Martin Elson
Property Without Propertarianism ended in a field: own the field, but no one owns the farmer. That leaves a question every economy answers each morning and almost none asks out loud. When the farmer brings in a crop with someone else’s tractor, on someone else’s land, for someone else’s wage, whose crop is it?
The answer is settled before the work begins. Whoever hired the work owns what it makes and owes what it costs: the product, every asset it creates and every liability it runs up. Not the landlord, not the lender, and not the people who did the work. This chapter argues that the answer is wrong by a standard libertarians already accept everywhere else, the one a court uses when it asks who did a thing. The people who do the work, together, should own what it produces and answer for what it uses up. That is the labor theory of property. It doesn’t rest on the labor theory of value, or on any theory of value at all.
The case for the employer’s claim
Locke put the hired hand into the founding text himself. Labor mixed with what was common makes it one’s own, he wrote, and then: “the turfs my servant has cut … become my property, without the assignation or consent of any body.” The servant’s labor counts as the master’s because the master has bought it. Labor makes property, and labor can be bought, so whoever buys labor makes property.
Modern economics keeps the same shape in other words. A firm buys inputs at market prices: hours of labor, hours of machine time, the use of land, the use of money. Whoever pays for them all keeps what is left. Paul Samuelson put the legal side plainly in his textbook: since slavery was abolished, a man “must rent himself at a wage.” A rental is a lease like any other, and someone who rents out a car for a day doesn’t own the trip. John Bates Clark added that in a competitive market each input is paid what its last unit adds, so nothing is taken from anyone (Property, Labor and Equality answers him on his own ground).
Three arguments explain why the leftover goes where it does. Frank Knight argued in 1921 that someone must carry the uncertainty: the residual claimant promises fixed payments to everyone else and takes whatever remains, profit or loss. Armen Alchian and Harold Demsetz argued in 1972 that in team production no one can measure who made what, so someone must watch for shirking, and the watcher works hardest if the residual is his, along with the power to hire and fire. Israel Kirzner added an ethic in two steps. The first is a claim about fact: profit comes from seeing an opportunity others missed. The second is a rule he offers as a moral intuition most people share, not as something proved: what you discover is yours, finders keepers. These are serious arguments. They explain why the firm has the shape it has.
Where it goes wrong
Only persons answer. Ask a court who robbed the bank and it won’t name the crowbar. However much the tools added, the burglar answers for the burglary. Lawyers call this imputation: legal responsibility is assigned to those whose intentional actions brought the result about. David Ellerman’s point is that the labor theory of property is the same principle applied to property instead of crime. Those who knowingly use up inputs to make an output are responsible for it, so they should own the output and owe for the inputs. Marx drew the line in Capital: the worst architect differs from the best bee because he builds the cell in his mind before he builds it in wax. Tools, land and money carry human action; they don’t originate it. Economics hides the difference by writing production as an engineering function, where an hour of a person and an hour of a machine sit side by side as “input services.” An engineer has no use for responsibility. A jury can’t do without it.
The whole product includes the bill. The first socialists before Marx, Thomas Hodgskin and William Thompson among them, demanded labor’s right to the whole product, and Anton Menger later wrote its history. They left the door open to a caricature that Samuelson walked through: give labor the whole product and nothing goes to property. But the whole product has a negative side: the materials used up, the machine worn down, the loan to repay. Workers who owned the output would also owe for every input. Lenders still get interest, landlords rent, suppliers their price. Nothing is confiscated. What changes is who stands at the center of the ledger, holding the product and owing the costs.
