What is value? Scholastics through Arrow-Debreu
For three centuries the answer was a real property of things — the labor and cost poured into them. Then it became a feeling at the margin, then a ranking you could read off a choice, then a vector of prices that only means anything relative to the whole economy at once. The story of how “value” lost its substance is the story of economics finding its method.
Als Debattengraph anzeigenHere is the whole thread the way the intellectual-lineage record holds it — one idea, the concept of value, traced across five eras from a 13th-century friar to a 1954 existence proof. The era-organized textbooks teach these figures in separate chapters; this is the single argument that runs through all of them. Click any node to see where it sits.
The spine the graph draws runs Aquinas → Smith → Ricardo → Marx → the 1871 marginalists → Marshall → Pareto, Hicks, Samuelson. It stops where the record stops, at Samuelson — the final rung, Arrow and Debreu’s 1954 proof, lives in this walkthrough’s prose, not yet in the graph. Open the full network in the intellectual-lineage graph.
The objective answer, and how far it could go
“Labour was the first price, the original purchase-money that was paid for all things. It was not by gold or by silver, but by labour, that all the wealth of the world was originally purchased.”
— Adam Smith, The Wealth of Nations, Book I, 1776
Here is the strange thing about that sentence. A century and a half before Smith wrote it, a group of theologians at the Spanish university of Salamanca had already located value somewhere else entirely — in common estimation, the price a thing fetches because of what buyers in a market are willing to give for it. Diego de Covarrubias wrote in 1554 that a good’s value depends “not on its nature but on the estimation of men, even if that estimation be foolish.” That is subjective value, three hundred years early. And then Smith reached straight past it for an objective anchor: labor, the real cost embodied in a good. The thread’s first move was not an advance. It was a reversal — and to see why a reversal was tempting, you have to see what the objective answer was reaching for.
The classical labor theory of value is a single, sturdy idea: the value of a good is governed by the quantity of labor needed to produce it. Smith called it “toil and trouble”; Ricardo sharpened it to the labor embodied directly and indirectly in a commodity. Around that core sits a simple dynamic. Natural price is the cost of production — wages, plus the going return on capital and rent. Market price oscillates above and below it as supply and demand wobble, but it is always pulled back toward the natural price as a center of gravity. Value, on this account, is not a mood. It is a fact about how much human effort a thing took.
Why reach for an objective anchor at all? Because the classical economists were not trying to explain why you bought one coat instead of two. They were trying to explain accumulation, distribution, and growth — how the surplus of a whole economy gets divided among workers, capitalists, and landlords, and whether that division can keep a society growing. For that you need a measuring rod that holds still while everything else moves: a unit of value independent of the very prices you are trying to explain. Labor looked like the one input common to everything humans make, so labor became the rod.
In the simplest classical form, the natural price of a good is the sum of its production costs — the wage bill plus the normal return on the capital advanced:
$$p_i = w\,\ell_i + (1+r)\,k_i$$where $\ell_i$ is labor and $k_i$ is capital used per unit of good $i$, $w$ the wage, and $r$ the rate of profit. The pure labor theory is the special case where the second term vanishes or is proportional to the first — value reduces to embodied labor $\ell_i$. The whole fight to come is about what happens when it doesn’t.
Think of value as something you could weigh on a scale. A table that took two days of work to make is “worth” twice a stool that took one day. The price you see in a shop bounces around with the season and the mood of buyers, but underneath it there is a real, solid number set by the labor that went in — and that solid number is what the price keeps returning to. The labor theory says: stop watching the bouncing price and look at the weight underneath.
That cost-of-production baseline is exactly the supply-side half of the modern apparatus — the demand side, which the classics underweighted, completes it. The intro-level home of the supply curve and cost-of-production logic is Economics Ch.2 §2.2 (Supply). For the lineage — how Smith founded the discipline on this measuring-rod project and where Ricardo drove it — see History of Economic Thought Ch.3 (Classical political economy); the subjective-value precursor the classics reached past is Ch.1 § The School of Salamanca.
