Andrew Kliman and the ‘Neo-Ricardian’ Attack on Marxism, Pt 1
[The following is the first of a two-part reply to a reader’s question. Since the reply had to be broken into two parts due to its length, part 2 will be posted two weeks after this part appears. My plan is to return to a monthly schedule after that.]
A while back a reader asked what I thought about the work of Andrew Kliman. Kliman is the author of a book entitled “Reclaiming Marx’s ‘Capital,'” published in 2007. In this book, Kliman, a professor of economics at Pace University, attempts to answer the claims by the so-called “neo-Ricardian” economists that Marx’s “Capital” is internally inconsistent. According to the “neo-Ricardians,” Marx was not successful in his attempts to solve the internal contradictions of Ricardo’s law of labor value.
The modern “neo-Ricardian” school is largely inspired by the work of the Italian-British economist and Ricardo scholar Piero Saffra (1898-1983). But elements of the “neo-Ricardian” critique can be traced back to early 20th-century Russian economist V. K. Dmitriev. Other prominent economists and writers often associated with this school include the German Ladislaus von Bortkiewicz (1868-1931) and the British Ian Steedman.
The Japanese economist Nobuo Okishio (1927-2003), best known for the “Okishio theorem”—much more on this in the second part of this reply—evolved from marginalism to a form of “critical Marxism” that was strongly influenced by the “neo-Ricardian” school.
In the late 20th century, the most prominent “neo-Ricardian” was perhaps Britain’s Ian Steedman. While Sraffa centered his fire on neoclassical marginalism, Steedman has aimed his at Marx. His best-known work is “Marx after Sraffa.” The “neo-Ricardian” attack on Marx centers on the so-called transformation problem and the Okishio theorem.
The Okishio theorem allegedly disproves mathematically Marx’s law of the tendency of the rate of profit to fall. The transformation problem is more fundamental than the Okishio theorem, since it involves the truth or fallacy of the law of labor value itself. I will therefore deal with the transformation problem in the first part of this reply and the Okishio theorem in the second part. However, Andrew Kliman seems to be more interested in the Okishio theorem for reasons that will soon become clear.
I have already dealt with the transformation problem in an earlier reply. But here I will take another look at it in the light of Kliman’s work.
The ‘Temporal Single-System Interpretation’
Kliman is a leader of a school that has developed over the last 40 years in response to the “neo-Ricardian” attack on Marx, called the “Temporal Single-System Interpretation,” or TSSI for short. As a supporter of TSSI, Kliman seeks to demonstrate that it is necessary to look at the formation of labor values dynamically in time. According to Kliman and the TSSI school, the main mistake that the “neo-Ricardians” make is that they seek to calculate input and output prices simultaneously.
Kliman claims that such simultaneous calculations of input and output prices inevitably lead to what he calls “physicalism.” Instead of calculating profits—surplus value—in terms of labor values, the “neo-Ricardians” end up calculating profits in terms of physical quantities of commodities. Or to use more traditional Marxist language, the “neo-Ricardians” end up calculating surplus value in terms of the use values of the commodities consumed by the consumers of surplus value, whether productively or unproductively, and not in terms of the labor values of these commodities.
As Ian Steedman, the author of “Marx After Sraffa,” likes to put it, the rate of profit in terms of prices of production is not determined by the (labor) value rate of profit. Instead, Steedman holds that the (labor) value rate of profit and the rate of profit in terms of prices of production—prices that equalize the rate of profit among the different branches of production—are not the same. Therefore, the rate of profit is not determined by the value rate of profit but rather by the physical conditions of production and the real wage. Steedman therefore came to the conclusion that the whole concept of labor value is meaningless and should be abandoned.
This is what Kliman means by the “physicalism” of the “neo-Ricardian” school. Since the “neo-Ricardians” reject all forms of the law of labor value—Ricardo’s as well as Marx’s—they are forced to reject Marx’s explanation of surplus value. To the extent the “neo-Ricardians” can be said to have a theory of surplus value at all, they see the surplus value as a surplus of physical use values that arise in the process of production.
Our modern “neo-Ricardians” turn the clock of economic thought back to not only before Marx but before Ricardo and before Adam Smith, back to the French Physiocrats of the 18th century. (1) The Physiocrats were the original “physicalists.” They saw surplus value arising physically in the sphere of agricultural production, a view that in their time was a considerable advance over the older view that surplus value arose in the sphere of trade or circulation.
The only real advance—leaving aside the mathematics—of today’s “neo-Ricardians” over their Physiocratic forbearers is that they don’t insist that profit arises only in the sphere of agriculture. However, even here, very much in the manner of the Physiocrats, they often use examples drawn from agriculture and indeed are forced to do so for reasons that will become clear below.
The importance of the issues at stake
Frederick Engels went so far as to say that socialism became a science only with the discovery by Marx that surplus value arose from the unpaid labor of the working class even if labor power was paid at its full value. If the “neo-Ricardians” are right, however, Marx’s theory of surplus value is invalid. Socialism is therefore no longer a science but once again a utopia. Therefore, the issues raised by Kliman are indeed central to economic science and to the future of Marxism. How well he answers the “neo-Ricardian” challenge is, however, another question altogether.
Crisis theory, Kliman and the Okishio theorem
While my posts to this blog center on crisis theory, the transformation problem and the Okishio theorem involve far more basic questions: the theory of value and surplus value and the determination of the rate of profit. This reply, therefore, does not deal with crisis theory but with value theory.
