personnel policies to govern the relationship. Contract law does not generally hold at-will employers accountable for their own policies as to working terms and conditions and fair treatment. This is the case even where the employer issues such policies to cultivate expectations among employees as to these matters in exchange for their loyalty and hard work.

At-will employment relationships are often formed without any written or oral contract.89 Recall that, unlike other agreements that the law recognizes as contractual, an employer does not have to commit itself to any contractual terms upfront to form an at-will employment relationship. The employer often specifies only a rate of pay. (And, as explained below, the at-will employer usually has a unilateral right to alter this rate prospectively.)

However, at-will employers often issue written policies to all employees or to a group of employees to govern them uniformly. Such policies may be in the form of a personnel manual, handbook, or other document. The materials may describe job duties, work rules, bonus schemes and other benefits, promotion ladders, and disciplinary procedures. The materials sometimes suggest that the employer will only terminate an employee for inadequate performance, for certain infractions, and/or only after following a set of procedures, like conducting an investigation and hearing.

Employees often (and reasonably) expect that these written policies will govern their work relationship, particularly where the employer expects employees to abide by rules in the documents. For example, where an employment manual lays out the details of a disciplinary process, an employee might argue that the employer’s failure to abide by this policy in terminating her was a breach of contract.

Even where there is no clear written policy, where the employer follows a consistent course of conduct, its employees often (and reasonably) expect that the employer will continue to adhere to the policy implicit in its treatment of them. For instance, a sales employee might argue that the employer’s regular practice in the timing of commission payments from the outset of the relationship reflects a term of the employment contract.

However, contract law does not provide a good basis for holding employers accountable for their policies. As noted above, courts look to the parties’ “reasonable expectations” to determine the terms of a contract. The concept of reasonable expectations also applies in determining whether the employee can hold the employer liable under the doctrine of promissory estoppel or “reliance.” This doctrine allows a party to recover when another’s unambiguous promise induces the party’s “reasonable” reliance on that promise to its detriment. The party relying on the promise can recover even though she did not offer a return promise or conduct sought by the promisor so as to form a unilateral or bilateral contract (Hillman 2014).90 Under these principles of interpretation, the at-will employee’s expectations or reliance regarding the employer’s policies is often “unreasonable.”

‘Reasonable expectations’
(1) The employer can often prove that the employee’s reliance or expectations were unreasonable by issuing a disclaimer.

For a time, courts seemed willing to enforce some employer practices and policies on the basis of “implied” contract theory (Stone 2007). In an implied contract, the contractual offer and acceptance are implicit and inferred from the parties’ conduct. For instance, an employer’s use of a certain formula over a course of years to calculate an employee’s commission payments might establish that the employee had a contractual right regarding the commission formula. Courts found that written policies, oral promises, and consistent practices could be the basis for implied contract claims by employees (Arnow-Richman 2009; Stone 2007).91 Courts even upheld limitations on the employer’s at-will authority on this basis, for instance, where a handbook promised job security or limited an employer’s right to dismiss employees arbitrarily (Stone 2007).

More recently, scholars have observed a contraction in the tendency of courts to find that employers have assumed enforceable obligations to at-will employees (Fineman 2008). They attribute this in part to employers learning to issue disclaimers to employees (Arnow-Richman 2009; Stone 2007). The disclaimers generally provide that the employer’s personnel policies, practices, and oral promises create no contractual obligations on the employer’s part. They often state that the relationship will remain at-will notwithstanding any oral statements or other personnel policies (Moss 2017; Arnow-Richman 2009).92 Thus, if an employer says that it has a policy not to terminate employees without following certain procedures, but also says that the relationship is at will, many courts will say it was unreasonable for the employee to believe that she was entitled to any sort of due process before termination.93 The Restatement (Third) of Employment finds that, despite a majority of jurisdictions being willing in theory to enforce employer policies under some circumstances, “All jurisdictions give considerable weight to the presence of a prominent disclaimer in the employer statement as evidence that the statement is not a binding commitment.”94

(2) Thinking that the employer limited its at-will authority is often ‘unreasonable’

Some scholars have argued that courts are particularly reluctant to find that an employer has contractually restricted its at-will authority (Bodie 2017; Summers 2000). Even where there is no disclaimer, courts have suggested that the at-will presumption makes it almost always unreasonable for an employee to rely on personnel policies or oral assurances as evidence that the employer limited its at-will authority.95 Some courts require exceptionally clear written evidence to find that the employer and employee agreed to something other than the at-will default (Bodie 2017).96 Some go so far as requiring that the employee offer something to the employer in addition to providing services under the employer’s direction, such as releasing the employer from a claim for damages.97 Otherwise, it is unreasonable for the employee to think that the employer had agreed to limit its at-will authority.98

(3) The employee’s expectations or reliance may be “unreasonable” for other reasons

As discussed in Part III, contractual promises must be somewhat clear and definite. Courts have found that the language in employer policies is too indefinite—to the point of being illusory—to constitute a contractual promise or the basis for a reliance claim.99 Where the employer expressly reserved a right to modify its personnel policies unilaterally at any time, courts have also found the employee’s expectation that the employer would abide by the policies—even where the policies were not modified—to be unreasonable.100 Some courts find that at-will employees cannot enforce written policies at all—any apparent promise by the employer to abide by its policies is illusory, because the employer can terminate the employee for any reason at any time.101 Even without a disclaimer then, some courts will find that it is unreasonable for an at-will employee to rely on an employer policy.102

Other obstacles to the contractual enforcement of employer policies
(1) Detrimental reliance

When bringing a claim of promissory estoppel, the employee may be unable to demonstrate that she relied to her detriment on an employer policy. In some jurisdictions, merely continuing to work for the employer following issuance of an employer policy is not enough.103 Some courts require that the employee demonstrate that she remained at work because of her reliance on the employer policy.104

(2) Unilateral and bilateral contracts: Bargaining for a promise or performance

Employer policies are difficult to enforce under a unilateral or bilateral contract analysis because the requirement that a return promise or performance be bargained for is normally missing.105

For an exchange of promises or an exchange of a promise for a performance to be contractually binding, the promisor must offer the promise in order to induce the return promise or performance, and this offer must in fact induce the return promise or performance. To meet the requirement of a contractual offer and acceptance, the offer cannot be in the form of a performance. To illustrate, if Fred bakes you cookies, and, in return, you promise Fred to pay him for the cookies, your promise is not enforceable: you did not make your promise to induce Fred to bake the cookies or to induce Fred to promise to bake cookies.

In the employment situation, employees are often unaware of the employer’s policies when they begin working, so it is not the case that the employer’s issuance of the policy induced the employee to work or continue working—it is difficult to say that the employee “accepts” the employer’s offer of its policies by beginning to work or continuing to work where the policies are disseminated through supervisors or by email.106 Likewise, the employer usually does not promulgate a written policy in response to the employee’s bargaining, such as where the employee threatens to quit in the absence of the policy.107 However, the employee does not make a contractual offer to the employer simply by working (baking cookies). Arguments that the employer’s assent to contractual obligations is evidenced by its course of conduct face similar problems under contract law.108

Where employers are obliged to follow their personnel policies: Departing from contract law

Where courts enforce policies on behalf of at-will employees against employers, they usually depart from a strict application of contract principles.109 For example, the Restatement (Third) of Employment notes that some jurisdictions enforce employer policies on “general estoppel principles,” without requiring that the employee demonstrate that she relied to her detriment on an employer policy. It recommends dropping the pretense that unilateral contract rules or the doctrine of detrimental reliance justifies their enforcement. The Restatement argues that the policies should be binding because the employer receives, and expects to receive, a benefit from the policies: The employer issues the policies because it expects them to contribute to an orderly and satisfied workforce; therefore, the employer should not be permitted to claim later that it was under no obligation to abide by them and that employees were unreasonable to expect as much.110

The Restatement summarizes the rule as follows with respect to policies limiting the employer’s at-will authority, and recommends the same rule for other employer policies:

Policy statements by an employer in documents such as employee manuals, personnel handbooks, and employment policy directives that are provided or made accessible to employees, whether by physical or electronic means, and that, reasonably read in context, establish limits on the employer’s power to terminate the employment relationship, are binding on the employer until modified or revoked.111

2. No enforceable expectations of certain or stable working terms or conditions

The failure of contract law to protect the expectations of at-will employees that employers will abide by their policies indicates that designating at-will employment as a contract does not disturb the employer’s power to make up the main terms of the relationship as it goes along. The point of a contract is to fix the parties’ obligations to one another within their bargain. However, as illustrated in Part III, at-will employment need not include a contractual bargain. The at-will employer can generally dictate working schedules, the pace of work, the kind of work, and other important conditions. It can alter these as it pleases.

3. No enforceable expectations that the employer will act honestly, rationally, or without opportunism

As explained in Part III, contract law imposes a duty on the contracting parties to act in good faith. In carrying out their contractual obligations, the parties should be honest with one another, and they should not be irrational or opportunistic in exercising contractual discretion.

Despite their designation as “contractual” parties, however, the duty of good faith does not give at-will employees a right to expect that the employer will act honestly, rationally, or without opportunism. This is because the applicability and scope of the duty depends on the underlying bargain. The duty of good faith is not a freestanding or extra-contractual obligation. It attaches to extant, particular contractual terms.112 In other words, parties do not have a general duty to act in good faith to one another by virtue of having entered a contract. They must act in good faith to one another only in the performance of their obligations. Thus, a federal appellate court explained that there is no duty of good faith in at-will employment, because at-will relationships do not have contractual terms to which the duty could be affixed.113 This ruling is consistent with contract doctrine and the common law of at-will employment. It highlights the weakness of contract law in protecting at-will employees from an employer’s abusive exercise of its power. 

