The question of value is one of the oldest and most prominent debates in economics, but also one that has been largely neglected by the mainstream in recent decades. For the classical political economists, value theory was the foundation on which the rest of their work sat: Smith, Ricardo and Marx all began their major respective works with an examination of value. Together, these three giants of political economy developed a theory of value that postulated that the normal exchange rate of commodities to one another was governed by the amount of labor embodied within that commodity. This approach to value was assailed in polemics by subjective value theorists such as Manger, Senior, Jevons and Walras who emphasized marginalism. In an attempt to reconcile the more cost centric supply-side account of value with marginalist demand side approach, Marshall formulated what would eventually become neo-classical economics. This theoretical framework has come to dominate, and the question of value has been deemed settled in economics outside of a few heterodox debates. The classical theory of value, besides the objections of the subjective value theory, was thought to be debunked by its own internal contradiction that came to light with Marx’s transformation problem. However, is this question really settled? There have been developments in classical political economy which may give the labor theory of value (LTV) the upper hand as a scientific theory of prices over the neo-classical economic framework, which has remained largely unchanged over the years in its conceptual basis. These theoretical developments, rooted in a probabilistic approach to equilibrium and economic variables, has far reaching implications for the field.
To start at the beginning, Adam Smith took on the question of prices and what determined them in the Wealth of Nations. Smith points out, after outlining the great efficiencies that the division of labor afforded, that the real value of goods bought in a market is the labor that the buyer is commanding in order to have his needs and wants met by other people. He also suggests that this formulation of value is not just an axiomatic one, and that at least in the “early and rude state of society” when labor was the sole input the value of commodities relative to one another was directly determined by the amount of labor normally used to create them. Thus, in such a situation, the price of the commodity would be exactly the same as the wages paid to the laborer for their work. This qualification of “normal” is also important here, as it applies not only to the labor involved but the prices generally. Smith is careful to point out that actually existing market prices may be above or below “natural prices” at any given time, those prices around which market prices gravitate as an equilibrium[1]. Smith’s model for the labor theory of value only applies to this rude state of society, his general theory of prices in the long term equilibrium is that of cost based around his theories of wages, profits and rent, a theory which is left ununified.
Ricardo takes Smith’s LTV and moves it a step forward. He says that the determination of value by labor isn’t equal to the value of labor itself, the exchange rate of goods is proportional to the amount of labor in those goods. Ricardo qualifies that LTV is only applicable to those goods which are reproduced by labor, and that while he rejects utility as the basis for understanding the magnitude of value, he maintains that a good having some demand, some utility, is a precondition for it to have value[2]. With some minor variations, this is the same framework that Marx would employ in his analysis of capitalism, and which he carried to its logical conclusion. The separation between the value of a commodity, which is determined by the labor embodied in it, and the value of the labor which created it, leads to the problem of “exploitation” in Marx. He points out that what a capitalist buys when they are paying a wage is a worker’s labor power, the ability for workers to work, but what they get as a use from this purchase is labor time and its this labor time which determines value[3] (Marx 135). Because the value of labor power is determined by the cost of the commodities required to reproduce the worker’s labor, the cost of subsistence, and the labor time is only how much the worker works to produce a unit of the commodity, this means that capitalists must work their workers to add more value than their cost of subsistence in order to have surplus value. Surplus value therefore determines the slice of the pie for profit and rent.
It may be worthwhile to reflect on why labor has such a central place in the classical school. Labor has a universal property to it; unlike other animals like bees or elephants, humans are capable of a wide variety of tasks and most humans can be trained to do a number of specialized tasks that they previously did not know how to do. Secondly, as an input, any commodity that is reproduced must have labor in it somewhere, and if we want to produce more of something generally we will need to use more of labor (Cockshott et al. 123). Not to mention, as Smith suggested, when we are buying goods on a market, we should generally think about it in terms of coordinating the social division of labor.
