The most argued comparison in this entire subject is conducted almost entirely by people who have not noticed that the two frameworks are answering different questions. One side declares that the marginal revolution refuted Marx. The other declares that marginalism describes appearances while the labour theory explains essences. Neither statement can be assessed, because neither side has specified the criterion on which the comparison is being made, and until a criterion is named the contest has no scoring system.

This article names the criterion, applies it across seven dimensions, identifies the narrow territory where the two frameworks genuinely compete for the same explanatory job, and delivers a verdict on that territory rather than a verdict on everything. It also disposes of a chronological error so widespread that it appears in university lecture notes: Marx did not reject marginal utility theory, because the first volume of Capital appeared in eighteen sixty-seven and the founding marginalist texts appeared four and seven years later. Whatever the labour theory of value is, it is not a response to marginalism, and the passages people cite as Marx’s dismissal of subjective valuation are directed at earlier writers.

The reader who finishes should be able to say precisely what each framework was built to explain, name the one question both of them answer, state which does that job better and on what evidence, and recognise the third framework that neither camp mentions because it embarrasses both. That last item is the reason most comparisons of this kind are unreliable: they present a two-way contest in a field that has had at least three serious contenders since the middle of the twentieth century.

The two positions, stated as their own defenders would state them

Comparison requires that both positions be recognisable to the people who hold them, which rules out most of the summaries in circulation.

The Marxian position, in its own terms, is that the value of a reproducible commodity is determined by the labour time socially necessary to produce it under prevailing conditions, and that this magnitude regulates exchange ratios not by fixing them but by acting as the centre around which competition drives them. Value is not a property a thing has by itself; it is a social relation that appears as a property of a thing, which is why the theory needs a separate account of the value form and of money. The framework’s object is a specific historical society in which production is private and dispersed while the products are socially interdependent, and its central question is how the total labour of such a society gets distributed among branches of production when nobody distributes it. The full statement belongs to the labour theory of value explained, which owns the definitional question in this series.

The marginalist position, in its own terms, is that the exchange ratio between two goods reflects the rate at which agents are willing to substitute one for the other at the margin, given their preferences and their constraints, and that in equilibrium this rate is equalised across all agents and equals the ratio of marginal costs of production. The framework’s object is the allocation of scarce means among competing ends, and its central question is what pattern of production and exchange emerges when agents optimise subject to constraint. Value in the sense of a substance is not part of the framework at all; the marginalist claim is that no such substance is needed, because the ratios can be derived from the interaction of preferences, endowments, and technology without positing anything behind them.

Two things follow immediately from putting the positions side by side properly. The first is that they do not disagree about a shared proposition; they take different objects. The second is that one of them denies the existence of the thing the other is theorising, which is not a disagreement about magnitude but about ontology, and disagreements of that kind are not settled by producing a better fit to price data.

What is the difference between the labour theory of value and the subjective theory of value?

The labour theory explains how a society allocates its total labour among branches of production without deciding to, and treats value as a social relation appearing as a property of things. The marginalist framework explains how given resources are allocated among competing uses by optimising agents, and denies that any substance underlies exchange ratios.

The chronology that disposes of a whole class of arguments

The founding marginalist texts appeared in the early eighteen seventies: William Stanley Jevons and Carl Menger published independently in eighteen seventy-one, and Leon Walras published the first part of his system three years later. The first volume of Capital had appeared in eighteen sixty-seven.

This ordering has consequences that are routinely ignored. Marx could not have been responding to marginal utility theory in the text where he sets out the value theory, because it did not exist. He lived until eighteen eighty-three and there is no engagement with the new framework in the published economic work, which is unsurprising given the state of his health and the condition of the manuscripts that became the second and third volumes. Anyone who writes that Marx dismissed marginal utility, or that he answered the subjectivists, is projecting a later argument backwards.

The corollary cuts the other way with equal force. The marginalists were not responding to Marx either. Jevons and Menger were addressing the classical tradition, principally Ricardo and John Stuart Mill, and the problems they identified with it were problems of that tradition rather than of Marx’s specific system. The idea that the marginal revolution was a reaction against socialism has been argued and is not supported by the sequence: the founding works engage classical political economy on technical grounds, and their authors did not have the third volume of Capital, which did not appear until eighteen ninety-four.

What did happen is that the two frameworks were brought into contact retrospectively, by a third generation. Bohm-Bawerk’s critique of eighteen ninety-six, which is the first substantial engagement from the marginalist side, was written after the third volume made the value-price relation explicit. The framing that treats the two theories as rival answers to one question is therefore not the framing either founder used. It is a later construction, and like most later constructions it flattens the differences that matter.

The paradox of value, and why only one framework needed to solve it

The shared inheritance is the puzzle Adam Smith set out early in the Wealth of Nations: water is essential and cheap, diamonds are useless and dear, so usefulness cannot be what determines exchange ratios. Smith’s own response was to separate value in use from value in exchange and to explain the latter by labour, which leaves the puzzle formally solved and substantively untouched, because nothing explains why usefulness has no role.

The marginalist solution is elegant and it is the framework’s founding achievement. Usefulness does determine exchange ratios, but the relevant magnitude is marginal utility rather than total utility: the value placed on a further unit, given the units already held. Water is abundant, so the last unit consumed satisfies a trivial want; diamonds are scarce, so the last unit satisfies an intense one. The paradox dissolves, and the dissolution generates a general apparatus for handling scarcity, substitution, and choice that turned out to be enormously productive across the whole of economics.

The Marxian response is different in kind and is usually misunderstood. Marx does not solve the paradox; he denies that it is a paradox by distinguishing use value from value at the outset and specifying their relation. Use value is a precondition of exchange, since nothing useless is bought, but it is not a measure of anything, because use values are qualitatively different and cannot be compared quantitatively. Water and diamonds satisfy incommensurable wants and there is no scale on which their usefulness can be ranked against each other. What can be compared quantitatively is the labour required to obtain them, and that is what the theory measures. The distinction and its consequences are worked out in use value, exchange value, and value, which owns the terminological question.

The instructive point is that the same puzzle produces a solution in one framework and a dissolution in the other, and the difference tells you exactly what each framework is doing. Marginalism treats the puzzle as a question about why agents pay what they pay, and answers it. The labour theory treats it as a question that arises only if you look for a measure in the wrong place, and redirects. Neither response is a refutation of the other, and treating the diamond and water case as a decisive argument against Marx, which happens constantly, mistakes a difference of question for a defeat.

The seven-dimension comparison

The table below is the article’s findable artifact. It compares the frameworks on seven dimensions rather than declaring a winner, and the final column is the one most comparisons omit: what would count as evidence against each.

Dimension	Labour theory of value	Marginalist framework
What it sets out to explain	How a society whose production is private allocates its total labour among branches, and what social form that allocation takes	How given resources are allocated among competing uses by agents optimising under constraint
Unit of analysis	The commodity as a social product, and the total social capital	The individual agent’s choice at the margin, aggregated into markets
Status of value	A social relation that appears as a property of things and requires money as its form of expression	Not a substance at all; exchange ratios are derived from preferences, endowments, and technology
Objective or subjective determination	Objective in the sense of independent of any individual’s valuation, being a magnitude of social labour	Subjective in the sense of grounded in preferences, though constrained by scarcity and technology
Treatment of distribution	Determined by the class relation between owners of means of production and sellers of labour power, prior to price formation	Determined by marginal products and factor scarcities within the same system that determines prices
Treatment of time and change	Historical and dynamic: technical change, accumulation, and crisis are internal to the framework	Comparative static in its core, with dynamics added as extensions rather than built in
What would count as evidence against it	Failure of aggregate identities under every consistent valuation scheme, or labour requirements tracking prices no better than arbitrary alternative inputs	Systematic and persistent violation of consistency conditions in choice, or the failure of equilibrium reasoning to predict adjustment

The table repays a slow reading of the first row, because everything else follows from it. Two frameworks with different explananda can both be internally successful, and the question of which is right does not arise until they are shown to be competing for the same job.

