Note: The citations are forthcoming, since I would like to add them formally...
I. The Problem
The word "value" is vague enough to sustain permanent disagreement between schools of political economy because everyone is talking about something slightly different. Marx understood value as abstract labour that is objectified in the commodity and the marginalists dissolved it into subjective utility at the margin. Later came physicalists in the line of Sraffa who derived relative prices from input-output structures. But if the question "what is value?" assumes that value is a single substance that can be "found," then this question is malformed from the start.
First, let's demonstrate that neither Marx (nor his critics) succeeded in finding a unitary substance, and that the strongest available defense of Marx's internal consistency actually supports our position.
Marx's Position
Marx establishes the difference between value and exchange value: exchange-value is just the observable ratio (like twenty yards of linen for one quote,) and Marx insists that this ratio is only the "form of appearance" of value proper that is behind it. Marx goes through layers of abstraction to find this substance: a coat is set equal to a pile of linen even though as useful things they have nothing to do with each other, so if different use-values are to be treated as equivalent to each other, then we need to abstract from both the use-value and also from the concrete labour that produced them. What is left? Abstract human labour without a specific form.
For Marx, this is what value is - abstract human labour, objectified in the commodity. Its magnitude can be determined with SNLT (socially necessary labour-time,) the average amount of labour currently needed to make the commodity under the current productive conditions. There are a few points we need to look at deeper: firstly, value is determined by current SNLT, not by historically expended labour. If wheat harvested last year required more labour than wheat harvested today then the old wheat's value falls to match the new. The market enforces this through producers who expend more than the average SNLT and cannot command a higher price from it, and the producers who expend less are not forced to charge less, so competition assembles the ratio around the social average.
Secondly, an hour of average labour "always yields the same amount of value, independently of any variations in productivity." When productivity doubles throughout an industry an hour of labour produces twice the output but the same amount of value, so each unit of output therefore has half its former value. This is how Marx theorises the observable fact that rising productivity tends to lower prices. It is also the foundation of his law of the tendential fall in the rate of profit (LTFRP): as capitalists replace workers with machines, more of each dollar of advanced capital goes to means of production and less to hiring workers. But it is the workers' living labour that adds all new value - so if labour-saving innovation reduces the living labour employed per dollar of capital while each hour of labour still adds the same value, the new value created per dollar of capital falls and with it, given a constant rate of exploitation, the rate of profit.
Thirdly, the commodity's value has two components: the value transferred from used-up means of production (constant capital) and the new value added by living labour. The surplus-value (amount by which the new value added exceeds the wages paid) is the only source of profit for Marx. The total profit in the economy equals the total surplus-value; the total price equals the total value; the aggregate price rate of profit equals the aggregate value rate of profit. These three "aggregate equalities" are his bases, and if they fail, we cannot really claim that exploitation is the source of profit.
Fourth (and this is where it will get controversial) not all labour is equal per hour. Skilled work must count as a multiple of unskilled which Marx addresses this through the "reduction of complex to simple labour," but he does not derive the conversion ratios in advance. He says the reduction is performed socially, by the market, "behind the backs of producers." So, the market is performing the conversion, but the labour theory of value is supposed to explain market prices - If we need the market to determine a variable that the theory requires in order to explain the market then things start to look worryingly circular.
Marx's defense would be that this misunderstands the method because he is not building a deductive system where every ratio must be calculable before exchange. This is valid, but notice what follows from this: if the conversion ratio is socially produced rather than discovered, if exchange generates it rather than revealing something already there, then "abstract labour objectified in the commodity" does not seem like a substance residing within things, it looks more like a social process of cost-determination that constrains the exchange ratio from below. And like that, the seeds of our semiotic approach to value are already inside of Marx's own logic.
The strongest defense of Marx
The circularity problem along and the broader set of internal inconsistency "charges" against Marx lead to a 100 year long controversy that Andrew Kliman's Reclaiming Marx's Capital (2007) decisively addressed. Kliman showed that every single alleged inconsistency in Marx's value theory comes from a single interpretive error of assuming simultaneous valuation. Simply, that is the stipulation that the per-unit price (or value) of each input must equal the per-unit price (or value) of the same good when produced as an output.
