Employment as a Socio-Legal Relationship

Karen Orren, Belated Feudalism (1991)

When the United Stated embarked upon full-scale industrialization in the decades following the Civil War, American labor relations were a remnant of the ancient order, in the sense that arrangements established in England in previous centuries were carried forward and enforced in law, often with only slight modification, to form the framework of relations between American employers and their employees.

The order of labor

At the most abstract level, “feudal” refers to the fact that the hierarchical relation of master and servant in nineteenth-century America was a remnant of the larger system of hierarchies that historically had extended up and down medieval society.

[B]eing a worker in late-nineteenth century America was still a legal status. By “status,” I refer to an established position in society conferred upon an individual that does not arise from any specific action or from a contract but from the individual’s personal characteristics. Nothing in American law directly stated that being a worker was a status or that there was a legal duty to work. Nevertheless, in every jurisdiction in the United States, not to work or be seeking work, if one was an able-bodied person without other visible means of support, was a crime, punishable by fine or imprisonment. Moreover, just as being a worker was a status, this crime, known as vagrancy, was one of the few acknowledged crimes of status in American law—that is, one that was not defined by an action or inaction taken in itself but was committed purely through one’s personal condition, by being a member of some predefined legal category. …

The Statutes of Labourers⊕During the Black Plague in England (1348-50), between 30-60% of the population died. The English Parliament enacted the Ordinance of Labourers (1349) and the Statute of Labourers (1351) to address the resulting labor shortage by imposing caps on wages, compelling all able-bodied men and women under age 60 (except those of independent means) to work, and restricting labor mobility.

presented the first comprehensive scheme in which the worker’s failure to work was suppressed through the apparatus of the criminal law. In the text, situated between the provision that victuals must be sold at reasonable prices and the provision that a laborer accepting more wages than customary must pay the surplus to the town, appears the following:

Item, because that many valiant [able-bodied] beggars, as long as they may live of begging do refuse to labour, giving themselves to idleness and vice, and sometime to theft and other abominations; none upon the said pain of imprisonment shall, under the colour of pity of alms, give anything to such, which may labour, or presume to favour them towards their desires, so that therby they may be compelled to labour for their necessary being.

Those statutes, which were enforced in the British colonies, went essentially unchanged into the eighteenth century, when they were grafted onto the laws of the new American statutes. The same configuration of meanings and policies in the old laws carried on into the nineteenth century. …

Another such principle was quicquid acquietur servo acquietur domino (“whatever is acquired by the servant is acquired by the master”). That principle was continued in the law of master and servant. A note in Hargrave’s eighteeth-century edition of Coke’s Institutes ⊕The Institutes of the Lawes of England are a series of legal treatises written by Sir Edward Coke, a prominent English barrister, judge and politician in the late sixteenth and early seventeenth centuries.

observed that the rule “about slaves holds in some degree in respect to apprentices and servants” and that it pertained with certainty to wages a worker earned from other employment when the master had given permission to the other employer without waiving the earnings.

By the late nineteenth century, employers in the United States could sue workers for breach of contract for their earnings in other employment only when it could be shown that the work had been done during hours and activities in which the plaintiff employer had been entitled to the worker’s efforts. In an 1877 Treatise on the Law of Master and Servant, however, Horace Wood⊕Wood’s treatise was an influential work on the law governing employment in the late 19th century. Among other things, it is widely cited as the primary source of the employment-at-will rule. See Jay M. Feinman, The Development of the Employment at Will Rule, 20 American Journal of Legal History 118, 125-27 (1976).

devoted three full pages to Hargrave’s “learned note” and concluded that under the rule of quicquid acquietur servo the employer could also retain waged earned by a worker in outside activities if the money somehow came into the employer’s hands. Moreover, even in its diluted American form, the law expressed the employer’s proprietary interest not only in the worker’s labor performed under the contract but also in all labor that might be performed by the worker’s person, under a different contract, The remedy went beyond simply dismissing the worker or deducting from his wages for so many hours; it extended to the outside earnings acquired, as if they (as an extension of the worker) belonged to the master. Such a principle would seem incongruous not simply with the market-model morality of enterprising employees, but even with the more sober depictions of nineteenth-century workers as earnest breadwinners for their families. It was eminently compatible, however, with other features, likewise ancient, of the hierarchical structure of employment relations during the period.

The judicial governance of master and servant

The links between past and present may be observed in the line of precedents used by American judges to decide the disputes between masters and servants and third parties that came before the courts. …

Regarding whether or not an employee might recover wages if asked by an employer to work on a Sunday, Wood proceeds for several pages citing precedents and authorities that include, among others, practice prior to the year 500, Edward the Confessor,⊕Edward the Confessor was the English king from 1042 to 1066.

The Mirror of Justices,⊕The Mirror of Justices was an Anglo-Norman law text published in 1642.

and Lord Coke, reaching the conclusion that the worker may not. The point is not only that the courts followed precedent, and the precedents were ancient, but also that there was a preestablished substance that constituted the legal relations between the parties, which the courts administered in the course of litigation.

As a remnant of feudalism, judicial regulation of labor relations in the nineteenth century maintained that comprehensive sense of governance, in which courts administered, and to a lesser extent, legislated, as well as adjudicated. Thus, for example, by the mid-nineteenth century, American courts, running ahead of the English, no longer permitted employers to beat their employees. Similarly, in the 1880s, judge-made law turned away from the assumption of an annual hiring, based on the English model and on the prevalence of agricultural labor, to a hiring “at-will.” Such modifications were no doubt responsible for the system’s survival for so long, preempting interference by other agencies of the state in unusually offensive or inconvenient circumstances. Even so, as in the case of at-will employment, the courts continued to prescribe labor relations, not derive them from contracts devised by the parties. …

Judicial governance of labor relations in the nineteenth century may best be seen in the details of cases that came regularly to the courts for decision and engaged central principles of the law. One principle was the property interest that the master legally had in the worker’s labor, and indeed in all labor performed by the worker’s person during the hours for which they contracted, as mentioned earlier in regard to the rule of quicquid acquietur servo. That property interest was further enforced in the action per quod servitium amisit (“by which the master lost the service”). In that action, the master might claim damages from a third party for injuries to his worker that resulted in loss of value of the worker’s services, much as if the injury had been to his chattel or machines or buildings. Blackstone, typically, found a parallel action regarding servants in ancient Athens; however, his discussion was more pertinent for its stress on the traditional nonreciprocity (in the per quod respect, as in others) of the employment contract: The servant, having no property in the “company, care, or assistance of the superior,” has no legal redress for injuries to the master.

[A] second principle in nineteenth-century labor law [was] the principle that a contract for labor was “entire.” Under that principle, a worker hired for a stated job or period of time was not legally entitled to be paid for any labor performed until the job or term was completed. If he or she quit work without legal cause, nothing could be recovered for the labor performed, unless the employee could prove that the contract had been wrongfully terminated by the employer. In that circumstance, wages could be sought by a quantum meruit (“for the amount owed”) suit or by a suit for damages for breach of contract. Legal recovery was difficult, not only because of the wide latitude given employers to discharge their workers but also because the courts were willing to accept as a justification for dismissal virtually any reason, even if that reason had not been stated or even known by the employer himself at the time the dismissal took place. If the suit was for damages, workers were required under the law to seek other employment after their dismissal, and the court normally would deduct any wages they received from what it required the defendant employer to pay.

This rule made it possible for employers to goad employees into quitting near the end of a term or pay period, and thereby benefit from their earlier labor without having to pay. But the “entire” contract principle was also basic to the full range of subject matters in dispute between nineteenth-century employers and employees, because very often other issues presented at law involved the question of whether or not a worker was entitled to, or had been unjustifiably denied, payment for his or her services. The principle was most important during the decades prior to the Civil War, when contracts definite as to term were more common than they would be later, but it retained its vitality after then as well, particularly with respect to salaried workers. The principle of hiring “at will,” under which either party could terminate the contract for any reason, began to take hold in the 1880s; but in the meantime many courts continued to apply the English rule of an assumed annual hiring, or else held that the pay period (by the week, month, etc.) would determine the point at which back wages could be accrued and recovered.

The province of work

The principles of hierarchy and obedience imply the existence of jurisdictional boundaries within which authority is organized and enforced. …

Jurisdiction was an attribute of the master’s property in the servant, just as was the claim over the servant’s labor. … To the extent that workplace relations under the nineteenth-century law of master and servant were governed by the courts as a unit, the combined workplaces constituted, concretely and not metaphorically, a single province of work within the larger territory of American society. …

Inside the province were the inhabitants: the employer and his employees. Those were real persons, not their alienated labor or functions. It was legal cause for immediate discharge to bring in an outside worker without the master’s consent to substitute for the one hired to do the work. Once having entered the contract for services, the worker was enlisted to be there and to perform during working hours, upon penalty of losing back pay for an infraction. Absense by mistake of a few days could cause discharge. The worker was not released from a tour of duty even under such circumstances as his or her accurate knowledge of the master’s imminent bankruptcy.