| Current: the employment rule (renting people) | Labor theory of property | |
|---|---|---|
| Owns the product | Whoever hired the work | The people who do the work, managers included |
| Owes for inputs used up | Whoever hired the work | The people who do the work |
| Workers get | A wage, counted as a cost | The residual, profit or loss |
| Lenders get | Interest or the residual, depending on what they hold | Interest, as agreed |
| Landlords and owners of tools get | Rent, as agreed | Rent, as agreed |
| Who directs the work | Managers answerable to the owners of capital | Managers answerable to the members |
The servant was the first exclusion. Locke’s turfs show the labor theory of property written with the hired hand already taken out. If labor makes property, the servant’s turfs are the servant’s. Locke could hand them to the master only by treating labor as a thing that changes owners. Yet labor can’t be delivered by anyone but the laborer; the master gets the turfs only because the servant cut them himself, under orders. The law admits as much when the work is a crime. A hired hand who breaks the law on the boss’s orders can’t plead the contract: both answer as partners. The same law, when the work is legal, gives the whole product to the boss. That is the pattern The Unfinished Exclusionstraces: a premise drawn wide, then stopped short where stopping suits the people drawing it.
The rental becomes a sale. Samuelson said that since slavery was abolished, human earning power can no longer be capitalized: no one can sell a person, or a claim on a person’s whole future work. About persons he was right. But something else can be sold, and is. Whoever holds the center of the ledger in a going firm holds a standing right to rent a team of people and keep what they make. That right can be bought and sold separately from the people, and every share of stock is a slice of it.
The clearest view of it comes when the team itself is the thing being bought. John Coyle and Gregg Polsky described the “acqui-hire” in 2013: large technology companies buying startups mainly to hire their engineers, with little interest in the products. Much of the price goes to the engineers, but the investors are paid too, for people they don’t own and who, under California law, could simply have walked out. In 2024 Microsoft hired most of Inflection AI’s staff of about seventy without buying the company. It paid Inflection about $620 million to license its models, money Inflection used to repay its investors, and about $30 million to waive legal claims over the hiring. Some of that bought real software. The $30 million bought one thing: permission to hire people nobody owned.
The claim is easy to overstate, so it helps to say what it isn’t. In these deals most of the money still goes to the workers, because they can leave, and that is exactly the difference between renting people and owning them. The gap between a company’s market price and its net assets isn’t all captured labor either; it also holds brands, patents and monopoly rents, much of them state privilege. What gets capitalized is narrower than “labor”: the expectation that people will keep working under the rental, together with the legal right to keep what they make while they do. That is a smaller claim than the old one, and a sturdier one. A right to keep the product of other people’s future work is bought and sold every day, on every stock exchange.
Value is a different question. Marx tried to answer the question with a theory of value: labor alone creates value, so profit is unpaid labor. That turned a question about ownership into one about prices, and on prices the marginalists had better answers. Ellerman suspects orthodox economists like it that way, since a Marx they can beat makes it look as if there is no other alternative. The labor theory of property says nothing about price. Pay every input its full market price and the question is still open: who owns the product? Marx and Clark disagree about wages, but both assume the product belongs to whoever owns the means of production. Ellerman calls that assumption the fundamental myth. Owning a machine gives you a claim to be paid for its use, not to what others make with it.
The strongest case against
Alchian and Demsetz make the best case, and part of it should be granted. Teams do need monitoring, and whoever monitors does need a stake. But nothing in their argument says the monitor must answer to the owners of capital rather than to the people being monitored. In a worker cooperative the members elect managers and can remove them, and every member shares the residual, so everyone has a reason to watch. The evidence is better than the theory predicts. Virginie Pérotin’s 2016 review of international studies found worker cooperatives at least as productive as conventional firms, at least as long-lived, and steadier employers, cutting pay before jobs in a downturn.
Henry Hansmann asks the sharper question: if cooperatives work so well, why are there so few? His answer in The Ownership of Enterprise (1996) is that collective decisions are costly. Investors want one thing, a return, while workers differ in age, skill and plans, and their quarrels cost money. That is fair, and partly true. But which forms win under today’s rules tells us which forms today’s rules favor. Law, finance and tax are built around the rental, and Property, Labor and Equality shows how much of that is state privilege. Survival can’t settle validity anyway. Robert Fogel and Stanley Engerman argued in Time on the Cross (1974) that slave plantations were more efficient than free farms. Their numbers are still disputed, and no one took them as a case for slavery.