The duty of good faith is antithetical to at-will employment (and to any other at-will relationship). The duty of good faith means that a contractual party cannot exercise its right to terminate an agreement dishonestly.114 In at-will employment, either party may terminate the agreement for any reason. For instance, the employer may terminate an at-will employee based on false accusations of employee wrongdoing. Employers are free to act dishonestly and opportunistically toward their at-will employees.

4. No enforceable expectations that the employer will abide by its binding commitments

Sometimes employers do assume binding obligations to at-will employees. As discussed above, in some jurisdictions and under some circumstances, the employer may be bound by personnel policies, including policies that change the employees’ at-will status. And the obligation to pay for completed work at a certain rate is usually treated as a contractual promise. However, contract law does not generally disturb the employer’s unilateral power to alter its commitments on a prospective basis. Contract law does regulate how parties can modify their contracts. Nonetheless, when courts apply these requirements to an employer’s attempt to modify obligations to its at-will employees—or to employees it now wants to consider at will—the usual conclusion is that these rules do not, in effect, apply. Contract law permits the employer to make prospective, unilateral changes whenever it wants, without even a minimal inquiry into the employer’s good faith. The employer may unilaterally eliminate a bonus to which it contractually committed, lower employee pay, or even declare that non-at-will employees are at-will employees going forward. It may act dishonestly, opportunistically, and irrationally in making these changes.

The usual rules for modifying contracts

The traditional rule—the “pre-existing duty rule”—is that the same requirements for entering a contract apply to modifying the contract: The party not seeking the change must assent, and the modified promise must be backed by something of value apart from what the party seeking the change already promised to do under the existing contract. In other words, the party seeking the change must provide additional consideration. For example, imagine that a buyer and seller have exchanged promises for the sale of a car for $3,000. Now imagine that, before the parties have performed, the seller decides that it wants $4,000 for the car. Two things need to happen under the traditional rule to modify the contract price: (1) the seller should offer something additional in exchange for the buyer’s promise to pay an extra $1,000 (e.g., perhaps free maintenance or a new navigation system), and (2) the buyer must agree to pay the $4,000. Otherwise, under the pre-existing duty rule, the seller breaches the contract if it refuses to give the buyer the car for the contract price of $3,000.

A more recent trend in contract law relaxes the pre-existing duty rule by permitting a contractual change without additional consideration if it is in good faith and the other party assents. For example, the UCC provides that an agreement modifying a contract for the sale of goods “needs no consideration to be binding,” but must be in good faith, meaning that it is sought for a “legitimate commercial reason.”115 The Restatement (Second) of Contracts provides that a “promise modifying a duty under a contract not fully performed on either side is binding…if the modification is fair and equitable in view of circumstances not anticipated by the parties when the contract was made.”116 Thus, the modification should also be prompted by unanticipated circumstances.

The requirements of additional consideration and good faith are intended to prevent the party seeking the change from subjecting the other party to a hold-up in bad faith. For example, imagine that A, a catering company, contracts with B to provide catering for B’s large gala. The morning of the gala, A tells B that it will only cater the gala for 50% more than the contract price. It is too late for B to find another caterer, so B may feel it has no choice but to pay the higher price to get the same services that A already promised.117 In this case, A held up B in bad faith.

Modifying at-will employment and non-at-will employment relationships governed by a binding policy

Disputes over changes to at-will employment relationships—and to those that are no longer at will due to a binding employer policy—arise in several circumstances, usually where the employer takes unilateral action. The employer might seek to alter the employee’s pay or benefits, or the employer might seek to retract a binding promise, for example, a promise that the employee will only be terminated for cause. Or the employer might seek to introduce a restrictive covenant or mandatory arbitration agreement (discussed in the section, “Imposing enforceable duties on at-will employees,” below).

The dominant approach is to permit the employer to modify the relationship on a prospective basis, unilaterally, upon reasonable notice. This applies whether the change is advantageous to the employee (for example, introducing job security) or adverse (for example, cutting employee pay). The Restatement (Third) of Employment describes the prevailing position as to modifying binding policies: “An employer may prospectively modify or revoke its binding policy statements if it provides reasonable advance notice of, or reasonably makes accessible, the modified statement or revocation to the affected employees.”118

Courts suggest that this approach follows the traditional rule that the modification must be supported by consideration and voluntarily accepted by the employee. For a modification that is beneficial to the employee, the employee voluntarily accepts the change and provides consideration by continuing to work for the employer instead of quitting. For changes beneficial to the employer, the employer provides consideration by not terminating the at-will employee right away. Since the employer has a legal right to terminate the at-will employee at any time, refraining from exercising this right gives the employer a right to insist that the employee abide by the new arrangement (e.g., accepting lower pay). Again, the employee voluntarily accepts the change by not quitting.

The claim that, by refraining from terminating the employee right away, the employer provides consideration to support modifying the relationship to its advantage has provoked some dispute: 

If the same at-will employment relationship continues, where is the consideration? The employer has relinquished nothing, since it retains exactly the same pre-existing right it always had to discharge the employee at any time, for any reason, for no reason, with or without cause. The employee has gained nothing, for he has not been given or promised anything other than that which he already had, which is “employment which need not last longer than the ink is dry upon [his] signature.”119

In many jurisdictions, courts rebut this challenge with the following explanation: The employer does not propose to change the employment contract but rather proposes a new contract.120 At-will employment is a one-day contract that ends and begins anew each day: “because employers can fire employees at any time with or without cause, every day of an at-will relationship is a new day.”121 So long as the employer does not terminate the employee right away, and the employee does not quit, the modification is valid.

Problems with the dominant approach under contract law

Neither attempt to rationalize the employer’s power to modify the agreement makes sense under contract law. That the dominant approach is a sui generis doctrinal innovation spurred by the peculiarity of employment illustrates the incompatibility between contract law and at-will employment: Courts are unable to apply contract law in an intelligible manner, deliver enforceable expectations to employees, and protect the employer’s power.122

Consider first the explanation in which at-will employment appears as one unilateral contract of indefinite duration. Where the employer changes the agreement to its advantage, we saw that it is difficult to find any consideration for the employee’s promise, since the employer can still terminate the employee at any time.

The explanation faces additional problems where the employer is seeking to rescind a promise. As explained in Part III, unilateral contracts are not formed until they are executed by one side: The promisee accepts the offer and completes its performance at the same time. At one time, the law permitted the promisor to revoke her promise at any time before the promisee completed performance. This is no longer the law. The current position is that once the promisee begins performance, the promisor may not withdraw its offer until the promisee has a reasonable opportunity to complete the performance.123 For example, let us say that I promise to pay you $100 for bicycling across a bridge. In response, you begin bicycling across the bridge. When you are halfway across, I tell you that I have changed my mind about paying you for crossing the bridge. I have violated the agreement.

Now apply this to at-will employment where the employer has promised some right or benefit to an employee but then seeks to retract it. For example, imagine that the employer makes a clear written promise to senior managers that they will have job security in the absence of a major disruption to its business, a promise that the court and employer acknowledge creates a binding obligation.124 It would seem that the reasonable expectation of the parties is that the employer was bargaining for the managers to work until retirement. If the courts were applying unilateral contract principles, the employer could not withdraw the benefit until the managers had a chance to complete their bargained-for performance. It could not terminate employees before retirement.125 Similarly, if the employer promises its sales employees to provide a 10% commission until they turn 60, it seems that the employer was bargaining for the employees to continue to work for it at least until age 60.

Another reason the dominant approach is hard to reconcile with contract law is that, unlike the bridge example or the examples provided in the previous paragraph, it is often difficult to understand what performance the employer bargained for in making a promise/offering a unilateral contract. For example, imagine the employer promises to pay a 10% commission to at-will sales employees. Two years later, the employer says that it is only going to pay a 5% commission going forward. The employees object. The employer might claim that, in promising to pay the commission, it was bargaining for the employees to work for it until it changed its mind about the commission, and that now it has changed its mind. Of course, this means there was no ex ante bargain. It cannot be that, whenever the employer decides to withdraw the promise, the precise amount of time has passed for which the employer initially bargained when it made its promise. Contract law does not accept ex post or retroactive bargains. Therefore, permitting the employer to alter the commission under the dominant approach does not make sense under unilateral contract theory. On the other hand, it is not clear from the promise what performance the employer was bargaining for, so it is unclear how the employees could complete their performance. Thus, it also makes little sense to try to impose a unilateral contract on this scenario.

Now consider the day-to-day or moment-to-moment rationale, where the employer can modify even its binding obligations prospectively, because it is simply proposing a new contract each day, hour, minute, etc. As noted in Part III, the common law does not, in other respects, treat at-will employment as a series of daily contracts. Rather, courts adopt this fiction on a selective basis where it generally favors the employer. Likewise, it is difficult to say either as a practical matter or doctrinal matter that the parties are continuously re-negotiating their contract based on some other arbitrary but short unit of time, like each hour or second. Recall that a promise to do or not do something “at will” is illusory under contract law and that the day-to-day fiction is meant to get around this problem. So, courts sometimes suggest that the employer is “promising” non-at-will employment on a moment-to-moment basis. Here, the court is just calling a performance a “promise”: Promises are future-regarding acts. Promising putatively non-at-will employment on a moment-to-moment basis is basically saying that the employer is performing—in exchange for the employee’s obedience, the employer is providing work to the employee, not promising it. This is not a contract. There is not even an implicit promise within the moment-to-moment construction. (Note we are here setting aside the problem that agreeing to follow another’s command is too indefinite to constitute a bargain—it would be a bargain not to have a bargain).