The different-question rule

The rule this article advances is the different-question rule: the labour theory explains how a society reproduces itself and distributes its surplus, and the marginal framework explains how given resources are allocated among competing uses, so a refutation of one by the other must first demonstrate that they are answering the same question, which almost no published refutation attempts.

Applied to the standard arguments, the rule disposes of most of them in a sentence each. The diamond and water case shows that marginal analysis explains a phenomenon the labour theory does not address; it does not show that the labour theory is wrong about what it does address. The observation that prices respond to demand shows that marginal analysis captures short-run movements the labour theory treats as oscillation; it does not touch the claim about what the oscillation is around. The observation that the labour theory cannot predict relative prices accurately shows that it performs badly at a job marginalism does well; it does not show that the social-form claim is false. And the Marxian retort that marginalism cannot explain the origin of profit shows that the marginal framework does not treat that as its question; it does not show that the framework fails on its own terms.

The rule has a sharper application that both camps dislike. If the frameworks answer different questions, then a great deal of what each camp says about the other is not merely wrong but unintelligible, and the intellectual energy spent on the contest has been largely wasted. That is an uncomfortable conclusion for anyone whose position depends on having won, and it explains why the rule is resisted from both directions.

Where the two frameworks genuinely compete

The rule is not a peace treaty. There is a real overlap, and identifying it is the only way to reach a verdict that means anything.

The overlap is the explanation of relative prices of reproducible commodities in the long run. Both frameworks make claims here. The labour theory says the regulating centre is determined by conditions of production, specifically by socially necessary labour time, modified by the equalisation of profit rates across capitals of different composition. The marginalist framework says relative prices in long-run competitive equilibrium equal ratios of marginal costs, which are themselves determined jointly with factor prices by the interaction of preferences, technology, and endowments. These are rival accounts of the same object and they can be assessed against each other.

A second and narrower overlap concerns the direction of causation between distribution and prices. The Marxian framework treats the wage as determined by the class relation and by the value of labour power, prior to and independently of the price system, and treats profit as the residual. The marginalist framework determines wages and profits within the same system as prices, as returns to factors according to marginal contribution. This is a genuine disagreement about structure, and it was the substance of the Cambridge capital controversies of the nineteen fifties and sixties, in which the coherence of an aggregate capital measure independent of distribution was successfully challenged. Reporting that dispute accurately matters: what was established was a problem for a particular aggregate formulation of marginal productivity theory, not a vindication of the labour theory of value, and the framework that gained most from it was neither of the two under comparison here.

A third overlap is often claimed and is not real. Both frameworks are said to explain the origin of profit. The marginal framework explains the return to capital as the marginal product of capital combined with time preference or waiting; the Marxian framework explains profit as appropriated surplus labour. These look like rival answers, and the reason they are not quite is that the two accounts define the question differently: one asks why a factor receives a return in equilibrium, the other asks where the aggregate surplus comes from before its distribution among claimants. A serious critic can insist that the two questions collapse into one, and that insistence is the strongest form of the objection this article has to answer.

The founders were not one school

Treating marginalism as a single doctrine is the mirror image of treating Marxism as one, and it produces the same distortions. The three founding works differ in method, in mathematical apparatus, and in what they take the theory to be about, and the differences generated two traditions that have been arguing with each other ever since.

Jevons built a theory of exchange on measurable utility, expressed in calculus, in which the ratio of exchange between two goods equals the inverse ratio of their final degrees of utility. His ambition was to make economics a mathematical science on the model of mechanics, and his treatment of labour is more nuanced than the caricature allows: he holds that cost of production determines supply, supply determines the final degree of utility, and the final degree of utility determines value, which is a chain rather than a flat denial that production conditions matter.

Menger built something different. His treatment is causal and verbal rather than mathematical, and he refused the calculus throughout his life. His central concepts are the imputation of value from consumption goods back to the goods of higher order used to produce them, and the role of knowledge, uncertainty, and time in economic action. Value in his account is not a measurable quantity of satisfaction but a judgement an agent makes about the significance of a particular good for their purposes. The Austrian tradition descends from this line, and its later hostility to equilibrium modelling and to aggregate reasoning is already implicit in the founding text.

Walras built the third thing: a system of simultaneous equations describing general equilibrium across all markets at once, with utility maximisation as the behavioural assumption and the equality of supply and demand in every market as the condition to be satisfied. The mainstream of twentieth-century price theory descends from this line, through the formalisation of general equilibrium in the middle of the century.

Three consequences follow for the comparison with Marx. First, the objection that marginalism ignores production is true of a caricature and false of all three founders, each of whom has an account of how conditions of production enter. Second, an argument against Austrian economics is not an argument against neoclassical price theory and vice versa, since the two traditions disagree about method, about equilibrium, about mathematics, and about whether aggregate concepts are meaningful at all. Third, the framing of a single subjective theory of value confronting a single labour theory of value is a construction that flattens two internally diverse traditions into a pair of slogans.

There is a further point worth making about Menger specifically, because it bears directly on the comparison. His theory of imputation faces the problem that a good of higher order is normally used to produce several different consumption goods, so the value imputed to it is not uniquely determined by any single one. This is the same class of difficulty as joint production in the classical tradition, and it was worked on for decades. Every framework has an unresolved technical core, and citing the opponent’s unresolved core as if it were unique to them is the standard move in this argument on both sides.

What each framework says about one concrete case

Abstract comparison persuades nobody. The most useful thing a comparison can do is run both frameworks over a single case and show where the accounts diverge, where they converge, and where one of them simply has nothing to say.

Take a firm in a competitive industry that installs a machine which halves the labour required per unit of output, while the industry’s other firms continue with the old technique.

The marginalist account runs as follows. The innovating firm’s marginal cost falls. At the prevailing market price, its profit-maximising output expands, since it now equates marginal cost to price at a higher quantity. Its economic profit rises above the normal return. That profit attracts imitation: other firms adopt the technique or new entrants appear, industry supply expands, and the market price falls until economic profit is competed away and price again equals the new, lower marginal cost. Consumers gain, the initial innovator earns a temporary rent, and the long-run outcome is a lower price reflecting the improved technique. Distribution enters through factor markets: labour released by the innovation is reallocated to wherever its marginal product is highest, and the wage adjusts accordingly.

The Marxian account runs alongside it and agrees about a surprising amount. The innovating firm produces at an individual value below the social value of the commodity, sells at or near the prevailing price, and captures a surplus profit, which Marx analyses explicitly as the mechanism driving the adoption of improved technique. As the technique generalises, the socially necessary labour time falls, the social value of the commodity falls, the surplus profit disappears, and the price settles at a lower level corresponding to the new conditions of production. That is the same sequence, and a reader encountering both accounts for the first time is entitled to ask what the disagreement is.

The disagreement appears when the questions are extended. The Marxian account adds that the cheapening of the commodity, if it enters the consumption of workers, reduces the value of labour power and therefore raises the rate of surplus value even with no change in the length or intensity of the working day, which is the mechanism of relative surplus value and which has no counterpart in the marginalist story because the marginalist story has no category for the value of labour power. It adds that the displaced workers form part of a reserve that exerts downward pressure on wages, whereas the marginalist account treats reallocation as an adjustment toward a new equilibrium rather than as a mechanism of class discipline. And it adds a claim about the aggregate: as this process repeats across the economy, the ratio of constant to variable capital tends to rise, which is where the crisis theory begins and where the marginalist framework has nothing corresponding at all.

