Marx’s discussions of value and price of course draw upon prior traditions in political economy, but they also represent the application of a more general materialist approach to the analysis of social class. The fundamental class-defining categories of necessary labor and surplus labor are applicable in all social circumstances, but the presence of capitalist commodity exchange forces particularization of those categories as part of a more elaborate accounting language with which to quantify and analyze exploitation and its consequences.
At the most general level, all societies produce outputs to meet various needs, combining non-human inputs (produced means of production, along with “nature”) and various human efforts. Social production in all forms thus requires some pattern of performance of labors, allocating efforts to different tasks, but it also inevitably generates some pattern of appropriation of these products, a social distribution of output (inherently a pattern of appropriation of the labors performed in producing the outputs). There is no reason to expect these patterns ever to be the same: an individual’s labors, if any, are sure to be qualitatively and quantitatively different from the labors that produce the products he/she acquires through social distribution outcomes. Understanding the differences but also the connections between the patterns of labor performance and labor appropriation is a task Marxism approaches through the organizing principle of class.
Even at this general level, labors performed and products produced can and should be understood as inextricably linked. The total/gross social product stands as the counterpart to all inputs, human and non-human, required to produce it, but if we analytically set aside both non-human inputs to production and the corresponding portion of total product that replaces those consumed physical inputs, the remainders also represent counterparts: the social net product (output beyond replacement) is associated simply with the new labors performed in production. Economic activity is thus a process of transforming human efforts, the various active labors that people engage in, into net outputs (sometimes but not always commodity outputs) that are the physical basis for social reproduction, the source of all consumption plus net investment. Whatever its apparent complexity and unique features, every economy is simply an ongoing process of transforming human labors into material rewards received. Non-labor inputs to production, in particular the physical means of production, are literally no more or less than that—the means by which society accomplishes this transformation.
Given this substantive social identity between the social net product and the labors performed in its production, Marx understands social distribution, which partitions the net product, as implicitly accomplishing a parallel partitioning of the aggregate labor that produced it. The terms “necessary labor” and “surplus labor,” universally applicable to any social situation, provide the foundation for conceptualizing class:
Assuming some form of social production to exist … a distinction can always be made between that portion of labour whose product is directly consumed individually by the producers and their families and—aside from the part which is productively consumed—that portion of labour which is invariably surplus-labour, whose product serves constantly to satisfy the general social needs no matter how this surplus-product may be divided, …[or] who may function as representative of these social needs.
(Marx 1967, 877–8, emphasis added)
Division of net output between reproduction of labor performers and the residual surplus product is paralleled by a conceptual partitioning of aggregate new labor between corresponding necessary and surplus portions. And while satisfaction of “social needs” might be accomplished in a collectively conscious and direct way, historically it more often occurs via various mechanisms enforcing class exploitation.
Theorizing exploitation quantitatively is most challenging in the presence of systematic commodity exchange. In capitalism, access to produced commodities is mediated by markets in which every transaction simultaneously involves acquisition of a portion of social output by the buyer and the receipt of an income by the seller. Marxism’s (unique) response is to count each commodity in two distinct though related ways. On the one hand, “value” accounting tallies flows of “doing,” of labor performed: here, the number attached to each commodity measures the human labor “put in” to create each good. On the other hand, “price” accounting (more generally, a study of “value-form”) tallies payments—revenue and income flows: each commodity’s price expresses not the labor it takes to produce it but the labor it “brings in” in exchange, so that the heterogeneous social product is measured by the labor-time expressed in the (money) incomes generated by each sale. Marx employs both “countings” in labor-time terms because a crucial aspect of what capitalist class relations are to Marx is summarized in the fact that these two simultaneous measures are both different and related. Each commodity “counts” both as a fraction of the total stream of income deriving from the social product and as a fraction of the total doing involved in producing that total product, and typically the two measures differ, even though the total product considered alternately as a sum of payments and a sum of doings is still the same total product.