However, in looking at Andrew Kliman’s work on the Internet I get the impression that his real concern is crisis theory—a subject that has been on everybody’s mind since the events of 2008 and their aftermath through which we are now living. In my blog, I have attempted to criticize crisis theory—the aspect of economic theory that seeks to explain why the capitalist economy is periodically hit by economic crises, such as the one that hit with such force in the fall of 2008, in the light of Marx’s value theory. I think Kliman is trying to do the same thing.
However, if the “neo-Ricardian” critique of Marxist value theory is correct, then both Kliman’s work and my own are misconceived. Kliman belongs to the school of crisis theory that explains crises by the tendency of the rate of profit to fall—the insufficient production of surplus value—as opposed to schools of thought that put the emphasis on the realization of surplus value. Kliman’s views on crisis theory are broadly related to the Grossman-Mattick school that I examined and criticized in my main posts. I will not repeat these criticism here but refer the reader to the post.
If Okishio really proved that there is no tendency of the rate of profit to fall, then the rug is pulled out from under the very school of crisis theory that Kliman supports. Therefore, Kliman, I believe, is really interested in the Okishio theorem and how it might be disproved. However, since the “neo-Ricardian” critique involves first and foremost an attack on Marx’s value theory, I must review the transformation problem in the light of Kliman’s proposed TSSI answer to the claims that Marx’s theory of value, like Ricardo’s, is inconsistent and therefore false.
The TSSI approach
According to Kliman, to understand Marx’s value theory we really must use the Temporal Single-System Interpretation, or TSSI, approach. According to Kliman, it is not only “neo-Ricardians,” avowed opponents of Marx’s value theory, but also certain Marxists—Anwar Shaikh, for example—who wrongly support the SSSI (Simultaneous Single-System Interpretation), which holds that the values of inputs and outputs are determined simultaneously.
For example, suppose a lathe is sold for $100,000 by an industrial capitalist to another industrial capitalist. For the industrial capitalist who produces the lathe, it is an output, but for the industrial capitalist who purchases the lathe it is an input. The lathe both as input and output has a price of $100,000.
Kliman holds, if I understand him correctly, that if we build economic models where the input and output prices are the same, two different rates of profit inevitably emerge: the rate of profit in terms of labor values and the rate of profit in terms of prices of production. The “neo-Ricardians” draw the conclusion that the only meaningful rate of profit is the rate of profit in terms of prices of production and throw out the value rate of profit as both redundant and meaningless. After all, no practical business people are interested in or even know the rate of profit in terms of value; they are only interested in the rate of profit in terms of price.
If the the rate of profit were the same in value terms and price of production terms, the “neo-Ricardians” reason, this would show that the value rate of profit determines the price of production rate of profit, and the law of labor value and Marx’s theory of surplus value would be correct. But since this is not the case, it is necessary to abandon Marx’s theory of labor value and surplus value.
Remember, if we can’t answer the “neo-Ricardian” criticisms, virtually all of Marx’s contributions to economic science vanish without a trace, and the claims of Marxist socialism to be a science are refuted.
Kliman holds that the Temporal Single-System Interpretation resolves the contradictions that the “neo-Ricardians” believe that they have located in Marxist value theory. According to Kliman and the TSSI school, what the “neo-Ricardians” leave out is the change of values over time.
Returning to the lathe example, assume that the price of $100,000 represents the labor value of the lathe. Suppose a way is found to produce an identical lathe with half the amount of labor than it took before. The value of a new lathe of the identical type will fall from $100,000 to only $50,000. For simplicity’s sake, let’s assume that the fall in the value of the lathe occurs immediately after the industrial capitalist who uses the lathe as an input has purchased it but before the industrial capitalist has begun to use the lathe up in production.
What is the value of the capital that our lathe-using industrial capitalist advanced when he or she purchased the lathe for $100,000. Is it $100,000 or $50,000? Kliman says it is $100,000, but according to him, “neo-Ricardians” calculating in physical terms will value the lathe at $50,000.
Corn models, or back to the Physiocrats
Kliman explains that the “neo-Ricardians” calculate the rate of profit not in terms of labor value but in terms of physical quantities. In order to do this, we have to compare the inputs to the outputs. The increase of the outputs in physical terms over the inputs is the physical rate of profit. But before we can compare the inputs and outputs quantitatively, the inputs and outputs must first be rendered qualitatively comparable.
In most industries, the use values of commodities that serve as inputs—fixed capital, raw materials and auxiliary materials such as energy—are very different than the use value of the commodity that is produced—the output. But let’s imagine a branch of capitalist industry that by way exception has the same commodity as both an input and output.
So in the spirit of the Physiocrats—the original “physicalists”—the “neo-Ricardians” take examples from agriculture where the inputs and outputs can with a considerable stretch be imagined to be identical. Imagine a capitalist corn farmer who uses corn both as input and output. (2) Another possible example would be a cattle rancher where cattle are both the input and output, but we will stick here to the more popular corn model.
I am assuming a capitalist corn farmer. (3) Don’t capitalist farmers by definition have to purchase the commodity labor power as inputs? If they don’t, they are not capitalist farmers. No problem, the “neo-Ricardians” proclaim, we will have our capitalist farmers pay the workers directly in corn. Presumably our workers have no other needs; they work without clothes, live out in open and eat only corn.
Our capitalist corn farmer begins with a definite quantity of corn, let’s say 100 bushels—an example taken from Kliman’s book. Some of the corn is used as seed corn, the rest is used to pay for the labor power of our workers. Our capitalist farmer employs no plows, no machinery, no motive power beyond human labor whether animal or fossil fuels or electricity. No fertilizer or pesticides of any kind or improved land is used. (4) Now, having made all these totally unrealistic abstractions, we have rendered both the inputs and outputs qualitatively identical making it possible to calculate the “profit” of our “capitalist” farmer in physical terms.
