“Is society a branch of physics?” asked the Abbé Mably, a minor nineteenth-century pamphleteer and philosophe. The absurd question serves very well to introduce a discussion of what modern economics is about and whether Karl Marx still has something to contribute to economic thought. For, essentially, economics has always answered Yes to the Abbé’s query. That is, it has always proceeded on the belief that there were enough regularities in the social process to enable a skilled observer to discover “laws” that described its movements, just as other laws described the motion of the planets in their orbits.

To be sure, economists have always recognized that there was a vast gulf between the unknowing planets and sentient human beings, and therefore they have never intended the laws to be as strict in the second case as in the first. Yet the gulf was not so wide as to destroy all similarity between the orderliness of the natural world and that of the social. For underneath the seeming disorder of the social universe, two processes could be discerned that imposed a degree of lawlike regularity on the events of economic society. One of these was the process of production itself—the actual technical sequences by which wheat became bread and grapes wine and iron ore steel. Although these sequences differed one from another, and although they changed over time with technological advance, nonetheless there seemed to be sufficient regularity, at least in the short run, so that we could talk of “laws” of production, such as diminishing returns or economies of scale or “coefficients of production” or “marginal elasticities of substitution”—all terms that describe the dependability of the productive element within the social universe.

The other order-bestowing element in the economic process concerned its human side, which is to say the behavior of workers and consumers and entrepreneurs. Clearly, this aspect of the underlying social orderliness could not be expected to demonstrate the same degree of invariance that is found in the physical world. Yet in the behavior of buyers and sellers there seemed to be a sufficient degree of repetitiveness so that we could talk of the “law” of supply and demand; and in the responses of consumers to changes in their incomes or of businessmen to changes in the interest rate other lawlike patterns emerged.

Thus from the very beginning, economists have striven for a picture of society in which the interaction of laws of production and behavior—production and behavior functions is the modern term—would describe the major economic events of the social system much as if it were a branch of physics. Moreover, by reducing the complexity of the real world to the simplicity of a “model” dominated by these two great functions, economists, like physicists, have sought to predict the path of motion of their system.

How successful has been this audacious intellectual effort? On the face of it, the achievement has been astonishing. Models of the economy are now so complex that they require the facilities of a computer and the techniques of difference equations, matrix algebra, LaGrangian multipliers, and the like. Sophistication, elegance, rigor—the criteria by which mathematics has traditionally been judged—are now the standards of economic theorizing. Not least, the success of modern economics can be read in the flattery of imitation paid to it by its sister disciplines of sociology and political science which now seek to build models similar to those of the economist. Certainly, when the intellectual history of our times is finally written, the creation of the edifice of modern “neo-classical” economics will occupy a central chapter in it.

The only question is, what will that chapter say about the usefulness and relevance of this extraordinary enterprise? Here I suspect the appraisal of the future will not be uncritically admiring. The theory of economics, magnificent to behold, is considerably less impressive to use. It is true that it has given us a rough picture of how the market system works, both in allocating its resources and in determining the level of overall output. But beyond this conception, which can be taught with ease to a college freshman, the ramifications of economics have produced singularly little. A rococo branch called welfare theory, for example, has not, to my knowledge, yet resulted in a single substantive proposal that has added significantly to the welfare of mankind. The beautifully finished portion called price theory fails to explain the pricing operations of the great corporations. International trade theory does not adequately account for the most important single fact about international trade—to wit, the failure of an international division of labor to shed its benefits on poor countries and rich countries alike. The theory of economic development does not tell underdeveloped countries how to grow.

Even the central achievement of twentieth-century economics—the elucidation of the forces that determine prosperity and recession—fails when we seek to foretell the fortunes of the economy a few months hence. No doubt economists reading these words will deem them vastly exaggerated, which perhaps they are. Yet it is surely an opinion not wholly at variance with mine that must have moved Kenneth Arrow, a well-known economist, to sum up the collected papers of Paul Samuelson, the most brilliant theorist of our generation, with these words: “A careful examination of the papers both on theory and on policy yields only the most oblique suggestion that neoclassical price theory is descriptive of the real world. Of course, there is no denial, but Samuelson’s attitude is clearly guarded and agnostic” (Journal of Political Economy, October, 1967).