The temporal single-system interpretation (TSSI) that Kliman defends makes two small modifications to Marx. First, valuation is temporal: input and output prices can differ, because inputs were purchased at one moment and outputs emerge later. Second, values and prices are determined interdependently within a single system, not in two separate "value" and "price" systems
These modifications are not ad hoc - Bortkiewicz's "correction" of Marx started the standard simultaneist reading, was fully aware that Marx's own conception was what Bortkiewicz called "successivist" (a causal chain "in which each link is determined in its composition and its magnitude, only by the preceding links.") Bortkiewicz attacked this conception explicitly and he instead supported the Walrasian view in which "the various economic factors or elements condition each other mutually"
The simultaneist interpretation of Marx is just the the imposition of a Walrasian framework onto a theory that was built on the opposite premise. When you impose this framework and contradictions appear then the contradictions belong to your imposition. Under the TSSI the circularity dissolves. Input prices are not unknowns to be solved for simultaneously with output prices since they are data, already determined from the previous period. The capital value advanced at the start of a production period is what was actually paid, at previously established prices. Then, production happens. Output values are determined by the SNLT at the time of production: the market does not need to know the conversion ratio of complex to simple labour in advance, because it works with already-settled input prices.
Marx's own reasoning was explicitly temporalist because he argued that profit "is produced before this division takes place, and before there can be any talk of it" - so profit exists prior to its distribution through competition. He argued that a commodity's price "is determined before it enters into circulation" - the sale realises the price but does not determine its magnitude. Kliman summarises the principle: the capital value actually advanced at the start of a period cannot be retroactively altered by subsequent price changes. "What's done is done."
Under this reading, all three aggregate equalities hold. The LTFRP is internally consistent. The transformation of values into prices of production does not generate the contradictions that Bortkiewicz claimed to have found. Kliman concludes from here: "the allegations of inconsistency are unproved... [and] implausible. When one interpretation makes the text make sense, while others fail to do so because they create avoidable inconsistencies within the text, it is not plausible that the latter interpretations are correct" (Kliman, Preface).
We accept this conclusion: Marx's value theory is internally consistent when read temporally and as a single system. This matters, because we are not going to build on the claim that "Marx was confused," but instead by asserting that what he correctly identified and what the TSSI vindicates the internal consistency of is one of two forces in the exchange ratio rather not the ratio itself.
Cost of Provision is a Constraint, Not a Substance
If we grand absolutely everything above, TSSI, that Marx's theory is internally consistent, that SNLT determines value when read temporally, that all the aggregate equalities hold, LTFRP is valid, etc. What follows? The cost of provision - all the expenditures needed to bring a commodity into existence, socially averaged under competition - constraints the exchange ratio from below.
If a producer consistently exchanges below what it cost to produce, production stops: competition undercuts producers above the averaged floor and eliminates those below it, so the ratio converges toward cost. This is what SNLT describes. It is a real material constraint and Marx saw it clearly. If cost of provision were the substance of value, if it constituted the exchange ratio rather than only constraining it, then the ratio would always sit at cost. But it plainly does not... An expensive watch does not exchange at the value of its metal and labour. 2-3 years ago, an NFT did not exchange at the cost of minting a digital file. Even in ordinary commodity markets the ratio floats above cost when it can and is forced down to cost only when there is pressure from competition.
So that means that something else is pulling the ratio above the floor. This other force is what we will call the acquisitive interpretant, which is the buyer's semiotically produced disposition to give up a certain quantity of one thing to acquire another. But before we introduce it let us deal with the rival account that claims to have dissolved the problem entirely.
Physicalism's Absurdities
The Sraffian tradition that started with "Production of Commodities by Means of Commodities" (1960), showed that relative prices and the rate of profit are derivable from the physical input-output structure of production without any reference to labour values. If the same ratios come out without the concept of value, what is value doing? This challenge can seem to be strong, but Kliman demonstrated that it rests on the same interpretive mistake as all other charges against Marx in this area: simultaneous valuation once more! Sraffa's derivation works only if input prices are constrained to equal output prices. Under this constraint, value is indeed redundant, but so is any idea that labour-time determines value, because simultaneous valuation makes labour-time irrelevant.
Kliman's corn model is the clearest illustration. Corn is produced using seed corn and labour. If simultaneous valuation is imposed - a bushel of seed corn must have the same value as a bushel of output corn - then "the value of a bushel of corn output must also be $5, no matter how much or how little the farmworkers have had to labor in order to produce it. They may have had to toil a thousand hours, or only ten hours - or not at all! It makes no difference." The corn's value is locked in by the constraint, regardless of the labour performed. There is "no meaningful sense in which the corn's value depends upon the amount of labor needed to produce it."