At the beginning of the growing season, our capitalist farmer begins with a “capital” of a hundred bushels of corn. He will use some of his corn as seed corn and some as wages to pay his workers. The capitalist farmer has begun with 100 bushels of corn and harvests, to use Kliman’s example, 120 bushels of corn. The mass of “profit,” calculated in physical terms, is 20 bushels of corn, and the rate of profit on an initial “capital” of 100 bushels of corn is 20 percent.
If the real wages rise—remember, all wages are paid in corn—this means the workers get higher corn wages. More corn would then have to be advanced to earn a “profit” of 20 bushels of corn. So the rate of “profit” would be lower, since the 20 bushels of “profit” would have to be calculated on a larger corn capital. Conversely if the real wage falls—the workers get less corn—the rate of profit rises, assuming a corn “profit” of 20 bushels, since the corn “profit” will be calculated on a smaller corn “capital.”
Assuming the capitalist farmer advances a capital of 100 bushels of corn and harvests 120 bushels of corn, we have a mass of profit of 20 bushels of corn and a rate of profit measured in physical terms of 20 percent. But, Kliman asks, what happens if the value of the corn falls because less labor time is necessary to produce the corn than before?
Instead of selling for $1 per bushel, let’s say the the price of corn falls to $.80 reflecting the new lower labor value of the corn. But wait a moment, where do prices come from? Haven’t we been dealing with a natural economy where only one product, corn, has been produced? But then again if we are dealing with a natural economy, how can we have a capitalist farmer at all since isn’t capitalism the highest stage of commodity production where labor power has become a commodity? (5) Never mind, we are in the dream world of the professional economist, and in this world this kind of nonsense is allowed.
Using our dream world dollars, the capitalist farmer lays out $100 worth of corn before the price drops and sells the corn for $100 after the price drops. We can measure the profit of the capitalist corn farmer in both physical terms—corn—and in terms of our dream world money.
Now, has the capitalist farmer made a profit or has he not? If we calculate in physical terms, the answer is yes, but if we calculate in value terms or in our special dream world dollars, the answer is no. Our capitalist farmer has just broken even, which is no good at all for a capitalist. Remember, like all capitalists, even dream world capitalists, our capitalist corn farmer must make a profit.
Professor Ian Steedman, calculating in physical terms, reassures the capitalist farmer that he has indeed made a profit. He started out with 100 bushels of corn and has ended up with 120 bushels of corn, a profit of 20 bushels. Next year, it will take only $80 to buy 100 bushels of seed corn, Professor Steedman explains. So after taking account of the negative rate of inflation—or deflation—the capitalist farmer has done quite well for himself.
If we were dealing with a natural economy that produces only use values, this would be true. But since we are supposedly dealing with a capitalist economy here—even if a “very simple one,” as the “neo-Ricardians” like to say—the capitalist farmer would send the professor packing, explaining that though Steedman may be an excellent professor of economics, he has no head for business. The capitalist corn farmer will explain to our Ian that in business we don’t calculate our profits in physical terms, we calculate our profits in money terms. (6)
In contrast to Steedman, Kliman using the TSSI approach, believes that the capitalist farmer advanced $100 in capital at the beginning of the corn growing season and not $80 in capital. Therefore, Kliman believes that the capitalist farmer better value his advanced capital at $100 not $80. The capitalist farmer would say that Kliman, unlike Steedman, seems to have the makings of a man of business. Though he speaks in the jargon of the professional economist, he at least understands that in business we have to have more money when we sell our commodities than we started out with or we have not made a profit.
Frankly, Kliman’s entire presentation of this point is complex and confusing because as we will soon see he lacks any meaningful theory of money and price. Kliman is trying to explain the capitalist economy as though money does not exist. He is not alone in this. Remember, his “orthodox” neo-classical marginalist colleagues hold that money is only a “veil” that can be abstracted away when explaining the operations of the capitalist economy. According to the neoclassical marginalists, money is a mere means of circulation—this is the basis of Say’s Law, which the (bourgeois) economists use to “prove” that economic crises of overproduction such as that of 2008 cannot occur. (7)
According to Marx—and on this point you won’t find a business person who is not a “Marxist”—the formula for capitalist production is M—C—P—C’—M’. It most certainly is not C—P—C’ like it is in the contrived models of the “neo-Ricardians.” Why isn’t C—P—C’—where we leave out money and look only at the real economy—a valid abstraction?
The production cycle must therefore begin with money and end with money. It is only by comparing the quantity of money that he ends up with to the quantity of money he started with that a real world capitalist can determine whether or not he has made a profit. This is why all attempts to explain capitalism with money abstracted, looking only at the real economy, are doomed from the start.
Kliman therefore draws the conclusion that if you use the SSSI approach—that is, if you ignore the changes of value through time—you end up calculating profits in physical terms, abandoning not only Marx’s theory of value but, far worse, abandoning reality.
But we don’t really need TSSI to avoid this error. All we need is Marx’s theory of exchange value as the form of value, or in plain language Marx’s theory of money, to avoid the error of calculating profits in “physical terms.” Unfortunately, as we will soon find out, Kliman has amputated this absolutely essential part of Marx’s theory of value.
Kliman thinks with TSSI the contradictions that the “neo-Ricardians” claim they have found in Marx’s economic work dissolve. Marx’s work is rendered internally consistent and the “neo-Ricardian” claims are refuted. According to Kliman, TSSI does not actually prove that Marx’s scientific findings are correct—they may or may not be. But if Marx’s work is internally consistent, then it is possible that it may be correct. This is all that Kliman attempts in “Reclaiming Marx’s ‘Capital.'” With the help of TSSI, we can reclaim “Capital” as a work that might actually explain the real world.
