American Economic Association

Understanding Capital: Marx's Economic Theory. by Duncan K. Foley Review by: John E. Roemer Journal of Economic Literature, Vol. 28, No. 4 (Dec., 1990), pp. 1727-1730 Published by: American Economic Association Stable URL: http://www.jstor.org/stable/2727452 . Accessed: 14/10/2013 01:13
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

American Economic Association is collaborating with JSTOR to digitize, preserve and extend access to Journal of Economic Literature.

http://www.jstor.org

This content downloaded from 152.2.176.242 on Mon, 14 Oct 2013 01:13:17 AM All use subject to JSTOR Terms and Conditions

Book Reviews
The new material that economists are likely to find of greatest interest is in Chapter 6, which is an eskay on nuclear safety, written with Elizabeth Nichols. One of the more interesting problems noted by the authors is that testing of nuclear reactors can often be counterproductive. One would like to ensure that the diesel generators for these reactors that might be called upon in an emergency will be able to provide sufficient back-up power; but, unfortunately, testing of these generators may actually make them less dependable after the test. Wildavsky focuses only briefly on the kinds of tradeoffs that are most prominent in treatments of risk issues by economists. In particular, Chapter 4 provides a discussion of the costs per life saved for various government policies. However, since this material is based on his citation of previous surveys of regulatory agencies' value of life that were published more than a decade ago and were based on regulations promulgated even earlier, it does not provide as up-to-date a treatment of these issues as is available elsewhere. From the standpoint of the author's major theme, this drawback is not essential since the primary intent is to document the multiple risk tradeoffs we face and the importance of economic growth to improvements in health. Readers will find here a diverse and eclectic series of examples to support this general view. Some economists might have wished that Wildavsky had gone even farther in his advocacy of tradeoffs. Since the U.S. Office of Managment and Budget now requires that all risk regulation agencies assess both the benefits and costs of their regulatory efforts, it may be that in some cases government risk regulation practices already recognize tradeoffs that go beyond those advocated here. Although it does not break new economic ground, this book provides a useful introduction to risk regulation issues for governmental officials and undergraduates. Furthermore, the fact that it has aroused considerable controversy among sociologists and political scientists suggests that there is a pressing need for additional risk management education. Reading Searching for Safety would be a good place for these people to start. W. KiP Viscusi Duke University
030
HISTORY OF ECONOMIC THOUGHT; METHODOLOGY

1727

Understanding capital: Marx's economic theory. By DUNCAN K. FOLEY. Cambridge, MA and London: Harvard University Press, 1986. Pp. viii, 183. $20.00, cloth; $8.50, paper. ISBN 0-674-92088-0, pbk. JEL 87-0376 In this succinct yet encompassing book, Duncan Foley presents his interpretation of Marx's economic theory, including his own resolution of some of the famous paradoxes or inconsistencies in the theory. The book should be comprehensible to the upper-division student and will be instructive to any reader, as it is the most codified and clear expression available of Foley's contribution to Marxist economic theory. Marx based his analysis of the capitalist economy on the labor theory of value (LTV)-inherited, with amendment, from Ricardo. His development of the LTV gave many insights into capitalism, but also produced some paradoxes and results that seem wrong when confronted with modern tools of economic analysis. The most common reaction of economists has been to discard Marxian economics because of these paradoxes or mistaken conclusions (Samuelson 1971). There have been, in my view, two notable minority reactions: (1) to admit that Marx's LTV is wrong in many aspects, but to show that some of his essential insights about capitalism can be demonstrated in models that do not assume the LTV (Morishima 1973; Cohen 1979; Roemer 1982, 1988; Elster 1985); and, (2) to retain the LTV as central, but to reinterpret it in a way that allows resolution of the paradoxes/mistakes. Foley's book is, I think, the best example of the latter approach. In what follows, I shall contrast these two approaches to Marxist economics on four questions. Property rights and exploitation Foley writes: "If you do not accept the postulate that labor produces the whole value added, you will not see much basis for the claim that wage-labor is exploitative. I think this is the main reason that the labor theory of value has fallen into disrepute among orthodox economists" (p. 39). Foley's solution is to stick to what he views as the Marxist guns, namely, the aforementioned postulate. The approach of Cohen (1979) and Roemer (1982, 1988) is to