The rate of profit becomes a purely physical quantity: if ten bushels of seed corn produce twelve bushels of output, the rate of profit is 20% - period. It is 20% whether the workers laboured a thousand hours or not at all, whether the corn is worth $5 or $500 per bushel. Value has been eliminated as a distinct concept and everything is determined by technology and real wages alone... this is what physicalism is. And so, naturally, it produces some very absurd results. Under simultaneous valuation, values can be negative even without joint production: if ten bushels of seed corn produce only nine bushels (a bad harvest), the simultaneous-valuation equation forces the corn's value to be negative when living labour is positive, or living labour to subtract value when the corn's value is positive
The first modern physicalist Dmitriev claimed in 1897 that profit could exist even if no living labour were performed at all, in a fully automated economy. But Kliman showed decisively in his work that Dmitriev assumed precisely what he needed to prove: that the machines would have a positive price. If no labour is performed and no value is created, the machines are free, the equation collapses, and the rate of profit is undefined.
Physicalism also implies that the rate of profit can never fall as a result of rising productivity. This is the Okishio theorem (1961) which for decades was thought to have disproved Marx's LTFRP. But Kliman showed that the theorem is logically invalid, it establishes its conclusion only by valuing inputs and outputs simultaneously, and so by "spiriting away" the deflationary impact of rising productivity. When a bushel of corn at the start of the year costs $5 but a bushel at the end costs $3.33, simultaneous valuation retroactively revalues the input at $3.33. One-third of the purchase price just "vanishes into thin air." This artificial reduction in the capital advanced is what makes the rate of profit appear to rise. Under temporal valuation where what's done is done and the capital advanced is what was actually paid, the deflationary impact of rising productivity pulls the rate of profit down... and this is exactly as Marx argued!
What matters for our purposes is the following: physicalism is what happens when you strip away both forces in the exchange ratio. There is no semiotically articulated cost of provision - just mechanical input-output quantities! There is no acquisitive interpretant - just physical surplus! Everything is reduced to a relationship between things, with no mediation. And this produces bizarre absurdities: negative values, profit without labour, the impossibility of falling profitability from rising productivity... these absurdities are evidence that the semiotic mediation on both sides of the ratio - the articulation of costs through social codes and the production of the buyer's disposition through semiotic processes - is very much real.
Marginalists
The subjective theory of value, developed by the marginalists gets remove the search for a substance since they just classify value as marginal utility, the satisfaction from one additional unit. There is nothing behind the exchange ratio to be found; the ratio simply reflects the point at which supply meets demand, with supply and demand both derived from individuals' subjective preferences. But if value is just marginal utility, why are exchange ratios patterned and stable? Why do similar commodities trade at similar prices across different buyers? The marginalist answer is that preferences are "revealed through choice." But is this really an explanation? It seems more like a refusal to explain anything. Where is the demand curve from? Why does it shift then? The theory treats these as exogenous. This is a mirror of the physicalist error, just as they derived supply-side quantities from physical data and saw conversion to value as automatic, as do marginalists obsevre demand-side qualities and treat the origin of demand as given, and neither touch the mediation that actually produces the magnitudes they work with.
Marx's own LTV has an explicit limitation in scope that bears on this point, because his theory pertains exclusively to commodity production, goods "produced for the purpose of being exchanged." When products are not produced as commodities, "the rates at which they exchange may depend exclusively upon the demand for them, or upon normative considerations, or upon customary rules". This is a concession from within the Marxist framework that indeed demand-side determination is real in some contexts, not just an illusion that we dissolve into labour-time.
So then why it is real? What produces it? How does it operate?
The Impasse
Marx has a theory of why exchange ratios are patterned (the cost of provision, socially averaged under competitive conditions, disciplines the ratio) but under the standard (simultaneist) interpretation his own substance becomes redundant: Sraffa can derive the same ratios without it. Under the TSSI, Marx's theory is rescued from redundancy, but only by making the determination explicitly temporal, which reveals that SNLT is a temporal constraint on the ratio - so a floor rather than the ratio itself. The marginalists have an observable measure of the other force (the buyer's willingness to pay), but no theory of where it comes from and no explanation for why it shifts.