The convergence in the first part of the case and the divergence in the second is the clearest single illustration of the different-question rule. On the question of what happens to prices and outputs after a cost-reducing innovation, the two accounts describe the same process in different vocabularies and reach compatible conclusions. On the questions of what the innovation does to the distribution between classes, to the position of displaced workers, and to the aggregate composition of capital over time, only one framework treats these as questions its apparatus is built to answer.

A critic can respond that the marginalist framework does have accounts of unemployment, of bargaining power, and of technical change embodied in growth models, and this is fair. The accurate statement is not that mainstream economics ignores these things but that it treats them as separate problems requiring separate models, whereas the Marxian framework generates them from the same apparatus that handles the price movement. Whether that unification is an achievement or a source of error is a genuine methodological disagreement, and it belongs to the wider comparison rather than to the value question.

The vocabulary that means different things in each framework

Most of the confusion in this argument is terminological, and the terminological problem is not that the words are hard but that they are shared. The same word carries different definitions in each framework, and a sentence that is true under one definition is false under the other.

Term	In the Marxian framework	In the marginalist framework
Value	A social relation, the magnitude of socially necessary labour, distinct from price and from usefulness	Usually a loose synonym for price or for the worth an agent places on a unit; no separate substance
Capital	A social relation of production, self-expanding value, not the physical means of production as such	A stock of produced means of production, or funds available for investment; a factor of production
Cost	The value of the inputs consumed, including the value of labour power	Opportunity cost, the value of the best alternative forgone, which may involve no outlay at all
Profit	The money form of appropriated surplus value, prior to its division into interest, rent, and enterprise profit	The return to the entrepreneur or to capital, decomposed into normal return and economic profit
Productivity of labour	Output per unit of labour time, a technical relation affecting the value of each unit	Marginal product of labour, the addition to output from an additional unit, which grounds the wage
Competition	The process that equalises the profit rate and forms prices of production, driving technical change	A market structure characterised by price-taking, many agents, and free entry
Demand	A condition of value realisation; effective demand backed by purchasing power	A schedule relating quantity to price, derived from constrained optimisation

The table is worth using actively rather than reading once. When an argument on this subject stalls, the cause is usually that one of these seven words is being used in both senses within a single exchange. The most damaging case is capital, because the Marxian claim that capital is a social relation and not a thing sounds like mysticism to someone using the factor-of-production definition, and the marginalist question about the marginal product of capital sounds like a category error to someone using the social-relation definition. Both reactions are correct given the respective definitions and neither is an argument.

The second most damaging case is productivity. A statement that wages should track productivity means one thing where productivity is the marginal product grounding factor payment, and something quite different where productivity is a technical ratio whose rise, other things equal, lowers the value of labour power and raises the rate of exploitation. The same empirical observation about a divergence between wage growth and output per hour is therefore read as a puzzle requiring explanation in one framework and as an expected result in the other.

The verdict on the overlap

On the narrow territory where the frameworks compete, the verdict has to be given separately for each of the two real overlaps, and the deciding factor has to be named.

On the explanation of long-run relative prices of reproducible commodities, the marginalist framework performs the job better, and the deciding factor is that it generates determinate predictions from observable data without requiring a magnitude that has to be computed under contested assumptions. The labour theory’s own mature statement concedes that prices of production diverge systematically from values, the transformation from one to the other cannot preserve both aggregate identities under simultaneous valuation, and the Sraffian result shows that the price vector and the profit rate can be determined without value magnitudes at all. A defender who wants to contest this verdict has to either reject simultaneous valuation, which is a live position, or relocate the theory’s claim away from price determination, which is the more common move and which concedes the overlap rather than winning it. The objections in full, with the replies stated at their strongest, are in criticisms of the labour theory of value, which owns that survey.

On the structure of distribution, the verdict runs the other way, and the deciding factor is the aggregation problem. The Marxian framework’s treatment of the wage as determined by a social relation prior to price formation is not vulnerable to the difficulties that the capital controversies exposed in aggregate marginal productivity accounts, because it does not require a measure of capital independent of distribution. This is a narrower victory than Marxian economists usually claim from it, since the critique was mounted by economists working in a classical rather than a Marxian framework, and since the disaggregated general equilibrium formulations do not depend on the aggregate capital measure at all. It is nevertheless a real point on a real overlap.

The overall verdict, therefore, is a split with a specified boundary rather than a scoreboard. If the question is how relative prices of reproducible goods are determined, the marginal framework wins. If the question is what determines the division of the product between wages and profits, the Marxian framework has the better structural account, and the marginalist alternative faces a genuine difficulty in its aggregate form. If the question is how a market society allocates social labour without deciding to, only one framework treats it as a question at all, and a framework that does not ask a question cannot be said to have lost it.

The third option neither camp mentions

The two-way framing survives because it suits both sides, and it conceals the most important development in this area since the eighteen seventies.

The classical surplus approach, of which Marx’s economics is a branch, did not die when the profession reorganised around marginal analysis. It was reconstructed in a non-Marxist form by Piero Sraffa in a book of nineteen sixty, which showed that given the technical conditions of production and one distributional variable, relative prices and the other distributional variable are determined simultaneously, with no appeal either to labour values or to marginal utility. That framework is the modern alternative to marginalism in price theory. It is not Marx, and it was the source of the most damaging technical criticism ever made of Marxian value theory, because it demonstrated that the classical results could be obtained without the value apparatus.

This matters for the comparison in two ways. First, a reader who concludes that rejecting marginalism commits them to the labour theory of value has been presented with a false alternative. Second, a reader who concludes that the labour theory’s difficulties vindicate marginalism has also been presented with a false alternative, since the classical revival inherits the surplus approach’s structural account of distribution while dropping the value magnitudes. The exchange between the two is worked through in Sraffa and Marxist economics.

The Austrian tradition constitutes a fourth position, and it is not the same as the mainstream marginalist one despite sharing the subjective starting point. The Austrian objection to Marx runs through the impossibility of economic calculation without market prices as much as through value theory proper, and the argument is not primarily about what determines exchange ratios but about what prices do informationally. The full argument belongs to Mises and Hayek against Marx rather than here, and conflating it with the neoclassical position is one of the more common errors in popular treatments, since Austrians reject the equilibrium apparatus that mainstream marginalism is built on.

What modern price theory actually uses

A comparison framed as Marx against marginalism invites the assumption that contemporary economics is Jevonsian. It is not, and the difference matters for anyone deciding which framework to invoke.

Cardinal utility, the measurable quantity of satisfaction that Jevons theorised, was largely abandoned in the nineteen thirties, when the demand theory was reconstructed on ordinal preferences and rates of substitution, dispensing with the need for any measurable utility magnitude. The revealed preference approach went further, deriving demand relations from observed choices without positing preferences as psychological entities at all. General equilibrium theory in its mature form derives prices from preferences, technology, and endowments without any of the utility-substance language that the popular presentation of the subjective theory of value relies on.

Two consequences follow. A critic of Marx who says value is subjective, meaning that value is determined by how much people want things, is stating a version of the framework that mainstream economics itself moved away from, and is exposed to the objection that preferences are formed rather than given. And a Marxist who attacks marginal utility theory as psychologistic is attacking a position the discipline vacated three generations ago, which is the mirror image of the anachronism about Marx and the marginalists with which this article began. The wider methodological comparison between the two research programmes, as opposed to the value-theoretic one handled here, is treated in Marxist versus mainstream economics.

Which theory of value do economists actually use?

Neither in the form the popular argument presents. Mainstream price theory works from ordinal preferences and revealed choice rather than measurable utility, and derives prices from preferences, technology, and endowments. Labour values appear in heterodox research programmes and in the empirical input-output literature rather than in standard price theory.

A second worked case: a general rise in wages

The innovation case showed convergence in the short run and divergence in the long. A rise in the general wage level shows the opposite pattern, and running it is the fastest way to see why the distributional overlap is the sharper of the two.