The objective theory at full strength — and its apex in Marx
Take the labor theory at its strongest before laying a finger on it. It has three real virtues, and they are not nothing. Labor is universal: it is the one input that goes into everything humans produce, so it offers a common denominator across goods that have nothing else in common. Labor is measurable: hours are countable in a way that “utility” in 1817 was not. And labor connects price to the real economy of effort rather than the fictive economy of whim — it says that behind every price tag there is a human being who did something hard, and that the price ought to answer to that. For a century these were not the assumptions of cranks. They were the working foundation of the most serious economics there was.
“The value of a commodity… is determined by the quantity of labour socially necessary for its production. The surplus-value… is the difference between the value the labour-power creates and the value of the labour-power itself.”
— Karl Marx, Capital, Volume I, 1867
Marx is the apex of the objective tradition, not a departure from it — he takes the labor theory the classics built and pushes it until it becomes a theory of exploitation with a formal claim attached. The worker, he argues, sells not labor but labor-power, and sells it at its reproduction cost: what it takes to keep a worker alive and able to return tomorrow. But labor-power, once bought, produces more value than it costs. That gap — surplus value — is where profit comes from, and it is extracted, not earned. And the 19th-century facts cooperate with the indictment: the industrial worker really did produce far more value than the wage reflected; the bargaining-power asymmetry between a factory owner and a worker with nothing to sell but their hours was real; and capitalism really did generate recurring crises that no market-clearing account predicted or explained. Read in 1867, or re-read after 2008, the pull is genuine. The objective tradition, here, is doing the work of moral and political inspection that the prettier later theories would politely decline to do.
“Labor creates all value. The profit the boss takes is value the workers made and didn’t get paid for.”
— a recurring claim across left-of-center politics, online and on the picket line
Does labor create all value?
It is the oldest slogan in economics and the easiest to half-believe. The labor theory of value gives it a rigorous form — and the rigorous form is exactly where it runs into trouble. The moral intuition survives the technical defeat; the price theory does not.
Where the objective answer broke — and for substantive reasons
The labor theory works only under conditions the world does not supply. Heterogeneous labor breaks the measurement: an hour of surgery is not an hour of ditch-digging, and the only way to make them commensurable is to weight them by skill — using prices the theory was meant to explain. Capital-intensive production breaks the proportionality: once goods use different ratios of labor to machinery, their prices stop tracking their labor content, and Ricardo knew it — he spent his last years hunting an “invariable measure of value” he never found, and Smith had already carried two unreconciled accounts (labor-commanded versus labor-embodied) without noticing they came apart.
Marx pushed the theory to the precise point where it snaps: the transformation problem. If value is embodied labor but profit rates equalize across industries with different capital intensities, then labor-values cannot be converted into the prices that actually prevail without contradiction. Bortkiewicz showed the gap formally in 1907; Ian Steedman’s Marx After Sraffa (1977) closed the technical case — you can compute prices and profits straight from the physical input-output data, and the labor-value step is not just wrong, it is redundant. And here is the honest part of the verdict: the labor theory was not dismissed for being left-wing. It was superseded for substantive, internal reasons that left-wing economists themselves worked out. John Roemer’s analytical Marxism (1981) then showed the punchline — the moral claim of exploitation survives perfectly well without the labor theory of value. What carries forward is distributional class analysis: bargaining power, the capital share, monopsony, the political economy of who captures the surplus. All of it is formalizable inside the mainstream apparatus. The measuring rod was discarded; the question it was built to answer was not. The distribution thread (forthcoming) is where that surviving residue gets followed to Piketty and after.
If value is not the labor poured in, what is it? In a single year — 1871 — three economists working independently in three countries gave the same answer, and it was the answer the Salamanca theologians had glimpsed three hundred years earlier. The difference was that this time they put calculus underneath it, and a paradox that had stumped everyone since Aristotle dissolved on contact.
The marginal revolution: value moves to the margin
“Value depends entirely upon utility… we have only to trace out carefully the natural laws of the variation of utility, as depending upon the quantity of commodity in our possession.”