Beyond the territory of the individual workplace itself, the inhabitants, both employer and employees, engaged in what might be referred to as interworkplace relations, that is, transactions with other employers and employees. Although easy movement across that boundary would seem natural in market-model societies, real crossings encountered fixed barriers. In the case of employees, for example, a kind of passport often was required to move from the hire of one employer to another, which was the testimonial letter. … In the late nineteenth century, testimonial letters were required to obtain employment in many major industries, and litigation arose over questions of whether or not previous employers were under legal obligation to furnish them, how complete and truthful they were required to be, and how many times they had to be provided. Another barrier to employees’ free movement was that in nineteenth-century America, as in fourteenth-century England, it was illegal to harbor servants of another employer, that is, to employ a worker while knowing that he or she was still under a contract of employment with someone else. It was not essential that it be shown that the new labor had actually commenced, or that there had been an intention to deprive the first employer of his worker’s services; the mere retention of the employee after learning that service was due the other employer was subject to common-law suit for damages.

Crossings of workplace boundaries met numerous other obstacles. One obstacle was the courts’ enforcement of private agreements whereby the employee, upon departing service, agreed not to enter in business competition with his or her former employer for a specified number or years or over a specific territory. Another was the enforcement of what employers claimed were implied (as well as express) contracts that employees would not pass on to other certain “trade secrets,” even though the machinery or processes were already known by others and were unprotected by patents. Although contracts like these are certainly familiar in the recent history of American commerce, in relations among businessmen they had a controversial status, being regarded as illegal restraints of trade, and were not enforced consistently until late in the nineteenth century. But contrast, in the master-servant context, such contracts had long been regarded as reasonable exceptions necessary for the educative and confidential relations of employment.

Of the several protective barriers surrounding the workplace, the most formidable in the law of master and servant was the provision against enticement. The Statutes of Labourers provided for both civil and criminal proceedings against any person who knowingly enticed or persuaded a servant away from his employment by another master. By 1355, an action of trespass on the case had developed; it provided an independent civil remedy of damages for those same infractions within the common law. The law of enticement is a vivid illustration of the persistence of ancient regulations into the modern period. In the leading antebellum case, Boston Glass Manufactory v. Binney (1827), plaintiffs based their argument on cases extending back as far as 1591, and the 1591 case had been based on precedents for the action of enticement dating from the fourteenth century.

The relations between the workplace and [the outside world] may be seen in the developing law of employer liability for injuries caused in the course of carrying on a trade or business. With respect to “strangers,” third parties outside the company, the rule of respondeat superior (“let the mater answer”) prevailed. That rule meant that the employer was liable for injuries inflicted through the fault of an employee performing his authorized duties, just as if the injury had been caused by the employer’s machine or animal. Because prior to that time employees had themselves been held liable, unless the particular negligent act in question had been specifically commanded or implied by the mater, we may speculate that the newer rule … was an adaptation to the more attenuated forms of management typical of larger companies.

On the other hand, … when an injury was inflicted on an employee through negligence of a fellow servant, the master was not responsible. That judgment was consistent with the idea that the law must protect the public; however, judges would not intrude in established master-servant relations to protect the employee.

Under the rules of liability, … employees stood in relation to one another as an employee would to a piece of machinery. The employer would be held liable for an injury inflicted by a fellow employee only if the employer had not taken due care to ensure that the employee at fault had been sufficiently skilled to perform the task, just as the employer would be held liable if he or she had neglected to care for a piece of machinery.

Here the worker may be seen to have had less protection than members of the public at large against identical injuries caused by identical accidents. By virtue of one’s status as an employee of the company in the course of whose business the injury occurred, one was unprotected against injuries for which the company would have assumed liability had they been inflicted on an ordinary member of the public. The evident injustice of the fellow-servant rule, which adversely affected the families and other associates of injured workers, as well as the workers themselves, would eventually lead to a major revolution in torts through the institution of workmen’s compensation laws. Those laws would bring new inroads into the domain of master and servant. However, at a time when court decisions in other areas of the law were moving in the direction of universal contracts, the effect of workers’ compensation laws was to enhance the status-based responsibilities of the employment relation.

The province and the republic

A final barrier against interference in the workplace was constitutional. Labor relations were bounded by the limits of legislative sovereignty. Regular payment of wages, and in money rather than scrip or credit; reasons for discharge, and wages due and service letters upon departure; removal of the fellow-servant defense against liability for injury; shorter hours of employment—these and other changes in the old law were obtained by workers and their political allies in legislation, through the activities of lobbying and elections. However, as is well known, in the majority of instances those statutes were overturned in review by the judiciary.

Within the broad doctrine of substantive due process, “liberty of contract” came closest to denying the validity of legislation per se. Among the subjects treated by the judges under the doctrine of liberty of contract, labor questions were in the forefront. The initial reference to “undue interference with men’s rights of making contracts” appeared as a dictum in an opinion of the Illinois Supreme Court on legislation specifying how coal should be weighed to calculate miners’ wages. The first decision to invalidate a statute as an unconstitutional infringement on liberty of contract was Godcharles v. Wigeman, overturning a Pennsylvania statute requiring iron mills to pay in cash. The first U.S. Supreme Court decision to invalidate a state statute based on liberty of contract was Lochner v. New York.

The opinions in the labor decisions indicate that the judges believed that what was at stake was no less than the moral order of things, not merely the formal division of powers or the privileges of favorite social groups. Their well-known opposition to “class” legislation was based not so much on a sense of insult to republican principles as on their fears that the entire system of society and politics faced imminent demolition should the relation of master and servant be upset. … In Lochner, Justice Peckham said that if an eight-hour law for bakers were condoned, personal liberty under the constitution would become “visionary”:

Not only the hours of employees, but the hours of employers, could be regulated, and doctors, lawyers, scientists, all professional men, as well as athletes and artisans, could be forbidden to fatigue their brains and bodies by prolonged hours of exercise, lest the fighting strength of the state be impaired.

Haskins v. Royster, 70 N.C. 600 (1874)

Rodman, J.

We take it to be a settled principle of law, that if one contracts upon a consideration to render personal services for another any third person who maliciously, that is, without a lawful justification, induces the party who contracted to render the service to refuse to do so, is liable to the injured party in an action for damages. It need scarcely be said that there is nothing in this principle inconsistent with personal freedom, else we would not find it in the laws of the freest and most enlightened States in the world. It extends impartially to every grade of service, from the most brilliant and best paid to the most homely, and it shelters our nearest and tenderest domestic relations from the interference of malicious intermeddlers. It is not derived from any idea of property by the one party in the other, but is an inference from the obligation of a contract freely made by competent persons.

We are relieved from any labor in finding authorities for this principle, by a very recent decision of the Supreme Court of Massachusetts, in which a learned and able Judge delivers the opinion of the Court. Walker v. Cronin, 107 Mass. 555.

That case was this: The plaintiffs declared in substance that they were shoemakers, and employed a large number of persons as bottomers of boots and shoes, and defendant unlawfully and intending to injure the plaintiff in his business, persuaded and induced the persons so employed to abandon the employment of the plaintiff, whereby plaintiff was damaged.

A second count says that plaintiff had employed certain persons named to make up stock into boots and shoes, and defendant well knowing, induced said persons to refuse to make and finish such boots and shoes.

I shall make no apology for quoting copiously from this opinion, because the high respectability of the Court, and the learning and care with which the question is discussed, make the decision eminently an authority.

This (the declaration) sets forth sufficiently (1) intentional and willful acts, (2) calculated to cause damage to the plaintiffs in their lawful business, (3) done with the unlawful purpose to cause such damage and loss, without right or justifiable cause on the part of the defendant (which constitutes malice), and (4) actual damage and loss resulting.

In all cases where a man has a temporal loss or damage by the wrong of another, he may have an action upon the case to be repaired in damages.” The intentional causing such loss to another, without justifiable cause, and with the malicious purpose to inflict it, is of itself a wrong.

Thus every one has an equal right to employ workmen in his business or service; and if by the exercise of this right in such manner as he may see fit, persons are induced to leave their employment elsewhere no wrong is done to him whose employment they leave, unless a contract exists by which such other person has a legal right to the further continuance of their services. If such a contract exists, one who knowingly and intentionally procures it to be violated, may be held liable for the wrong, although he did it for the purpose of promoting his own business.

Every one has a right to enjoy the fruits and advantages of his own enterprise, industry, skill, and credit. He has no right to be protected against competition; but he has a right to be free from malicious and wanton interference, disturbance or annoyance. If disturbance or loss come as a result of competition, or the exercise of like rights by others, it is damnum absque injuria, unless some superior right by contract or otherwise is interfered with. But if it come from the merely wanton or malicious acts of others, without the justification of competition of the service of any interest or lawful purpose, it then stands upon a different footing, and falls within the principle of the authorities first referred to.

It is a familiar and well established doctrine of the law upon the relation of master and servant, that one who entices away a servant, or induces him to leave his master, may be held liable in damages therefor, provided there exists a valid contract for continued service known to the defendant. It has sometimes been supposed that the doctrine sprang from the English statute of laborers, and was confined to menial service. But we are satisfied that it is founded upon the legal right derived from the contract, and not merely upon the relation of master and servant, and that it applies to all contracts of employment, if not to contracts of every description.

It is suggested, (for we did not have the benefit of an argument for the defendant,) that in the present case the contract between the plaintiff and Eastwood and Wilkerson is unreasonable and therefore void. We cannot suppose it to be contended that this Court, or any Court, when there is no suggestion of fraud, can inquire whether the reward agreed to be paid to a workman is the highest that he might have got in the market, and to declare the contract void, or to make a new one if it thought not to be the highest. No Court can make itself the guardian of persons sui juris. That would be an assumption inconsistent with their freedom. We suppose the objection to a point to that part of the contract which is, in substance, that if either party of the second part, or any person for whom they contract, shall misbehave in the opinion of the party of the first part, such misbehaving party shall quit the premises and forfeit to the party of the first part all his interest in the common crop.