Knight’s argument from risk proves less than it seems to. Risk can be shared and priced: lenders, insurers and holders of nonvoting shares all bear risk, and none of them owns the product. Workers already carry the risk they can least spread out, a livelihood tied to one firm. Kirzner’s finders keepers fails at a different point. Seeing an opportunity is work, and under the labor theory a founder who works in the firm shares in what it makes. What discovery can’t cover is the work other people do once the opportunity is found. You can find a field. You can’t find the harvest. A finder who wants to keep the whole harvest can bring it in himself.
Where this lands
Who owns what work produces is not a question about price. It is the question a court asks: who did it? Persons act and things don’t, so the people who work in an enterprise, managers included, should own what it makes and owe for everything it uses up. Those who supply land, tools and money are paid what they agreed, as lenders and lessors, not as owners of other people’s work. This takes nothing from anyone. It is the rule libertarians already apply to crime, applied to property. Locke wrote the premise and then took the servant out. Put him back, and the labor theory of property needs no theory of value at all.
That closes Part VI and leaves one question open. If the people who make something own it, what should they charge for it? Cost the Limit of Price opens Part VII by pulling apart three claims that often travel together, this chapter’s among them, and then takes up Josiah Warren’s answer: no more than it cost.
Sources
- David Ellerman, “The Labor Theory of Property” (1985). Labor as the only factor that can be responsible; the juridical principle of imputation; the whole product, negative side included. on ellerman.org. A later version against marginal productivity theory is “The Labour Theory of Property and Marginal Productivity Theory” (Economic Thought 5, no. 1, 2016): 19–36, PDF.
- David Ellerman, Property and Contract in Economics: The Case for Economic Democracy (Blackwell, 1992). The labor theory of property and the fundamental myth. copy here (PDF)
- David Ellerman, “Reframing the Labor Question: On Marginal Productivity Theory and the Labor Theory of Property.” Marx as the convenient alternative; Rothstein on capital’s right of command. original (PDF)
- John Locke, Second Treatise of Government (1689), ch. 5, §28. The turfs the servant cut. Public domain.
- Paul A. Samuelson, Economics (10th ed., McGraw-Hill, 1976), p. 52. Human earning power forbidden to be capitalized; renting oneself at a wage.
- Karl Marx, Capital, vol. 1 (1867), ch. 7. The architect and the bee; instruments of labor as conductors of activity.
- Thomas Hodgskin, Labour Defended against the Claims of Capital (1825) and The Natural and Artificial Right of Property Contrasted (1832). Labor’s claim to its product. Public domain.
- Anton Menger, The Right to the Whole Produce of Labour (1886; English translation, Macmillan, 1899). The history of the claim. Public domain.
- John Bates Clark, The Distribution of Wealth (Macmillan, 1899). Marginal productivity theory.
- Frank H. Knight, Risk, Uncertainty and Profit (Houghton Mifflin, 1921). The residual claimant as bearer of uncertainty.
- Armen A. Alchian and Harold Demsetz, “Production, Information Costs, and Economic Organization” (American Economic Review 62, no. 5, 1972): 777–95. Team production and the monitor as residual claimant.
- Israel M. Kirzner, Discovery, Capitalism, and Distributive Justice (Basil Blackwell, 1989). Finders keepers.
- Henry Hansmann, The Ownership of Enterprise (Harvard University Press, 1996). Why worker ownership is rare: the costs of collective decisions.
- Virginie Pérotin, What Do We Really Know about Worker Co-operatives? (Co-operatives UK, 2016; first given at the ILO–ICA Research Conference, Antalya, 2015). Productivity, survival, investment and employment compared. 2015 conference version
- Robert W. Fogel and Stanley L. Engerman, Time on the Cross: The Economics of American Negro Slavery(Little, Brown, 1974). The efficiency of slave agriculture, still disputed.
- John F. Coyle and Gregg D. Polsky, “Acqui-hiring” (Duke Law Journal 63, 2013): 281–346. Startups bought mainly to hire their engineers; how the price is split between engineers and investors. article
- Microsoft and Inflection AI, March 2024: most of the staff hired, about $620 million paid to license Inflection’s models (used to repay its investors) and about $30 million to waive claims over the hiring. Reported by Reuters and The Information; summary in The Batch. summary