Carlo D’Ippoliti
In a state of society, however, in which the industrial system is entirely founded on purchase and sale . . . the question of Value is fundamental. Almost every speculation respecting the economical interests of a society thus constituted implies some theory of Value: the smallest error on that subject infects with corresponding error all our other conclusions; and anything vague or misty in our conception of it creates confusion and uncertainty in everything else.
Mill [1848] 1871: 456
For some heterodox economics schools values are different from prices, while others regard such distinction as metaphysical. This difference of approach is related to different opinions on what function(s) the price system takes on in a capitalist economy. By discussing these different functions (allocative, distributive, and informative), this chapter shows that price and value theory holds analytical centrality in all economic paradigms, both heterodox and mainstream, to the point of defining contemporary heterodox (and thus orthodox) economics.
Indeed, the concepts of price and value, and thus price theory, directly descend from an economist’s general picture of the working of the economy as a whole. To use Schumpeter’s (1954) terminology, I refer here to the ‘pre-analytical vision,’ that is the first, ideologically inspired step of economic theorization. While not yet articulated in concepts and variables, a vision contains some elements (and significantly, not others) and preliminary implicit hypotheses on the causal links between these elements—it thus crucially shapes the subsequent phases of theorization. Since price theory descends from the pre-analytical vision of the working of the economy as a whole, it inherits its analytical centrality from the general character of this vision.
The literature on the theory of value and prices is vast and ever expanding, and no attempt is made here to provide a comprehensive review of it.1 Rather, I will focus on the two main visions of the working of the economy that, with alternate fortunes, have developed throughout the history of economic thought, which I shall call the ‘marketplace vision’ and the ‘market-system vision.’ In turn, these are related to two main approaches to price theory: the subjective and the objective approach, respectively. I argue that all heterodox traditions hold some variant of the market-system vision of the economy and the objective approach to prices (with the exception of Austrian economics and some strands of ecological and evolutionary economics).
It appears that sharing the same pre-analytical vision, thus the same or compatible theories of value, are necessary conditions to assess the compatibility between schools of economic thought, if not the potential for their integration (D’Ippoliti & Roncaglia 2015). This may appear surprising to some, given that heterodox economics seem to reflect mainstream economics’ bipartition of a mathematically sophisticated ‘high theory,’ where prices and values are explicitly considered, and applications of ‘lowbrow theory,’ which often even make recourse to marginalist supply-and-demand mechanisms.
Thus, it seems necessary to investigate the possibility for a wide-tent pluralist approach within the boundaries of the market-system view of the economy and the associated objective concept of prices. This will be the object of the present chapter.
In part, difficulties in the debate on value and price theories derive from the adoption of the same terms with different meanings or conceptual underpinnings. Thus, it is first of all necessary to clarify the terminology used in this chapter. This terminology does not reflect the one adopted by any single school, but it appears as necessary in order to compare different, occasionally incommensurable approaches.2
For our purposes, we can use the terms object, good, commodity, and service as interchangeable, unless explicitly noted. Since the Middle Ages, scholars recognize two different aspects of an object’s ability to meet some human desire: one, a property of the object; and the other, an idiosyncratic feeling of the individual. With respect to both, we shall refer to this ability as the object’s value in use. As is well known, mainstream economics assumes that such property is quantifiable through a cardinal or ordinal utility function. In turn, value in use is assumed to directly or indirectly determine a second sort of value, referred to as value in exchange. Such exchange-value is the degree to which a commodity performs a second function, that is, it can be given or taken in exchange for something else. The notion that value in use measures or causes value in exchange will be referred to here as the subjective approach to value, because it postulates that value derives from introspection.
By contrast, the British classical political economists, and most heterodox economists after them, consider the existence of some value in use as a simple precondition for the object to have some value in exchange. But value in use is not generally considered a measure of value in exchange and/or it does not exert a quantifiable influence on it. The cause or measure of value in exchange must be sought elsewhere, in variables external to the individual; thus, this approach will be referred to here as the objective approach to value. This term descends from the history of economic thought on this subject, which is mostly concerned with variables that may be objectively determined, pertaining to the technology, distribution of income, and institutions of a certain society.
Indeed, at least since Petty, we have a distinct view that the difficulty to supply must be the real determinant of the value in exchange of a commodity, its ‘natural price.’ Since at the time money itself was either a commodity, such as gold and silver, or its value was directly proportional to the quantity of a commodity, it was obvious that money was not an adequate unit of measurement because its own value was subject to change. Petty put forward a theory of natural prices based on production costs (often simplified as a labor theory of value or, according to some historians of thought, a land-and-labor theory of value). After Petty, Cantillon proposed a land theory of value. Later, the British classical political economists almost unanimously employed a labor theory of value, alternatively based on the notions of labor commanded or labor employed. The evolution of the theory of value in classical political economy as well as its more recent reconstruction by Piero Sraffa are recounted in Chapter 7 (by Sinha) in this volume. Subsequent developments, as is well known, see neoclassical economists rescuing the earlier subjective approach in a new, mathematically sophisticated way. In contrast, within contemporary heterodoxy, different schools have developed different concepts and analyses of prices and values, while most heterodox schools agree on employing some form of the objective approach.
In the subjective approach, an object’s value in exchange is its price. Each agent has a reservation price, that is the maximum value that a buyer is willing to pay in order to obtain it, or the minimum value that a seller is willing to accept in order to part with it. The market price is the equilibrium value that does not induce any buyer or seller to change her behavior (see next section for a more detailed definition).