What resolves the impasse is the recognition that value is not a substance to be located - it is not in the commodity, not in abstract labour, not in the buyer's mind... but it is a ratio of two forces, whose ceiling force determines mentally that this quantity of one thing stands for that quantity of another. The first force is the cost of provision: the totality of expenditures required to bring the commodity into existence, temporally prior to exchange. The second force is the acquisitive interpretant: the buyer's semiotically produced disposition to give up a certain quantity of one thing to acquire another. Neither is value itself. Value is the ratio between them.
II. The Two Forces
All social reality is mediated through human consciousness, which functions through systems of signs. So the coordination of social action, the organisation of production, the stabilisation of expectations - all of these operate semiotically. See here.
This apparatus is what makes the second force in the exchange ratio - the acquisitive interpretant - visible as a produced effect rather than an exogenous given.
Let us reestablish the two forces of the ratio:
The First Force: Cost of Provision
The cost of provision is the totality of expenditures a producer must make in order to bring a commodity into existence and also to market (c+v). These expenditures are temporally prior to the exchange, they are already determined, already incurred, before the commodity enters circulation. They constrain the exchange ratio from below: if a producer consistently exchanges below cost, production ceases. Under competition the cost of provision converges to the social average. Competition undercuts producers above the averaged floor and destroys anyone below it. This convergence is the real content of Marx's concept of SNLT, so, not an abstract substance - just the competitive discipline that pulls the exchange ratio toward the averaged cost. The TSSI makes the temporal character of this constraint clear - input prices are data, already settled, and the capital advanced is what was actually paid. This floor is also evidently historical.
But the cost of provision has two layers, and the second is where semiotic mediation enters even on the supply side as well:
The first layer is brute material constraint: you need iron ore to make steel, calories to sustain labour, electricity for machinery. These are non-negotiable, material reality cannot really be narrated away. A famine will strike regardless of the semiotic structure, a shortage of rare earth metals constrains production regardless of the subsemiosphere's self-understanding. This is the layer that keeps all social analysis "honest." Codes can mediate how material constraints are interpreted and responded to. However they cannot abolish the constraints themselves.
The second layer is the articulating of the cost. What counts as a cost? How are costs measured? What inputs are deemed "necessary"? Whether a producer "needs" to provide health insurance, meet environmental standards, pay a living wage, obtain regulatory approval... these are determined by the codes of the subsemiospheres within which production occurs. A pharmaceutical company's cost of provision includes years of regulatory compliance, this is not because the molecules require it, but because the codes of the medical-regulatory subsemiosphere require it. If we just change the codes and the cost of provision changes, even if the material inputs remain identical.
This means that even the supply-side constraint is (partially) needing in a semiotic framework too. The brute material layer is not but the articulation of cost is. This is why the "same" commodity can have different costs of provision in different social environments, and why cost structures shift historically without any change in the underlying material requirements.
The monetary expression of labour-time (MELT) - the conversion factor between labour-hours and money is itself a social determination. Kliman notes that it is the economy-wide ratio of total money price to total labour-time value. This ratio is produced by the totality of exchanges in the economy, which means it is produced by the interaction of both forces - cost of provision and acquisitive interpretant - across all markets simultaneously. The MELT is not given prior to exchange because it emerges from the ratio-system generally.
The Second Force: The Acquisitive Interpretant
The acquisitive interpretant is the buyer's disposition to give up a certain quantity of one thing in order to acquire another. It is the Peircean interpretant specified for the context of exchange. Marginalists observe it and call it "marginal utility" or "preference." But calling it a preference and treating it as exogenous is what prevents a healthy understanding.... what produced this interpretant? Which codes? Within which subsemiosphere? Through what mechanisms of sign-reproduction?
The entire distance between cost and price is the space in which this industry operates. Consider the variation across commodities. In bread: the representamen is the loaf, the object is nourishment, the acquisitive interpretant is the disposition to exchange for it. This interpretant is anchored heavily in material constraint (biological need naturally dominates.) Codes still mediate (what counts as food, what is acceptable to eat, what grain is considered a staple), but the interpretant does not collapse when fashions shift. There is a floor beneath the floor: the body's need to eat sustains the acquisitive interpretant independently of any code.
In a luxury watch: the representamen is the watch itself, the object is not timekeeping (a cheaper watch keeps better time) but status, a social message readable within specific subsemiospheres. The acquisitive interpretant - the disposition to give up ten thousand dollars - is produced almost entirely by codes of luxury branding, social hierarchy, peer signalling, and the semiotics of craftsmanship. Change the codes and the interpretant collapses. The ratio falls toward cost of provision, it falls all the way down towards what the metal and labour are worth as material inputs.