C. Imposing enforceable duties on at-will employees

Another consequence of categorizing at-will employment as a contractual relationship is to make it easier for employers to impose enforceable obligations on employees. Employers generally rely on their power of termination to procure the at-will employee’s compliance with its rules and commands. However, in some situations, often where the employment relationship has already ended, the employer seeks the law’s assistance. The most common of these situations is where the employer wants to hold an employee to a restrictive covenant or mandatory arbitration agreement. Employers often require employees to sign arbitration agreements and restrictive covenants as a condition of employment. Another scenario where the employer seeks affirmative assistance from the law is where it wants to hold an employee liable for breaching the duty of loyalty.126

1. Restrictive covenants and arbitration agreements

Restrictive covenants restrict the employee’s actions following termination. For example, a restrictive covenant might prohibit an ex-employee from soliciting any of the employers’ customers or other employees, or it might prohibit an ex-employee from dealing with anyone who was a customer of the employer during the employment relationship. Common restrictive covenants include noncompete, nonsolicitation, and nondealing agreements. Noncompete agreements prohibit the ex-employee from engaging in a line of work or kind of business in competition with the employer. A few states do not enforce noncompete agreements, and a few have banned their use for certain low-wage employees (Prescott, Bishara, and Starr 2016; Hahn and Beck 2019).127 However, over one-quarter of employees in the private sector are subject to noncompete agreements (Colvin and Shierholz 2019). An employee who breaks a valid restrictive covenant can be subject to damages or an injunction.

Mandatory arbitration agreements require employees to arbitrate employment disputes instead of bringing them to court. These agreements often prohibit employees from bringing collective claims as well, meaning that the employee must bring any claims against the employer as an individual.128 The Supreme Court’s interpretation of the Federal Arbitration Act makes arbitration agreements widely enforceable against employees, even where the agreement requires the employee to arbitrate civil rights claims and other statutory rights.129 The Supreme Court has made it clear that it views most arbitration agreements to be voluntary.130 Courts enforce these agreements by refusing to hear disputes covered by a valid arbitration agreement. An estimated 55% of U.S. workers are subject to mandatory arbitration agreements (Colvin 2018).

Construing at-will employment as a contractual relationship makes it easier for employers to impose binding restrictive covenants131 and arbitration agreements on employees unilaterally. In many courts, the rationale is the same as that for permitting the employer to modify its binding obligations: The employer provides consideration for the employee’s “promise” to comply with the agreement by not terminating the employee right away, i.e., through an implicit “promise” of at-will employment. Again, it does not matter that the employer may terminate, or in fact terminates, the employee shortly thereafter.132 The rationale applies whether the employer imposes the agreement at the beginning, middle, or even end of the employment relationship. Where the employer imposes the agreement in the middle of the relationship, some courts again resort to the day-to-day construction to explain this outcome: The employer did not modify the agreement, but proposed a new contract.133 A majority of jurisdictions permit employers to impose restrictive covenants on employees unilaterally under these rationales (Arnow-Richman 2016, 439).134

Although it still provides a rationale in many instances, the contractual designation of at-will employment may be less critical in enabling the employer to impose arbitration agreements on at-will employees compared with restrictive covenants.135 The employer’s return promise to arbitrate disputes constitutes consideration for the employee’s promise to arbitrate.136 The court need not assert that the employer’s provision of employment on an at-will basis constitutes a contractual promise. Nor does it need to depict at-will employment as a series of daily contracts to find contractual consideration. According to Arnow-Richman (2016, 445), most arbitration agreements do include a promise by the employer to arbitrate, given the employer sacrifices little in making this promise.137

2. Duty of loyalty

Some states impose a nonreciprocal duty on employees to be “loyal” to the employer (Selmi 2012). In these jurisdictions, employers can sue and hold employees liable for breaching that duty, which is applicable primarily where an employee begins to compete with her employer while still employed. While employed, an employee may take “preparatory steps” to begin competing with the employer after leaving the employer. For example, the employee may prepare for a job interview with a competitor of her employer or apply for a business license to start a new business in competition with the employer. The employee cannot begin competing while still employed, however. For instance, the employee cannot solicit the employer’s customers or co-workers (Selmi 2012; Fisk and Barry 2012). The employer is not subject to a reciprocal duty of loyalty to the employee. It may, for example, require the employee to train another worker to become the employee’s replacement, and it may sabotage the employee’s work or the employee’s outside vocation.

Construing at-will employment as a contract provides some rationale for imposing the duty of loyalty. The duty attaches to agents in all principal-agent relationships, and it is therefore not unique to employment. However, principal-agent relationships are generally formed through contracts. Thus, imposing the duty on at-will employees goes beyond agency law. Some states do not recognize a duty of loyalty in at-will employment relationships, because they do not see at-will employment as a contractual relationship in this regard (Selmi 2012).

3. Escape valves do not justify designating at-will employment a contractual relationship

Contract law includes a few doctrines intended to relieve vulnerable parties of their contractual obligations, such as the doctrine of unconscionability. The purpose of the unconscionability doctrine is to prevent “oppression” and “unfair surprise” to the weaker party, for example, where an imbalance in bargaining power leads the weaker party to agree to harsh terms hidden in the fine print.138

The doctrine of unconscionability is sometimes helpful to at-will employees in relieving them of obligations the employer has imposed via a restrictive covenant or arbitration agreement. This does not, however, demonstrate that the contractual classification of at-will employment benefits employees, for at least two reasons: 

  1. As explained above, the contractual designation provides a legal rationale for permitting employers to impose these obligations on at-will employees in the first instance.
  2. Courts do not, and cannot in the case of arbitration agreements, regularly find that employee obligations under these agreements are unconscionable. The doctrine of unconscionability is not meant to re-allocate contractual risks and burdens that reflect the imbalance of power between the parties. In the arbitration context, the Supreme Court has held that “mere inequality in bargaining power…is not a sufficient reason to hold that arbitration agreements are never enforceable in the employment context.”139 The Supreme Court has also limited the room for both courts and legislatures to apply unconscionability principles to arbitration agreements.140

Part VI. Legitimizing employer power

Rather than temper the employer’s whim, trying to impose a contractual framework on at-will employment obscures, even as it reinforces, the employee’s subjugation to capital. It lends the legitimacy of law to the employer’s exercise of power over the employee.

Calling at-will employment a contract is a post hoc legitimization of the employer’s exercise of power over the employee. The defining features of at-will employment are the opposite of the defining features of an enforceable contract. An enforceable contract requires a commitment by at least one party. By definition, an at-will relationship lacks this commitment. An enforceable contract requires that the parties arrive at a bargain—some minimal agreement as to what they are going to exchange or do for one another at the outset of the relationship. In contrast, the employee agrees only to subject her very faculty for agreement to the employer’s disposal. For no other kind of contract does the law assume and permit that the bargain was not to have a bargain.

Contract law’s broad notion of assent tends to rationalize the employer’s power to make up the terms of the deal as the parties go along. Historically, designating at-will employment as a contract enabled courts to rationalize the employer’s unfettered control on the pretext that the employee implicitly assented to everything the employer might do as the relationship proceeded (Tomlins 1993). Courts still rely on this line of reasoning, as illustrated by a more recent judicial defense of the at-will presumption: “The employee usually feels free to leave and take another job if it presents a more desirable opportunity. Similarly, the employer generally feels free to discharge the employee if he no longer wants his services. The at-will presumption is simply a legal recognition of the parties’ normal expectations. In the vast majority of cases, it is consistent with the parties’ intent.”141 Since contracts are almost always “voluntary,” the employee would not have agreed to enter this contract if she did not assent to the employer’s authority; and, the moment she does not, she would express her non-assent by quitting.

We saw that the interpretative principles of contract law do not guarantee any enforceable expectations to the at-will employee apart from being paid for completed work; instead, applying these principles rationalizes the employer’s power as the reasonable expectations of both employer and employee. Contract law declares the realized power of the employer as the parties’ mutual expectations agreed to ex ante.

We also see this legitimization of the employer’s power in disputes over employer attempts to modify an agreement or impose binding obligations on employees. As noted, the main approaches are to construe at-will employment as a day-to-day sequence of contracts or to invent a new doctrine of unilateral contracts to justify the employer’s power. These approaches tend to ensure that the employer will be effectively exempt from the rules of contractual modification. The meaningful application of the duty of good faith and other requirements for contractual modification depends upon the existence of a contractual bargain. By definition, at-will employment need not include a bargain that the law would otherwise recognize as “contractual.” The point of at-will employment is not to commit the employer to a bargain.

In other words, the law resorts to various doctrinal contortions and empirical fictions to say, “See, at-will employment is contractual.” After this herculean effort, the law turns around and says, “but, at-will employment is such a peculiar contract that the usual ways in which we would supervise a contract are inapplicable here; we defer to the employers.” The formal exercise of applying these rules to the at-will employment “contract” renders them meaningless. Why undertake a cumbersome analysis, requiring a set of sui generis doctrinal contrivances, if the inevitable conclusion is that even the duty of good faith—a duty that in theory applies to all contracting parties—does not apply to at-will employers? Construing at-will employment as a unilateral contract or series of one-day contracts respects neither doctrine nor social practice. The main consequence is to legitimize the employer’s power.