In the objective approach, several different positions have been put forward. As will be discussed in the next sections, some economists did not (and do not) think that the price at which each transaction takes place is liable to theorization. Others do not object to the theorization of market prices, but see them as very volatile or anyway not the only relevant quantity. In both cases, economists following the objective approach refer to different concepts—that is, value (as opposed to market price), natural price, normal price, long-run price, production price, or several other concepts, depending on the specific theory. What all these concepts have in common is that they represent the abstract, social characteristics of an object’s value in exchange, as opposed to the vagaries of the turbulent capitalist economy that affect its volatile—possibly random—market price. Thus, these terms denote different notions of regulating prices, that is, values that are relevant for the analysis at hand. For example, theories in which the accumulation of capital is a fundamental driver of economic activity, will have regulating prices as the values that firms use when making investment decisions; or theories in which free entry and exit from an industry are key concepts will exhibit regulating values as those that are assumed to prevail under competitive conditions (these may or may not coincide with the former example).
Thus, in the rest of this chapter ‘value’ and ‘price,’ in the specific sense of regulating price, will be used as synonymous, when it is not relevant to making a distinction between approaches in which a specific notion of regulating price is being adopted. In any case, both value and (regulating) price will be assumed to be different from market prices, which may or may not qualify as theoretical variables.
Adopting this terminology, price theory encompasses both value theory (that is, the analysis of the determinants of regulating prices) and the analysis of the relation between values and market prices.
Many social thinkers tend to think of the market as a marketplace, a distinct point in space and time in which one-shot exchanges take place. Archetypal images of this vision may be the stock exchange, a flea market, or a street market. From this perspective, each individual comes to the marketplace already endowed with money, goods, and/or the ability to work (we do not ask where this endowment comes from), engages in one or more transactions, and leaves the marketplace. If we further assume that goods are perishable (for example, fish), suppliers will not leave the market until their endowments are exhausted, thus the idea of equilibrium as ‘market clearing’ emerges. In such a setup, the value in exchange, or the market price of a particular good is determined by the interplay of the exogenously given quantity brought to the market and buyers’ preferences. Adding the assumption of the law of one price, as a consequence of competition among buyers and among sellers separately, and the assumption that no transaction takes place at a price different from the equilibrium one, we get a theory of price explained by supply and demand.
Today, among the heterodox schools only Austrian economics and some strands of ecological and evolutionary economics start from this pre-analytical vision; as a consequence, Austrians consistently stick to a subjective theory of value (though not necessarily the marginalist theory, as shown for example by Menger), while the ecological and evolutionary schools exhibit some internal variation, as hinted to below.
Contrary to the view of the market as a marketplace, others conceptualize the market as a social system of recurring exchanges, in a society based on the division of labor and social stratification. Archetypal images of this vision are Cantillon’s description of the mutual dependence of town and countryside, the nationalist ideal of an autarchic economy, or the asymmetric relations of the global North and South in the contemporary world economy. For the sake of simplicity, at least at the pre-analytical stage of theorizing it is pictured that productive units produce only one or a few kinds of commodities, and thus everybody needs to continuously carry out a number of transactions in order to obtain the vast assortment of goods and services that they need. Notably, such transactions include firms, which need to buy their means of production in the market. Thus, according to this vision the production and distribution of commodities are intertwined, and the crucial analytical problem is how such a decentralized society can survive and even grow (in a sustainable manner, if we adopt an ecological point of view). While in the marketplace vision the given data are initial endowments and fixed preferences, in the market-as-a-social-system vision the main givens are technology and the social organization of production. If we assume competition (but the assumption can be modified, as discussed further below), prices are determined in the process of production in which a unit’s product is another one’s means of production. Thus a system of prices has to be such that no productive unit buys its means of production at too high a cost (thus incurring losses) or too low (thus benefiting of exceptional profits). Otherwise, competition between capitalists would lead some firms out of business (in the former case), or would attract more firms in the industry (in the latter).
The difference in the respective focuses on the sphere of exchange in the marketplace view and the sphere of production in the market-system view, led some authors to denote this as the defining feature of the two main economic paradigms of mainstream and heterodox economics. However, it must be stressed that the differences are pervasive and concern the pre-analytical vision, conceptualization, and formalization of the functioning of the economy, and are not limited to the greater or lesser focus on some aspects of it. For this reason, most heterodox economists object to a research program that seeks compatibility between mainstream economics and specific heterodox theories.
In the marketplace view, prices play a crucial allocative role. While the ‘equilibrium’ price depends on the equality of supply and demand, each quantity in turn depends on prices, positively (the quantity supplied) and negatively (the quantity demanded). The problem of simultaneous interdependence is solved through the conceptualization of price equations, and the bulk of the mainstream literature on ‘general equilibrium’ theory is devoted to showing that a solution to a system of equations representing market clearing in all markets exists (although such a solution may lack crucial desired characteristics such as uniqueness and stability). Under these conditions, prices’ allocative role consists in inducing firms and consumers to change their demand and supply in response to discrepancies between prices and the agents’ subjective assessment of the value of each good. Although prices are surreptitiously equalized across individuals with the assumption of a no-arbitrage condition (Mirowski 1989: 236), prices are defined at the individual level, as a measure of her desires (or ‘preferences’) in relation to scarce means. At the aggregate level, prices are thus indexes of the overall scarcity of a good or service (given its best production technology) relative to the market demand for it; and aggregates such as an economy’s total value added are a measure of an economy’s capacity to satisfy (unlimited) human wants.
In mainstream economics, such a subjective evaluation is expressed by firms’ marginal costs and consumers’ marginal utility. However, since the very beginning of the so-called marginalist revolution, many were aware that individual desires are incommensurable and probably not even liable to precise quantification: this was for example Menger’s position. Moreover, historical evidence suggests that even in market societies and even abstracting from non-market provision of goods and services, prices do not necessarily play an allocative role: more often than not, it is quantities that adjust to accommodate demand and supply shocks, with prices kept fixed by regulation, convenience, or convention (Braudel 1967). Thus, a branch of ‘neo-Austrian’ economics emerged, based on the notion of out-of-equilibrium dynamics, in which prices are fixed for a meaningful time interval. Finally, while not immediately apparent at the time, the ‘socialist calculation debate’ in the 1930s highlighted the political relevance of understanding prices as the carriers of diffused information that is not available to any single market participant (Hayek 1948). These and possibly other causes of dissent from mainstream economics led to a different concept of prices and their role within Austrian economics, that is the role of transferring information processed by markets.3
This different notion of prices is one of the main reasons why some Austrian economists identify themselves as ‘heterodox.’ However, the Austrians’ adherence to a marketplace view and its associated notion of prices as determined by the equilibrium of supply and demand are the reasons why other heterodox economists consider Austrian economics a variant of mainstream economics (D’Ippoliti & Roncaglia 2015).