Returning to Marx and Ricardo, there comes a problem as they further develop their theories of value: profit is generally accounted for in terms of money and in a cost based theory of prices, profit must be accounted for above costs. Ricardo and Marx, along with many other thinkers even more recently, presume a uniform rate of profits across all industries. The reason for this assumption is articulated clearly by Ricardo:
It is to be understood that I am speaking of profits generally. I have already remarked, that the market price of a commodity may exceed its natural or necessary price, as it may be produced in less abundance than the new demand for it requires. This, however, is but a temporary effect. The high profits on capital employed in producing that commodity, will naturally attract capital to that trade; and as soon as the requisite funds are supplied, and the quantity of the commodity is duly increased, its price will fall, and the profits of the trade will conform to the general level.[4]
This creates a problem. The ordinary LTV is premised on the idea that profit is the result of surplus value generated by active labor as opposed to other inputs, so the more capital an industry has, the more the capitalist pays up front, and the less they will receive in profits as a ratio to costs. Ricardo dismisses this problem as largely being unimportant and believed that LTV survives despite it. Marx however, attempted to address this problem in Volume 3 of Capital, a volume which was put together posthumously from his notes. To solve the issue, Marx attempted to transform the labor values of commodities into prices of production by effectively relegating the issue of surplus value to that of a national accounting identity. However, even in making this admission he largely failed in the overall attempt: it was later discovered that Marx’s postulations of total price equaling total value were mathematically incompatible with his idea that total profit equals total surplus value so long as the assumption of uniform profits held.
Given these serious issues, let us examine the contemporaneous opponents to LTV. Jevons, Walras, Senior and Manger. Each of them proposed that value was to be understood as the result of marginal utility. That is, they believed that the value of a good was solely determined by the utility, use or desire for it and its relative scarcity. These arguments are nothing new, Aristotle spoke of use values, the framework of moral law before Smith was largely the same, and Smith himself pointed out that this is largely what governed price movements in the short term with his discussion of “effectual” supply and demand. What these thinkers did was simply formalize the issue with the addition of marginalism, that rational actors think on the margin and demand curves are governed by a diminishing rate of marginal utility. The principal thrust of these theorists in regards to value, besides Walras, may be that the costs associated with supplying a good to market are not the basis for value, but the consequence of it. However, this argument also has big problems: it cannot reckon with the simple fact that when it is easier to acquire a good the price of it goes down even if the demand and utility remain the same.
I single out Walras among the other subjective value theorists because for him, the major issue was not marginal utility, but general equilibrium, and he only adopted marginal utility as a means of completing his framework for general equilibrium[5]. This difference is important as when this was combined with the updates of Marshall’s neo-classical framework general equilibrium became the only basis with which the neo-classical system could generate a theory of prices without reference to prices themselves.
What might become clear with the differences in the approach between marginal utility and the cost-based theories of the classical school is the former lends itself well towards short term analysis while the latter lends itself well for long term analysis. Smith, as was mentioned, already pointed out that in the short term price was governed by utility and scarcity alone, while all the LTV writers were focused on an “natural prices” as a long term equilibrium. At stake was what value was, if value was merely the same as actually observed prices, then the marginalists appeared to have the upper hand. The actual market clearing prices should logically be determined by the supply and the demand at any given moment. But the classical political economists also asked “what determines the point at which supply and demand meet,” in other words, what determines the equilibrium. It was this equilibrium, the “normal, average and natural” price that the classical theorists considered to be value.
Marshall’s project is one that attempts to formally unite these two methods of analysis in acknowledging that while prices are determined by costs in the long term, they are merely a function of supply and demand in the short term. Unlike the subjective value theorists he goes through a thorough analysis of costs that determine the supply curve in the Marshallian cross; even in the short term the supply can usually be adjusted according “prime costs” those costs besides “supplemental costs”. These are those costs which are variable in the short term such as labor and raw materials, while opposed to the costs of land, buildings and machines[6].
