Cost the Limit of Price

Martin Elson

Three old ideas travel together on the libertarian left, and they are often taken for one. The first says who owns what a firm produces. The second says what a thing should sell for. The third says what borrowed money should cost. You can hold any one of them without the others, and each has to be argued on its own. This page takes the second, Warren’s, gives it its strongest objection, and says what I’d keep, for now. Its companion, What Prices Carry, takes up the rules behind the prices we actually pay, and the third idea, interest.

This is not Econ 101. If you took it, you will recognize the ground: supply and demand, prices as signals, what happens under a price ceiling. Then you will see it pointing somewhere your textbook never went, at the property rules the textbook takes as given and rarely questions.

Three claims, not one

David Ellerman’s labor theory of property is the first. The people who work in a firm are responsible for what it produces and what it uses up, so they should own the product and answer for the inputs. Property, Labor and Equalitymakes that case. It is a claim about ownership, and it says nothing about price. A worker co-op that sells at whatever the market will bear satisfies it completely.

Josiah Warren’s cost the limit of price is the second. No one should be paid more for a thing than it cost them to provide. That is a claim about exchange, and it is this page’s subject.

The mutualist case against interest is the third. Pierre-Joseph Proudhon, William Greene, Benjamin Tucker and, today, Kevin Carson argue that interest is mostly a monopoly price for credit, held up by laws that limit who may issue money. That is a claim about banking, and What Prices Carry weighs it.

The separation matters because Ellerman’s own frame lets a democratic firm borrow and pay interest. Labor hires capital and pays rent on it. Renting things is fine; renting people is not. So the claim that a co-op paying interest reproduces the alienist structure at one remove is not Ellerman’s. It is a mutualist claim, and it has to stand on mutualist arguments. Nothing in the inalienist test settles it. A reader can accept Ellerman and reject Warren, or accept Warren and still think interest is a fair price for waiting.

Warren came back from Robert Owen’s commune at New Harmony convinced that holding everything in common had failed, and in 1827 he opened a store in Cincinnati to test the opposite idea. Goods sold at what he had paid for them, plus a small charge for the store’s upkeep. His own time behind the counter the customer paid for separately, in a labor note: a written promise of an equal stretch of the customer’s own work, timed by a clock on the wall. The Time Store ran for about three years, and he later tried the principle in whole villages, Utopia in Ohio and Modern Times on Long Island. True Civilization (1863) sets out the theory in its fourth chapter, and Stephen Pearl Andrews gave it its fullest statement in The Science of Society (1851).

Warren’s cost is not Marx’s labor time. Cost, he wrote, is the pain, discomfort or sacrifice we bear in what we do. An hour of dirty or dangerous work costs more than an hour of pleasant work, and should fetch more. Value, how much the buyer wants the thing, is a different matter, and Warren thought letting it set prices was the root of speculation, booms and bankruptcies. Land in its natural state cost no one anything, so it should have no price at all.

The word that carries the weight is limit. A limit can be read two ways. It can be a ceiling, held by rule: no one may charge more than cost. Or it can be where prices end up when no one is protected from competition. Warren meant the second. Profit and usury, he held, would disappear under perfectly free competition, not by decree. That is the reading this page defends, and the section on the strongest case against shows why the first reading fails.

Read that way, the principle is less strange than it sounds. Mainstream economics says something close: in a competitive market, price tends toward cost and profit above a normal return is competed away. The quarrel is over what counts as cost. The textbook counts rent on land, interest on money and royalties on patents as payments to factors of production. The mutualist says much of that is a toll paid for a legal privilege, and that a free market would shrink it.