I would say that if we replace TSSI (8) with Marx’s complete theory of value, which includes his analysis of exchange value and money as the form of value, Marx’s economic theory is perfectly consistent. In my opinion, the entire “neo-Ricardian” critique is based on the failure to understand Marx’s value theory. I think we can go further and say that the empirical evidence is overwhelming that Marx’s economic theory is in fact true.
It is extremely unfortunate that Kliman wrote “Reclaiming Marx’s ‘Capital'” without fully mastering the theory whose internal consistency he seeks to defend.
In my earlier post, I indicated that I thought that Anwar Shaikh’s very different approach to the transformation problem was correct. Unlike Kliman, Shaikh in his work on the transformation problem employs Marx’s real theory of value (9) and not the amputated version that Kliman uses in his work. Let’s review the transformation problem and its whole meaning for value theory. Then we can compare the approaches of Marx, Shaikh and Kliman.
A brief review of the transformation problem
The transformation problem has deep roots in classical political economy. It centers on the relationship between values determined by the quantity of labor socially necessary to produce commodities and the tendency of free competition in a capitalist economy to equalize the rate of profit among capitals of different organic compositions and turnover periods in such a way that capitals of equal sizes earn equal rates of profit in equal periods of time.
Ricardo, who developed the concept of labor value further and more consistently than anyone before Marx, was stumped by the apparent contradiction between his law of labor value, which holds that the value of commodities is determined by the quantity of labor socially necessary to produce them, and the tendency of free competition to equalize the rate of profit that is yielded by equal capitals in equal periods of time.
Since capitals invested in different branches of production have different durabilities (Ricardo did not have any conception of the organic composition of capital) and if commodities sell at prices determined by labor values, the rate of profit will be unequal between different industries. But Ricardo, like virtually all other economists both before and after him, realized that if rates of profit are higher in some branches of industry than in others, the capitalists always in search of the highest possible profit will move their capital out of branches that are making less than the average rate of profit and into branches that are making more than the average rate.
However, this will mean that the axes around which market prices fluctuate—the prices of production—will deviate from the prices that would prevail if commodities sold at prices that were directly proportional to labor values. This was a particular problem for Ricardo, since unlike Marx he made no distinction between values and prices of production.
While to Marx values represented definite quantities of abstract human labor measured in some unit of time embodied in commodities, prices in contrast were definite quantities of the use value of the money commodity—for example, weights of gold.
Therefore, Ricardo’s theory implied that something other than the quantity of socially necessary labor was determining the values of commodities. Ricardo frankly acknowledged the contradiction in his theory of value and his inability to resolve it. He hoped that a future economist would find the answer to this apparent contradiction. And this is how things stood when Ricardo died prematurely of an ear infection in 1823.
None of Ricardo’s bourgeois followers were able to resolve the apparent contradiction between Ricardo’s labor law of value and the tendency of free competition to equalize rates of profit among different industries. This helped lead to what Marx called the disintegration of the Ricardian school.
Political economy made its great turn away from the law of value determined by the quantity of socially necessary labor needed to produce a commodity—that is, it committed suicide as a science—and eventually arrived at marginalism—the view that the value of commodities is determined by the scarcities of commodities relative to subjectively determined human needs. (10)
Marx’s solution to the transformation problem and the ‘neo-Ricardian’ critique of it
First Marx raised the whole question to a higher level through his distinction between constant capital, which merely preserves the value of capital, and variable capital—labor power—which alone produces new value, including surplus value. If commodities sell at prices that are directly proportional to their values, industries with an above-average organic of capital—assuming equal turnover times—will make a lower than average rate of profit, while industries with a lower than average organic composition should make a higher than average rate of profit.
Yet Marx agreed with the classical economists that free competition would tend to equalize the rate of profit among different industries. This would transform values into prices of production that would equalize the rate of profit among industries with different organic compositions of capital. As a result, the average prices around which market prices fluctuate—prices of production, or production prices for short—would inevitably diverge from values.
Marx’s greatly improved theory of value helped clarify how this could occur without invalidating the law of labor value. In a commodity producing economy, the values of commodities inevitably take the form of rates of exchange between different commodities—that is, exchange values.
Therefore, the value of a commodity is not expressed directly in terms of quantities of abstract labor measured in some unit of time but in terms of the use value of another commodity. Except in the most primitive stages of commodity production, the commodity that measures the values of commodities in terms of its own use value is a special money commodity. For thousands of years, the main money commodity has been the precious metal gold.
Assuming gold bullion is the money commodity, exchange values are measured in terms of the quantities of gold bullion as a material use value. The unit of measure of gold bullion is weight. Exchange value expressed in terms of quantities of the money commodity—weights of gold—is nothing other than price. This is true not only of concrete market prices but also of the abstractions that we call prices of production.
It is perfectly possible for the value of a commodity to express itself in a weight of gold—price—that has a different value than the value of the commodity whose value the gold is measuring. Indeed, Marx makes clear not only is this possible, it is the rule.
Shaikh’s concept of direct price
Anwar Shaikh has clarified what Marx meant when he talked about commodities selling at their values. A commodity, strictly speaking, does not sell at a “value” but always at a price. If the quantity of abstract human labor embodied in a commodity and the quantity of abstract human labor embodied in the gold—or whatever commodity serves as money—with which it exchanges is identical, the commodity is selling at a price that directly reflects its value, or direct price for short. We can contrast direct prices to both prices of production, which equalize profit rates and form the axes around which market prices fluctuate, and market prices, the prices we actually pay in the supermarket, for example.