While Marshall presents himself as a continuation of the same project as Smith and Ricardo, he distances himself from key parts of their framework. Firstly, he establishes “economics” as something distinct from political economy, starting with the title of his book. Whereas political economy was the study of the capitalist economic and political system, its tendencies, laws and characteristics, Marshall, in the guise of expanding the focus of the classical school, actually narrowed it down to the study of “free enterprise” and its variations. This would be a field of study consumed with the operation of markets alone and the way humans interacted with them in the abstract. Marshall also rejects key parts of the classical school of thought, namely LTV, and the notion of equilibrium that holds “average,” “ordinary,” “normal,” and “natural” prices to be the same. In contrast to this notion of equilibrium, Marshall begins from an understanding of a mechanical balance between desire and effort being equilibrium, and he suggests that equilibrium of normal prices is only the special case of equilibrium of normal supply and demand. This is different from the classical approach, which took the normal price to be the equilibrium price. Marshall, quite dramatically, dismisses the idea that any equilibrium can be found even in the long term average price, believing that the complexities and disruptions of the real world would make this tendency impossible to be observed. In order to investigate this long term equilibrium he takes the classical notion of the “stationary state” which initially referred to a stable economic situation which we were slowly moving towards and instead uses it as an ideal ceteris paribus case[7]. He notes that this state doesn’t require perfect staticness, as the average of values can be maintained in a system so long as the dynamics that produce the whole range of values remains the same. He suggests that this average can be considered representative of firms in an economy, and that in such a state where everything were held equal costs would necessarily govern prices both in the long and short term. The essence of this analytical device and its purpose is expressed in the following comment:
But an answer may be given here to the objection that since “the economic world is subject to continual changes, and is becoming more complex, … the longer the run the more hopeless the rectification ” : so that to speak of that position which value tends to reach in the long run is to treat “variables as constants.” (Devas, Political Economy, Book IV. ch. v.) It is true that we do treat variables provisionally as constants. But it is also true that this is the only method by which science has ever made any great progress in dealing with complex and changeful matter, whether in the physical or moral world.[8]
Here, Marshall commits a grave and yet very common error. The last sentence of his defense of the stationary state and his use of ceteris paribus is categorically false, for it is the case in both the “physical and moral world” that science has made immense progress using non-constant variables. In actuality, the physics of gasses, which deals with the energy of loosely interacting atoms, has been home to a very different kind of analysis. The statistical mechanics of Maxwell and Boltzmann show that in order to understand such systems, rather than assuming constant mechanical properties of the atoms and molecules, you must use a stochastic variable that represents that probability distribution of the energy levels of the different particles.
This is the premise of one of the first works of “econophysics”, Laws of Chaos, a book written by Israeli Marxian economists Farjoun and Machover. In it, they point out that while notions of equilibrium and natural prices are necessarily statistical phenomena, and their statistical basis is expressed in writing in nearly all major works, this statistical nature is immediately excluded in any of the formal mathematical models[9]. Indeed, this is exactly how Marshall proceeded, for while he points out that such constants can represent constantly shifting situations due to them maintaining the same distribution of values, he leaves out any discussion or mathematical notion of how these distributions could affect his models.
The difference between these two understandings of equilibrium is subtle, but signifigant. Farjoun and Machover discovered that any model that does not take such considerations is in effect committing a fallacy, they point out “a mathematical relation that holds among variable quantities does not, in general, hold between their respective averages[10]”. The prime example they give is that of profit rates, for while two firms may have identical average rates of profit over a given period of time, it may also be the case that at any given point in time the actually observed rates of profit between the two firms is far apart. Attempting to tax both of the firms according to their average profit rates or their actual profit rates on a year by year basis will thus produce widely different results.