Three ways to set a price

Three ways of setting a price get run together, so it helps to set them side by side. Market pricing starts from the buyer: the seller asks what people will pay and charges that, whatever the thing cost. Surge pricing, and the airline seat that cost three times what the one beside it did, are the pure cases. Cost-plus starts from the seller: take what the thing cost and add a markup, often a fixed percentage. Cost-price, Warren’s rule, also starts from cost, but adds nothing beyond the seller’s own real costs, overhead and time included, and shows the buyer the figures. The first two sound different, and a business can describe itself either way, but in practice they meet. Whatever the method, the price is cost plus some markup, and the markup is as large as the seller can make stick. Only the third sets the markup at zero and puts the cost where the buyer can see it.

So the useful question is not which label a seller uses but what decides the size of the markup. Two things decide it: how much competition the seller faces, and how much the buyer knows.

The second is easy to miss. If one shop sells a thing for five dollars and another a mile away sells the same thing for six, the six-dollar shop survives only because some buyers don’t know about the other one, or can’t spare the time to find out. George Stigler argued in 1961 that the spread of prices for the same good in the same place measures how ignorant buyers are. Peter Diamond went further in 1971: if checking another shop costs buyers even a little, every seller can charge close to what a monopolist would, however many sellers there are. The reverse shows up when searching gets cheap. As comparison websites spread for term life insurance in the late 1990s, Jeffrey Brown and Austan Goolsbee found that prices fell by roughly 8 to 15 percent.

Not every difference in margin is a squeeze. The shop a mile away may pay more rent, stay open later or know its stock better, and those are real costs. The trouble is that a buyer can’t tell an honest margin from a squeeze when costs are hidden. Anyone who has walked into a shop with no price tags, and watched the person at the counter size them up before naming a figure, knows what Warren was against. His answer at the Time Store was to show the cost: in his own description of the store (1829), the keeper showed the bills for everything he bought “to public view so that the cost of every article may be known to all,” and his own margin was his time, paid by the clock. Cost the limit of price is a rule about honesty, and its defenders put it more bluntly. Tucker called interest, rent and profit the three forms of usury, and Proudhon had already called property theft.

So where is the border? A markup that pays for rent, service or the seller’s time is a cost by another name. A markup that lasts only because the buyer can’t see the cost, or can’t go elsewhere, is something else. It is not robbery in the strict sense, since no one is forced to buy. But it is a taking: payment extracted for nothing the seller supplied. The test is the question Start Here asks of every arrangement, can the person leave?, plus the one still being weighed as a third line: can they see what is being done to them? What Prices Carry sets out the four ways the leaving half usually fails.

Capitalist price, socialist price

Has anyone spoken of a socialist price? Yes, in two senses that point in opposite directions.

The first is Soviet. Price formation under “socialism” was a whole field of Soviet economics, and its answer was cost-plus. Planners set wholesale prices at the average cost of the branch plus a markup decided in Moscow. In the 1960s Soviet economists argued at length over how big that markup should be and what it should be proportional to, wages, materials or capital, and the 1967 price reform set it so that every branch earned roughly the same rate of profit. The socialist price of the twentieth century, then, was cost-plus set by a ministry. The markup survived. It went to the state. That is why critics on the left, anarchists among the first, called the system state capitalism: the owner changed, and the toll for owning stayed.

The second socialist price is older. Tucker called himself a socialist, and in “State Socialism and Anarchism” he traced socialism to a single principle that Warren, Proudhon and Marx shared: labor is the true measure of price, or in Warren’s words, cost the limit of price. From it, Tucker wrote, follows that the natural wage of labor is its product, and that anyone with income from another source takes it from that wage, in three forms: interest, rent and profit. Proudhon’s name for the taking was the droit d’aubaine, the right of increase: income that comes to an owner simply for owning. On this reading socialism is a claim about what a price may pay for, not about who sets it. Marx handed the job to a state; Warren and Proudhon handed it to markets freed of privilege.