— William Stanley Jevons, The Theory of Political Economy, 1871
Jevons published that in England in 1871. The same year, in Vienna, Carl Menger published his Grundsätze, locating value in subjective want-satisfaction. Three years later, in Lausanne, Léon Walras published the first volume of his Éléments, turning the same idea into a system of equations for a whole economy. Three men, three countries, no coordination, one insight — arriving at almost the same moment. And the insight cracked open a paradox that had defeated every objective theory of value since Aristotle: water is essential to life and nearly free; diamonds are useless and dear. If value is utility, why is the useful thing cheap? If value is labor, why is the labor-cheap thing dear? Nobody could answer it — until you asked the question at the margin.
The move is small and total at once. Value is set not by the total usefulness of a good but by the usefulness of its last unit — its marginal utility — and marginal utility falls as you consume more. Now the paradox dissolves in a sentence. Water is abundant, so the last glass you consume does almost nothing for you, and its marginal utility — hence its value — is near zero, no matter how vital water is in total. Diamonds are scarce, so the marginal diamond is doing a great deal of work, and its value is high. Total usefulness was never the right variable. The margin was.
The optimizing consumer spreads spending so that the marginal utility per dollar is equal across all goods — the equimarginal condition:
$$\frac{MU_x}{p_x} = \frac{MU_y}{p_y} = \cdots = \lambda$$If one good gave more utility per dollar than another, you would shift spending toward it — lowering its marginal utility — until the ratios equalized. At the optimum, relative prices equal relative marginal utilities. Value has become a fact about the margin of exchange, not the substance of the good.
You are dying of thirst in a desert. The first glass of water is worth almost anything to you. The second is worth a lot. By the tenth you are sloshing, and an eleventh is worth nothing — you would not cross the road for it. Water did not change. Your situation did. “What is water worth?” has no answer; “what is one more glass worth, right now, to you?” has a precise one. That is the whole revolution: stop asking what things are worth in general and start asking what the last one is worth at the margin.
That marginal apparatus — preferences, marginal utility, the consumer’s optimum — is the foundation of modern microeconomics, taught today in Economics Ch.5 §5.1 (Preferences and Utility). For the lineage — the three simultaneous 1871–74 arrivals and what each saw — see History of Economic Thought Ch.5 (The marginalist revolution); Menger’s more radical, anti-mathematical subjectivism founds a distinct branch traced in Ch.6 (The Austrian tradition).
The cardinal machinery, argued at its strongest
The idea worth defending here is not the labor theory — that one is behind us. It is the early marginalists’ own cardinal conception of utility, the very thing the next rung will strip out. Jevons and Walras wrote as if utility were a measurable quantity, like temperature: a number of “utils” you could in principle add up, compare across people, and do arithmetic on. To a modern eye trained on ordinal preferences this looks like an embarrassing metaphysical commitment. It was nothing of the kind — in 1871 it was the right call. Cardinal utility made the calculus run cleanly: if utility is a smooth measurable function, you can differentiate it, set marginal utilities equal to prices, and get demand curves with a genuine microfoundation underneath them. It let Jevons write the equations of exchange and let Walras do something no one had ever done — write down a system of equations for an entire economy at once, every market linked to every other through prices.
“The theory of economics… must be mathematical simply because it deals with quantities. Wherever the things treated are capable of being greater or less, there the laws and relations must be mathematical in nature.”
— William Stanley Jevons, The Theory of Political Economy, preface, 1871
This is why no one rushed to abandon the cardinal scaffolding. It was not a mistake waiting to be corrected; it was the natural language of the revolution, the thing that let the new subjective theory become a quantitative science rather than a philosophical mood. Walras’s general-equilibrium dream — the whole economy as one solvable system — was only thinkable because he was willing to treat utility as a measurable magnitude he could write into equations. The cardinal machinery did real, load-bearing work. That it turned out to be carrying more weight than the result required is the discovery of the next generation, and it is no knock on the people who built the engine that they used a heavier frame than hindsight needed.
The loop the scholastics opened, finally closed
The marginal revolution closed a circle three centuries wide. Value is subjective, set at the margin of exchange — exactly what Salamanca’s “common estimation” had gestured at — but now with formal apparatus underneath it, demand curves derived rather than asserted, the diamond-water paradox resolved rather than waved away. This is the operative concept’s first mature form, and it is a genuine resolution of a genuine problem the objective tradition could not crack. But the victory rode in on machinery heavier than it needed. Nothing in a demand curve actually requires that utility be cardinal, summable, or comparable across people — a demand curve needs only that you can rank options, not that you can measure how much you like them. The cardinal “utils” are excess weight the theory is still carrying. That redundancy is the next rung’s target, and the surprise is that cutting it makes the theory stronger, not weaker.