Farjoun and Machover began their inquiry into economic equilibrium by approaching the transformation problem, and the issue of uniform profit rates especially. Neo-Ricardian economists such as Sraffa have attempted to preserve such assumptions in their theory of prices, at the expense of the LTV which was relegated to a special case of the Ricardian framework. Farjoun and Machover challenge this assumption, and not just on the grounds that it is impractical for reality to align with a theoretical equilibrium. Rather, they suggest that a theory of competition that posits uniform rates of profit is horribly incomplete. This theoretical objection is based on three ideas, one mathematical and two economic. The mathematical objection is based on the stability of uniform profit rates at equilibrium; it is entirely possible that such an assumption may not be approximately true when equilibrium is approximately achieved, even if it is true in a perfectly competitive equilibrium. The economic objection is premised on the nature of competition and empirical fact. They note that the distribution of profit rates is at least as wide as other important economic variables such as wages to capital invested, while other variables which receive far less examination, such as the ratio of wages to profits are much more narrowly distributed[11] .
The final economic objection is far more far reaching; they take aim at the metaphor consistently used by economists from Smith to Marshall to describe equilibrium, that of a pendulum moving back and forth. The self-correcting mechanism of prices and, in this case, profits, is based on the premise that once at the equilibrium only an external force can move the pendulum away from its stable point. This force which drives the automatic market response to changes in supply and demand is competition. But hold on, isn’t it also the case that competition is a force that not only causes prices and profits to fall towards equilibrium, but also away from them? They describe a though experiment, assuming an economy where prices and/or profits were set at their equilibrium levels for every commodity and every firm, all else being equal, would you expect the force of competition to have zero impact on those prices and profits? The answer is no, competition would almost certainly cause some firms to change their prices, even at the expense of profits, to out-compete other firms.
While I described Farjoun and Machover as a project in “econophysics” they are rarely recognized for their extremely early work in this area, and more recently most of the mainstream work in the field has been focused on finance, mostly due to computer science and physics students working for financial firms[12]. However, while the connection to physics and the math of Maxwell and Boltzmann is made explicit here, it also seems to be the case that the marginalists and the neo-classical school were all founded on classical Newtonian physics, in particular as the extension of the rationalist, individualist project represented by them. In this sense, Farjoun and Machover and their few students represent a very unique niche of “classical econophysics”, one which examines the foundations of the whole of economic thought with a relatively new lens of formally stochastic and probabilistic equilibrium.
Returning to value, it is also the case that any thorough theory of value must also be a theory of distribution, as the refinements between Ricardo and Smith suggest. Marshall advances a theory that utilizes Ricardo’s theory of differential rents in the context of factor costs which today is referred to as “marginal productivity”. This theory, which he terms “quasi rent”, suggests that the costs associated which each factor, such as fixed capital or labor, is determined by the least productive unit employed. In this sense, it is also premised on a diminishing marginal productivity of a given factor as it is scaled up while all else is held equal[13]. With factor prices dealt with, all that remains is profit, which is an altogether trickier matter. As for Ricardo himself, besides his theory of rent and subsistence wages, he considered profit merely to be a residual between prices and costs. Marx described a multitude of factors which may determine profit and surplus value, including the length and intensity of the working day.
Considering that Marshall based his dismissal of LTV, like Senior before him, on his theory of profits as the reward for abstinence, this should also be a key part of his theory of distribution. Specifically, Marshall says that it is absolutely wrong to consider goods to be the result of labor alone, rather, they are the result of “labor and waiting[14]”. Unlike other authors, Marshall doesn’t do much to distinguish between money interest on loans and profit, nor, for that matter, profits from ownership of capital and wages to supervisory workers/managers. This creates some conceptual issues, most notably that his defense of profit as a reward borderlines a normative argument rather than a scientific one, which is precisely what he accuses theorists such as Marx of. Afterall, with different structural institutions within firms and forms of property, such incentives need not appear like profit, as the difference between costs and prices. What’s more, if we are to take seriously theoretical and empirical models of a falling rate of profit, or even the near zero interest global interest rate levels of a couple years ago, it is obvious that there are situations where there may be no or nearly no such reward. But to treat his position generously we can associate it with the ideas of Kalecki and Keynes which reveal that, as a national accounting identity, savings should be equal to investment and profit[15]. Even with this identity, there is not much in Marshall’s accusation that refutes Marx’s claims about LTV. This is because Marx, like Ricardo, only claims that prices are proportional to labor time, unless Marshall could show that profit is uniform or disproportionate to the labor embodied in goods, he is not disputing the central thesis that labor time is the determinate of value.