Note how contract law’s loose version of assent and its indifference to the equity of the bargain help make sense of these rationales for the employer’s power to modify the agreement or impose a binding duty on an at-will employee. The employee “assents” to the employer’s modification by not quitting. It might seem that a decision made under the ultimatum “your job or my way” is not a choice made under free circumstances. Under contract law’s extremely broad definition, however, the employee’s decision not to quit is voluntary. Also, under these approaches, the employer provides consideration by not terminating the employee right away or by putatively hiring the employee for one day. It likewise might seem like a rather inequitable deal for the employee to give up, for example, in exchange for a day of work, her annual bonus or even her right to be terminated only for cause. The employer’s abstention from terminating the employee before the ink dries on an arbitration agreement might likewise seem quite an insubstantial thing to support the employee’s relinquishment of her right to go to court. However, contract law is not generally concerned with the adequacy of consideration.

By trying to construct at-will employment as a contractual relationship and apply contract law rules, the courts offer a legal justification for the employer’s extra-contractual power to make up and change the terms of the relationship as it goes along. Contract law is used both to explain the employer’s power to terminate the employee whenever it wants for any reason, but also to rationalize the employer’s power to commit the employee to obligations that outlive the employment relationship. The law launders employer power into lawful authority.

Part VII. Solutions?

We should surrender the pretense that at-will employment is contractual. The contractual classification dresses up the employer’s power as lawful authority, making it more impervious to legislative or judicial abrogation. However, it is not enough to simply shed the contractual label and for the common law to stop recognizing at-will employment as a contractual relationship. After all, contract law does not give employers power over workers—capital does.

How then do we counter the power of capital over the lives of workers? Legislation has been critical in protecting the civil, political, and human rights of workers, and it is worth considering new statutory protections and better enforcement of existing ones. However, with a partial exception explained below, this is an inadequate solution. The panoply of statutes regulating employment does not confront the employer’s overall power. The statutes mainly carve out discrete exceptions to the employer’s use of its power. For instance, under anti-discrimination statutes, employers cannot use their power to discriminate on the basis of sex.

At-will employment is aberrant among contracts. Contract law does not otherwise recognize the open-ended authority of the at-will employer as compatible with a contractual relationship—it is a “magisterial authority” (Tomlins 1993, 284–85). Individuals do not normally enter these anti-contractual agreements unless they are without other property to bring to the market to make a living. It is a disparity of bargaining power that leads individuals into at-will employment. Within the syntax of contract law, the main features of these relationships are discernible only ex post. They are the outcomes of power struggles waged by the parties, not the products of ex ante commitments. In the absence of the workers having collective power, the employer often wins these struggles.

Therefore, reviving and protecting worker rights to organize and collectively bargain should be both the primary means of removing at-will employment from the governance of contract law and of countering the power of capital. In particular, we need much more worker organization and collective bargaining to constrain the employer’s power—both so workers can impose fairer terms of exchange and restrict the employer’s discretion.

The NLRA governs the legal status of unions and collective bargaining in the U.S., with some exceptions.142 It mandates that an employer bargain in good faith over the terms and conditions of employment with its employees’ chosen representative. The mandate is coupled with a right for employees to take collective action against employers to gain bargaining leverage in these negotiations without facing certain forms of employer retaliation.

Rather than target discrete aspects of the employer’s power, the NLRA protects the employee’s right to counter this power through collective action and bargaining. Congress did not intend the NLRA to be a radical instrument, and legislative amendments, case law, and administrative interpretations have unduly restricted workers’ NLRA rights and left many of them ambiguous and/or unstable (Mishel, Rhinehart, and Windham 2020); nonetheless, the underlying principle is worth fighting for—through collective power, workers can restrict the employer’s arbitrary power over their lives. They can subject employer power to negotiation.

Collective worker power is a necessary precondition for contract law or contractual principles to play an effective and humane role in employment. Where the employer and employees approach one another with more proportional bargaining power, a contract-like agreement such as a collective bargaining agreement (CBA) can play a useful role in governing their relationship. CBAs generally include fairer terms of exchange than the employee could obtain through bargaining in isolation, such as higher wages. CBAs also partially contractualize the employment relationship. They usually restrict the employer’s at-will authority, thus adding a promissory element to the relationship. They restrict the employer’s discretion in other ways as well, for instance, by regulating assignments, the intensity of the work, working hours, and employee discipline. In contrast to contract law and the common law generally, workers can use CBAs to impose enforceable limits on the employer’s power over the disposal of their capacity for purposive action.143

How to revive the NLRA to better empower workers, and whether other models for building and exercising collective power are more appropriate than the NLRA model for some workers today, is beyond the scope of this paper. The point is that, once organized, workers have a more realistic chance of confronting the power of capital.


At-will employment is simply unintelligible in contractual terms. Yet the law insists that it is a contractual relationship, requiring courts to attempt to use contract law to resolve disputes over the parties’ rights and obligations within this relationship.

This does little to help employees. We might not expect that designating at-will employment as a contractual relationship would lead to more equitable terms and conditions of work. The purpose of contract law is to protect and effectuate the “reasonable” expectations of the parties as to the meaning of their bargain, not to ensure an equitable exchange. And, indeed, designating at-will employment as a contractual relationship does not really disturb the imbalance of power between employees and employers.

Designating at-will employment as a contract does not even deliver on the basic promise of contract law. It does little to help employees secure legal recognition of employer policies and promises as enforceable obligations. The employer can often determine the parties’ “reasonable expectations,” for instance, by requiring employees to sign a disclaimer acknowledging that the employer has made no contractual commitments. And, should the at-will employer assume an enforceable obligation, it can modify it prospectively at any time upon notice. Thus, the at-will employer is obligated to the employee only to the extent that it desires (with the exception of its duty to pay for completed work) and only for the time that the employer desires its obligations to remain in effect. Thus, trying to treat at-will employment as a contract does not ensure that employees will have stable or predictable working conditions. It does not ensure that the employer will treat employees honestly or nonarbitrarily, let alone fairly.

Courts have been unable to both apply contract law to at-will employment in a coherent manner and to also rationalize the employer’s power. The ability of contract law to deliver enforceable expectations depends upon the very features that at-will employment lacks—a commitment and a definite bargain entered ex ante. By insisting that at-will employment is a contract then, the common law leads courts to engage in tortured doctrinal analyses that have the effect, paradoxically, of protecting and rationalizing the anti-contractual features of at-will employment—the employer’s power to terminate employees at will, to determine the terms and conditions of employment as it goes along, and to do these things via dishonest, arbitrary, and opportunistic means. It even protects and rationalizes the employer’s power to subject at-will employees to binding obligations, such as arbitration and noncompete agreements, while in effect exempting the employer from following the usual legal rules.

Even labeling at-will employment as a contract has undesirable consequences. Affixing the contractual label tends also to associate at-will employment with the ideology of “freedom of contract” in the eyes of adjudicators and policymakers. This makes it more difficult as a political matter to regulate the relationship via judicial and legislative means.

Why then do we persist in the fallacy of designating at-will employment as a contractual relationship? At-will employees are subject to the employer’s power, and the law both disguises and legitimizes this power by treating their subjugation as a contractual relationship. Construing at-will employment as a contractual relationship rationalizes this power as legal authority and disguises political questions as legal issues.

We must drop the pretense that at-will employment is a contractual relationship. The most promising way of removing at-will employment from the governance of contract law and countering the employer’s private rule is through collective worker power.


I am grateful to Larry Mishel for providing the opportunity to undertake this project and for his patience and creative and intellectual support along the way. I also thank Aditi Bagchi, Michael Harper, and Wilma Liebman and for their detailed and thoughtful feedback on earlier drafts.


1. This is the fact pattern underlying the Supreme Court’s decision in Epic Systems v. Lewis, 138 S. Ct. 1612 (2018).

2. This presumption is also in the Louisiana Civil Code, La. C.C. Art. 274. Louisiana is the only U.S. state that is not a common law jurisdiction.

3. This paper focuses on at-will employment, but parts of the analysis also apply to some employment relationships that are not at will. I indicate where this is the case.

4. This is the fact pattern underlying the Supreme Court’s decision in Epic Systems v. Lewis, 138 S. Ct. 1612 (2018).

5. For purposes of this paper, I confine the analysis to the 50 states and Washington D.C.

6. To illustrate common law doctrines, the paper uses case law and the Restatement (Second) of Agency (Am. Law Inst. 1958), Restatement (Third) of Agency (Am. Law Inst. 2006), Restatement (Second) of Contracts (Am. Law Inst. 1981), and Restatement (Third) of Employment (Am. Law Inst. 2015). The Restatements are compilations by jurists, legal scholars, and practitioners that summarize a certain body of law. Although the Restatements do not have legal force like case law, courts will sometimes cite them to illustrate points of law. For instance, the Supreme Court has often used the Restatement (Second) of Agency to distinguish employment from independent contracting. E.g., Cmty. for Creative Non-Violence v. Reid, 490 U.S. 730, 752 (1989).

7. Every state and Washington D.C. have codified a version of the Uniform Commercial Code (UCC) (Am. Law Inst. & Unif. Law Comm’n 2002), a set of model laws for commercial transactions. To illustrate contract law principles, this paper looks in particular at UCC Article 2, which covers the sale of goods. Louisiana is the only state that has not adopted Article 2.

8. Federal statutes generally apply to all private-sector employment relationships. States and municipalities may enact their own laws to cover aspects of employment that are not preempted by federal law. Most federal statutes do not preempt states and localities from enacting their own laws. These include the Fair Labor Standards Act (FLSA) (29 U.S.C. §§ 201–219), which establishes a right to the federal minimum wage and overtime pay, and federal anti-discrimination law. In contrast, the National Labor Relations Act (NLRA) (29 U.S.C. §§ 151-169), which regulates the legal status of trade unions and collective bargaining in the private sector, has a broad preemptive effect (Chamber of Commerce of the United States of Am. v. City of Seattle, 890 F.3d 769 (9th Cir. 2018)). The Occupational Safety and Health Act (OSHA) also has a preemptive effect. State and local laws must not provide less protection to workers than federal statutes. They sometimes regulate matters not covered by federal law. For instance, a California statute provides paid family leave (7 Cal. Unemp. Ins. Code §§ 3300–3308 (2003)), while the federal Family and Medical Leave Act (FMLA) (29 §§ U.S.C. 2602–2654) provides only unpaid leave.