Marx used direct prices in his analysis of the origin of surplus value in the unpaid labor of the working class. He explained that this is the case even if workers are paid the full value of their labor power, or what comes to exactly the same thing, even if the workers sell their labor power at its direct price. Marx stressed that if you cannot explain surplus value on the basis of commodities selling at their values—or direct prices—you cannot explain surplus value at all.
Marx also used direct prices in his famous diagrams of simple and expanded reproduction. Only in Volume III of “Capital”—quite late in the work—did Marx introduce prices of production that equalize the rate of profit among branches of industry with different organic compositions of capital. Prices of production play a crucial role in Marx’s analysis of differential and absolute ground rent, which is also covered in Volume III of “Capital.” We cannot analyze differential and absolute rent only with direct prices. In order to analyze ground rent, we need prices of production.
Marx’s solution to the transformation problem
In Volume III of “Capital,” Marx presented a model in which commodity prices instead of selling at their direct prices—an assumption that Marx had made up until that point—sell instead at prices of production that produce an equal rate of profit among five branches of production with differing organic compositions of capital.
Marx’s model shows how a system of prices that directly reflect values is transformed into a system of prices that equalizes the rate of profit among the branches of production with different organic compositions of capital. Furthermore, in Marx’s model both the rate and mass of profit for the five branches taken as a whole are equal in terms of labor values, direct prices and prices of production.
What Marx’s solution to the transformation problem demonstrates
Though it seems after the transformation from direct prices into prices of production that both constant and variable capital are equally productive of surplus value, in reality only variable capital produces surplus value or profit.
Did Marx make a mistake?
Starting with von Bortkiewicz, “neo-Ricardians” have claimed that Marx made a mistake in his solution to the “transformation problem.” These critiques point out that Marx’s model does not transform the input prices from values—or direct prices—to prices of production. In Marx’s solution to the transformation problem, the industrial capitalists buy their inputs at values—direct prices—but sell them at prices of production. In Marx’s model the individual commodities are partially transformed from direct prices to prices of production, and the rate of profit is not affected. Whether you calculate the rate of profit in terms of direct prices, or in terms of prices of production, you get exactly the same rate of profit that you get if you calculate the rate of profit directly in terms of value.
In the text, Marx indicated his solution was an incomplete one. A full solution requires the transformation of the inputs into prices of production. Therefore, Marx didn’t make a mistake but left us an incomplete calculation that points the way to a full solution.
Completing the transformation of direct prices into prices of production
The full solution requires what mathematicians and computer programmers call iteration. After the first calculation, you repeat it again with the prices of production derived from the first calculation serving as the input prices in the second calculation. After you repeat it a number of times, the input prices and output prices become completely consistent with one another.
The transformation of direct prices into prices of production is now complete. Both inputs and outputs sell at their price of production. Not only that but if you assume that all commodities produced re-enter the reproductive process—that is, all commodities are what the “neo-Ricardians” call “basic commodities”—the rate of profit in terms of values will always be equal to the rate of profit in terms of prices of production.
While Anwar Shaikh accepts this solution, Andrew Kliman strongly rejects it. If we embrace this solution, according to Kliman, we will end up calculating profits in physical quantities rather than in values, and Marx’s whole theory of value including his theory of profit and surplus value will go out the window.
I don’t agree with Kliman here. I do strongly agree with Andrew Kliman that calculating profits in “physical terms” is a tremendous mistake that both “neo-Ricardians” as well as the marginalists make. But I don’t think the problem lies in the iterative-simultaneous solution to the transformation problem. Instead, I believe that left-wing “neo-Ricardians” fall into the error of calculating profit in physical terms rather than in terms of the unpaid labor of the working class, because they fail to distinguish between value and the form of value—exchange value. That is, it lies in incorrect “non-commodity” theories of money.
One has not fully mastered Marx’s theory of value if one only understands that the values of commodities are determined by the amounts of abstract labor needed to produce them. It is also necessary to understand exchange value as the form of value, which leads to Marx’s theory of money and price. As we will see below, Kliman lacks a theory of money—and therefore of price—worthy of the name. With such an incomplete theory of value you cannot, in my opinion, answer the “neo-Ricardians.”
But why is this so? If we simply have to complete Marx’s partial mathematical calculation, why do “neo-Ricardians” have a case at all?
The problem arises from the fact that the rate and mass of profit are equal in terms of direct prices only so long as all the commodities re-enter the process of reproduction—or are “basic commodities” in “neo-Ricardian” jargon. It doesn’t matter whether the commodities in question enter into the process of production as fixed capital, raw materials or auxiliary materials, or whether they enter into the process of production as items of personal consumption of the productive (of surplus value) workers. As long as this is true, any gains that an individual capitalist makes by selling his or her commodities above their values is exactly counter-balanced by the extra cost of inputs that must be purchased by other capitalists. Since we are assuming that all commodities function as inputs, anything some capitalists gain by selling their commodities above value is lost by other capitalists buying the input above value.
The converse is true as well. Any loss that an individual capitalist suffers from selling his or her commodity below value is gained by some other capitalist buying that commodity as an input below its value. Therefore, the total mass and rate of profit remain unaffected by the transformation of values, or more strictly direct prices, into prices of production. Therefore, the rate of profit in terms of values will exactly equal the rate of profit in terms of prices of production.
So far so good. Even the amputated version of Marx’s theory of value employed by Kilman is sufficient.
But this is true only as long as all commodities enter the process of reproduction. But what, the “neo-Ricardians” ask, about the case of luxury commodities, the commodities consumed only by the capitalists—or weapons consumed by the capitalist state? Once these “non-basic commodities” are taken into account, the mass and rate of profit will differ somewhat when they are measured in terms of values on one hand and prices of production on the other. The equality between the rate of profit in terms of value and the rate of profit in terms of prices of production will now only be approximately true, not exactly true as before.