It is noteworthy that the recent ‘information revolution’ within mainstream economics gave centrality to the notions of markets as information processors and prices as information carriers. This is made evident by the increasing relevance of the notion of informational efficiency (especially, of financial markets, following the works by Eugene Fama) and its complicated relation with allocative efficiency (Stiglitz & Grossman 1980).4 Consequently, a slow convergence of the price concept may be taking place among the schools that start from a marketplace vision, and the mainstream approach may become increasingly similar to the Austrian.
In the market-system view, the notion of market clearing is absent and prices play a completely different role—that is, coordinating capitalist production by distributing the resources necessary for the continuation of production and the surplus among industries and between social classes. Here, the schematic idea is that production in a capitalist society requires the prior expenditure of money to buy the means of production and to pay for workers’ wages (constituting ‘capital’); capitalists advance money capital with the view of gaining it back with a profit through sales (the ‘realization’ of capital). Prices determine the aggregate value of the means of production and the revenues, and the purchasing power of wages, profits, and rents. Since both exchanges among productive units (and between workers and firms) and exchanges of final consumption goods take place in markets, any transfer of purchasing power must take place through exchange-values.5
In this approach too, even if exchange-values are by definition relative, it makes sense to compute aggregate values of capital, profit, or the economic surplus. These quantities represent a collection of heterogeneous commodities and services, aggregated by weighting them for their price. In intuitive terms, the price here is an index of difficulty of production in a society with particular technology and social institutions that determine the functional distribution of income.6
All heterodox traditions hold some variant of the market-system vision of the economy and the objective approach to prices; as explained above, those that do not should probably not be defined as heterodox, from a theoretical point of view. However, the peculiar position of (American) institutional economics should be singled out. As Jo (2016) summarizes, early institutionalists and Veblen himself occasionally recur to simple supply and demand reasoning, perhaps because of its hegemonic role in US academic economics. Truly, basic pillars of institutional thought, such as socially constructed preferences and artificially created scarcity, highlight the role of the agency of dominant elites and going concerns that are obviously alternative to the mechanistic and impersonal price mechanism posited by marginalist economics. But subjective factors play a significant role in some institutional analyses of prices, leading to questions as to these authors’ adherence to the objective paradigm. This is true, for example, of the influences of ‘ceremonial efficiency’ on prices and wages, or the strong role attributed to habits and conventions.
These factors are not totally absent in other objective approaches—most heterodox economists agree that, for example, the determination of the socially necessary labor to produce a commodity and the level of the subsistence wage are historically contingent, and affected by institutional factors. However, the issue is methodological insofar as the assumption of general mechanisms of economic processes “is not the way institutionalists explain economic activities. There is no universally ordained institutional economics. All institutional theories are (to be) historically contingent due to the specificity of social institutions and the heterogeneity of agency therein” (Jo 2016: 338). If one views that the institutional approach should have a distinct theory for each historical case, then it is not possible to enlist institutionalists in either the subjective or the objective camp. If instead we define (varieties of) capitalism as a sufficiently specific spectrum of validity for a certain theory—that is, for the identification of a general mechanism—then it should be possible to develop an institutional theory of prices. Jo (2016) makes such an attempt, but as a matter of fact he ends up delineating an institutional theory of pricing, rather than a price theory.
Apart from these methodological issues, four main conceptual issues seem to be most relevant in assessing the potential for mutual compatibility between heterodox theories of value: (i) the relation between values in exchange and relative prices, with the associated concepts of natural, production, and market prices; (ii) the role and scope of the labor theory of value; (iii) the definition and inclusion in the analysis of what counts as ‘labor’; and (iv) the mutual relation between the determination of quantities and prices.
The first systematic treatments of value as determined only by objective forces may be traced back to Petty, Cantillon, and later the physiocrats and the British classicals. The fact that they often held different conceptual views may be a source of continuing differences between contemporary approaches.
Petty wrote at a time when a strong quantitative turn was pervading natural sciences, with the early developments of classical mechanics. He notoriously set himself to only talk “in Terms of Number, Weight, or Measure,” expressly leaving aside the subjective variables “that depend upon the mutable Minds, Opinions, Appetites and Passions” (quoted in Roncaglia 2005: 56). This drive for quantification led Petty to first inquire into adequate subjects of quantitative analysis. In his ‘Dialogue of Diamonds,’ Petty sketches the procedure through which an imaginary diamond dealer traces back each specific diamond—with its size, weight, color, etc.—to an ideal diamond, by means of correction coefficients. In turn, there will be an ideal market for the ideal diamond, in which the law of one price prevails. Therefore commodities and markets, thus market prices, are but abstractions, useful to establish a link between the seemingly disconnected exchanges in a given society. Each transaction is affected by a myriad of erratic elements (for example, bargaining power) and it would be impossible to develop a quantitative theory of what may be considered a random collection of disconnected facts.
Most classical political economists shared the idea that this conceptualization of the market price is not liable to precise quantitative theorization. Such a concept of the market price begs a question: what then is systematic or permanent and thus liable for quantitative theorization in market exchanges? Many later writers interpret Adam Smith’s analogy of gravitation as implying that market prices turbulently and erratically move over time, but always exhibit a long-run tendency to approach the respective natural prices. Smith defined natural prices differently from Petty (as well as from Cantillon and the physiocrats), but their analytical role is the same with all these authors—that is, regulating capitalist economic activity in a society.