It is said that these provisions make the plaintiff a judge in his own cause, which the law will not allow, and that they are manifestly so oppressive and fraudulent as to avoid the whole contract. This proposition will be found on examination to go much too far even as between the parties to the contract, and to have no application as between one of the parties and a malicious intermeddler, as the defendant must, in this stage of the case, be considered.

It is not necessary to decide what would be the effect of such a stipulation in an action on the contract between the parties to it. But as there seems to be some misconception of the law of such a case, and as although there are numerous authorities on the question, it is not yet of “familiar learning” in our Courts, a few observations will more conveniently lead us to the question actually presented.

The authorities are conclusive that the parties to a contract, if there be no fraud or concealment of the interest, may agree to make a person interested, or even one of the parties an arbitrator to decide all controversies which may arise under the contract, and such agreement will be valid and effectual.

These authorities unquestionably establish that such stipulations are not void or voidable, even as between the parties, and it has never been supposed or contended that they made the whole contract void; as even if void themselves, they are clearly separable from the other parts. Either party, therefore, could maintain an action on this contract.

It is important however to notice, that none of these authorities goes to the length of holding, that if after the contractors had duly performed all or a part of the work, the plaintiff had mala fide, or without lawful cause, discharged them, they could not recover upon the contract. The power attempted to be reserved cannot have any greater effect than to make the discharge prima facie lawful, if so much as that.

Contracts with such stipulations as we find in the present, are not to be commended as precedents. Such stipulations are unusual; they answer no useful purpose, and suggest an intent (perhaps in this case untruly) to take some improper advantage, and to exact from the employees a degree of personal deference and respect, beyond that civil and courteous deportment which every man owes to his fellow in every relation in life. To this extent, a mutual duty is implied in every contract which creates the relation of master and servant. If the servant fails in due respect, the master may discharge him, and so, if the master fails, the servant will be justified in quitting the employment.

Again it is suggested, that the contractors of the second part in this contract are croppers, and not servants. By cropper, I understand a laborer who is to be paid for his labor by being given a proportion of the crop. But such a person is not a tenant, for he has no estate in the land, nor in the crop until the landlord assigns him his share. He is as much a servant as if his wages were fixed and payable in money.

It is unnecessary to discuss the question whether one who maliciously persuaded a tenant to abandon his holding, would not be liable in damages for such officious intermeddling.

But whatever may be the effect of the provisions commented on, as between the parties to the contract, the authorities are clear and decisive that a person in the situation of the defendant, can take no advantage from them. As the case now stands, he cannot pretend to play the part of a chivalrous protector of defrauded ignorance. For the present at least, he must be regarded as a malicious intermeddler, using the word malicious in its legal sense.

There is a certain analogy among all the domestic relations, and it would be dangerous to the repose and happiness of families if the law permitted any man, under whatever professions of philanthropy or charity, to sow discontent between the head of the family and its various members, wife, children and servants. Interference with such relations can only be justified under the most special circumstances, and where there cannot be the slightest suspicion of a spirit of mischief-making, or self interest.

To enable a plaintiff to recover from one who entices his servant, it is sufficient to show a subsisting relation of service, even if it be determinable at will. In Keane v. Boycott, the plaintiff sued a recruiting officer for enticing his servant. The servant was an infant and had been a slave in St. Vincents where he indentured himself to serve the plaintiff for five years. The indenture of course was void upon a double ground, but the Court held the plaintiff entitled to recover. “The defendant in this case had no concern in the relation between the plaintiff and his servant; he dissolved it officiously, and to speak of his conduct in the mildest terms, he carried too far his zeal for the recruiting service.”

We are of opinion that the complaint sets forth a sufficient cause of action.

Pollock v. Williams, 322 U.S. 4 (1944)

MR. JUSTICE JACKSON delivered the opinion of the Court.

Appellant Pollock questions the validity of a statute of the State of Florida making it a misdemeanor to induce advances with intent to defraud by a promise to perform labor and further making failure to perform labor for which money has been obtained prima facie evidence of intent to defraud. It conflicts, he says, with the Thirteenth Amendment to the Federal Constitution and with the antipeonage statute enacted by Congress thereunder. Claims also are made under the due process and equal protection clauses of the Fourteenth Amendment which we find it unnecessary to consider.

Pollock was arrested January 5, 1943, on a warrant issued three days before which charged that on the 17th of October, 1942, he did “with intent to injure and defraud under and by reason of a contract and promise to perform labor and service, procure and obtain money, to-wit: the sum of $5.00, as advances from one J.V. O’Albora, a corporation, contrary to the statute in such cases made and provided, and against the peace and dignity of the State of Florida.” He was taken before the county judge on the same day, entered a plea of guilty, and was sentenced to pay a fine of $100 and in default to serve sixty days in the county jail. He was immediately committed.

On January 11, 1943, a writ of habeas corpus was issued by the judge of the circuit court, directed to the jail keeper, who is appellee here. Petition for the writ challenged the constitutionality of the statutes under which Pollock was confined and set forth that “at the trial aforesaid, he was not told that he was entitled to counsel, and that counsel would be provided for him if he wished, and he did not know that he had such right. Petitioner was without funds and unable to employ counsel. He further avers that he did not understand the nature of the charge against him, but understood that if he owed any money to his prior employer and had quit his employment without paying the same, he was guilty, which facts he admitted.” The Sheriff’s return makes no denial of these allegations, but merely sets forth that he holds the prisoner by virtue of the commitment “based upon the judgment and conviction as set forth in the petition.” The Supreme Court of Florida has said that “undenied allegations of the petition are taken as true.”

The Circuit Court held the statutes under which the case was prosecuted to be unconstitutional and discharged the prisoner. The Supreme Court of Florida reversed. It read our decisions in Bailey v. Alabama and Taylor v. Georgia to hold that similar laws are not in conflict with the Constitution in so far as they denounce the crime, but only in declaring the prima facie evidence rule. It stated that its first impression was that the entire Florida act would fall, as did that of Georgia, but on reflection it concluded that our decisions were called forth by operation of the presumption, and did not condemn the substantive part of the statute where the presumption was not brought into play. As the prisoner had pleaded guilty, the Florida court thought the presumption had played no part in this case, and therefore remanded the prisoner to custody. An appeal to this Court was taken and probable jurisdiction noted.

Florida advances no argument that the presumption section of this statute is constitutional, nor could it plausibly do so in view of our decisions. It contends, however, (1) that we can give no consideration to the presumption section because it was not in fact brought into play in the case, by reason of the plea of guilty; (2) that so severed the section denouncing the crime is constitutional.

I.

These issues emerge from an historical background against which the Florida legislation in question must be appraised.

The Thirteenth Amendment to the Federal Constitution, made in 1865, declares that involuntary servitude shall not exist within the United States and gives Congress power to enforce the article by appropriate legislation. Congress on March 2, 1867, enacted that all laws or usages of any state “by virtue of which any attempt shall hereafter be made to establish, maintain, or enforce, directly or indirectly, the voluntary or involuntary service or labor of any persons as peons, in liquidation of any debt or obligation, or otherwise,” are null and void, and denounced it as a crime to hold, arrest, or return a person to the condition of peonage.

Clyatt v. United States was a case from Florida in which the Federal Act was used as a sword and an employer convicted under it. This Court sustained it as constitutional and said of peonage: “It may be defined as a status or condition of compulsory service, based upon the indebtedness of the peon to the master. The basal fact is indebtedness. Peonage is sometimes classified as voluntary or involuntary, but this implies simply a difference in the mode of origin, but none in the character of the servitude. The one exists where the debtor voluntarily contracts to enter the service of his creditor. The other is forced upon the debtor by some provision of law. A clear distinction exists between peonage and the voluntary performance of labor or rendering of services in payment of a debt. In the latter case the debtor, though contracting to pay his indebtedness by labor or service, and subject like any other contractor to an action for damages for breach of that contract, can elect at any time to break it, and no law or force compels performance or a continuance of the service.”

Then came the twice-considered case of Bailey v. Alabama, in which the Act and the Constitution were raised as a shield against conviction of a laborer under an Alabama act substantially the same as the one before us now. Bailey, a Negro, had obtained $15 from a corporation on a written agreement to work for a year at $12 per month, $10.75 to be paid him and $1.25 per month to apply on his debt. In about a month he quit. He was convicted, fined $30, or in default sentenced to hard labor for 20 days in lieu of the fine and 116 days on account of costs. The Court considered that the portion of the state law defining the crime would require proof of intent to defraud, and so did not strike down that part; nor was it expressly sustained, nor was it necessarily reached, for the prima facie evidence provision had been used to obtain a conviction. This Court held the presumption, in such a context, to be unconstitutional.

Later came United States v. Reynolds in which the Act of 1867 was sword again. Reynolds and Broughton were indicted under it. The Alabama Code authorized one under some circumstances to become surety for a convict, pay his fine, and be reimbursed by labor. Reynolds and Broughton each got himself a convict to work out fines and costs as a farm hand at $6.00 per month. After a time each convict refused to labor further and, under the statute, each was convicted for the refusal. This Court said, “Thus, under pain of recurring prosecutions, the convict may be kept at labor, to satisfy the demands of his employer.” It held the Alabama statute unconstitutional and employers under it subject to prosecution.

