Comments on Milanovic on Marx

By Fred Moseley

I am a Marxist economist (Professor of Economics, Mount Holyoke College) and I appreciate Branko Milanovic's open-mindedness and his efforts in a recent post on his blog to educate economists who often have a crude and superficial misunderstanding of Marx’s labor theory of value.  

For context for my comments on Milanovic, I will first say a few words about my interpretation of Marx’s labor theory of value (LTV).  In my view, Marx’s LTV is primarily a macro theory and the main question addressed in Marx’s macro LTV is the determination of the total profit (or surplus-value) produced in the capitalist economy as a whole.  Profit is the main goal of capitalist economies and should be a key variable in any theory of capitalism.  Marx’s theory of the total profit is that profit is the difference between the value produced by workers and the wages they are paid, i.e. that profit is produced by the “surplus labor” of workers.

I argue that Marx’s “surplus labor” theory of profit has very significant and wide-ranging explanatory power.  Marx’s theory provides straight-forward and robust explanations of the following important phenomena of capitalist economies:  conflicts between capitalists and workers over wages, and over the length of the working day, and over the intensity of labor (i.e. how hard workers work, which determines in part how much value they produce); endogenous technological change (in order to reduce necessary labor and increase surplus labor and surplus-value); increasing concentration of capital and income(i.e. increasing inequality); the trend and fluctuations in the rate of profit over time; and endogenous cycles due to fluctuations in the rate of profit rate of profit.  (A more complete discussion of the explanatory power of Marx’s theory of profit is provided in my “Marx's Economic Theory: True or False? A Marxian Response to Blaug's Appraisal,” in Moseley (ed.), Heterodox Economic Theories:  True or False?, Edward Elgar, 1995).

This wide-ranging explanatory power of Marx’s surplus labor theory of profit is especially impressive when compared to mainstream economics.  In mainstream macroeconomics, there is no theory of profit at all; profit (or the rate of profit) is not even a variable in the theory!  I was shocked when I realized in graduate school this absence of profit in mainstream macro, and am still shocked that there is no effort to include profit.  Indeed, DSGE models go in the opposite direction and many models do not even have firms!

Mainsteam micro does have a theory of profit (or interest) – the marginal productivity theory of distribution – but it is a weak and largely discredited theory.  Marginal productivity theory has been shown by the capital controversy and other criticisms to have insoluble logical problems (the aggregation problem, reswitching, cannot integrate intermediate goods, etc.).  And marginal productivity theory has very meager explanatory power and explains none of the important phenomena listed above that are explained by Marx’s theory.  

Milanovic agrees that Marx’s LTV is primarily a macro theory, but he interprets it in this post as only the assumption that “sum of values will be equal to sum of production prices”.  And he continues:  “The former is an unobservable quantity so Marx’s contention is not falsifiable.  It is therefore an extra-scientific statement that we have to take on faith.  

I argue, to the contrary, that Marx’s macro LTV is primarily a theory of profit and my conclusion that Marx’s theory is the best theory of profit we have is not based on faith but is instead based on the standard scientific criterion of empirical explanatory power.  It is much more accurate to say that marginal productivity theory is accepted by mainstream economists on faith, as Charles Ferguson famously said in his conclusion to the capital controversy.

Now to my comments on Milanovic's three main points:  

1.  Milanovic's main point is that the LTV is often misinterpreted as a simple micro theory that assumes that the prices of individual commodities are proportional to the labor-times required to produce them.  Milanovic argues that is not true in a capitalist economy because of the equalization of the profit rate across industries with unequal ratios of capital to labor, so that according to Marx’s theory, long-run equilibrium prices are determined by the equation:  w + d + rK, where w is wages, d is depreciation and r is the economy-wide rate of profit (missing in this equation is the cost of intermediate goods, but I will ignore this).  

Milanovic emphasizes that Walras and Marshall had essentially the same equation for long-run equilibrium prices.  I agree that all three theories of long-run equilibrium prices have this same form, but there is an important difference.  Marx’s theory provides a logically rigorous theory of the rate of profit in this equation (based on his theory of the total profit discussed above) and Walras and Marshall just take the rate of profit as given, disguised as an “opportunity cost”, and thus provides no theory of profit at all.  Therefore, I think Marx’s theory of long-run equilibrium prices is superior to Walras’ and Marshall’s in this important sense.

2.  Milanovic's second main point is that Marx’s theory of long-run equilibrium prices are “clearly very, very far from derisive statements that the labor theory of value means that people are just paid for their labor input regardless of what is the ‘socially necessary labor’ required to produce a good.”  I presume that this derisive statement means that workers produce more value than they are paid and thus are exploited in capitalism.  But Branko is mistaken about this.  Marx’s theory of long-run equilibrium prices is based on his macro theory of profit according to which the source of profit is the surplus labor of workers.  This conclusion is indeed derisive and that is the main (non-scientific) reason that Marx’s theory of profit is rejected by mainstream economists in spite of its superior explanatory power.

I know from previous correspondence that Milanovic understands well Marx’s “exploitation” theory of profit, but he seems to overlook the connection between Marx’s micro theory of prices of production and his macro theory of profit.

3.  Milanovic's third point is that Marx’s labor theory of value is most helpful in understanding pre-capitalist economies and the relation between capitalism and non-capitalist economies today.  I argue, to the contrary, that Marx’s labor theory of value and profit is the best theory we have to understand the most important phenomena of capitalist economies, including 21st century capitalism.

It would be one thing if mainstream economics had a robust theory of profit with significant explanation power.  But it has almost no theory of profit.  Therefore it would seem to be appropriate from a scientific point of view that Marx’s surplus labor theory of profit should be given more serious consideration.

Thanks again to Milanovic and I look forward to further discussion.