A Note on the Value-Form School
The claim that value "only exists in the process of exchange" might appear to align this analysis with the value-form understanding... this strand holds value is "established in the market rather than having existence prior to it" But this view is rather superficial.
The value-form position (if we take it as an interpretation of Marx) makes "the theory that value is determined by labor-time meaningless" because if abstract labour only becomes abstract when the product is subsequently sold, then value is created by the sale and labour-time determines nothing at all. The temporal sequence of determination is violated because labour must be abstract in production for it to create value.
But we don't make this mistake as we do not claim that value is "established" in the market, rather that the exchange ratio is pulled between two forces, both of which are prior to the specific exchange event. The cost of provision is temporally prior: it is what was already spent, at already-determined prices. The acquisitive interpretant is semiotically prior - it is produced by codes that exist and operate within subsemiospheres before the specific buyer encounters the specific commodity.
The exchange ratio is where the two forces meet, but they do not originate there. The market is the site where the ratio is realised but it is still not where its determinants are created. This is an important difference because the value-form school looks at the social mediation of value but cannot specify what that mediation does to the ratio. It says the social form matters but cannot say how. Our framework specifies exactly what the social form does: it produces the acquisitive interpretant within subsemiospheres and it articulates the cost of provision through codes that determine what counts as a necessary expenditure.
III. Where the Ratio Settles
Where the exchange ratio lands depends on which force is stronger and the dynamics of the ratio follow from this dependence with precision. When the cost of provision is high and the market is competitive, the ratio clusters near cost: competition will undercut the producers above the averaged floor and destroy those below it. The acquisitive interpretant is present (buyers still need to be willing to pay) but competitive pressure prevents the ratio from floating far above cost. This is basically just the world Marx described and his analysis is the approximation under these conditions. The LTV is not wrong here since it is a local description of what happens when the cost-of-provision force dominates.
When the cost of provision is low and the acquisitive interpretant is high then the ratio floats far above cost. The distance between cost and price is held up by codes. Here lie the speculative assets, fashion, branded pharmaceuticals, platform monopolies... The STV is not "wrong" here because it is a local description of what happens when the acquisitive interpretant dominates. Also, when the cost of provision is high and the acquisitive interpretant is low, the thing stops being produced, because the ratio stops sustaining production.
When the cost of provision falls due to technological progress, the floor drops. If the acquisitive interpretant remains unchanged, the gap between floor and ratio widens and a larger proportion of the exchange ratio gets dependant on the now higher acquisitive interpretant. This makes the ratio structurally more shaky even if it has not moved since more of it is held up by social forces that can easily shift while the material constraint beneath it has receded.
It should be noted the law of the tendential fall in the rate of profit is actualy a description of the dynamics within the cost-of-provision-dominated regime. When labour-saving technological change occurs in competitive commodity production then the cost of provision per unit falls (productivity rises, prices or the rate of inflation decline). Kliman showed this is internally consistent when we read it temporally, the deflationary impact of rising productivity pulls the rate of profit down even as physical output per unit of input rises as the capital advanced is valued at the prices that were actually paid, not revalued at cheaper replacement costs. Bortkiewicz himself noticed in 1907 that his simultaneist formulae prevented falling prices from lowering the rate of profit, and instead of recognising this as a hole in his model he just declared Marx "wrong" and the Okishio theorem systematised the error.
Crises
Economic crises can originate from either side of the ratio. A cost-side crisis occurs when productivity shifts lower the floor, competition disciplines the ratio downward, capital values are destroyed. This is (roughly) what Marx was describing. The mechanism is temporal in that capital was advanced at higher prices, output must be sold at lower prices, the difference is a loss. The crisis restores profitability by destroying the capital values that accumulated during the preceding period: "the tendency of the rate of profit to fall is constantly overcome by way of crises"
An interpretant-side crisis occurs when that which sustains the acquisitive interpretant destabilises, so such as whe important actors exit the subsemiosphere, the narrative loses coherence, confidence breaks, and the interpretant collapses. The ratio falls toward cost of provisionb but if cost of provision is near zero, there is no floor to catch the fall. This explains the NFT collapse, bursting of speculative bubbles, etc.
To be continued...