In contemporary analyses, the ‘econophysics’ literature and some heterodox economists take up this conceptualization of the market price and argue for the use of a stochastic approach to account for its randomness. Significantly, though, this way the market price again becomes a variable liable to precise quantification. Nonetheless, the use of statistical distributions does not necessarily imply that the notion of regulating prices is useless (see, for example, Shaikh 2016). Ultimately, natural or production prices, or labor values—all instances of prices that regulate capitalist activity in a given society—may still be relevant in this approach, since they are the reference point for capitalists or other agents who make decisions.
According to Marshall’s interpretation of Smith, natural prices regulate the long-run tendencies of production, in that natural prices correspond to the sum of wages, profits, and rents paid at their natural rate (with some ambiguities and risks of circular reasoning). This long-run equilibrium is reached because competition between capitalists equalizes profit rates across industries (as competition between workers and rentiers equalizes wage and rent rates). Contrary to this interpretation, Andrews (2015: 266) brings up J.S. Mill’s interpretation of natural prices as logical requirements: they are the prices that allow production to continue unchanged. Not only market prices need not gravitate around them, but also production may actually not continue, and indeed it is very unlikely for it to remain unchanged in the face of capital accumulation, technological innovations, and social developments.
There is bidirectional causality between conceptual analyses of contemporary price theories and historical investigations. For example, Adam Smith’s and the other classical political economists’ views on natural prices have often been re-interpreted (or clarified, depending on the point of view) and debated within the Sraffian school. At the same time, Sraffian and other historical analyses help clarify the conceptual foundations of contemporary approaches to prices. This example is especially pertinent because there is a clear connection between the concept of natural prices and Sraffa’s (1960) production prices. Some Sraffians (for example, Garegnani, Pasinetti, and Schefold) interpret production prices as long-run end positions of a gravitation process driven by the competition among capitalists, while others (for example, Roncaglia, Sen, and Arena) hold a ‘snapshot interpretation,’ according to which production prices express the instantaneous conditions of reproduction of a capitalist system at a specific moment. The former conceptualization of production prices is in obvious continuity with Marshall’s (and possibly Ricardo’s) interpretation of Smith’s natural prices, whereas the latter is in continuity with Mill’s.
This conceptual divergence has relevance for the potential complementarity between the Sraffian approach and other heterodox approaches (see Aspromourgos 2004; Lavoie 2011). Some claim—but this is a subject of ongoing controversy—that the assumption that production prices represent long-run equilibrium positions implies an assumption of constant returns to scale. This latter assumption is in turn incompatible with the conceptualization of production by several other heterodox schools, which typically posit increasing returns. The snapshot interpretation seems more able to conceptually accommodate not only increasing returns to scale, but also—at least in principle—radical uncertainty (which has always been connected to profits in the economic literature), which is crucial from a Post Keynesian perspective. Conversely, the notion of production prices as long-run end positions is historically closer to the Marxian approach. In Marxian treatment, however, labor values are more important than prices, and thus the debate on the compatibility of Sraffa’s approach with the Marxian one has rather revolved around the relation between labor values and production prices, the so-called transformation problem, which is dealt with in the next section.
Thus, while the debate around the compatibility of various price theories usually revolves around the mathematical properties of formal models (such as the high number of exogenous variables in Sraffa’s systems, or their degrees of freedom), the conceptual underpinnings of these models seem equally, if not more, important.
Piero Sraffa’s 1960 Production of Commodities by Means of Commodities sparked successive rounds of debate on the relation and compatibility between his analysis and Marx’s. Towards the end of the 1970s, the majority view seems to have been that Marx’s labor values may be given a consistent treatment that is not incompatible with Sraffa’s analysis; however, production prices are independent of Marx’s labor values and the latter are possibly not even needed to discuss class struggle (see Steedman 1977). These results may be a cause for the subsequent independence of Sraffian economics from Marxian economics, ‘the no-communication monologues’ (Bellofiore 2012). Recent archival work on both Marx’s and Sraffa’s papers is igniting new debates, shifting focus from the ‘transformation problem’ to the similarity and potential filiation of Sraffa’s system with Marx’s reproduction schemes (and from Quesnay’s Tableau Économique ). Part of the reason for this shift is that, as informed by the ‘new interpretation’ of Marx’s theory, many Marxists now hold the labor theory of value only at the macroeconomic level.7
However, here too conceptual differences could be more relevant than formal compatibility. For example, in a simple Sraffa system the labor embodied may be computed through a ‘reduction to dated quantities of labor.’ However, even ignoring the existence of a commodity residual, whose size depends on the rate of profits, this procedure only shows how much labor would have been necessary to produce a certain commodity, had the current technology been used since the very beginning of production—that is, the production of the commodities’ means of production, and the means of production of the commodities’ means of production, and so on. This is not just different from what Marx was interested in, but it is indeed a very un-Marxian hypothesis, far away from his evolutionary approach (Mirowski 1989: 182).
On conceptual grounds, there is a more obvious sense in which the Marxian theory of value is complementary to, but has not been substituted by, Sraffa’s systems. As was made clear within the Cambridge capital controversies, ‘capital’ (and surplus, and thus profits) has two meanings: a heterogeneous collection of commodities and a homogeneous monetary value (Cohen & Harcourt 2003). In the latter sense, reference to an absolute value such as labor embodied has a clear conceptual sense, independent of relative prices.
Moreover, Marxian economists consider living labor (and the difference between labor and labor power) to explain the ‘cause’ of surplus value—that is, surplus labor; it consists of all labor activities exerted for a time over and above what is necessary to produce the goods used for workers’ consumption. Even in some Sraffian treatments, labor is considered to be the only ultimate source of exchange-value and thus it retains a conceptual significance independently of production prices (see, for example, Pasinetti 1981).