Within fifty years economists would strip out everything about utility you could not observe — the summing, the comparing, the cardinal magnitude itself — and discover that the theory of value lost nothing and gained rigor. You don’t need to know how much someone wants coffee. You only need to watch them choose it.
The formalization: ordinal utility, revealed preference, and Marshall’s scissors
“We might as reasonably dispute whether it is the upper or the under blade of a pair of scissors that cuts a piece of paper, as whether value is governed by utility or cost of production.”
— Alfred Marshall, Principles of Economics, 1890
Marshall wrote that in 1890 to end a quarrel: the classics said value came from cost, the marginalists said it came from utility, and Marshall’s answer was that the question was malformed. Both blades cut. Demand (utility) and supply (cost) jointly determine value, and asking which one “really” sets the price is like asking which scissor-blade does the cutting. But the deeper move of this era was quieter and stranger. Starting with Pareto around 1900, running through Hicks and Allen in 1934 and arriving at Paul Samuelson’s revealed preference in 1938, economists asked a radical question: how much of the marginalist apparatus can you keep if you throw away everything about utility you cannot actually measure?
The answer turned out to be: all of it. You can demote utility from a cardinal magnitude to a mere ranking — an ordering that says “I prefer A to B” without ever claiming “by 4.7 utils” — and every result survives. Indifference curves need only that the consumer can rank bundles; the slope of an indifference curve, the marginal rate of substitution, is all the apparatus requires, and it is purely ordinal. Then Samuelson went one step further and removed even the introspection: revealed preference recovers the entire ranking from observed choices. If you bought A when B was affordable, you revealed that you prefer A — no peering into anyone’s head required. Value became a relation among rankings and prices, fully observable in behavior.
At the consumer’s optimum the marginal rate of substitution equals the price ratio — the ordinal condition that replaces the cardinal one:
$$MRS_{xy} = \frac{MU_x}{MU_y} = \frac{p_x}{p_y}$$and the weak axiom of revealed preference (WARP) ties it to behavior: if bundle $A$ is chosen when $B$ is affordable, then $B$ is never chosen when $A$ is affordable. No utility number appears anywhere — only choices and the prices that constrain them.
You don’t need to know how much someone likes coffee, or whether their liking is “more” than your liking of tea. You only need to watch the till. If, when coffee and tea cost the same, they reach for coffee, they have told you everything the theory needs — they prefer coffee at those prices. The whole machinery of value runs on choices you can see, not feelings you have to guess. The theory got lighter and saw further at the same time.
The ordinal consumer — preferences, indifference curves, the consumer’s problem — is taught in Economics Ch.5 §5.2 (The Consumer’s Problem), and revealed preference gets its rigorous treatment in Ch.11 §11.2 (Revealed Preference); the welfare reading of supply-and-demand — consumer and producer surplus, the thing Marshall’s scissors measure — is Ch.3 (Elasticity and welfare). The lineage of the ordinal turn — Pareto, Hicks, Samuelson, and the von Neumann-Morgenstern completion — lives in History of Economic Thought Ch.5 (Pareto, welfare, and the formalization).
Marshall’s scissors, argued at full strength
Give Marshall’s partial-equilibrium method its full due before the final rung sets it aside. Marshall analyzes one market at a time, holding everything else fixed: take the market for tea, draw supply and demand, find the price and quantity where they cross, and read off the welfare. It sounds like a simplification, and it is — and it is also the single most deployed apparatus in all of economics, then and now. One market at a time is tractable. It answers the questions economists actually get asked — what happens to the price of bread if the harvest fails, who bears a tax, how much surplus a tariff destroys — without requiring you to solve the entire economy first. Marshall kept a cardinal flavor too, in his consumer surplus, and that was a feature: it let him measure the gains from trade in a market in pounds and pence, which is exactly what a policy-maker needs.