On this basis, Marxism derives some basic statements at the heart of the class analysis of capitalism. First, not all who receive incomes engage in performance of labor; it then follows that if some who do not engage in the “doing” of labor nonetheless receive payments allowing command over produced output, it must be the case that others who do engage in “doing” are receiving less than (an equivalent for) their full “doing.” In Marx’s simpler language, rewards to non-doers presuppose “surplus labor.” Applied to the situations of the aged, the disabled, the very young, the point is obvious; applied to non-doers of a different sort—recipients of capitalist incomes reflecting ownership titles rather than labor performed—the point may be less obvious but no less inescapable. Non-labor incomes and the access to goods/services they provide presuppose the existence of surplus or unpaid labor and the corollary existence of a surplus product (output beyond that required to reproduce the performers of labor). When actual performers of labor are neither individual nor collective participants in appropriating such surplus labor, Marx called this exploitation.
Second, as noted, even though the total product considered as the sum of doings is identical to the total product considered as the sum of payments, this need not be so for individual commodities. Hence, a commodity’s value (the total doing or congealed social labor-time it represents as an output of production) need not be identical to its price (the social labor time it represents in an equivalent exchange, paid in the form of money by the buyer). If so, then labor-time is redistributed by the act of exchange—the divergence between commodity values and prices enforces transfers of labor-time, including surplus labor, such that the labor-time represented by the incomes realized from the commodity’s sale diverges from the labor-time originally performed to produce those incomes. In general, then, the capitalist distribution of income (and thus also the allocation of outputs purchased with those incomes) inherently presupposes redistributions of labor-time. And, incorporating the prior point, the profits of capital realized from sales represent a redistribution of the unpaid surplus labor-time of labor performers.
Marx, in fact, structures Capital to examine a sequence of such redistributive transfers, initially from performers of wage labor to employers who capture the revenue from the sale of output, but then subsequently, via market pricing, from less to more productive firms within each industry, from one industry to another based on inter-industry competition or monopoly power, to entities external to commodity production in the form of rents or interest on debt, to governments via taxes, and so on, layer upon layer. A capitalist economy distributes rewards through multiple and overlapping market and (sometimes) non-market redistributions of the labors, the paid and especially unpaid “doings,” that make all income possible. Every distribution of income, and product, is thus simultaneously a redistribution of labor, and value, and Marx’s theory of value with its juxtapositioning of the categories of value and price is the accounting language designed to make this understanding of class visible and quantifiable in a consistent fashion.
Summarizing the basic accounting claims that emerge from within this vision: given appropriate units for commensurating labors on a common scale (see “Abstract Labor,” this volume) and for translating between labor-time and money measures, it should be the case that:
1 the value of total product equals the price of total product (whether both are measured in money terms or both directly in terms of labor-time);
2 the value added by performance of current labor equals, because it is the source of, the aggregate income represented by the social net product;
3 aggregate profit in the broad sense (income available to property ownership) expresses and thus equals unpaid surplus labor in the form of surplus-value (new value created by labors performed, net of payments made to performers of the labor that created that new value).
All three accounting relations must hold simultaneously to strictly sustain the vision of a capitalist economy as a class-structured site for multiple redistributions of unpaid as well as paid labor-time. Value is created by current labor, and value and surplus-value are then “conserved” even as they are redistributed via the pricing of commodities in exchange (of course, value may also be “destroyed” in various ways, though not as a consequence simply of normal commodity exchange).
Over the years since Marx, competing perspectives have emerged concerning the categories of value and price, perspectives that offer different conclusions as to whether, and to what extent, these statements can hold consistently and simultaneously, and why and to what degree it matters if some or all of these accounting claims are deemed inconsistent, or simply dropped as unnecessary.