The issue of the (degree of) compatibility between the Sraffian system and Marxian economics concerns formal models as well. For example, a relevant issue comes up when trying to determine which distributive variable is taken as exogenously given in Sraffa’s equations. The approach closest to Marx’s (and the classics’) treatment is to assume the wage as given. It is usually assumed to be at the level of reproduction of the workforce, implying that the given wage is not necessarily the minimum necessary for subsistence but what is socially necessary in a certain society at a certain time. Such focus on the reproduction of workers has been considered as an obvious potential link with feminist literature (Picchio 1992). An approach closer to Post Keynesian economics, inspired by Sraffa’s (1960) own treatment, is to take the profit rate as given, possibly determined by the interest rate and thus highlighting the relevance of financialization in contemporary capitalism. A third hypothesis, embraced for example by Sylos Labini, is that either of the given variables is itself variable; whether it is given or variable depends on the historical specificity of the society we wish to analyze. This last approach is obviously the closest to institutional economics.
Finally, the exogenous givens of Sraffa’s systems, be they explicit or implicit, concern not only distribution but also technology. This point is especially relevant to the extent that the definition of industries, outputs, and input coefficients implicitly define ‘productive’ and ‘unproductive’ labor, as will be discussed in the next section.
As with ‘natural,’ the adjective ‘productive’ has not been a value-free term in the history of economic thought. Smith’s critique of the physiocrats’ assumption that only agricultural labor is productive is well known. Yet, Smith too struggled with the concept of productive labor. Until today, it is fair to say that contemporary heterodox schools do not have a shared understanding of what ‘productive’ means. From the point of view of price theory one definition is that only the labor that contributes to the production of surplus value should be considered as productive. This may or may not encompass all labor exerted in society, depending on one’s conceptualization of some aspects that are often at the margins of the pre-analytical vision of the market system, even though they involve growing shares of the population. Thus, from a Marxian perspective all wage labor involved in capitalist production (though possibly not in the sphere of circulation) should be considered productive. Evidently, this may exclude non-wage labor or employment in non-capitalist sectors of the economy.
From a Sraffian perspective, even though it obviously depends on production coefficients and output levels, each industry’s productivity can only be computed given the distributive variables and production prices (because the means of production and outputs are heterogeneous quantities). That is to say, every industry’s productivity depends on that of other industries. In this perspective, more relevant concepts are the viability and/or self-reproducibility of an industry, and thus all labor that contributes to the economic system’s reproduction and accumulation should be considered as productive. A further complication in both the Marxian and the Sraffian approaches is that, as explained in the previous section, natural and production prices are regulating prices—that is, they are not the selling prices of every firm in an industry and not even necessarily of any firm. Rather, regulating prices connote what is ‘socially necessary’—that is, the values taken as a reference point by the capitalists for their new investments. Thus, even in capitalist production there is ‘waste,’ bearing unproductive work to the extent that not all firms adopt the socially necessary technique of production at all times. Waste is all the more relevant under noncompetitive market conditions, when capitalist firms incur non-socially necessary expenditures such as the costs of marketing, advertising, etc. (Sylos Labini [1956] 1962; Baran & Sweezy 1966).
Exceptions or variants of the above definition of productive labor are found in both mainstream and heterodox economic literature. Austrians, adopting a marketplace vision of the economy, consider all employed labor as productive, because all labor produces value in use. Post Keynesians define productivity on the basis of value added rather than surplus labor, even though many Post Keynesians consider finance as ‘unproductive.’
Ultimately, all models (and even non-formalized theories) must include an infinite number of ceteris paribus clauses, since no one-to-one theoretical mapping of reality is possible. What is kept out of the analysis is an indication of which factors are considered to be of secondary or no relevance to the determination of a certain phenomenon. Thus, the industries or sectors that are ignored by price theory imply that certain kinds of work are deemed of secondary relevance to the production of surplus value and/or the reproduction of the economy—they are implicitly assumed to be unproductive work. From this point of view, at least three elements that are usually excluded (or only included in the ceteris paribus clauses) seem to deserve a closer investigation than has so far been done in extant literature on price theory.
A first element is the public sector. Its usual exclusion from price theory may be due to implicit hypotheses of ‘first approximation’ in formal models. But at the conceptual level it confirms the prejudice that publicly provided goods and services are deemed to be irrelevant for the production of surplus value, the accumulation of capital, and the reproduction of society (contra, see Corsi & D’Ippoliti 2013).
A second element is the financial sector. Its characteristic exclusion from the analysis of prices cannot be considered as an acceptable approximation, because the financial sector necessarily affects the interest rate and possibly the profit rate (and the wage rate), therefore the value of the surplus. Employment in the financial sector may be considered unproductive if finance is considered to belong to the circulation or the consumption sphere (Barba & de Vivo 2012). Some assume a ‘banking sector’ that earns the same rate of profit as all other industries, which then determines an interest rate that is conceptualized as the price of banks’ output (Shaikh 2016). Further complications arise from the consideration of who pays and who receives interest payments, with interest alternatively considered as a deduction from the pool of profits or an internal transfer among capitalists. Indeed, recent research shows that in a Sraffian system the impact of bank lending on the distribution of income can be non-trivial (Panico et al. 2012).
Finally, a third crucial element is unpaid labor, especially housework and unpaid care labor. The issue has obviously been raised by feminist economists. It should be noted again that the definition of productive and unproductive labor has systemic consequences for the definition and value of the product and of the surplus product, as well as for their distribution; thus this issue should be of general interest for all economists and cannot be ignored even on first approximation (Picchio 1992). If households are considered as productive because they are necessary for the reproduction of the system, several conceptual issues emerge, which have not yet found a definitive solution. According to Todorova (2009), it is possible to conceptualize unpaid work in terms of capitalist firms’ payment to workers of a wage bill smaller than that which would grant the reproduction of the labor force, that is Todorova’s ‘warranted wage bill.’ However, the attribution of a value to unpaid work (that is, the difference between the actual wage bill and the warranted wage bill) assumes that there is a common unit of measurement—that is, given the current heterodox approaches to price theory, a market value for unpaid work. The problem of comparability between paid and unpaid labor is solved within marginalist economics by tracing back both kinds of labor to their subjective value in use (disutility). But in the heterodox economic approaches, value, though defined differently by each school, is generally not a matter of individual appreciation: its definition is inherently related to the capitalist mode of production (be it labor values, prices of production, etc.). Thus, the introduction of a non-capitalist mode of production in the theoretical system calls for yet another redefinition of an appropriate concept and a measure of value, through which heterogeneous quantities pertaining to the different parts of the system (the capitalist and the non-capitalist) could be compared straightforwardly.