This is the synthesis that worked — the competent, deployable theory of value that ran applied economics for a century. And there is a genuine, unresolved tension sitting at its edge, one the discipline reports rather than settles: Marshall’s partial equilibrium versus Walras’s general equilibrium, the practical tool versus the foundational claim. Marshall held everything else fixed because it let him get answers; Walras refused to, because he thought “everything else” was precisely what set the price. Both are right about different things. Partial equilibrium is the right working picture for one market; general equilibrium is the right foundation for the concept of value. The next rung is where holding-everything-else-fixed stops being a convenience and becomes the thing that has to give. The von Neumann-Morgenstern extension of this same preference apparatus into choice under uncertainty — value-as-preference becoming rationality-as-preference — is the spine of the rationality thread, which reads the very same ordering from a different angle.
The theory got stronger by throwing things away
The ordinal turn is a genuine economy of assumptions, and that is the surprising part: the theory of value became stronger by dropping the unobservable cardinal scaffolding the marginalists rode in on. Value is now a relation among rankings and prices, recoverable from choices you can watch — no introspection, no interpersonal utility comparisons, no metaphysics of “utils.” Marshall’s scissors remain the right working picture for any one market. But “one market, everything else fixed” cannot be the foundation of a value concept, because what “everything else” is doing is exactly what determines this market’s prices — the price of tea depends on the price of coffee, of sugar, of labor, of land, in an unbroken web. The relational idea, taken seriously, demands a system, not a partial view. It demands that you solve every market at once. And the moment you try, you hit the question that stopped Walras cold.
Walras had already seen the whole picture in 1874: value is not a property of a good, or even of a market, but of the entire system of markets solved simultaneously. The only problem was that he could not prove such a system has a solution at all — that prices exist at which every market clears at once. For eighty years the dream had no proof. In 1954, two mathematicians supplied it.
Arrow-Debreu: value as the equilibrium price vector
“The problem of the existence of a competitive equilibrium… The present paper gives a positive answer: under suitable assumptions, there exists a set of prices at which the demand for each commodity equals its supply.”
— Kenneth Arrow & Gérard Debreu, “Existence of an Equilibrium for a Competitive Economy,” Econometrica, 1954
That sentence is the intellectual mountaintop of the whole thread. Walras had written down the equations of a whole-economy equilibrium in 1874 but could only count equations against unknowns and hope — he had no proof that a solution existed. For eighty years the central claim of general equilibrium was an article of faith. In 1954, Kenneth Arrow and Gérard Debreu proved it rigorously, using a fixed-point theorem from topology: a complete-markets competitive economy does have a price vector at which every market clears simultaneously. Walras’s dream was vindicated, eighty years late, and the concept of value reached its purest form. Value is no longer a property a good carries around. It is the price vector that makes the entire system fit together.
In general equilibrium, value is the equilibrium price vector $p^*$ — the list of prices, one per good, at which, given everyone’s endowments and preferences and the available technology, the total amount each good demanded equals the total supplied, in every market at once. At those prices, each consumer is at their optimum (price equals marginal rate of substitution) and each producer is at theirs (price equals marginal cost), and all the plans of all the people are mutually consistent. Value has completed its journey from a substance to a relation: a good’s “value” is just its coordinate in $p^*$, and that coordinate means nothing on its own — it is defined only relative to the whole web of endowments, technology, and other prices that fix it.
Write $z(p)$ for the economy’s excess-demand vector (demand minus supply, good by good). An equilibrium price vector $p^*$ is one at which no market has positive excess demand:
$$z(p^*) \le 0, \qquad p^* \cdot z(p^*) = 0$$Arrow and Debreu prove such a $p^*$ exists by showing the price-adjustment map has a fixed point (via Kakutani’s theorem) — a price vector that maps to itself, i.e. one nobody wants to move away from. The math is the depth here, not the point: what matters is that existence is proved, not assumed.
Imagine everyone in the economy writing down their plans — what they’d buy and sell at every possible set of prices. For most price lists the plans clash: too many people want bread, nobody wants the labor to bake it. Value is whatever single set of prices makes all the plans fit together at once, with no leftover wants and no leftover goods. Arrow and Debreu’s achievement was proving that such a magic price list always exists — that the clashing plans of millions of strangers can, in principle, be made to mesh by prices alone, with no one in charge.