In Taylor v. Georgia the Federal Act was again applied as a shield, against conviction by resort to the presumption, of a Negro laborer, under a Georgia statute in effect like the one before us now. We made no effort to separate valid from invalid elements in the statute, although the substantive and procedural provisions were, as here, in separate, and separately numbered, sections. We said, “We think that the sections of the Georgia Code upon which this conviction rests are repugnant to the Thirteenth Amendment and to the Act of 1867, and that the conviction must therefore be reversed.” Only recently in a case from Northern Florida a creditor-employer was indicted under the Federal Act for arresting a debtor to peonage, and we sustained the indictment. United States v. Gaskin.

These cases decided by this Court under the Act of 1867 came either from Florida or one of the adjoining states. And these were but a part of the stir caused by the Federal Antipeonage Act and its enforcement in this same region. This is not to intimate that this section, more than others, was sympathetic with peonage, for this evil has never had general approval anywhere, and its sporadic appearances have been neither sectional nor racial. It is mentioned, however, to indicate that the Legislature of Florida acted with almost certain knowledge in designing its successive “labor fraud” acts in relation to our series of peonage decisions. The present Act is the latest of a lineage, in which its antecedents were obviously associated with the practice of peonage. This history throws some light on whether the present state act is one “by virtue of which any attempt shall hereafter be made” to “enforce involuntary servitude,” in which event the Federal Act declares it void.

In 1891, the Legislature created an offense of two elements: obtaining money or property upon a false promise to perform service, and abandonment of service without just cause and without restitution of what had been obtained. In 1905, this Court decided Clyatt v. United States, indicating that any person, including public officers, even if acting under state law, might be guilty of violating the Federal Act. In 1907, the Florida Legislature enacted a new statute, nearly identical in terms with that of Alabama. In 1911, in Bailey v. Alabama, this Court held such an act unconstitutional. In 1913, the Florida Legislature repealed the 1907 act, but re-enacted in substance the section denouncing the crime, omitting the presumption of intent from the failure to perform the service or make restitution. In 1919, the Florida Supreme Court held this act, standing alone, void under the authority of Bailey v. Alabama. Whereupon, at the session of 1919, the present statute was enacted, including the prima facie evidence provisions, notwithstanding these decisions by the Supreme Court of Florida and by this Court. The Supreme Court of Florida later upheld a conviction under this statute on a plea of guilty, but declined to pass on the presumption section, because, as in the present case, the plea of guilty was thought to make its consideration unnecessary. The statute was re-enacted without substantial change in 1941. Again in 1943 it was re-enacted despite the fact that the year before we held a very similar Georgia statute unconstitutional in its entirety.

II.

The State contends that we must exclude the prima facie evidence provision from consideration because in fact it played no part in producing this conviction. Such was the holding of the State Supreme Court. We are not concluded by that holding, however, but under the circumstances are authorized to make an independent determination.

What the prisoner actually did that constituted the crime cannot be gleaned from the record. The charge is cast in the words of the statute and is largely a conclusion. It affords no information except that Pollock obtained $5 from a corporation in connection with a promise to work which he failed to perform, and that his doing so was fraudulent. If the conclusion that the prisoner acted with intent to defraud rests on facts and not on the prima facie evidence provisions of the statute, none are stated in the warrant or appear in the record. None were so set forth that he could deny them. He obtained the money on the 14th of October, 1942, and the warrant was not sought until January 2, 1943. Whether the original advancement was more or less than $5, what he represented or promised in obtaining it, whether he worked a time and quit, or whether he never began work at all are undisclosed. About all that appears is that he obtained an advancement of $5 from a corporation and failed to keep his agreement to work it out. He admitted those facts and the law purported to supply the element of intent. He admitted the conclusion of guilt which the statute made prima facie thereon. He was fined $20 for each dollar of his debt, and in default of payment was required to atone for it by serving time at the rate of less than 9¢ per day.

Especially in view of the undenied assertions in Pollock’s petition we cannot doubt that the presumption provision had a coercive effect in producing the plea of guilty. The statute laid its undivided weight upon him. The legislature had not even included a separability clause. Of course the function of the prima facie evidence section is to make it possible to convict where proof of guilt is lacking. No one questions that we clearly have held that such a presumption is prohibited by the Constitution and the federal statute. The Florida Legislature has enacted and twice re-enacted it since we so held. We cannot assume it was doing an idle thing. Since the presumption was known to be unconstitutional and of no use in a contested case, the only explanation we can find for its persistent appearance in the statute is its extra-legal coercive effect in suppressing defenses. It confronted this defendant. There was every probability that a law so recently and repeatedly enacted by the legislature would be followed by the trial court, whose judge was not required to be a lawyer. The possibility of obtaining relief by appeal was not bright, as the event proved, for Pollock had to come all the way to this Court and was required, and quite regularly, to post a supersedeas bond of $500, a hundred times the amount of his debt. He was an illiterate Negro laborer in the toils of the law for the want of $5. Such considerations bear importantly on the decision of a prisoner even if aided by counsel, as Pollock was not, whether to plead guilty and hope for leniency or to fight. It is plain that, had his plight after conviction not aroused outside help, Pollock himself would have been unheard in any appellate court.

In the light of its history, there is no reason to believe that the law was generally used or especially useful merely to punish deceit. Florida has a general and comprehensive statute making it a crime to obtain money or property by false pretenses or commit “gross fraud or cheat at common law.” These appear to authorize prosecution for even the petty amount involved here. We can conceive reasons, even if unconstitutional ones, which might lead well-intentioned persons to apply this Act as a means to make otherwise shiftless men work, but if in addition to this general fraud protection employers as a class are so susceptible to imposition that they need extra legislation, or workmen so crafty and subtle as to constitute a special menace, we do not know it, nor are we advised of such facts.

We think that a state which maintains such a law in face of the court decisions we have recited may not be heard to say that a plea of guilty under the circumstances is not due to pressure of its statutory threat to convict him on the presumption.

As we have seen, Florida persisted in putting upon its statute books a provision creating a presumption of fraud from the mere nonperformance of a contract for labor service three times after the courts ruled that such a provision violates the prohibition against peonage. To attach no meaning to such action, to say that legally speaking there was no such legislation, is to be blind to fact. Since the Florida Legislature deemed these repeated enactments to be important, we take the Legislature at its own word. Such a provision is on the statute books for those who are arrested for the crime, and it is on the statute books for us in considering the practical meaning of what Florida has done.

In the view we take of the purpose and effect of this prima facie evidence provision it is not material whether as matter of state law it is regarded as an independent and severable provision.

III.

We are induced by the evident misunderstanding of our decisions by the Florida Supreme Court, in what we are convinced was a conscientious and painstaking study of them, to make more explicit the basis of constitutional invalidity of this type of statute.

The undoubted aim of the Thirteenth Amendment as implemented by the Antipeonage Act was not merely to end slavery but to maintain a system of completely free and voluntary labor throughout the United States. Forced labor in some special circumstances may be consistent with the general basic system of free labor. For example, forced labor has been sustained as a means of punishing crime, and there are duties such as work on highways which society may compel. But in general the defense against oppressive hours, pay, working conditions, or treatment is the right to change employers. When the master can compel and the laborer cannot escape the obligation to go on, there is no power below to redress and no incentive above to relieve a harsh overlordship or unwholesome conditions of work. Resulting depression of working conditions and living standards affects not only the laborer under the system, but every other with whom his labor comes in competition. Whatever of social value there may be, and of course it is great, in enforcing contracts and collection of debts, Congress has put it beyond debate that no indebtedness warrants a suspension of the right to be free from compulsory service. This congressional policy means that no state can make the quitting of work any component of a crime, or make criminal sanctions available for holding unwilling persons to labor. The federal statutory test is a practical inquiry into the utilization of an act as well as its mere form and terms.

Where peonage has existed in the United States it has done so chiefly by virtue of laws like the statute in question. Whether the statute did or did not include the presumption seems to have made little difference in its practical effect. In 1910, in response to a resolution of the House of Representatives, the Immigration Commission reported the results of an investigation of peonage among immigrants in the United States. It found that no general system of peonage existed, and that sentiment did not support it anywhere. On the other hand, it found sporadic cases of probable peonage in every state in the Union except Oklahoma and Connecticut. It pointed out that “there has probably existed in Maine the most complete system of peonage in the entire country,” in the lumber camps. In 1907, Maine enacted a statute, applicable only to lumber operations but in its terms very like the section of the Florida statute we are asked to separate and save. The law was enforcible in local courts not of record. The Commission pointed out that the Maine statute, unlike that of Minnesota and the statutes of other states in the West and South, did not contain a prima facie evidence provision. But as a practical matter the statute led to the same result.

The fraud which such statutes purport to penalize is not the concealment or misrepresentation of existing facts, such as financial condition, ownership of assets, or data relevant to credit. They either penalize promissory representations which relate to future action and conduct or they penalize a misrepresentation of the present intent or state of mind of the laborer. In these “a hair perhaps divides the false and true.” Of course there might be provable fraud even in such matters. One might engage for the same period to several employers, collecting an advance from each, or he might work the same trick of hiring out and collecting in advance again and again, or otherwise provide proof that fraud was his design and purpose. But in not one of the cases to come before this Court under the antipeonage statute has there been evidence of such subtlety or design. In each there was the same story, a necessitous and illiterate laborer, an agreement to work for a small wage, a trifling advance, a breach of contract to work. In not one has there been proof from which we fairly could say whether the Negro never intended to work out the advance, or quit because of some real or fancied grievance, or just got tired. If such statutes have ever on even one occasion been put to a worthier use in the records of any state court, it has not been called to our attention. If this is the visible record, it is hardly to be assumed that the off-the-record uses are more benign.