Folbre (1982) finds such a common unit of measurement in the ‘labor embodied’ measure of value, on the grounds that the concept of ‘socially necessary labor’ may apply to both paid and unpaid work equally. Folbre’s claim is that Marx considered the labor theory of value as only applicable to capitalist production because in that case competition eliminates inefficient producers; however, it can be assumed that other social forces may induce households to produce efficiently too. Thus in her view establishing a clear meaning of ‘socially necessary household labor’ is required. Besides the acceptance of this specific approach, it is noteworthy here that this procedure drastically modifies the very concepts of ‘labor embodied’ and ‘value.’
One last open issue is the relationship between prices and quantities. This is perhaps the most complicated and controversial of the four issues considered here. It is virtually impossible to adequately summarize the vast number of positions and related issues in a single chapter, since the issue at stake spans from economic methodology to the history of economic thought, to formal models. Indeed, heterodox economists’ views vary on the level at which the link (if any) between prices and quantities should be sought.
At the macroeconomic level, new interpretations of Marxian economics take quantities into account in the determination of labor values (Bellofiore 2012). Similarly, in the Sraffian literature, Pasinetti developed his model of structural dynamics, which considers the interplay of effective demand and production prices, before the publication of Sraffa (1960) and later further developed it in Pasinetti (1981). Conversely, Roncaglia & Tonveronachi (2014) are skeptical about the search for an all-encompassing formal model of the whole economy; rather they argue for the usefulness of Keynes’ ‘short causal chains’ in economic modeling, and the opportunity to limit the search for compatibility between models at the conceptual level.
The approach looking for (and requiring) conceptual compatibility without a necessary formal compatibility inevitably produces less elegant and comprehensive models, but it allows room for (conceptual) integration of several disparate approaches.8 However, many heterodox economists maintain that seeking theoretical integration at the formal modeling level (as for example attempted by Lee & Jo 2011) is necessary to establish a consistent theoretical alternative to the mainstream.
With regard to Marshallian ‘partial equilibrium,’ two main issues arise. The first is the role of demand and the associated issue of economies of scale, obviously relevant in Post Keynesian literature, for example. In general, classical political economists did not ignore the role of demand. Not only did they consider demand variations as a cause of continuous, transitory market perturbations, as in Adam Smith’s gravitation metaphor, but they also perceived a systematic nexus between prices and demand. For example J.-B. Say noticed that as the price of a commodity increases, the number of consumers who can afford to buy it falls (Chapter 21, by Corsi & Guarini in this volume). Similarly J. S. Mill, incorrectly accused of having ‘corrupted’ the classical theory of value, highlighted that if firms buy their means of production in advance of production and sales, and if they face a financial constraint, then as the price of a means of production increases, firms will demand less of it (without implying substitutability between various means of production; see D’Ippoliti 2011). These insights are a long shot from a demand-and-supply theory of the determination of market prices, which is founded on a completely different pre-analytical vision and a formal model of market clearing based on supply and demand curves.
The second issue that arises at the industry and/or firm level is the relation between values, or natural or production prices, and firms’ pricing procedures. For historical reasons, in the literature the need to establish a clear connection between the two has been explicitly linked to the reciprocal relation of Sraffian and Post Keynesian economics (Lavoie 2011). But the connection is relevant for Marxian economics and other approaches as well. For example, Post Keynesian economists tend to consider prices as determined by a mark-up on costs (Lee 1998). This is only compatible with Sraffa’s production prices—when interpreted as long-run positions—if by costs we imply the ‘normal costs’ that are associated with the best technology in the industry and a ‘normal’ degree of capacity utilization, and if the mark-up comprises a competitively determined ‘pure remuneration of capital’ that cannot be fixed independently by the firms but has to relate to the economic system as a whole (Aspromourgos 2004).
More research on these analytical constraints is surely warranted, but once again it seems appropriate to draw attention to the need for mutual conceptual compatibility. For example, Sylos Labini ([1956] 1962) and in the Marxian tradition Baran & Sweezy (1966) raised the issue that all-pervading oligopolistic or monopolistic market forms are not just a partial equilibrium issue: they have pervasive systemic consequences (for example on the pace and diffusion of technological progress). This view is patently at odds with the long-run conception of production prices, based on the notion of free entry and exit from an industry. The degree to which the two conceptions can be reconciled, for example with the addiction of an exogenously determined ‘extra-profit’ vector that is dependent on firms’ market power, is still a matter of controversy.
In short, the link between theories of prices and theories of output levels directly calls into question how best to conceptualize the general pre-analytical vision described above. Thus, issues such as whether and how the same vision can represent different historical varieties of capitalism— for example, the competitive, oligopolistic, and money-manager stages of capitalism—become crucial.
By way of conclusion, it seems fitting to briefly describe a recent stream of literature spinning off from the debates on price theory. For several reasons, mainstream economics has increasingly become a field of applied statistics with increasingly less attention to economic concepts. By the same token we observe an empirical turn in heterodox studies on price theory. Institutional constraints such as journal and department rankings or hiring and promotion procedures may explain this trend. But possibly new computational possibilities and theoretical reasons are relevant factors too. As is well known, the Cambridge capital controversies address the theoretical inconsistency of marginalist economic theory. Immediately thereafter, mainstream economists ignored the results of that debate often on the grounds of lack of empirical evidence (or even empirical relevance, from the mainstream viewpoint; see Cohen & Harcourt 2003). Thus, some Marxian and Sraffian economists may deem it necessary to continue their critique on this new level.