The marginalist account, in its competitive form, treats a general wage rise as either a movement along a demand schedule, in which case employment falls as firms substitute toward other inputs and reduce output, or as a reflection of increased labour productivity, in which case employment need not fall because the marginal product has risen. Which of the two applies is an empirical matter about the source of the increase. Relative prices change according to how labour-intensive each industry is, since a wage rise increases costs more in labour-intensive branches, and the pattern of price changes follows from the pattern of factor intensities.

The Marxian account agrees about the direction of the relative price changes and derives them differently. A rise in wages reduces the rate of surplus value at given productivity, and since prices of production incorporate the general rate of profit, a fall in that rate changes the prices of production in a determinate pattern: commodities produced with a higher ratio of constant to variable capital become relatively cheaper, and labour-intensive commodities become relatively dearer. This is the same qualitative result as the marginalist account produces from factor intensities, and the convergence is not an accident, since both are tracking the same underlying technical relations.

The divergence is about causation and about what is being explained. In the marginal framework, the wage is determined within the price system by the interaction of labour supply and labour demand, and a general wage rise is something that requires an explanation in terms of shifts in those schedules. In the Marxian framework, the wage is determined by the value of labour power and by the balance of class forces, both of which sit outside the price system, and the effect on prices is a consequence rather than a puzzle. Neither framework can be evaluated on this case by looking at the price movements, because they agree about those. The test, if there is one, concerns whether the wage is better explained as a factor price or as the outcome of a social relation, and that is not a question about value theory at all.

This is the finding that most repays carrying away from the comparison. The two frameworks agree about more of the observable price behaviour than either camp admits, and they disagree about the direction of explanation, about what is taken as given, and about which magnitudes are determined inside the system. Disagreements of that kind are real, but they are not settled by pointing at prices.

Does this comparison matter outside economics?

It does, in three places, and each is a reason a non-economist might need the distinction clearly drawn.

In political philosophy, the labour theory of value is often confused with a labour theory of property, the claim that mixing one’s labour with something generates a rightful claim to it. That claim descends from Locke and it is not Marx’s. Marx does not argue that workers deserve the full product because their labour created it, and he attacks the socialist writers who did argue that, on the ground that the wage relation involves no violation of exchange and that a demand for exchange at labour values is a demand for the market to work properly rather than for it to be abolished. A reader who imports the desert argument into Marx will make errors in both directions, treating the value theory as a moral premise it is not and treating its refutation as a moral defeat it would not be.

In sociology and cultural analysis, the vocabulary of value has spread far beyond price, and the phrase creating value is now used for activities with no obvious relation to either framework. The distinction worth keeping is between an account of what determines exchange ratios and an account of what people find worthwhile, since the two frameworks under comparison are both about the former and neither is a theory of significance in the broader sense. Writing that treats the marginalist framework as a philosophy of subjective worth is making the same slide in the opposite direction.

In the analysis of contemporary work, the distinction determines what questions get asked. If value originates in labour, then unpaid or unrecognised labour is a candidate source of value that the accounts fail to register, and the analytical task is to trace where it goes. If exchange ratios reflect marginal valuations under constraint, then unpaid activity is not a source of value but a cost borne privately, and the analytical task is different. Neither framing is neutral about what counts as a finding, which is a reason to be explicit about which one is in use rather than sliding between them.

How the two-way framing was built

The contest as it is now conducted is a construction with a traceable history, and knowing that history explains why the framing is so resistant to correction.

The first stage was polemical. The engagement between the marginalist and Marxian frameworks began not with the founders but with the generation that followed, when the third volume of Capital made the value-price relation explicit and provided a target. The exchange that resulted set the pattern: a critique concentrating on internal consistency, a reply concentrating on levels of abstraction, and no agreement on what would count as settling the matter.

The second stage was pedagogical. As economics professionalised, the labour theory of value entered the curriculum as a historical stage superseded by the marginal framework, presented in a compressed form suitable for a chapter in a history of thought course. That compression produced the version of the theory most critics attack, in which prices are proportional to labour times, which is the version the third volume denies. Meanwhile, on the other side, socialist education presented marginalism as bourgeois apologetics designed to obscure exploitation, which produced the version most Marxists attack, in which value is whatever anyone happens to feel like paying. Two caricatures were installed in two curricula, and each has been reproduced ever since by people who learned the subject from them.

The third stage was political. During the decades when the world was divided between systems claiming these two intellectual inheritances, the value question acquired a significance out of all proportion to its analytical content, and defending or attacking a theory of exchange ratios became a way of defending or attacking a social order. That association has outlasted the circumstances that produced it.

The fourth stage is the one shaping the argument now, and it is driven by the economics of online publishing. The query rewards a confident verdict, so pages that deliver one accumulate links and attention, while pages that specify criteria and reach a split verdict do not. The result is a body of writing in which both camps are represented by their most assertive rather than their most careful advocates, and in which the classical revival, which cannot be recruited to either side, is nearly invisible.

What each framework gave up along the way

Both traditions abandoned positions their founders held, and the abandonments are informative because each camp tends to attack the other’s discarded version while defending its own revised one.

Marginalism gave up cardinal utility, the measurable quantity of satisfaction that the founding texts treated as the substance of value, replacing it with ordinal rankings and rates of substitution, and later with the derivation of demand from observed choice alone. It gave up the innocence of the aggregate production function, at least among those who followed the capital controversies, where the possibility of measuring capital independently of distribution was seriously damaged. And it largely gave up the ethical gloss that some early expositors attached to marginal productivity, the claim that factors receive what they contribute and therefore what they deserve. Modern statements of the theory are careful to present marginal productivity as a positive account of factor pricing under specified conditions rather than as a normative justification, and a critic who attacks the ethical version is attacking a position most of the discipline has disclaimed.

The Marxian tradition gave up the proportionality of prices to labour values, which the third volume itself abandons. Much of it gave up the claim that Marx’s own transformation procedure is complete, accepting the technical demonstration that inputs cannot be left untransformed, though the temporal reading contests this. A substantial part of it gave up the claim that the value theory is necessary for the exploitation conclusion, following the analytical reconstruction. And the tradition’s most sophisticated wing gave up the price-explanatory ambition altogether, relocating the theory’s content to the analysis of social form.

The pattern is symmetrical and neither side likes noticing it. Each framework has a founding version that is vulnerable and a revised version that is more defensible and claims less. Each camp attacks the other’s founding version and defends its own revised one. An argument conducted on those terms cannot terminate, because the two sides are not addressing the same objects at any point.

A note on the words objective and subjective

These labels do more damage than any other pair in the comparison, and it is worth being explicit about why.

Calling the labour theory objective suggests that value is a physical property of the object, measurable by inspection, in the way that mass is. That is not the claim. Socially necessary labour time is a social magnitude, determined by prevailing conditions of production across a whole branch, and it changes when techniques change anywhere in the branch, without anything happening to the individual object. A commodity produced yesterday can lose value overnight because a technique elsewhere improved. Whatever that is, it is not a physical property, and the fetishism argument in the first chapter is precisely about the illusion that it is one.

Calling the marginalist theory subjective suggests that prices reflect whims, and that a sufficiently enthusiastic buyer determines the price. That is not the claim either. Marginal valuations are constrained by income, by the availability of substitutes, and by the technology that determines what it costs to supply the next unit, and in long-run competitive equilibrium the framework produces prices equal to marginal costs, which are conditions of production. A framework that arrives at cost-determined prices in the long run is not one in which value is whatever anyone feels.

The more accurate labels would be that the labour theory determines value at the level of the social relations of production while the marginal framework determines it at the level of individual choice under constraint. That formulation is duller and it will not fit on a placard, which is roughly why the older pair persists. A student who uses the accurate formulation in an answer, and explains why the conventional labels mislead, has demonstrated something that no amount of confident assertion demonstrates.