In fact, Marshall goes on to do the exact opposite. Seemingly unaware of the issue of uniform rates of profit in relation to Marx’s transformation problem, he goes into great detail in the section almost immediately after he objects to Marx’s LTV about how profit rates are proportional to the capital intensity of a given industry[16]. Specifically, he cites how higher upfront capital costs will lead to lower profits, and how industries with low prime costs relative to their supplemental costs will have lower profits due to occasional production in the short run being driven by only the prime costs. This would seem to support Marx’s argument that profit rates and surplus value are only created by the “living labor” or variable capital, as those industries with the highest worker to capital ratio or lowest capital intensity would have the highest profits. This seems to be exactly what the empirical research shows, confirming Marshall’s suspicions about profit, as well as Marx’s Capital Vol 1 formulations and Forjoun and Machover’s assertions[17].
Marx himself anticipated attacks in the vein of Marshall’s “labor and waiting” and his response was rooted in a thought experiment, tracking a hypothetical capitalist attempting to produce a profit. This capitalist bought all the raw materials and machines required to produce yarn and paid his workers a subsistence wage. The capitalist then had the workers work a certain amount of time that happened to be equivalent to the amount of time required to add enough value, make enough yarn, equal to the value required for their necessary subsistence. In this scenario, no profit could be produced, regardless of whether the capitalist sacrificed something in order to provide the capital, or whether what he did was socially useful[18]. In order for the capitalist to make a profit, he must have the workers produce more than the consumption required for their reproduction, this is a fundamental fact about how capitalism works and which is difficult to deny. Marx’s point is quite simple, the creation of this surplus is a choice, and specifically, it is a choice about how long and how hard laborers work, hence why he describes it as exploitation.
In addition, it’s also been pointed out that if we are to do a total accounting of value based on cost, profit may also be understood as the cost required for the reproduction of the capitalist class[19]. Whether capitalists are saving/investing or consuming with their money, each of these things are necessary to ensure that capitalism, as a class system, continues to reproduce itself. One the one hand, it is true that capitalists choose whether to consume or to save, and how much they save determines the magnitude of profit for the whole economy. But if they choose only to consume, there will no longer be a capitalist class. Thus, on the macro level, and especially for the biggest capitalists who would likely have quite some difficulty consuming all their money in the first place, there must be savings. If not, the system would fail to reproduce and capitalists would loose their class status along with the luxury of not having to work to survive.
Thus, the capitalist class is confronted with two choices which determine the distribution in an economy. They must choose how much to save, which will determine the magnitude of profit on a macro scale, but not value in terms of relative exchange values. And they must choose how hard and how long to drive workers, partially determining the rate of surplus value and thus the value and profit of a given commodity and industry, respectively.
As I write, I find myself curious to the extent of differences between surplus value produced via workers and savings, that is the differences between profit ex ante and ex post, something which may be dependent on a theory of money and inflation over the course of a production cycle. This is, however, beyond the scope of this essay, even if it is extremely pertinent to the violation of Say’s Law and Marxian crisis theory.
It should be clear by now that each of these different theories of value should generate testable predictions, as far as the classical vs neo-classical schools go. What few propositions the pure subjective value theorists concocted, such as Jevon’s “two fundamental propositions[20]”, do not seem falsifiable to any signifigant degree. With regard to the serious schools of thought: neo-classicals predict that marginal productivity is the basis for distribution among factors, while classicals predict that prices should be proportional to the labor time used in production.
The empirical research on marginal productivity for the basis of income is decidedly mixed. Early studies which only fit a Cobb-Douglas production function to manufacturing data found only small deviations from marginal utility predictions. Later studies using more sophisticated methods, however, found anywhere from moderate to high deviations, usually with capital receiving higher than expected income, or in places with high organized labor power such as France, labor receiving more income[21].