Influenced by the Ricardian mathematician Bortkiewicz, much 20th-century commentary interpreted commodity value in a fashion that guarantees inconsistency among the three accounting relations. This traditional approach (Bortkiewicz 1907, Sweezy 1942, Seton 1957) sees Marx, like Ricardo, conceiving commodity value as determined strictly by the technical conditions of production: the (“dead”) labor contribution from consumed means of production is expressed simply by the labor technically required to produce these elements of constant capital; adding the value added by (“living”) labor-time performed by current workers yields output value. Later contributors (e.g., Pasinetti 1977, Steedman 1977) in this vein formalized this approach to values as “vertically-integrated labor coefficients”: λ = l[I – A] –1.1 Values (λ) formalized in this fashion as physically-embodied quantities of labor are determined simply by technology (per-unit input coefficients, A and l), without reference to the scales of the various outputs, the conditions governing distribution and exchange, or even the form of the social relations of production. On the other hand, capitalist prices—conceived initially as exchange ratios between commodities—necessarily involve reference to the distribution of income (typically via a specification of wages in some form, often a “real wage bundle” of commodities needed to reproduce labor-power b) and the criteria for exchange (typically, a competitive rate of profit (r) on advanced capital); prices expressed on any absolute scale require some further normalizing constraint. This reading of Marx thus premises two distinct (“dual”) systems of equations, a “value system” in which technological conditions alone determine the quantity of labor embodied in each output, and a separate “price system” in which additional information on distribution (e.g., b) allows determination of exchange ratios; a subsequent condition typically incorporating information on output levels (x) then allows prices to be “normalized” on some absolute scale so that value and price magnitudes can be aggregated and compared.
This “neo-Ricardian” (NR) approach to value allows only limited satisfaction of the accounting relations above—limited, due to now well-known inconsistencies arising from any scale normalization for prices in the presence of a competitive rate of profit. One can impose any particular scale constraint desired (enforcing, say, equality of aggregate values and prices), but in general this does not permit the satisfaction of other conditions (for example, equality of aggregate surplus-value and profit). Only under quite specific and restrictive assumptions is it possible to fulfill simultaneously all three accounting claims above.2 To critics of Marx, this problem of inconsistency arises because of a deeper problem: reference to value and surplus-value is unnecessary if the question is simply to specify the price structure consistent with a competitive distribution of profit. If reference to value is redundant in that project, the point of basing analysis on an accounting in labor-time terms is itself undermined—the entire Marxian conception of exploitation based on the appropriation of surplus labor is called into question.
The neo-Ricardian critique has provoked various Marxian responses, all of which, in one way or another, emphasize the crucial contribution made by Marx’s focus on labor as the source and substance of value and the basis for a class accounting of the quantitative dimensions of capitalism. In particular, though, the established conception of value embedded in the NR approach has become a watershed, a dividing point from which alternative approaches diverge. Three broad forms of response can be distinguished. Some commentators have affirmed the conception of value as labor directly and indirectly embodied on the basis of the existing technology, while defending the central importance of “labor-content” in understanding price phenomena, irrespective of the inconsistencies among accounting relations it entails. Others have effectively abstained, implicitly accepting the embodied-labor conception but directing attention elsewhere, in particular to the usefulness of a labor-time accounting of the distribution of value-added. Still others have rejected that embodied-labor conception of value as a misspecification, instead affirming a different reading of value that allows consistency among the basic accounting claims.
Affirmation of the embodied-labor conception entails acceptance that not all the basic accounting relations can hold simultaneously. Some (e.g., Shaikh 1984) see this not as a problem but rather as a structural consequence of the “relative autonomy” of circulation; if the equality of aggregate price and aggregate value (“direct price”) is presumed, the implied general deviation between total surplus-value and total profit then simply reflects the flow of the portion of surplus-value used for capitalist consumption out of the “circuit of capital” and into the “circuit of revenue” (with price-value deviations at the commodity level showing up in the aggregate as transfers between circuits that entail a loss or gain of actual profit relative to surplus-value (“direct profit”)) (Shaikh 1984, 54). But these deviations are argued to be empirically rather small; referring to various studies, his own and by others, Shaikh (1998) finds that both cross-sectional and temporal variations in prices are highly correlated with variations in value. Value, he argues, thus dominates in the empirical regulation of prices, even though labor-embodied value does not consistently satisfy all of the theoretical accounting relationships implied by Marx’s vision of capitalism.