Money and the transformation problem
Why is this? In my opinion, to understand why this is so you have to understand both the role of money as the measure of the values of commodities and its role as the standard of price.
When we talk about the prices of all commodities, we are by definition leaving one commodity out—the one commodity in the capitalist economy that does not have a price. And what commodity by definition has no price? The money commodity. Since the money commodity serves as the standard of price, it itself cannot have a price. Only if we imagine that money is not a commodity can we talk about the prices of all commodities. Let N equal the total quantity of commodities. The total sum of commodity prices will always leave one commodity out. We can add up the prices only of N – 1 commodities.
Suppose that on average the capitalists buy their luxury commodities at prices of production that are above values. This will mean that the capitalists as a whole will have a rate of profit that in terms of prices is slightly higher than the rate of profit they will have in terms of values. But what they gain by selling these commodities above their values they lose as buyers. Conversely, if they buy their luxury commodities at production prices that are below their values, what they lose as sellers they gain back as buyers of these same luxury commodities.
Therefore, when we calculate in terms of prices of production, value seems to be produced or destroyed in the process of circulation, because we are leaving out the value of the money commodity. Throwing up their arms, the “neo-Ricardians” give up on value at this point and return to dealing only with physical use values.
But once we take into account the value of the money commodity—the one commodity that has no price—the apparent creation or destruction of value in circulation disappears. It is a mere money illusion. Therefore, at the end of the day the value the capitalists get to consume, whether unproductively as items of personal consumption or productively as means of production, is nothing else but the surplus value produced by the working class minus the surplus product embodied in the gold that is used as money, since strictly speaking money is not consumed.
Not understanding that they are dealing with a money illusion, Ian Steedman imagines that he and his fellow “neo-Ricardians” have finally overthrown the whole concept of labor value and surplus value as the unpaid labor of the working class.
Why didn’t Marx complete his solution to the transformation problem, though in the text he indicates that he was aware of all the problems raised by the “neo-Ricardians” over the last 100 years? This is, of course, a matter of speculation. Remember, Marx was working in the days before those marvelous devices we use today both to write and carry out complex arithmetic calculations, when we have to, called computers. And remember, Volume III of “Capital” is only a draft. Perhaps Marx would have gone through the laborious—in those pre-computer days—arithmetical calculations if he had lived long enough to complete his work.
But I think there is a more fundamental reason why Marx left his work as it was. Once we bring in the role of luxury commodities—the non-basic goods in “neo-Ricadian” terminology—we further obscure the fact that only variable capital—living labor—creates surplus value. Once luxury commodities are brought in, the appearance that constant capital—dead labor—can also produces surplus value—the appearance that the “neo-Ricardians” fall hook, line and sinker for—is deepened.
As part of his fundamental method, Marx generally brings in the complicating factors—beyond indicating to the careful reader that he is aware of them—only as needed. This is why he left the entire subject of the prices of production and the “transformation problem” to Volume III of “Capital.” There he had to deal with it, if only because the theory of ground rent required it.
And what does Steedman—and other “neo-Ricardians”—put in the place of Marx’s theory of value and surplus value, which they first failed too understand and then abandoned? Simply the “commonsense” (11) observation that the rate of profit is determined by the physical conditions of production and the real wage. The “physicalism” that Kliman so correctly criticizes in the “neo-Ricardians” arises therefore in my opinion not from their simultaneously determining input and output prices of production but rather from their failure to understand the relationship between value, exchange value, money and price.
Like Kliman, they first amputated Marx’s theory of value. The “neo-Ricardians” then proceeded to refute the amputated theory of value that they have put in place of Marx’s real theory of value. Kliman’s mistake is that he attempts to defend the falsified amputated theory of value that the “neo-Ricardians” have put in place of Marx’s full, genuine theory of value.
In reality, profits are measured in a physical unit but not in the physical units of commodities as a whole—imagine trying to perform such a calculation in the real world as opposed to the corn worlds of the “neo-Ricardian” dreamscape—but in the physical units of the money commodity—weights of gold.
‘MELT’—Kliman’s wrong theory of value, exchange value, money and price
“In recent years,” Kliman writes, “owing largely to the work of Alejandro Ramos … the term monetary expression of labor-(MELT) has become popular. If each hour of socially necessary labor adds $60 of new value … the MELT is $60/hr.”
In a footnote, Kliman further clarifies the MELT theory of money. The MELT, he explains, “is the reciprocal of the amount of labor a unit of money commands [emphasis added—SW].” “It is also the economy-wide ratio of the total money price of output to the total labor-time value of output.”
According to Kliman, the “ratio of the total money price of output to the total labor-time value of output” [emphasis added—SW] determines the value of the total quantity of money and its individual unit. This allows the price of a particular commodity—such as a Big Mac or Macintosh computer, for example—to deviate from its labor value. But at any time, in contrast to Marx’s theory of value, the sum total of all prices must equal the sum total of values by definition. Like in all theories of “non-commodity” money, money here simply reflects—or “commands”—the value of the commodities that it helps to circulate.
Perhaps like many others, Kliman is confused by the fact that today’s paper dollars, unlike in the past, are not legally convertible into a fixed amount of gold. This gives rise to the illusion that the “real value of a dollar” derives from the commodities it circulates and not from its relationship to gold, the special money commodity. This is indeed the commonsense view defended by the upholders of “orthodox—neo-classical—political economy, but it is not the view of Marx.
According to Marx, money is a counter-value to the value of the commodity it is measuring and circulating. It must have a value of its own that is separate from the commodity whose value it is both measuring and circulating. This is why money must always be a commodity.