It is a mistake to believe that in dealing with statutes of this type we have held the presumption section to be the only source of invalidity. On the contrary, the substantive section has contributed largely to the conclusion of unconstitutionality of the presumption section. The latter in a different context might not be invalid. Indeed, we have sustained the power of the state to enact an almost identical presumption of fraud, but in transactions that did not involve involuntary labor to discharge a debt. James-Dickinson Farm Mortgage Co. v. Harry. Absent this feature any objection to prima facie evidence or presumption statutes of the state can arise only under the Fourteenth Amendment, rather than under the Thirteenth. In deciding peonage cases under the latter this Court has been as careful to point out the broad power of the state to create presumptions as it has to point out its power to punish frauds. It “has frequently recognized the general power of every legislature to prescribe the evidence which shall be received, and the effect of that evidence in the courts of its own government. In the exercise of this power numerous statutes have been enacted providing that proof of one fact shall be prima facie evidence of the main fact in issue; and where the inference is not purely arbitrary and there is a rational relation between the two facts, and the accused is not deprived of a proper opportunity to submit all the facts bearing upon the issue, it has been held that such statutes do not violate the requirements of due process of law.” Bailey v. Alabama. But the Court added that “the State may not in this way interfere with matters withdrawn from its authority by the Federal Constitution or subject an accused to conviction for conduct which it is powerless to proscribe.” And it proceeded to hold that the presumption, when coupled with the other section, transgressed those limits, for while it appeared to punish fraud the inevitable effect of the law was to punish failure to perform labor contracts.

In Taylor v. Georgia both sections of the Act were held unconstitutional. There the State relied on the presumption to convict. But it was not denied that a state has power reasonably to prescribe the prima facie inferences to be drawn from circumstantial evidence. It was the substance of the crime to establish which the presumption was invoked that gave a forbidden aspect to that method of short-cutting the road to conviction. The decision striking down both sections was not, as the Supreme Court of Florida thought, a casual and unconsidered use of the plural. Mr. Justice Byrnes knew whereof he spoke; unconstitutionality inhered in the substantive quite as much as in the procedural section and no part of the invalid statute could be separated to be salvaged. Where in the same substantive context the State threatens by statute to convict on a presumption, its inherent coercive power is such that we are constrained to hold that it is equally useful in attempts to enforce involuntary service in discharge of a debt, and the whole is invalid.

It is true that in each opinion dealing with statutes of this type this Court has expressly recognized the right of the state to punish fraud, even in matters of this kind, by statutes which do not either in form or in operation lend themselves to sheltering the practice of peonage. Deceit is not put beyond the power of the state because the cheat is a laborer nor because the device for swindling is an agreement to labor. But when the state undertakes to deal with this specialized form of fraud, it must respect the constitutional and statutory command that it may not make failure to labor in discharge of a debt any part of a crime. It may not directly or indirectly command involuntary servitude, even if it was voluntarily contracted for.

From what we have said about the practical considerations which are relevant to the inquiry whether any particular state act conflicts with the Antipeonage Act of 1867 because it is one by which “any attempt shall hereafter be made to establish, maintain or enforce” the prohibited servitude, it is apparent that we should not pass on hypothetical acts. Reservation of the question of the validity of an act unassociated with a presumption now, as heretofore, does not denote approval. The Supreme Court of Florida has held such an act standing alone unconstitutional. A considerable recorded experience would merit examination in relation to any specific labor fraud act. We do not enter upon the inquiry further than the Act before us.

Another matter deserves notice. In Bailey v. Alabama it was observed that the law of that state did not permit the prisoner to testify to his uncommunicated intent, which handicapped him in meeting the presumption. In Taylor v. Georgia, the prisoner could not be sworn, but could and did make a statement to the jury. In this Florida case appellee is under neither disability, but is at liberty to offer his sworn word as against presumptions. These distinctions we think are without consequence. As Mr. Justice Byrnes said in Taylor v. Georgia, the effect of this disability “was simply to accentuate the harshness of an otherwise invalid statute.”

We impute to the Legislature no intention to oppress, but we are compelled to hold that the Florida Act of 1919 as brought forward on the statutes as §§ 817.09 and 817.10 of the Statutes of 1941 are, by virtue of the Thirteenth Amendment and the Antipeonage Act of the United States, null and void. The judgment of the court below is reversed and the cause is remanded for further proceedings not inconsistent with this opinion.

Elizabeth Anderson, Private Government (2017)

Most workers in the United States are governed by dictatorships in their work lives. Usually, those dictatorships have the legal authority to regulate workers’ off-hour lives as well—their political activities, speech, choice of sexual partner, use of recreational drugs, alcohol, smoking, and exercise. Because most employers exercise this off-hours authority irregularly, arbitrarily, and without warning, most workers are unaware of how sweeping it is. Most believe, for example, that their boss cannot fire them for their off-hours Facebook postings, or for supporting a political candidate their boss opposes. Yet only about half of U.S. workers enjoy even partial protection of their off-duty speech from employer meddling. Far fewer enjoy legal protection of their speech on the job, except in narrowly defined circumstances. Even where they are entitled to legal protection, as in speech promoting union activity, their legal rights are often a virtual dead letter due to lax enforcement: employers determined to keep out unions immediately fire any workers who dare mention them, and the costs of litigation make it impossible for workers to hold them accountable for this.

Employees are pervasively subject to private government, as I have defined it. Why is this so? As far as the legal authority of the employer to govern employees was concerned, the Industrial Revolution did not mark a significant break. Legally speaking employers have always been authoritarian rulers, as an extension of their patriarchal rights to govern their households.

The Industrial Revolution moved the primary site of paid work from the household to the factory. In principle, this could have been a liberating moment, insofar as it opened the possibility of separating the governance of the workplace from the governance of the home. Yet industrial employers retained their legal entitlement to govern their employees’ domestic lives. In the early twentieth century, the Ford Motor Company established a Sociological Department, dedicated to inspecting employees’ homes unannounced, to ensure that they were leading orderly lives. Workers were eligible for Ford’s famous $5 daily wage only if they kept their homes clean, ate diets deemed healthy, abstained from drinking, used the bathtub appropriately, did not take in boarders, avoided spending too much on foreign relatives, and were assimilated to American cultural norms.

Workers today might breathe a sigh of relief, except that most are still subject to employer governance of their private lives. In some cases, this is explicit, as in employer-provided health insurance plans. Under the Affordable Care Act (ACA), employers may impose a 30 percent premium penalty on covered workers if they do not comply with employer-imposed wellness programs, which may prescribe exercise programs, diets, and abstinence from alcohol and other substances. In accordance with this provision, Penn State University recently threatened to impose a $100 per month surcharge on workers who did not answer a health survey that included questions about their marital situation, sexual conduct, pregnancy plans, and personal finances. In other cases, employer authority over workers’ off-duty lives is implicit, a by-product of the employment-at-will rule: since employers may fire workers for any or no reason, they may fire them for their sexual activities, partner choice, or any other choice workers think of as private from their employer, unless the state has enacted a law specifically forbidding employer discrimination on these grounds. Workplace authoritarianism is still with us.

The pro-market egalitarian aspiration toward nearly universal self-employment aimed to liberate workers from such governance by opening opportunities for nearly everyone to become their own boss. Why did it fail? Why are workers subject to dictatorship? Within economics, the theory of the firm is supposed to answer this question. It purports to offer politically neutral, technical, economic reasons why most production is undertaken by hierarchical organizations, with workers subordinate to bosses, rather than by autonomous individual workers. The theory of the firm contains important insights into the organization of production in advanced economies. However, it fails to explain the sweeping scope of authority that employers have over workers. What is worse, its practitioners sometimes even deny that workers lie under the authority of their bosses, in terms that reflect and reinforce an illusion of workers’ freedom that also characterizes much of public discourse. Both the theory of the firm, and public discourse, are missing an important reality: that workers are subject to their employers’ private government.

The pro-market egalitarian dream failed in part due to economies of scale. The technological changes that drove the Industrial Revolution involved huge concentrations of capital. A steam-powered cotton mill, steel foundry, cement or chemical factory, or railway must be worked by many hands. The case is no different for modern workplaces such as airports, hospitals, pharmaceutical labs, and computer assembly factories, as well as lower-tech workplaces such as amusement parks, slaughterhouses, conference hotels, and big-box retail stores. The greater efficiency of production using large, indivisible capital inputs explains why few individual workers can afford to supply their own capital. It explains why, contrary to the pro-market egalitarian hope, the enterprises responsible for most production are not sole proprietorships.