The empirical correlation between embodied-labor values and prices is a central focus for others who reject the neo-Ricardian critique of value analysis for different reasons. Farjoun and Machover (F&M) (1983) argue for a probabilistic approach to price theory. Capitalist competition does indeed direct capital flows to areas of high profitability, but, they argue, this generates a tendency not for a uniform “equilibrium” rate of profit on capital but rather “a sort of dynamic equilibrium with a characteristic distribution of the various rates of profit among various capitals” (F&M 1983, 18). Dispensing with any requirement for profit rate uniformity allows a probabilistic approach not merely to rates of profit but also to prices as random variables. F&M pose unit price as a random variable related statistically rather than deterministically to the “labor content” of a commodity (effectively identical to the labor-embodied measure of value), justifying this with several Marx-like arguments about the centrality of labor. Empirical studies in this tradition (Cockshott and Cottrell 1998, Frohlich 2013) offer evidence that labor-content functions as a significant “attractor” for market price, to a degree roughly equivalent to that of production prices calculated in the neo-Ricardian manner. F&M argue that their approach leads to a “reconstructed” Marxian theory in which “labour value … must play a central role”—the apparent “contradictions” in accounting “arise not from the notion of labour-value, but from the insistence on a deterministic connection between labour-values and individual … prices and rates of profit” (F&M 1983, 137). Even without a fully consistent accounting structure of the sort that “deterministic” readings of capitalist competition seek, one can apparently still assert that “through the disorderly mechanism of competition, [the surplus labour-content created in the economy] is carved up among all firms, and appears as their profits” (F&M 1983, 156). Accounting consistency is downplayed as an overly “deterministic” concern, with empirical statistical applicability elevated instead as the crucial criterion.
A second, different response to the NR critique also stresses empirical applicability, although correspondence to macroeconomic data rather than statistical correlation is its focus. What is now commonly labeled the “new solution” (NS) approach3 (Duménil 1980, Foley 1982, Mohun 2004) basically accepts the labor-embodied conception of value, but directs attention away from relationships involving commodity value (individual or aggregate), towards those describing distributive shares.
The broad Marxian vision described above presumes that current performance of labor creates the new value available to distribute as social net income; simultaneously, the portions of that income representing profit (initially encompassing all forms of property income) derive simply from redistributions of the surplus-value created by the surplus labor of current workers. The NS approach is able on its own terms to sustain the simultaneous validity and consistency of these two relationships; what allows this to hold, despite the NR critique, is a reinterpretation of surplus-value as the result of unpaid labor, where it is the payment for labor-power, rather than the labor embodied in the commodities effectively received through that payment, that defines paid labor as distinct from unpaid labor.
To clarify: the new solution envisions two distinct accounting realms, a value realm with labor-time as the unit of account, and the price realm where magnitudes are measured in terms of money prices (P). These realms are of course related, and translation from one realm to the other occurs via a scalar often referred to as the “monetary expression of labor-time” or “melt” (m). The value created by new living labor can be expressed in money (as “money value added” MVA) by multiplying total hours worked (lx) by m. MVA is by definition the result of measuring the social net product with prices (Py), which in turn expresses the social net income available to distribute. Jointly, these imply that the melt (m) is = Py/lx, a definition quite deliberately based on the social net product (y).
Of course, MVA breaks down into total wages plus total profit. And total value added by new labor (whether measured in labor-time or money) breaks down into necessary labor plus surplus labor. The new solution then interprets surplus labor as unpaid labor in the following sense: total surplus labor expressed in money (and therefore also monetary surplus-value) equals MVA minus the money paid for total labor-power, i.e., (mlx – Pblx). Since this is definitionally identical to (Py – Pblx), which is total money profit, it follows that total profit equals total surplus-value, whether measured in money (mlx – Pblx) or directly as unpaid labor-time (lx –[Pb/m]lx), where Pb/m here expresses the “paid portion” of an hour worked (the money wage translated via m into a measure of paid labor; Pb/m is the wage share of social net income).4
This is clearly a break from the prior general understanding of surplus-value, since aggregate surplus-value in this NS sense is conceived without direct reference to the values of individual commodities. Yet, since commodity value is still conceived in the traditional labor-embodied sense, the neo-Ricardian critique is not fully bypassed: the monetary value of the net product (mλy), derived from the new value added by living labor (lx), is indeed equal to the price of that net product (Py), but the monetary value of the total gross product (mλx) does not equal its price (Px).