This content downloaded from 152.2.176.242 on Mon, 14 Oct 2013 01:13:17 AM All use subject to JSTOR Terms and Conditions

1728

Journal of Economic Literature, Vol. XXVIII (December 1990)

of workers. A facet of the LTV was thereby maintained. Foley, however, thinks Morishima gives up too much of the LTV. Foley's procedure is to treat two parameters as given: the money wage, w, and the hours of labor per dollar value added in the economy-call this m, Foley's "value of money." He adds an equation stating that the total value-added (in dollars) per unit labor expended should equal m:
m[(p
-

view value added as produced by both labor and capital, and to say that the distribution of that part of it which is surplus to the current wages of workers depends upon property rights. In this view, labor is exploited in capitalism not because some of the product that it alone produces is expropriated from it, but rather because the property rights that give capitalists a right to part of the product jointly produced by labor and capital are illicit. I think that as Foley frames it, the charge of exploitation is not credible, because his key underlying postulate is not credible. Preserving the LTV and the transformation problem In Marx's view, at the highest level of abstraction, prices should be proportional to the labor embodied in commodities. (This is one facet of the LTV.) If there are several sectors of production, this assumption is inconsistent with the equalization of profit rates across sectors, an equilibrium condition driven by competition of capital. The transformation problem consisted in showing how some of the other facets of the LTV could be maintained while deriving a theory of prices in which profit rates are indeed equalized. Consider a two sector economy, producing steel and wheat, with a Leontief technology in which steel and labor are inputs into both wheat and steel. The usual representation of the price equations are:
Ps= (1 + r)(p,a, + wl,) (1)

p,aS)S

+ (Pw -

pa)W]

= ls

+ lW,

(3)

where (S,W) is the total production of (steel, wheat). (This is a version of Foley's equation (6.7), adapted to the present notation.) Given S, W, m, and w, prices and the profit rate are now determined. Foley argues for the Marxist pedigree of this procedure because it preserves what he believes to be an important facet of the LTV, namely that the amount of labor per dollar valued added is preserved when 'transforming' prices from ones proportional to labor values to ones that equalize the rate of profitas given by (1), (2), and (3'). I cannot understand the economic motivation for Foley's (3'). It seems to me to be a deus ex machina whose purpose is to maintain certain claims of the LTV. I can see a general equilibrium model of a capitalist economy in which equation (3) appears (as the budget constraint of a worker); but I do not see a general equilibrium model in which (3') expresses some real constraint on economic activity. The falling rate of profit (FRP) Ricardo argued for a falling rate of profit in a corn economy based upon diminishing returns to labor in the corn industry. Marx proposed a FRP theory that did not depend upon diminishing returns, but rather upon technological change involving increased capital inputs. Suppose there is a Leontief technology (constant returns) for producing corn, whose (corn, labor) input coefficients are (a, 1). Okishio (1961) argued that technological change with a constant real wage of b units of corn per unit labor would necessarily cause the rate of profit tv rise. Briefly, if with the old technology we have: p= (1 + r) (pa + wl) pb = w (4) (5)

Pw= (1 + r)(p,a. + wl.),

(2)

where r is the rate of profit, (a, Is) are the (steel, labor) inputs into the production of steel, etc. There are thus far two equations in four unknowns (ps, p. r, w). Normalization of prices can reduce this to three unknowns. How should the system be closed? Morishima's answer was to assume a real wage of b (in wheat, say) for workers, and to close the system with: w = pwb. (3)

His argument that this procedure has a good Marxist pedigree was that this theory of prices and the profit rate preserved (as a theorem) the statement that a positive profit r is only consistent with a positive rate of exploitation