9. Karl Polanyi (1957) theorized that money and land were also “fictitious” commodities.

10. To sell completed work, as opposed to selling the ability to work, it is not necessary that the exchange occur only after the work is completed, as in the programming example or the case of a carpenter who creates a chair that she later sells to a customer. In most cases of the production of personal services, the exchange at least in part overlaps with the work. For example, a doctor running a private practice in part exchanges her work as she examines or treats a patient. The recipient of the service is a necessary part of the productive process; however, the doctor is not placing her ability to work under the authority of the patient.

11. This notion appears implicit in the 13th Amendment’s prohibition of both slavery and involuntary servitude: Selling this faculty on a permanent basis is akin to slavery, because you cannot purchase a person’s capacity for purposive action without purchasing the person.

12. Restatement (Second) of Agency § 220(1) (Am. Law Inst. 1958); Restatement (Third) of Agency § 7.07 (Am. Law Inst. 2006); Restatement (Third) of Employment § 1.01 (Am. Law Inst. 2015).

13. There is a common misconception that the employer’s property gives it an ownership right over the job, and that this, in turn, “gives the employer the right to impose any requirement on the employee, give any order and insist on obedience, change any term of employment, and discard the employee at any time” (Summers 2000, 78). In the collective bargaining context, employers and arbitrators often insisted that employers had “reserved authority” over any aspect of the work relationship and enterprise governance that was not expressly qualified in the collective bargaining agreement (Young 1963, 247; Atleson 1982–1983, 95–96). Yet, property rights in nonlabor factors of production (e.g., capital, plant, equipment, and intellectual property) do not give the employer a property right over the job. Nor do they give the employer at-will authority or a right to direct, monitor, and discipline workers. The right to exclude others is a basic property right, but an owner can qualify and limit this right by contracting to permit others access to its property. The employer, like any other property owner, retains no absolute right to exclude others where this breaches its contractual obligations. The employer can likewise establish measures to protect its property when it grants access to others. However, an owner has no property right to require that others affirmatively cooperate with it, according to the owner’s terms, to help the owner valorize its property (Chamberlain 1951, 315).

14. Only Montana and Washington, D.C., through legislation, have changed this presumption. Restatement (Third) of Employment § 2.01 comment b (Am. Law Inst. 2015). Other statutes limit the reasons for which an employer may terminate employees, including at-will employees.

15. Id. at § 2.01.

16. Upton v. JWP Businessland, 425 Mass. 756 (1997).

17. The federal FLSA requires that employees be paid time-and-a-half for hours worked above their regular hours.

18. Upton, 425 Mass. 756.

19. Martin v. Capital Cities Media, Inc., 354 Pa. Super. 199 (1986).

20. Id. at 223.

21. Allen v. totes/Isotoner Corp., 123 Ohio St.3d 216 (2009). State and federal statutory anti-discrimination law prohibits some terminations related to lactation.

22. E.g., Murray v. Warren Pumps, LLC, 821 F.3d 77, 81 (1st Cir. 2016). The tort of wrongful discharge limits some terminations related to internal whistleblowing. As in the lactation case, any termination protections for at-will employees depend upon the development of statutory and tort law, not contract law.

23. An OSHA rule gives workers a right to refuse to perform dangerous work, but it is not regularly and decisively enforced (Berkowitz and Sonn 2020).

24. Four states—California, Colorado, New York, and North Dakota—have statutes that prohibit employers from terminating employees for their lawful activities outside of work. Cal. Labor Code §§ 96(k) (1999), 98.6 (2005); Colo. Rev. Stat. § 24-34-402.5 (1990); N.Y. Labor Law § 201-d; (2004) N.D. Cent. Code §§ 14-02.4-01 et seq. (1993). Another eight states protect employees from termination on the basis of using lawful products outside of work hours (ALI 2020). Few states have broad statutory prohibitions on discriminating against private employees on the basis of political beliefs or political activities outside of work (Mateo-Harris and Ogletree Deakins 2016). In most states, private employers may still punish employees for not participating in employer-sponsored political activities (Hertel-Fernandez 2017; Harvard Law Review 2014).

25. E.g., Frankel v. Warwick Hotel, 881 F. Supp. 183 (E.D. Pa. 1995) (upholding termination of at-will employee where employee refused to divorce his wife).

26. E.g., Bammert v. Don’s Super Valu, Inc., 646 N.W.2d 365 (Wis. 2002) (upholding termination where the employee’s husband, a patrol officer, arrested the employer’s spouse for drunk driving).

27. Restatement (Third) of Employment § 5.01 (Am. Law Inst. 2015).

28. Id.

29. Id. at § 5.02.

30. Restatement (Second) of Contracts § 1 (Am. Law Inst. 1981).

31. Id. at § 77.

32. Courts sometimes recognize this. E.g., White v. Roche Biomedical Laboratories, Inc., 807 F. Supp. 1212, 1219–20 (D.S.C. 1992) aff’d, 998 F.2d 1011 (4th Cir. 1993) (“a promise of employment for an indefinite duration with no restrictions on the employer’s right to terminate is illusory since an employer who promises at-will employment has the right to renege on that promise at any time for any reason”).

33. Restatement (Third) of Employment § 2.01 (Am. Law Inst. 2015).

34. Rekal Co. v. PGT Indus., Inc., No. 8:13-CV-1433-T-33TGW, 2013 WL 5487370, at *4 (M.D. Fla. Sept. 30, 2013) (citing Johnson Enterprises of Jacksonville, Inc. v. FPL Group, Inc., 162 F.3d 1290 (11th Cir. 1998)).

35. E.g., Johnson Enterprises, 162 F.3d 1290.

36. E.g., Haines v. City of New York, 41 N.Y.2d. 769 (Ct. App. N.Y. 1977).

37. U.C.C. § 2-309 (Am. Law Inst. & Unif. Law Comm’n 2002).

38. Under some circumstances, promises to make a gift are enforceable, and some jurisdictions recognize contracts under seal. Restatement (Second) of Contracts § 90 comment f, 4 3 Stat. Note (Am. Law Inst. 1981). These exceptions are not relevant to at-will employment.

39. Id. at § 34(2); Greene v. Oliver Realty, Inc., 363 Pa. Super. 534 (Penn. 1987).

40. U.C.C. § 2-204(3) (Am. Law Inst. & Unif. Law Comm’n 2002).

41. See Sun Printing & Publishing Association v. Remington Paper & Power Co., 139 N.E. 470, 471 (N.Y. 1923).

42. Mid-S. Packers, Inc. v. Shoney’s, Inc., 761 F.2d 1117, 1121-22 (5th Cir. 1985).

43. In theory, there is a distinction between “implication,” in which the court inserts new language into an agreement, and “interpretation,” where the court interprets language already in the agreement. The distinction is often difficult to discern in practice. The “reasonable expectations” principle applies to both implication and interpretation.

44. See Restatement (Second) of Contracts (Am. Law Inst. 1981) § 204 (“a term which is reasonable in the circumstances is supplied by the court”).

45. U.C.C. § 2-305.

46. Contract law provides guidance to courts regarding where they can look for gap fillers and what they can insert into a gap. For example, the UCC ranks the places courts should look for missing terms—the first place is the history of the contractual relationship, then prior dealings between the parties, and then industry custom. U.C.C. § 1-303. As to what to imply, contract law provides some standard gap fillers, such as the place for delivery in a sale-of-goods contract where the destination is not specified. U.C.C. § 2-504.

47. Centronics Corp. v. Genicom Corp., 562 A.2d 187 (N.H. 1989).

48. Id.

49. U.C.C. § 2-306(2).

50. Wood v. Lucy, Lady Duff-Gordon, 222 N.Y. 88 (1917); Restatement (Second) of Contracts § 77 (Am. Law Inst. 1981).

51. U.C.C. § 2-306.

52. Restatement (Second) of Contracts § 205. See also U.C.C. § 1-304.

53. E.g., Centronics Corp. v. Genicom Corp., 562 A.2d 187 (N.H. 1989).

54. Morin Building Products Co. v. Baystone Construction, Inc., 717 F.2d 413, 415 (7th Cir. 1983).

55. U.C.C. § 2-305(2).

56. The problem of a missing ex ante bargain is less of a problem in non-at-will employment agreements. An employment agreement with a definite term or “for cause” provision still appears to require the employee to submit her faculty for purposive action to the employer’s authority. Thus, like the at-will agreement, it appears too open-ended to be a contract. However, where an employer terminates an employee for “cause” or for “breaching” the agreement, and the employee contests this termination, the court must imply some content to the agreement to determine whether the employer in fact had “cause” or was entitled to terminate the employee before the agreement expired. Courts interpret these terms as constraining the reasons the employer can deem to be “cause” for termination or to constitute a “breach.” If the employer could declare that the employee had breached the agreement for “any reason” the employer felt like at the time (e.g., because the employee refused to divorce her husband at the employer’s request), the term would mean little. Thus, courts imply terms into non-at-will employment agreements that outline a faint bargain.

57. Where employees lack collective power and face difficult labor markets, we see evidence that employers have taken advantage of this open-endedness at the macro level in the widening gap between worker pay and productivity (Mishel and Bivens 2015).