But economies of scale do not explain why production is not managed by independent contractors acting without external supervision, who rent their capital. One could imagine a manufacturing enterprise renting its floor space and machinery and supplying materials to a set of self-employed independent contractors. Each contractor would produce a part or stage of the product for sale to contractors at the next stage of production. The final contractor would sell the finished product to wholesalers, or perhaps back to the capital supplier. Some New England factories operated on a system like this from the Civil War to World War I. They were superseded by hierarchically organized firms. According to the theory of the firm, this is due to the excessive costs of contracting between suppliers of factors of production. In the failed New England system, independent contractors faced each other in a series of bilateral monopolies, which led to opportunistic negotiations. The demand to periodically renegotiate rates led contractors to hoard information and delay innovation for strategic reasons. Independent contractors wore out the machinery too quickly, failed to tightly coordinate their production with workers at other stages of production (leading to excess inventory of intermediate products), and lacked incentives to innovate, both with respect to saving materials and with respect to new products.

The modern firm solves these problems by replacing contractual relations among workers, and between workers and owners of other factors of production, with centralized authority. A manager, or hierarchy of managers, issues orders to workers in pursuit of centralized objectives. This enables close coordination of different workers and internalizes the benefits of all types of innovation within the firm as a whole. Managers can monitor workers to ensure that they work hard, cooperate with fellow workers, and do not waste capital. Because they exercise open-ended authority over workers, they can redeploy workers’ efforts as needed to implement innovations, replace absentees, and deal with unforeseen difficulties. Authority relations eliminate the costs associated with constant negotiation and contracting among the participants in the firm’s production. To put the point another way, the key to the superior efficiency of hierarchy is the open-ended authority of managers. It is impossible to specify in advance all of the contingencies that may require an alteration in an initial understanding of what a worker must do. Efficient employment contracts are therefore necessarily incomplete: they do not specify precisely everything a worker might be asked to do.

While this theory explains why firms exist and why they are constituted by hierarchies of authority, it does not explain the sweeping scope of employers’ authority over workers in the United States. It does not explain, for example, why employers continue to have authority over workers’ off-duty lives, given that their choice of sexual partner, political candidate, or Facebook posting has nothing to do with productive efficiency. Even worse, theorists of the firm appear not to even recognize how authoritarian firm governance is. Major theorists soft-pedal or even deny the very authority they are supposed to be trying to explain.

Consider Ronald Coase, the originator of the theory of the firm. He acknowledges that firms are “islands of conscious power.” The employment contract is one in which the worker “agrees to obey the directions of an entrepreneur.” But, he insists, “the essence of the contract is that it should only state the limits to the powers of the entrepreneur.” This suggests that the limits of the employer’s powers are an object of negotiation or at least communication between the parties. In the vast majority of cases, outside the contexts of collective bargaining or for higher-level employees, this is not true. Most workers are hired without any negotiation over the content of the employer’s authority, and without a written or oral contract specifying any limits to it. If they receive an employee handbook indicating such limits, the inclusion of a simple disclaimer (which is standard practice) is sufficient to nullify any implied contract exception to at-will employment in most states. No wonder they are shocked and outraged when their boss fires them for being too attractive, for failing to show up at a political rally in support of the boss’s favored political candidate, even because their daughter was raped by a friend of the boss.

What, then, determines the scope and limits of the employer’s authority, if it is not a meeting of minds of the parties? The state does so, through a complex system of laws—not only labor law, but laws regulating corporate governance, workplace safety, fringe benefits, discrimination, and other matters. In the United States, the default employment contract is employment-at-will. There are a few exceptions in federal law to this doctrine, notably concerning discrimination, family and medical leave, and labor union activity. For the most part, however, at-will employment, which entitles employers to fire workers for any or no reason, grants the employer sweeping legal authority not only over workers’ lives at work but also over their off-duty conduct. Under the employment-at-will baseline, workers, in effect, cede all of their rights to their employers, except those specifically guaranteed to them by law, for the duration of the employment relationship. Employers’ authority over workers, outside of collective bargaining and a few other contexts, such as university professors’ tenure, is sweeping, arbitrary, and unaccountable—not subject to notice, process, or appeal. The state has established the constitution of the government of the workplace: it is a form of private government.

Resistance to recognizing this reality appears to be widespread among theorists of the firm. Here, for example, is what Armen Alchian and Harold Demsetz say in their classic paper on the subject:

It is common to see the firm characterized by the power to settle issues by fiat, by authority, or by disciplinary action. This is delusion. The firm has no power of fiat, no authority, no disciplinary action any different in the slightest degree from ordinary market contracting between any two people. I can “punish” you only by withdrawing future business or by seeking redress in the courts for any failure to honor our exchange agreement. That is exactly all that any employer can do. He can fire or sue, just as I can fire my grocer by stopping purchases from him or sue him for delivering faulty products. What then is the content of the presumed power to manage and assign workers to various tasks? Exactly the same as one little consumer’s power to manage and assign his grocer to various tasks. To speak of managing, directing, or assigning workers to various tasks is a deceptive way of noting that the employer continually is involved in renegotiation of contracts on terms that must be acceptable to both parties. Telling an employee to type this letter rather than to file that document is like telling a grocer to sell me this brand of tuna rather than that brand of bread. I have no contract to continue to purchase from the grocer and neither the employer nor the employee is bound by any contractual obligations to continue their relationship.

Alchian and Demsetz appear to be claiming that wherever individuals are free to exit a relationship, authority cannot exist within it. This is like saying that Mussolini was not a dictator, because Italians could emigrate. While emigration rights may give governors an interest in voluntarily restraining their power, such rights hardly dissolve it.

Alternatively, their claim might be that where the only sanctions for disobedience are exile, or a civil suit, authority does not exist. That would come as a surprise to those subject to the innumerable state regulations that are backed only by civil sanctions. Nor would a state regulation lack authority if the only sanction for violating it were to force one out of one’s job. Finally, managers have numerous other sanctions at their disposal besides firing and suing: they can and often do demote employees; cut their pay; assign them inconvenient hours or too many or too few hours; assign them more dangerous, dirty, menial, or grueling tasks; increase their pace of work; set them up to fail; and, within very broad limits, humiliate and harass them.

Perhaps the thought is that where consent mediates the relationship between the parties, the relationship cannot be one of subordination to authority. That would be a surprise to the entire social contract tradition, which is precisely about how the people can consent to government. Or is the idea that authority exists only where subordinates obey orders blindly and automatically? But then it exists hardly anywhere. Even the most repressive regimes mostly rely on means besides sheer terror and brainwashing to elicit compliance with their orders, focusing more on persuasion and rewards.

Alchian and Demsetz may be hoodwinked by the superficial symmetry of the employment contract: under employment-at-will, workers, too, may quit for any or no reason. This leads them to represent quitting as equivalent to firing one’s boss. But workers have no power to remove the boss from his position within the firm. And quitting often imposes even greater costs on workers than being fired does, for it makes them ineligible for unemployment insurance. It is an odd kind of countervailing power that workers supposedly have to check their bosses’ power, when they typically suffer more from imposing it than they would suffer from the worst sanction bosses can impose on them. Threats, to be effective, need to be credible.

The irony is that Alchian and Demsetz are offering a theory of the firm. The question the theory is supposed to answer is why production is not handled entirely by market transactions among independent, self-employed people, but rather by authority relations. That is, it is supposed to explain why the hope of pro-market pre–Industrial Revolution egalitarians did not pan out. Alchian and Demsetz cannot bear the full authoritarian implications of recognizing the boundary between the market and the firm, even in a paper devoted to explaining it. So they attempt to extend the metaphor of the market to the internal relations of the firm and pretend that every interaction at work is mediated by negotiation between managers and workers. Yet the whole point of the firm, according to the theory, is to eliminate the costs of markets—of setting internal prices via negotiation over every transaction among workers and between workers and managers.

Alchian and Demsetz are hardly alone. Michael Jensen and William Meckling agree with them that authority has nothing to do with the firm; it is merely a nexus of contracts among independent individuals. John Tomasi, writing today, continues to promote the image of employees as akin to independent contractors, freely negotiating the terms of their contract with their employers, to obtain work conditions tailor-made to their idiosyncratic specifications. While workers at the top of the corporate hierarchy enjoy such freedom, as well as a handful of elite athletes, entertainers, and star academics, Tomasi ignores the fact that the vast majority of workers not represented by unions do not negotiate terms of the employer’s authority at all. Why would employers bother, when, by state fiat, workers automatically cede all liberties not reserved to them by the state, upon accepting an offer of work?

Not just theorists of the firm, but public discourse too, tend to represent employees as if they were independent contractors. This makes it seem as if the workplace is a continuation of arm’s-length market transactions, as if the labor contract were no different from a purchase from Smith’s butcher, baker, or brewer. Alchian and Demsetz are explicit about this, in drawing the analogy of the employment relation with the customer–grocer relation. But the butcher, baker, and brewer remain independent from their customers after selling their goods. In the employment contract, by contrast, the workers cannot separate themselves from the labor they have sold; in purchasing command over labor, employers purchase command over people.

What accounts for this error? The answer is, in part, that a representation of what egalitarians hoped market society would deliver for workers before the Industrial Revolution has been blindly carried over to the post–Industrial Revolution world. People continue to deploy the same justification of market society—that it would secure the personal independence of workers from arbitrary authority—long after it failed to deliver on its original aspiration. The result is a kind of political hemiagnosia:⊕Hemiagnosia, also known as hemispatial neglect, is a neuropsychologial condition, resulting from brain damage after a stroke or injury, in which a person loses awareness of objects and stimuli on one side of their body. Oliver Sacks described one such case in his book, The Man Who Mistook His Wife for a Hat.

like those patients who cannot perceive one-half of their bodies, a large class of libertarian-leaning thinkers and politicians, with considerable public following, cannot perceive half of the economy: they cannot perceive the half that takes place beyond the market, after the employment contract is accepted.