Advertisement

Why is it that modern economic theory presents the spectacle of superb intellectual achievement without much social relevance? To my mind there are two reasons. One lies in the difficulties of reducing the real world—both in its technical and in its behavioral aspects—to reliable patterns with which we can then construct dependable models. It is one thing to ascribe an underlying “lawlike” character to the processes of production and to the responses of the economic actors, and quite another to reduce these activities to mathematical functions. In the case of production, for example, we encounter enormous difficulties in devising mathematical functions that will accurately account for the constantly changing nature of technology. And this difficulty is compounded by the even more intractable problem of finding functional representations of human behavior. No doubt, for instance, men tend to buy less when prices rise and to buy more when prices fall. Yet, on occasion, they will do just the opposite, as when they expect a price rise or price fall to continue—in which case their self-interest bids them to buy more in a hurry in the first case, and to hold off in the second.

Hence the inherent complexities of the production process and the vagaries of human behavior may well set limits to the predictive possibilities of economic theorizing, and these limits may account for much of the gap that exists between economic theory and economic reality. Yet, however much these difficulties explain the inaccuracy of economic theorizing, they do not account for its irrelevance. I have already mentioned the failure of price theory to explain the behavior of the large corporation and the gap between the theory and the reality of international trade. Now I must point out other areas of economic life over which modern economic theory passes in virtual silence. The distribution of wealth, for example, is a central economic fact about which it is mute. The effects on the distribution of income attributable to the process of growth is another, so that economics gives us no hint of the disturbances and frictions produced by long-run economic advance. The effect of a constantly improving technology on the level of employment is similarly ignored, so that today the theory of technological unemployment is in much the same shape as was the theory of mass unemployment in the days before Keynes. The nature of class interests in a capitalist system is not mentioned in any textbook, so that nothing in the nature of political or social constraints confines the free movement of the economic model.

In all of this, it will be noted, there is a common denominator. This is the systematic exclusion of matters that might connect the functional model with the pressures and resistances of the political world. This exclusion, which accounts for so much of the irrelevance of economics, is by no means accidental. Rather, it results from a fundamental failure of vision on the part of the modern model-builders, who do not see that the social universe that they are attempting to reproduce in a set of equations is not and cannot be adequately described by functional relationships alone, but must also and simultaneously be described as a system of privilege. In other words, if a model of economic reality is to be relevant, it must portray both the functional relationships peculiar to the provisioning process and those stemming from the clash of interests generated by this very functional process itself.

Is it possible to construct a system that is at one and the same time a portrayal of functional relationships and of privilege? There is one such system, Marxian economics—that vast terra incognita over which the average economics student flies while en route to the oral examination (where it may be mentioned as part of the History of Economic Doctrines) and at which he never again casts a glance. For what is unique about the Marxian system is that the categories, both of production and of behavior, into which it disaggregates the world are considerably different from those of the neoclassical system. On the production side, for example, Marxism lays great stress on the necessity for a “fit” between the output of the capital goods sector and that of the consumer goods sector, a relationship that is unnoticed in neoclassical economics where the aggregate output of all sectors is stressed rather than the relationship between them. Similarly, on the behavioral side, Marx approaches the problem of describing the great “human” functions by building up a picture of the actions of producers—that is, workers and capitalists—rather than by analyzing the activities of buyers, ie., of consumers and investors. In different words, the Marxian analysis breaks down the total flow of economic activity into layers of costs, wages, and profits rather than into the slices of consumption and investment characteristic of the Keynesian approach.

Advertisement

The result of the special categories of abstraction imposed by the Marxian view is to bring into the foreground a number of matters that fail to appear in neoclassical analysis, in particular the instability of the economy stemming from a failure of its productive components to interlock, and the changing division of the social product among the classes—profit receivers and wage earners—that compete for it. Now it should be said immediately that the manner in which classical Marxian analysis performs its task of constructing a model of society is very awkward and occasionally downright wrong. The “laws of motion” that it discerns within the capitalist system depend on rigid assumptions about the way in which technology permits labor to be combined with capital and loses sight of the central effect of productivity in changing the real shares of wage earners in the final product. Worse yet, as a means of explaining the price mechanism by which the system is coordinated, Marxian economics is hopelessly clumsy: if one examines the efforts of the more liberal Soviet or Czech economists to create a rational pricing system using Marxian concepts, and compares these efforts with the results obtained by nonMarxian price theory, the contrast is like that between a dull cleaver and a sharp scalpel.*

Why then bother with Marxian economics when, as virtually every economist will tell you, it is “wrong”? The reason is that, unlike neoclassical analysis, which is “right,” the Marxian model has in surfeit the quality of social relevance that is so egregiously lacking in the other. The neoclassical model has rigor, but, alas, also mortis. The Marxian model has relevance, but, alas, also mistakes. The answer, then, is clear. Marxian insights must be married to neoclassical techniques to produce an economic theory that is both elegant and consistent as a model and freighted with meaning as a theory of society.

