Andrew J. Kliman 139 Value in Process : A Reply to Naples As Naples notes, there is much about which we agree, especially the validity of Marx s own account of the value-price transformation. Moreover, I welcome her stress on the general issue of value conservation, rather than on the special case of uniform profitability. This serves to redirect attention to the class relations that remain intact despite reallocation of already extracted surplus-value. Her serious examination of contentious issues, however, demands not only praise but an equally serious and similarly focused reply. Such debates are needed if the Marxian critique of political economy is to continue to develop in opposition to the various economic orthodoxies. Historical Class Process vs. Technocratic Fetishism Kliman and McGlone s (1988; hereafter K&M) results contradict the view, held by Naples, Shaikh and the host of transformation problem solutions that find Marx guilty of inconsistency, that changes in values depend solely upon changes in production conditions. Why? We eschewed both In response to Michele Naples critique, the author suggests that the dispute revolves around different concepts of value. Exploring the contrast between Marx s concept of value as a selfmoving substance and technological determinist concepts, this paper further develops Kliman and McGlone s view of the value-price transformation.
140 Capital & Class 51 the technological determinist embodied labour theory of value in which a commodity s value is, by definition, the direct and indirect labour needed to (re)produce it under current production conditions and its market-centred abstract labour alternative. Instead, we argued that Marx s is a production-centred value theory of labour applicable solely to capitalist production s inverted relations of dead to living labour. Marx (1977 p.128) identifies abstract labour as the social substance embodied in commodities values. One can twist and turn a use-value forever without finding any (concrete) labour lurking within. It is thus fetishistic to regard labour embodiment as a suprahistorical technological reality, i.e., as the expenditure of labour common to all production. Embodied labour is instead a phantom-like objectivity that arises from the peculiar social character of the labour that produces [commodities] (ibid. pp.128, 165, emphases added). As Marx (1964 pp.122 23) wrote in Alienated Labour : The worker puts his life into the object, and his life no longer belongs to himself but to the object The alienation of the worker in his product means not only that his labour becomes an object, assumes an external existence, but that it exists independently, outside himself, and alien to him, and that it stands opposed to him as an autonomous power. The life which he has given to the object sets itself against him as an alien and hostile force. That which is called embodiment in Capital is here referred to as life that belongs to the object, labour that exists independently, outside himself, and life given to the object. It should be clear that Marx s embodied labour theory is a theory of abstract, alienated labour. Because labour embodiment is, in this view, a peculiar social process, not a technological requirement as such, the labour embodied in a commodity need not equal the amount of labour needed to (re)produce it. Rather than being analytically derived from current production functions, value is conceived as a self-moving substance value in process (Marx 1977 p.256), extracted in historical time through the capitalist production process.
Value in Process 141 Value As a Self-Moving Substance While Naples argues that K&M s model fails to conserve real value, we agree that the historically determined quantum of already existing value is conserved (redistributed but unaltered in magnitude) by exchange. Yet the redistribution of the substance of value implies that some commodities now embody more of it than before, some less more or less of the labour already extracted is now alienated in them. Naples fails to recognise that, precisely because redistribution does not alter the sum of value in existence, the capital advanced to production does not cease to be a sum of [ real ] value merely because it differs from the values of its material elements, as K&M wrote. The changed labour embodiments remain quanta of the same real value substance, and each commodity s price of production is nothing but the monetary expression of its (changed) labour embodiment. Precisely because value is conserved, furthermore, it is the labour actually embodied in the constant capital, not the imaginary amount that would have been embodied in the absence of value redistribution, that is conserved preserved and transferred to the value of subsequently produced commodities. As a result, total value (constant capital plus living labour) can change even when production conditions do not. Were it otherwise, Marx s theory that the constant capital does not add to (nor subtract from) value could not be sustained. 1 And, as Naples notes perceptively, we treated variable capital and surplus-value in an identical manner, as actual, not imaginary, labour embodiments. 2 In our view, then, value real value consists always and only of the labour extracted from workers in production and embodied in commodities. Once it is recognised that K&M and the discussion above have traced the self-movement of the value substance itself, not merely changes in its monetary expression, Naples distinction between real and nominal value becomes otiose. 3 By the implicit analogy with real use-values and their nominal monetary expression, a nominal change in value seems to mean that an aggregate of use-values costs more or less labour. But so does a real change in value: when mechanisation causes use-values to cost less labour, we say their (real) value has fallen. While Naples would evidently wish to categorise as nominal changes in value only those caused by