This tendency was reinforced by a narrowing of egalitarian vision in the transition to the Industrial Revolution. While the Levellers and other radicals of the mid-seventeenth century agitated against all kinds of arbitrary government, Thomas Paine mainly narrowed his critique to state abuses. Similarly, the Republican Party kept speaking mainly on behalf of the interests of businesspeople and those who hoped to be in business for themselves, even after it was clear that the overwhelming majority of workers had no realistic prospect of attaining this status, and that the most influential businesspeople were not, as Lincoln hoped, sole proprietors (with at most a few employees, the majority of whom were destined to rise to self-employed status after a few years), but managers in large organizations, governing workers destined to be wage laborers for their entire working lives. Thus, a political agenda that once promised equalizing as well as liberating outcomes turned into one that reinforced private, arbitrary, unaccountable government over the vast majority.

Finally, nineteenth-century laissez-faire liberals, with their bizarre combination of hostility toward state power and enthusiasm for hyperdisciplinary total institutions, attempted to reconcile these contradictory tendencies by limiting their focus to the entry and exit conditions of the labor contract, while blackboxing what actually went on in the factories. In fact, they did drive a dramatic improvement in workers’ freedom of entry and exit. Under the traditional common law of master and servant, employees were bound to their employers by contracts of one year (apprentices and indentured servants for longer), could quit before then only on pain of losing all their accrued wages, and were not entitled to keep wages from moonlighting. Other employers were forbidden to bid for their labor while they were still under contract. Workers were liberated from these constraints over the course of the nineteenth century.

This liberation, as is well-known, was a double-edged sword. Employers, too, were liberated from any obligation to employ workers. As already noted, the worst the workers could do to the boss often involved suffering at least as much as the worst the boss could do to them. For the bulk of workers, who lived at the bottom of the hierarchy, this was not much of a threat advantage, unless it was exercised collectively in a strike. They had no realistic hope under these conditions for liberation from workplace authoritarianism.

No wonder a central struggle of British workers in the mid-nineteenth century was for limits on the length of the working day—even more than for higher wages. This was true, even though workers at this period of the Industrial Revolution were suffering through “Engels’s pause”—the first fifty to sixty years of the Industrial Revolution during which wages failed to grow. My focus, like theirs, is not on issues of wages or distributive justice. It is on workers’ freedom. If the Industrial Revolution meant they could not be their own bosses at work, at least they could try to limit the length of the working day so that they would have some hours during which they could choose for themselves, rather than follow someone else’s orders.

That was an immediate aim of European workers’ movements in the mid-nineteenth century. As the century unfolded, workers largely abandoned their pro-market, individualistic egalitarian dream and turned to socialist, collectivist alternatives—that is, to restructuring the internal governance of the workplace. The problem was that the options open to workers consisted almost exclusively of private governments. Laissez-faire liberals, touting the freedom of the free market, told workers: choose your Leviathan. That is like telling the citizens of the Communist bloc of Eastern Europe that their freedom could be secured by a right to emigrate to any country—as long as they stayed behind the Iron Curtain. Population movements would likely have put some pressure on Communist rulers to soften their rule. But why should Leviathan set the baseline against which competition took place? No liberal or libertarian would be satisfied with a competitive equilibrium set against this baseline, where the choice of state governments is concerned. Workers’ movements rejected it for nonstate governments as well.

To their objection, libertarians and laissez-faire liberals had no credible answer. Let us not fool ourselves into supposing that the competitive equilibrium of labor relations was ever established by politically neutral market forces mediated by pure freedom of contract, with nothing but the free play of individuals’ idiosyncratic preferences determining the outcome. This is a delusion as great as the one that imagines that the workplace is not authoritarian. Every competitive equilibrium is established against a background assignment of property rights and other rights established by the state. The state supplies the indispensable legal infrastructure of developed economies as a kind of public good, and is needed to do so to facilitate cooperation on the vast scales that characterize today’s rich and sophisticated economies. Thus, it is the state that establishes the default constitution of workplace governance. It is a form of authoritarian, private government, in which, under employment-at-will, workers cede all their rights to their employers, except those specifically reserved for them by law.

Freedom of entry and exit from any employment relation is not sufficient to justify the outcome. To see this, consider an analogous case for the law of coverture, which the state had long established as the default marriage contract. Under coverture, a woman, upon marrying her husband, lost all rights to own property and make contracts in her own name. Her husband had the right to confine her movements, confiscate any wages she might earn, beat her, and rape her. Divorce was very difficult to obtain. The marriage contract was valid only if voluntarily accepted by both parties. It was a contract into subjection, entailing the wife’s submission to the private government of her husband. Imagine a modification of this patriarchal governance regime, allowing either spouse to divorce at will and allowing any clause of the default contract to be altered by a prenuptial agreement. This is like the modification that laissez-faire liberals added to the private government of the workplace. Women would certainly have sufficient reason to object that their liberties would still not be respected under this modification, in that it preserves a patriarchal baseline, in which men still hold virtually all the cards. It would allow a lucky few to escape subjection to their husbands, but that is not enough to justify the patriarchal authority the vast majority of men would retain over their wives. Consent to an option within a set cannot justify the option set itself.

I do not claim that private governments at work are as powerful as states. Their sanctioning powers are lower, and the costs of emigration from oppressive private governments are generally lower than the costs of emigration from states. Yet private governments impose a far more minute, exacting, and sweeping regulation of employees than democratic states do in any domain outside of prisons and the military. Private governments impose controls on workers that are unconstitutional for democratic states to impose on citizens who are not convicts or in the military.

The negative liberties most workers enjoy de facto are considerably greater than the ones they are legally entitled to under their employers. Market pressures, social norms, lack of interest, and simple decency keep most employers from exercising the full scope of their authority. We should care nevertheless about the insecurity of employees’ liberty. They work in a state of republican unfreedom, their liberties vulnerable to cancellation without justification, notice, process, or appeal. That they enjoy substantially greater negative liberty than they are legally entitled to no more justifies their lack of republican liberty than the fact that most wives enjoyed greater freedoms than they were legally entitled to justified coverture—or even coverture modified by free divorce.

Suppose people find themselves under private government. This is a state of republican unfreedom, of subjection to the arbitrary will of another. It is also usually a state of substantial constraints on negative liberty. By what means could people attain their freedom? One way would be to end subjection to government altogether. When the government is a state, this is the anarchist answer. We have seen that when the government is an employer, the answer of many egalitarians before the Industrial Revolution was to advance a property regime that promotes self-employment, perhaps even to make self-employment a nearly universally accessible opportunity, at least for men. This amounts to promoting anarchy as the primary form of workplace order.

The theory of the firm explains why this approach cannot preserve the productive advantages of large-scale production. Some kind of incompletely specified authority over groups of workers is needed to replace market relations within the firm. However, the theory of the firm, although it explains the necessity of hierarchy, neither explains nor justifies private government in the workplace. That the constitution of workplace government is both arbitrary and dictatorial is not dictated by efficiency or freedom of contract, but rather by the state. Freedom of contract no more explains the equilibrium workplace constitution than freedom to marry explained women’s subjection to patriarchy under coverture.

In other words, in the great contest between individualism and collectivism regarding the mode of production, collectivism won, decisively. Now nearly all production is undertaken by teams of workers using large, indivisible forms of capital equipment held in common. The activities of these teams are governed by managers according to a centralized production plan. This was an outcome of the Industrial Revolution, and equally much embraced by capitalists and socialists. That advocates of capitalism continue to speak as if their preferred system of production upholds “individualism” is simply a symptom of institutional hemiagnosia, the misdeployment of a hopeful preindustrial vision of what market society would deliver as if it described our current reality, which replaces market relations with governance relations across wide domains of production.

Workers in the nineteenth century turned from individualistic to collectivist solutions to workplace governance because they saw that interpersonal authority—governments over groups of workers—was inescapable in the new industrial order. If government is inescapable or necessary for solving certain important problems, the only way to make people free under that government is to make that government a public thing, accountable to the governed. The task is to replace private government with public government.

When the government is a state, we have some fairly good ideas of how to proceed: the entire history of democracy under the rule of law is a series of experiments in how to make the government of the state a public thing, and the people free under the state. These experiments continue to this day.

But what if the government is an employer? Here matters are more uncertain. There are four general strategies for advancing and protecting the liberties and interests of the governed under any type of government: (1) exit, (2) the rule of law, (3) substantive constitutional rights, and (4) voice. Let us consider each in turn.

Exit is usually touted as a prime libertarian strategy for protecting individual rights. By forcing governments to compete for subjects, exit rights put pressure on governments to offer their subjects better deals. “The defense against oppressive hours, pay, working conditions, or treatment is the right to change employers.” Given this fact, it is surprising how comfortable some libertarians are with the validity of contracts into slavery, from which exit is disallowed. In their view, freedom of contract trumps the freedom of individuals under government, or even the freedom to leave that government. While contracts into slavery and peonage are no longer valid, other contractual barriers to exit are common and growing. Noncompete clauses, which bar employees from working for other employers in the same industry for a period of years, have spread from technical professions (where nearly half of employees are subject to them) to jobs such as sandwich maker, pesticide sprayer, summer camp counselor, and hairstylist. While employers can no longer hold workers in bondage, they can imprison workers’ human capital. California is one of the few states that prohibit noncompete clauses. As the dynamism of its economy proves, such contractual barriers to exit are not needed for economic growth, and probably undermine it. There should be a strong legal presumption against such barriers to exit, to protect workers’ freedom to exit their employers’ government.