Is one framework more scientific than the other?

Neither, on any usable criterion, and the question is usually a way of asking which one the speaker prefers. Both have a protected core that adherents do not abandon in the face of anomalies, both generate auxiliary hypotheses to absorb disconfirming results, and both have produced substantial bodies of empirical work.

The charge levelled at the Marxian framework is that its central magnitude is unobservable and that every divergence between prediction and observation can be attributed to a countervailing factor. The charge is fair against certain defences and not against the framework as stated, which has implications that could fail, principally about aggregate identities and about how well labour requirements track relative prices compared with alternative input measures.

The charge levelled at the marginalist framework is that its core is a set of consistency conditions on choice rather than a claim about the world, that documented violations of those conditions have produced revisions to the behavioural assumptions rather than abandonment of the optimising apparatus, and that the framework’s central objects, preferences, are inferred from the behaviour they are invoked to explain. That charge is also fair and is also insufficient to condemn the framework, since inferring unobservables from observables is how a great deal of successful science proceeds.

The honest position is that both are research programmes rather than testable propositions, that programmes are assessed over time by whether they generate novel results or merely absorb anomalies, and that this assessment is a judgement rather than a calculation. Anyone who tells a student that one framework is scientific and the other is ideology has substituted an insult for the judgement.

Five misreadings that dominate this comparison

The first misreading is that Marx rejected marginal utility. Chronologically impossible for the founding text, unsupported in the later writing, and usually based on passages attacking earlier utility-based accounts.

The second is that the subjective theory says value is whatever anyone feels. It says nothing of the kind. The claim concerns marginal rates of substitution under constraint, and constraint does most of the work: an agent’s willingness to pay is bounded by income and by the alternatives available, so the framework is not a theory of unconstrained desire. Presenting it as such makes it easy to ridicule and makes the ridicule worthless.

The third is that the two are mirror images, one a supply-side theory and the other a demand-side theory. Both frameworks incorporate supply and demand. The labour theory treats them as governing oscillation around a centre determined by conditions of production; the marginalist framework treats supply as reflecting marginal costs, which are conditions of production, and demand as reflecting marginal valuations. Neither is one-sided, and the supply-versus-demand framing has no basis in either text.

The fourth is that the profession’s adoption of the marginal framework was a refutation event. The reorientation was under way for two decades before the best-known critique of Marx appeared, was driven by the analytical power of the new apparatus for questions the classical approach handled poorly, and was not preceded by any demonstration that the classical approach was internally inconsistent.

The fifth is that accepting one framework requires rejecting all the political conclusions associated with the other. Nothing in marginal analysis entails any particular distributive politics, as the socialist economists who used the apparatus in the calculation debate demonstrated. And the Marxian claim that surplus is appropriated from unpaid labour can be constructed without the value theory, as the analytical Marxists showed. Value theory and politics are far more loosely coupled than either camp’s rhetoric suggests.

The strongest objection to the different-question defence

The different-question rule has to face its own strongest objection or this article fails its own standard.

The objection is this. A theory of value exists to explain exchange ratios. That is what the term has meant in economics since the classical period, that is what Ricardo was doing, that is what Marx was doing when he opened his major work with the analysis of the commodity and the question of what makes two dissimilar things commensurable. If the labour theory cannot explain relative prices better than the alternative, it has failed at the one task a value theory is for. The different-question defence, on this objection, is not a discovery about explananda but a retreat constructed after the failure, and its timing gives it away: it becomes prominent in the second half of the twentieth century, after the transformation problem and the redundancy result had done their work.

This objection is serious and it is partly right. A great deal of twentieth-century Marxist economics did claim price-predictive content and did not deliver it, and the empirical programme that tried to deliver it produced results whose interpretation remains contested. Any defender who presents the social-form reading as the tradition’s settled position is misrepresenting the history.

The reply is textual and it has to be stated carefully. The different-question reading is not invented late; the material for it is in the founding text. The opening chapter’s treatment of the commodity is explicitly an inquiry into why the products of private labour take the form of exchangeable things, the fetishism section is about the social relation appearing as a property of objects, and the third volume states that prices of production diverge from values systematically. The dating objection therefore fails as stated, though a weaker version survives: the reading existed in the text and was not what most of the tradition emphasised until the price-predictive project ran into trouble. Both things are true, and saying so is more honest than either camp’s account.

Where Marx breaks with Ricardo, and why the comparison depends on it

A comparison that treats the labour theory of value as one doctrine running from Smith through Ricardo to Marx will misidentify what the marginalist framework is competing with. Marx inherited the labour principle and changed it in two ways that matter here.

The first change is the distinction between labour and labour power. Ricardo and the classical school treated the wage as the price of labour, which produces an immediate difficulty: if labour is a commodity whose value is determined by the labour required to produce it, the definition is circular. Marx’s move is to say that what the worker sells is not labour but the capacity to labour, which is a commodity like any other with a value determined by what it costs to reproduce, while labour itself is the use of that capacity and is not sold at all. The difference between the value of labour power and the value that labour creates is surplus value. This distinction, developed in the sixth chapter of the first volume of Capital, is what turns a theory of relative prices into a theory of exploitation, and it has no counterpart in Ricardo.

The second change is the treatment of value as a social form rather than only a magnitude. Ricardo asks what regulates exchange ratios and answers with embodied labour. Marx asks a prior question, which is why the products of labour take the form of exchangeable things at all, why social labour has to be validated through money, and what kind of society requires that its members relate to one another through the movement of prices. That question does not appear in classical political economy and it does not appear in the marginalist framework either, which takes exchange, money, and prices as the setting within which agents optimise.

For the comparison, both changes matter. The first means that an objection to Ricardo’s version does not automatically reach Marx’s, and that the exploitation claim rests on a distinction rather than on price proportionality. The second means that the different-question rule has textual grounding in Marx specifically, not in the classical tradition generally. Ricardo really was doing price theory and can be assessed as a competitor to marginalism on that ground. Marx was doing price theory as part of something else, and the something else is where the two frameworks stop competing.

A defender should not overplay this. The price-theoretic component is genuinely present in Marx, the third volume takes it seriously enough to construct a transformation procedure, and generations of Marxist economists treated it as the framework’s core. Saying that Marx was doing something other than price theory is accurate only if it means something in addition to price theory, and the addition does not repair the price theory’s difficulties.

The calculation debate is where the argument actually went

The value-theoretic dispute is often presented as the main event, and for most of the twentieth century it was not. The argument that mattered practically, and that engaged the best minds on both sides, concerned whether a planned economy could allocate resources rationally without market prices.

The case against, developed by Ludwig von Mises in the interwar period and extended by Friedrich Hayek, is that rational allocation requires prices for producer goods, that such prices arise only from exchange between owners, and that a planning board therefore faces not a computational difficulty but the absence of the information a computation would require. Hayek’s version shifts the emphasis from calculation to knowledge: the relevant information is dispersed, tacit, and continually changing, and the price system is a mechanism for economising on how much of it any agent needs to know.

The case for, developed principally by Oskar Lange and Abba Lerner, accepted the marginalist framework entirely and argued within it that a planning board could simulate the market by instructing managers to set price equal to marginal cost and adjusting prices in response to observed shortages and surpluses. The point worth registering for this comparison is that the socialist side of the calculation debate used marginal analysis rather than the labour theory of value, and used it well. That fact is fatal to the assumption that value theory determines political conclusions, and it is why anyone who treats marginalism as inherently pro-capitalist has a historical problem to explain.

The debate’s relevance here is that it relocated the real disagreement. If the question is whether markets are necessary, the labour theory of value contributes very little to the answer, and the arguments that do the work concern information, incentives, and the dispersal of knowledge. Anyone who arrives at the value comparison expecting it to settle the question of economic organisation has come to the wrong argument. The calculation dispute has its own treatment in this series, and readers should go there rather than expecting a theory of value to carry that weight.