This content downloaded from 152.2.176.242 on Mon, 14 Oct 2013 01:13:17 AM All use subject to JSTOR Terms and Conditions

Book Reviews
and then a new technology (a',l') appears, then capitalist corn growers will introduce it, ceteris paribus, if it lowers unit costs, i.e., if: pa' +wl' <pa+wl. (6)

1729

After the technology is thoroughly introduced, the new equilibrium is characterized by: p' = (1 + r') (p'a' + w'l') p'b = w'. (7) (8)

These equations imply r' > r. Note the assumption of constancy of the real wage in (5) and (8). Foley retains (4) as the fundamental relation of the price and the profit rate, and (6) as the criterion for rational capitalist innovation, but replaces (5) with: 1/(1 - a) p which states that the labor embodied per dollar of corn is a constant. (That is, the "value of money," m, is constant.) With a given money wage w, and given m, (4) and (9) determine p and r. After the technological innovation (satisfying (6)), the new equilibrium equations are: p" = (1 + r")(p"a ' + wl ') = m /(-a) p't Now (4) and (9) imply that: -1. r= a I+(a+ -a)wm (12) (10) (11) M (9)'

power of labor and technical change. (Foley agrees that neither procedure can replace a full theory.) I see the Okishio assumption of a constant real wage as grounded in a possible scenario of a real economy: if labor is abundant relative to capital in a fixed coefficient technology, then the real wage will arguably be bid down to some fixed real alternative that workers have outside the capitalist sector (e.g., what they could earn returning to the subsistence farm). This real wage will remain fixed with technological change in the capitalist sector (assuming, always, a relative scarcity of capital). I see no such economic scenario justifying the assumption of a nondecreasing value of labor power. Socialism Foley is quite insistent that Marx viewed socialism as a society in which money and commodity production (production for exchange, not use) were not predominant (p. 163). Perhaps so. Yet Foley nowhere challenges Marx's view, that is, nowhere suggests that socialism might be compatible with a market economy and commodity production. In my view, socialism-which means the democratic control of the social surplus (profits) by the mass of people who produce it, rather than by a class of capitalists who come to own it under relations of private property in capital-is compatible with a market economy. Public ownership, that is to say, is compatible with a market (though not a laissez-faire) economy, in which profits are allocated to citizens by a political process, and the government intervenes to influence the pattern of investment. If I were to believe what appears to be Foley's view, that socialism is incompatible with commodity production, then I would conclude socialism is either impossible or grossly inefficient in any complex economy. This issue was not so imminent when Foley was writing his book as it is now, after the Fall of 1989. Does Foley conceive of a feasible socialism? Finally, Chapter 5 on the reproduction of capital is an original and transparent reconstruction of Marx's views on the possibility of balanced growth in a capitalist economy. With Foley's help, Marx emerges as a brilliant pioneer of the two sector growth model. This chapter

It immediately follows that wm < 1 (for r > 0). From (12), we see that if a increases to a', and wm does not decrease, then the rate of profit falls: that is, r" < r. Now wm is the "value of labor-power, " the number of hours embodied in the corn the worker can purchase with his wage (p. 36). What Foley has shown is that, if the value of labor-power does not decrease, then rational capitalist innovation will cause the rate of profit to fall. To be methodologically acceptable, both Okishio's assumption of a constant real wage (as technology changes) and Foley's assumption of a constant (or nondecreasing) value of labor power should be short cuts to a full general equilibrium analysis, or class-struggle analysis, of the relationship between the purchasing

This content downloaded from 152.2.176.242 on Mon, 14 Oct 2013 01:13:17 AM All use subject to JSTOR Terms and Conditions