142 Capital & Class 51 factors other than technical change, the concept of a nominal change refers to the effect which is identical in both cases. Naples appears to regard inflation (or deflation) as a necessary consequence of the redistribution of value in K&M s illustration probably because she regards the changes in value aggregates themselves as purely nominal ones. Yet, although I have argued that the redistribution of value will typically bring about changes, over time, in the aggregate values of commodities, this does not entail any changes in the price level. If the monetary expression of value remains constant, then a rise (fall) in aggregate value would indeed imply a corresponding amount of price inflation (deflation). K&M s illustration did indeed assume a constant monetary expression of value, but only for simplicity. It seems more plausible that the nominal price level is determined by forces apart from and independent of the transformation of values into prices of production. If this is the case, then the redistribution of value would lead to changes in aggregate value but leave the price level unchanged, implying an alteration in the monetary expression of value. The Constant Price-Expression of Living Labour Naples implies that K&M s model is inconsistent because it requires the price-expression of living labour (PELL) to be constant, though other elements of price change. Despite her emphasis on sequential as against simultaneous causation, she fails to recognise that neither the substance of value nor its monetary expression can be redistributed through exchange until after it has been extracted and embodied in commodities. The asymmetry in K&M s treatment of dead and living labour is simply a temporal one. Nor did the constancy of PELL constitute a concealed invariance postulate, as Naples implies. First, it was not imposed as an additional postulate, but necessarily results from our assumptions that both living labour and the monetary expression of value were constant over time. Second, the latter assumption implied no special constraint on PELL, but applied equally to the other elements of price as well which did change over time. 4 Third, Naples suggestion that constant value-added in price terms imposes a price level in our
Value in Process 143 illustration cannot be true: although PELL remained constant, the price level changed over time. Given the corresponding values, the price level (and indeed every monetary sum) were determined by the assumption of a constant monetary expression of value. Thus, if Naples uses the term invariance postulate to refer to any condition that sets the price level, this was ours. When K&M criticised invariance postulates, however, we used the term to refer to the imposition of a specific, yet arbitrarily dictated, equality between value and price aggregates, without which the value and price systems of Marx s critics would remain entirely unrelated. In contrast, ours was purely a simplifying assumption. Had we assumed a different monetary expression of value, or allowed it to vary over time, the general value and price rates of profit would remain equal, and the aggregate value/price and surplus-value/profit proportionalities would still obtain, in every period. Naples Alternative Nonequilibrium Model Naples statements that circulation [for K&M] causes the value of labour-power and unit constant capital and that we allow circulation to claim precedence over production in determining values, could mislead readers into thinking we hold that circulation processes create value. Because such displacement of the antagonism between capital and labour in production would have serious theoretical and practical consequences, I believe this issue must be treated with absolute precision. K&M in fact depicted the redistribution of value in the sphere of circulation as only one factor that can effect changes in the magnitude of particular value components and one that does not alter the total amount of value already extracted or produce any new value. All new value was generated in production, through the renewed extraction of living labour from workers. Yet Naples regards this as the precedence of circulation over production, and argues that K&M s model fails to conserve real value. The reason is clearly that she believes not only that the redistribution of value should conserve already existing value, but also that alternative distributions of value should have absolutely no effect on subsequent value aggregates. In this conception, production takes precedence
144 Capital & Class 51 not when the capital-labour antagonism is made central to theory, but when technology is privileged: real values should be determined solely by production conditions. While Naples believes that her view is consistent with Marx, and K&M s is not, Marx argues to the contrary that the cost price [can] diverge from the value sum of the elements of which this component of the price of production is composed, even in the case of commodities that are produced by capitals of average composition. Let us assume that the average composition is 80 c + 20 v. the 80 c may be greater or less than the value of c, the constant capital, since this c is composed of commodities whose prices of production are different from their values. The 20 v [variable capital] can similarly diverge from its value, if the spending on wages on consumption involves commodities whose prices of production are different from their values (Marx 1981 