The empirical route and what it could settle

Both frameworks make claims that could in principle be tested, and the comparison would be far more tractable if the testing were straightforward. It is not, and the reasons are worth stating because they explain why the argument has run for so long without resolution.

Testing the labour theory’s price claim requires computing labour requirements per unit of output across industries, which is done from national input-output tables, and then comparing the resulting vector with observed prices. The programme exists, it has produced results across many countries, and the results are contested on statistical grounds concerning whether the correlation reflects a relation between price and labour content or merely the fact that both vectors are scaled by industry size. That dispute is the subject of a dedicated article in this cluster and is genuinely unresolved.

Testing the marginalist framework is harder in a different way. The core apparatus is not a prediction about magnitudes but a set of consistency conditions on choice, and the framework’s empirical content lies in the models built from it rather than in the apparatus itself. Violations of the consistency conditions have been documented extensively in experimental work, and the response within the discipline has been to revise the behavioural assumptions rather than to abandon the optimisation framework, which critics describe as protecting the core and defenders describe as normal scientific practice.

What an empirical result could settle is narrow. If labour requirements tracked relative prices no better than arbitrary alternative input measures after proper normalisation, the labour theory’s price claim would be in serious trouble, and its defenders would have to retreat entirely to the social-form reading. If they tracked substantially better, the marginalist account would face a question about why an input measure with no role in its theory predicts so well. Neither result would touch the questions about distribution, exploitation, or social form, which is a reminder that the empirical route addresses the overlap and only the overlap.

Could evidence ever settle this comparison?

Only on the narrow overlap. Data on how well labour requirements track relative prices bears on the one job both frameworks claim. It cannot settle disputes about the origin of surplus, the structure of distribution, or the social form of labour allocation, because the frameworks do not make competing measurable claims there.

Six questions that decide which framework a reader needs

The practical output of the different-question rule is a set of diagnostic questions. Each one, answered honestly, points at a framework, and the exercise is more useful than any general verdict.

Is the question about how an agent responds to a change in constraints, prices, or incentives? The marginal apparatus was built for it and there is no reason to reach for anything else. Is the question about why a particular good’s price moved this quarter, who bears a tax, or what happens to quantities when a subsidy is withdrawn? Same answer.

Is the question about why the wage is what it is, understood not as a factor price but as the outcome of a relation between people who own means of production and people who do not? The Marxian apparatus treats that as its subject, and the marginal framework will answer it by reference to marginal products, which presupposes rather than explains the institutional arrangement.

Is the question about the origin of aggregate profit, as distinct from its distribution among claimants? Only one framework poses that question in that form.

Is the question about what happens over time to the composition of employment, the concentration of ownership, or the frequency of crises? The Marxian framework generates these from its core apparatus; the marginal framework addresses them through separate models, which is a real methodological difference rather than a failure.

Is the question about whether resources are being allocated efficiently in a defined sense, and whether an intervention improves on that allocation? The marginal framework has a developed welfare apparatus and the Marxian framework does not, which is sometimes presented as a Marxian virtue and is at least a fact about what each can do.

Is the question about why the products of labour appear as commodities with prices at all, rather than being allocated some other way? Only one framework treats this as a question rather than a starting point, and a reader whose interest is here should not expect the other to be responsive.

Why this comparison is conducted so badly

It is worth asking why an argument with this much attention has produced so little that is useful, because the answer is diagnostic rather than merely deflationary.

The first reason is that the comparison functions as a proxy. Very few of the people arguing about the determination of exchange ratios are interested in the determination of exchange ratios. They are arguing about whether the existing distribution of income is justified, and the value theory has become the flag under which that argument is conducted. This makes the technical questions rhetorically loaded and makes concessions costly, which is exactly the condition under which arguments do not converge.

The second reason is asymmetric expertise. The strongest technical objections to the labour theory of value come from a literature that requires linear algebra to follow, and the strongest objections to the marginalist framework come from a literature on aggregation and capital that requires equal preparation. Most public argument is conducted by people who have read neither, using arguments inherited from secondary sources that flatten both.

The third reason is that both camps benefit from the two-way framing. Excluding the classical revival lets each present the other’s difficulties as its own vindication. A three-way comparison would require the Marxian side to answer why the classical results can be obtained without value magnitudes, and would require the marginalist side to answer why a non-utility framework produces determinate prices and distribution.

The corrective is not to be even-handed for its own sake. It is to insist on the criterion. An argument that names the question, names the standard of assessment, and names what would change the verdict is worth having whichever side it reaches, and an argument that does none of these has not begun.

Which framework should a given reader use

The answer depends entirely on the question being asked, which is the practical form of the different-question rule.

A reader who wants to understand why a particular good’s price moved, how a tax will be borne, what happens to quantities when a constraint binds, or how agents respond to changed incentives should use the marginal framework, because it was built for exactly those problems and no alternative handles them as economically. A Marxist who refuses the apparatus on principle when analysing a price change is refusing a tool for reasons that have nothing to do with the analysis.

A reader who wants to understand why a society organises production through exchange at all, what the wage relation is and why it is a relation between classes rather than a factor price, how the division of the product between wages and profits is structured, or what accumulation does to the composition of employment over time should use the Marxian framework, because the marginal apparatus takes as given the institutions that these questions are about.

A reader who wants a determinate account of prices and distribution that does not rest on either utility or labour values should look at the classical revival, which is the option most comparisons omit.

And a reader who wants to argue about which is right should first specify the question, because the answer changes with it, and an argument conducted without that specification cannot terminate.

How this comparison is examined

This pairing is an examination staple across political theory, sociology, and history of economic thought papers, and it is examined in a predictable form: a question inviting a comparison, an evaluation, or a judgement about which account of value is more convincing. The band is usually determined by whether the candidate compares on stated criteria or narrates the two theories in sequence.

The one distinction that earns marks here is between the explananda. A candidate who writes that the two theories offer competing answers to the same question is describing the popular argument; a candidate who writes that the labour theory addresses the allocation of social labour and its form while the marginal framework addresses allocation of scarce means under optimisation, and then identifies long-run relative prices as the one genuine overlap, has established the terms on which any evaluation can proceed. Everything after that is easier to write and easier to mark.

The standard trap is the chronology. A candidate who writes that Marx rejected marginal utility has made a factual error that a knowledgeable marker will notice, and it is entirely avoidable. The second trap is treating the diamond and water paradox as a decisive refutation of Marx; it is a demonstration that the marginal framework solves a puzzle the labour theory dissolves differently, and a candidate who says so has turned a stock example into an evaluation point. The third is the assumption that modern economics uses cardinal utility, which is a generation and a half out of date.

For a worked structure, the strongest shape is to define both frameworks precisely, identify the overlap, evaluate on the overlap with a named deciding factor, and then note the third position. Question types, marks logic, and answer skeletons for this material belong to the exam and essay guide for Capital Volume One, which owns that guidance rather than this article, and candidates should use the two together: the structure from there, the comparative content from here. Building a set of comparison notes that keep each criterion attached to its evidence is exactly the kind of task VaultBook’s study and citation tools are for, and this topic rewards that discipline more than most, because the marks come from the criteria rather than from the content.

Running this as a seminar or classroom debate

The debate motion that works is not the one people reach for. The motion that value is objective rather than subjective produces a shouting match, because it invites students to defend a political allegiance under an economic label. The motion that produces argument is narrower and duller on paper: that the two theories are answering different questions and cannot be ranked. Students have to establish what each framework claims before they can attack or defend the motion, which is the work that needed doing.