1730

Journal of Economic Literature, Vol. XXVIII (December 1990)

ment relating "value" to "growth," his reasoning resembled a form of verbal algebra, which invited the application of explicit mathematical techniques. That these reformulations have turned out to vary considerably (some are critically reviewed in Morishima's book), suggests that Ricardo may have had in mind more than one model. Before the 1950s, probably the most remarkable contribution on "value" theory was made by Dmitriev, the Russian economist, whose obscure work formed the acknowledged basis of Ladislaus von Bortkiewicz' long paper on "Value Calculation and Price Calculation in Marx," a paper that shed as much light on Ricardo as on Marx and incidently did not fail to do justice to J. S. Mill's early essays on political economy. While Bortkiewicz' paper is still of interest, there is little doubt that a large part of the recent revival of interest in Ricardo's economics is due to Piero Sraffa, especially as the result of his introduction to Volume 1 of Ricardo's Collected Works, edited with the assistance of Maurice Dobb. Since its publication in 1960, Sraffa'sbook, Production of Commodities by means of Commodities, has been a further landmark in this area. While Sraffa'sattempted reconstruction of Ricardo's value analysis has been very influential, Sraffa left out the explicitly dynamic components of Ricardo's economics. That job was performed to a certain extent by Luigi Pasinetti's valuable article published in 1960. However, Pasinetti's mathematical analysis was simplified by a number of assumptions, such as a "fix wage" hypothesis, absence of a production lag in agriculture, and adherence to a labor theory of value. Morishima has generalized Pasinetti's dynamic analysis in a number of useful directions but most notably by relaxing the assumption of a labor theory of value as well as by explicating the time structure involved in Ricardo's production theory. This has made it possible to have an extensive discussion of the "machinery" question in Ricardo, which deviates from the presentation given by John Hicks in Capital and Time by stressing "interindustrial
repercussions."

is, I think, the single best rendition in English of what Capital, Volume II, was trying to accomplish. Understanding Capital is clearlywritten, and the economic models are no more complex than necessary. It displays at work a talented economic theorist and a serious historian of economic thought. I have proposed some challenges, as an innovative work deserves no less, and because I believe, but cannot here amplify upon, that one's views on these matters, and on the real nature of capitalism and socialism, are closely intertwined.
JOHN E. ROEMER

University of California, Davis

REFERENCES

G. A. 1979. "The LabourTheory of Value and the Concept of Exploitation,"Philosophyand Public Affairs, 1979, 7, pp. 338-60. ELSTER, JON. Making sense of Marx. Cambridge: CambridgeU. Press, 1985.
COHEN,
MORISHIMA, OKISHIO,

MicHio.

Marx's economics. Cambridge:

CambridgeU. Press, 1973.

N. "Technical Change and the Rate of

Profit,"Kobe U. Econ. Rev., 1961, 7, pp. 86-99. ROEMER, J. E. A general theory of exploitationand class. Cambridge, MA: Harvard U. Press, 1982. . Free to lose:An introductionto Marxisteconomic philosophy. Cambridge, MA: Harvard U. Press, 1988. Nothe Marxian SAMUELSON, PAUL. "Understanding tion of Exploitation:A Summaryof the So-Called Problem Between MarxianValues Transformation and Competitive Prices,"J. Econ. Lit., June 1971, 9(2), pp. 399-431. Ricardo's economics: A general equilibrium theory of distribution and growth. By MICHIO New York and Melbourne: MORISHIMA. Cambridge University Press, 1989. Pp. viii, 254. $39.50. ISBN 0-521-36630-5. JEL 90-0044 Attempts at mathematical reformulation of Ricardo's economics have a history stretching back over more than one hundred and fifty years. Starting with William Whewell's first attempt in the 1830s, beginning to be better appreciated these days, down to Morishima's present book, theorists of great distinction have tried to dissect Ricardo's classic Principles of Political Economy and Taxation in a mathematical fashion. There are good reasons for this. Apart from the complexity of Ricardo's argu-

Morishima's book seeks to establish a direct intellectual lineage between Ricardo, Marx, and Walras. His analysis on this point would not find favor with economists who draw a rigid

This content downloaded from 152.2.176.242 on Mon, 14 Oct 2013 01:13:17 AM All use subject to JSTOR Terms and Conditions