Therefore, there is always the possibility that the counter-value might differ from the value of the commodity that is circulating in any given case. Indeed, Marx in many places makes clear not only that this might be so, it almost certainly will be so in every real world case.
If we apply Marx’s concept of money and token money to a dollar bill, the apparent value of the dollar bill stems not from the commodities it purchases like our commonsense economists proclaim, but rather from the value of the amount of gold it actually exchanges for on the world market at this particular moment in time.
Thanks to the Internet, I can tell you exactly what the value of a dollar bill is in terms of gold at the very instant you are reading this and not the instant that I am writing this. The gold value is at this instant the reciprocal of the dollar price of gold found on the Web site kitco.com. That is, a U.S. dollar now represents 1/[the dollar price of gold as reported at Kitco.com] of an ounce of gold. Under the average current conditions of production now needed to produce this quantity of gold, a given amount of abstract human labor is required. The value a U.S. dollar bill represents is therefore exactly the quantity of abstract human labor that is necessary to produce 1/[dollar price of gold as reported at Kitco.com] of an ounce of gold under the prevailing conditions of production.
If the dollar price of gold were to be stabilized—the international gold standard were to be restored—the amount of gold that a dollar represents would cease to fluctuate. But the amount of abstract human labor that a dollar represents through the commodity gold would still fluctuate, though less than it does at present. The amount of human labor measured in some unit of time that it takes to produce the given weight of gold bullion that would define the dollar under the new international gold standard would continue to fluctuate in response to the ever-changing conditions of production in the gold mining and refining industries.
Unlike the MELT theory of non-commodity money, Marx’s theory of money contains the possibility—indeed the virtual certainty—that the sum total of money prices will not actually equal the sum total of direct prices at any given point in time. Indeed, Marx hints at this in the very first volume of “Capital.” In Volume I Chapter I of “Capital,” Marx writes, “Jacob doubts whether gold has ever been paid for at its full value.”
Marx does not indicate that he agrees with the now long-forgotten Jacob on this point, but he doesn’t indicate that he disagrees with him either. Let’s assume for the sake of argument that our Mr. Jacob was correct. If gold has never been paid for at its full value and if gold is the money commodity, this would mean that the sum total of commodity prices have always exceeded the sum total of values, or more precisely the sum total of direct prices.
Why does Marx even bring up the obscure Jacob’s opinion that gold has never been “paid for at its full value” in the very first chapter of “Capital”? Is it because the Jacob quoted was so well known? Perhaps he was in the 19th century when Marx wrote “Capital,” though I doubt it. Isn’t Marx warning us that though he will assume as a general rule throughout the first and second volumes that prices equal values—direct prices—and that therefore the sum total of all commodity prices equal the sum total of all direct prices, this in fact might not necessarily be true?
Indeed, it might never be true in the real world. Marx is warning us at the very beginning of his great work against the very mistakes the supporters of the MELT concept including Kliman have fallen into. If you disregard this warning by the author of “Capital” at the very beginning of “Capital,” you are headed for big trouble later on.
Kliman’s inconsistent theory of value and money
Kliman himself senses there is something missing in the concept of MELT. “Prices,” he wrote in an article on the 2008 economic crisis, “have indeed consistently risen in relationship to the real values of goods and services….” Presumably, he is referring to the period since the end of the 1930s Depression. I don’t believe this is factually true, as I explained in my main posts, but this is not the point I want to make here.
If the MELT theory of money is true, how could prices ever rise above the “real values of goods and services” [Kliman should have said commodities—SW] at all—or fall below them for that matter?” According to MELT, isn’t the sum of all prices equal by definition to the sum of all values?
Yet here when faced with a concrete problem of an economic crisis, Kliman instinctively—and this is to his credit—throws MELT into the melting pot where it belongs, and talks about commodity prices rising above the values of all commodities. How is this possible, though, unless we have a money commodity that measures the values of commodities in terms of its own use value?
If Kliman follows his instinct here, I think he will find the correct answer to the “neo-Ricardians” as well as greatly improve his own grasp of crisis theory.
Kliman’s lack of a solution to the transformation problem
Kliman’s rejects, as we saw, Anwar Shaikh’s solution to the transformation problem. Shaikh’s solution—and Shaikh provides quotes from Marx’s text indicating that Marx was heading in the same direction—is that the rates of profit in terms of values and in terms of prices of production are only approximately equal, not exactly equal. The rate of profit in terms of money—prices of production—is a somewhat distorted image of the value rate of profit. Still, the price of production rate of profit and ultimately the rate of profit realized in terms of money in the real world are in the long run ultimately governed by the value rate of profit as I explained in my reply on the transformation problem.
If I understand Kliman correctly, he seems to be saying that the value rate of profit and the money rate of profit are equal by definition—the MELT theory of money. But when he has to face the concrete problems of changes in the industrial cycle and economic crises, he is obliged to tacitly throw MELT out the window. It is clear that he is confused by the whole phenomena of almost uninterrupted inflation of nominal currency prices that has occurred throughout the capitalist world since 1933 and the relationship of these rising prices in currency terms and underlying labor values. The common age-old phenomena of inflation caused by the devaluation of token money—paper money or in earlier times currency made of base metals—against gold is simply beyond the ability of MELT to explain.
Kliman therefore doesn’t offer, as far as I can see, a meaningful explanation of the transformation problem, let alone provide a convincing refutation of “neo-Ricardian” criticisms of Marx. If Kliman’s defense of Marx was the best we had as far as the transformation problem is concerned, we would have to admit that Marxist theory was in a bad way. Perhaps we would even have to acknowledge the correctness of the “neo-Ricardian” critique and concede the collapse of scientific socialism in its Marxist form.