New solution authors regard this as a minor defect, outweighed by the perceived gain in terms of empirical applicability. While the new solution approach to “price” originated in response to the theoretical issues involving prices in the presence of a uniform competitive profit rate, the same accounting is possible using any actual market price vector; it is this that then allows, for example, the NS use of NIPA data in Marxian macro analyses. Still, that gain comes with a corresponding loss in terms of consistency with the broader Marxian theoretical vision. NS accounting is not capable of fully sustaining the vision of capitalist pricing as multi-layered redistribution of labor-time, since inequality of mλx and Px means that aggregate labor-time accounting does not “add up”: aggregate value is not conserved in the process of its redistribution via the pricing structure. As a result, while NS analyses regularly employ the concept of surplus-value or unpaid labor as the aggregate from which profit and the various distributions out of profit derive, they make very little use of the concept of value as distinct from price, either at the level of the individual commodity or at the aggregate level.
A third broad strain of thinking instead focuses on the importance of theoretical consistency, arguing for an understanding of value and price categories that does fully sustain the coherence of the overarching vision of capitalist pricing as the redistribution of paid and unpaid labor-time. While there is considerable diversity among the particular perspectives here, a common element is rejection of the traditional labor-embodied (λ) conception of commodity value in favor of an alternative specification with a broader set of determinants. These various approaches exemplify what some have called a “single system” (SS) interpretation, as distinct from the analytically separate or “dual” systems of value and price equations of the neo-Ricardian critique (and most other Marxian responses).
The original “single-system” interpretation was presented by Wolff, Roberts, and Callari (1982) and later developed by Roberts (1988, 1997, 2009).5 WRC deliberately set aside the question of the melt and monetary measures of value and price variables, focusing instead on prices (p) and profits (π) denominated on the same scale and in the same labor-time units as value and surplus-value (given a value for the melt, however specified, monetary measures of prices and values are necessarily proportional to these labor-time measures, and have the same relationships). Market exchange values are understood to fluctuate around a “center of gravity” reflecting what Marx calls “equivalent” exchange. “Gravitation” here comes about because non-equivalent exchange will provoke adjustments to supply and demand, and the resulting price adjustments keep exchange ratios “tethered to” or centered on equivalence. Crucially, though, the particular “equivalence rule” defining that center of gravity depends on particular social circumstances—equivalent exchange is not simply assumed for all cases to require equal quantities of value changing hands (for example, behaviors that create pressures for a competitively equalized rate of profit enforce a social equivalence rule different from the value-equivalent exchange of Capital 1).
In general (i.e., outside the special circumstances assumed in Capital 1), the labor-time attached to a commodity in (equivalent) exchange will differ from the direct and indirect labor-time required to reproduce it, and the crucial premise of the SS interpretation is that, when this is the social norm, it is the former rather than the latter that carries forward as the relevant measure of its labor-content. This premise can be seen as broadening the narrower insight at the core of the New Solution. The NS accounts for variable capital and its contribution to output value by what is paid for labor-power (wages), yet constant capital is still tallied by the labor technically embodied in (required for the reproduction of) the commodities that are later productively consumed; in contrast, single-system approaches treat both constant and variable capital symmetrically: each is counted by the labor represented by the payment made for them, thus tallying them as the paid labor portions of output value.
So the SS specification of value (v) for a competitive capitalist economy à la Capital 3 differs from λ-value: v = pA + l, or v = pK + s (where K = A + bl, the commodity measure of consumed constant and variable capital). Clearly v and p are approached differently from the separate and distinct “dual” systems of the prior literature. Note, though, that despite appearances, this equation for value does not imply, as some have argued, that a solution for v in this sense requires a prior solution for p (see Roberts 2009); moreover, expressed here as production price p in terms of labor-time, price is not some causally different and dimensionally incommensurable measure alien to Marxian value accounting, but is instead, as Marx claimed, “a definite social manner of expressing the amount of labour bestowed upon an object” (Marx 1967, 82) (see Roberts 1997, where it is shown that production prices p represent the labor embodied in commodities under technologically average or ‘standard’ conditions).