The books under review all contribute to this end. Fred M. Gottheil’s Marx’s Economic Predictions is a succinct, crystal-clear statement of the basic “laws” of the Marxian system; I do not know of a better exposition of the content of Marxian economics or a more lucid critique. Unfortunately Mr. Gottheil felt impelled to go beyond this excellent heuristic purpose, and to extract and enumerate every future-oriented statement, direct or implied, that Marx ever made, finally arriving at 172 such “predictions.” Since these range from the major trends that Marx did indeed see as central to capitalism, such as the tendency of the rate of profit to fall, to trivial obiter dicta (Gottheil’s No. 45: “The English proletariat will get nowhere until the Irish revolution is accomplished”), the effect is to confuse Marxian economics as a very large-scale model of social and economic evolution with Marxism as a kind of fortune-telling operation. I note that this book, whose expository excellence I cannot praise too highly, emerged from a doctoral dissertation. That explains a lot.

Murray Wolfson has undertaken a more ambitious task in A Reappraisal of Marxian Economics. This is to criticize the Marxian system not only for the shortcomings of internal consistency that Gottheil has highlighted, but as an economic philosophy based on propositions that are essentially metaphysical and therefore unfalsifiable. I am only half convinced by Wolfson’s argument, but—as with the Gottheil book—I am not interested in arguing its theses here. Rather, I welcome Wolfson’s book, because, like Gottheil’s, this is a successful attempt to “translate” Marxian economics into the language and symbols of modern Western economics (isoquants and all) so that a bridge is built between Marx’s thought and that of the non-Marxian world, and the possibility of mutual enlightenment is thereby opened. Let me merely add that I found particularly interesting Wolfson’s efforts to rescue the “metaphysical” notion of the labor theory of value by applying it not to the determination of prices, where it does indeed lead to terrible problems, but to the overall division of the product within society where the theory can be restated in substantive terms of class bargaining. Whether or not this particular recon-struction of Marxism can be faulted, it illustrates the kinds of questions that open themselves to investigation once one employs a different set of conceptual tools.

Last, let me mention a collection of essays edited by David Horowitz, Marx and Modern Economics. Here is an absolutely first-rate compendium of articles, by such luminaries as Wassily Leontief. Joan Robinson, Martin Brofenbrenner, Oskar Lange, Lawrence Klein, and others, which again serves the purpose of seeking a fusion between Marxian insight and neoclassical technique. The essay by Klein, for example (originally printed in the Journal of Political Economy for April, 1947), points out that the famous Marxian equations establishing the relations between the capital goods and the consumer goods branches of the economy are concerned with definition rather than with behavior, and are therefore not very satisfactory as a starting point for analysis, but that it is possible to reconstruct the Marxian system along more dynamic lines with results that anticipate, and in some cases go beyond, those of the Keynesian model. The argument is unfortunately complicated (and quite mathematical), and I must refer the interested reader to the essay in question. The point, however, is that work such as Klein’s makes it clear that the aridity of so much of neoclassical theorizing does not lie in its use of highly abstract models, but in the nature of the relationships which are explored through these models.

The Horowitz book, like the others, is far from simple for the noneconomist, but it would be food and drink for a student in an advanced undergraduate or intermediate graduate course. Will it be used there? I should dearly like to see some such reading made part of the standard education of the economist. It is just because society is not a branch of physics that we must ask of economics a special kind of prediction called foresight. I do not myself believe that Marxism has very accurate foresight, but I am convinced that neoclassical economics has none at all. The hope is that in combining the Marxian angle of vision, with its emphasis on class structure and privilege, with the fierce insistence of neoclassical techniques on consistency and clarity, that we might have the beginnings of a new and fruitful chapter in the effort to comprehend the social universe.

This Issue

December 5, 1968