So there are two questions that usually get run together. One is who sets the price, a market or a planner. The other is what the price pays for: work, sacrifice and real costs, or ownership as such. In the sense this site opposes, capitalism is a system of prices that pay ownership as such, a toll for holding title. A socialist price in Tucker’s sense pays for everything the seller supplied and nothing for the title. The Soviet price fails the second test, with a planner setting it. Today’s market price fails it too, through the four ways a market gets bent that What Prices Carry sets out. What this site defends is the remaining corner, the one Gary Chartier and Charles Johnson call socialist ends by market means: a market sets the price, and open entry and visible costs push ownership’s share toward zero.

That gives the border asked about earlier a name. Gratuitous appropriation is the part of a price that pays for nothing supplied: not work, not real cost, not risk borne. The strongest reply comes from Frank Knight. Profit, he argued in 1921, is the reward for bearing uncertainty that no one can insure, and someone has to bear it. Israel Kirzner adds that profit rewards spotting what others missed. Both are real, and neither is a taking. Warren’s cost already counts sacrifice, so honestly borne risk is a cost, and spotting what others missed is the profit from knowing that What Prices Carrydefends. What neither defends is a return that outlasts the risk and the discovery and is kept by a fence. Here is the test, set in advance: if, where entry is open and costs can be seen, owners still collect well above what their work and risk would earn, then Tucker was wrong and ownership is supplying something this page hasn’t named.

The same split runs the other way. A price can pay ownership for something nobody supplied, and it can also leave out costs that somebody did bear. The old phrase for it is profits privatized, costs socialized: call it a capitalist price with a socialist cost. The owner keeps the margin, and the bill for what the price left out goes to neighbors, taxpayers or whoever comes later. Arthur Pigou named the first kind in 1920: a factory whose smoke dirties the town’s washing sells its goods below their real cost, because part of that cost is paid in laundry and lungs. The second kind shows up in a crisis. In October 2008 Congress authorized up to $700 billion to rescue banks whose gains in the boom had stayed private, and a bank that expects to be rescued borrows more cheaply than one that doesn’t, so the guarantee pays out even in good years.

For Warren this is the same fault seen from the other side. His cost counts every sacrifice a thing takes, so a price that leaves some of them on people who never agreed to bear them is not at cost. It is cheap for the buyer and rich for the owner, and someone with no say pays the difference. The freed-market reading offers the same fix it offers for the toll: no rescue for losses, liability for harm done, and every cost on the bill where the buyer can see it. A market that let owners keep the gains and pass on the losses would not be a free one. It would be the fence again, built this time around the costs.

The strongest case against

Hold a price at cost when demand runs ahead of supply and you get a queue. That is the objection to cost-price at full strength, and the evidence for it is strong. By cost-price I mean selling at what a thing cost to provide, with no margin on top: no profit, and no rent or interest beyond the real cost of the land or credit used. Since every price is cost plus a markup, cost-price differs from cost-plus pricing and from market pricing only in the size of the markup, which is zero, and in the fact that the buyer can see the cost.

In 1994 San Francisco extended rent control to small apartment buildings built before 1980. Rebecca Diamond, Timothy McQuade and Franklin Qian followed what happened. Covered tenants were much likelier to stay put, which is what the law was for. But landlords cut the rental housing they offered by 15 percent, converting buildings to condos or redeveloping them, and rents across the city rose by about 5 percent. So the tenants already in covered flats gained, and people who came to the city later paid for it: with fewer flats to rent, they faced higher rents. The authors put about two-fifths of the total cost on those later arrivals.

The Soviet Union ran the experiment on a whole economy. János Kornai called it a shortage economy: queues, waiting lists for flats and cars measured in years, buyers taking whatever was on the shelf, firms hoarding inputs against the next shortfall. A queue rations by free time and connections, not by need. And a price held down by rule stops telling anyone to produce more.

These shortages were chronic, not occasional. Any version of cost-price that holds prices at cost and lets the extra demand show up as queues is the version this evidence refutes.