In this SS approach, values (v) and prices (p) emerge from the same social process of competitive production and exchange, and each reflects the same set of basic determining factors (technologies-in-use, output scales, distribution arrangements, along with the social form of equivalence relations—the ruling form of market competition). Values and prices are different as measures—one approaches the commodity as a consequence of the various flows of “doing” necessary to reproduce it, the other as a source of income representing a claim on a portion of the aggregate of “doings” expressed in the social product—but each can be derived from the same data without explicit reference to the other (Roberts 1997; 2009).
Given the v-value conception, all three accounting requirements noted above for the Marxian vision are satisfied here (vx = px, lx = py, and sx = (l – pb)x = πx); the conceptual consistency of these measures allows the labor-time accounts to “add up” in a way that conserves value as commodities move through (equivalent) exchange. In this SS approach, the presence of consistent measures of value and price makes visible the class nature of capitalist distribution: every distribution of commodities and revenue is inherently a redistribution of the (paid and unpaid) labor-time “bestowed upon” those commodities.
A similar but ultimately quite distinct “temporal” viewpoint emphasizes the time dimension in its SS analysis. The “temporal single system interpretation” (TSSI) (Kliman and McGlone 1988, Freeman 1995, Kliman 2007) is concerned not with the structure of equivalent exchange at each moment in time but with the effects produced by changes in the basic data that play out through time. Rather than seeking the particular set of price (and value) relationships which, when applied to both inputs and outputs, allows for exchange equivalence at each moment, TSSI takes the set of input prices as given (valuation of constant and variable capital inputs is treated as historical data, given when they enter the production process), and the resulting output prices are then derived based on current circumstances, with no expectation that the latter will match the former. The rate of profit derived is then differently specified from the rest of the literature, with its denominator expressing “capital” based on the prior (rather than current) period’s reproduction cost of capital.
TSSI thus explicitly rejects the simultaneous treatment of price determination that is central to the SS approach in WRC and most other treatments of profitability as well (although, for any given set of basic data, the TSS solutions for values and prices follow a temporal sequence that converges to the same results as the WRC approach). TSS authors claim a greater textual fidelity to Marx, and see their treatment as validated by its congruence with other aspects of Marx’s work, especially the Marxian tendency of the rate of profit to fall: TSSI authors vigorously argue that only the TSS approach to value and price guarantees the dynamic relation between rising labor productivity and falling profitability said to be Marx’s signature conclusion (Kliman 2007).
A further variant of the SS approach, the “Macro-Monetary” interpretation (Moseley 2015), stresses Marx’s focus on money-capital advances (the advance of “M” which becomes M´ after capitalist production and sale) and argues that constant and variable capital must be taken as predetermined money magnitudes already expressed in terms of the relevant set of prices. The stress on monetary measurement in this approach is so all-encompassing that the measure and even the concept of value as a quantity of social labor-time largely disappears as an analytical category, with only the quantity of current living labor employed (l) remaining as a significant variable not derived in or taken to be pre-given in money terms.
Despite specific differences, SS interpretations share a common premise that the constant capital component of output value is impacted by the payment conditions through which input commodities are acquired, rather than exclusively determined by technical production conditions. It is this different fundamental understanding of labor-time in relation to value that allows each of these single-system approaches to argue that Marx’s vision stands up to the charge of accounting inconsistency.
And more generally, all the various responses to the NR critique understand and deploy the basic Marxian categories of value and price to defend the fundamental class-analytic insight concerning exploitation and its consequences: unpaid surplus labor performed by workers is appropriated as surplus-value and then redistributed complexly though market and non-market mechanisms.
1 Space constraints allow formal consideration of only the simplest case: single-product industries (no joint production) with no fixed capital, uniform turnover, and homogeneous labor, all of which is taken to be productive labor. A customary notation then defines: A, matrix of commodity input coefficients per-unit output; l, vector of labor input coefficients per-unit output; b, vector of commodities required to reproduce each unit of labor-power (the “real wage bundle”); x, vector of gross outputs produced; y = [I – A]x, vector of net outputs; λ, vector of commodity “value” understood as vertically-integrated labor coefficients; P, vector of prices expressed in “money”; p, vector of prices expressed in labor-time.
2 See Steedman (1977), or the summary in Roberts (1997); Steedman also enumerates other circumstances that critics see as sources of inconsistency or indeterminacy in Marxian categories (e.g., fixed capital, joint production, heterogeneous labor, etc.).