The rule of law is a complex ideal encompassing several protections of subjects’ liberties: (a) Authority may be exercised only through laws duly passed and publicized in advance, rather than arbitrary orders issued without any process. (b) Subjects are at liberty to do anything not specifically prohibited by law. (c) Laws are generally applicable to everyone in similar circumstances. (d) Subjects have rights of due process before suffering any sanctions for noncompliance. Not all of these protections, which were devised with state authority in mind, can be readily transferred to the employment context. Most of the solutions to problems the state must address involve regulations that leave open to individuals a vast array of options for selecting both ends and means. By contrast, efficient production nearly always requires close coordination of activities according to centralized objectives, directed by managers exercising discretionary authority. This frequently entails that the authority of managers over workers be both intensive (limiting workers to highly particular movements and words, not allowing them to pursue their own personal objectives at work or even to select their own means to a prescribed end) and incompletely specified. The state imposes traffic laws that leave people free to choose their own destinations, routes, and purposes. Walmart tells its drivers what they have to pick up, when and where they have to deliver it, and what route they have to take. In addition, managers need incompletely specified authority to rapidly reassign different tasks to different workers to address new circumstances. Finally, excessively costly procedural protections against firing also discourage hiring. All these obstacles to applying rule-of-law protections in the workplace empower employers to abuse their authority, subject workers to humiliating treatment, and impose excessive constraints on their freedom.

At the same time, it is easy to exaggerate the obstacles to imposing rule-of-law protections at work. Larger organizations generally have employee handbooks and standard practice guides that streamline authority along legalistic lines. Equal protection and due process rights already exist for workers in larger organizations with respect to limited issues. A worker who has been sexually harassed by her boss normally has recourse to intrafirm procedures for resolving her complaint. Such protections reflect a worldwide “blurring of boundaries” among business, nonprofit, and state organizations, which appears to be driven not simply by legal changes, but by cultural imperatives of scientific management and ideas of individual rights and organizational responsibilities. Some but not all of these managerial developments are salutary. They are proper subjects of investigation for political theory, once we get beyond the subject’s narrow focus on the state.

A just workplace constitution should incorporate basic constitutional rights, akin to a bill of rights against employers. To some extent, the Fair Labor Standards Act, anti-discrimination laws, and other workplace regulations already serve this function. A workers’ bill of rights could be strengthened by the addition of more robust protections of workers’ freedom to engage in off-duty activities, such as exercising their political rights, free speech, and sexual choices. Similar protections for employee privacy could be extended in the workplace during work breaks. The Occupational Safety and Health Administration (OSHA) prohibitions of particularly degrading, dangerous, and onerous working conditions can be viewed as part of a workers’ bill of rights. Nabisco once threatened its female production line workers with three-day suspensions for using the bathroom, and ordered them to urinate in their clothes instead. It was only in 1998 that OSHA issued a regulation requiring employers to recognize workers’ right to use a bathroom, after cases such as Nabisco’s aroused public outrage. Workers in Europe are protected from harassment of all kinds by anti-mobbing laws. This gives them far more robust workplace constitutional rights than workers in the United States, who may be legally harassed as long as their harassers do not discriminate by race, gender, or other protected identities in choosing their victims.

There are limits, however, to how far a bill of rights can go in protecting workers from abuse. Because they prescribe uniformity across workplaces, they can at best offer a minimal floor. In practice, they are also grossly underenforced for the least advantaged workers. Furthermore, such laws do not provide for worker participation in governance at the firm level. They merely impose limits on employer dictatorship.

For these reasons, there is no adequate substitute for recognizing workers’ voice in their government. Voice can more readily adapt workplace rules to local conditions than state regulations can, while incorporating respect for workers’ freedom, interests, and dignity. Just because workplace governance requires a hierarchy of offices does not mean that higher officeholders must be unaccountable to the governed, or that the governed should not play any role in managerial decision-making. In the United States, two models for workers’ voice have received the most attention: workplace democracy and labor unions. Workplace democracy, in the form of worker-owned and -managed firms, has long stood as an ideal for many egalitarians. While much could be done to devise laws more accommodating of this structure, some of its costs may be difficult to surmount. In particular, the costs of negotiation among workers with asymmetrical interests (for example, due to possession of different skills) appear to be high.

In the United States, collective bargaining has been the primary way workers have secured voice within the government of the workplace. However, even at its peak in 1954, only 28.3 percent of workers were represented by a labor union. Today, only 11.1 percent of all workers and 6.6 percent of private sector workers are represented. Although laws could be revised to make it easier for workers to organize into a union, this does not address difficulties inherent to the U.S. labor union model. The U.S. model organizes workers at the firm level rather than the industry level. Firms vigorously resist unionization to avoid a competitive disadvantage with non-unionized firms. Labor unions also impose inefficiencies due to their monopoly power. They also take an adversarial stance toward management—one that makes not only managers but also many workers uncomfortable. At the same time, they often provide the only effective voice employees have in workplace governance.

It is possible to design a workplace constitution in which workers have a non-adversarial voice in workplace governance, without raising concerns about monopolization. The overwhelming majority of workers in the United States would like to have such a voice: 85 percent would like firm governance to be “run jointly” by management and workers. In the United States, such a constitution is illegal under the National Labor Relations Act, which prohibits company unions. Yet this structure is commonplace in Europe. Germany’s system of codetermination, begun in the Weimar era and elaborately developed since World War II, offers one highly successful model.

It is not my intention in this lecture to defend any particular model of worker participation in firm governance. My point is rather to expose a deep failure in current ways of thinking about how government fits into Americans’ lives. We do not live in the market society imagined by Paine and Lincoln, which offered an appealing vision of what a free society of equals would look like, combining individualistic libertarian and egalitarian ideals. Government is everywhere, not just in the form of the state, but even more pervasively in the workplace. Yet public discourse and much of political theory pretends that this is not so. It pretends that the constitution of workplace government is somehow the object of voluntary negotiation between workers and employers. This is true only for a tiny proportion of privileged workers. The vast majority are subject to private, authoritarian government, not through their own choice, but through laws that have handed nearly all authority to their employers.

Workman v. United Parcel Service, Inc., 234 F.3d 998 (7th Cir. 2000)

Posner, Circuit Judge

This is a diversity suit, governed by Indiana law and resolved in favor of the defendant on summary judgment, for breach of contract and promissory estoppel. The plaintiff is an employee of UPS who claims that the company made a binding promise not to demote him without just cause and broke its promise.

On the merits, the plaintiff relies for both his contractual claim and his claim of promissory estoppel on a handbook that UPS gives its employees explaining its employment policies. Under the law of many states, such a handbook can create a binding contract if it contains clear promissory language that makes the handbook an offer that the employee accepts by continuing to work after receiving it. Indiana has yet to decide whether to follow these states. We need not speculate about whether it will. Even if we assume it will, and even if the UPS handbook could, as we doubt, be interpreted to contain a clear promise not to demote an employee except for cause, the plaintiff’s contractual claim is extinguished by the statement in the handbook that “this Policy Book is not a contract of employment and does not affect your rights as an employee of UPS.”

Such a disclaimer, if clear and forthright, as it is here, is a complete defense to a suit for breach of contract based on an employee handbook. Since an employer is under no legal obligation to furnish its employees with a statement of its employment policies, we cannot think of a basis for holding that any statement it does give them has to be legally binding. The only effect of such a rule would be to extinguish employee handbooks.

We are mindful of cases that hold that it is not enough for the handbook to disclaim creating an employment contract; it must state in addition that the employee can be terminated at the will of the employer.

The decisions that refuse to give effect to the short-form disclaimer strike us as paternalistic in the extreme. Employment at will is the norm in the United States. An employee therefore has no reason to presume that he has tenure, and a disclaimer that a handbook creates a contract is a clear statement that if he is fired he can’t sue for breach of contract. What more is needed? But there was more here, enough more perhaps to satisfy the courts that rendered the decisions we just cited: the statement that the handbook gives the employee no rights.

One might wonder what function an employee handbook serves if it does not create enforceable obligations. The answer is that it conveys useful information to the employee. And more—for to the extent that it does contain promises, even if not legally binding ones, it places the employer under a moral obligation, or more crassly gives him a reputational incentive, to honor those promises. Such promises may not be worth as much to the promisee as a promise that the law enforces, but they are worth more than nothing, and it is nothing that the employee can expect if employers must choose between nothing and giving up employment at will.

A disclaimer that is effective against a claim of breach of contract is also effective, we believe, against a claim of promissory estoppel. The function of the doctrine of promissory estoppel is to provide an alternative basis to consideration for making promises legally enforceable. A promise can be legally binding because it is supported by consideration or because it induces reasonable reliance, but in either case the promisor is free by a suitable disclaimer to deny any legally binding effect to the promise. To put this differently, consideration or reliance is a necessary but not a sufficient condition of the enforceability of a promise. Another necessary condition is that the promise be worded consistently with its being intended to be enforceable. Because of the disclaimer, that condition was not fulfilled in this case.