In the new empirical turn, at least two strands of literature should be distinguished. One may be labeled a positivist research program, looking at the data in order to falsify and/or compare theories. Thus, statistical investigations of the empirical relevance of paradoxes in the marginalist capital theory increasingly make use of input-output tables (for a review see Schefold 2016).9 Similarly, the transformation problem has been investigated by trying to measure how much prices diverge from labor values (for example, Shaikh 2016). Perhaps the most copious branch of this literature is devoted to the empirical investigation of Marx’s laws of tendency of capitalism, in particular that of the falling rate of profit (see Basu 2015). The other strand of literature tries to apply the theory in search of a quantification of the main variables involved and/or a refinement of the theory itself by means of induction from the data (see, for example, Mariolis & Tsoulfidis 2015).
In all these exercises, a crucial difficulty is finding an acceptable empirical quantity with which to measure, or at least approximate, the theoretical variables. For example, Basu (2015) describes how ‘Marxian national accounts’ have to be produced before empirical analyses of Marxian theory can be carried out. Similarly, as Shaikh (2016) notes, the use of time-series data may be tricky because the notion of reproduction of the system is instantaneous—that is, the surplus (and thus profits) should be measured using both input and output prices of the same period, whereas firms (and national accounts) measure profits on the basis of historic costs.
Most efforts are usually made to correct national accounting aggregates, for example transforming value added into surplus value by deducting workers’ necessary expenses. However, crucial difficulties lie in the quantification of prices and/or values too. For example, since Phillips (1966) it is obvious that waste (in the sense of non-socially necessary expenditure) should be deducted too. Moreover, since different theoretical approaches embrace different concepts of regulating prices, they need different empirical methods. For example, the straightforward use of observed input-output matrixes proposed as if they represented Sraffian systems is highly problematic, because it implies the assumption that the average technology used in each industry is the best technology given prices and income distribution, as well as several other assumptions concerning fixed capital and capacity utilization. Evidently, acceptance of these assumptions depends on one’s opinion on the concept of long-run positions and of the convergence process, and those Sraffians who withhold the snapshot view will consider empirical testing of the theory virtually impossible.
In conclusion, the burgeoning empirical literature on price theory is showing great potential for cross-fertilization and even integration across several heterodox approaches. But as is the case of the methodological and theoretical literatures, conceptual, analytical, and empirical hurdles still remain. Given these cognitive costs and the obviously lower professional rewards, if marginalism was correct, all (utility-maximizing) economists would prefer to be mainstream.
I am grateful to Scott Carter, Lynne Chester, Massimo Cingolani, Tae-Hee Jo, Maria Cristina Marcuzzo, Alessandro Roncaglia, and Bertram Schefold for comments and criticisms on previous drafts of this chapter. All remaining errors are mine.
1 Recent surveys on the topic may be found, for example, in the June 2015 Symposium of the Review of Radical Political Economics; the 2012 Special Issue, Vol. 47, No. 2, of the Cambridge Journal of Economics; or the three volumes edited by Levrero, Palumbo, and Stirati (Levrero et al. 2013).
2 Most of the terms and conceptualization used in this chapter are drawn from Roncaglia (2005). However, in several cases historians of economic thought have not yet found a definitive agreement on the interpretation of the thought of several scholars cited here, and even heterodox scholars may disagree with some of the interpretations put forward in what follows.
3 This conceptualization of price best applies to Hayek. The early Austrians (such as Böhm-Bawerk) and later Hicks were rather concerned with the role of prices in establishing the necessary link between economic activities through time. The emphasis on the methodological and analytical role of time is fully pertinent to the theory of value, and recognized as such by many heterodox approaches (see Cohen & Harcourt 2003).
4 Similarly, prices increasingly play an overall different, purely methodological role in mainstream microeconomics, most clearly highlighted by the ‘Chicago price theory’ (Weyl 2015) and its burgeoning empirical applications. Here prices are a mere mathematical formulation with no specific conceptual meaning: they descend from the conceptualization of any social issue as an optimization problem, in which a market price or a shadow price is mathematically the adjustment variable leading to equilibrium. Accordingly, there can be ‘implicit’ prices for anything, irrespective of the assumption of actual or potential exchanges.
5 From this point of view, Mirowski (1989) defines the objective approach as a ‘substance’ theory of value since it conceptualizes value as a property of the commodities (and labor power) exchanged. This view best fits the Marxian approach, while several Sraffian economists would object to it and, instead, conceptualize exchange-value as a property of the whole system of production.
6 However, even if prices are measures of relative value in exchange, their changes affect the absolute size of aggregate values because they denote changes in the technology and/or in the distribution of income (Shaikh 2016).
7 It thus retains relevance for the quantification, for example, of the surplus. However, its role as a theory of prices is lost.
8 For example, Pasinetti (1981) is bound to take a more restrictive approach, founding his analysis on the view that not only production prices but also the distributive variables are ‘natural’ in the sense that they are independent of social institutions and can be determined at the level of pure logic: a position that obviously runs counter to institutionalist insights.
9 These exercises typically find few examples of capital paradoxes, and gave birth to a literature dedicated to the explanation of the supposed limited empirical relevance of the Cambridge critique of the marginalist theory of value. However, to some extent these results are driven by the very data chosen since, as shown by Bharadwaj (1970), the maximum number of switching points between two hypothetical systems is equal to the number of different basic commodities (without double counting).
Andrews, D. 2015. ‘Natural price and the long run: Alfred Marshall’s misreading of Adam Smith.’ Cambridge Journal of Economics, 39 (1): 265–279.
Aspromourgos, T. 2004. ‘Sraffian research programmes and unorthodox economics.’ Review of Political Economy, 16 (2): 179–206.
Baran, P.A. & Sweezy, P. 1966. Monopoly Capital: An Essay on the American Economic and Social Order. New York: Monthly Review Press.