3 The label “new solution” reflects the initial focus on rebutting the NR critique of Marx’s volume 3 “solution” for a general rate of profit via “transformation” of values into production prices, but the approach is more accurately understood as a Marxian accounting framework applicable whatever the form of pricing.
4 NS authors typically take as distributional data not a real-wage bundle b but a scalar wage in money terms w, where w takes the place of Pb in the above relations.
5 Recent extensions of this general approach include Kristjanson-Gural (2009) and Olsen (2013); Roberts (1997) takes up the more complex cases not formally considered here.
Bortkiewicz, L. von. 1952 [1907]. “Value and Price in the Marxian System.” International Economic Papers 2: 5–60.
Cockshott, W. P. and A. Cottrell. 1998. “Does Marx Need to Transform?” In Marxian Economics: A Reappraisal, Vol. 2, R. Bellofiore, ed., 70–85. New York: St. Martin’s Press.
Duménil, G. 1980. De la valeur aux prix de production. Paris: Economica.
Farjoun, E. and M. Machover. 1983. Laws of Chaos: A Probabalistic Approach to Political Economy. London: Verso.
Foley, D. 1982. “The Value of Money, the Value of Labor Power, and the Marxian Transformation Problem.” Review of Radical Political Economics 14(2): 37–47.
Freeman, A. 1995. “The Psychopathology of Walrasian Marxism.” In Marx and Non-equilibrium Economics, A. Freeman and G. Carchedi, eds., 1–28. Cheltenham: Edward Elgar.
Frohlich, N. 2013. “Labour Values, Prices of Production and the Missing Equalization of Profit Rates: Evidence from the German Economy.” Cambridge Journal of Economics 37(5): 1107–1126.
Kliman, A. 2007. Reclaiming Marx’s ‘Capital’: A Refutation of the Myth of Inconsistency. Lanham, MD: Lexington Books.
Kliman, A. and T. McGlone. 1988. “The Transformation Non-problem and the Non-transformation Problem.” Capital and Class 35: 56–83.
Kristjanson-Gural, D. 2009. “Post-Structural Logic in Marx’s Theory of Value.” Rethinking Marxism 21(1): 14–33.
Marx, K. 1967 [1865]. Capital, Volume 3. New York: International Publishers.
Mohun, S. 2004. “The Labour Theory of Value as Foundation for Empirical Investigations.” Metroeconomica 55(1): 75–95.
Moseley, F. 2015. Money and Totality. Leiden: Brill.
Olsen, E. 2013. “Similarities and Differences in Two Marxian Theories of Value.” Working paper, University of Missouri, Kansas City.
Pasinetti, L. 1977. Lectures on the Theory of Production. New York: Columbia University Press.
Roberts, B. 1988. “What is Profit?” Rethinking Marxism 1(1): 136–151.
Roberts, B. 1997. “Embodied Labor and Competitive Prices: A Physical Quantities Approach.” Cambridge Journal of Economics 21(4): 483–502.
Roberts, B. 2009. “Ricardo : Standard Commodity :: Marx : ___?___.” Review of Political Economy 21(4): 589–619.
Seton, F. 1957. “The Transformation Problem.” Review of Economic Studies 24(3): 149–160.
Shaikh, A. 1984. “The Transformation from Marx to Sraffa.” In Ricardo, Marx, Sraffa: The Langston Memorial Volume, E. Mandel and A. Freeman, eds., 43–84. London: Verso.
Shaikh, A. 1998. “The Empirical Strength of the Labour Theory of Value.” In Marxian Economics: A Reappraisal, Vol. 2, R. Bellofiore, ed., 225–251. New York: St. Martin’s Press.
Steedman, I. 1977. Marx After Sraffa. London: New Left Books.
Sweezy, P. 1970 [1942]. The Theory of Capitalist Development. New York: Monthly Review Press.
Wolff, R., B. Roberts, and A. Callari. 1982. “Marx’s (not Ricardo’s) ‘Transformation Problem’: A Radical Reconceptualization.” History of Political Economy 14(4): 564–582.