The Arrow-Debreu model as taught — Walrasian equilibrium, existence, and the two welfare theorems that bracket it — lives in Economics Ch.11 §11.5 (General Equilibrium: Walrasian Equilibrium). The lineage from Walras’s 1874 system through the 1954 existence proof — and the Cambridge capital controversy that marks where the apparatus stops — lives in History of Economic Thought Ch.5 § Walras: every market at once and the formalization section that follows. (The value-lineage graph above stops at Samuelson; this 1954 rung and the Cambridge controversy are carried here in prose, ahead of the timeline.)
The objective tradition returns — as the questions the proof cannot answer
Now the thread doubles back on itself. The tradition worth taking seriously again is the very objective one Stage 1 superseded — not its discarded answers, but its questions, which come back at full strength precisely where Arrow-Debreu falls silent. First: is there a real value beneath the price vector? The existence proof defines value only relative to a given endowment distribution — who owns what at the start. Change the endowments and every equilibrium price changes; the rich man’s yacht and the poor man’s bread trade at prices that encode the initial distribution of wealth. The price vector is exquisitely silent on whether that distribution is just. The old objective question — is there a value a thing ought to have, beneath the price it fetches? — is not refuted by the proof. It is placed outside the proof’s frame.
Second: what about value that markets don’t price? The Arrow-Debreu theorem holds only when markets are complete — when there is a market for everything, every good, every contingency, every period. Externalities, public goods, and unpriced ecological services are, by construction, the things for which markets are missing, and so they live exactly in the gap the theorem assumes away. The clean air a factory fouls, the climate the carbon loads, the knowledge a discovery spills — the model does not say these have zero value; it says it has nothing to say about them, because the markets that would price them do not exist. The externalities-and-public-goods apparatus that economics built to handle this is in Economics Ch.4 (Market failures) — a whole chapter dedicated to the cases the existence proof excludes.
Third, the live descendants. These are not museum questions. The socialist-calculation debate — Mises and Hayek arguing that without market prices a planner cannot compute value at all, against Lange and Lerner answering that a planner could simulate the equilibrium — is a direct fight over whether the price vector contains information that nothing else can reconstruct. And ecological economics presses the second question hardest: it argues that the most important values — a stable climate, biodiversity, a habitable planet — are systematically the ones markets are worst at registering, and that a value concept silent on the unpriced is dangerously incomplete as a guide to what a society should do. The objective tradition’s questions did not die at the marginal revolution. They went underground and surfaced again, inside the mainstream, as its hardest open problems. The Austrian calculation argument is traced in History of Economic Thought Ch.6 § Mises and the calculation problem; the post-2008 heterodox frontier where these objective-value questions resurface is in Ch.17 (Modern pluralism).
None of this is a refutation of Arrow-Debreu, and saying so honestly is the point. The existence proof is exactly what it claims to be: a demonstration that decentralized price coordination can be coherent, that the clashing plans of millions of strangers can in principle mesh through prices with no one in charge. That is a profound and durable result. What it is not is a description of any real economy, and it never pretended to be. The endowments it takes as given, the complete markets it assumes, the externalities it excludes — these are the boundary of the frame, drawn in the open. The relational concept of value is the right answer to what value is in a market economy. The objective tradition’s surviving questions are about what value markets fail to register — a different question the proof was never built to answer, and the place the old tradition still does its work.
“The market knows the price of everything and the value of nothing.” / “If it mattered, there’d be a market for it.”
— the two slogans the surviving objective questions get compressed into
Do markets price everything, or nothing that matters?
The two slogans are mirror-image overstatements, and the truth between them is exactly the gap Arrow-Debreu drew in the open: markets price what has a market, and the most important things often don’t.