58. An exception is where the agreement expressly states that breach of the provision is a basis for liability. See Restatement (Third) of Employment § 9.07 (Am. Law Inst. 2015).

59. McGrane v. Reader’s Digest Association, Inc., 822 F. Supp. 1044, 1050 (S.D.N.Y. 1993).

60. Greene v. Oliver Realty, Inc., 363 Pa. Super. 534 (Penn. 1987).

61. Restatement (Third) of Employment §2.07 (Am. Law Inst. 2015).

62. Id.

63. Id.

64. Id.

65. As discussed in Part V, the employer can often avoid assuming any contractual duties to at-will employees.

66. E.g., Copeco, Inc. v. Caley, 91 Ohio App.3d 474, 478 (1992).

67. Restatement (Second) of Contracts § 231 (Am. Law Inst. 1981). See also Hamer v. Sidway, 124 N.Y. 538 (1891).

68. The common law imposes some limits on what the parties can do during negotiations. For example, the claim of misrepresentation provides a remedy to an individual who enters a contract due to her reasonable reliance on another party’s misrepresentation of a material fact.

69. Restatement (Second) of Contracts § 45 (Am. Law Inst. 1981).

70. Restatement (Third) of Employment § 3.01 (Am. Law Inst. 2015).

71. Note that a sign advertising lemonade for $2 is unlikely to be a contractual offer or promise. Advertisements are normally invitations to bargain under contract law.

72. U.C.C. § 2-314.

73. In economic terms, the parties would be incurring unnecessary and high “transaction costs” in trying to make the deal each time, including the costs involved in searching the market for a contractual party and negotiating a deal. Thus, it may be cheaper to do business in a different way, for instance, through a long-term contract or as a firm (Williamson 1981).

74. Several factors have played a role in this transformation, including de-unionization, deregulation and lax enforcement, technological changes, and changing patterns of competition related to financialization and globalization (Kalleberg 2011; Weil 2014).

75. E.g., 765 S.W.2d 497 (Tex. App. 1989), writ denied (Dec. 13, 1989).

76. Restatement (Second) of Contracts § 79 (Am. Law Inst. 1981).

77. Without a provision requiring notice before termination, the promise would be illusory.

78. E.g., Haines v. City of New York, 41 N.Y.2d. 769 (Ct. App. N.Y. 1977).

79. Lochner v. New York, 198 U.S. 45 (1905).

80. In a bilateral contract it is backed by the other party’s return promise, and in a unilateral contract it is backed by the other party’s performance. Restatement (Second) of Contracts § 79 (Am. Law Inst. 1981).

81. Id.; Greene v. Oliver Realty, Inc., 363 Pa. Super. 534 (1987).

82. Moral obligations and sentiments do not meet the requirement of consideration. Restatement (Second) of Contracts § 71; Stone v. Lynch, 312 N.C. 739, 325 S.E.2d 230, 233. For example, if you promise “eternal gratitude” in exchange for your roommate’s promise to pay your rent next month, your roommate’s promise is not backed by consideration.

83. E.g., Restatement (Second) of Contracts § 79 comment c.

84. Courts may look at the fairness of the exchange for other purposes. The court may deny the non-breaching party a remedy of specific performance where there is a “gross disparity” in consideration. Unfair terms might also be evidence of misrepresentation or fraud. Id. at § 79.

85. The main exceptions that vitiate the voluntariness of an agreement are incapacity, duress (physical coercion or threats), undue influence, and unconscionability.

86. One interpretative principle, however, tends to favor the weaker party: where a provision in a standard-form contract is ambiguous, the court should construe the provision against the party that drafted it. E.g., ConFold Pac., Inc. v. Polaris Indus., Inc., 433 F.3d 952 (7th Cir. 2006).

87. A contract of adhesion is an agreement drafted by the stronger party with little or no negotiation and presented to the weaker party on a take-it-or-leave-it basis (Radin 2013).

88. Collective bargaining agreements cover only a small percentage of private-sector workers today (Mishel, Rhinehart, and Windham 2020).

89. It is difficult to find recent data on the percentage of U.S. employees that have a written agreement. Research from the 1990s indicates that very few employers used written agreements (Verkerke 1995).

90. Restatement (Second) of Contracts § 90.

91. E.g., Thompson v. St. Regis Paper Co., 102 Wash.2d 219 (1984).

92. E.g., Lobosco v. N.Y. Tel. Co./NYNEX, 96 N.Y.2d 312, 317 (2001); Bisig v. Time Warner Cable, Inc., 940 F.3d 205, 215, 216 (6th Cir. 2019).

93. A disclaimer communicated to employees will sometimes prevent the court from finding that the employer has assumed any enforceable obligations, despite other promises or assurances the employer may have provided. E.g., Trabing v. Kinko’s, Inc., 2002 WY 171 (2002) (employee’s knowledge of at-will policy defeated claim that employer should have followed disciplinary procedure in handbook before terminating her). But see McDonald v. Mobil Coal Producing, Inc., 789 P.2d 866, 869–70, on rehearing 820 P.2d 986 (Wyo. 1991) (disclaimer in handbook stating that it was not a contract did not prevent employee from making a viable claim). In some courts, a disclaimer is not dispositive evidence of the absence of enforceable obligations, but it may be enough to keep the issue from the jury. See Gargasz v. Nordson Corp., 68 Ohio App.3d 149 (1991) (finding that employee was unreasonable to expect employer to follow disciplinary sequence laid out in handbook).

94. § 2.05.

95. E.g., Martin v. Capital Cities Media, Inc., 354 Pa. Super. 199, 213 (1986); Restatement (Third) of Employment Law § 2.03.

96. Restatement (Third) of Employment Law § 2.03.

97. E.g., Spacesaver Sys., Inc. v. Adam, 212 Md. App. 422, aff’d, 440 Md. 1 (2014). The objective theory of contract interpretation leaves room for judicial policy making. In some cases evaluating employer assurances of job security, it is clear that the court’s determination of what is reasonable reflects its policy preferences. These preferences are sometimes based on the unproven economic assumption that employers must have nearly complete control over the enterprise for efficiency purposes.

98. E.g., Hawley v. Dresser Industries, Inc., 737 F. Supp. 445, 463, 464 (Ohio 1990). In Ohio, the courts presume that promises of “permanent” employment mean at-will employment, and that it would be unreasonable for an employee to rely on any such promise, even a 59-year-old employee. Where the parties are incorporated (and thus potentially immortal) entities, it may well be within the parties’ reasonable expectations that a “permanent” agreement means only an indefinite agreement (terminable on notice). It is difficult to see why this must be the case where one party is mortal.

99. E.g., Martin, 354 Pa. Super. at 213 (1986) (finding that employer’s handbook was an “aspirational statement,” and that the “vagueness and the generalities coupled with the employer’s reservation of power to unilaterally alter the handbook’s terms would lead a reasonable at-will employee to interpret its distribution as an informational guideline”).

100. E.g., Neri v. Ross-Simons, Inc., 897 A.2d 42, 48 (R.I. 2006).

101. Restatement (Third) of Employment § 2.05 (Am. Law Inst. 2015).

102. Id. Individuals have also sought damages where they relied upon an employer’s unfulfilled promise to hire them, for instance, by turning down another job offer or moving across the country. Some jurisdictions will not permit individuals to recover for relying on promises of at-will employment. See White v. Roche Biomedical Laboratories, Inc., 807 F. Supp. 1212, 1219–20 (D.S.C. 1992) aff’d, 998 F.2d 1011 (4th Cir. 1993) (“reliance on a promise consisting solely of at-will employment is unreasonable as a matter of law since such a promise creates no enforceable rights in favor of the employee other than the right to collect wages accrued for work performed”); Slate v. Saxon, Marquoit, Bertoni & Todd, 166 Or. App. 1 (2000).

103. E.g., McKenny v. John V. Carr & Son, Inc., 922 F. Supp. 967, 979 (D. Vt. 1996).

104. E.g., Baker v. City of SeaTac, 994 F. Supp.2d 1148, 1160 (W.D. Wash. 2014).

105. Restatement (Third) of Employment § 2.05. Despite these challenges, employees occasional win. E.g., Presto v. Sequoia Systems, Inc., 633 F. Supp. 1117, 1120 (enforcing on basis that the employer should have known that the employee would rely on its promises).

106. Restatement (Third) of Employment § 2.05.

107. Id.

108. Where an employee seeks to prove the employer orally promised employment that was to last more than a year, the “statute of frauds” can pose obstacles to enforcing the promise in some courts (Bodie 2017). The statute of frauds requires contracts that cannot by their terms be performed within a year to be in writing and signed to be enforceable.

109. Restatement (Third) of Employment § 2.05 (Am. Law Inst. 2015).

110. Id. at § 2.05, comment b.

111. Id. at § 2.05. Even this relaxed approach to contractual rules does not ensure that employers will be held to their policies. Jurisdictions that follow the Restatement approach retain the requirement that the employee’s expectations must be “reasonable.” As discussed, the employee’s expectations are often “unreasonable” because the employer’s superior bargaining power did not enable the employee to negotiate for more balanced, clear, and binding terms.

112. E.g., Racine & Laramie v. Department of Parks and Recreation, 14 Cal.Rptr.2d 335 (Cal. App. 4 Dist. 1992).

113. Liu v. Amway Corp., 347 F.3d 1125, 1138 (9th Cir. 2004).

114. E.g., Anthony v. Atlantic Group, Inc., 909 F.Supp.2d 455 (S.C. 2012) (“To impose a duty on employers to exercise reasonable care…would eviscerate the employment-at-will presumption.”)