The misconception to preempt is that this is a left-versus-right disagreement. It is worth stating early that socialist economists used marginal analysis extensively in the calculation debate, that the most damaging technical critique of Marxian value theory came from a framework sympathetic to the left, and that nothing in marginal analysis entails any distributive conclusion. Students who arrive believing that the theory of value determines one’s politics will produce weak arguments in both directions, and the fastest correction is the historical one.

The extract that resolves the most confusion is Smith’s water and diamonds passage, given without commentary, followed by the question of how each framework handles it. Students discover for themselves that one framework answers the question and the other refuses it, and that discovery does more work than an hour of exposition. A follow-up that pays off is to ask what each framework says about the price of a good produced by an inefficient firm, which separates the labour theory from the fairness intuition and separates marginal cost from average cost in one move.

What a writer or researcher should verify

Three claims should never be repeated without checking. The first is any statement about what Marx said concerning marginal utility; the safe form is to give the publication dates and let the reader draw the conclusion. The second is any characterisation of what economists believe, which no survey establishes and which invites correction; the safe form is to describe what standard price theory uses, which is checkable from any graduate text. The third is the claim that the marginal revolution was a response to Marx or to socialism, which is a contested thesis in the history of economic thought rather than an established fact, and which the publication sequence does not support.

For sourcing, the founding marginalist texts are distinct works by distinct authors with different emphases, and treating them as one doctrine loses the differences that matter, particularly between the Austrian line and the equilibrium line that descends from Walras. On the Marxian side, the value theory is stated in the first volume and modified in the third, and the third volume is an editorial construction assembled from manuscripts Marx did not prepare for publication, which should be stated whenever the argument turns on what he concluded. Anyone assembling a comparison file should keep each claim attached to its work, its part, and its date, because in this particular dispute the chronology is doing analytical work and a misplaced date changes the argument. One further verification habit repays the effort here more than in most topics: check whether a source is describing the founding version of a framework or its revised version, since both traditions abandoned positions their founders held and the two camps routinely attack the discarded version while defending the revised one. A citation that establishes what Jevons held about measurable utility establishes nothing about what a contemporary price theorist holds, and a citation that establishes what a nineteenth-century socialist held about exchange at labour values establishes nothing about Marx, who attacked exactly that position. The safest general practice is to name the author, the work, and the claim rather than the school, because school labels in this argument have become containers for whatever the writer wants to attack.

What is settled and what remains open

Settled: the frameworks were not built as rivals, the chronology rules out any direct exchange between the founders, and the two-way framing is a later construction. Settled: the diamond and water paradox is not an argument against the labour theory, and the supply-versus-demand contrast has no basis in either framework. Settled: mainstream price theory does not use the cardinal utility that popular accounts of the subjective theory describe.

Open: whether the aggregate-versus-disaggregate distinction saves marginal productivity theory from the aggregation critique, which specialists continue to dispute. Open: whether the social-form claim is a genuine explanatory achievement or an unfalsifiable residue, which is the question the different-question rule ultimately turns on. Open: whether the empirical relation between labour requirements and relative prices constitutes evidence for the labour theory once industry size is normalised out, which is the subject of the empirical evidence on the labour theory of value and the one place where this comparison could in principle be settled by data rather than by argument.

The reason this comparison generates so much heat and so little movement is that it is conducted as a contest between worldviews when it is a contest between frameworks with different objects and one narrow shared territory. Naming the territory does not end the argument. It makes the argument possible, and it converts a dispute that has generated more than a century of noise, and a great deal of writing that mistakes vehemence for analysis, into three separate questions with different answers and different standards of evidence. The reader who can name those three questions is in a better position than most of the people writing confidently about this on either side, and the position costs nothing beyond the willingness to say that a framework won here, lost there, and was never entered in the contest somewhere else, which is what an honest verdict on a comparison of this kind will nearly always look like when the criteria have been stated in advance.

What neither framework explains

A comparison that only lists strengths against strengths misses the most useful finding, which is that both frameworks share two blind spots, and that the shared blind spots are where the interesting work in economics has been done by people outside both traditions.

The first is the formation of preferences and needs. The marginal framework takes preferences as given and asks what follows; this is a deliberate methodological choice and its defenders are explicit that explaining where preferences come from is not economics. The Marxian framework takes needs as historically formed and says so, but its analytical apparatus does very little with the claim: the value of labour power is said to include a historical and moral element determining the customary standard of living, and that element is then held constant while the analysis proceeds. Both frameworks therefore treat as exogenous something that advertising, social comparison, credit, and institutional change all visibly shape. Neither is well equipped to say what happens when the wants themselves are produced.

The second is the natural world. Both frameworks treat nature as a free input or as a constraint entering through prices. The labour theory explicitly holds that natural materials contribute use value but not value, since no labour was expended in producing them, which is internally consistent and leaves the framework with no measure of depletion. The marginal framework prices natural resources through scarcity and ownership, which handles depletion where property rights and markets exist and handles it not at all where they do not. Neither framework generates ecological limits from its own apparatus, and the substantial literature that addresses this has had to modify or supplement both.

Naming the shared gaps is not a way of declaring the comparison void. It is a corrective to the assumption that the choice between the two exhausts the available positions. A reader who is dissatisfied with both is not thereby committed to incoherence; they are in the company of most of the interesting work done in economics over the past half century, which has proceeded either by modifying one framework substantially or by building something else.

There is a final observation worth making about how the comparison should end. The question of which theory of value is correct has the grammar of a question with an answer, which is what makes it so attractive and so unproductive. The question of what each framework is for, what it can be held to, and what it leaves out has the grammar of a research programme. The second question is harder to argue about at a party and it is the one that produces anything. A reader who leaves able to say what each framework was built to explain, where the two genuinely collide, which one is better on that narrow ground and why, and what neither of them addresses, has extracted everything this comparison contains.

Frequently Asked Questions

Q: What is the difference between the labour theory of value and the subjective theory of value?

They take different objects. The labour theory asks how a society whose production is private and dispersed allocates its total labour among branches, and holds that socially necessary labour time regulates exchange ratios of reproducible commodities while value itself is a social relation appearing as a property of things. The subjective or marginalist framework asks how agents allocate scarce means among competing ends, and derives exchange ratios from marginal rates of substitution under constraint, denying that any substance underlies them. The frameworks genuinely compete on the explanation of long-run relative prices and on the structure of distribution, and almost nowhere else.

Q: Did Marx ever respond to marginal utility theory?

No. The first volume of Capital appeared in eighteen sixty-seven and the founding marginalist works appeared in eighteen seventy-one and eighteen seventy-four, so the value theory was set out before the framework existed. Marx lived until eighteen eighty-three and there is no engagement with the new approach in the published economic writing. Passages cited as his dismissal of subjective valuation are directed at earlier utility-based accounts in classical political economy. The first substantial contact between the two frameworks came from the marginalist side, after the third volume appeared, and the retrospective framing of the two as rival answers to one question was constructed by later commentators rather than by either founder.

Q: Which theory of value do modern economists use?

Standard price theory does not use either in the form popular arguments present. Cardinal utility, the measurable satisfaction that the founding marginalist texts theorised, was largely abandoned in favour of ordinal preferences and rates of substitution, and revealed preference approaches derive demand from observed choices without treating preferences as psychological quantities at all. Prices in general equilibrium formulations follow from preferences, technology, and endowments. Labour values appear in heterodox research programmes and in the empirical input-output literature rather than in mainstream price determination. Anyone asserting what proportion of economists holds any position should be asked which survey establishes it.

Q: Does the diamond and water paradox disprove Marx?

No, and the example is misapplied. The marginal framework solves the puzzle by distinguishing marginal from total utility: water is abundant, so its last unit satisfies a trivial want. The labour theory dissolves the puzzle rather than solving it, by separating use value from value at the outset. Use values are qualitatively different and cannot be ranked on a common scale, so usefulness was never a candidate measure; what can be compared quantitatively is the labour required to obtain reproducible goods. Water and diamonds differ enormously in that respect. The case shows the frameworks handle a shared inheritance differently, which is not the same as one refuting the other.