p.309). The crucial point is that the feedback of prices on subsequent value aggregates can affect even capitals of average composition. For if such capitals can be affected, then so can the total capital, of which they are but miniature versions. If, moreover, the cost price as a whole can diverge from its corresponding value, the deviations affecting c and v are not mutually compensating. Hence, Marx s text does not support Naples claim that the redistribution of value must leave subsequent value aggregates unaltered. Even according to her own concepts of value and its redistribution, such conservation is not generally possible. Naples (1989) concludes otherwise only because she examines a special case which conforms to two conditions: (1) the input and output values of capital-goods are equal, and (2) balanced growth the ratio of capital-goods to wage-goods is the same for inputs as for outputs. Table 1, which presents two examples of her transformation model, will help make this clear. Departments I, II, and III produce capital-goods, wagegoods, and luxury goods, respectively. 5 Given Naples concepts, redistribution does conserve values intertemporally in example A, which conforms to conditions (1) and (2). The output prices of Departments I and II, 234 and 126, become the C and V of the next period and, since Naples would argue that
Value in Process 145 Table 1. Naples Nonequilibrium Transformation Procedure Example Department Constant Capital Variable Capital Surplus Value output Value Price Profit Price-value ratio real profit A: Balanced Growth C V S I 80 50 50 180 234 104 1.300 95.0 II 60 60 60 180 126 6 0.700 6.0 III 40 70 70 180 180 70 1.000 79.0 total 180 180 180 540 540 180 180.0 I 80 55 55 190 247 112 1.300 104.5 B: Unbalanced Growth II 60 60 60 180 126 6 0.700 6.0 III 40 65 65 170 167 62 0.982 69.5 total 180 180 180 540 540 180 180.0 Notes: 1. All figures except ratios are socially-necessary labour-hours. 2. C = constant capital; V = variable capital; S = surplus-value. 3. To calculate real profit, C and V are revalued in accordance with the pricevalue ratios. the 126 V is only the nominal value of labour-power, but that its real value remains 180, surplus-value (S) remains 180. Total value is conserved in Naples sense, because C+V+S equals 540 (234+126+180), as it would had redistribution not occurred. In example B, however, condition (2) is violated. If the total product is realised, 6 the next period s C and V are 247 and 126 and, since realisation of the total product implies unchanged extraction of living labour, Naples would again say that S remains 180. Here, total value is not conserved in Naples sense, because C+V+S equals 553 (247+126+180), while it would have equalled 550 (190+180+180) had redistribution not occurred. In fact, in example B and in general, the value aggregates can remain unaltered only if no redistribution occurs! Naples model suffers from such flaws because her comparative static idea of conservation clashes with the dynamic, historical process of valuation into which she imports it. Her formalist methodology prevented her from undertaking the dialectical tracing of value in process needed to comprehend how value is both conserved and not conserved unaltered by any act of exchange but changing over time. Naples instead tries to impose upon the data, by means of a priori invariance postulates, her preconception that value is determined by production conditions alone. Her model in this respect replicates the method, and the technological determinist
146 Capital & Class 51 concept of value, of traditional transformation problem solutions. And just as they sever all intrinsic links between values and prices which differ from values, Naples model must do so as well or fall into self-contradiction. The Rate(s) of Profit Naples statement that K&M agree with Shaikh (1977) that the Neo-Ricardian model is ultimately right on the magnitude of the profit rate [and Marx] is ultimately wrong is completely unjustified. Her less extravagant claim that we tried to reconcile the Marxian and Sraffian profit rates is also incorrect. By tracing value in process, we showed that the general profit rate is always the ratio of surplus-value to the capital-value advanced, even when it differs from the profit rate corresponding to the technologically determined value system of the transformation problem, and even when the mathematical tautologies of the Sraffian equilibrium price system obtain. Hence, rather than trying to reconcile Marx s procedure with Sraffian solutions a la Shaikh, we were undermining the Sraffian model. It is true that our illustration depicted a convergence of the Marxian profit rate to the Sraffian rate. Yet once our fundamental conception of value and price determination our general model is disentangled from that particular illustration, such convergence does not validate Naples claims. Although she recognises that simple reproduction, constant technology, and uniform profitability are special assumptions bearing no intrinsic relation to the value-price transformation, she nonetheless conflates an illustration that employed these assumptions with our general model. She cannot have it both ways. Thus, while K&M presented an illustration that converged to static equilibrium (again, to demonstrate that the Sraffian solutions fail to refute Marx s transformation procedure), this does not mean that our model reconciles Marx with the Sraffians, even in a quantitative sense. Any apparent reconciliation was only an incidental byproduct of the illustration, because the model vindicates Marx s procedure and results without relying on i.e., even in the absence of that demonstration and the specific assumptions that produced convergence.