Where the thread landed — and the limit it found
One verdict has to be delivered straight, because it is the place the apparatus genuinely meets its edge: the Cambridge capital controversy. Through the 1950s and 60s, Cambridge UK — Joan Robinson, Piero Sraffa, Pierangelo Garegnani — argued that you cannot measure aggregate “capital” as a quantity independent of the prices and interest rate it is supposed to help explain, because the value of a stock of machines depends on the very profit rate the theory uses capital to determine. The phenomenon of reswitching — the same technique becoming optimal, then suboptimal, then optimal again as the interest rate falls — showed the aggregate production function can misbehave. And here is the honest verdict: Cambridge UK won the technical points. Paul Samuelson, the leading figure on the other side, conceded it in 1966 in plain terms. The mainstream did not restructure anyway — not out of stubbornness, but because Sraffa’s positive program explained less, not more, and the aggregation problem bites only at the level of economy-wide capital aggregates, while most of modern economics lives at the individual-market level where the apparatus is undisturbed. The limit is real and it is precise: it marks where the apparatus stops, not that it is wrong.
So where did the thread land? Value is now a relational-equilibrium quantity — the price vector that clears a complete-markets economy — and that is the operative apparatus of modern economics. It won, and it deserves to have won: it resolved the value paradox the objective theories choked on, and unlike the labor theory it does not break on heterogeneous labor, capital intensity, or demand. But it won by being more productive — by explaining and predicting more — not by being unfalsified or by answering every question worth asking. The objective tradition’s questions survive as honest critique inside the mainstream itself: general equilibrium’s silence on the justice of endowments and on unpriced value is exactly where externality theory, public-goods theory, the socialist-calculation debate, and ecological economics do their work. The concept of value traveled from a moral property of things, through a subjective magnitude at the margin, to a pure relation the whole system fixes — and at each step it answered its predecessor’s unanswered problem and bequeathed a new one. The thread did not end the value question. It relocated it — out of the substance of goods and into the structure of markets, and then, at the edges, back out again into the questions markets cannot reach.
One thread, four rungs
Trace the whole arc and the concept of value loses its substance one layer at a time. Objective value: value is a real property of a good, the labor and cost embodied in it — Smith’s measuring rod, Ricardo’s embodied labor, Marx’s surplus value pushed to the transformation problem, where the theory of price snapped while the theory of distribution survived. Marginal value: value is a subjective magnitude at the margin of utility — the 1871 revolution that dissolved the diamond-water paradox and vindicated what Salamanca had glimpsed, carrying a cardinal-utility scaffolding it did not need. Ordinal/relational value: value is a relation among rankings and prices, recoverable from choices alone — the formalization that made the theory stronger by throwing away everything unobservable, with Marshall’s scissors as the working tool. Equilibrium-relational value: value is the price vector that clears the whole system at once — Walras’s 1874 dream, proved to exist by Arrow and Debreu in 1954, value defined only relative to the entire web of endowments, technology, and markets.
This is not a progress march where each step is simply better than the last. Each rung solved a real problem its predecessor created, and each created the problem the next would answer. And the landing is calibrated, not triumphal. The relational-equilibrium concept won as the operative apparatus — it is more productive, it does not break where the labor theory broke, and it is what modern economics actually runs on. But the objective tradition’s questions outlived its answers. “Is there a real value beneath the price vector?” and “what about value markets don’t price?” were not refuted by Arrow-Debreu; they were placed outside its frame, and they return — inside the mainstream — as externalities, public goods, the socialist-calculation debate, and ecological economics. The relational concept answers what value is in a market economy; it is silent on what value markets fail to register, and that silence is where the oldest questions in economics still live.
A closing note on universality. In the 14th century, outside the entire Salamanca-to-Smith chain of influence, Ibn Khaldun wrote in the Muqaddimah that “the value of labor” is realized in the value of goods — a labor-value intuition reached independently, centuries early, in a different civilization. The value question is universal; people everywhere have asked what things are really worth. What was particular — what only became possible inside the specific Salamanca → Smith → Marx → marginalist → Arrow-Debreu lineage — was the formal answer, the apparatus that turned the question into a solvable system. This value thread is one of several that branch off the same Ricardo-to-marginalist-to-equilibrium spine: the distribution thread follows the surplus where Marx left it, toward Piketty and modern inequality; the rationality thread follows the preference ordering into choice under uncertainty and the behavioral break; the market-structure thread reads the same Arrow-Debreu model not as value but as how markets coordinate and can be designed.