115. U.C.C. § 2-209 (Am. Law Inst. & Unif. Law Comm’n 2002).

116. § 89.

117. The modified agreement might be in good faith, however, where the caterer requests 10% above the contract price due to an unexpected product recall or embargo that abruptly increases the price of truffles the buyer has requested for the party. This would be in good faith due to the unanticipated circumstances.

118. Restatement (Third) of Employment § 2.06 (Am. Law Inst. 2015).

119. Lake Land Emp. Grp. of Akron, LLC v. Columber, 804 N.E.2d 27, 34 (2004) (dissent) (quoting Kadis v. Britte, 244 N.C. 154, 163 (1944)). In a minority of jurisdictions, the employer’s abstention from terminating the employee is inadequate consideration to support the employer’s unfavorable modification. The employer must offer something in addition, like a promotion, new benefit, or improved working conditions. E.g., Hopper v. All Pet Animal Clinic, Inc., 861 P.2d 531, 541 (Wyo. 1993). The basis of the requirement is that the employer’s implied promise of continued employment on an at-will basis is illusionary. Given that contract law is unconcerned with the “adequacy” of consideration, it is generally easy for employers to meet the requirement of additional consideration. The employer could, for instance, provide an additional vacation day (Arnow-Richman 2009, 155). Where the additional benefit depends on continued at-will employment, however, it would also seem to be illusionary. There are two other variations on the dominant approach worth mention: (1) A few courts find that the employer’s abstention from terminating the employee constitutes consideration for an unfavorable change only if the employer does not terminate the employee for some time following the change, for instance, until the employee decides to leave her job. E.g., Zellner v. Conrad, 183 A.D.2d 250, 589 N.Y.S.2d 903 (App. Div. 1992). While this approach is fairer to the employee, it is difficult to justify under contractual rules, since it implies an ex post bargain as to the period of employment the parties were bargaining for. An illusory promise does not become less illusory under contract law where the party carries it out in part. See Curtis 1000, Inc. v. Suess, 24 F.3d 941 (7th Cir. 1994) (noting the ex post bargain issue as a problem, but suggesting that requiring the employer to continue employment for a “substantial period” shows that the employer was not acting fraudulently in promising at-will employment). (2) The second variation requires that the binding promise the employer wants to retract be in effect for a “reasonable time” before the retraction. Asmus v. Pac. Bell, 23 Cal.4th 1, 14, 999 P.2d 71 (2000). This is likewise more fair to employees but hard to explain under contract law.

120. E.g., Summits 7, Inc. v. Kelly, 2005 VT 97, 10 (2005) (“the presentation of a noncompetition agreement [for example] is, in effect, a proposal to renegotiate the at-will relationship”).

121. Id.

122. The reasonable notice requirement, for example, has no place under unilateral contract law and seems to be an attempt to create a unilateral-bilateral hybrid. Given these issues, the Restatement (Third) of Employment § 2.06 (Am. Law Inst. 2015) makes no attempt to reconcile the dominant position with contract law. Instead, it suggests the at-will employer’s power of modification is not restricted by contract law since contract law does not form the basis for finding that its policy statements are binding in the first place.

123. Restatement (Second) of Contracts § 45 (Am. Law Inst. 1981).

124. This is roughly the fact pattern underlying the California Supreme Court decision Asmus v. Pacific Bell, 23 Cal.4th 1 (2000).

125. Sometimes courts will find that a clear bargained-for-performance is implicit in the promise and prevent the employer from retracting the promise prematurely. E.g., McCaskey v. California State Auto. Assn., 189 Cal.App.4th 947 (2010).

126. Another kind of provision that employers sometimes seek to enforce against employees is a prohibition against disclosing trade secrets. Also, high-level employees are subject to fiduciary duties.

127. California’s Civil Code makes virtually all noncompete agreements unenforceable against employees (Cal. Bus. & Prof. Code § 16600). California law does not, however, prohibit employers from using noncompete agreements. Research suggests that this distinction is consequential. A recent study found little evidence of a correlation between the law on the enforceability of noncompete agreements and the rate that employers use noncompete agreements, including in California (Prescott, Bishara, and Starr 2016).

128. The Supreme Court recently found that these agreements do not violate an employee’s right under the NLRA to take collective action. Epic Systems v. Lewis, 138 S. Ct. 1612 (2018).

129. Gilmer v. Interstate/Johnson Lane Corp., 500 US 20 (1991).

130. AT&T Mobility LLC v. Concepcion, 563 U.S. 333, 346–47, 131 S. Ct. 1740, 1750 (2011).

131. Note that this discussion only deals with the issue of contractual formation and modification. To be binding, restrictive covenants must also meet public policy requirements. For instance, in evaluating noncompete agreements, courts apply a test of “reasonableness” to determine whether the agreement is an unreasonable restraint of trade and unduly limits the employee’s ability to earn a living. Hopper v. All Pet Animal Clinic, Inc., 861 P.2d 531, 541 (Wyo. 1993). In some jurisdictions, construing at-will employment as a contractual relationship may help employers surmount the public policy obstacle to enforcing noncompete agreements against former at-will employees. To not impose an unreasonable trade restraint, among other things, the agreement must be “ancillary to an otherwise valid transaction or relationship” (Restatement (Second) of Contracts § 187). In many jurisdictions, the “otherwise valid transaction or relationship” does not have to be an enforceable contract. Therefore, at-will employment satisfies the requirement. In Texas, however, a nonenforceable agreement does not satisfy this requirement. And, in other jurisdictions, construing at-will employment as a contract appears to make judges more comfortable with finding that the noncompete is enforceable. E.g., Mattison v. Johnston, 152 Ariz. 109 (1986).

132. See Marine Contractors Co. v. Hurley, 365 Mass. 280, 310 N.E.2d 915 (1974). Unlike in the context of altering employer policies, however, the employer does not need to provide reasonable notice before imposing an arbitration agreement or restrictive covenant.

133. See infra Part V.B. for a discussion of why these rationales are inadequate as a matter of contract law.

134. In contrast to restrictive covenants, regarding arbitration agreements imposed in the middle of an employment relationship, Arnow-Richman (2016, 444) has found that courts are more likely to require the agreement to include a reciprocal promise by the employer to arbitrate before holding the employee to the agreement. According to Arnow-Richman (2016), most arbitration agreements do include an employer promise.

135. See, e.g., Epic Systems, 138 S. Ct. 1612.

136. See Patterson v. Nine Energy Service, LLC, 330 F.Supp.3d 1280, 1308 (D.N.M. 2018).

137. Where the parties do not exchange promises (i.e., where the employer does not promise anything in return), most jurisdictions find that the employer provides consideration by providing at-will employment or even by considering the employee for hire. E.g., Coup v. Scottsdale Plaza Resort, LLC, 823 F.Supp.2d 931, 943 (D.Ariz 2011) (employer’s implied promise of continued at-will employment was consideration for arbitration agreement); Martindale v. Sandvik, Inc., 173 N.J. 76, 87, 800 A.2d 872, 879 (2002) (upholding arbitration agreement in application for employment). 

138. Restatement (Second) of Contracts § 208.

139. Gilmer v. Interstate/Johnson Lane Corp., 500 U.S. 20, 33 (1991); see also Brondyke v. Bridgepoint Educ., Inc. 985 F.Supp.2d 1079 (S.D.Iowa 2013).

140. AT&T Mobility LLC v. Concepcion, 563 U.S. 333, 346–47, 131 S. Ct. 1740, 1750 (2011).

141. Greene v. Oliver Realty, Inc., 363 Pa. Super. 534 (1987).

142. The NLRA does not cover public-sector employees or employees covered under the Railway Labor Act.

143. Although the NLRA makes CBAs legally enforceable, they differ from contracts in several respects. The CBA is between a collective and the employer. And its terms often resemble tariff-like regulations more than specific contractual duties and rights. Some have described the CBA as akin to a constitution for the workplace (MacNeil 1977–1978).


Anderson, Elizabeth. 2017. Private Government: How Employers Rule Our Lives (and Why We Don’t Talk About It). Princeton University Press.

Arnow-Richman, Rachel. 2009. “Response to Working Group on Chapter 2 of the Proposed Restatement of Employment Law: Putting the Restatement in Its Place Labor Law Group.” Employee Rights and Employment Policy Journal 13, no. 1: 143–58.

Arnow-Richman, Rachel. 2016. “Modifying at-Will Employment Contracts.” Boston College Law Review 57, no. 2: 427–506.

Atleson, James B. 1982–1983. “Management Prerogatives, Plant Closings, and the NLRA.” New York University Review of Law and Social Change 11, no. 1: 83–124.

Bagenstos, Samuel R. 2020a. “Consent, Coercion, and Employment Law.” Harvard Civil Rights-Civil Liberties Law Review 55, no. 2: 409–66.

Bagenstos, Samuel. 2020b. Lochner Lives On: Lochner Presumption of Equal Power Lives in Labor Law and Undermines Constitutional, Statutory, and Common Law Workplace Protections. Economic Policy Institute.

Berkowitz, Deborah, and Paul Sonn. 2020. Protecting Worker Safety & Health in the COVID Crisis: A State & Local Model Policy Response. National Employment Law Center.

Bivens, Josh, and Ben Zipperer. 2018. The Importance of Locking in Full Employment for the Long Haul. Economic Policy Institute.

Bodie, Matthew T. 2017. “The Best Way Out Is Always Through: Changing the Employment at-Will Default Rule to Protect Personal Autonomy.” University of Illinois Law Review 2017, no. 1: 223–68.