Indeed, once the specific assumption of constant technology is dropped, K&M s model demonstrates not a convergence, but a systematic divergence, of the Marxian and Sraffian profit rates. Assuming continuous mechanisation instead of constant technology, but employing the same conception of value in process put forth in K&M and reviewed here, I (Kliman 1988) have demonstrated that the former may fall under conditions in which the latter must rise; if extraction of living labour fails to increase over time, the former must eventually fall to zero despite any and all increases in productivity and the rate of exploitation. Hence, in opposition to the Okishio Theorem (Okishio 1961), which employs the Sraffian model in its attempt to disprove Marx s theory that mechanisation causes the rate of profit to fall, our model vindicates Marx here as well. These results stem from the fundamental irreconcilability of the concept of value in process and the static equilibrium, technological determinist character of the Sraffian model. Our model is likewise independent of the assumption of uniform profitability. 7 We followed Marx in abstracting from the real inequality in profit rates simply because we sought to defend his account of the value-price transformation. Because we did not seek to advance over Marx or to model the actual operation of the capitalist economy, the uniform profitability assumption should not be construed as a claim that it is attained in reality. I thus find Naples recognition that discrepancies between the Marxian and Sraffian profit rates can persist due to nonuniform profitability a welcome advance in the literature, not a challenge to K&M s model. Yet uniform profitability is theoretically possible, and its attainment under conditions of constant technology will produce the convergence of the Marxian and Sraffian profit rates, even if it is the reproduction cost-based profit rate (Naples real rate) instead of the historical cost-based profit rate (Naples nominal rate) that is equalised. 8 & 9 For instance, had K&M assumed the uniformity of the former rather than the latter, the convergence of relative prices and the general real rate of profit to the Sraffian equilibrium would have been immediate, but our results would have remained identical in all other important respects. It is thus difficult to understand why Naples places such great emphasis on the contrast between the two rates, or why she implies that our model depends crucially on the assumption of a uniform nominal rate. Value in Process 147
148 Capital & Class 51 She does raise thought-provoking questions concerning the legitimacy of this assumption. I cannot resolve this issue definitively here, but offer only a few comments. The profit rate that capitalist competition tends to equalise, real or nominal, depends on what capitalists seek to maximize. Naples appears to suggest that they quest after luxury goods. If that is correct, then competition will, as she notes, tend to yield uniform real profit rates. Yet equal real profit rates generally imply that capitals realise disproportionate amounts of actually existing surplus-value. Thus, if capitalists quest after surplus-value, wealth in the abstract, and finance luxury consumption from residual proceeds after requisite outlays for capital have been made, as depicted in K&M, then competition will tend to yield uniform nominal profit rates. There should be little doubt concerning Marx s views on this matter. He insisted repeatedly that the appropriation of ever more wealth in the abstract is the sole driving force Use-values must therefore never be treated as the immediate aim of the capitalist (Marx 1977 p.254). And in the only passage in which, to my knowledge, he addresses the question of real and nominal profit explicitly, Marx writes that abstract enrichment is solely expressed as such in money, and quotes Thomas Chalmers approvingly: The great object of the monied capitalist, in fact, is to add to the nominal amount of his fortune (Marx 1988 p.100). Conclusion K&M showed that Marx s account of the value-price transformation no longer seems internally inconsistent once it is conceived as a dialectical self-development in which value becomes price of production, rather than as a failed attempt to reconcile self-contained systems of values and prices in a formalistic manner. Although Naples has echoed the charge of internal inconsistency against us, not Marx, this reply has shown that dialectics vs. analytical formalism remains at issue. By tracing value as a self-moving substance, unencumbered by technological determinist preconceptions imposed as a priori axioms, the reply has revealed the cogency and consistency of our model and helped elucidate its radical break from Sraffianism. In contrast, while Naples Nonequilibrium model