Barba, A. & de Vivo, G. 2012. ‘An ‘unproductive labour’ view of finance.’ Cambridge Journal of Economics, 36 (6): 1479–1496.
Basu, D. 2015. A selective review of recent quantitative empirical research in Marxist political economy. University of Massachusetts Amherst, Department of Economics Working Paper 2015–05. Available from https://www.umass.edu/economics/publications/2015–05.pdf [Accessed August 30, 2016]
Bharadwaj, K. 1970. ‘On the maximum number of switches between two production systems.’ Schweizerische Zeitschrift für Volkwirthschaft und Statistik, 106 (4): 409–429.
Bellofiore, R. 2012. ‘The ‘tiresome objector’ and Old Moor: a renewal of the debate on Marx after Sraffa based on the unpublished material at the Wren Library.’ Cambridge Journal of Economics, 36 (6): 1385–1399.
Braudel, F. 1967. Civilisation matérielle, économie et capitalisme, XVe–XVIIIe siècle. Vol. 1: Les structures du quoti-dien. Paris: Colin.
Cohen, A.J. & Harcourt, G.C. 2003. ‘Retrospectives: whatever happened to the Cambridge capital theory controversies?’ The Journal of Economic Perspectives, 17 (1): 199–214.
Corsi, M. & D’Ippoliti, C. 2013. ‘The productivity of the public sector: a classical view.’ PSL Quarterly Review, 66 (267): 403–434.
D’Ippoliti, C. 2011. Economics and Diversity. London and New York: Routledge.
D’Ippoliti, C. & Roncaglia, A. 2015. ‘Heterodox economics and the history of economic thought,’ in: T.-H. Jo & Z. Todorova (eds.), Advancing the Frontiers of Heterodox Economics: Essays in Honor of Frederic S. Lee. London: Routledge, 21–38.
Folbre, N. 1982. ‘Exploitation comes home: a critique of the Marxian theory of family labour.’ Cambridge Journal of Economics, 6 (4): 317–329.
Hayek, F.A. 1948. ‘The meaning of competition,’ in: Individualism and Economic Order. Chicago: University of Chicago Press, 92–106.
Jo, T.-H. 2016. ‘What if there are no conventional price mechanisms?’ Journal of Economic Issues, 50 (2): 327–344.
Lavoie, M. 2011. ‘Should Sraffian economics be dropped out of the Post-Keynesian school?’ Économies et Sociétés, 44 (7): 1027–1059.
Lee, F.S. 1998. Post Keynesian Price Theory. Cambridge: Cambridge University Press.
Lee, F.S. & Jo, T.-H. 2011. ‘Social surplus approach and heterodox economics.’ Journal of Economic Issues, 45 (4): 857–875.
Levrero, E.S., Palumbo, A., & Stirati, A. 2013. Sraffa and the Reconstruction of Economic Theory: Volume One. Basingstoke, UK: Palgrave Macmillan.
Mariolis, T. & Tsoulfidis, L. 2015. Capital theory ‘paradoxes’ and paradoxical results: resolved or continued? MPRA Working Paper 68214. Available from https://mpra.ub.uni-muenchen.de/68214/ [Accessed August 30, 2016]
Mill, J.S. [1848] 1871. Principles of Political Economy, With Some of Their Applications to Social Philosophy. London: Parker.
Mirowski, P. 1989. More Heat than Light: Economics as Social Physics, Physics as Nature’s Economics. Cambridge: Cambridge University Press.
Panico, C., Pinto, A., & Anyul, M.P. 2012. ‘Income distribution and the size of the financial sector: a Sraffian analysis.’ Cambridge Journal of Economics, 36 (6): 1455–1477.
Pasinetti, L.L. 1981. Structural Change and Economic Growth: A Theoretical Essay on the Dynamics of the Wealth of Nations. Cambridge: Cambridge University Press.
Phillips, J.D. 1966. ‘Appendix: estimating the economic surplus,’ in: P.A. Baran & P. Sweezy (eds.), Monopoly Capital: An Essay on the American Economic and Social Order. New York: Monthly Review Press, 369–391.
Picchio, A. 1992. Social Reproduction: The Political Economy of the Labour Market. Cambridge: Cambridge University Press.
Roncaglia, A. 2005. The Wealth of Ideas: A History of Economic Thought. Cambridge: Cambridge University Press.
Roncaglia, A. & Tonveronachi, M. 2014. ‘Post-Keynesian, post-Sraffian economics: an outline,’ in: D.B. Papadimitriou (ed.), Contributions to Economic Theory, Policy, Development and Finance: Essays in Honor of Jan A. Kregel. Basingstoke, UK: Palgrave Macmillan.
Schefold, B. 2016. Marx, the production function and the old neoclassical equilibrium: workable under the same assumptions? With an appendix on the likelihood of reswitching and of Wicksell effects. Centro Sraffa Working Paper 19. Available from http://www.centrosraffa.org/public/6c384fc3-ba7a-4b1d-858c-75db58468eec.pdf [Accessed August 30, 2016]
Schumpeter, J.A. 1954. History of Economic Analysis, edited by E. Boody Schumpeter. New York: Oxford University Press.
Shaikh, A. 2016. Capitalism: Competition, Conflict, Crises. New York: Oxford University Press.
Sraffa, P. 1960. Production of Commodities by Means of Commodities. Cambridge: Cambridge University Press.
Steedman, I. 1977. Marx after Sraffa. London: NLB.
Stiglitz, J. & Grossman, S. 1980. ‘On the impossibility of informationally efficient markets.’ American Economic Review, 70 (3): 393–408.
Sylos Labini, P. [1956] 1962. Oligopoly and Technical Progress. Cambridge, MA: Harvard University Press.
Todorova, Z. 2009. Money and Households in a Capitalist Economy. Cheltenham, UK: Edward Elgar.
Weyl, E.G. 2015. ‘Price theory.’ Journal of Economic Literature, forthcoming. Available from http://ssrn.com/abstract=2444233 [Accessed August 30, 2016]