Bureau of Labor Statistics. 2020. “The Employment Situation—July 2020.” News Release. U.S. Department of Labor.

Chamberlain, Neil W. 1951. Collective Bargaining. McGraw-Hill.

Colvin, Alexander J.S. 2018. The Growing Use of Mandatory Arbitration: Access to the Courts Is Now Barred for More than 60 Million American Workers. Economic Policy Institute.

Colvin, Alexander J.S., and Heidi Shierholz. 2019. Noncompete Agreements: Ubiquitous, Harmful to Wages and to Competition, and Part of a Growing Trend of Employers Requiring Workers to Sign Away Their Rights. Economic Policy Institute.

Corbin, A.L. 1995. Corbin on Contracts: A Comprehensive Treatise on the Working Rules of Contract Law. West Publishing.

Doellgast, Virginia. 2012. Disintegrating Democracy at Work: Labor Unions and the Future of Good Jobs in the Service Economy. Cornell University Press.

Doussard, Marc. 2013. Degraded Work: The Struggle at the Bottom of the Labor Market. University of Minnesota Press.

Epstein, Richard A. 1984. “In Defense of the Contract at Will.” University of Chicago Law Review 51, no. 4: 947–82.

Epstein, Richard A. 1992. Forbidden Grounds: The Case Against Employment Discrimination Laws. Harvard University Press.

Farnsworth, E. Allen. 1990. Farnsworth on Contracts. Little, Brown.

Fineman, Jonathan. 2008. “The Inevitable Demise of the Implied Employment Contract.” Berkeley Journal of Employment and Labor Law 29, no. 2: 345–404.

Fisk, Catherine, and Adam Barry. 2012. “Contingent Loyalty and Restricted Exit: Commentary on the Restatement of Employment Law Papers from the American Bar Foundation.” Employee Rights and Employment Policy Journal 16, no. 2: 413–64.

Fox, Alan. 1974. Beyond Contract: Work, Power and Trust Relations. Faber.

Gershon, Ilana. 2011. “‘Neoliberal Agency.’” Current Anthropology 52 no. 4: 537–55.

Hahn, Erika, and Russell Beck. 2019. “Noncompete Misconceptions May Be Inhibiting Reform.” Law360, December 17.

Hale, Robert L. 1923. “Coercion and Distribution in a Supposedly Non-Coercive State.” Political Science Quarterly 38, no. 3: 470–94.

Hale, Robert L. 1943. “Bargaining, Duress, and Economic Liberty.” Columbia Law Review 43, no. 5: 603–28.

Harvard Law Review. 2014. “Citizens United at Work: How the Landmark Decision Legalized Political Coercion in the Workplace.” Harvard Law Review 128, no. 2: 669–90.

Hertel-Fernandez, Alexander. 2017. “American Employers as Political Machines.” Journal of Politics 79, no. 1: 105–17.

Hillman, Robert A. 2014. “Drafting Chapter 2 of the ALI’s Employment Law Restatement in the Shadow of Contract Law: An Assessment of the Challenges and Results Symposium: Assessing the Restatement of Employment Law: Essay.” Cornell Law Review 100, no. 6: 1341–68.

Hollister, Matissa N., and Kristin E. Smith. 2014. “Unmasking the Conflicting Trends in Job Tenure by Gender in the United States, 1983–2008.” American Sociological Review 79, no. 1: 159–81.

Irani, Lilly. 2015. “Difference and Dependence Among Digital Workers: The Case of Amazon Mechanical Turk.” South Atlantic Quarterly 114, no. 1: 225–34.

Kalleberg, Arne. 2011. Good Jobs, Bad Jobs: The Rise of Polarized and Precarious Employment Systems in the United States, 1970s to 2000s. Russell Sage Foundation.

Katz, Lawrence F., and Alan B. Krueger. 2016. The Rise and Nature of Alternative Work Arrangements in the United States, 1995-2015. National Bureau of Economic Research.

Kristal, Tali. 2013. “The Capitalist Machine: Computerization, Workers’ Power, and the Decline in Labor’s Share Within U.S. Industries.” American Sociological Review 78, no. 3: 361–89.

Lambert, Susan J., Peter J. Fugiel, and Julia R. Henly. 2014. Schedule Unpredictability Among Early Career Workers in the US Labor Market: A National Snapshot. University of Chicago.

Lazear, Edward P. 1979. “Why Is There Mandatory Retirement?” Journal of Political Economy 87, no. 6: 1261–84.

MacNeil, Ian R. 1977–1978. “Contracts: Adjustment of Long-Term Economic Relations Under Classical, Neoclassical, and Relational Contract Law.” Northwestern University Law Review 72, no. 6: 854–905.

Marx, Karl. 2005. Karl Marx: Selected Writings. Edited by David McLellan. 2nd ed. Oxford University Press.

Mateo-Harris, Gray I., and Ogletree Deakins. 2016. “Politics in the Workplace: A State-by-State Guide.” SHRM, October 31.

Mishel, Lawrence, and Josh Bivens. 2015. Understanding the Historic Divergence Between Productivity and a Typical Worker’s Pay: Why It Matters and Why It’s Real. Economic Policy Institute.

Mishel, Lawrence, Lynn Rhinehart, and Lane Windham. 2020. Explaining the Erosion of Private-Sector Unions: How Corporate Practices and Legal Changes Have Undercut the Ability of Workers to Organize and Bargain. Economic Policy Institute.

Morris, Richard B. 1946. Government and Labor in Early America. Columbia University Press.

Moss, Scott A. 2017. “The Value of the Statement of Employment Law, Based on 50-State Empirical Analyses and the Importance of Clarifying Disputed Issues—But with Caveats About the Restatement’s Imperfect Work Product.” Employee Rights and Employment Policy Journal 21, no. 2: 409–58.

Offe, Claus, and Helmut Wiesenthal. 1980. “Two Logics of Collective Action: Theoretical Notes on Social Class and Organizational Form.” Political Power and Social Theory 1: 67–115.

Orren, Karen. 1991. Belated Feudalism: Labor, the Law, and Liberal Development in the United States. Cambridge University Press.

Polanyi, Karl. 1957. The Great Transformation. 1st Beacon paperback ed. Beacon Press.

Prassl, Jeremias. 2018. Humans as a Service: The Promise and Perils of Work in the Gig Economy. Oxford University Press.

Prescott, J.J., Norman D. Bishara, and Evan Starr. 2016. “Understanding Noncompetition Agreements: The 2014 Noncompete Survey Project.” Michigan State Law Review 2016, no. 2: 369–464.

Radin, Margaret Jane. 2013. Boilerplate: The Fine Print, Vanishing Rights, and the Rule of Law. Princeton University Press.

Schwab, Stewart J. 1993. “Life-Cycle Justice: Accommodating Just Cause and Employment at Will.” Michigan Law Review 92, no. 1: 8–62.

Selmi, Michael. 2012. “Restatement’s Supersized Duty of Loyalty Provision, the Papers from the American Bar Foundation.” Employee Rights and Employee Policy Journal 16, no. 2: 395–412.

Selznick, Philip. 1969. Law, Society, and Industrial Justice. Russell Sage Foundation.

Silver, Beverly J. 2008. Forces of Labor: Workers’ Movements and Globalization Since 1870. Cambridge Studies in Comparative Politics. Cambridge University Press.

Steinfeld, Robert J. 1991. The Invention of Free Labor: The Employment Relation in English and American Law and Culture, 1350–1870. University of North Carolina Press.

Stone, Katherine V. W. 2004. From Widgets to Digits. Cambridge University Press.

Stone, Katherine V. W. 2007. “Revisiting the At-Will Employment Doctrine: Imposed Terms, Implied Terms, and the Normative World of the Workplace.” Industrial Law Journal 36, no. 1: 84–101.

Summers, Clyde W. 2000. “Employment at Will in the United States: The Divine Right of Employers.” University of Pennsylvania Journal of Labor and Employment Law 3, no. 1: 65–86.

Summers, Robert S. 1968. “‘Good Faith’ in General Contract Law and the Sales Provisions of the Uniform Commercial Code.” Virginia Law Review 54, no. 2: 195–267.

Tomlins, Christopher L. 1992. “Law and Power in the Employment Relationship.” In Labor Law in America: Historical and Critical Essays, edited by Christopher L. Tomlins and Andrew J. King, 71–98. Johns Hopkins University Press.

Tomlins, Christopher L. 1993. Law, Labor, and Ideology in the Early American Republic. Cambridge University Press.

Uber (n.d.). “How Much Can Drivers Make with Uber?” (website). Accessed February 8, 2021.

VanderVelde, Lea. 2017. “Where Is the Concept of Good Faith in the Restatement of Employment? The Labor Law Group Symposium on the Restatement of Employment Law.” Employee Rights and Employment Policy Journal 21, no. 2: 335–76.

Verkerke, J. Hoult. 1995. “An Empirical Perspective on Indefinite Term Employment Contracts: Resolving the Just Cause Debate.” Wisconsin Law Review 1995, no. 4: 837–918.

Weber, Max. 1961 (1923). General Economic History. Frank H. Knight trans., Greenberg.

Weil, David. 2014. The Fissured Workplace: Why Work Became So Bad for So Many and What Can Be Done to Improve It. Harvard University Press.

Weiler, Paul C. 1990. Governing the Workplace: The Future of Labor and Employment Law. Harvard University Press.

Williamson, Oliver. 1981. “The Economics of Organization: The Transaction Cost Approach.” American Journal of Sociology 87: 548–77.

Young, Stanley. 1963. “The Question of Managerial Prerogatives.” ILR Review 16, no. 2: 240–53.

See more work by Julia Tomassetti