The answer has to be the freed-market one. Warren’s limit is not a ceiling. When demand jumps, the price is free to rise, and the rise is the signal Friedrich Hayek described: it calls in more producers. Cost the limit of price is the claim that, where entry is open, prices fall back toward cost, and that what stays above cost is mostly a toll. That is a prediction, and it can be checked where barriers fall. When a drug’s patent expires and generic makers enter, prices drop. The FDA found that with one generic maker the price ran about 39 percent below the brand’s, and with six or more, over 95 percent below. Where a barrier stays, the price stays.

Two hard questions remain. Patents are defended as the way drug research gets paid for, and the freed-market answer owes an account of how discovery is funded without them. And Warren’s natural land, which costs no one anything, is still scarce, so something must decide who uses which piece: occupancy and use, a land value tax, or commons governance. Property Without Propertarianism takes up the second.

What I’d keep

Warren’s principle is a prediction about what prices do once privilege is removed, not a rule to enforce. Read that way it survives the strongest case against it, and it names what it opposes: the part of a price that pays for nothing supplied. Ellerman settles who owns the product; Warren says what the product may sell for. Whether real prices ever get there depends on the rules they are set under, and that is where What Prices Carry picks up.

Sources

Josiah Warren, True Civilization an Immediate Necessity (1863), ch. 4. The principle in Warren’s own words. Public domain. original
Josiah Warren, “A Brief Outline of Equitable Commerce” (1852). Cost as pain, discomfort or sacrifice; value kept apart from cost; natural land with no price. original
Josiah Warren, “Plan of the Cincinnati Labor for Labor Store” (Mechanics’ Free Press, 1829). Warren’s own account of the Time Store: posted bills, the clock, labor notes. Public domain. original
Stephen Pearl Andrews, The Science of Society (1851), part 2, “Cost the Limit of Price.” The fullest statement, which Warren endorsed. original
George J. Stigler, “The Economics of Information” (Journal of Political Economy 69, no. 3, 1961): 213–25. Price dispersion as a measure of buyers’ ignorance.
Peter A. Diamond, “A Model of Price Adjustment” (Journal of Economic Theory 3, no. 2, 1971): 156–68. With any search cost, sellers can charge near the monopoly price.
Jeffrey R. Brown and Austan Goolsbee, “Does the Internet Make Markets More Competitive? Evidence from the Life Insurance Industry” (Journal of Political Economy 110, no. 3, 2002): 481–507.
Benjamin R. Tucker, “State Socialism and Anarchism: How Far They Agree, and Wherein They Differ” (1886). Socialism as the cost principle; interest, rent and profit. Copy here (post 360).
Pierre-Joseph Proudhon, What Is Property? (1840), ch. 4. The droit d’aubaine: rent, interest and profit as the owner’s right of increase.
Morris Bornstein, “The Soviet Price Reform Discussion” (Quarterly Journal of Economics 78, no. 1, 1964): 15–48. The Soviet debate over cost-plus markups.
Gary Chartier and Charles W. Johnson, eds., Markets Not Capitalism (Autonomedia, 2011). Socialist ends, market means; the senses of “capitalism.”
Frank H. Knight, Risk, Uncertainty and Profit (Houghton Mifflin, 1921). Profit as the reward for bearing uninsurable uncertainty.
Israel M. Kirzner, Competition and Entrepreneurship (University of Chicago Press, 1973). Profit from alertness.
Arthur C. Pigou, The Economics of Welfare (Macmillan, 1920). Costs a producer leaves on others, the factory smoke example.
Rebecca Diamond, Timothy McQuade and Franklin Qian, “The Effects of Rent Control Expansion on Tenants, Landlords, and Inequality: Evidence from San Francisco” (American Economic Review 109, no. 9, 2019): 3365–94. original
János Kornai, Economics of Shortage (North-Holland, 1980). Chronic shortage under price-controlled planning.
US Food and Drug Administration, “Generic Competition and Drug Prices: New Evidence Linking Greater Generic Competition and Lower Generic Drug Prices” (2019). Prices by number of generic makers.