Q: Are the two theories answering the same question?

Only on a narrow territory. The labour theory’s central question is how the total labour of a society with private production gets distributed among branches and what social form that distribution takes. The marginal framework’s central question is what pattern of production and exchange emerges when agents optimise subject to scarcity. Those are different questions. They overlap on the explanation of long-run relative prices of reproducible commodities, where both make claims, and on the structure of distribution, where they disagree about whether wages and profits are determined inside or outside the price system. Any evaluation should be confined to those overlaps and stated as such.

Q: Which theory better explains real prices?

The marginal framework, on the narrow question of relative prices of reproducible goods in the long run, and the deciding factor is that it yields determinate results from observable data without requiring magnitudes computed under contested assumptions. The labour theory’s own mature statement concedes systematic divergence between prices and values, the transformation between them cannot preserve both aggregate identities under simultaneous valuation, and prices can be determined from technical conditions and one distributional variable without value magnitudes at all. Defenders contest this either by rejecting simultaneous valuation or by relocating the theory’s claim away from price determination, which concedes the overlap rather than winning it.

Q: What do Austrians say about Marx’s value theory?

The Austrian position is not identical to the mainstream marginalist one, and conflating them is a common error since Austrians reject the equilibrium apparatus that mainstream price theory is built on. The best-known Austrian critique of Marx targets the relation between the first and third volumes of Capital, arguing that the system abandons its own foundation when it concedes that commodities exchange at prices of production rather than at values. The separate and arguably more important Austrian argument concerns economic calculation and holds that prices carry dispersed information that no planning procedure can reproduce, which is a claim about what prices do rather than about what determines them.

Q: Is the subjective theory of value provable?

Not in the sense of being demonstrated from first principles, and its defenders generally do not claim it is. It is assessed by the coherence of its assumptions and by the empirical adequacy of the models built on it, particularly whether observed choice behaviour satisfies the consistency conditions the framework requires. Those conditions have been tested extensively and violations are well documented, which has generated substantial revisions within the framework rather than its abandonment. What can be said is that a theory grounded in preferences is difficult to falsify decisively, since preferences are not directly observed, which is the mirror image of the unfalsifiability charge levelled at the labour theory.

Q: Can you accept both theories of value at once?

On the different-question reading, yes, and several economists effectively do. Nothing prevents using marginal analysis for questions about how agents respond to changed constraints while using the Marxian framework for questions about the wage relation, the origin of aggregate surplus, and the social form of labour allocation. The combination becomes incoherent only where the two make rival claims about the same object, principally long-run relative prices, and a position that accepts both has to decide that overlap. Whether the different-question reading is a legitimate interpretation of the labour theory or a retreat constructed after its price-predictive difficulties is itself contested.

Q: When did the marginal revolution happen?

The founding works appeared in the early eighteen seventies, with William Stanley Jevons and Carl Menger publishing independently in eighteen seventy-one and Leon Walras publishing the first part of his system in eighteen seventy-four. The reorientation of the discipline around the new apparatus took decades rather than years and was substantially advanced before the best-known critique of Marx appeared in the eighteen nineties. Describing it as a revolution is a retrospective label; the term marginal revolution was popularised much later by historians of economic thought, some of whom dispute whether the changes were simultaneous or revolutionary enough to warrant the name.

Q: Is value objective or subjective?

The question is less clean than it sounds, because the two frameworks mean different things by both words. The Marxian claim is that value is objective in the sense of being independent of any individual’s valuation, since it is a magnitude of social labour, while also being social rather than physical, so it is not a natural property of the object either. The marginalist claim is that exchange ratios arise from preferences, which are subjective in origin, but are constrained by scarcity and technology, which are not. Asserting that value is subjective is therefore not an argument; it is a statement of allegiance to a framework whose own content is more complex.

Q: Why is Sraffa relevant to this comparison?

Because he demonstrated a third possibility that both camps prefer to leave out. His framework showed that given technical conditions of production and one distributional variable, relative prices and the other distributional variable are determined, without labour values and without marginal utility. That result revived the classical surplus approach in a non-Marxist form, which means rejecting marginalism does not commit anyone to the labour theory of value. It also supplied the strongest technical objection ever made to Marxian value theory, since it showed the classical results could be obtained without value magnitudes. A two-way comparison that omits this option misrepresents the field.

Q: Did the marginal revolution happen because of Marx?

The thesis has been argued and the publication sequence does not support it. The founding marginalist works engaged classical political economy, principally Ricardo and Mill, on technical grounds concerning demand, scarcity, and the determination of exchange ratios. Their authors did not have the third volume of Capital, which appeared two decades later, and the first volume had made limited impact outside socialist circles when they wrote. Marx’s system became a target for marginalist writers in the generation after the founders, most visibly in the eighteen nineties. Presenting the reorientation as an ideological reaction is a contested interpretation and should be labelled as one.

Q: How should a student structure a comparison answer on this topic?

Define both frameworks in terms of what each sets out to explain rather than in terms of their conclusions, since the explananda differ and everything follows from that. Identify the overlap explicitly, which is long-run relative prices of reproducible commodities and the structure of distribution. Evaluate on the overlap with a named deciding factor rather than declaring a general winner. Note the classical revival as a third position, which very few candidates do and which signals wider reading. Avoid the chronological error about Marx and marginal utility, avoid presenting the diamond and water case as decisive, and avoid describing modern economics as based on measurable utility.

Q: What is the strongest argument against saying the theories answer different questions?

That a theory of value exists to explain exchange ratios, that this is what the term has meant since the classical period, that Marx opened his major work by asking what makes dissimilar things commensurable, and that a framework which cannot explain relative prices has therefore failed at its defining task. On this objection the different-question defence is a retreat constructed after the transformation problem and the redundancy result did their damage. The objection is partly right: much twentieth-century Marxist economics did claim price-predictive content. It fails as stated, because the material for the social-form reading is in the founding text rather than invented later.

Q: Does rejecting marginalism commit you to the labour theory of value?

No, and this false alternative is the most consequential error in popular treatments. The classical surplus approach was reconstructed in a non-Marxist form in the mid twentieth century, determining relative prices and distribution from technical conditions and one distributional variable without appeal to utility or to labour values. Post-Keynesian, institutionalist, and other heterodox programmes reject core marginalist assumptions without adopting the labour theory. The framing of the argument as a binary serves both camps rhetorically, since each can present the other’s difficulties as its own vindication, and it survives largely for that reason rather than on the merits.

Q: What should a teacher do when this becomes a political argument in class?

Establish early that the value-theoretic question and the distributive politics are more loosely connected than students assume, using the historical evidence rather than assertion. Socialist economists deployed marginal analysis extensively in the calculation debate. The most damaging technical criticism of Marxian value theory came from an economist sympathetic to the political project. The exploitation claim can be constructed without the labour theory at all. Once students see that the alignment they assumed does not hold historically, the discussion moves from allegiance to argument, and a debate motion phrased as whether the two theories can be ranked at all works better than one phrased as objective versus subjective.

Q: How do the two frameworks differ on distribution?

This is the second genuine overlap and the more interesting one. The Marxian framework treats the wage as determined by the class relation and by what it costs to reproduce labour power, prior to and independently of price formation, with profit as the residual. The marginalist framework determines wages and profits inside the same system that determines prices, as returns corresponding to marginal contributions. The aggregate version of the second position faced a serious difficulty in the capital controversies, since measuring capital independently of distribution proved problematic. That dispute is often overclaimed by Marxian writers: it damaged a particular aggregate formulation rather than vindicating the labour theory of value.