Value in Process 149 also successfully challenges the criticisms lodged against Marx s transformation procedure in some respects, especially insofar as such criticisms rest on the assumption of uniform profitability, it ultimately suffers either from internal inconsistency or an inability to account for prices that deviate from values. Ted McGlone is not responsible for any statements contained herein, even when they refer to jointly held views. I wish to thank him, along with Michele Naples, Alfredo Saad-Filho, the editors of Capital&Class, and participants in the 1992 Eastern Economic Association panel on State Capitalism and the 1990s for their helpful comments. Acknowledgement 1. If, for example, 20 labour-hours were originally embodied in constant capital but, after redistribution, only 15 labour-hours are now embodied, and if 10 living labour-hours are added, the actual value of output is 25. Only by maintaining that the use of constant capital creates 5 extra labour-hours or by denying that price movements redistribute value can one claim that the value of output is 30. 2. Naples is not quite correct that K&M s model is unique in making the rate of exploitation endogenous to the pricing procedure ; Giussani (1991 92) has recently proposed a substantially identical model. Also, after our paper was accepted, Glick and Ehrbar (1987) published an iterative version of the new solution which similarly recognises that variable capital and surplus-value are actual labour embodiments. Yet, retaining an ahistorical, technological determinist conception of value, they fall into inconsistency by failing to recognise that constant capital is likewise the labour actually embodied in capital-goods and thus reject Marx s account as self-contradictory. 3. Naples is correct to criticise K&M s imprecise discussion of money. Despite our mention of the value of money, we intended to posit the existence only of a money of account, not commodity money (and we did not assume that gold is money, as Naples states). This terminological imprecision has no bearing on the other issues under contention. I refer throughout this reply to money of account that expresses value. Notes
150 Capital & Class 51 4. That aggregate prices underwent change, despite a constant relation between the monetary unit and each labour-hour of value, should be sufficient to demonstrate that the price level changes were likewise changes in real value. 5. I also follow Naples in imposing two invariance postulates: total price is the monetary expression of total value, and total real profit (total price minus reproduction costs) is the monetary expression of total surplus value. For simplicity, Table 1 ignores the monetary expression of value, depicting the labour-time equivalents of prices and profits. 6. An infinite number of growth combinations would make its realisation possible. One possibility is that Department I expands by 45%, II contracts by 46.67%, and III expands by 5%. 7. Although Naples states that K&M motivated this assumption in terms of its association with equilibrium, the passage to which she refers merely argued that, while prices which assure equal profitability require the equilibration of supplies and demands, they do not require stationary prices. 8. Naples (1989) finds uniform profitability to be impossible, not because she assumes capitalists seek the highest real profit rate, as some readers of her present critique may infer, but because she imposes invariance postulates that prevent the redistribution of value from altering the value of the capital advanced. 9. Although Naples uses the same terminology in both cases, the controversy over which profit rate (if either) should be assumed to be uniform is unrelated to the question of whether K&M s values are real or nominal. References Giussani, P. (1991 92) The determination of prices of production in International Journal of Political Economy, Vol.21 No.4, pp.67 86. Glick, M. and H. Ehrbar (1987) The transformation problem: an obituary in Australian Economic Papers, Vol.26 No.4, pp.294 317. Kliman, A.J. (1988) The profit rate under continuous technological change in Review of Radical Political Economics, Vol.20 Nos.2&3, pp.283 89. Kliman, A.J. and T. McGlone (1988) The transformation nonproblem and the non-transformation problem in Capital&Class 35, Summer, pp.56 83.
Value in Process 151 Marx, K. (1964) Alienated labour in T.B. Bottomore [ed.] Karl Marx: Early writings. McGraw-Hill, New York. (1977) Capital: a critique of political economy, Vol.I. Vintage, New York. (1981) Capital: a critique of political economy, Vol.III. Vintage, New York. (1988) Karl Marx, Frederick Engels: collected works, Vol.30. Progress Publishers, Moscow. Naples, M.I. (1989) A radical economic revision of the transformation problem in Review of Radical Political Economics, Vol.21 Nos.1&2, pp.137 58. Okishio, N. (1961) Technical changes and the rate of profit in University Economic Review 7, pp.85 99. Shaikh, A. (1977) Marx s theory of value and the transformation problem in J. Schwartz [ed.] The Subtle Anatomy of Capitalism. Goodyear, Santa Monica, CA.