In contrast, in studies of input-output analysis where wage bills were used to estimate amount of hours worked it was found that labor values were incredibly closely correlated (94.2 to 98.6%) with prices by sector, and this result was confirmed across many different countries. Using the same analysis for other widely used inputs does not produce anywhere close to the same level of correlation[22].
It was mentioned in passing that general equilibrium theory, unlike Marshallian cross analysis, was capable of explaining prices without reference to prices themselves, and thus saves the neo-classical school from a potentially embarrassing criticism of circular logic. This form of analysis, however, suffers from large difficulties in achieving a unique and stable equilibrium solution. Attempts to operationalize this kind of analysis in the 70s and 80s by computing solutions to general equilibrium have largely fallen out of fashion and are only used as methods of forecasting the effects of certain policy proposals. Because of this particular use of the models, there hasn’t really been any attempt to empirically test them, and those models which have attempted dynamic simulation are not grounded in empirical reality. Whether such a system can produce workable predictions on value remains to be seen.
The new developments in classical econophysics, a school of thought which began in 1983 with the publication of Laws of Chaos and was updated with theoretical developments and empirical research by a new wave of writers with Classical Econophysics in 2009, have largely been overlooked by the mainstream. However, in applying statistical mechanics frameworks to equilibrium and looking back at the very foundations of economic thought, they are effectively on the cutting edge of theoretical developments in political economy. More serious attention should be directed at these issues and hopefully more research will be done to settle these questions. The potential impact of classical econophysics on all manner of important economic fields is enormous, as are their innovations in value theory. Knowing the distribution of profit rates in an economy, for example, can be used to easily calculate what the effect of average earning growth or other external shocks would be on the profitability of firms on the margin; a way to model crisis like never before. It is effectively the most radically simple and most radically accurate form of materialist analysis we can conduct on capitalist economies.
[1] Adam Smith, Wealth of Nations Chapter v
[2] David Ricardo, Principles of Political Economy Chapter 1. On Value
[3] Karl Marx, Capital Volume I Chapter 7: The Labour Process and the Process of Producing Surplus Value
[4] David Ricardo Principles of Political Economy Batoche, 2001.Pg 77
[5] Jaffé, William. “MENGER, JEVONS AND WALRAS DE‐HOMOGENIZED.” Economic Inquiry
[6] Alfred Marshall, Principles of Economics Palgrave Macmillan, 2013.Pg 299
[7] Alfred Marshall, Principles of Economics Chapter III
[8] Alfred Marshall, Principles of Economics Palgrave Macmillan, 2013. Pg 315
[9] Farjoun and Machover, Laws of Chaos Chapter Two
[10] Farjoun and Machover, Laws of Chaos Verso, 1983. Pg 30
[11] Farjoun and Machover, Laws of Chaos Chapter One
[12] spiritofcontradiction.eu/rowan-duffy/2013/02/06/interview-paul-cockshott-on-econophysics-and-socialism
[13] Alfred Marshall, Principles of Economics Palgrave Macmillan, 2013. Pg 351
[14] Alfred Marshall, Principles of Economics Palgrave Macmillan, 2013. Pg 487
[15] Foster, Gladys P. “Keynes and Kalecki on Saving and Profit: Some Implications.”Journal of Economic Issues, vol. 24, no. 2, 1990, pp. 415
[16] Alfred Marshall, Principles of Economics Chapter VIII
[17] Nils Fröhlich, Labour values, prices of production and the missing equalisation tendency of profit rates: evidence from the German economy, Cambridge Journal of Economics, Volume 37, Issue 5, September 2013
[18] www.marxists.org/archive/marx/works/download/pdf/Capital-Volume-I.pdf Pg 134
[19] http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.646.4495
[20] Jaffé, William. “MENGER, JEVONS AND WALRAS DE‐HOMOGENIZED.” Economic Inquiry Pg 517
[21] Biewen, Martin, and Constantin Weiser. “An Empirical Test of Marginal Productivity Theory.” Applied Economics 46.9 (2014)
[22] Cockshott et al, Classical Econophysics. Chapter 10.4
bunkermundo 
Leave a comment