Fortunately, in addition to the hints of Marx himself, and the power of Marxist theory in general, we have the work of Anwar Shaikh, which points us toward the real solution to the transformation problem, which has haunted economics in one form or another since the days of Adam Smith.
To be continued
1 I don’t much like the term “neo-Ricardian,” because the “neo-Ricardians” reject Ricardo’s greatest contribution to economic science, his labor law of value. To be sure, Ricardo’s theory of value was not without its internal contradictions, or lacking in consistency as Kliman would put it. But this was not all bad. As Marx put it, its very contradictions contained the seeds of further development, in contrast to the sterility of post-Ricardian schools of political economy.
Ricardo himself began with building “corn models,” where he calculated rent—surplus value—in physical terms in corn. But Ricardo realized the inadequacy of these models and turned instead to valuing commodities in terms of the quantity of labor necessary to produce them. He realized that the (capitalist) economy involved human beings engaged in production and exchanging the products of their labor.
When Ricardo was unable to resolve the apparent contradictions of his law of labor value, he sensed that they could be resolved, even if he was not able to do it.
The “neo-Ricardians,” in contrast, faced with the apparent contradictions of the law of labor value and misunderstanding and rejecting Marx’s solution to the contradictions, turned in the opposite direction back to the corn models and the calculation of surplus value in physical terms that Ricardo left behind in his mature work. In my opinion, the “neo-Ricardians” represent a huge regression not only in relationship to Marx but in relationship to Ricardo as well. It is really a great injustice to call this school “neo-Ricardian.” It would perhaps be better to refer to them as “neo-Physiocrats.”
In respect to Ricardo’s memory, I place the term “neo-Ricardians” in quotation marks in this text.
The Physiocrats were a school of economists that arose in France before the Great French Revolution. In their day, they represented a great advance for the emerging science of bourgeois political economy. Their greatest contribution was to locate the origin of surplus value in the sphere of production and not the sphere of circulation. Their advance remains valid even if they made the mistake of seeing only agriculture as productive of surplus value and measuring value in physical terms rather than in terms of quantities of human labor. They were, after all, living in what was still a pre-industrial agricultural society and were impressed by the ability of seed corn to produce a greater amount of corn—a physical surplus that they, much like today’s “neo-Ricardians,” confused with surplus value.
But in justice to the Physiocrats, they unlike today’s “neo-Ricardians,” lived before not only Marx but also Ricardo, and performed their work before Adam Smith, who was both a pupil and critic of the Physiocrats. One of their members, Dr. Francois Quesnay (1694-1774) in his “Economic Tableau” developed the first diagram of capitalist reproduction. It was not attempted again by another economist before Marx, whose own work on reproduction was inspired by Quesnay’s work.
4 I suspect if he asked a real capitalist farmer what he thought of the corn models, the capitalist farmer would answer that those professors of economics never put in an honest day’s work in a backyard garden let alone on a real farm! Certainly, workers employed in agriculture would react this way. Their daily work would demonstrate to them what nonsense these corn models really are. And they surely would realize that they need more than “corn” to remain alive and reproduce the next generation of agricultural workers.
5 Kliman should have called the “neo-Ricardians” to order right here. Unfortunately, he doesn’t, so we are forced to try to follow the mad logic of this dream world conjured up in the minds of our “neo-Ricardian” economists.
6 At this point, I suspect that even Ian Steedman would say that maybe calculating in physical terms isn’t such a great idea after all. But I admit I am only speculating here. Maybe the now-retired professor really does live only on corn, though photographs of him, if I recall, directly show him wearing clothes, so probably he does not live on only corn after all.
7 The fact that the crisis of 2008 and many other such crises, beginning with the crisis of 1825, have occurred itself proves that neo-classical marginalism or any other economic theory that treats money this way is wrong. Again, Kliman should have called the “neo-Ricardians” to order on this point, but since he himself is totally at sea on the subject of money and price, he does not.
8 I am not saying there is no truth in TSSI. I am saying that the truth that is contained in the TSSI approach is fully subsumed within Marx’s complete theory of value. However, if you have only Kliman’s amputated theory of labor value, TSSI does not really answer the “neo-Ricardian” criticisms.
9 This is not to say that Shaikh has consistently applied Marx’s value theory in all his work. When he attempted to demonstrate empirically a falling rate of profit, he in my opinion made an error when he calculated the rate of profit during the inflationary decade of the 1970s after inflation—that is, in real (physical) terms.
He should have calculated the rate of profit during the 1970s in terms of the use value of the money commodity—in terms of weights of gold bullion. If Shaikh had done this, he would have found that profits were strongly negative between 1970 and the Volcker Shock, after which they dramatically recovered and became strongly positive once again. In order to determine whether or not the empirical data demonstrates a falling rate of profit, he would have to average the huge losses that occurred in terms of real money—gold bullion—in the 1970s with the huge profits that were made in the years following the Volcker Shock.
But in his work on the transformation problem, which is what I am concerned with here, Shaikh employed Marx’s real theory of value, exchange value, money and price, and prices of production. Therefore, unlike Kliman, he arrived at the correct solutions.
10 We shouldn’t forget that while the very real logical contradictions in Ricardo’s law of labor value played a role in the demise of classical political economy, the fundamental reason why the post-Ricardian (bourgeois) economists were not able to continue Ricardo’s work was the growing intensity of the class struggle between the capitalist class and the working class. A further development of Ricardo’s theory would have laid bare the real origins of surplus value—profit—in the unpaid labor of the working class, and there was no political possibility of doing this on the basis of bourgeois political economy.